Citation
The effects of pyridoxine 5'-β-D-glucoside on the metabolic utilization of pyridoxine in rats and humans

Material Information

Title:
The effects of pyridoxine 5'-β-D-glucoside on the metabolic utilization of pyridoxine in rats and humans
Creator:
Gilbert, Joyce Ann
Publication Date:
Language:
English
Physical Description:
ix, 106 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Dosage ( jstor )
Enzymes ( jstor )
Food ( jstor )
Liver ( jstor )
Metabolism ( jstor )
Oxidases ( jstor )
Phosphates ( jstor )
Pseudomyxoma peritonei ( jstor )
Rats ( jstor )
Vitamins ( jstor )
Dissertations, Academic -- Food Science and Human Nutrition -- UF
Food Science and Human Nutrition thesis Ph. D
Human beings -- Physiology ( lcsh )
Rats -- Physiology ( lcsh )
Vitamin B6 -- Physiological effect ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1991.
Bibliography:
Includes bibliographical references (leaves 92-105).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Joyce Ann Gilbert.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
027670519 ( ALEPH )
26272875 ( OCLC )

Downloads

This item has the following downloads:


Full Text











THE EFFECTS OF PYRIDOXINE 5'-0-D-GLUCOSIDE
ON THE METABOLIC UTILIZATION OF PYRIDOXINE
IN RATS AND HUMANS


















By

JOYCE ANN GILBERT


















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1991


UtasvRSITYOFJLORIlDA SI(iIUES
































This work is dedicated to my parents, Helen and Reginald,

who showed me life has no boundaries, only opportunities.















ACKNOWLEDGEMENTS


I thank Drs. Lynn Bailey, Peggy Borum, Robert Cousins and

James Cerda for allowing me the previlage of studying with the

best scientists in their respective fields.

Special thanks go to my advisor and research mentor, Dr.

Jesse Gregory, for his example of integrity and dedication in

all endeavors.

I also thank Doris Sartain and my collegues in the

"Yellow Lab" for their endless patience and sensitivity.




























iii















TABLE OF CONTENTS



ACKNOWLEDGEMENTS . ................. iii

LIST OF TABLES . .................. . vi

LIST OF FIGURES . ................. . vii

Abstract . . . . . . . . . . . . . . . . . . . . .. viii

CHAPTER 1
INTRODUCTION . .................. .. 1

CHAPTER 2
LITERATURE REVIEW ......... .......... 4
Bioavailability of Vitamin B-6 .......... 4
Pyridoxine-f-D-Glucoside ............. 15
Factors Affecting the Bioavailability of Vitamin
B-6 . ...... . . ..... ...... 21
Enzymatic Interconversion of B-6 Vitamers . . . . 31

CHAPTER 3
PYRIDOXINE-5'-f-D-GLUCOSIDE AFFECTS THE METABOLIC
UTILIZATION OF PYRIDOXINE IN RATS . . . . . . . . . .. 38
Introduction ................... 38
Materials and Methods . ............. 39
Protocol . . . ............... 39
Forms of vitamin B-6 . ........... . 40
Sample preparation . ........... . 41
HPLC equipment and analysis . ....... . 42
Measurement of radioactivity . ....... 43
Statistical analysis . ........... 43
Results . .................. . . . 44
Discussion . .................. . 48

CHAPTER 4
EFFECTS OF PYRIDOXINE-5'-f3-D-GLUCOSIDE ON THE
METABOLIC UTILIZATION OF PYRIDOXINE IN HUMANS . . . . 55
Introduction . ................ . . 55
Materials and Methods . . . . . . . . . . . ... . 56
Synthesis of forms of vitamin B-6 . . . . . . 56
Protocols of trials with human subjects . . . 57
Analytical methods . ............ 59


iv









Mass spectral analysis of deuterium-labeled
4PA . ...... . .... . ... . . . 59
Statistical analysis . ........... 60
Results ...... .... ..... ... ..- * . . . . 61
Vitamin B-6 nutritional status of subjects 61
Stable-isotopic trials . ......... . 61
Discussion . .................. . 65

CHAPTER 5
EFFECTS OF PYRIDOXINE-5'-p-D-GLUCOSIDE ON THE
IN VITRO KINETICS OF PYRIDOXAL KINASE AND PYRIDOXAMINE
(PYRIDOXINE)-5'PHOSPHATE OXIDASE IN RAT LIVER . ... . 69
Introduction . . . . . . . . .......... 69
Materials and Methods . ............. 70
Protocol . ....... . ....... 70
Forms of vitamin B-6 . .... ....... . 71
Pyridoxal kinase . . . . . . . . . . . . . . 72
Pyridoxamine (pyridoxine) 5' phosphate
oxidase . .. ............. 75
Statistical analysis . ........... 77
Results . ................... . . 77
PL kinase . . .. ............. . 77
PMP (PNP) oxidase . . . . . . . . . . . . .. 79
Discussion . ................ . . . 83

CHAPTER 6
SUMMARY AND CONCLUSIONS . ......... ..... . 86

LITERATURE CITED . ................. . 92

BIOGRAPHICAL SKETCH . ............... . 106























v














LIST OF TABLES


Table 3.1 Liver B-6 vitamer distribution and total
liver 14C in rats administered varying levels of PN-
glucoside in the dose (Experiment 1).* . ..... 44

Table 3.2 Urinary 4PA and total urinary 14C in rats
administered varying levels of PN-glucoside
(Experiment 1)* . . . . . . . . . . . . . . . . . 45

Table 3.3 Liver B-6 vitamer distribution and total
liver 14C in rats administered varying levels of PN
or PN-glucoside. (Experiment 2)* . ..... . . 46

Table 3.4 Urinary 4PA and total urinary 14C in rats
administered varying levels of PN or PN-glucoside.
(Experiment 2)* . . . . . . . . . . . . . . . .. 47

Table 4.1 Indicator of vitamin B-6 nutritional status
of human subjects 24 h prior to each trial' . . . 62

Table 4.2 Molar isotopic ratio of urinary d24PA.' . . 63

Table 4.3 Percentage of d2PN (5Amol) dose excreted as
urinary d24PA . . . . . . . . . . . . . . . . . . 64

Table 4.4 Total urinary 4PA excretion.1 . . . . . . . 65

Table 5.1 Purification of pyridoxal kinase from rat
liver. . . . . . . . . . . . . . . . . . .. .. . 78

Table 5.2 Kinetic parameters of pyridoxal kinase with
varying concentrations of PN-glucoside. ..... 79

Table 5.3 Purification of pyridoxamine (pyridoxine)
phosphate oxidase from rat liver. . . . . . . . . 80

Table 5.4 Kinetic parameters of pyridoxamine
(pyridoxine) phosphate oxidase with varying
concentrations of PN-glucoside. ...... ... . 80





vi














LIST OF FIGURES


Figure 5.1 Double reciprocal plot for the
phosphorylation of PN by PL kinase from rat liver
in the presence of various levels of PN-
glucoside. ............... . . .... . . . 81

Figure 5.2 Double reciprocal plot for the conversion of
PMP to PLP by PMP (PNP) oxidase from rat liver in
the presence of various levels of PN-glucoside. . 82




































vii














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

THE EFFECTS OF PYRIDOXINE 5'-P-D-GLUCOSIDE
ON THE METABOLIC UTILIZATION OF PYRIDOXINE
IN RATS AND HUMANS

By

Joyce Ann Gilbert

December, 1991


Chairperson: Dr. Jesse F. Gregory III
Major Department: Food Science and Human Nutrition

A major form of vitamin B-6 in plant-derived foods is 5'-

O- (-D-glucopyranosyl) pyridoxine (PN-glucoside) . Previous

studies have shown that PN-glucoside is poorly available as a

source of vitamin B-6 in rats and undergoes incomplete

utilization in humans. The present research was conducted to

determine whether unlabeled PN-glucoside affects the metabolic

utilization of simultaneously administered isotopically

labeled pyridoxine (PN) in rats and humans. In addition, the

in vitro effect of PN-glucoside on the activity of enzymes in

the vitamin B-6 metabolic pathway, specifically pyridoxal-

kinase and pyridoxamine (pyridoxine)-phosphate oxidase was

determined.

Experimental results with rats given [14C]PN indicated

that urinary excretion of 14C increased significantly with


viii








increasing dose of PN-glucoside, while hepatic 14C decreased

significantly as the PN-glucoside dose increased. The

proportion of hepatic 14C-labeled pyridoxal, PN, and

pyridoxamine decreased whereas hepatic pyridoxine phosphate

and pyridoxal phosphate increased in proportion to the PN-

glucoside dose. In addition, the concentration of urinary

[14C] 4-pyridoxic acid (4PA), relative to total urinary 14C,
decreased as the dose of PN-glucoside increased. Stable

isotopic methodology was employed to determine whether PN-

glucoside affected the metabolic utilization of simultaneously

administered deuterium-labeled PN (d2PN) in humans.

Experimental results showed that twenty-four hour urinary

excretion of 4PA was decreased significantly with increasing

dose of PN-glucoside. The percentage of ingested d2PN

excreted as d24PA showed an inverse relationship, that was

statistically significant, in proportion to the PN-glucoside

dose. In vitro enzyme assays indicated that PN-glucoside had

no significant effect on the activity of partially purified

pyridoxal-kinase and pyridoxamine (pyridoxine) phosphate

oxidase. These results provide evidence that PN-glucoside

weakly retards the metabolic utilization of nonglycosylated

forms of vitamin B-6. However, the effect of PN-glucoside on

PN is not due to the direct effect of PN-glucoside on the

enzymes PL kinase and PMP (PNP) oxidase.





ix














CHAPTER 1
INTRODUCTION





Vitamin B-6 has often been referred to as the protein

vitamin because of its association with amino acid metabolism

and the influence of dietary intakes of protein on vitamin B-6

requirements (1-3). Adequate vitamin B-6 nutriture is

essential to health through its numerous roles in the body as

the active coenzyme form, pyridoxal 5'-phosphate (PLP). A

large majority of these roles involves metabolism of amino

acids. These functions include, among others, nonoxidative

decarboxylation of amino acids, transamination,

desulfhydration, and enzymes affecting reactions of amino acid

side chains. Such PLP dependent enzymes are important

functional complexes in the biosynthesis and catabolism of

essential and non-essential amino acids and some provide a

connection between the amino acid and intermediates of

carbohydrate metabolism. Recent research in animal models has

suggested a role for PLP in the modulation of hormone actions

(4-5).

Nutritional status with respect to vitamin B-6 is

influenced by the amount of the vitamin ingested, the extent


1








2

of absorption and metabolic utilization of the B-6 compounds

in the diet, and the specific requirements of the individual.

Although vitamin B-6 is widely distributed in nature, the

amount in food is relatively small. The American diet often

contains 1 - 2 mg per day (2). Marginal vitamin B-6

nutritional status has been well documented in several

segments of the American population including adolescents,

pregnant women and the elderly (6-15). However, vitamin B-6

deficiency with apparent clinical symptoms is not widespread

in the general population. It is significant that over 50% of

the subjects of the USDA 1977-78 Nationwide Food Consumption

Survey consumed less than 70% of the Recommended Dietary

Allowance (RDA) for vitamin B-6 (2). The 1985 Continuing

Survey of Food Intakes of Individuals indicated that only 27%

of women consumed 70% or more of the RDA for vitamin B-6 (12).

Women of all ages and the elderly have been especially

prevalent in the group consuming less than the RDA for vitamin

B-6. Vitamin B-6 is required in amino acid metabolism and

therefore, the RDA for this vitamin is related to protein

intake. Twice the RDA for protein, which is considered the

upper boundary of acceptable level of protein intake, was used

to establish the RDA for vitamin B-6. The RDA for vitamin B-6

was revised using a dietary ratio of 0.016 mg vitamin B-6/g

protein.

Our present knowledge of the essential role of vitamin B-

6 in preserving health emphasizes the importance of








3

establishing an accurate RDA for vitamin B-6. Establishing an

adequate RDA for vitamin B-6 is a complex task. The

appropriateness of the RDA for vitamin B-6 is dependent on

accurate information on the content of the vitamin in foods

consumed, the enzymes involved in the metabolism of the

various forms of vitamin B-6, as well as information regarding

vitamin B-6 bioavailability.














CHAPTER 2
LITERATURE REVIEW



Bioavailability of Vitamin B-6


The concept of bioavailability is most appropriate when

considering the extent of intestinal absorption and metabolic

utilization and any potential antagonistic effects of all

naturally occurring dietary forms of the vitamin. Hence, net

bioavailability is defined as the portion of total dietary

vitamin B-6 that is biologically active following intestinal

absorption (16). Vitamin B-6 as it occurs in mammalian

tissues and fluids encompasses three interconvertible 3-

hydroxy-2-methylpyridine compounds including pyridoxine (PN),

pyridoxal (PL), pyridoxamine (PM), their corresponding 5'-

phosphoryl derivatives (PNP, PLP, and PMP respectively) and

the excretory catabolite 4-pyridoxic acid (4PA) (17). Vitamin

B-6 in nature is found in phosphorylated forms and closely

associated with proteins. In foods, PL, PN, and PM are found

as both phosphorylated and nonphosphorylated forms. PL, PM,

PLP, and PMP are the predominant forms in animal products,

while PN, in its free vitaminic form and as the glycoside

(discussed in a later section) are the major forms in most

plant foods (18).


4








5

To quantitate total vitamin B-6 in most food, test

microorganisms are generally utilized. The most common

organism used is Saccharomyces uvarum. Test microorganisms

demonstrate low biological activity of vitamin B-6 in its

bound (pyridoxyl-amino acid and amines) glycosylated and

phosphorylated forms until they are released by acid

hydrolysis. However, this acid treatment of food is not

entirely representative of the digestive process in the

gastrointestinal tract.

Results of early studies of intestinal absorption of

vitamin B-6 indicated that absorption occurs by a simple

passive diffusion mechanism (19-22). A number of studies

indicate that the rate of uptake by intestinal tissue of the

B-6 vitamers, PN, PM, and PL increased proportionally to the

dose over a wide range of intralumenal concentrations (19,23-

31). Transport of PN across the intestinal wall occurred by

diffusion, independent of accumulation within the tissue.

There is no evidence of a saturable transport system either in

vivo (19,23-24) or in vitro (23,26-31). These data also

indicate that transport across the intestinal wall from the

mucosal to the serosal side of the non-phosphorylated B-6

vitamers to be quantitatively similar. PN in the free and

conjugated forms is rapidly absorbed by the intestine directly

or following hydrolysis by intestinal enzymes. PL and PM are

present in food primarily as the 5'-phosphates and require

hydrolysis by lumenal phophatases. The phosphorylated B-6








6

vitamers, PLP and PMP are well absorbed following enzymatic

dephosphorylation, although these vitamers are slowly absorbed

as the intact 5'-phosphate esters (28-30). Studies have

indicated that the transport of the B-6 vitamers into the

intestinal mucosal cell involves a saturable enzymatic

process, identified as intracellular phosphorylation by PL

kinase (EC 2.7.1.35) (24, 28-29). Most ingested vitamin B-6

is absorbed by the jejunum in the nonphosphorylated forms, PL,

PN, and PM. Tissues, especially liver, rapidly take up

circulating vitamin B-6, where the phophorylated B-6 vitamers

are hydrolyzed by plasma membrane phosphatases and enter the

cells by a facilitated process and diffussion followed by

metabolic trapping (28-29). PN is phosphorylated to PNP by PL

kinase which has been detected in all mammalian tissues

investigated (32). PNP is then converted to PLP by flavin-

dependent PNP oxidase (EC 1.4.3.5) (33) which, in contrast to

the widely distributed PL kinase, is found in few tissues,

mainly liver, erythrocytes, kidney, and brain (34-35). PLP

can be transformed to PMP by transamination or hydrolyzed to

PL by phosphatases (17). PL is then converted to pyridoxic

acid (4PA) by aldehyde oxidase (18) or again to PLP by PL

kinase. PLP is bound by cellular protein or released into

plasma, by the cell, as PLP, PL or 4PA. The analysis of

kinase and phosphatase enzymes in liver tissue indicates

similar activities. The relative activities of these enzymes

account for the accumulation of the 5'-phosphate compounds;








7

however, it also allows for the dephosphorylation of PLP that

is not protein bound and its release from the liver as PL or

oxidized to 4PA (31). Regulation of vitamin B-6 metabolism

may also occur through the conversion of PNP and PMP to PLP,

which is highly sensitive to product inhibition (30).

Recent advances in analytical techniques have enabled

collection of more precise data concerning the quantity and

various forms of vitamin B-6 compounds present in the diet.

However, present knowledge of the bioavailability of vitamin

B-6 is still not sufficient enough to determine an accurate

assessment of adequate dietary intake.

The bioavailability of a nutrient such as vitamin B-6 is

primarily determined experimentally by comparing the

concentration of vitamin B-6 that is biologically active to

total vitamin B-6 ingested (16). Traditionally biologically

available vitamin B-6 has been determined through animal

bioassays (36). Human bioassays, to determine biologically

active vitamin B-6 in food, have been performed, but are

generally lengthy procedures with limited precision.

Significant advances and recent applications in isotopic

methodology offers a useful alternative to bioassay for the

sutdy of bioavailability of vitamin B-6 (36). Results of

studies utilizing rat bioassays to compare animal products

with plant-derived products have indicated that the

bioavailability of vitamin B-6 in animal products is greater

than that of plant-derived products (36). In general, a poor








8

correlation was observed in these samples when comparing rat

growth and plasma PLP concentration. These results were

difficult to interpret because of the potential effect of diet

composition on the synthesis of vitamin B-6 by intestinal

microflora. The lack of agreement between rat growth and

plasma PLP, suggests that data from bioassays to determine

biologically available vitamin B-6 in food sources,

specifically those in which the test diets may differ greatly

from the reference diet in the type of carbohydrate, are

somewhat equivocal (37). The bioavailability of vitamin B-6

has been studied in somewhat more detail than many of the

other vitamins. However, few generalizations can be stated

regarding the overall bioavailability of the B-6 vitamers or

the factors influencing it.

Essential to measuring the bioavailability of vitamin B-6

in a food source is the determination of the total vitamin B-6

content in the food. This is complicated by the interaction

of the vitamin with food components. These interactive

compounds may or may not be biologically available. The

interactions of vitamin B-6 with food components and with

other vitamin B-6 compounds are important considerations since

they may represent sources of available vitamin B-6 in the

diet. Experimental data regarding the bioavailability of

naturally occurring vitamin B-6 in foods were initially

reported by Sarma et al. (34). The authors utilized a

comparison of rat bioassay and Saccharomyces uvarum assay








9

results for a variety of plant and animal products. Liver

fractions, whole wheat and yellow corn exhibited

bioavailability values of between 65% and 70% compared with an

apparent value of approximately 100%. These results were the

first indication of a wide variation and potentially

incomplete bioavailability of vitamin B-6 in cereal grains and

other food products (34).

A comparison of rat growth bioassays and Saccharomyces

uvarum data for dried beef, lima beans, non-fat dry milk and

whole wheat flour yielded a good correlation between the

assays with small differences obtained for the flour and milk

products (35). Nelson et al. (38) examined the rate of

intestinal absorption of the vitamin B-6 in orange juice in

human subjects. A triple lumen perfusion technique was

employed to determine the relative absorption of naturally

occurring vitamin B-6 in orange juice. Their results

demonstrated a significant decrease in rate of absorption of

vitamin B-6 in orange juice (42%) than that of synthetic B-6

vitamers in saline (67%) and saline-glucose solutions (79%).

The lower rate of absorption of the naturally occurring

vitamin B-6 in orange juice was thought to be a result of

interactions of the vitamin with naturally occurring food

components in the orange juice (38). A follow-up study

conducted by Nelson et al. (39) investigated the nature and

extent of binding of different forms of vitamin B-6 in orange

juice. The study suggested extensive and equal binding of








10

both PL and PN in orange juice to a small dialyzable molecule

which is heat stable and non-protein in nature. The

interaction of vitamin B-6 with this component of orange juice

was suggested as being responsible for the lower rate of

absorption of naturally occurring vitamin B-6 in orange juice.

Leklem and coworkers (40) were the first to employ a

bioassay with human subjects to examine the effect of dietary

fiber on the bioavailability of the vitamin. The study

investigated the bioavailability of vitamin B-6 in white and

whole wheat bread. Significant differences in the

bioavailability of B-6 vitamers were not apparent when plasma

PLP and erythrocyte aminotransferase data were examined.

However, fecal vitamin B-6 and urinary 4PA excretion data

suggested incomplete utilization of the vitamin in the whole

wheat bread. Frequent attempts have been made to calculate

the vitamin B-6 balance of B-6 vitamers and 4PA in bioassays

employing human subjects (41).

Similar experiments in which the use of data concerning

fecal vitamin B-6 have shown little validity. Unabsorbed

vitamin B-6 from dietary sources as well as microbiologically

synthesized B-6 vitamers are contained in fecal material. The

composition of the diet will influence the microbial

contribution to fecal vitamin B-6 which may affect the

apparent vitamin B-6 nutriture of the subject. The fact that

the intestinal microorganisms produce vitamin B-6 is well

established (41-46). However, the availability of vitamin B-6








11

to the host organism, especially in the absence of coprophagy,

has been unclear (45,47). Since several animals can develop

a vitamin B-6 deficiency even when coprophagy is allowed,

metabolic utilization of microbial vitamin B-6 must be miminal

(48). In animals exposed to conditions of nutritional

deficiency, coprophagy may increase sufficiently to become a

significant source of nutrients. This may explain the results

of Ikeda et al. (49-50), who observed that germfree rats tend

to be more susceptible to vitamin B-6 deficiency than rats

with normal populations of intestinal microorganisms. Hughes

et al. (51) reported that in addition to the intestinal

microflora, nutritionally deprived animals may also benefit

from other environmental sources of microbial vitamins such as

microbial activity on the floor of the pen. As indicated by

these studies, there exists no compelling evidence that

vitamin B-6 can be absorbed from the large intestine in

amounts sufficient to make a detectable contribution to the

daily intake.

A study (52) involving human bioassay in measuring

bioavailability of vitamin B-6 in beef compared with soybeans

suggested the vitamin B-6 in soybeans to be less available

compared to that contained in beef. It has been suggested that

the presence of nondigestible polysaccharides and lignin

components of dietary fiber of the plant derived foods is

responsible for differences in the availability of vitamin B-6

in plant foods compared with animal products (40). A variety








12

of physical and chemical properties of dietary fiber suggest

the possibility of binding or entrapment of the B-6 vitamers,

which may influence intestinal absorption. Schultz and Leklem

(53) did a study comparing vegetarian versus nonvegetarian

women. They observed that although the vegetarian women

consumed more crude fiber than the nonvegetarian women, there

was no significant differences between these two groups for

plasma PLP, urinary 4PA, and urinary vitamin B-6. Therefore,

it was concluded that there appeared to be no adverse effect

of fiber on the bioavailability of vitamin B-6 between these

groups (53). It has been reported that human intestinal

microflora produce vitamin B-6 (54). Several bioavailability

studies have indicated that ingestion of diets high in dietary

fiber or carbohydrate leads to significant increases in

microbial synthesis of vitamin B-6 (55-56).

Tarr et al. (57), evaluated a typical American mixed diet

to its bioavailability of vitamin B-6. They reported 70%

bioavailability of vitamin B-6 based on urinary 4PA and plasma

PLP concentration relative to PN in a formula diet. The

authors speculated that thermal processing was potentially

responsible for the incomplete bioavailability of vitamin B-6

because canned goods, including both animal and plant foods,

composed much of the mixed diet. A study by Nguyen and

Gregory (36) employed rat bioassays to examine the

bioavailability of vitamin B-6 in selected foods as influenced

by thermal processing. The effects of food composition and








13

thermal processing on the relative bioavailability of vitamin

B-6 in beef, spinach, potato and cornmeal were assessed. The

results further demonstrated that diet composition and food

processing must be considered in evaluating the

bioavailability of vitamin B-6.

The instability of certain B-6 vitamers during food

processing and storage may contribute to losses of the

nutritional quality of foods with respect to vitamin B-6.

Processing and storage of foods have been found to have

various effects on vitamin B-6 with losses decreasing the

adequacy of food products as sources of dietary vitamin B-6.

A deficiency of vitamin B-6 in infants who consumed a non-

fortified, heat sterilized canned infant formula, led to

extensive research, during the 1950's, concerning the effects

of food processing on the bioavailability of vitamin B-6 (58).

Hassinen et al. (59) demonstrated that the two main

naturally occurring B-6 vitamers in milk, PM and PL, were much

less stable than added synthetic PN. The research of

Tomarelli et al. (60) reported that the retorting of milk and

infant formula induced large losses in the availability of

naturally occurring vitamin B-6 in these products.

Very few studies have evaluated the thermal stability of

vitamin B-6 in low moisture food systems. Vitamin

fortification is common practice in the food industry and will

more than likely continue into the future with more and more

types of products being fortified. Vitamin B-6, in the form








14

of PN-HC1, is added to many breakfast cereals at levels of

25% to 100% of the United States Recommended Daily Allowance

(USRDA) per ounce. Gregory et al. (61-62), using a dehydrated

model food system, simulating breakfast cereals, indicated

that the roasting and storage of low moisture food systems

resulted in losses of 50% to 70% of the added PN,PM,and PLP.

The remaining vitamin B-6 was found to be fully available as

determined by rat bioassay using growth, feed efficiency,

erythrocyte aspartate aminotranferase activity (AspAT) and in

vitro (AspAT) coenzyme stimulation. Gregory (63) also

examined vitamin B-6 bioavailability in rice-based, PN-

fortified cereal and non-fat dry milk. The non-fat milk and

rice base breakfast cereal samples were analyzed for vitamin

B-6 by microbiological, HPLC, and rat bioassay procedures.

The results indicated that the vitamin B-6 of the non-fat dry

milk was fully available, while the vitamin B-6 availability

in the cereal product was comparatively low. The apparent

losses of the PN in the fortified cereal was explained by a

first order kinetic model described by Evans et al. (64).

These studies (61,63-64) indicated that food fortified with PN

is susceptible to significant degradation under certain

processing and storage conditions and that the bioavailability

of the remaining vitamin B-6 may not be complete. The above

results suggest that thermal processing of foods does not

induce nutritionally important losses in the bioavailability

of vitamin B-6. At the very least, research to date indicates








15

that any possible adverse effects of thermal processing and

storage on the bioavailability of vitamin B-6 would not be of

sufficient magnitude to explain the incomplete utilization of

the vitamin observed in human subjects (57,65).


Pvridoxine-fB-D-Glucoside


A conjugated form of pyridoxine, 5'-0-(P-D-

glucopyranosyl) pyridoxine (PN-glucoside) has been found to be

a major naturally occurring form of vitamin B-6 in many fruits

and vegetables of human diets. Scudi et al. (66) first

reported the presence of conjugated forms of the vitamin B-6

in 1942 to occur in rice bran. Some thirty-five years later,

Yasumoto et al. (67) reported that the PN component of the

conjugated form of vitamin B-6 in rice bran was in a 1:1 ratio

with glucose and identified this glycosylate as a 5'-

glucopyranosyl-derivative of PN. Several other studies (68-

69) have indicated the synthesis of this glucoside in pea

seedlings as well as the formation of other PN conjugates in

lesser concentrations than the 5'-pyridoxine-glucoside

derivative. Following the first isolation and structural

identification of 5'-0-(P-D-glucopyranosyl)pyridoxine (PN-

glucoside) from rice bran (67), the possible nutritional

significance of glycosylated forms of vitamin B-6 has been a

subject of intense study. Evidence for the presence of

conjugated forms of vitamin B-6 existing as a variable

proportion of the total vitamin B-6 of plant-derived foods has








16

been observed as well as evidence supporting the widespread

existence of P-glucoside conjugates of B-6 vitamers in plant

tissues (68). PN-glucoside appears to be formed in plants by

enzymatic transglycosylation (70-73). The glucose moiety of

PN-glucoside occurs at the 5' and 4' positions of the PN ring;

however, the presence of the 5'-isomer as the primary

naturally occurring form of PN-glucoside suggests specificity

for this position by the transglycosylation reaction in most

plant tissue (67). These studies provide evidence that

conjugates of vitamin B-6 exist as a substantial proportion of

the total vitamin B-6 in plant-derived foods (66-69). The

results of Kabir et. al (65) indicated the existance of 3-

glucopyranosyl conjugates in a wide variety of plant products.

The importance of PN-glucoside as a bioavailable source

of vitamin B-6 in humans was first studied by Kabir et al.

(74). The investigators determined the urinary and fecal

excretion patterns of glycosylated vitamin B-6 when different

foods that contained the naturally occurring PN-glucoside were

fed to human subjects. It was observed that an inverse

relationship existed between plant derived PN-glucoside (as

percent of total B-6) and the net B-6 bioavailability in

humans. The investigators suggested that the proportion of

the glucosylated vitamin B-6 in the diet may be useful as an

index of vitamin B-6 bioavailability.

Kabir et al. (65) developed a microbiological assay

procedure for quantitating glycosidic conjugated forms of PN








17

as well as total vitamin B-6 in food. The PN-glucoside, when

measured indirectly by hydrolysis, was reported to occur

widely in plant foods, ranging from 0% to 82% of the total

amount of vitamin B-6 present (74-76).

Studies by Iwami and Yasumoto (76) indicated that

intestinal absorption of PN-glucoside was similar to that of

free PN. However, Ink et al. (77) administered a single, oral

dose of radiolabeled PN and PN-glucoside in an alginate gel

diet using a combination of intrinsic and extrinsic labeling

in the intact rat. They reported that the absorption of the

glucoside was similar to that of unconjugated PN, but the

absorbed PN-glucoside was metabolically utilized less

effectively as vitamin B-6 than was PN. Iwami et al. (76)

measured the relative disappearance from the intestinal

perfusate of PN and a purified 4'-derivative (rather than the

naturally occurring 5'-derivative). In the rat, Ink et al.

(77) observed the relative metabolism of the intrinsically and

extrinsically labeled glucoside to have less than 40% of the

bioavailability of PN.

The nutritional properties of these conjugated forms of

vitamin B-6 are not fully understood at the present time.

Although PN-glucoside comprises 25% to 75% of the total

vitamin B-6 in many plant tissues (67,75) its bioavailability

in humans continues to be studied. PN-glucoside has been

found to undergo intestinal absorption in the intact form.

Initial studies employing bioassays with human subjects (40)








18

indicated that the proportion of the total vitamin B-6 which

was present as PN-glucoside correlated closely with the

bioavailability of the vitamin in a variety of foods examined,

including tuna, peanut butter and whole wheat bread. These

results suggested that PN-glucoside would not be available to

humans for intestinal absorption and metabolic utilization.

However, further studies by Bills et al. (78) indicated that

this observed correlation was inconsistent when additional

foods were examined. As discussed earlier, Nelson et al. (38)

reported that the vitamin B-6 from orange juice, a large

portion of which is in the conjugated form, exhibited a slower

rate of absorption during intestinal perfusion of humans than

PN from control solutions. Tsuji et al. (79) observed that

synthetic PN-glucoside exhibited vitamin B-6 activity which

was approximately equivalent to that of PN in a rat bioassay.

In contrast, Trumbo et al. (80) utilizing a rat bioassay

reported that PN-glucoside isolated from alfalfa sprouts

exhibited only 10% to 30% bioavailability relative to PN in

the rat. Studies in rats using radiolabeled PN-glucoside

indicate effective absorption of this B-6 compound and that

the portion of PN-glucoside not hydrolyzed to metabolically

active PN, is rapidly excreted in the urine in the intact form

(77,81). Ink et al. (77) reported the extent of absorption of

radiolabeled PN-glucoside from diets containing intrinsically

enriched alfalfa sprouts was similar, but quantitatively less








19

than that of purified labeled PN-glucoside, suggesting an

inhibitory effect of plant tissues.

Stable-isotopic techniques and methodologies evaluate the

bioavailability of purified deuterium-labeled (d2) PN-

glucoside in human subjects (82). In a study that used stable

isotopic techniques to analyze the bioavailability of orally

administered extrinsically enriched oatmeal containing labeled

PN-glucoside, it was reported that the labeled PN-glucoside

was approximately 58% of that of labeled-PN fed under

identical conditions (82). Although the ability metabolically

to utilize PN-glucoside appears to have a substantial level of

individual variability among human subjects, the

bioavailability of PN-glucoside in human subjects has been

observed to be substantially greater than that of the rat

(82). Further development of the methodology and application

of stable-isotopic techniques will continue to provide greater

clarification of the factors affecting bioavailability of

vitamin B-6.

The above section is a review of the current knowledge

concerning the bioavailability of the vitamin B-6 in foods.

There appears to be a general agreement between the rat

bioassay and the Saccharomyces uvarum assay method for many

foods, although the range of apparent bioavailability relative

to PN was quite large for the rat. These comparative data

provided the first indication that the bioavailability in

certain plant foods may be less than complete. The factors








20

responsible for the differences in apparent bioavailability

of vitamin B-6 are not clear.

Human subjects have also been used in researching the

bioavailability of vitamin B-6 in specific foods and mixed

diets. Bioassays with humans indicated that the overall

bioavailability of vitamin B-6 in a typical American diet is

reasonably high although incomplete. A study evaluating free-

living vegetarian and non-vegetarian women who consumed

quantities of total vitamin B-6 equivalent to those in a mixed

diet found no significant differences between groups with

respect to vitamin B-6 status when comparing plasma PLP levels

(53). This suggests that the bioavailability of vitamin B-6

in vegetarian diets was similar to that in a mixed diet.

Andon et al. (84) reported that a mean of 2.5% of the total

vitamin B-6 present in breast milk of omnivorous lactating

women was in the form of PN-glucoside. Although a mean of 15%

of the total vitamin B-6 in the diet was present as PN-

glucoside, there appeared to be no correlation between the

amount of glycosylated B-6 in the diet and the percent PN-

glucoside in breast milk over the range examined. In

constrast, a study by Reynolds et al. (85) reported that the

percentage of dietary vitamin B-6 present as PN-glucoside was

equivalent to the amount of PN-glucoside found in the breast

milk of lactating Nepalese vegetarian women. Gregory and Ink

(75) did not observe any PN-glucoside in the breast milk of

six lactating American females including three








21

lactoovovegetarians. Differences in the vitamin B-6

composition of the breast milk reported in these studies are

unclear. The factors responsible for the incomplete

bioavailability of vitamin B-6 have not been determined.


Factors Affecting the Bioavailabilitv of Vitamin B-6


Several factors are known to affect the bioavailability

of vitamin B-6 in food products. These include the formation

of certain reaction products during food processing and

storage, fiber type and the quantity present in the food

source, and the presence of PN-glucoside in plant foods. The

results of previously discussed studies suggested that thermal

processing and storage of foods may adversely affect

bioavailability of vitamin B-6 through the formation of

reaction products with amino acid residues of the

proteinatious portion of the food (36,56-57).

The bioavailability of vitamin B-6 from animal products

approaches 100% for most foods. The biological activity of

vitamin B-6 from plant derived products is generally lower.

A series of studies concerning the nutritional quality of

military rations provided the initial impetus to consider the

adverse effect of thermal preservation of canned foods.

Register et al. (83) reported that rats which subsisted

wholely on either of two homogenized combat rations required

supplemental vitamin B-6 to sustain normal growth even though

the diets contained enough vitamin B-6 for adequate rat








22

growth. Harding et al. (86) fed human subjects canned rations

which had been stored for twenty months at 100 OF. They

observed marginal vitamin B-6 deficiency in the subjects even

though the rations provided apparently adequate amounts of the

vitamin. Inadequate vitamin B-6 nutriture was most likely a

result of an effect of thermal processing. It is reasonable to

consider that marginal limitations in essential amino acid

content of the ration may have contributed to nutrient

insufficiency. The addition of vitamin B-6 to diets deficient

in certain essential amino acids has been shown to produce a

growth response (87). However, the reason for this apparent

incomplete bioavailability of vitamin B-6 in these canned

rations is unclear. Several infants who consumed a non-

fortified, heat sterilized canned infant formula were found to

be severely vitamin B-6 deficient and suffer with

neurochemically induced convulsions. Extensive research was

subsequently initiated to study the chemical behavior of

vitamin B-6 in milk products (58).

The differing stability properties of the various vitamin

B-6 compounds have been of interest since Hassinen et al. (59)

observed that PM and PL, the naturally occurring forms of

vitamin B-6 in milk, were much less stable than added PN.

Employing microbiological assays, Hassinen (59) demonstrated

that PM and PL added to milk were degraded at the same rate as

naturally occurring B-6 vitamers. Added PN exhibited greater

stability than either PM or PL during thermal processing of








23

canned milk and formula products. Through the use of rat

bioassays and microbiological analyses, Tomarelli et al. (60)

evaluated the bioavailability of vitamin B-6 in milk, spray-

dried and heat sterilized infant formula products. They

concluded that approximately half of the vitamin B-6 present

in retorted milk or infant formula was biologically

unavailable. Davies et al. (88) used rat and chick growth

bioassay methods to estimate the bioavailability of vitamin B-

6 in raw and canned milk. In contrast to the study of

Tomarelli (60), Davies et al. (88) reported the relative loss

of vitamin B-6 in heat processing to be equal as determined by

Saccharomyces uvarum, rat and chick bioassay methods. This

suggests that thermal sterilization in the production of

canned evaporated milk does not adversely affect the

bioavailability of vitamin B-6. Further evaluation of the

data reported by Tomarelli (60) indicated that the

bioavailability of the naturally occurring vitamin B-6 in

thermally sterilized unfortified milk was only 15% less than

that of PN fortified milk samples (37), which further

substantiated the results of Davies et al. (88).

Lushbough et al. (89) examined the bioavailability of

vitamin B-6 in meats and the effect of cooking using

Saccharomvces uvarum and rat growth bioassay methods. The

relative losses of vitamin B-6 during the cooking process were

nearly identical for every product tested. These data suggest

little effect of cooking on the bioavailability of the vitamin








24

B-6 in meats. Studies conducted by Ink et al. (90) used a

combination of intrinsic and extrinsic labeling of liver and

muscle tissue with radiolabeled forms of vitamin B-6. These

data indicated that thermal processing caused partial

destruction of vitamin B-6 in liver and muscle, but had little

or no effect on the bioavailability of the remaining vitamin

B-6.

The bioavailability of vitamin B-6 in foods from plant

origin has been proposed to be adversely influenced by dietary

fiber (36-37). The effect of various forms of dietary fiber

on vitamin B-6 bioavailability has been examined in both

animals and humans.

The effect of dietary fiber on the bioavailability of B-6

vitamers was first studied by Leklem et al. (40) who employed

a human bioassay to examine the utilization of vitamin B-6 in

white verse whole wheat bread. Their data indicated 10% lower

availability of the vitamin B-6 which was attributed to

reduced intestinal absorption by the fiber components. The

results of these studies suggested that the vitamin B-6 of

soybeans was approximately 5% to 10% less available than beef

vitamin B-6. These data have suggested that the presence of

nondigestible components of dietary fiber, may reduce the

bioavailability of the vitamin B-6. Physical properties

inherent of dietary fiber may influence the absorption of the

vitamin through entrapment or alteration of the viscosity of

the intestinal contents (16).








25

The presence and type of purified dietary fiber have only

a minor effect on the bioavailability of the B-6 vitamers (91-

93). Nuygen et al. (91) evaluated the potential for physical

binding of vitamin B-6 using a variety of native and modified

polysaccharides and lignin under conditions similar to the

human intestine. In vitro binding of the B-6 vitamers by

these fiber components did not occur. Differences in

viscosity may have influenced the rate of diffusion, although

studies by Machida and Nagai (93) indicated that a reduced

rate of absorption does not induce a reduction in net vitamin

B-6 absorption. Rat and chick bioassays have been employed to

evaluate the effects of selected dietary fiber components on

the absorption of vitamin B-6 (91,93). The results of these

bioassays indicated no inhibitory effects of these dietary

fiber components on the bioavailability of vitamin B-6. The

results of these studies along with the results of studies

with human subjects (39,56) suggest that dietary fiber has

little effect on the bioavailability of vitamin B-6 in foods.

The slight effect of dietary fiber components on the

bioavailability of vitamin B-6 does not fully account for the

lower bioavailability of B-6 vitamers from plant food products

relative to animal sources.

Many vitamin B-6 antimetabolites have been identified

and determined to be chemical reaction products formed from

vitamin B-6 during the thermal processing or storage of foods.

The potential conversion of various B-6 vitamers to these








26

antagonistic compounds has been proposed to explain certain

observed effects of food processing or storage on the

bioavailability of vitamin B-6 in certain foods (94-96). In

studying the stability of the B-6 vitamers in food products,

PN was found to be the most stable vitamer, followed by PL,

PM, PMP, and the least stable B-6 vitamer being PLP (88).

Much research has focused on the potential for PL and PLP to

react with such food components as proteins, amino acids and

reducing sugars resulting in the formation of degradation

products of limited bioavailability.

Several studies have reported the characterization of the

by-products resulting from the interaction of PL and PLP with

amino groups of proteins in food products (96-102). The

chemical reaction of PL or PLP with cysteine or other

sulfhydryl amino groups has been proposed as a mechanism

responsible for the lowered bioavailability of vitamin B-6

reported in thermally sterilized milk products (96-98). One

such product produced from the heat induced reaction of

cysteine and PL has been identified as bis-4-pyridoxyl

disulfide (97). Later studies by Srncova et al. (99)

supported the reactivity of milk protein sulfhydryl groups

with PL when high concentrations of PL were present. The

spontaneous reaction of PL or PLP with various thiols and

aminothiols results in the formation of thiohemiacetal and

thiazolidine complexes (94). These chemical complexes are

readily dissociable and therefore would not impair the








27

bioavailability of these B-6 vitamers. Gregory et al. (100-

102) conducted a series of experiments to examine further the

possible influence of thermal processing and storage on the

net bioavailability of vitamin B-6 in foods. When milk

containing radiolabeled PL or PLP was subjected to heat

sterilization, analysis using HPLC revealed no formation of

bis-4-pyridoxal disulfide (103). The losses of the B-6

vitamers PL and PLP were due to the reductive binding of the

aldehydes of these vitamers to food protein as e-

pyridoxyllysyl residues (103). The formation of �-

pyridoxyllysine has been identified as a mechanism of the loss

of vitamin B-6 during thermal processing or low moisture

storage of proteinaceous model food systems (100-101) and

various meat and dairy food products (16). Similar results

were reported in intrinsically enriched chicken liver and

muscle tissues (103). The phosphopyridoxyllysyl complex

accelerated the onset and enhanced the severity of vitamin B-6

deficiency symptoms. The effect of e-pyridoxyllysine on the

bioavailability of vitamin B-6 in foods is a function of the

ratio of total vitamin B-6 content of the diet and the

addition product to the other B-6 vitamers. The

bioavailability of the vitamin of diets which contain e-

pyridoxyllysine but which are adequate in vitamin B-6 content

would most likely be high (102). Whether naturally occurring

amounts of vitamin B-6 are low or conditions present during

food processing produce E-pyridoxyllysine, the antimetabolite








28

effects of E-pyridoxyllysine generally decreases the apparent

bioavailability of vitamin B-6 by competitive inhibition of PL

kinase (103). It is reasonable to surmise from the above

studies that a similar antagonistic effect of E-

pyridoxyllysine may have been responsible for the severe

deficiencies observed in infants with compromised vitamin B-6

nutriture and fed diets composed entirely of non fortified

canned infant formulas (60). The mechanisms of action for an

antimetabolite, such as e-pyridoxyllysine, includes inhibition

of any of the enzymes in the metabolic pathway to interconvert

the B-6 vitamers. Other mechanisms of vitamin B-6

antimetabolites are structural analogues of the vitamin and

inhibition of the active coenzyme function of PLP (96).

The complex chemical identity and nutritional properties

of other B-6 vitamer derivatives found in foods have not been

fully determined. The formation of the degradation product,

6-hydroxy-pyridine from the relatively stable B-6 vitamer PN,

has been reported in thermal treatment of various fruits and

vegetables (104). The generation of hydroxyl radicals

apparently mediated the hydroxylation of PN which was

associated with the oxidative degradation of ascorbic acid

(105). Gregory and Leatham (106) reported that 6-hydroxy-PN

has no vitamin B-6 or antivitamin B-6 activity. It appears

that hydroxylation of PN at the 6 position is not a

significant mechanism responsible for the loss of vitamin B-6

activity in foods.








29

A number of derivatives of PN-glucoside have been

isolated and identified. Tadera et al. (73,107) reported

several additional PN-glycosides to be minor components of the

total vitamin B-6 in pea seedlings and rice bran (108).

Assuming that PN-glucoside esters occur naturally and in

great enough quantities, it is unclear whether these glucosyl

compounds contribute significantly to the total vitamin B-6 in

foods and, if so how they may impact vitamin B-6 nutriture

when consumed. However, it is worth mentioning that PN-

glucoside has been the only glycosylated form of vitamin B-6

detected in a variety of foods utilizing HPLC methodology

(71,77). Tadera et al. (108) have observed the presence of a

vitamin B-6 conjugate termed "B-6X". The authors described

this compound as PN-glucoside esterified to an organic acid

based on its response in microbiological assays following

sequential alkaline treatment and hydrolysis with beta-

glucosidase (108).

The nutritional properties of these conjugated forms of

vitamin B-6 are not fully understood at the present time. The

nutritional significance of these glycosylated vitamin B-6

compounds is related to their potentially limited

bioavialability. As earlier stated, PN-glucoside comprises 5%

to 80% of the total vitamin B-6 in many plant tissues (67,75).

Its bioavailability for humans is approximately 58% relative

to that of PN when administered orally (82). PN-glucoside has

been found to undergo intestinal absorption via hydrolysis of








30

the beta-glycosidic bond of PN-glucoside by mucosal beta-

glucosidase releasing free PN. Initial studies employing

bioassays with human subjects (40) indicated that the

proportion of the total vitamin B-6 which was present as PN-

glucoside correlated inversely with the bioavailability of the

vitamin in the foods examined (tuna, peanut butter, and whole

wheat bread). These results suggested that PN-glucoside would

not be biologically available to humans. However, studies by

Bills et al. (78) indicated that this observed correlation was

inconsistent when additional foods were examined and that the

quantity of PN-glucoside present in a food product could

function as a predictor of the bioavailability of vitamin B-6

from that food source. Stable-isotopic techniques have been

used to evaluate the bioavailability of purified deuterium-

labeled (d2) PN-glucoside in human subjects (82). In these

studies, the bioavailability of orally administered

extrinsically enriched oatmeal containing labeled PN-glucoside

was similar to that of labeled-PN fed under identical

conditions (82). Although metabolic utilization of PN-

glucoside appears to have a substantial level of individual

variability among human subjects, the bioavailability of PN-

glucoside in human subjects has been observed to be

substantially greater than that of the rat (82). Further

application of stable-isotopic techniques provide greater

clarification of the factors affecting bioavailability of

vitamin B-6.








31

The bioavailability of vitamin B-6 compounds is a

function of their extent of absorption and metabolic

utilization, as reflected by conversion to active coenzymatic

forms. It is important to assess the bioavailability of

vitamin B-6 as consumed in a typical mixed diet. The

assessment of vitamin B-6 status in populations is contingent

on identifying and quantifying the different forms of the

vitamin occurring naturally in foods consumed, determining the

potential interactions between these forms and the relative

bioavailability of the different vitamers, and their possible

interactions.

Our present knowledge of the essential role of vitamin

B-6 in preserving health emphasizes the importance of

establishing an accurate intake for vitamin B-6. The adequacy

of intake is difficult to assess critically without accurate

information regarding the amount of the vitamin in foods

consumed, the effect of various vitamin B-6 compounds on the

enzymes involved in the metabolism of various forms of vitamin

B-6, as well as information regarding vitamin B-6

bioavailability.


Enzymatic Interconversion of B-6 Vitamers


There are several enzymatic reactions responsible for the

interconversion of the B-6 vitamers. The liver serves as the

center of vitamin B-6 metabolism and contains the enzymes

necessary for the metabolic interconversion of the B-6








32

compounds. Following intestinal absorption, the

dephosphorylated B-6 vitamers diffuse into the liver and are

converted to the coenzyme form of the vitamin, PLP (109-110).

A portion of the PLP is released into the circulation and

constitutes the major source of plasma PLP (111).

The dephosphorylated B-6 vitamers, PL, PN, and PM are

phosphorylated by the same kinase enzyme, pyridoxal kinase (EC

2.7.1.35) (PL kinase). The 5'-phosphates of PN and PM are

subsequently oxidized by pyridoxamine (pyridoxine) 5'-

phosphate oxidase (EC 1.4.3.5) (PMP (PNP) oxidase) to yield

PLP. This coenzymatic form is bound to apoproteins,

transported to blood or dephosphorylated by phosphatases back

to PL (112). Further metabolism of PL results in

rephosphorylation by PL kinase or oxidation, by aldehyde

oxidase, to 4PA, the catabolic end product of vitamin B-6

metabolism. Vitamin B-6 in its coenzymatic form PLP or as

PLP-dependent enzymes, functions in many metabolic processes.

These include, amino acid metabolism (3), neurochemical

function (113-115), and modulation of hormone action (4-5).

Therefore, it is important that the cellular levels of PLP be

properly regulated.

As stated above, PL kinase is responsible for the

postabsorptive phosphorylation of PL, PN, and PM with

relatively equal efficiency. The result is the formation of

PLP from PL and upon further oxidation of PN and PM by PMP

(PNP) oxidase these two vitamers are also converted to the








33

coenzymatic form (109-110). The PLP that is not released into

circulation is either bound to cellular proteins or

dephosphorylated to PL. The vitamin B-6 taken up by the liver

is predominately maintained in the phosphorylated form (109-

110).

There exists a cycle of phosphorylation/dephosphorylation

between PL and PLP which is catalyzed by PL kinase and

phosphatase enzymes (109). Under physiological conditions PL

kinase has approximately a 10 fold greater activity than

phosphatases (112). The degradative enzyme aldehyde oxidase

has an activity greater than the kinase and effectively

competes for PL. In order to maintain a steady state of

intracellular PLP a form of metabolic trapping occurs. The

coenzyme is bound to protein which protects it from hydrolysis

and hence, determines the relative amounts of PLP and PL and

creates a shift of equillibrum toward a balanced state (112).

Considering that PLP is required as a conenzyme for a

number of enzymes, it is not surprising that PL kinase is

widely distributed in tissues. This enzyme has been detected

in practically all tissues tested, including liver, kidney,

brain, and muscle (30). PL kinase has also been detected in

erythrocytes, however, the enzyme differs in its properties

compared with other tissue PL kinase (116). Since the liver

is the major site of vitamin B-6 metabolism it is generally

the tissue selected for the study of the enzymes in the B-6

vitamer interconversion pathway (116). It is important to








34

note that in comparative assessment of either of the enzymes

PL kinase or PMP (PNP) oxidase from different sources, the

levels and forms (apo or holo) of the enzyme is markedly

influenced by factors such as species, age, sex and dietary

status of the animal (110). Therefore, the following

discussion regarding the properties of these two enzymes will

be limited to rat liver as the source for the enzymes, unless

otherwise noted.

Hepatic PL kinase has a molecular weight of approximately

60,000 and a pH optimum between 5.5 and 6.0 (117). PL kinase

from all sources studied requires a divalent cation for

activity, with the greatest activity in liver PL kinase seen

with Zn2+ (30). The enzyme also requires monovalent cations

with activation greatest in the presence of Kr (117).

Although similar to PL kinase in tissue distribution and

molecular weight, PMP (PNP) oxidase differs significantly in

other physical and kinetic properties. The molecular weight

of oxidase is approximately 54,000, each having two identical

subunits of 27,000. PMP (PNP) oxidase occurs widely

distributed in both tissues and cells (118-119). Pogell (32)

was the first to describe PMP (PNP) oxidase as having a

requirement for a flavin cofactor in the oxygen-dependent

conversion of PMP to PLP. Later Wada and Snell (31) observed

that the same oxidase was responsible for the conversion of

both PNP and PMP to PLP. The investigators reported that the

partially purified enzyme required FMN as the specific








35

flavocoenzyme and demonstrated product inhibition as well as

inhibition by some phosphorylated analogs of vitamin B-6

compounds (31).

The relative reactivity of PMP(PNP) oxidase with varying

substrates is pH dependent (118). Within the physiological pH

range both PMP and PNP are suitable substrates for liver PMP

(PNP) oxidase. Although at slightly alkaline pH the Vx for

PNP is higher, the observed velocity is decreased due to

greater substrate inhibition by PNP even at low substrate

concentrations (112). The Km values demonstrate that at pH

7.0 PMP and PNP bind with similar affinity, whereas PMP binds

more tightly at the usual assay pH of 8.0 compared to PNP

(118).

Many compounds have been indicated as inhibitors of the

metabolism of vitamin B-6. When assessing the site of action

of these antagonists it is reasonable to suspect that such

compounds may act on those enzymes responsible for the

formation of the active coenzyme form PLP. It is well

documented that liver PL kinase is inhibited by analogues

containing a 4-formyl group which resemble PL in their

affinity for PL kinase, condensation products formed from PL

and hydroxylamine, O-substituted hydroxylamines, hydrazine and

substituted hydrazines (120). PL kinase has also been shown

to be sensitive to inhibition by degradation products such as

pyridoxyllysine (121).








36

It is well known that the product PLP, inhibits the

activity of PMP(PNP) oxidase (112). As a result, other

phospho-B-6 analogues have been evaluated (118). It was

discovered that a dianionic charge on the 5'-position is

necessary for binding of the substrate analogue and

subsequently inhibition of the enzyme (122). Inhibition of

PMP (PNP) oxidase also occurs when alterations are introduced

in the structure of the flavin coenzyme, FMN (118). In as

much as both PL kinase and PMP(PNP) oxidase have substrate

reactivity with structural analogues, it is not unlikely that

conjugated forms of vitamin B-6 would inhibit these enzymes.

Conflicting data have been reported concerning the extent

of bioavailability of PN-glucoside in vitamin B-6 metabolism.

Most experimental results in rats have indicated 20-30% net

bioavailability of PN-glucoside relative to PN. The

bioavailability of PN-glucoside, although incomplete, is

substantially greater in humans than in rats. While the

incomplete utilization of PN-glucoside has been clearly

established in rats and humans, its potential interactive

effects in vitamin B-6 metabolism have not been examined

previously.

The purpose of this research was to demonstrate the

effects PN-glucoside has on the metabolic utilization of

vitamin B-6. The effort to evaluate these effects is an

important one, when considering the overall nutritional status

of individuals with respect to vitamin B-6. Being aware of









37

the multitude of roles vitamin B-6 plays in the body and its

impact on the overall health of individuals, it is imperative

that information be collected concerning how naturally

occurring forms of vitamin B-6 potentially interact with one

another.














CHAPTER 3
PYRIDOXINE-5'-3-D-GLUCOSIDE AFFECTS THE
METABOLIC UTILIZATION OF PYRIDOXINE IN RATS


Introduction


A conjugated form of pyridoxine (PN) was first isolated

from rice bran and identified as 5'-0-(P-D-

glucopyranosyl)pyridoxine (PN-glucoside) (67). Analysis of a

variety of plant-derived foods by HPLC has shown that PN-

glucoside comprised 5-80 % of the total vitamin B-6 in many

fruits and vegetables (75). Similar results were obtained by

the use of a microbiological assay procedure for the

glycosidic conjugate of pyridoxine (65).

Conflicting data have been reported concerning the

bioavailability of PN-glucoside in vitamin B-6 metabolism.

Studies in rats have indicated that purified PN-glucoside is

relatively well absorbed but undergoes little metabolic

utilization and is rapidly excreted in intact form (77,80).

Most experimental results in rats have indicated 20-30 % net

bioavailability of PN-glucoside relative to PN (77,80-81). In

constrast, Tsuji et al. (79) observed nearly 100 %

bioavailability of PN-glucoside relative to PN in the rat. The

bioavailability of PN-glucoside, although incomplete, is

substantially greater in humans than in rats (82). The mean


38








39

bioavailability of orally administered PN-glucoside is

approximately 58% relative to PN. While the incomplete

utilization of PN-glucoside has been clearly established in

rats and humans, its potential interactive effects in vitamin

B-6 metabolism have not been examined previously.


Materials and Methods


Protocol


Two studies were conducted to evaluate the in vivo

utilization of [14C]PN in the presence of PN-glucoside. In

each of the studies eighteen male Sprague-Dawley rats (Crl:CD

(SD) BR) from Charles River Breeding Laboratories (Wilmington,

MA), weighing 200-300 g, were housed individually in stainless

steel metabolism cages in animal quarters maintained at 24�10C

with a 12-h light/dark cycle. All procedures for the care and

treatment of the experimental animals were in accordance with

the National Institutes of Health Guidelines.

The rats were fed ad libitum a casein-sucrose (20%:60%)

based diet, adequate in all micronutrients including 7 mg PN

HCl/kg.d for eight days (80). At the end of a seven day

acclimation period the rats were randomly assigned to one of

three treatment groups. On day eight of studies 1 and 2,

following a twelve hour fast, the rats were administered a

dose of 166.5 MBq (4.5 gCi) [14C]PN, equivalent to 240 nmol of

PN. In study 1 the rats received a simultaneous dose of

either 0, 36, or 72 nmol of unlabeled PN-glucoside (treatment








40

groups 1,2, and 3 respectively). In study 2 the rats received

a simultaneous dose of either 0 or 72 nmol unlabeled PN-

glucoside or 21.6 nmol unlabeled PN (treatment groups 1,2, and

3 respectively). The 21.6 nmol dose of PN was selected to

provide 30% of the PN-glucoside treatment, based on an assumed

30% net bioavailability of PN-glucoside relative to PN (4-6).

Doses were administered in 1.5 mL sterile H20 by gavage.

Urine was collected for the ensuing 24 h into foil-covered

flasks to avoid photochemical degradation of vitamin B-6

compounds. The urine collection funnels and flasks were

rinsed with water and rinses pooled with collected urine;

after dilution to 50 mL, urine was stored at -20" C until

analysis. The rats were killed by decapitation following

brief anesthesia, and livers were rapidly excised. The tissue

was divided into 2 g samples, then stored at -20" C until

analysis. All procedures were performed under GEF40GO gold

lights that emit a wavelength between 500 nm and 750 nm to

prevent photochemical degradation of B-6 vitamers.


Forms of vitamin B-6


PN-glucoside was prepared by biological synthesis using

the propagation of alfalfa sprouts germinated in the presence

of PN-HCl obtained commercially (Sigma Chemical Company, St.

Louis, MO) (80). The [4,5-14C]pyridoxine hydrochloride 684.5

KBq/nmol (18.5 Ci/mol) was a gift from Hoffmann-LaRoche








41

(Nutley, NJ) with a purity of greater than 98%, as determined

by ion-pair reverse phase HPLC.


Sample preparation


Urine samples were deproteinated by centrifugal

ultrafiltration with micropartition tubes (YMT membrane

filters; Amicon, Danvers, MA) (80). Urinary 4PA analysis was

by reverse-phase HPLC, as described below. Fractions (0.5 mL)

were collected by using an ISCO Cygnet Fraction Collector

(ISCO, Lincoln, NB). Each filtered sample was decolorized

(81) and an aliquot (100 AiL) was then counted for total

radioactivity. Liver tissue was minced and homogenized in 6

mL of 4.3 mol/L trichloroacetic acid with a Polytron

Homogenizer (Brinkman Instruments, Westburg, NY), and then

centrifuged for 20 min. at 12,000 x g (6). The supernatant

was partitioned against an equal volume of diethyl ether to

extract the trichloroacetic acid and filtered with 0.45 Am

membrane filters (Gelman Sciences, Inc., Ann Arbor, MI). The

sample was aerated with nitrogen gas, then analyzed for

vitamin B-6 by HPLC as described below, and an aliquot of the

supernatant from each sample was analyzed for radioactivity.

Decolorization of the extract was performed as above prior to

liquid scintillation spectrometry.








42

HPLC eauipment and analysis


The separation of radiolabeled B-6 compounds in the urine

and liver tissue was accomplished by ion-pair reverse-phase

HPLC (82). Chromatographic analysis was performed with a

Rainin HP/HPX Drive Module (Rainin Instrument, Woburn, MA).

All of the above analysis utilized a loop injection valve

(Model 904-2,Altex), a fluorometric detector (Model LS-5,

Perkin-Elmer, Norwalk, CT) and an electronic integrator (Model

3388A, Hewlett-Packard, Avondale, PA). Excitation and

emission wavelengths were 295nm (5nm slit width) and 405nm

(5nm slit width) respectively. Two mobile phases were employed

in a gradient elution procedure using an Ultrasphere IP 5 Atm

C-18 4.6 mm x 25 cm column (Beckman Instruments, San Ramon,

CA). Mobile phase A contained 0.033 mol/L phosphoric acid and

8 mmol/L 1-octanesulfonic acid, adjusted to pH 2.2 with 6

mol/L KOH. Mobile phase B contained 0.033 mol/L phosphoric

acid and 3.4 mol/L acetonitrile, adjusted to pH 2.2 with 6

mol/L KOH and no ion-pairing reagent. Fractions (0.5 mL) were

collected using an ISCO Cygnet Fraction Collector (ISCO,

Lincoln, NB). The identity of the PLP and PNP peaks was

confirmed by monitoring the formation of pyridoxal (PL) and

PN, respectively, by HPLC following incubation of the PLP and

PNP fractions in 0.1 mol/L HC1 at 80� C for 3 h.








43

Measurement of radioactivity


HPLC fractions and extracts of tissue were measured for

radioactivity using a commercial scintillation fluid

(ScintiVerse LC, Fisher Scientific, Orlando, FL) and a liquid

scintillation spectrophotometer (Beckman LS 2800 Beckman

Instruments, San Ramon, CA). A quench curve for 14C was used

for conversion of cpm to dpm.


Statistical analysis


In experiment 1 and 2 the distribution of vitamin B-6

compounds and total radioactivity between groups was compared

by the method of least squares analysis of variance using

general linear model procedures of SAS (123) following a log

tranformation of the data to normalize variance (123). Data

are reported as least squares mean � the pooled standard error

of the least squares mean (SEM). In experiment 1, orthogonal

contrasts were made to examine the linear and quadratic

effects of different doses of PN-glucoside on dependent

variables. The quadratic effects were not significant for

any dependent variable. Values with p<0.05 were considered

statistically significant and were all for linear effects of

PN-glucoside unless otherwise stated. In expreriment 2 two

orthogonal contrasts were used to compare treatment means.

Contrast one compared group 2 with group 1 and 3; contrast two

compared group 1 with group 3.








44

Table 3.1 Liver B-6 vitamer distribution and total liver 14C
in rats administered varying levels of PN-glucoside in the
dose (Experiment 1).*

LEVEL OF PN-GLUCOSIDE IN THE DOSE

Vitamers 0 nmol 36 nmol 72 nmol SEM p<0.05

% liver radioactivity
PLP 17.5 37.3 51.6 7.5 #
PNP 0.0 3.1 16.4 3.4 #
PMP 46.7 36.1 32.0 8.0 #
PL 16.9 14.5 0.0 4.8 +
PN 15.2 7.8 0.0 3.5 +
PM 3.6 1.2 0.0 0.6 +

% of dose
Total 14C 20.0 17.0 12.3 4.0 +

Values based on n=6 are means and pooled standard error
of the mean (SEM). Log transformations were performed on
data prior to analysis of variance. The quadratic effects
were not significant for any dependant variable.
# Significant linear effect, p<0.05.
+ Significant decreased linear effect, p<0.05.


Results


These experiments were designed to investigate the effect

of unlabeled PN-glucoside on the metabolism, in vivo

retention, and excretion processes of simultaneously

administered [14C]PN. The distribution of B-6 vitamers in

liver tissue, total urinary [14C]4PA, and total urinary 14C

were determined to evaluate the utilization of [14CJPN in the

presence of PN-glucoside. Data were expressed as a percentage

of total radioactivity administered in the dose or as a

percentage of total ([4C] B-6 in the tissue.








45

Table 3.2 Urinary 4PA and total urinary 14C in rats
administered varying levels of PN-glucoside (Experiment 1)*


0 nmol 36 nmol 72 nmol
PN-glucoside PN-glucoside PN-glucoside SEM p<0.05

% urinary radioactivity
4PA 37.6 12.7 10.9 5.1 +
% of dose
14C 26.9 35.1 37.3 12.3 #


* Values based on n=6 are means and pooled standard error
of the mean (SEM). Log transformation of the data was
performed prior to the analysis of variance. The
quadratic effects were not significant for any dependant
variable.
* Significant linear effect, p<0.05.
+ Significant decreased linear effect, p<0.05.


The results of study 1 indicated statistically

significant differences (p<0.05) in the quantity and

distribution of labeled vitamin B-6 compounds in liver tissue

and urine among groups fed varying levels of PN-glucoside

(Tables 3.1 and 3.2). A significant linear relationship was

observed between phosphorylated vitamin B-6 compounds

pyridoxal 5'-phosphate (PLP) and pyridoxine 5'-phosphate (PNP)

in liver tissue and the amount of administered PN-glucoside.

Total hepatic 14C decreased linearly (p<0.05) with varying

amounts of PN-glucoside (Table 3.1). The proportion of

hepatic nonphosphorylated 14C-labeled PN, PL and pyridoxamine

(PM) decreased linearly with the amount of PN-glucoside (Table

3.1).

Urinary excretion of total '4C was showed a significant

linear effect (p<0.05) among groups receiving varying doses of








46

Table 3.3 Liver B-6 vitamer distribution and total liver 14C
in rats administered varying levels of PN or PN-glucoside.
(Experiment 2)*


LEVELS OF PN-GLUCOSIDE LEVELS OF PN

Vitamers 0 nmol 72 nmol 21.6 nmol SEM
Group 1 Group 2 Group 3

% liver radioactivity
PLP 31.5 58.4a 31.5 3.4
PNP 0.0 14.4a 0.0 0.1
PMP 41.4 27.2a 40.1 1.0
PL 13.3 0.08 14.3 0.4
PN 6.3 0.08 6.0 0.2
PM 7.5 0.0 a 8.1 0.4
% of dose
Total 14C 15.1 10.0b 15.0 0.0


* Values based on n=6 are means and pooled standard error
of the means (SEM). Log transformation of the data was
performed prior to analysis of variance. The quadratic
effects were not significant for any dependant variable.
a Significantly different at p<0.05 for linear effects of
PN-glucoside in the dose.
b Significantly different at p<0.05 for decreased linear
effect.


PN-glucoside (Table 3.2). In addition, the concentration of

urinary [14C]4PA, relative to total urinary 14C, and total

urinary [14C] 4PA decreased linearly with the amount of

administered PN-glucoside (Table 3.2).

Experiment 2 was conducted to determine if the results of

experiment 1 were due to a specific effect of PN-glucoside or

simply to an increase in the total amount of available PN

(derived from PN-glucoside). The experimental design was the

same, except on day eight, following a twelve hour fast, the

rats were administered a simultaneous dose of 166.5 MBq (4.5








47

IMCi) [14C]PN, while treatment groups 1 and 2 received 0 nmol

and 72 nmol unlabeled PN-glucoside, respectively. Treatment

group 3 received 21.6 nmol unlabeled PN based on an assumed

30% net bioavailability of PN-glucoside relative to PN (77,80-

81). Statistical analysis by orthogonal contrasts of the data

from this study indicated no statistically significant linear

effects between the quantity or distribution of labeled

vitamin B-6 compounds in liver tissue and urine and treatment

groups one and three (Table 3.3). However, there was a

significant linear effect (p<0.05) in the relative

concentration of phosphorylated vitamin B-6 compounds PLP,





Table 3.4 Urinary 4PA and total urinary 14C in rats
administered *varying levels of PN or PN-glucoside.
(Experiment 2)


LEVELS OF PN-GLUCOSIDE LEVELS OF PN

0 nmol 72 nmol 21.6 nmol SEM

% urinary radioactivity
4PA 35.5 14.2b 32.7 12.3
% of dose
14C 30.0 46.0 27.0 4.0
----------------------------------------------------

Values based on n=6 are means and pooled standard error
of the mean (SEM). Log transformation of the data was
performed prior to analysis of variance. The quadratic
effects were not significant for any dependant variable.
a Significantly different at p<0.05 for linear effects of
PN-glucoside in the dose.
b Significantly different at p<0.05 for decreased linear
effects.








48

PNP, and pyridoxamine 5'-phosphate (PMP) in liver tissue,

while hepatic 14C decreased linearly (p<0.05) (Table 3.3). The

proportion of hepatic nonphosphorylated forms of [14C]PN, PL

and PM was significantly less among treatment group 2

(p<0.05) than groups 1 and 3. Urinary excretion of total 14C

also was significantly greater (p<0.05) in group two relative

to groups one and three (Table 3.4). In addition, the

concentration of urinary 4PA, relative to total urinary 14C,

was significantly lower (p<0.05) in group two as compared to

groups 1 and 3 (Table 3.4). The results of experiment 2

indicate that the effects of PN-glucoside observed in

experiment 1 were specific for this vitamin B-6 derivative and

were not due to isotope dilution effects of the unlabeled

compound administered.


Discussion


The focus of this research was to determine the direct

effects of PN-glucoside on [14C]PN administered in a

simultaneous dose, similar to what would occur when consuming

a mixed diet. The ratios of PN-glucoside to nonglycosylated

vitamin B-6 (14C-labeled plus dietary) were in the range of 8%

to 16%, which is consistent with observations of typical human

diets (82). Since our interest was in determining the fate of

labeled-vitamin B-6 compounds in the presence of PN-glucoside

the data were expressed as percent of the total radioactivity

in the dose. This enabled the determination of the direct








49

effect of PN-glucoside on the metabolic utilization of [14C]PN

and distribution of the 14C-labeled B-6 metabolites in the

liver 24 h post-dose when the two vitamers are consumed

together. Expression of the data in this manner reflects the

direct effect of PN-glucoside on the distribution of B-6

vitamers and urinary excretion of 4PA as well as interactions

between different naturally occurring forms of vitamin B-6

found in foods consumed simultaneously in a typical mixed

diet. The bioavailability of vitamin B-6 compounds is a

function of their extent of absorption and metabolic

conversion to active coenzymatic forms. Pyridoxine that is

absorbed from the intestinal tract is concentrated initially

in the liver (124) and is the sole source of plasma PLP (125).

Thus, liver plays a central role in the overall metabolism of

vitamin B-6 (126).

The major transformations in hepatic vitamin B-6

metabolism involve phosphorylation catalyzed by pyridoxal

kinase, oxidation of PNP and PMP by pyridoxine (pyridoxamine)

5'-phosphate oxidase (PNP oxidase), along with interconversion

of PLP and PMP through transamination reactions (109-110,127-

129). The principal forms of vitamin B-6 in liver are PLP and

PMP (129-131). PNP is usually present in only trace

quantities because of its rapid oxidation to PLP (130,132-

133). The non-phosphorylated B-6 vitamers constitute less

than 10% of the total vitamin B-6 content in the liver (131).

In this study, the concentration of [14C]PNP increased linearly








50

with the quantity of unlabeled PN-glucoside administered

(p<0.01). This suggests a mechanism involving a possible

inhibition of PNP oxidase. Investigations of possible direct

inhibitory effects of PN-glucoside on PNP oxidase are in

progress. Comparatively high levels (36%) of the non-

phosphorylated B-6 vitamers were observed in the livers of the

rats in group 1, which may have been due to several factors.

The vitamin B-6 requirement for the rat is 6-7 mg PN/kg diet

(135). The experimental diet contained 7 mg PN/kg diet with

the rats consuming, on the average, 70 Ag vitamin B-6 per day.

The dose of labeled-PN contained 240 nmol (49.3 gg) [14C]PN.

Hence, during the 24 h post dose period, rats received 49.3 ug

vitamin B-6 above their requirement. In addition, a single

observation of tracer distribution at a time point 24 h post-

dose, would not necessary reflect steady state concentrations

of endogenous vitamin B-6 in the liver. These factors would

account for greater than normal levels of B-6 metabolites in

liver.

The compensating changes observed in the proportions of

hepatic [14C]PMP and [14C]PLP (Table 3.1) cannot be readily

explained. Unexpectedly, these changes are consistent with

those observed when comparing the metabolism of [14C]PN in

vitamin B-6 adequate verses deficient rats (81). These

results indicate that the administration of unlabeled PN-

glucoside in this study caused changes in metabolic patterns








51

which paralleled, in one respect, those seen in vitamin B-6

deficiency.

The metabolic pathway for the degradation of PLP involves

enzymatic hydrolysis of the phosphate ester bond, and the

oxidation of PL to 4PA (132-133). As a terminal product of

vitamin B-6 metabolism, urinary 4PA reflects the in vivo

metabolic utilization of the vitamin. In these studies, the

extent of conversion of administered [14C]PN to [14C]4PA is

indicative of reduced utilization of [14C)PN which is inversly

related to the dose of PN-glucoside. This study also shows a

negative correlation between percent radioactivity in the

urine as [14C]4PA and the quantity of PN-glucoside

administered. The data demonstrate that total liver 14C was

inversly related to the PN-glucoside dose while total urinary

14C was directly proportional to the amount of PN-glucoside

administered.

The intracellular concentration of PLP in liver tissue is

tightly regulated which does not permit the excess

accumulation of this coenzyme (133). In the liver, newly

synthesized PLP and PMP are contained in compartments that

have a rapid rate of turnover (132). These small and rapidly

mobilized pools are poorly miscible with the endogenous

coenzyme pools (132). The newly synthesized [14C]PLP and

[14C]PMP in the liver would be rapidly metabolized by

hepatocytes and degraded to [14C]4PA and excreted in the urine,

or transported to the circulation as [14C]PL, and [14C]PLP.








52

These B-6 vitamers could then be taken-up by erythrocytes

and/or tissues. This would account for the observed results

of liver 14C and urinary 14C in response to PN-glucoside dose

seen in this study. When rats were given a quantity of

unlabeled PN equivalent of 30% of the PN-glucoside dose, the

effect on the metabolic utilization and distribution of other

forms of vitamin B-6 differed significantly from the group

administered the PN-glucoside dose.

In previous studies in this laboratory, the absorption of

PN-glucoside was 50% relative to PN in rats fed a diet

adequate in vitamin B-6 when administered as an alginate gel

(77) but was nearly equivalent to PN when given in solution

(81). Analysis of urine showed that most of the absorbed PN-

glucoside was excreted in the intact form, suggesting low

bioavailability (4,6). Previous findings by Trumbo et al.

(80) showed poor utilization of PN-glucoside relative to PN on

the basis of growth and plasma PLP concentration, which

indicated that PN-glucoside has a low biological availability

as vitamin B-6 in the rat. However, PN derived from

hydrolysis of PN-glucoside, can enter into vitamin B-6

metabolic pathways of the liver to produce other forms of

vitamin B-6 (77,81).

Kabir et al. (65) observed with human subjects, an

inverse relationship between the percentage PN-glucoside (of

total vitamin B-6) and overall vitamiln B-6 bioavailability in

food. The investigators suggested that the proportion of the








53

glycosylated vitamin B-6 in the diet may be useful as an index

of vitamin B-6 bioavailability. The nutritional significance

of the effect of PN-glucoside on the metabolic utilization of

other forms of vitamin B-6 is based on the fact that PN-

glucoside is a major naturally occurring form of vitamin B-6

in many fruits and vegetables of human diets (65,75). The

results of the present studies suggest that PN-glucoside is

nutritionally significant in two ways. As shown previously,

PN-glucoside can act as a source of partially available

vitamin B-6 (77,81-82). In addition, this study has shown

that PN-glucoside can act as a weak antagonist that may hinder

the utilization of PN and possibly other forms of the vitamin.

These results indicate that PN-glucoside alters the

metabolism and in vivo retention of [14C]PN in the rat and that

PN-glucoside may retard the utilization of nonglycosylated

forms of vitamin B-6. PN-glucoside represents a significant

proportion of the total vitamin B-6 in many plant derived

foods (75). Therefore, the results of this study are

important in assessing the bioavailability of vitamin B-6 in

a typical mixed diet. The assessment of vitamin B-6 status in

human populations is contingent on identifying and quantifying

the content and distribution of the different forms of the

vitamin found in foods, determining the potential interactions

between these forms and the relative bioavailability of the

different vitamers. If it is found that PN-glucoside affects

the metabolic utilization of PN similarly in humans, it is








54

conceivable that the vitamin B-6 status in humans would not be

accurately reflected by current food consumption data.

Studies evaluating the metabolic interactions of PN-glucoside

on PN in humans are currently in progess.














CHAPTER 4
EFFECTS OF PYRIDOXINE-5'-fj-D-GLUCOSIDE ON THE
METABOLIC UTILIZATION OF PYRIDOXINE IN HUMANS


Introduction


A conjugated form of vitamin B-6, first isolated from

rice bran was identified as 5'-0-(P-D-

glucopyranosyl)pyridoxine (PN-glucoside) (67). Analysis of a

variety of plant-derived foods indicated that PN-glucoside is

abundant in many fruits and vegetables and the only

significant glycosylated form of vitamin B-6 present in the

foods examined (65,71,75). Assessing the nutritional status

of a person with respect to vitamin B-6 is a function of the

amount and form of vitamin B-6 in food, the bioavailability,

which is the extent of intestinal absorption and metabolic

utilization of the B-6 compounds in the diet and the specific

requirement of the individual.

Previous studies concerning the bioavailability of

vitamin B-6 in rats indicated that purified PN-glucoside is

relatively well absorbed compared to PN, undergoes little

metabolic utilization, and is rapidly excreted (2,77,80-81).

These results indicated 20%-30% net bioavailability of PN-

glucoside relative to PN. The extent of in vivo hydrolysis of

the glycosidic bond, rather than intestinal absorption, was


55








56

the limiting step in the utilization of PN-glucoside as

vitamin B-6 in the rat (2,77,80). A recent study in our

laboratory has shown that PN-glucoside alters the metabolism

and in vivo retention of simultaneously administered PN in the

rat (136).

A stable-isotopic study in humans found the

bioavailability of PN-glucoside to be 58% of that of free PN,

which is greater than that found in the rat (82). In an

investigation of lactating women, Andon et al. (84) reported

that dietary PN-glucoside contributed little to vitamin B-6

nutriture. These findings indicate the need for additional

data concerning factors affecting the utilization of PN-

glucoside and the potential effect of PN-glucoside on the

metabolic utilization of vitamin B-6 by humans consuming a

mixed diet. The study reported here was conducted to

determine the effect of PN-glucoside on the metabolic

utilization of PN in humans through the use of stable-isotopic

methods.


Materials and Methods


Synthesis of forms of vitamin B-6


The deuterium-labeled form of vitamin B-6 used in this

study was [5'-C2H2OH]pyridoxine (d2PN) which was prepared in

our laboratory as described by Coburn et al. (108). Mass

spectral analysis indicated that the d2PN species comprised

66% of the d2PN preparation, with a majority of the remainder








57

in the monodeutero (dl) form. All procedures were performed

under yellow-light to prevent photochemical degradation of B-6

vitamers.

PN-glucoside was prepared from PN-HC1 using biological

synthesis as previously described (80). Twelve grams of

alfalfa seeds were germinated in the presence of 62 mg of PN-

HC1 (297 Amol) obtained commercially (Sigma Chemical Company,

St. Louis, MO) followed by purification by cation-exchange

(80) and gel filtration chromatography on Sephadex G-10

(Pharmacia). The yield of purified PN-glucoside was 237 Amol

(80%). Chromatographic analysis of the purified PN-glucoside

preparation confirmed the absence of free (nonglycosylated) PN

and other forms of vitamin B-6.


Protocols of trials with human subiects


The study was conducted using the same adult men (n=6;

22-32 y) in each of three trials (Table 4.1). All subjects

were in good health and exhibited normal blood chemistry and

hematological values. The subjects were in normal vitamin B-6

status as determined by plasma pyridoxal phosphate (PLP) and

urinary 4-pyridoxic acid (4PA) concentration, erythrocyte

aspartate aminotransferase (AspAT) activity, and erythrocyte

AspAT stimulation by in vitro addition of PLP, as judged by

previously published criteria (137-138). The procedures for

selection of subjects and the experimental protocol were








58

approved by the University of Florida Institutional Review

Board and an informed consent was obtained from each subject.

Three trials were conducted to compare the in vivo

metabolic utilization of d2PN using the same group of six

subjects. The experimental period lasted seven weeks

consisting of three trials, each lasting one week with a two

week wash out period between each trial. A self-selected diet

was consumed for the duration of the study.

On the first day of each trial the subjects collected a

24-h urine sample into foil-covered polyethylene containers

and kept it refrigerated during the collection period. After

an overnight fast, a five ml blood sample was collected into

a Vacutainer brand tube (Becton Dickinson, Rutherford, NJ)

containing ethylenediaminetetraacetic acid (EDTA). Following

these sample collections the subjects consumed a single oral

dose of 5 gmol d2PN and either 2Amol PN or lmol PN + limol

PN-glucoside or 2Amol PN-glucoside (trials 1, 2, and 3

respectively) (Table 4.2) contained in 250 ml of apple juice.

The doses of unlabeled PN and PN-glucoside were selected to

provide and equal molar amount (2Amol) of either PN, PN+PN-

glucoside, or PN-glucoside. Urine collection was continued

for the following 48-h period. Urine was analyzed for

creatinine, total 4PA, and d24PA. The blood samples were

centrifuged and the plasma collected was analyzed for plasma

PLP and the erythrocytes were used to determine hemoglobin

concentration and AspAT activity.








59

Analytical methods


Urinary 4PA and plasma PLP were determined by reverse-

phase HPLC procedures with fluorometric detection (48,109).

Erythrocyte AspAT activity was measured by a

spectrophotometric assay procedure (138) with commercially

available reagents (Sigma Chemical, St. Louis, MO). Urinary

creatinine was determined based on the method of Heinegard and

Tiderstrom (139), while hemoglobin in erythrocyte hemolysates

was determined spectrophotometrically as the

cyanomethemoglobin derivative (140).


Mass spectral analysis of deuterium-labeled 4PA


Urinary d24PA was determined by gas chromatography-mass

spectrometry (GCMS) following isolation of 4PA from urine

samples by cation exchange chromatography (Bio-Rad AG 50W-X8,

100-200 mesh, H� form) and reverse-phase HPLC (Whatman,

Partisil 10 ODS-3 Magnum 9 column, 9mm i.d. x 25 cm) as

described by Gregory et al.(82). GCMS was performed in the

electron capture negative ionization mode (Model 4500 GCMS

system, Finnigan MAT, San Jose, CA) with a DB-5 capillary gas

chromatographic column (J&W Scientific, Folsom, CA) and

methane as a reagent gas. The derivatization (141) was

performed by dissolving the dried 4PA sample in 0.5 ml of 1:1

(v/v) solution of pyridine:acetic anhydride, heated at 100� C

for 90 min to form the 3-acetyl-4PA-lactone which is

evaporated to dryness under a stream of nitrogen. The 3-








60

acetyl-4PA-lactone was dissolved in 50 Al ethyl acetate and 2

1l of the resulting solution injected into the GCMS injection

port. As this 4PA derivative eluted from the GC column,

electron capture negative ionization yielded two anions at m/z

207 and 209 for the do and d2 4PA species, respectively.

Selected-ion monitoring at m/z 207 and 209 permitted

simultaneous measurement of the do and d2 4PA species during

GCMS analysis. A series of 4PA standards of known molar

ratios of do and d2 species were prepared and analyzed to

facilitate quantitative analysis. Calibration curves were

constructed to relate the ratios of observed GCMS peak areas

to actual molar isotope ratios of the d2/d0 4PA species. The

urinary excretion of d24PA was calculated using the isotope

ratio determined by GCMS, the concentration of total urinary

4PA as determined by reverse-phase HPLC, and total urine

volume.


Statistical analysis


The urinary excretion of d24PA was expressed as a

percentage of the administered dose of labeled PN. The

excretion of d24PA (as a percentage of administered dose),

total urinary 4PA, and vitamin B-6 nutritional status

parameters (see below) was compared by the method of least

squares analysis of variance using general linear model

procedures of SAS (123), following a log transformation of the

data to normalize variance when necessary. Data for d24PA and








61

total urinary 4PA are reported as least squares mean � the

pooled standard error of the least squares mean (SEM).

Orthogonal contrasts were made to examine the linear and

quadratic effects of different doses of PN-glucoside on

dependent variables. The quadratic effects were not

significant for any dependent variable. Values with p<0.05,

unless otherwise stated, were considered statistically

significant and were all for linear effects of the amount of

PN-glucoside (123).


Results


Vitamin B-6 nutritional status of subjects

The principal criteria used for assessing the vitamin B-6

status of the subjects were plasma PLP, urinary 4PA excretion

per 24-h, and erythrocyte aspartate aminotransferase

stimulation by in vitro addition of PLP. As shown in Table

4.1, mean values for these subjects were within the proposed

guidelines for vitamin B-6 nutritional adequacy (137-138).

These data indicate the adequate vitamin B-6 status of these

subjects prior to the administration of labeled vitamin B-6 in

each trial, which reflects the adequacy of their self-selected

diets prior to and between experiments. Subjects demonstrated

no statistical significant changes between trials for these

vitamin B-6 nutritional status parameters.









62
Table 4.1 Indicator of vitamin B-6 nutritional status of
human subjects 24 h prior to each trial1.

TRIAL URINARY 4PA PLASMA PLP ERYTHROCYTE
Amol/24 H nmol/L AspAT
STIMULATION
BY PLP



TRIAL 1 10.8�1.9 86.9�7.4 35.8�4.5
(5/mol d2PN
+ 2Amol PN)
(n=6)

TRIAL 2 8.0�1.8 82.4�10.3 36.4�6.3
(5Amol d2PN +
lmol PN
+ lrmol PN-GLUCOSIDE)
(n=6)

TRIAL 3 7.4�0.7 83.9�9.8 16.7�5.9
(5Amol d2PN +
2Amol PN-GLUCOSIDE)
(n=6)


1Mean and SEM, n=6; Blood and urine collections were made
24 h prior to administration of labeled forms of vitamin
B6. Proposed guidelines for adequacy of vitamin B6 status
are: urinary 4PA excretion, >5Amol/24 h (ref. 139); Plasma
PLP, >50 nmol/l (ref. 139); And in vitro stimulation of
erythrocyte aspartate aminotransferase (AspAT) by added
PLP, < 50% (ref. 140).







Stable-isotopic trials


This study was conducted to determine the effect of PN-

glucoside on the metabolic utilization of d2PN by evaluating

the excretion of d24PA following the ingestion of an oral dose

of d2PN while increasing the dose of PN-glucoside across the

three trials. The detection of urinary d24PA was conclusive









63

Table 4.2 Molar isotopic ratio of urinary d24PA.'

EXPERIMENTAL TRIALS


SUBJECT TRIAL 1 TRIAL 2 TRIAL 3
(5gimol d2PN + (5Amol d2PN + (5gmol d2PN
2jAmol PN) lmol PN + l mol 2Amol PN-
(n=6) PN-GLUCOSIDE) GLUCOSIDE)
(n=6) (n=6)

d2/dO
0-24 hr

1 0.20�.001 0.09�.002 0.07�.010
2 0.07�.001 0.l0�.001 0.12�.003
3 0.07�.003 0.08�.003 0.09�.001
4 0.17�.002 0.15�.010 0.10�.001
5 0.13�.001 0.01�.001 0.07�.001
6 0.10�.001 0.08�.001 0.12�.002
mean 0.12�.021 0.09�.018 0.09�.009

d2/dO
24-48 hr

1 0.05�.000 0.l0�.001 0.06�.000
2 0.09�.001 0.06�.000 0.03�.000
3 0.l0�.000 0.l0�.000 0.03�.000
4 0.07�.001 0.l0�.000 0.04�.001
5 0.09�.000 0.l0�.000 0.06�.000
6 0.05�.000 0.19�.001 0.07�.000
mean 0.07�.007 0.11�.017 0.05�.006


'Mean and SEM, n=6.
Values are average of three injections per sample.



evidence of the utilization of d2PN in vitamin B-6 metabolism.

Deuterium-labeled 4PA was detected in urine over 48-h post-

dose. The d24PA comprised a small portion of the total

urinary 4PA using this experimental design. However, the

isotope ratio observed was measurable by the GCMS method

employed. Twenty-four hours following the administration of








64

Table 4.3 Percentage of d2PN (5jAmol) dose excreted as urinary
d24PA.

EXPERIMENTAL TRIALS


TIME TRIAL 1 TRIAL 2 TRIAL 3
(H) (5tmol d2PN (5jAmol d2PN + (5Asmol d2PN
+ 2Amol PN) ljmol PN + limol 2Amol PN-
(n=6) PN-GLUCOSIDE) GLUCOSIDE)
(n=6) (n=6)


(Mmol) (jmol) (Amol) (SEM)


0-24 46.5 27.2 24.3 12.7

24-48 37.1 28.6 17.0 12.6

0-48 83.6 55.8 41.3 17.7


'Values based on n=6 are means and pooled standard error of
the mean (SEM). The quadratic effects were not
significant for any dependent variables. The values in
table are significantly different at p<0.05 for linear
effects of the amount of PN-glucoside.


the isotopically labeled d2PN, the mean d2/d0 molar isotopic

ratio of labeled 4PA was 0.12 � 0.02, 0.09 � 0.02, and 0.09 �

0.01 (mean � SEM) for trial 1, 2, and 3 respectively (Table

4.2). These were not significantly different. The 24-48 hr

molar isotopic ratios are also not significantly different

(Table 4.2).

The group mean values for excretion of d24PA as a

percentage of the administered d2PN dose between trials 1 and

3 for the 24-h post-dose period decreased linearly with the

amount of PN-glucoside (Table 4.3). Total 4PA excreted over

a 24-h and 48-h post-dose period showed group mean values for








65

Table 4.4 Total urinary 4PA excretion.1

EXPERIMENTAL TRIALS


TIME TRIAL 1 TRIAL 2 TRIAL 3
(H) (5AImol d2PN (5gimol d2PN + (5Amol d2PN
+ 2Amol PN) lg.mol PN + ltmol 2Amol PN-
(n=6) PN-GLUCOSIDE) GLUCOSIDE)
(n=6) (n=6)


(Mmol) (j1mol) (Mumol) (SEM)


0-24 26.0 13.5 10.3 3.9

24-48 15.5 10.5 11.5 4.82

0-48 41.5 24.0 21.8 7.63


'Values based on n=6 are means and pooled standard error of
the mean (SEM). The quadratic effects were not
significant for any dependent variables. The values in
table are significantly different at p<0.05 for linear
effects of the amount of PN-glucoside.
2 p<0.5

3 p<0.08


trials 1 and 3 to decrease linearly with the amount of PN-

glucoside (Table 4.4). These results indicate that the

percentage of d2PN dose excreted as d24PA showed an inverse

relationship in proportion to the PN-glucoside dose and that

total 4PA decreased linearly with the amount of PN-glucoside.


Discussion


Because PN-glucoside is a major form of vitamin B-6 in

plant derived foods, its influence on the utilization of








66

nonglycosylated species of the vitamin is nutritionally

significant. These effects, which may be minor in a mixed

diet containing both animal and plant derived foods, may be

much more pronounced in vegetarian diets. Kabir et al. (65)

showed an inverse relationship between the content of PN-

glucoside in a food and the net bioavailability in humans.

Our present knowledge of the essential role of vitamin B-

6 in preserving health emphasizes the importance of

establishing an accurate Recommended Dietary Allowance (RDA)

for vitamin B-6. A vitamin B-6 deficiency with apparent

clinical symptoms is rare in the general population, but the

vitamin B-6 requirement has not been clearly determined. It

is significant that over 50% of those evaluated in the 1977-78

USDA Food Consumption Survey consumed only 70% of the 1980 RDA

for vitamin B-6. Females and the elderly were especially

prevalent in the group consuming only 70% of the RDA (2). The

1985 Continuing Survey of Food Intakes of Individuals

indicated that only 27% of women consumed 70% or more of the

RDA for vitamin B-6 (143-144). Although the range of intakes

is large, substantial portions of the population consume

apparently marginally adequate amounts of vitamin B-6.

Adequate vitamin B-6 nutriture is essential to health through

the multiplicity of roles of its active coenzymatic form

pyridoxal 5'-phosphate (PLP).

The catabolic pathway for PLP involves enzymatic

hydrolysis of the phosphate ester bond, and the subsequent








67

oxidation of pyridoxal (PL) to 4PA (130-132). As a terminal

metabolite of vitamin B-6 metabolism, urinary 4PA reflects the

in vivo metabolic utilization of the vitamin. The deuterated

form of PN used in this study is labeled at a metabolically

stable position (i.e., 5'-CH2OH) and, therefore, retains the

label throughout absorption, metabolism, transport, and

excretion (82).

The results of this study demonstrate a negative

correlation between the quantity of PN-glucoside administered

and the amount of urinary d24PA derived from oral d2PN. The

data also showed that an inverse relationship between the

amount of total urinary 4PA excreted and the amount of PN-

glucoside in the dose. These findings are consistent with

those observed when comparing the metabolism of [14C]PN in the

presence of increasing doses of PN-glucoside in the rat (136).

The PN derived from hydrolysis of PN-glucoside enters into

vitamin B-6 metabolism to produce other forms of vitamin B-6,

including 4PA (77,81). However, the PN-glucoside that is not

hydrolyzed to PN may not be totally inert but instead appears

to affect the metabolism and interconversion of the B-6

vitamers. The possible inhibition by PN-glucoside of PL

kinase and/or PNP-oxidase, two enzymes in the metabolic

interconversion of B-6 vitamers, is discussed in chapter 5.

The results of these isotopic studies provide evidence that

PN-glucoside alters the metabolism of PN in humans and rats by








68

partially retarding the utilization of nonglycosylated B-6

compounds.

The appropriateness of the RDA is difficult, if not

impossible to assess without accurate information concerning

the content and forms of B-6 compounds in foods consumed, and

accurate data regarding vitamin B-6 bioavailability. This

research provides evidence of interactions amoung various B-6

compounds found in a typical mixed diet through an influence

of PN-glucoside on the utilization of PN. Such an interaction

represents another factor involved in the bioavailability of

vitamin B-6 and must be considered in fully understanding the

adequacy of vitamin B-6 intake. Future research will address

the nutritional significances of these interactions amoung the

vitamin B-6 compounds and the importance of these compounds in

assessing the nutritional status of the population for this

vitamin.














CHAPTER 5
EFFECTS OF PYRIDOXINE-5'-3-D-GLUCOSIDE ON THE
IN VITRO KINETICS OF PYRIDOXAL KINASE AND
PYRIDOXAMINE (PYRIDOXINE)-5'PHOSPHATE OXIDASE
IN RAT LIVER






Introduction


There are a number of enzymatic reactions in the metabolic

pathway responsible for the interconversion of vitamin B-6

compounds. In particular, PL kinase and PMP (PNP) oxidase are

considered the two most important catalytic enzymes in the

formation of the coenzymatic form of vitamin B-6, PLP (144).

Although other tissues contribute to vitamin B-6 metabolism,

the liver is thought to serve as the center of interconversion

of B-6 vitamers to PLP (109-110).

The B-6 vitamers are postabsorptively taken-up by the

liver and converted to PLP. The non-phosphorylated vitamers

are converted to their 5'-phosphorylated forms by a single

kinase (144). Following phosphorylation, PMP and PNP are

oxidized by PMP (PNP) oxidase to form PLP. This active

coenzymatic form of vitamin B-6 can then either be bound by

cellular apoproteins, released into circulation as PLP, or

dephosphorylated by alkaline phosphatase to PL (144). The PL


69








70

can then itself be released by the cell, oxidized to 4PA, the

metabolic end product of B-6 catabolism, or rephosphorylated

to form PLP.

Previously discussed studies in chapters 3 and 4 provided

evidence that PN-glucoside quantitatively alters the

metabolism and retention of administered PN (136). These

studies indicated that unlabeled PN-glucoside affects the

metabolic utilization of a simultaneous dose of labeled PN

(136). These data showed a linear relationship between the

dose of PN-glucoside and hepatic radiolabeled PNP and PLP

(136). Hepatic 14C PN, PL, and PM were inversely proportional

to the dose of PN-glucoside.

The data from these studies led us to postulate that the

effect on vitamin B-6 metabolism exhibited by PN-glucoside may

be caused by a direct affect of the glycosylated vitamer on

the action of either PL kinase, PMP (PNP) oxidase, or both.

The purpose of this study was to determine the effect of PN-

glucoside on the in vitro catabolic activity of PL kinase and

PMP (PNP) oxidase partially from purified rat liver.


Materials and Methods


Protocol


Two studies were conducted to evaluate the effect of PN-

glucoside on the reaction rates of the enzymes PL kinase and

PMP (PNP) oxidase. In each of the studies six male Sprague

Dawley rats (Crl:CD(SD)BR) from Charles River Breeding








71

Laboratories (Wilmington, MA), weighing 200-300 g, were housed

individually in stainless steel cages in animal quarters

maintained at 24�1" C with a 12 h light/dark cycle. All

procedures for the care and treatment of the experimental

animals were in accordance with the National Institutes of

Health Guidelines.

The rats were fed ad libitum a casein-sucrose (20%:60%)

based diet, adequate in all micronutrients including 7 mg PN

HCl/kg diet for seven days (80). At the end of the seven day

acclamation period the rats were killed by lethal injection of

sodium pentobarbital (W.A. Butler Co., Columbus, OH). The

livers were rapidly excised and stored at -20� C until

analysis. The enzyme assay was performed at the temperature

and reaction time determined from preliminary experiments and

pH as shown in method of Merrill and Wang (116). Four varying

concentrations of PN-glucoside ranging from 0 AM to 150 AM and

six varying substrate concentrations were used. Preliminary

experiments were also performed to determine the linear

ranges of the assays with respect to enzyme concentration and

reaction time.


Forms of vitamin B-6


PN-glucoside was prepared by biological synthesis using

the propagation of alfalfa sprouts germinated in the presence

of PN-HC1 (Sigma Chemical Co., St. Louis, MO) (80). The [4,5-
3H]PN-HCl 684.5 KBq/nmol (Hoffmann-LaRoche, Nutley, NJ) with








72

a purity of greater than 98%, as determined by ion-pair

reverse phase HPLC. The pyridoxamine-phosphate-HCl and

pyridoxal-5'-phosphate were obtained from (Sigma Chemical Co.,

St. Louis, MO).


Pvridoxal kinase

Extraction. Pyridoxal kinase was extracted from the

liver tissue of six rats using the procedure of Merrill et al.

(112). The tissue was stored at -20� C and processed within

72 h following removal from the rat. To each 1 g of liver, 4

ml of 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

(Hepes) (Sigma Chemical, Co., St. Louis, MO) at pH 7.4 and 25�

C, was added. The mixture was homogenized 30-45 s with a

Polytron PT 20 (Brinkman Instruments, Westbury, NY).> The

resulting homogenates were centrifuged (Model L, type 40

rotor, Beckman Instruments, Pala Alto, CA) at 105,000 x g for

1 h at 4� C. The supernatant was collected and stored at -20�

C.

Purification. The extracted PL kinase was purified

using the column chromatography method employed by Cash et al.

(145). Preparation of the 4-pyridoxyl-Sepharose as an

affinity support was adapted from the method used by Kwok and

Churchich (146). Seven grams of aminohexyl-Sepharose

(Sepharose AH, Pharmacia, Ltd. Piscataway, NJ) was pretreated

and washed with water. The Sepharose AH slurry was added to

a solution of 50 ml water containing 1 g PL-HC1 (Sigma

Chemical Co., St. Louis, MO) adjusted to pH 7.0 by addition of








73

NaOH. The slurry was shaken 16 h in the dark at ambient

temperature and then placed in an ice bath with the dropwise

addition of ice cold borohydride until disappearance of the

yellow color. During this period the slurry was shaken and 7%

acetic acid was added to prevent the pH from rising above 9.0.

The slurry was adjusted to pH 6.0 with 7% acetic acid and

packed into a 1.6 x 15 cm chromatography column. The column

was rinsed with 3M KC1 followed by water prior to use. All

rinsing and eluting buffers contained 2mM potassium phophate

(Fisher Scientific, Fairlawn, NJ) at pH 7.0 and 10mM

glutathione.

The 105,000 x g supernatant was applied to the PL-

Sepharose column and washed through with 50 ml of 100 mM KC1

buffer. The column was then rinsed with 500 ml 400mM KC1

followed by 50 ml of 100mM KC1 buffer. The PL kinase was then

eluted and adsorbed directly onto a Bio-Gel HTP

hydroxylapatite column (Bio Rad Laboratories, Richmond, CA),

previously equilibrated with buffer containing no KC1, with

500 ml 10 mM PN adjusted to pH 7.0 in buffer and rinsed with

25 ml 100mM KC1 buffer. The hydroxyapatite column was eluted

with 100 ml of a linear gradient of 2-300mM potassium

phosphate at pH 7.0. Approximately 3 ml fractions were

collected and the active fractions pooled. The enzyme

preparation was then tested for glucosidase activity using the

method of Daniels et al. (147) and for phosphatase activity

employing the PL kinase assay (discussed below) using PLP as








74

substrate followed by an inorganic phosphate assay kit (Sigma

Chemical Co., St. Louis, MO) and measuring the percent

conversion of PLP to PL.

Kinase assay. The assay procedure used was that of

Merrill and Wang (116). All reagents and buffers were at pH

5.75 and the procedure was conducted in GEF40GO gold lights

that emit at a wavelength between 500 nm and 750 nm to prevent

photodegradation of pyridoxyl compounds. The assay mixture

contained 10 AL each of 0.2M potassium phosphate, 10mM ATP (

Sigma Chemical Co., St. Louis, MO), 0.8mM ZnCl2, 0.6mM KC1,

0.3mM (20iCi/mmol) [3H]PN diluted with unlabeled PN to yield

the appropriate specific activity, and 100 Mg of enzyme

extract with enough water to make a final volume of 100 AL.

The mixture was incubated for 30 min at 370C and the reaction

halted with 0.5 ml ice cold 10 mM ammonium formate. The

samples were transfered to columns which contained 0.2 ml of

DEAE-cellulose (Sigma Chemical Co., St. Louis, MO) . The

columns were rinsed with 12 ml of 10mM ammonium formate,

followed by 2 ml water and eluted with 2 ml of 0.5 M KC1. A

2 ml fraction of the final elution was collected and added to

15 ml ScintiVerse II (Fisher Scientific, Fair Lawn, NJ) and

counted for total radioactivity. The reaction rates were

calculated using the following equation of Merrill and Wang

(116):

cpmreaction - cpmtie 0
Enzyme units = ___________________ X 2 X quench factor
Specific Activity
(enzyme preparation)








75

Pyridoxamine (pyridoxine) 5' phosphate oxidase


Extraction. The rat liver PMP (PNP) oxidase was

extracted by the method of Kazarinoff and McCormick (119).

Liver tissue was homogenized in 0.02M potassium phosphate

buffer containing 0.lmM mercaptoethanol, pH 7.0 and a ratio of

50 g tissue to 200 ml buffer. The homogenate was centrifuged

at 18,000 x g for 30 min and the supernatant adjusted to pH

5.0 with 2N acetic acid while stirring constantly. After 10

min the precipitate was removed by centrifugation at 18,000 x

g for 15 min. Potassium chloride (3.6 g) was added to the

supernatant and the volume brought up to 400 ml with water.

The pH was readjusted to 5.0 with 2N acetic acid.

Purification. The PMP (PNP) oxidase was extracted from

6 rat livers using the method of Kazarinoff and McCormick

(119). The clear supernatant resulting from the extraction

procedure was heated to 50 �C and held for 8 min then plunged

into ice and cooled to 20 �C and centrifuged at 18,000 x g for

30 min. The supernatant was applied to a DEAE-A50 column

(Sigma Chemical Co., St. Louis, MO) which had been prepared

by swelling overnight in 0.1 M potassium phosphate (pH 8.0).

A linear gradient established between 0.1 and 0.2 M potassium

phosphate at pH 8.0 was used to elute the column. Fractions

were collected and read at A280. The active fractions were

pooled. Solid ammonium sulfate (Sigma Chemical Co., St.

Louis, MO) (22.8 g/100 ml, 30% saturation) was added. After

30 min at 4�C the precipitate was removed by centrifugation at








76

18,000 x g for 30 min and addition of ammonium sulfate (15.2

g/100 ml, 50% saturation). After 60 min the solution was

centrifuged and the active precipitate was dissolved in one

tenth the initial volume of water. This solution was dialyzed

against 0.02 M potassium phosphate buffer for 24 h and

centrifuged to remove the precipitate. The clear yellow

supernatant was applied to a column of Sephadex G-100 (Sigma

Chemical Co., St. Louis, MO) equilibrated with 0.02 M

phosphate buffer and eluted with the same buffer. Fractions

were collected and read at an A280 and the active fractions

pooled.

Oxidase assay. PMP (PNP) oxidase activity was measured

by the method of Wada and Snell (31) using phenylhydrazine as

the color development agent. The assay mixture contained 1 ml

of ImM PMP solution in 0.2M Tris-HCl, pH 8.0, 10-200 units of

enzyme (one unit of oxidase enzyme catalyzes the formation of

Inmol PLP/mL/min at 37 �C), and enough Tris buffer to bring

the total volume to 3.5 ml. The mixture was gently shaken for

30 min at 37 �C. The reaction was stopped by the addition of

0.3 mL of 100% (w/v) trichloroacetic acid. There was no

formation of a precipitate therefore, the entire supernatant

was used for the color reaction. The 3 mL supernatant was

placed in a clean test tube, 0.2 mL phenylhydazine reagent (2

g dissolved in 100 mL 10ON H2SO4) was added and the mixture was

heated for 20 min at 60 "C. The reaction mixture was then

cooled to room temperature and the absorbance read at 410 nm








77

verses a reagent blank. The activity was expressed in nmol

PLP/mL enzyme/h.


Statistical analysis


The kinetic parameters Km and Vmx were calculated by non

linear regression using the EZ-FIT program (149). Confidence

intervals for each Km and Vx, were calculated by the equation:

b, � t.975 X SE, where b = Km or Vmx values (123). Overlapping

confidence intervals are not significantly different at

p<0.05.


Results


These studies were conducted to determine the effects of

PN-glucoside on two enzymes, PL kinase and PMP (PNP) oxidase,

in the metabolic pathway responsible for the interconversion

of vitamin B-6 compounds. These catalytic enzymes were

investigated, in vitro, by evaluating the rates of reaction in

the presence and absence of PN-glucoside. The purification

methods used for rat liver PL kinase and PMP (PNP) oxidase

yielded an enzyme preparation acceptable for kinetic analysis.


PL kinase

Partial purification of the enzyme by the above technique

resulted in a 243-fold increase in specific activity relative

to the crude homogenate (Table 5.1). Preliminary experiments

indicated that the rate of formation of [14C]PNP was linear

over varying enzyme concentrations and the rate of








78
Table 5.1 Purification of pyridoxal kinase from rat liver.




Fraction Vol Total Protein Units/mg
Purification (ml) Units (mg/ml) protein (fold)

Supernatant 80 509 0.25 2.54 ---

PL-Sepharose 6 188 0.05 628 247
/Hydroxyapatite


Units are nmol PNP/min/ml at 37"C in phosphate buffer pH
5.75.
50 g rat liver was used.



phosphorylation was constant over at least a 60 min period

under the assay conditions outlined above. The rate of

phosphorylation of PN in this study indicated a Km of 21 AM

which is in general agreement with the previously reported

value of 25 iM measured under similar conditions (9). In the

presence of varying concentrations of PN-gluoside the Km value

at all levels was not significantly different from that

observed with 0 A.M PN-glucoside (Table 5.2). The Vmx was 16.8

nmol of PNP/h/mg of protein and did not differ significantly

over the range of PN-glucoside concentrations (Table 5.2).

The kinetic parameters (Figure 5.1) were calculated from

analysis using the EZ-FIT Program (149).








79

Table 5.2 Kinetic parameters of pyridoxal kinase with varying
concentrations of PN-glucoside.

PN-glucoside Km V
concentration (AM) (nmol PNP/h/mg protein)
(iM)

0 23�2.0 0.29�.01
50 22�2.2 0.28�.01
100 21�2.4 0.28�.01
150 20�3.0 0.27�.01






PMP (PNP) oxidase


The purification scheme yielded a preparation of 24-fold

increase in enzyme purity compared to the crude homogenate

(Table 5.3). The conversion of the substrate, PMP, to PLP has

previously been shown to exhibit a Km value of 25AM (8). The

present study indicated a similar Km closely related value of

23AM. Values for Km of the substrate in the presence of all

concentrations of PN-glucoside were not significantly

different (Table 5.4). The maximum velocity values in this

study did not differ signifcantly from the control value of

1.65 nmol PLP/min/mg protein (Table 5.4). All kinetic

parameters (Figure 5.2) were calculated using the EZ-FIT

Program (149).








80

Table 5.3 Purification of pyridoxamine (pyridoxine) phosphate
oxidase from rat liver.

Fraction Vol Total Protein Units/
Purification (ml) Units (mg/ml) mg protein (fold)

Supernatant 200 11,300 15.0 3.77 ----
Acid ppt. 250 7,100 4.5 6.25 1.65
Ammonium
Sulfate 25 3,267 12.1 10.8 2.86
DEAE Sephadex 30 10,100 4.2 78.3 20.77
Sephadex G-100 45 6,961 1.7 91.1 24.16


Units are nmol PNP/min/ml at 37�C in phosphate buffer pH
5.75.
50 g of rat liver was used.


















Table 5.4 Kinetic parameters of pyridoxamine (pyridoxine)
phosphate oxidase with varying concentrations of PN-glucoside.

PN-glucoside Km Vx
concentration (AM) (nmol PLP/h/mg
(AM) protein)


0 23�3.1 1.65�0.10
50 27�1.0 1.69�0.09
100 24�2.2 1.63�0.11
150 24�2.0 1.70�0.06










12







0- 6

"o 4- *OuM PNG
E A 50uM PNG
2- * 100uM PNG
> U 150uM PNG
0 I I
0.000 0.020 0.040 0.060 0.080 0.100 0.120
-1 -1
(Pyridoxine), uM






Figure 5.1 Double reciprocal plot for the phosphorylation of PN by PL kinase from rat
liver in the presence of various levels of PN-glucoside. H











3.00
- 2.75 -
2.50
E 2.25
S2.00--
E 1.75
n 1.50
-J
n 1.25 0 0 uM PN-glucoside
S 1.00 Ar 50 uM PN-glucoside
E 3 100 uM PN-glucoside
: 0.75- V 150 uM PN-glucoside
9 0.50
> 0.25
0.00
0 14 28 42 56 70 84 98 112 126 140
-1 -1
(PYRIDOXAMINE- PHOSPHATE) uM





Figure 5.2 Double reciprocal plot for the conversion of PMP to PLP by PMP (PNP) oxidase
from rat liver in the presence of various levels of PN-glucoside.








83




Discussion


An earlier study from this laboratory demonstrated that

the conjugated vitamin B-6 compound, PN-glucoside, affected

the metabolic utilization of PN when administered in a

simultaneous dose, in rats (136). The concentration of

phosphorylated forms of vitamin B-6 increased in proportion to

the quantity of PN-glucoside. These data further indicated a

strong possibility that the effect PN-glucoside exhibited on

vitamin B-6 metabolism may be a direct effect on one or both

of the two enzymes responsible for the interconversion of B-6

vitamers to the active coenzyme form, PLP. Hence, the present

study was undertaken to determine the in vitro effects of PN-

glucoside on the reaction rates of PL kinase and PNP (PNP)

oxidase.

The results showed that neither PL kinase nor PMP (PNP)

oxidase were effected by the concentrations of PN-glucoside

used in the study. The kinetic parameters, KY and Vax, were

within the expected ranges for rat liver and demonstated no

significant differences when measured under conditions of

varying substrate and PN-glucoside concentrations.

When conducting in vitro kinetic studies consideration

should be given to any potential differences that may occur

under in vivo conditions. In the rat, vitamin B-6

postabsorptively diffuses into the liver where it is








84

metabolically trapped via phosphorylation. The vitamin is

maintained predominantly in phosphorylated forms due mainly to

protein binding and a ten fold greater activity of PL kinase

than phosphatase. It has been shown that hydrolysis of PLP is

inhibited by substrate and PMP (112).

Rat liver PMP (PNP) oxidase is inhibited by its product,

PLP, and the presence of PLP increases both the Km and Vw

(148). The increase in Vx may be a result of the action of

the more active rat liver phosphatases or an inability to

accurately measure initial velocity due to product inhibition

at early reaction times (148). Phosphatase more rapidly

hydrolyzes PLP when this product is free rather than protein

bound (144). The partially purified enzyme preparation used

in this study showed minimal phosphatase activity.

Approximately 20% of the PLP in vivo is in the free or loosely

bound form and the remaining PLP is bound to proteins.

Therefore, in vivo product inhibition may differ significantly

from in vitro inhibition. However, studies have shown that

crude rat liver homogenate compared with pure enzyme is

inhibited to a similar extent (148). Whether or not PLP

inhibits PMP (PNP) oxidase is dependent on the concentrations

of available substrate and product formed. Greater substrate

inhibition has been reported with PNP, in concentrations as

small as 5 Am, due to a reduced enzyme-substrate complex that

reacts less efficently with oxygen (112). This has

significant consideration, since the presence of increasing








85

concentration of PN-glucoside during the metabolism of PN

causes proportional increases in the concentration of liver

PNP (136).

In summary, the results from this study indicate that, in

vitro, PN-glucoside has no direct effect on PL kinase or

PMP(PNP) oxidase. These data, provide evidence that these

biosynthetic enzymes are not directly involved in the changes

in distribution of B-6 vitamers caused by ingested PN-

glucoside, observed in the previous study (136).














CHAPTER 6
SUMMARY AND CONCLUSIONS



Since the isolation and identification of PN-glucoside

from rice bran (67) much focus has been on the comparative

bioavailability of this glycosylated form and other vitamin B-

6 compounds. A number of studies spanning the last decade has

provided a great deal of information regarding the biological

availability of vitamin B-6 relative to its multiple forms.

As discussed in previous chapters, between 5%-80% of the

vitamin B-6 found in a variety of plant derived foods was in

the glycosylated form and both food processing and diet

composition are important considerations when evaluating the

bioavailability of vitamin B-6. Also the net bioavailability

of PN-glucoside relative to PN is less than 40% in rats

(75,77,90). Enzymatic hydroxylation of PN does not exhibit

antivitamin B-6 activity, however other degradative products

such as pyridoxallysine have been shown to be vitamin B-6

analogues and act as competitive inhibitors to enzymes in the

vitamin B-6 metabolic pathway.

The use of stable iotopic methodology showed PN-glucoside

to be 58% as biologically active as PN in humans (41). This

study indicated that although PN-glucoside bioavailability was



86








87

incomplete, it is approximately two fold greater than

previously found in rats.

The utilization of stable isotopic methodology in humans

continues to contribute information regarding the absorption,

metabolism and utilization of vitamin B-6 in humans. In an

effort to advance our understanding of vitamin B-6 nutriture

in humans it is necessary to gather information regarding the

dietary forms of vitamin B-6, their bioavailabililty and the

enzymes responsible for the interconversion of B-6 vitamers.

It is equally important to determine any potential interaction

between multiple forms of B-6 compounds and the effect these

may have on metabolic utilization of B-6 vitamers when

consumed together.

The purpose of the present research was to determine the

effect of PN-glucoside on the metabolic utilization of PN when

administered in a simultaneous dose in rats and humans. The

adventive of stable isotopic techniques allowed for the

evaluation in human subjects. The effect of PN-glucoside on

enzymes responsible for control of the conversion of vitamin

B-6 compounds to the active coenzyme form, PLP was also

determined.

The research discussed in chapters 3-5 was divided into

three separate but interconnected studies. The experimental

design of each study was contingent on the findings of the

prior study or studies. Initially, an animal study was

performed to determine if any interaction between PN-glucoside








88

and PN occurred, which affected metabolic utilization. The

premise of these studies was to simulate ingestion of a mixed

diet by determining the effect of a simultaneous dose of the

two B-6 compounds. Since the interest was in evaluating a

direct affect of PN-glucoside on PN metabolic utilization, the

rats received labeled-PN and varying concentrations of

unlabeled-PN-glucoside. The results indicated an effect of

PN-glucoside on the hepatic metabolism of PN. There existed

a linear relationship between the quantity of phosphorylated

B-6 vitamers and increasing concentrations of PN-glucoside.

It was also noted that the concentration of PNP, a B-6 vitamer

not detected in control rat livers, was directly proportional

to the concentration of PN-glucoside in the dose. The reason

for this finding is presently unclear. These data indicated

a negative correlation between the amount of 4PA detected in

urine and the concentration of PN-glucoside administered in

the dose. A similar correlation to that of 4PA existed for

non-phosphorylated B-6 vitamers in relation to PN-glucoside

concentration in the dose. These results indicated that the

PN-glucoside which was not hydolyzed and therefore

metabolically active as PN, is not biologically inert.

Instead this intact glycosylated B-6 compound exhibits an

effect on the utilization of vitamin B-6 in the rat.

As previously discussed, studies showed that the

bioavailability of PN-glucoside, although incomplete, was

greater in humans than rats. In consideration of this








89

observation, the effect of PN-glucoside on vitamin B-6

metabolism seen in rats would be significant, but of a lesser

magnitude in humans, assuming that intact PN-glucoside is the

active compound.

A second study was designed using stable isotopic methods

to determine the effect of PN-glucoside on metabolism of PN.

Deuterium-labeled [5'-C2H2OH]pyridoxine (d2PN) was used in the

study. The d2PN is metabolically stable and remains attached

to the PN molecule throughout catabolism. Therefore,

metabolic utilization of d2PN may be followed and ultimately

quantified as the urinary catabolite d24PA. The results of

this study indicated a negative correlation between the

quantity of total 4PA excreted and the concentration of PN-

glucoside in the dose.

Review of the results of these two studies showed that

PN-glucoside, when administered simultaneously with PN, has an

effect on the metabolic utilization of vitamin B-6 and that

PN-glucoside which is not hydrolyzed by intestinal mucosal

glucosidase, exhibits some biological activity regarding

vitamin B-6 metabolism. To determine which vitamin B-6

metabolic process was affected by PN-glucoside, study three

was proposed. Data from chapter 3 indicated increased

quantities of phosphorylated B-6 vitamers, specifically PLP

and PNP. These results indicated that either of the enzymes

responsible for controlling the level of phosphorylated B-6

compounds may be affected by PN-glucoside. Study three was








90

designed to determine the in vitro effects of PN-glucoside on

PL kinase and PMP (PNP) oxidase. The results indicated no

direct effect of PN-glucoside on either enzyme as determined

by no significant differences observed between the kinetic

parameters, NK and V.. with varying concentrations of the

glycosylated vitamin B-6 compound. Further studies are needed

to assess potential effects of PN-glucoside on other enzymes

in the interconversion pathway of vitamin B-6 metabolism.

These experimenal results provide evidence that PN-

glucoside exhibits a quantitative effect on the metabolic

utilization of PN. Although the biochemical mechanism

responsible for this effect are presently uncertain, it is

evident that consideration in the study of vitamin B-6

bioavailability must incorporate the potential interactions

among B-6 compounds present in foods generally consumed in a

mixed diet.

The information from these studies is consequential in

relation to current USDA Dietary Guidelines, which are used to

convey recommendations to the general population. Present

nutrition education is promoting the consumption of greater

quantities of plant derived foods. Food consumption data have

shown that there are several sub-populations at nutritional

risk for vitamin B-6, specifically pregnant females,

adolescents, and the elderly.those individuals undergoing

growth or aging.








91

The predominant form of vitamin B-6 in many fruits and

vegetables is PN-glucoside and research has indicated that

not only is this vitamin B-6 compound biologically less

available as a usable source of vitamin B-6 it also effects

the metabolic utilization of other B-6 vitamers. In

development of recommendations for dietary allowances and

dietary food intake patterns it is necessary to consider not

only vitamin B-6 availability and biological activity, but

interactions that occur between different vitameric forms and

how these interactions effect bioavialability of the nutrient.




Full Text
25
The presence and type of purified dietary fiber have only
a minor effect on the bioavailability of the B-6 vitamers (91-
93). Nuygen et al. (91) evaluated the potential for physical
binding of vitamin B-6 using a variety of native and modified
polysaccharides and lignin under conditions similar to the
human intestine. In vitro binding of the B-6 vitamers by
these fiber components did not occur. Differences in
viscosity may have influenced the rate of diffusion, although
studies by Machida and Nagai (93) indicated that a reduced
rate of absorption does not induce a reduction in net vitamin
B-6 absorption. Rat and chick bioassays have been employed to
evaluate the effects of selected dietary fiber components on
the absorption of vitamin B-6 (91,93). The results of these
bioassays indicated no inhibitory effects of these dietary
fiber components on the bioavailability of vitamin B-6. The
results of these studies along with the results of studies
with human subjects (39,56) suggest that dietary fiber has
little effect on the bioavailability of vitamin B-6 in foods.
The slight effect of dietary fiber components on the
bioavailability of vitamin B-6 does not fully account for the
lower bioavailability of B-6 vitamers from plant food products
relative to animal sources.
Many vitamin B-6 antimetabolites have been identified
and determined to be chemical reaction products formed from
vitamin B-6 during the thermal processing or storage of foods.
The potential conversion of various B-6 vitamers to these


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE EFFECTS OF PYRIDOXINE 5' -/3-D-GLUCOSIDE
ON THE METABOLIC UTILIZATION OF PYRIDOXINE
IN RATS AND HUMANS
By
Joyce Ann Gilbert
December, 1991
Chairperson: Dr. Jesse F. Gregory III
Major Department: Food Science and Human Nutrition
A major form of vitamin B-6 in plant-derived foods is 5'-
0-(/3-D-glucopyranosyl) pyridoxine (PN-glucoside) Previous
studies have shown that PN-glucoside is poorly available as a
source of vitamin B-6 in rats and undergoes incomplete
utilization in humans. The present research was conducted to
determine whether unlabeled PN-glucoside affects the metabolic
utilization of simultaneously administered isotopically
labeled pyridoxine (PN) in rats and humans. In addition, the
in vitro effect of PN-glucoside on the activity of enzymes in
the vitamin B-6 metabolic pathway, specifically pyridoxal-
kinase and pyridoxamine (pyridoxine)-phosphate oxidase was
determined.
Experimental results with rats given [UC]PN indicated
that urinary excretion of KC increased significantly with
viii


83
Discussion
An earlier study from this laboratory demonstrated that
the conjugated vitamin B-6 compound, PN-glucoside, affected
the metabolic utilization of PN when administered in a
simultaneous dose, in rats (136). The concentration of
phosphorylated forms of vitamin B-6 increased in proportion to
the quantity of PN-glucoside. These data further indicated a
strong possibility that the effect PN-glucoside exhibited on
vitamin B-6 metabolism may be a direct effect on one or both
of the two enzymes responsible for the interconversion of B-6
vitamers to the active coenzyme form, PLP. Hence, the present
study was undertaken to determine the in vitro effects of PN-
glucoside on the reaction rates of PL kinase and PNP (PNP)
oxidase.
The results showed that neither PL kinase nor PMP (PNP)
oxidase were effected by the concentrations of PN-glucoside
used in the study. The kinetic parameters, and V^, were
within the expected ranges for rat liver and demonstated no
significant differences when measured under conditions of
varying substrate and PN-glucoside concentrations.
When conducting in vitro kinetic studies consideration
should be given to any potential differences that may occur
under in vivo conditions. In the rat, vitamin B-6
postabsorptively diffuses into the liver where it is


89
observation, the effect of PN-glucoside on vitamin B-6
metabolism seen in rats would be significant, but of a lesser
magnitude in humans, assuming that intact PN-glucoside is the
active compound.
A second study was designed using stable isotopic methods
to determine the effect of PN-glucoside on metabolism of PN.
Deuterium-labeled [5'-C2H2OH]pyridoxine (d2PN) was used in the
study. The d2PN is metabolically stable and remains attached
to the PN molecule throughout catabolism. Therefore,
metabolic utilization of d2PN may be followed and ultimately
quantified as the urinary catabolite d24PA. The results of
this study indicated a negative correlation between the
quantity of total 4PA excreted and the concentration of PN-
glucoside in the dose.
Review of the results of these two studies showed that
PN-glucoside, when administered simultaneously with PN, has an
effect on the metabolic utilization of vitamin B-6 and that
PN-glucoside which is not hydrolyzed by intestinal mucosal
glucosidase, exhibits some biological activity regarding
vitamin B-6 metabolism. To determine which vitamin B-6
metabolic process was affected by PN-glucoside, study three
was proposed. Data from chapter 3 indicated increased
quantities of phosphorylated B-6 vitamers, specifically PLP
and PNP. These results indicated that either of the enzymes
responsible for controlling the level of phosphorylated B-6
compounds may be affected by PN-glucoside. Study three was


80
Table 5.3 Purification of pyridoxamine (pyridoxine) phosphate
oxidase from rat liver.
Fraction
Purification
Vol
(ml)
Total
Units
Protein
(mg/ml)
Units/
mg protein
(fold)
Supernatant
200
11,300
15.0
3.77
Acid ppt.
Ammonium
250
7,100
4.5
6.25
1.65
Sulfate
25
3,267
12.1
10.8
2.86
DEAE Sephadex
30
10,100
4.2
78.3
20.77
Sephadex G-100
45
6,961
1.7
91.1
24.16
Units are nmol PNP/min/ml at 37C in phosphate buffer pH
5.75.
50 g of rat liver was used.
Table 5.4 Kinetic parameters of pyridoxamine (pyridoxine)
phosphate oxidase with varying concentrations of PN-glucoside.
PN-glucoside V|nax
concentration (/liM) (nmol PLP/h/mg
(juM) protein)
0
233.1
1.650.10
50
271.0
1.690.09
100
242.2
1.630.11
150
242.0
1.7010.06


30
the beta-glycosidic bond of PN-glucoside by mucosal beta-
glucosidase releasing free PN. Initial studies employing
bioassays with human subjects (40) indicated that the
proportion of the total vitamin B-6 which was present as PN-
glucoside correlated inversely with the bioavailability of the
vitamin in the foods examined (tuna, peanut butter, and whole
wheat bread). These results suggested that PN-glucoside would
not be biologically available to humans. However, studies by
Bills et al. (78) indicated that this observed correlation was
inconsistent when additional foods were examined and that the
quantity of PN-glucoside present in a food product could
function as a predictor of the bioavailability of vitamin B-6
from that food source. Stable-isotopic techniques have been
used to evaluate the bioavailability of purified deuterium-
labeled (d2) PN-glucoside in human subjects (82) In these
studies, the bioavailability of orally administered
extrinsically enriched oatmeal containing labeled PN-glucoside
was similar to that of labeled-PN fed under identical
conditions (82) Although metabolic utilization of PN-
glucoside appears to have a substantial level of individual
variability among human subjects, the bioavailability of PN-
glucoside in human subjects has been observed to be
substantially greater than that of the rat (82) Further
application of stable-isotopic techniques provide greater
clarification of the factors affecting bioavailability of
vitamin B-6.


49
effect of PN-glucoside on the metabolic utilization of [UC]PN
and distribution of the 14C-labeled B-6 metabolites in the
liver 24 h post-dose when the two vitamers are consumed
together. Expression of the data in this manner reflects the
direct effect of PN-glucoside on the distribution of B-6
vitamers and urinary excretion of 4PA as well as interactions
between different naturally occurring forms of vitamin B-6
found in foods consumed simultaneously in a typical mixed
diet. The bioavailability of vitamin B-6 compounds is a
function of their extent of absorption and metabolic
conversion to active coenzymatic forms. Pyridoxine that is
absorbed from the intestinal tract is concentrated initially
in the liver (124) and is the sole source of plasma PLP (125).
Thus, liver plays a central role in the overall metabolism of
vitamin B-6 (126).
The major transformations in hepatic vitamin B-6
metabolism involve phosphorylation catalyzed by pyridoxal
kinase, oxidation of PNP and PMP by pyridoxine (pyridoxamine)
5'-phosphate oxidase (PNP oxidase), along with interconversion
of PLP and PMP through transamination reactions (109-110,127-
129). The principal forms of vitamin B-6 in liver are PLP and
PMP (129-131). PNP is usually present in only trace
quantities because of its rapid oxidation to PLP (130,132-
133). The non-phosphorylated B-6 vitamers constitute less
than 10% of the total vitamin B-6 content in the liver (131).
In this study, the concentration of [UC]PNP increased linearly


36
It is well known that the product PLP, inhibits the
activity of PMP(PNP) oxidase (112) As a result, other
phospho-B-6 analogues have been evaluated (118). It was
discovered that a dianionic charge on the 5'-position is
necessary for binding of the substrate analogue and
subsequently inhibition of the enzyme (122). Inhibition of
PMP (PNP) oxidase also occurs when alterations are introduced
in the structure of the flavin coenzyme, FMN (118). In as
much as both PL kinase and PMP (PNP) oxidase have substrate
reactivity with structural analogues, it is not unlikely that
conjugated forms of vitamin B-6 would inhibit these enzymes.
Conflicting data have been reported concerning the extent
of bioavailability of PN-glucoside in vitamin B-6 metabolism.
Most experimental results in rats have indicated 20-30% net
bioavailability of PN-glucoside relative to PN. The
bioavailability of PN-glucoside, although incomplete, is
substantially greater in humans than in rats. While the
incomplete utilization of PN-glucoside has been clearly
established in rats and humans, its potential interactive
effects in vitamin B-6 metabolism have not been examined
previously.
The purpose of this research was to demonstrate the
effects PN-glucoside has on the metabolic utilization of
vitamin B-6. The effort to evaluate these effects is an
important one, when considering the overall nutritional status
of individuals with respect to vitamin B-6. Being aware of


93
11. Kirksey A., Keaton K., Abernathy R.P. (1978). Vitamin
B-6 Nutritional Status of a group of Female
Adolescents. Am J Clin Nutr 31: 946-954.
12. U.S. Department of Agriculture. (1986) Nationwide
Food Consumption Survey Continuing Survey of Food
Intakes by Individuals, Low Income Women 19-25 Years
and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII
Report No. 85-2. U.S. Department of Agriculture,
Hyattsville, MD.
13. Shane B., Contractor S.F. (1975). Assessment of
Vitamin B-6 Status. Studies on Pregnant Women and Oral
Contraceptive Users. Am J Clin Nutr 28: 739-747.
14. Roepke J.L.B., Kirksey A. (1979). Vitamin B-6
Nutriture During Pregnancy and Lactation. Am J Clin
Nutr 28: 146-156.
15. Pao E.M., Mickle S.J. (1981). Problem Nutrients in
the United States. Food Technol 35(9): 58-69.
16. Gregory J.F., Kirk J.R. (1981). The Bioavailability
of Vitamin B-6 in Foods. Nutri Rev 39(1): 1-8.
17. Turner J.M. (1961). Pyridoxal Phosphate Breakdown by
an Alkaline-Phosphatase Preparation. Biochem J 80:
663-668.
18. Leklem J.E. (1988). Vitamin B-6: Of Reservoirs,
Receptors and Requirements. Nutr Today Sept/Oct: 4-
10.
19. Booth C.C., Brain M.C. (1962). The Absorption of
Tritium Labelled Pyridoxine Hydrochloride in the Rat.
J Physiol 164: 282-294.
20. Scudi J.V., Unna K., Antopol W. (1940). A Study of
Urinary Excretion of Vitamin B-6 by a Colorimetric
Method. J Biol Chem 135: 371-376.
21. Scudi J.V., Koones H.F., Keresztesy J.C. (1940).
Urinary Excretion of Vitamin B-6 in the Rat. Proc Soc
Exptl Biol Med 43: 118-122.
22. Yamada K., Tsuji M. (1968). Transport of Vitamin B-6
in Human Erythrocytes. J Vitaminol (Kyoto) 14: 282-
294.
23. Middleton H.M. (1979). In Vivo Absorption and
Phosphorylation of Pyridoxine-HCl in Rat Jejunum.
Gastroenterol 76: 46-49.


2
of absorption and metabolic utilization of the B-6 compounds
in the diet, and the specific requirements of the individual.
Although vitamin B-6 is widely distributed in nature, the
amount in food is relatively small. The American diet often
contains 1 2 mg per day (2) Marginal vitamin B-6
nutritional status has been well documented in several
segments of the American population including adolescents,
pregnant women and the elderly (6-15). However, vitamin B-6
deficiency with apparent clinical symptoms is not widespread
in the general population. It is significant that over 50% of
the subjects of the USDA 1977-78 Nationwide Food Consumption
Survey consumed less than 70% of the Recommended Dietary
Allowance (RDA) for vitamin B-6 (2) The 1985 Continuing
Survey of Food Intakes of Individuals indicated that only 27%
of women consumed 70% or more of the RDA for vitamin B-6 (12) .
Women of all ages and the elderly have been especially
prevalent in the group consuming less than the RDA for vitamin
B-6. Vitamin B-6 is required in amino acid metabolism and
therefore, the RDA for this vitamin is related to protein
intake. Twice the RDA for protein, which is considered the
upper boundary of acceptable level of protein intake, was used
to establish the RDA for vitamin B-6. The RDA for vitamin B-6
was revised using a dietary ratio of 0.016 mg vitamin B-6/g
protein.
Our present knowledge of the essential role of vitamin B-
6 in preserving health emphasizes the importance of


88
and PN occurred, which affected metabolic utilization. The
premise of these studies was to simulate ingestion of a mixed
diet by determining the effect of a simultaneous dose of the
two B-6 compounds. Since the interest was in evaluating a
direct affect of PN-glucoside on PN metabolic utilization, the
rats received labeled-PN and varying concentrations of
unlabeled-PN-glucoside. The results indicated an effect of
PN-glucoside on the hepatic metabolism of PN. There existed
a linear relationship between the quantity of phosphorylated
B-6 vitamers and increasing concentrations of PN-glucoside.
It was also noted that the concentration of PNP, a B-6 vitamer
not detected in control rat livers, was directly proportional
to the concentration of PN-glucoside in the dose. The reason
for this finding is presently unclear. These data indicated
a negative correlation between the amount of 4PA detected in
urine and the concentration of PN-glucoside administered in
the dose. A similar correlation to that of 4PA existed for
non-phosphorylated B-6 vitamers in relation to PN-glucoside
concentration in the dose. These results indicated that the
PN-glucoside which was not hydolyzed and therefore
metabolically active as PN, is not biologically inert.
Instead this intact glycosylated B-6 compound exhibits an
effect on the utilization of vitamin B-6 in the rat.
As previously discussed, studies showed that the
bioavailability of PN-glucoside, although incomplete, was
greater in humans than rats. In consideration of this


31
The bioavailability of vitamin B-6 compounds is a
function of their extent of absorption and metabolic
utilization, as reflected by conversion to active coenzymatic
forms. It is important to assess the bioavailability of
vitamin B-6 as consumed in a typical mixed diet. The
assessment of vitamin B-6 status in populations is contingent
on identifying and guantifying the different forms of the
vitamin occurring naturally in foods consumed, determining the
potential interactions between these forms and the relative
bioavailability of the different vitamers, and their possible
interactions.
Our present knowledge of the essential role of vitamin
B-6 in preserving health emphasizes the importance of
establishing an accurate intake for vitamin B-6. The adeguacy
of intake is difficult to assess critically without accurate
information regarding the amount of the vitamin in foods
consumed, the effect of various vitamin B-6 compounds on the
enzymes involved in the metabolism of various forms of vitamin
B-6, as well as information regarding vitamin B-6
bioavailability.
Enzymatic Interconversion of B-6 Vitamers
There are several enzymatic reactions responsible for the
interconversion of the B-6 vitamers. The liver serves as the
center of vitamin B-6 metabolism and contains the enzymes
necessary for the metabolic interconversion of the B-6


29
A number of derivatives of PN-glucoside have been
isolated and identified. Tadera et al. (73,107) reported
several additional PN-glycosides to be minor components of the
total vitamin B-6 in pea seedlings and rice bran (108).
Assuming that PN-glucoside esters occur naturally and in
great enough quantities, it is unclear whether these glucosyl
compounds contribute significantly to the total vitamin B-6 in
foods and, if so how they may impact vitamin B-6 nutriture
when consumed. However, it is worth mentioning that PN-
glucoside has been the only glycosylated form of vitamin B-6
detected in a variety of foods utilizing HPLC methodology
(71,77). Tadera et al. (108) have observed the presence of a
vitamin B-6 conjugate termed "B-6X". The authors described
this compound as PN-glucoside esterified to an organic acid
based on its response in microbiological assays following
sequential alkaline treatment and hydrolysis with beta-
glucosidase (108) .
The nutritional properties of these conjugated forms of
vitamin B-6 are not fully understood at the present time. The
nutritional significance of these glycosylated vitamin B-6
compounds is related to their potentially limited
bioavialability. As earlier stated, PN-glucoside comprises 5%
to 80% of the total vitamin B-6 in many plant tissues (67,75).
Its bioavailability for humans is approximately 58% relative
to that of PN when administered orally (82) PN-glucoside has
been found to undergo intestinal absorption via hydrolysis of


42
HPLC equipment and analysis
The separation of radiolabeled B-6 compounds in the urine
and liver tissue was accomplished by ion-pair reverse-phase
HPLC (82). Chromatographic analysis was performed with a
Rainin HP/HPX Drive Module (Rainin Instrument, Woburn, MA).
All of the above analysis utilized a loop injection valve
(Model 904-2,Altex), a fluorometric detector (Model LS-5,
Perkin-Elmer, Norwalk, CT) and an electronic integrator (Model
3388A, Hewlett-Packard, Avondale, PA). Excitation and
emission wavelengths were 295nm (5nm slit width) and 405nm
(5nm slit width) respectively. Two mobile phases were employed
in a gradient elution procedure using an Ultrasphere IP 5 /m
C-18 4.6 mm x 25 cm column (Beckman Instruments, San Ramon,
CA) Mobile phase A contained 0.03 3 mol/L phosphoric acid and
8 mmol/L 1-octanesulfonic acid, adjusted to pH 2.2 with 6
mol/L KOH. Mobile phase B contained 0.033 mol/L phosphoric
acid and 3.4 mol/L acetonitrile, adjusted to pH 2.2 with 6
mol/L KOH and no ion-pairing reagent. Fractions (0.5 mL) were
collected using an ISCO Cygnet Fraction Collector (ISCO,
Lincoln, NB) The identity of the PLP and PNP peaks was
confirmed by monitoring the formation of pyridoxal (PL) and
PN, respectively, by HPLC following incubation of the PLP and
PNP fractions in 0.1 mol/L HC1 at 80 C for 3 h.


104
133. Snell E.E., Haskell B.E. (1971) The Metabolism of
Vitamin B-6. In: Comprehensive Biochemistry, Chapter
I Section C. (Florkin M., Stotz E.H., eds.) Vol. 21
Elsevier, Amsterdam. 47-71.
134. Merrill A.H. Jr., Henderson J.M. (1990) Vitamin B-6
Metabolism by Human Liver. In: Vitamin B-6
(Dakshinamurti K., ed.) Vol. 585 Annals of the New
York Academy of Sciences, New York. 110-117.
135. Beaton G.H., Cheney M.C. (1965) Vitamin B-6
Requirements of the Male Albino Rat. J Nutr 87: 125-
134.
136. Gilbert J.A., Gregory J.F. (1991) The Effect of
Pyridoxine-5 '-/3-D-glucoside on the Metabolic
Utilization of Pyridoxine in the Rat. FASEB J 1:A586
(abs. 1250).
137. Sauberlich H.E., Dowdy R.P., Skala J.H. (1974) Vitamin
B-6. In: Laboratory Tests for the Assessment of
Nutritional Status. CRC Press, Cleveland, OH. 37-49.
138. Committee on Enzymes of the Scandinavian Society for
Clinical Chemistry and Clinical Physiology (1974)
Recommended Methods for the Determination of Four
Enzymes in Blood. Scand J Clin Lab Invest 33: 291-
306.
139. Heinegard D., Tiderstrom G. (1973) Determination of
Serum Creatinine by a Direct Colorimetric Method. Clin
Chim Acta 43: 305-310.
140. Crosby W., Munn J.I., Furth F.W. (1954)
Standardization in a Method for Clinical
Hemoglobinometry. U.S. Armed Forces Med J 5: 693-703.
141. Hachey D.L., Coburn S.P., Brown L.T., Erbelding W.F.,
DeMark B., Klein P.D. (1985) Quantitation of Vitamin
B-6 in Biological Samples by Isotope Dilution Mass
Spectrometry. Anal Biochem 151: 159-168.
142. U.S. Department of Agriculture. (1986) Nationwide
Food Consumption Survey, Continuing Survey of Food
Intakes by Individuals, Low-Income Women 19-50 Years
and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII
Report No. 85-2. U.S. Department of Agriculture,
Hyattsville, Md.


LIST OF TABLES
Table 3.1 Liver B-6 vitamer distribution and total
liver 14C in rats administered varying levels of PN-
glucoside in the dose (Experiment 1) .* 44
Table 3.2 Urinary 4PA and total urinary 14C in rats
administered varying levels of PN-glucoside
(Experiment 1)* 45
Table 3.3 Liver B-6 vitamer distribution and total
liver 14C in rats administered varying levels of PN
or PN-glucoside. (Experiment 2)* 46
Table 3.4 Urinary 4PA and total urinary 14C in rats
administered varying levels of PN or PN-glucoside.
(Experiment 2)* 47
Table 4.1 Indicator of vitamin B-6 nutritional status
of human subjects 24 h prior to each trial1. ... 62
Table 4.2 Molar isotopic ratio of urinary d24PA.1 . 63
Table 4.3 Percentage of d2PN (5unol) dose excreted as
urinary d24PA 64
Table 4.4 Total urinary 4PA excretion.1 65
Table 5.1 Purification of pyridoxal kinase from rat
liver 78
Table 5.2 Kinetic parameters of pyridoxal kinase with
varying concentrations of PN-glucoside 79
Table 5.3 Purification of pyridoxamine (pyridoxine)
phosphate oxidase from rat liver 80
Table 5.4 Kinetic parameters of pyridoxamine
(pyridoxine) phosphate oxidase with varying
concentrations of PN-glucoside 80
vi


71
Laboratories (Wilmington, MA) weighing 200-300 g, were housed
individually in stainless steel cages in animal quarters
maintained at 241 C with a 12 h light/dark cycle. All
procedures for the care and treatment of the experimental
animals were in accordance with the National Institutes of
Health Guidelines.
The rats were fed ad libitum a casein-sucrose (20%:60%)
based diet, adequate in all micronutrients including 7 mg PN
HCl/kg diet for seven days (80) At the end of the seven day
acclamation period the rats were killed by lethal injection of
sodium pentobarbital (W.A. Butler Co., Columbus, OH). The
livers were rapidly excised and stored at -20 C until
analysis. The enzyme assay was performed at the temperature
and reaction time determined from preliminary experiments and
pH as shown in method of Merrill and Wang (116) Four varying
concentrations of PN-glucoside ranging from 0 ¡jlK to 150 xM and
six varying substrate concentrations were used. Preliminary
experiments were also performed to determine the linear
ranges of the assays with respect to enzyme concentration and
reaction time.
Forms of vitamin B-6
PN-glucoside was prepared by biological synthesis using
the propagation of alfalfa sprouts germinated in the presence
of PN-HC1 (Sigma Chemical Co., St. Louis, MO) (80). The [4,5-
3H]PN-HC1 684.5 KBq/nmol (Hoffmann-LaRoche, Nutley, NJ) with


47
iCi) [UC]PN, while treatment groups 1 and 2 received 0 nmol
and 72 nmol unlabeled PN-glucoside, respectively. Treatment
group 3 received 21.6 nmol unlabeled PN based on an assumed
30% net bioavailability of PN-glucoside relative to PN (77,80-
81) Statistical analysis by orthogonal contrasts of the data
from this study indicated no statistically significant linear
effects between the quantity or distribution of labeled
vitamin B-6 compounds in liver tissue and urine and treatment
groups one and three (Table 3.3). However, there was a
significant linear effect (p<0.05) in the relative
concentration of phosphorylated vitamin B-6 compounds PLP,
Table 3.4 Urinary 4 PA and total urinary UC in rats
administered varying levels of PN or PN-glucoside.
(Experiment 2)*
LEVELS OF PN-GLUCOSIDE LEVELS OF PN
0 nmol
72 nmol
21.6 nmol
SEM
% urinarv
radioactivitv
4 PA
35.5
14.2b
32.7
12.3
UC
%
of dose
30.0
46.0a
27.0
4.0
Values based on n=6 are means and pooled standard error
of the mean (SEM) Log transformation of the data was
performed prior to analysis of variance. The quadratic
effects were not significant for any dependant variable.
a Significantly different at p<0.05 for linear effects of
PN-glucoside in the dose.
b Significantly different at p<0.05 for decreased linear
effects.


LITERATURE CITED
1. Canhaxn J.E., Baker E.M., Harding R.S., Sauberlich H.E.,
Plough I.C. (1969). Dietary Protein Its
Relationship to Vitamin B-6 Reguirements and Function.
Ann NY Acad Sci 166: 16-29.
2. Food and Nutrition Board (1989). Recommended Dietary
Allowances. Natl Acad Sci., Washington DC.
3. Meister A. (1965). The Function of Vitamin B-6 in
Amino Acid Metabolism. In: Biochemistry of the Amino
Acids, 2nd Ed., Vol. 1 Academic Press, New York. 375-
412.
4. Bender D.A. (1987). Oestrogens and Vitamin B-6 Actions
and Interactions. World Rev Nutr Diet 51: 140.
5. Bunce G.E., Vessal M. (1987). Effect of Zinc and/or
Pyridoxine Deficiency upon Oestrogen Retention and
Oestrogen Receptor Distribution in the Rat Uterus. J
Steroid Biochem 26: 303.
6. Rose C.S., Gyorgy P., Butler M. (1976). Age
Differences in Vitamin B-6 Status in 617 Men. Am J
Clin Nutr 29: 847-853.
7. Schuster K., Bailey L.B., Mahan C.S. (1984). Effect
of Maternal Pyridoxine HCL Supplementation on the
Vitamin B-6 Status of Mother and Infant and on
Pregnancy Outcome. J Nutr 114: 977-988.
8. Brophy M.H., Suteri P.K. (1975). Pyridoxal Phosphate
and Hypertensive Disorders of Pregnancy. Am J Obstet
Gynecol 121:1075-1079.
9. Ranke E., Tauber A., Horonick B. (1960). Vitamin B-6
Deficiency in the Aged. J Gerontol 15: 41-44.
10. Driskell J.A. (1978). Vitamin B-6 Status of the
Elderly. In: Human Vitamin B-6 Requirements. Natl
Acad Sci., Washington DC. 252-256.
92


61
total urinary 4PA are reported as least squares mean the
pooled standard error of the least squares mean (SEM).
Orthogonal contrasts were made to examine the linear and
quadratic effects of different doses of PN-glucoside on
dependent variables. The quadratic effects were not
significant for any dependent variable. Values with p<0.05,
unless otherwise stated, were considered statistically
significant and were all for linear effects of the amount of
PN-glucoside (123).
Results
Vitamin B-6 nutritional status of subjects
The principal criteria used for assessing the vitamin B-6
status of the subjects were plasma PLP, urinary 4PA excretion
per 24-h, and erythrocyte aspartate aminotransferase
stimulation by in vitro addition of PLP. As shown in Table
4.1, mean values for these subjects were within the proposed
guidelines for vitamin B-6 nutritional adequacy (137-138).
These data indicate the adequate vitamin B-6 status of these
subjects prior to the administration of labeled vitamin B-6 in
each trial, which reflects the adequacy of their self-selected
diets prior to and between experiments. Subjects demonstrated
no statistical significant changes between trials for these
vitamin B-6 nutritional status parameters.


72
a purity of greater than 98%, as determined by ion-pair
reverse phase HPLC. The pyridoxamine-phosphate-HCl and
pyridoxal-51-phosphate were obtained from (Sigma Chemical Co.,
St. Louis, MO).
Pvridoxal kinase
Extraction. Pyridoxal kinase was extracted from the
liver tissue of six rats using the procedure of Merrill et al.
(112). The tissue was stored at -20 C and processed within
72 h following removal from the rat. To each 1 g of liver, 4
ml of 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
(Hepes) (Sigma Chemical, Co., St. Louis, MO) at pH 7.4 and 25
C, was added. The mixture was homogenized 3 0-45 s with a
Polytron PT 20 (Brinkman Instruments, Westbury, NY).> The
resulting homogenates were centrifuged (Model L, type 40
rotor, Beckman Instruments, Pala Alto, CA) at 105,000 x g for
1 h at 4 C. The supernatant was collected and stored at -20
C.
Purification. The extracted PL kinase was purified
using the column chromatography method employed by Cash et al.
(145). Preparation of the 4-pyridoxyl-Sepharose as an
affinity support was adapted from the method used by Kwok and
Churchich (146). Seven grams of aminohexyl-Sepharose
(Sepharose AH, Pharmacia, Ltd. Piscataway, NJ) was pretreated
and washed with water. The Sepharose AH slurry was added to
a solution of 50 ml water containing 1 g PL-HC1 (Sigma
Chemical Co., St. Louis, MO) adjusted to pH 7.0 by addition of


20
responsible for the differences in apparent bioavailability
of vitamin B-6 are not clear.
Human subjects have also been used in researching the
bioavailability of vitamin B-6 in specific foods and mixed
diets. Bioassays with humans indicated that the overall
bioavailability of vitamin B-6 in a typical American diet is
reasonably high although incomplete. A study evaluating free-
living vegetarian and non-vegetarian women who consumed
quantities of total vitamin B-6 equivalent to those in a mixed
diet found no significant differences between groups with
respect to vitamin B-6 status when comparing plasma PLP levels
(53). This suggests that the bioavailability of vitamin B-6
in vegetarian diets was similar to that in a mixed diet.
Andn et al. (84) reported that a mean of 2.5% of the total
vitamin B-6 present in breast milk of omnivorous lactating
women was in the form of PN-glucoside. Although a mean of 15%
of the total vitamin B-6 in the diet was present as PN-
glucoside, there appeared to be no correlation between the
amount of glycosylated B-6 in the diet and the percent PN-
glucoside in breast milk over the range examined. In
constrast, a study by Reynolds et al. (85) reported that the
percentage of dietary vitamin B-6 present as PN-glucoside was
equivalent to the amount of PN-glucoside found in the breast
milk of lactating Nepalese vegetarian women. Gregory and Ink
(75) did not observe any PN-glucoside in the breast milk of
six lactating American females including three


ACKNOWLEDGEMENTS
I thank Drs. Lynn Bailey, Peggy Borum, Robert Cousins and
James Cerda for allowing me the previlage of studying with the
best scientists in their respective fields.
Special thanks go to my advisor and research mentor, Dr.
Jesse Gregory, for his example of integrity and dedication in
all endeavors.
I also thank Doris Sartain and my collegues in the
"Yellow Lab" for their endless patience and sensitivity.
iii


75
Pyridoxamine (pyridoxine) 51 phosphate oxidase
Extraction. The rat liver PMP (PNP) oxidase was
extracted by the method of Kazarinoff and McCormick (119).
Liver tissue was homogenized in 0.02M potassium phosphate
buffer containing O.lmM mercaptoethanol, pH 7.0 and a ratio of
50 g tissue to 2 00 ml buffer. The homogenate was centrifuged
at 18,000 x g for 30 min and the supernatant adjusted to pH
5.0 with 2N acetic acid while stirring constantly. After 10
min the precipitate was removed by centrifugation at 18,000 x
g for 15 min. Potassium chloride (3.6 g) was added to the
supernatant and the volume brought up to 400 ml with water.
The pH was readjusted to 5.0 with 2N acetic acid.
Purification. The PMP (PNP) oxidase was extracted from
6 rat livers using the method of Kazarinoff and McCormick
(119). The clear supernatant resulting from the extraction
procedure was heated to 50 C and held for 8 min then plunged
into ice and cooled to 20 C and centrifuged at 18,000 x g for
30 min. The supernatant was applied to a DEAE-A50 column
(Sigma Chemical Co., St. Louis, MO) which had been prepared
by swelling overnight in 0.1 M potassium phosphate (pH 8.0).
A linear gradient established between 0.1 and 0.2 M potassium
phosphate at pH 8.0 was used to elute the column. Fractions
were collected and read at A28Q. The active fractions were
pooled. Solid ammonium sulfate (Sigma Chemical Co., St.
Louis, MO) (22.8 g/100 ml, 30% saturation) was added. After
30 min at 4C the precipitate was removed by centrifugation at


103
124. Serebro H.A., Solomon H.M., Johnson J.H., Hendrix, T.R.
(1966) The Intestinal Absorption of Vitamin B-6
Compounds by the Rat and Hamster. Bull Johns Hopkins
Hosp 119: 166-171.
125. Lumeng L., Li T.K. (1974) Vitamin B-6 Metabolism in
Chronic Alcohol Abuse: Pyridoxal Phosphate Levels in
Plasma and the Effects of Acetaldehyde on Pyridoxal
Phosphate Synthesis and Degradation in Human
Erythrocytes. J Clin Invest 53: 693-704.
126. Lumeng L., Li T.K., Lui A. (1985) The Interorgan
Transport and Metabolism of Vitamin B-6. In: Vitamin
B-6, Its role in Health and Disease. (Reynolds R.D.,
Leklem J.E., eds.) Alan R. Liss Inc., New York, on
Dietary Allowances, Food and Nutrition Board, National
Research Council. Natl Acad Sci, Washington, D.C. 35-
54.
127. Johansson S., Lindstedt S., Tiselius H.G. (1968)
Metabolism of [3H8]pyridoxine in Mice. Biochemistry
7: 2327-2332.
128. Contractor S.F., Shane B. (1971) Metabolism of [14C]
Pyridoxol in the Pregnant Rat. Biochim Biophys Acta
230: 127-136.
129. Tryfiates G.P., Saus F.L. (1976) Metabolism of
Pyridoxine in the Liver of Vitamin B-6-Deficient Rats.
Biochim Biophys Acta 451: 333-341.
130. Bain J.A., Williams H.L. (1960) Concentrations of B-6
Vitamers in Tissues and Tissue Fluids. In: Inhibition
in the Nervous System and Gamma-aminobutyric Acid,
(Roberts E., Baxter C.F., Harreveld A.V., Wiersma
C.A.G., Adey W.R., Killiam K.F., eds.) Pergamon
Press, New York.
131. Lyon J.B., Porter J. (1960) The Distribution of
Vitamin B-6 in the Tissues of two inbred Strains of
Mice fed Complete and Vitamin B-6-Deficient Rations. J
Biol Chem 237: 1989-1991.
132. Lumeng L., Li T.K. (1978) Validation of the Diagnostic
Value of Plasma Pyridoxal 5'phosphate Measurements in
Vitamin B-6 Nutrition of the Rat. J Nutr 108: 545-
553.


65
Table 4.4 Total urinary 4PA excretion.1
EXPERIMENTAL TRIALS
TIME
(H)
TRIAL 1
(5/mol d2PN
+ 2/mol PN)
(n=6)
TRIAL 2
(5/mol d2PN +
1/zmol PN + 1/tmol
PN-GLUCOSIDE)
(n=6)
TRIAL 3
(5/mol d2PN
2/mol PN-
GLUCOSIDE)
(n=6)
(/mol)
(/mol)
(/mol)
(SEM)
0-24
26.0
13.5
10.3
3.9
24-48
15.5
10.5
11.5
4.82
0-48
41.5
24.0
21.8
7.63
Values based on n=6 are means and pooled standard error of
the mean (SEM) The quadratic effects were not
significant for any dependent variables. The values in
table are significantly different at p<0.05 for linear
effects of the amount of PN-glucoside.
2 p<0.5
3 p<0.08
trials 1 and 3 to decrease linearly with the amount of PN-
glucoside (Table 4.4). These results indicate that the
percentage of d2PN dose excreted as d24PA showed an inverse
relationship in proportion to the PN-glucoside dose and that
total 4PA decreased linearly with the amount of PN-glucoside.
Discussion
Because PN-glucoside is a major form of vitamin B-6 in
plant derived foods, its influence on the utilization of


19
than that of purified labeled PN-glucoside, suggesting an
inhibitory effect of plant tissues.
Stable-isotopic techniques and methodologies evaluate the
bioavailability of purified deuterium-labeled (d2) PN-
glucoside in human subjects (82) In a study that used stable
isotopic techniques to analyze the bioavailability of orally
administered extrinsically enriched oatmeal containing labeled
PN-glucoside, it was reported that the labeled PN-glucoside
was approximately 58% of that of labeled-PN fed under
identical conditions (82) Although the ability metabolically
to utilize PN-glucoside appears to have a substantial level of
individual variability among human subjects, the
bioavailability of PN-glucoside in human subjects has been
observed to be substantially greater than that of the rat
(82). Further development of the methodology and application
of stable-isotopic techniques will continue to provide greater
clarification of the factors affecting bioavailability of
vitamin B-6.
The above section is a review of the current knowledge
concerning the bioavailability of the vitamin B-6 in foods.
There appears to be a general agreement between the rat
bioassay and the Saccharomvces uvarum assay method for many
foods, although the range of apparent bioavailability relative
to PN was quite large for the rat. These comparative data
provided the first indication that the bioavailability in
certain plant foods may be less than complete. The factors


63
Table 4.2 Molar isotopic ratio of urinary d24PA.1
EXPERIMENTAL TRIALS
SUBJECT
TRIAL 1
(5/mol d2PN +
2/xmol PN)
(n=6)
TRIAL 2
(5/mol d2PN +
1/mol PN + 1/mol
PN-GLUCOSIDE)
(n=6)
TRIAL 3
(5/mol d2PN
2/mol PN-
GLUCOSIDE)
(n=6)
1
0.2 0.001
d2/d0
0-24 hr
0.091.002
0.071.010
2
0.07.001
0.101.001
0.121.003
3
0.07.003
0.081.003
0.091.001
4
0.17.002
0.151.010
0.101.001
5
0.131.001
0.011.001
0.071.001
6
0.101.001
0.081.001
0.121.002
mean
0.121.021
0.091.018
0.091.009
1
0.051.000
d2/d0
24-48 hr
0.101.001
0.061.000
2
0.091.001
0.061.000
0.031.000
3
0.101.000
0.101.000
0.031.000
4
0.071.001
0.101.000
0.041.001
5
0.091.000
0.101.000
0.061.000
6
0.051.000
0.191.001
0.071.000
mean
0.071.007
0.111.017
0.051.006
1Mean and SEM, n=6.
Values are average of three injections per sample.
evidence of the utilization of d2PN in vitamin B-6 metabolism.
Deuterium-labeled 4PA was detected in urine over 48-h post
dose. The d24PA comprised a small portion of the total
urinary 4PA using this experimental design. However, the
isotope ratio observed was measurable by the GCMS method
employed. Twenty-four hours following the administration of


CHAPTER 5
EFFECTS OF PYRIDOXINE-5' -/3-D-GLUCOSIDE ON THE
IN VITRO KINETICS OF PYRIDOXAL KINASE AND
PYRIDOXAMINE (PYRIDOXINE)-5'PHOSPHATE OXIDASE
IN RAT LIVER
Introduction
There are a number of enzymatic reactions in the metabolic
pathway responsible for the interconversion of vitamin B-6
compounds. In particular, PL kinase and PMP (PNP) oxidase are
considered the two most important catalytic enzymes in the
formation of the coenzymatic form of vitamin B-6, PLP (144).
Although other tissues contribute to vitamin B-6 metabolism,
the liver is thought to serve as the center of interconversion
of B-6 vitamers to PLP (109-110).
The B-6 vitamers are postabsorptively taken-up by the
liver and converted to PLP. The non-phosphorylated vitamers
are converted to their 5'-phosphorylated forms by a single
kinase (144). Following phosphorylation, PMP and PNP are
oxidized by PMP (PNP) oxidase to form PLP. This active
coenzymatic form of vitamin B-6 can then either be bound by
cellular apoproteins, released into circulation as PLP, or
dephosphorylated by alkaline phosphatase to PL (144) The PL
69


99
78. Bills N.D., Leklem J.E., Miller L.T. (1987). Vitamin
B-6 Bioavailability in Plant Foods is Inversily
Correlated with Percent Glycosylated Vitamin B-6. Fed
Proc 46: 1487.
79. Tsuji H., Okada J., Iwami K., Yasumoto K., Mitsuda H.
(1977). Availability of Vitamin B-6 and Small
Intestinal Absorption of Pyridoxine-Beta-D-Glucoside in
Rats. Vitamins 51: 153-159.
80. Trumbo P.R., Gregory J.F., Sartain D.B. (1988).
Incomplete Utilization of Pyridoxine-beta-Glucoside as
Vitamin B-6 in the Rat. J Nutr 118: 170-175.
81. Trumbo P.R., Gregory J.F. (1988). Metabolic
Utilization of Pyridoxine-beta-Glucoside in Rats:
Influence of Vitamin B-6 Status and Route of
Administration. J Nutr 118: 1336-1342.
82. Gregory J.F., Trumbo P.R, Bailey L.B, Toth J.P,
Baumgartner T.G., Cerda J.J. (1991). Bioavailability
of Pyridoxine-51 -/3-D-Glucoside Determined in Humans by
Stable-Isotopic Methods. J Nutr 121: 177-186.
83. Register U.D., Lewis U.J., Ruegamer W.R., Elvehuem C.A.
(1950). Studies on the Nutritional Adequacy of Army
Combt Rations. J Nutr 40: 281-294.
84. Andn M.B., Reynolds R.D., Moser-Veillon P.B., and
Howard M.P. (1989). Dietary Intake of Total and
Glycosylated Vitamin B-6 and the Vitamin B-6
Nutritional Status of Unsupplimented Lactating Women
and Their Infants. Am J Clin Nutr 50: 1050-1058.
85. Reynolds R.D. (1988). Bioavailability of Vitamin B-6
from Plant Foods. Am J Clin Nutr 48: 863-867.
86. Harding R.S., Plough I.C., Friedemann T.E. (1959).
The Effect of Storage on the Vitamin B-6 Content of a
Packaged Army Ration, with a Note on the Human
Requirement for the Vitamin. J Nutr 68: 328-331.
87. Sauberlich H.E. (1961). Studies on the Toxicity and
Antagonistism of Amino Acid for Weaning Rats. J Nutr
74: 289-297.
88. Davies M.K., Gregory M.E., Henry K.G. (1959). The
Effect of Heat on the Vitamin B-6 of Milk. J Dairy Res
26: 215-220.


58
approved by the University of Florida Institutional Review
Board and an informed consent was obtained from each subject.
Three trials were conducted to compare the in vivo
metabolic utilization of d2PN using the same group of six
subjects. The experimental period lasted seven weeks
consisting of three trials, each lasting one week with a two
week wash out period between each trial. A self-selected diet
was consumed for the duration of the study.
On the first day of each trial the subjects collected a
24-h urine sample into foil-covered polyethylene containers
and kept it refrigerated during the collection period. After
an overnight fast, a five ml blood sample was collected into
a Vacutainer brand tube (Becton Dickinson, Rutherford, NJ)
containing ethylenediaminetetraacetic acid (EDTA). Following
these sample collections the subjects consumed a single oral
dose of 5 /mol d2PN and either 2/mol PN or 1/mol PN + 1/mol
PN-glucoside or 2/mol PN-glucoside (trials 1, 2, and 3
respectively) (Table 4.2) contained in 250 ml of apple juice.
The doses of unlabeled PN and PN-glucoside were selected to
provide and equal molar amount (2/mol) of either PN, PN+PN-
glucoside, or PN-glucoside. Urine collection was continued
for the following 48-h period. Urine was analyzed for
creatinine, total 4PA, and d24PA. The blood samples were
centrifuged and the plasma collected was analyzed for plasma
PLP and the erythrocytes were used to determine hemoglobin
concentration and AspAT activity.


28
effects of e-pyridoxyllysine generally decreases the apparent
bioavailability of vitamin B-6 by competitive inhibition of PL
kinase (103) It is reasonable to surmise from the above
studies that a similar antagonistic effect of e-
pyridoxyllysine may have been responsible for the severe
deficiencies observed in infants with compromised vitamin B-6
nutriture and fed diets composed entirely of non fortified
canned infant formulas (60). The mechanisms of action for an
antimetabolite, such as e-pyridoxyllysine, includes inhibition
of any of the enzymes in the metabolic pathway to interconvert
the B-6 vitamers. Other mechanisms of vitamin B-6
antimetabolites are structural analogues of the vitamin and
inhibition of the active coenzyme function of PLP (96).
The complex chemical identity and nutritional properties
of other B-6 vitamer derivatives found in foods have not been
fully determined. The formation of the degradation product,
6-hydroxy-pyridine from the relatively stable B-6 vitamer PN,
has been reported in thermal treatment of various fruits and
vegetables (104) The generation of hydroxyl radicals
apparently mediated the hydroxylation of PN which was
associated with the oxidative degradation of ascorbic acid
(105). Gregory and Leatham (106) reported that 6-hydroxy-PN
has no vitamin B-6 or antivitamin B-6 activity. It appears
that hydroxylation of PN at the 6 position is not a
significant mechanism responsible for the loss of vitamin B-6
activity in foods.


9
results for a variety of plant and animal products. Liver
fractions, whole wheat and yellow corn exhibited
bioavailability values of between 65% and 70% compared with an
apparent value of approximately 100%. These results were the
first indication of a wide variation and potentially
incomplete bioavailability of vitamin B-6 in cereal grains and
other food products (34).
A comparison of rat growth bioassays and Saccharomvces
uvarum data for dried beef, lima beans, non-fat dry milk and
whole wheat flour yielded a good correlation between the
assays with small differences obtained for the flour and milk
products (35). Nelson et al. (38) examined the rate of
intestinal absorption of the vitamin B-6 in orange juice in
human subjects. A triple lumen perfusion technique was
employed to determine the relative absorption of naturally
occurring vitamin B-6 in orange juice. Their results
demonstrated a significant decrease in rate of absorption of
vitamin B-6 in orange juice (42%) than that of synthetic B-6
vitamers in saline (67%) and saline-glucose solutions (79%).
The lower rate of absorption of the naturally occurring
vitamin B-6 in orange juice was thought to be a result of
interactions of the vitamin with naturally occurring food
components in the orange juice (38). A follow-up study
conducted by Nelson et al. (39) investigated the nature and
extent of binding of different forms of vitamin B-6 in orange
juice. The study suggested extensive and equal binding of


48
PNP, and pyridoxamine 51-phosphate (PMP) in liver tissue,
while hepatic 14C decreased linearly (p<0.05) (Table 3.3). The
proportion of hepatic nonphosphorylated forms of [14C]PN, PL
and PM was significantly less among treatment group 2
(p<0.05) than groups 1 and 3. Urinary excretion of total 14C
also was significantly greater (p<0.05) in group two relative
to groups one and three (Table 3.4). In addition, the
concentration of urinary 4PA, relative to total urinary 14C,
was significantly lower (p<0.05) in group two as compared to
groups 1 and 3 (Table 3.4). The results of experiment 2
indicate that the effects of PN-glucoside observed in
experiment 1 were specific for this vitamin B-6 derivative and
were not due to isotope dilution effects of the unlabeled
compound administered.
Discussion
The focus of this research was to determine the direct
effects of PN-glucoside on [14C]PN administered in a
simultaneous dose, similar to what would occur when consuming
a mixed diet. The ratios of PN-glucoside to nonglycosylated
vitamin B-6 (14C-labeled plus dietary) were in the range of 8%
to 16%, which is consistent with observations of typical human
diets (82) Since our interest was in determining the fate of
labeled-vitamin B-6 compounds in the presence of PN-glucoside
the data were expressed as percent of the total radioactivity
in the dose. This enabled the determination of the direct


3
establishing an accurate RDA for vitamin B-6. Establishing an
adequate RDA for vitamin B-6 is a complex task. The
appropriateness of the RDA for vitamin B-6 is dependent on
accurate information on the content of the vitamin in foods
consumed, the enzymes involved in the metabolism of the
various forms of vitamin B-6, as well as information regarding
vitamin B-6 bioavailability.


60
acetyl-4PA-lactone was dissolved in 50 /I ethyl acetate and 2
IjlI of the resulting solution injected into the GCMS injection
port. As this 4PA derivative eluted from the GC column,
electron capture negative ionization yielded two anions at m/z
207 and 209 for the dQ and d2 4 PA species, respectively.
Selected-ion monitoring at m/z 207 and 209 permitted
simultaneous measurement of the dQ and d2 4PA species during
GCMS analysis. A series of 4PA standards of known molar
ratios of dQ and d2 species were prepared and analyzed to
facilitate quantitative analysis. Calibration curves were
constructed to relate the ratios of observed GCMS peak areas
to actual molar isotope ratios of the d2/dQ 4PA species. The
urinary excretion of d24PA was calculated using the isotope
ratio determined by GCMS, the concentration of total urinary
4PA as determined by reverse-phase HPLC, and total urine
volume.
Statistical analysis
The urinary excretion of d24PA was expressed as a
percentage of the administered dose of labeled PN. The
excretion of d24PA (as a percentage of administered dose),
total urinary 4PA, and vitamin B-6 nutritional status
parameters (see below) was compared by the method of least
squares analysis of variance using general linear model
procedures of SAS (123), following a log transformation of the
data to normalize variance when necessary. Data for d24PA and


14
of PN-HC1, is added to many breakfast cereals at levels of
25% to 100% of the United States Recommended Daily Allowance
(USRDA) per ounce. Gregory et al. (61-62), using a dehydrated
model food system, simulating breakfast cereals, indicated
that the roasting and storage of low moisture food systems
resulted in losses of 50% to 70% of the added PN,PM,and PLP.
The remaining vitamin B-6 was found to be fully available as
determined by rat bioassay using growth, feed efficiency,
erythrocyte aspartate aminotranferase activity (AspAT) and in
vitro (AspAT) coenzyme stimulation. Gregory (63) also
examined vitamin B-6 bioavailability in rice-based, PN-
fortified cereal and non-fat dry milk. The non-fat milk and
rice base breakfast cereal samples were analyzed for vitamin
B-6 by microbiological, HPLC, and rat bioassay procedures.
The results indicated that the vitamin B-6 of the non-fat dry
milk was fully available, while the vitamin B-6 availability
in the cereal product was comparatively low. The apparent
losses of the PN in the fortified cereal was explained by a
first order kinetic model described by Evans et al. (64).
These studies (61,63-64) indicated that food fortified with PN
is susceptible to significant degradation under certain
processing and storage conditions and that the bioavailability
of the remaining vitamin B-6 may not be complete. The above
results suggest that thermal processing of foods does not
induce nutritionally important losses in the bioavailability
of vitamin B-6. At the very least, research to date indicates


10
both PL and PN in orange juice to a small dialyzable molecule
which is heat stable and non-protein in nature. The
interaction of vitamin B-6 with this component of orange juice
was suggested as being responsible for the lower rate of
absorption of naturally occurring vitamin B-6 in orange juice.
Leklem and coworkers (40) were the first to employ a
bioassay with human subjects to examine the effect of dietary
fiber on the bioavailability of the vitamin. The study
investigated the bioavailability of vitamin B-6 in white and
whole wheat bread. Significant differences in the
bioavailability of B-6 vitamers were not apparent when plasma
PLP and erythrocyte aminotransferase data were examined.
However, fecal vitamin B-6 and urinary 4PA excretion data
suggested incomplete utilization of the vitamin in the whole
wheat bread. Frequent attempts have been made to calculate
the vitamin B-6 balance of B-6 vitamers and 4PA in bioassays
employing human subjects (41).
Similar experiments in which the use of data concerning
fecal vitamin B-6 have shown little validity. Unabsorbed
vitamin B-6 from dietary sources as well as microbiologically
synthesized B-6 vitamers are contained in fecal material. The
composition of the diet will influence the microbial
contribution to fecal vitamin B-6 which may affect the
apparent vitamin B-6 nutriture of the subject. The fact that
the intestinal microorganisms produce vitamin B-6 is well
established (41-46). However, the availability of vitamin B-6


22
growth. Harding et al. (86) fed human subjects canned rations
which had been stored for twenty months at 100 F. They
observed marginal vitamin B-6 deficiency in the subjects even
though the rations provided apparently adequate amounts of the
vitamin. Inadequate vitamin B-6 nutriture was most likely a
result of an effect of thermal processing. It is reasonable to
consider that marginal limitations in essential amino acid
content of the ration may have contributed to nutrient
insufficiency. The addition of vitamin B-6 to diets deficient
in certain essential amino acids has been shown to produce a
growth response (87). However, the reason for this apparent
incomplete bioavailability of vitamin B-6 in these canned
rations is unclear. Several infants who consumed a non-
fortified, heat sterilized canned infant formula were found to
be severely vitamin B-6 deficient and suffer with
neurochemically induced convulsions. Extensive research was
subsequently initiated to study the chemical behavior of
vitamin B-6 in milk products (58).
The differing stability properties of the various vitamin
B-6 compounds have been of interest since Hassinen et al. (59)
observed that PM and PL, the naturally occurring forms of
vitamin B-6 in milk, were much less stable than added PN.
Employing microbiological assays, Hassinen (59) demonstrated
that PM and PL added to milk were degraded at the same rate as
naturally occurring B-6 vitamers. Added PN exhibited greater
stability than either PM or PL during thermal processing of


73
NaOH. The slurry was shaken 16 h in the dark at ambient
temperature and then placed in an ice bath with the dropwise
addition of ice cold borohydride until disappearance of the
yellow color. During this period the slurry was shaken and 7%
acetic acid was added to prevent the pH from rising above 9.0.
The slurry was adjusted to pH 6.0 with 7% acetic acid and
packed into a 1.6 x 15 cm chromatography column. The column
was rinsed with 3M KC1 followed by water prior to use. All
rinsing and eluting buffers contained 2mM potassium phophate
(Fisher Scientific, Fairlawn, NJ) at pH 7.0 and lOmM
glutathione.
The 105,000 x g supernatant was applied to the PL-
Sepharose column and washed through with 50 ml of 100 mM KC1
buffer. The column was then rinsed with 500 ml 400mM KC1
followed by 50 ml of lOOmM KCl buffer. The PL kinase was then
eluted and adsorbed directly onto a Bio-Gel HTP
hydroxylapatite column (Bio Rad Laboratories, Richmond, CA),
previously equilibrated with buffer containing no KCl, with
500 ml 10 mM PN adjusted to pH 7.0 in buffer and rinsed with
25 ml lOOmM KCl buffer. The hydroxyapatite column was eluted
with 100 ml of a linear gradient of 2-300mM potassium
phosphate at pH 7.0. Approximately 3 ml fractions were
collected and the active fractions pooled. The enzyme
preparation was then tested for glucosidase activity using the
method of Daniels et al. (147) and for phosphatase activity
employing the PL kinase assay (discussed below) using PLP as


27
bioavailability of these B-6 vitamers. Gregory et al. (100-
102) conducted a series of experiments to examine further the
possible influence of thermal processing and storage on the
net bioavailability of vitamin B-6 in foods. When milk
containing radiolabeled PL or PLP was subjected to heat
sterilization, analysis using HPLC revealed no formation of
bis-4-pyridoxal disulfide (103). The losses of the B-6
vitamers PL and PLP were due to the reductive binding of the
aldehydes of these vitamers to food protein as e-
pyridoxyllysyl residues (103). The formation of e-
pyridoxyllysine has been identified as a mechanism of the loss
of vitamin B-6 during thermal processing or low moisture
storage of proteinaceous model food systems (100-101) and
various meat and dairy food products (16). Similar results
were reported in intrinsically enriched chicken liver and
muscle tissues (103). The phosphopyridoxyllysyl complex
accelerated the onset and enhanced the severity of vitamin B-6
deficiency symptoms. The effect of e-pyridoxyllysine on the
bioavailability of vitamin B-6 in foods is a function of the
ratio of
total vitamin
B-6
content of
the diet and
the
addition
product to
the
other B-6
vitamers.
The
bioavailability of the vitamin of diets which contain e-
pyridoxyllysine but which are adequate in vitamin B-6 content
would most likely be high (102). Whether naturally occurring
amounts of vitamin B-6 are low or conditions present during
food processing produce e-pyridoxyllysine, the antimetabolite


76
18,000 x g for 30 min and addition of ammonium sulfate (15.2
g/100 ml, 50% saturation). After 60 min the solution was
centrifuged and the active precipitate was dissolved in one
tenth the initial volume of water. This solution was dialyzed
against 0.02 M potassium phosphate buffer for 24 h and
centrifuged to remove the precipitate. The clear yellow
supernatant was applied to a column of Sephadex G-100 (Sigma
Chemical Co., St. Louis, MO) equilibrated with 0.02 M
phosphate buffer and eluted with the same buffer. Fractions
were collected and read at an A280 and the active fractions
pooled.
Oxidase assay. PMP (PNP) oxidase activity was measured
by the method of Wada and Snell (31) using phenylhydrazine as
the color development agent. The assay mixture contained 1 ml
of ImM PMP solution in 0.2M Tris-HCl, pH 8.0, 10-200 units of
enzyme (one unit of oxidase enzyme catalyzes the formation of
lnmol PLP/mL/min at 37 C), and enough Tris buffer to bring
the total volume to 3.5 ml. The mixture was gently shaken for
30 min at 37 C. The reaction was stopped by the addition of
0.3 mL of 100% (w/v) trichloroacetic acid. There was no
formation of a precipitate therefore, the entire supernatant
was used for the color reaction. The 3 mL supernatant was
placed in a clean test tube, 0.2 mL phenylhydazine reagent (2
g dissolved in 100 mL 10N H2S04) was added and the mixture was
heated for 20 min at 60 C. The reaction mixture was then
cooled to room temperature and the absorbance read at 410 nm


21
lactoovovegetarians. Differences in the vitamin B-6
composition of the breast milk reported in these studies are
unclear. The factors responsible for the incomplete
bioavailability of vitamin B-6 have not been determined.
Factors Affecting the Bioavailabilitv of Vitamin B-6
Several factors are known to affect the bioavailability
of vitamin B-6 in food products. These include the formation
of certain reaction products during food processing and
storage, fiber type and the quantity present in the food
source, and the presence of PN-glucoside in plant foods. The
results of previously discussed studies suggested that thermal
processing and storage of foods may adversely affect
bioavailability of vitamin B-6 through the formation of
reaction products with amino acid residues of the
proteinatious portion of the food (36,56-57).
The bioavailability of vitamin B-6 from animal products
approaches 100% for most foods. The biological activity of
vitamin B-6 from plant derived products is generally lower.
A series of studies concerning the nutritional quality of
military rations provided the initial impetus to consider the
adverse effect of thermal preservation of canned foods.
Register et al. (83) reported that rats which subsisted
wholely on either of two homogenized combat rations required
supplemental vitamin B-6 to sustain normal growth even though
the diets contained enough vitamin B-6 for adequate rat


56
the limiting step in the utilization of PN-glucoside as
vitamin B-6 in the rat (2,77,80). A recent study in our
laboratory has shown that PN-glucoside alters the metabolism
and in vivo retention of simultaneously administered PN in the
rat (136).
A stable-isotopic study in humans found the
bioavailability of PN-glucoside to be 58% of that of free PN,
which is greater than that found in the rat (82) In an
investigation of lactating women, Andn et al. (84) reported
that dietary PN-glucoside contributed little to vitamin B-6
nutriture. These findings indicate the need for additional
data concerning factors affecting the utilization of PN-
glucoside and the potential effect of PN-glucoside on the
metabolic utilization of vitamin B-6 by humans consuming a
mixed diet. The study reported here was conducted to
determine the effect of PN-glucoside on the metabolic
utilization of PN in humans through the use of stable-isotopic
methods.
Materials and Methods
Synthesis of forms of vitamin B-6
The deuterium-labeled form of vitamin B-6 used in this
study was [51-C2H2OH]pyridoxine (d2PN) which was prepared in
our laboratory as described by Coburn et al. (108). Mass
spectral analysis indicated that the d2PN species comprised
66% of the d2PN preparation, with a majority of the remainder


39
bioavailability of orally administered PN-glucoside is
approximately 58% relative to PN. While the incomplete
utilization of PN-glucoside has been clearly established in
rats and humans, its potential interactive effects in vitamin
B-6 metabolism have not been examined previously.
Materials and Methods
Protocol
Two studies were conducted to evaluate the in vivo
utilization of [UC]PN in the presence of PN-glucoside. In
each of the studies eighteen male Sprague-Dawley rats (Crl:CD
(SD) BR) from Charles River Breeding Laboratories (Wilmington,
MA) weighing 200-300 g, were housed individually in stainless
steel metabolism cages in animal quarters maintained at 241C
with a 12-h light/dark cycle. All procedures for the care and
treatment of the experimental animals were in accordance with
the National Institutes of Health Guidelines.
The rats were fed ad libitum a casein-sucrose (20%:60%)
based diet, adequate in all micronutrients including 7 mg PN
HCl/kg*d for eight days (80) At the end of a seven day
acclimation period the rats were randomly assigned to one of
three treatment groups. On day eight of studies 1 and 2,
following a twelve hour fast, the rats were administered a
dose of 166.5 MBq (4.5 nCi) [UC]PN, equivalent to 240 nmol of
PN. In study 1 the rats received a simultaneous dose of
either 0, 36, or 72 nmol of unlabeled PN-glucoside (treatment


57
in the monodeutero (d^ form. All procedures were performed
under yellow-light to prevent photochemical degradation of B-6
vitamers.
PN-glucoside was prepared from PN-HCl using biological
synthesis as previously described (80). Twelve grams of
alfalfa seeds were germinated in the presence of 62 mg of PN-
HCl (297 /mol) obtained commercially (Sigma Chemical Company,
St. Louis, MO) followed by purification by cation-exchange
(80) and gel filtration chromatography on Sephadex G-10
(Pharmacia). The yield of purified PN-glucoside was 237 /mol
(80%). Chromatographic analysis of the purified PN-glucoside
preparation confirmed the absence of free (nonglycosylated) PN
and other forms of vitamin B-6.
Protocols of trials with human subjects
The study was conducted using the same adult men (n=6;
22-32 y) in each of three trials (Table 4.1). All subjects
were in good health and exhibited normal blood chemistry and
hematological values. The subjects were in normal vitamin B-6
status as determined by plasma pyridoxal phosphate (PLP) and
urinary 4-pyridoxic acid (4PA) concentration, erythrocyte
aspartate aminotransferase (AspAT) activity, and erythrocyte
AspAT stimulation by in vitro addition of PLP, as judged by
previously published criteria (137-138). The procedures for
selection of subjects and the experimental protocol were


This dissertation was submitted to the Graduate Faculty
of the College of Agriculture and to the Graduate School and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy. I *
of tjfr
December 1991
/vt'>
Dean, College of
Agriculture
Dean, Graduate School


3.00 j
2.75--
2.50 --
cn
£
2.25--
c
2.00-
E
\
1.75--
a.
_j
1.50--
CL
1.25-
O
£
1.00-
0.75 -
0.50
>
0.25 -
0.00-
O uM PN-glucoside
50 uM PN-glucoside
V 100 uM PNglucsido
150 uM PNglucsido
+
4-
+
+
+
14 28 42 56 70 84 98 112
-1 -1
(PYRID0XAMINE-PH0SPHATE) uM
126 140
Figure 5.2 Double reciprocal plot for the conversion of PMP to PLP by PMP (PNP) oxidase
from rat liver in the presence of various levels of PN-glucoside. n>


106
Joyce Ann Gilbert was born in Washington, D.C. and
raised in Woodbridge, VA. Upon graduation from high school,
Joyce attended the University of South Carolina where she
received a Bachelor of Science degree in biology. Following
a brief respite from academia, Joyce returned to Clemson
University where she received a Master in Nutritional
Science and became a registered dietitian. After working in
clinical and research dietetics as well as in food
management, Joyce accepted an instructor position in the
Department of Food Science and Human Nutrition at the
University of Florida. Joyce relinquished her faculty
position to pursue a Ph.D. in human nutrition. Joyce has
accepted an assistant professor position at the Pennsylvania
State University upon completion of her dissertation. When
not engaged in the sciences, Joyce enjoys athletics, both as
a participant and fan, art and riding her Harley Davidson.


67
oxidation of pyridoxal (PL) to 4PA (130-132). As a terminal
metabolite of vitamin B-6 metabolism, urinary 4PA reflects the
in vivo metabolic utilization of the vitamin. The deuterated
form of PN used in this study is labeled at a metabolically
stable position (i.e., 5'-CH2OH) and, therefore, retains the
label throughout absorption, metabolism, transport, and
excretion (82).
The results of this study demonstrate a negative
correlation between the quantity of PN-glucoside administered
and the amount of urinary d24PA derived from oral d2PN. The
data also showed that an inverse relationship between the
amount of total urinary 4PA excreted and the amount of PN-
glucoside in the dose. These findings are consistent with
those observed when comparing the metabolism of [UC]PN in the
presence of increasing doses of PN-glucoside in the rat (136).
The PN derived from hydrolysis of PN-glucoside enters into
vitamin B-6 metabolism to produce other forms of vitamin B-6,
including 4PA (77,81). However, the PN-glucoside that is not
hydrolyzed to PN may not be totally inert but instead appears
to affect the metabolism and interconversion of the B-6
vitamers. The possible inhibition by PN-glucoside of PL
kinase and/or PNP-oxidase, two enzymes in the metabolic
interconversion of B-6 vitamers, is discussed in chapter 5.
The results of these isotopic studies provide evidence that
PN-glucoside alters the metabolism of PN in humans and rats by


35
flavocoenzyme and demonstrated product inhibition as well as
inhibition by some phosphorylated analogs of vitamin B-6
compounds (31).
The relative reactivity of PMP(PNP) oxidase with varying
substrates is pH dependent (118). Within the physiological pH
range both PMP and PNP are suitable substrates for liver PMP
(PNP) oxidase. Although at slightly alkaline pH the V for
PNP is higher, the observed velocity is decreased due to
greater substrate inhibition by PNP even at low substrate
concentrations (112). The values demonstrate that at pH
7.0 PMP and PNP bind with similar affinity, whereas PMP binds
more tightly at the usual assay pH of 8.0 compared to PNP
(118) .
Many compounds have been indicated as inhibitors of the
metabolism of vitamin B-6. When assessing the site of action
of these antagonists it is reasonable to suspect that such
compounds may act on those enzymes responsible for the
formation of the active coenzyme form PLP. It is well
documented that liver PL kinase is inhibited by analogues
containing a 4-formyl group which resemble PL in their
affinity for PL kinase, condensation products formed from PL
and hydroxylamine, O-substituted hydroxylamines, hydrazine and
substituted hydrazines (120). PL kinase has also been shown
to be sensitive to inhibition by degradation products such as
pyridoxyllysine (121).


40
groups 1,2, and 3 respectively). In study 2 the rats received
a simultaneous dose of either 0 or 72 nmol unlabeled PN-
glucoside or 21.6 nmol unlabeled PN (treatment groups 1,2, and
3 respectively). The 21.6 nmol dose of PN was selected to
provide 30% of the PN-glucoside treatment, based on an assumed
30% net bioavailability of PN-glucoside relative to PN (4-6).
Doses were administered in 1.5 mL sterile H20 by gavage.
Urine was collected for the ensuing 24 h into foil-covered
flasks to avoid photochemical degradation of vitamin B-6
compounds. The urine collection funnels and flasks were
rinsed with water and rinses pooled with collected urine;
after dilution to 50 mL, urine was stored at -20 C until
analysis. The rats were killed by decapitation following
brief anesthesia, and livers were rapidly excised. The tissue
was divided into 2 g samples, then stored at -20 C until
analysis. All procedures were performed under GEF40G0 gold
lights that emit a wavelength between 500 nm and 750 nm to
prevent photochemical degradation of B-6 vitamers.
Forms of vitamin B-6
PN-glucoside was prepared by biological synthesis using
the propagation of alfalfa sprouts germinated in the presence
of PN-HC1 obtained commercially (Sigma Chemical Company, St.
Louis, MO) (80). The [4,5-uC]pyridoxine hydrochloride 684.5
KBq/nmol (18.5 Ci/mol) was a gift from Hoffmann-LaRoche


LIST OF FIGURES
Figure 5.1 Double reciprocal plot for the
phosphorylation of PN by PL kinase from rat liver
in the presence of various levels of PN-
glucoside 81
Figure 5.2 Double reciprocal plot for the conversion of
PMP to PLP by PMP (PNP) oxidase from rat liver in
the presence of various levels of PN-glucoside. 82
vii


97
56. Lindberg A.S., Leklem J.E., Miller L.T. (1983). The
Effect of Wheat Bran on the Bioavailability of Vitamin
B-6 in Young Men. J Nutr 113: 2578-2586.
57. Tarr J.B., Tamura T., Stokstad L.R. (1981).
Availability of Vitamin B-6 and Pantothenate in an
Average American Diet in Man. Am J Clin Nutr 34:
1328-1337.
58. Coursin D.B. (1954). Convulsive Seizures in Infants
with Pyridoxine Deficient Diet. J Am Med Assoc 154:
406-408.
59. Hassinen J.B., Durbin G.T., Bernhart F.W. (1954). The
Vitamin B-6 Content of Milk. J Nutr 53:249-257.
60. Tomarelli R.M., Spence E.R., Bernhart F.W. (1955).
Biological Availability of Vitamin B-6 in Heated Milk.
J Agrie Food Chem 3: 338-341.
61. Gregory J.F., Kirk J.R. (1978). Assessment of
Roasting Effects on Vitamin B-6 Stability and
Bioavailability in Dehydrated Food Systems. J Food Sci
42: 1585-1589.
62. Gregory J.F., Kirk J.R. (1978). Assessment of Storage
Effects on Vitamin B-6 Bioavailability in Dehydrated
Food Systems. J Food Sci 43: 1801-1808.
63. Gregory J.F. (1980). Bioavailability of Vitamin B-6
in Nonfat Dry Milk and a Fortified Rice Breakfast
Cereal Product. J Food Sci 45:84-86.
64. Evans S.R., Gregory J.F., Kirk J.R. (1981). Thermal
Degradation Kinetics of Pyridoxine Hydrochloride in
Dehydrated Model Food Systems. J Food Sci 46: 555-
558, 563.
65. Kabir H., Leklem J., Miller L.T. (1983). Measurement
of Glycosylated Vitamin B-6 in Foods. J Food Sci 48:
1422-1425.
66. Scudi J.V. (1942). Conjugated Pyridoxine in Rice Bran
Concentrates. J Biol Hem 145: 637-639.
67. Yasumoto K., Tsuji H., Iwami K., Mitsuda H. (1977).
Isolation from Rice Bran of a Bound Form of Vitamin B-6
and its Identification as 5'-0-(beta-D-
Glucopyranosyl)Pyridoxine. Agrie Biol Chem 41: 1061-
1067.


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Professor of Food Science
and Human Nutrition
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Lynri B. Bailey
Professor of Food Science
and Human Nutriton
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Peggy R. Borum
Professor of Food Science
and Human Nutrition
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Robert J. Cousins
Boston Family Professor of
Human Nutrition
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
a dissertation for the degree of Doctor of Philosophy.
Professor of Medicine
as


13
thermal processing on the relative bioavailability of vitamin
B-6 in beef, spinach, potato and cornmeal were assessed. The
results further demonstrated that diet composition and food
processing must be considered in evaluating the
bioavailability of vitamin B-6.
The instability of certain B-6 vitamers during food
processing and storage may contribute to losses of the
nutritional quality of foods with respect to vitamin B-6.
Processing and storage of foods have been found to have
various effects on vitamin B-6 with losses decreasing the
adequacy of food products as sources of dietary vitamin B-6.
A deficiency of vitamin B-6 in infants who consumed a non-
fortified, heat sterilized canned infant formula, led to
extensive research, during the 1950's, concerning the effects
of food processing on the bioavailability of vitamin B-6 (58) .
Hassinen et al. (59) demonstrated that the two main
naturally occurring B-6 vitamers in milk, PM and PL, were much
less stable than added synthetic PN. The research of
Tomarelli et al. (60) reported that the retorting of milk and
infant formula induced large losses in the availability of
naturally occurring vitamin B-6 in these products.
Very few studies have evaluated the thermal stability of
vitamin B-6 in low moisture food systems. Vitamin
fortification is common practice in the food industry and will
more than likely continue into the future with more and more
types of products being fortified. Vitamin B-6, in the form


THE EFFECTS OF PYRIDOXINE 5'-/3-D-GLUCOSIDE
ON THE METABOLIC UTILIZATION OF PYRIDOXINE
IN RATS AND HUMANS
By
JOYCE ANN GILBERT
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1991
UNiOSITYOF FLORIDA LIERARJES


51
which paralleled, in one respect, those seen in vitamin B-6
deficiency.
The metabolic pathway for the degradation of PLP involves
enzymatic hydrolysis of the phosphate ester bond, and the
oxidation of PL to 4PA (132-133). As a terminal product of
vitamin B-6 metabolism, urinary 4PA reflects the in vivo
metabolic utilization of the vitamin. In these studies, the
extent of conversion of administered [UC]PN to [UC]4PA is
indicative of reduced utilization of [14C]PN which is inversly
related to the dose of PN-glucoside. This study also shows a
negative correlation between percent radioactivity in the
urine as [UC]4PA and the quantity of PN-glucoside
administered. The data demonstrate that total liver 14C was
inversly related to the PN-glucoside dose while total urinary
14C was directly proportional to the amount of PN-glucoside
administered.
The intracellular concentration of PLP in liver tissue is
tightly regulated which does not permit the excess
accumulation of this coenzyme (133). In the liver, newly
synthesized PLP and PMP are contained in compartments that
have a rapid rate of turnover (132). These small and rapidly
mobilized pools are poorly miscible with the endogenous
coenzyme pools (132). The newly synthesized [UC]PLP and
[14C]PMP in the liver would be rapidly metabolized by
hepatocytes and degraded to [14C]4PA and excreted in the urine,
or transported to the circulation as [14C]PL, and [14C]PLP.


23
canned milk and formula products. Through the use of rat
bioassays and microbiological analyses, Tomarelli et al. (60)
evaluated the bioavailability of vitamin B-6 in milk, spray-
dried and heat sterilized infant formula products. They
concluded that approximately half of the vitamin B-6 present
in retorted milk or infant formula was biologically
unavailable. Davies et al. (88) used rat and chick growth
bioassay methods to estimate the bioavailability of vitamin B-
6 in raw and canned milk. In contrast to the study of
Tomarelli (60), Davies et al. (88) reported the relative loss
of vitamin B-6 in heat processing to be equal as determined by
Saccharomvces uvarum, rat and chick bioassay methods. This
suggests that thermal sterilization in the production of
canned evaporated milk does not adversely affect the
bioavailability of vitamin B-6. Further evaluation of the
data reported by Tomarelli (60) indicated that the
bioavailability of the naturally occurring vitamin B-6 in
thermally sterilized unfortified milk was only 15% less than
that of PN fortified milk samples (37), which further
substantiated the results of Davies et al. (88).
Lushbough et al. (89) examined the bioavailability of
vitamin B-6 in meats and the effect of cooking using
Saccharomvces uvarum and rat growth bioassay methods. The
relative losses of vitamin B-6 during the cooking process were
nearly identical for every product tested. These data suggest
little effect of cooking on the bioavailability of the vitamin


THE EFFECTS OF PYRIDOXINE 5'-/3-D-GLUCOSIDE
ON THE METABOLIC UTILIZATION OF PYRIDOXINE
IN RATS AND HUMANS
By
JOYCE ANN GILBERT
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1991
UNiOSITYOF FLORIDA LIERARJES

This work is dedicated to my parents, Helen and Reginald,
who showed me life has no boundaries, only opportunities.

ACKNOWLEDGEMENTS
I thank Drs. Lynn Bailey, Peggy Borum, Robert Cousins and
James Cerda for allowing me the previlage of studying with the
best scientists in their respective fields.
Special thanks go to my advisor and research mentor, Dr.
Jesse Gregory, for his example of integrity and dedication in
all endeavors.
I also thank Doris Sartain and my collegues in the
"Yellow Lab" for their endless patience and sensitivity.
iii

TABLE OF CONTENTS
ACKNOWLEDGEMENTS iii
LIST OF TABLES vi
LIST OF FIGURES vii
Abstract viii
CHAPTER 1
INTRODUCTION 1
CHAPTER 2
LITERATURE REVIEW 4
Bioavailability of Vitamin B-6 4
Pyridoxine-/3-D-Glucoside 15
Factors Affecting the Bioavailability of Vitamin
B-6 21
Enzymatic Interconversion of B-6 Vitamers .... 31
CHAPTER 3
PYRIDOXINE-5 -/3-D-GLUCOSIDE AFFECTS THE METABOLIC
UTILIZATION OF PYRIDOXINE IN RATS 38
Introduction 38
Materials and Methods 39
Protocol 39
Forms of vitamin B-6 40
Sample preparation 41
HPLC equipment and analysis 42
Measurement of radioactivity 43
Statistical analysis 43
Results 44
Discussion 48
CHAPTER 4
EFFECTS OF PYRIDOXINE-5 '-/3-D-GLUCOSIDE ON THE
METABOLIC UTILIZATION OF PYRIDOXINE IN HUMANS .... 55
Introduction 55
Materials and Methods 56
Synthesis of forms of vitamin B-6 56
Protocols of trials with human subjects ... 57
Analytical methods 59
iv

Mass spectral analysis of deuterium-labeled
4 PA 59
Statistical analysis 60
Results 61
Vitamin B-6 nutritional status of subjects 61
Stable-isotopic trials 61
Discussion 65
CHAPTER 5
EFFECTS OF PYRIDOXINE-5' -/3-D-GLUCOSIDE ON THE
IN VITRO KINETICS OF PYRIDOXAL KINASE AND PYRIDOXAMINE
(PYRIDOXINE)-5'PHOSPHATE OXIDASE IN RAT LIVER 69
Introduction 69
Materials and Methods 70
Protocol 70
Forms of vitamin B-6 71
Pyridoxal kinase 72
Pyridoxamine (pyridoxine) 5' phosphate
oxidase 75
Statistical analysis 77
Results 77
PL kinase 77
PMP (PNP) oxidase 79
Discussion 83
CHAPTER 6
SUMMARY AND CONCLUSIONS 86
LITERATURE CITED 92
BIOGRAPHICAL SKETCH 106
V

LIST OF TABLES
Table 3.1 Liver B-6 vitamer distribution and total
liver 14C in rats administered varying levels of PN-
glucoside in the dose (Experiment 1) .* 44
Table 3.2 Urinary 4PA and total urinary 14C in rats
administered varying levels of PN-glucoside
(Experiment 1)* 45
Table 3.3 Liver B-6 vitamer distribution and total
liver 14C in rats administered varying levels of PN
or PN-glucoside. (Experiment 2)* 46
Table 3.4 Urinary 4PA and total urinary 14C in rats
administered varying levels of PN or PN-glucoside.
(Experiment 2)* 47
Table 4.1 Indicator of vitamin B-6 nutritional status
of human subjects 24 h prior to each trial1. ... 62
Table 4.2 Molar isotopic ratio of urinary d24PA.1 . 63
Table 4.3 Percentage of d2PN (5unol) dose excreted as
urinary d24PA 64
Table 4.4 Total urinary 4PA excretion.1 65
Table 5.1 Purification of pyridoxal kinase from rat
liver 78
Table 5.2 Kinetic parameters of pyridoxal kinase with
varying concentrations of PN-glucoside 79
Table 5.3 Purification of pyridoxamine (pyridoxine)
phosphate oxidase from rat liver 80
Table 5.4 Kinetic parameters of pyridoxamine
(pyridoxine) phosphate oxidase with varying
concentrations of PN-glucoside 80
vi

LIST OF FIGURES
Figure 5.1 Double reciprocal plot for the
phosphorylation of PN by PL kinase from rat liver
in the presence of various levels of PN-
glucoside 81
Figure 5.2 Double reciprocal plot for the conversion of
PMP to PLP by PMP (PNP) oxidase from rat liver in
the presence of various levels of PN-glucoside. 82
vii

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE EFFECTS OF PYRIDOXINE 5' -/3-D-GLUCOSIDE
ON THE METABOLIC UTILIZATION OF PYRIDOXINE
IN RATS AND HUMANS
By
Joyce Ann Gilbert
December, 1991
Chairperson: Dr. Jesse F. Gregory III
Major Department: Food Science and Human Nutrition
A major form of vitamin B-6 in plant-derived foods is 5'-
0-(/3-D-glucopyranosyl) pyridoxine (PN-glucoside) Previous
studies have shown that PN-glucoside is poorly available as a
source of vitamin B-6 in rats and undergoes incomplete
utilization in humans. The present research was conducted to
determine whether unlabeled PN-glucoside affects the metabolic
utilization of simultaneously administered isotopically
labeled pyridoxine (PN) in rats and humans. In addition, the
in vitro effect of PN-glucoside on the activity of enzymes in
the vitamin B-6 metabolic pathway, specifically pyridoxal-
kinase and pyridoxamine (pyridoxine)-phosphate oxidase was
determined.
Experimental results with rats given [UC]PN indicated
that urinary excretion of KC increased significantly with
viii

increasing dose of PN-glucoside, while hepatic 14C decreased
significantly as the PN-glucoside dose increased. The
proportion of hepatic 14C-labeled pyridoxal, PN, and
pyridoxamine decreased whereas hepatic pyridoxine phosphate
and pyridoxal phosphate increased in proportion to the PN-
glucoside dose. In addition, the concentration of urinary
[14C] 4-pyridoxic acid (4PA) relative to total urinary 14C,
decreased as the dose of PN-glucoside increased. Stable
isotopic methodology was employed to determine whether PN-
glucoside affected the metabolic utilization of simultaneously
administered deuterium-labeled PN (d2PN) in humans.
Experimental results showed that twenty-four hour urinary
excretion of 4PA was decreased significantly with increasing
dose of PN-glucoside. The percentage of ingested d2PN
excreted as d24PA showed an inverse relationship, that was
statistically significant, in proportion to the PN-glucoside
dose. In vitro enzyme assays indicated that PN-glucoside had
no significant effect on the activity of partially purified
pyridoxal-kinase and pyridoxamine (pyridoxine) phosphate
oxidase. These results provide evidence that PN-glucoside
weakly retards the metabolic utilization of nonglycosylated
forms of vitamin B-6. However, the effect of PN-glucoside on
PN is not due to the direct effect of PN-glucoside on the
enzymes PL kinase and PMP (PNP) oxidase.
ix

CHAPTER 1
INTRODUCTION
Vitamin B-6 has often been referred to as the protein
vitamin because of its association with amino acid metabolism
and the influence of dietary intakes of protein on vitamin B-6
requirements (1-3). Adequate vitamin B-6 nutriture is
essential to health through its numerous roles in the body as
the active coenzyme form, pyridoxal 5'-phosphate (PLP). A
large majority of these roles involves metabolism of amino
acids. These functions include, among others, nonoxidative
decarboxylation of amino acids, transamination,
desulfhydration, and enzymes affecting reactions of amino acid
side chains. Such PLP dependent enzymes are important
functional complexes in the biosynthesis and catabolism of
essential and non-essential amino acids and some provide a
connection between the amino acid and intermediates of
carbohydrate metabolism. Recent research in animal models has
suggested a role for PLP in the modulation of hormone actions
(4-5).
Nutritional status with respect to vitamin B-6 is
influenced by the amount of the vitamin ingested, the extent
1

2
of absorption and metabolic utilization of the B-6 compounds
in the diet, and the specific requirements of the individual.
Although vitamin B-6 is widely distributed in nature, the
amount in food is relatively small. The American diet often
contains 1 2 mg per day (2) Marginal vitamin B-6
nutritional status has been well documented in several
segments of the American population including adolescents,
pregnant women and the elderly (6-15). However, vitamin B-6
deficiency with apparent clinical symptoms is not widespread
in the general population. It is significant that over 50% of
the subjects of the USDA 1977-78 Nationwide Food Consumption
Survey consumed less than 70% of the Recommended Dietary
Allowance (RDA) for vitamin B-6 (2) The 1985 Continuing
Survey of Food Intakes of Individuals indicated that only 27%
of women consumed 70% or more of the RDA for vitamin B-6 (12) .
Women of all ages and the elderly have been especially
prevalent in the group consuming less than the RDA for vitamin
B-6. Vitamin B-6 is required in amino acid metabolism and
therefore, the RDA for this vitamin is related to protein
intake. Twice the RDA for protein, which is considered the
upper boundary of acceptable level of protein intake, was used
to establish the RDA for vitamin B-6. The RDA for vitamin B-6
was revised using a dietary ratio of 0.016 mg vitamin B-6/g
protein.
Our present knowledge of the essential role of vitamin B-
6 in preserving health emphasizes the importance of

3
establishing an accurate RDA for vitamin B-6. Establishing an
adequate RDA for vitamin B-6 is a complex task. The
appropriateness of the RDA for vitamin B-6 is dependent on
accurate information on the content of the vitamin in foods
consumed, the enzymes involved in the metabolism of the
various forms of vitamin B-6, as well as information regarding
vitamin B-6 bioavailability.

CHAPTER 2
LITERATURE REVIEW
Bioavailabilitv of Vitamin B-6
The concept of bioavailability is most appropriate when
considering the extent of intestinal absorption and metabolic
utilization and any potential antagonistic effects of all
naturally occurring dietary forms of the vitamin. Hence, net
bioavailability is defined as the portion of total dietary
vitamin B-6 that is biologically active following intestinal
absorption (16). Vitamin B-6 as it occurs in mammalian
tissues and fluids encompasses three interconvertible 3-
hydroxy-2-methylpyridine compounds including pyridoxine (PN) ,
pyridoxal (PL), pyridoxamine (PM), their corresponding 5'-
phosphoryl derivatives (PNP, PLP, and PMP respectively) and
the excretory catabolite 4-pyridoxic acid (4PA) (17). Vitamin
B-6 in nature is found in phosphorylated forms and closely
associated with proteins. In foods, PL, PN, and PM are found
as both phosphorylated and nonphosphorylated forms. PL, PM,
PLP, and PMP are the predominant forms in animal products,
while PN, in its free vitaminic form and as the glycoside
(discussed in a later section) are the major forms in most
plant foods (18).
4

5
To quantitate total vitamin B-6 in most food, test
microorganisms are generally utilized. The most common
organism used is Saccharomvces uvarum. Test microorganisms
demonstrate low biological activity of vitamin B-6 in its
bound (pyridoxyl-amino acid and amines) glycosylated and
phosphorylated forms until they are released by acid
hydrolysis. However, this acid treatment of food is not
entirely representative of the digestive process in the
gastrointestinal tract.
Results of early studies of intestinal absorption of
vitamin B-6 indicated that absorption occurs by a simple
passive diffusion mechanism (19-22) A number of studies
indicate that the rate of uptake by intestinal tissue of the
B-6 vitamers, PN, PM, and PL increased proportionally to the
dose over a wide range of intralumenal concentrations (19,23-
31). Transport of PN across the intestinal wall occurred by
diffusion, independent of accumulation within the tissue.
There is no evidence of a saturable transport system either in
vivo (19,23-24) or in vitro (23,26-31). These data also
indicate that transport across the intestinal wall from the
mucosal to the serosal side of the non-phosphorylated B-6
vitamers to be quantitatively similar. PN in the free and
conjugated forms is rapidly absorbed by the intestine directly
or following hydrolysis by intestinal enzymes. PL and PM are
present in food primarily as the 5'-phosphates and require
hydrolysis by lumenal phophatases. The phosphorylated B-6

6
vitamers, PLP and PMP are well absorbed following enzymatic
dephosphorylation, although these vitamers are slowly absorbed
as the intact 5'-phosphate esters (28-30). Studies have
indicated that the transport of the B-6 vitamers into the
intestinal mucosal cell involves a saturable enzymatic
process, identified as intracellular phosphorylation by PL
kinase (EC 2.7.1.35) (24, 28-29). Most ingested vitamin B-6
is absorbed by the jejunum in the nonphosphorylated forms, PL,
PN, and PM. Tissues, especially liver, rapidly take up
circulating vitamin B-6, where the phophorylated B-6 vitamers
are hydrolyzed by plasma membrane phosphatases and enter the
cells by a facilitated process and diffussion followed by
metabolic trapping (28-29). PN is phosphorylated to PNP by PL
kinase which has been detected in all mammalian tissues
investigated (32). PNP is then converted to PLP by flavin-
dependent PNP oxidase (EC 1.4.3.5) (33) which, in contrast to
the widely distributed PL kinase, is found in few tissues,
mainly liver, erythrocytes, kidney, and brain (34-35). PLP
can be transformed to PMP by transamination or hydrolyzed to
PL by phosphatases (17). PL is then converted to pyridoxic
acid (4PA) by aldehyde oxidase (18) or again to PLP by PL
kinase. PLP is bound by cellular protein or released into
plasma, by the cell, as PLP, PL or 4PA. The analysis of
kinase and phosphatase enzymes in liver tissue indicates
similar activities. The relative activities of these enzymes
account for the accumulation of the 5'-phosphate compounds;

7
however, it also allows for the dephosphorylation of PLP that
is not protein bound and its release from the liver as PL or
oxidized to 4PA (31). Regulation of vitamin B-6 metabolism
may also occur through the conversion of PNP and PMP to PLP,
which is highly sensitive to product inhibition (30).
Recent advances in analytical techniques have enabled
collection of more precise data concerning the quantity and
various forms of vitamin B-6 compounds present in the diet.
However, present knowledge of the bioavailability of vitamin
B-6 is still not sufficient enough to determine an accurate
assessment of adequate dietary intake.
The bioavailability of a nutrient such as vitamin B-6 is
primarily determined experimentally by comparing the
concentration of vitamin B-6 that is biologically active to
total vitamin B-6 ingested (16). Traditionally biologically
available vitamin B-6 has been determined through animal
bioassays (36). Human bioassays, to determine biologically
active vitamin B-6 in food, have been performed, but are
generally lengthy procedures with limited precision.
Significant advances and recent applications in isotopic
methodology offers a useful alternative to bioassay for the
sutdy of bioavailability of vitamin B-6 (36). Results of
studies utilizing rat bioassays to compare animal products
with plant-derived products have indicated that the
bioavailability of vitamin B-6 in animal products is greater
than that of plant-derived products (36). In general, a poor

8
correlation was observed in these samples when comparing rat
growth and plasma PLP concentration. These results were
difficult to interpret because of the potential effect of diet
composition on the synthesis of vitamin B-6 by intestinal
microflora. The lack of agreement between rat growth and
plasma PLP, suggests that data from bioassays to determine
biologically available vitamin B-6 in food sources,
specifically those in which the test diets may differ greatly
from the reference diet in the type of carbohydrate, are
somewhat equivocal (37). The bioavailability of vitamin B-6
has been studied in somewhat more detail than many of the
other vitamins. However, few generalizations can be stated
regarding the overall bioavailability of the B-6 vitamers or
the factors influencing it.
Essential to measuring the bioavailability of vitamin B-6
in a food source is the determination of the total vitamin B-6
content in the food. This is complicated by the interaction
of the vitamin with food components. These interactive
compounds may or may not be biologically available. The
interactions of vitamin B-6 with food components and with
other vitamin B-6 compounds are important considerations since
they may represent sources of available vitamin B-6 in the
diet. Experimental data regarding the bioavailability of
naturally occurring vitamin B-6 in foods were initially
reported by Sarma et al. (34). The authors utilized a
comparison of rat bioassay and Saccharomvces uvarum assay

9
results for a variety of plant and animal products. Liver
fractions, whole wheat and yellow corn exhibited
bioavailability values of between 65% and 70% compared with an
apparent value of approximately 100%. These results were the
first indication of a wide variation and potentially
incomplete bioavailability of vitamin B-6 in cereal grains and
other food products (34).
A comparison of rat growth bioassays and Saccharomvces
uvarum data for dried beef, lima beans, non-fat dry milk and
whole wheat flour yielded a good correlation between the
assays with small differences obtained for the flour and milk
products (35). Nelson et al. (38) examined the rate of
intestinal absorption of the vitamin B-6 in orange juice in
human subjects. A triple lumen perfusion technique was
employed to determine the relative absorption of naturally
occurring vitamin B-6 in orange juice. Their results
demonstrated a significant decrease in rate of absorption of
vitamin B-6 in orange juice (42%) than that of synthetic B-6
vitamers in saline (67%) and saline-glucose solutions (79%).
The lower rate of absorption of the naturally occurring
vitamin B-6 in orange juice was thought to be a result of
interactions of the vitamin with naturally occurring food
components in the orange juice (38). A follow-up study
conducted by Nelson et al. (39) investigated the nature and
extent of binding of different forms of vitamin B-6 in orange
juice. The study suggested extensive and equal binding of

10
both PL and PN in orange juice to a small dialyzable molecule
which is heat stable and non-protein in nature. The
interaction of vitamin B-6 with this component of orange juice
was suggested as being responsible for the lower rate of
absorption of naturally occurring vitamin B-6 in orange juice.
Leklem and coworkers (40) were the first to employ a
bioassay with human subjects to examine the effect of dietary
fiber on the bioavailability of the vitamin. The study
investigated the bioavailability of vitamin B-6 in white and
whole wheat bread. Significant differences in the
bioavailability of B-6 vitamers were not apparent when plasma
PLP and erythrocyte aminotransferase data were examined.
However, fecal vitamin B-6 and urinary 4PA excretion data
suggested incomplete utilization of the vitamin in the whole
wheat bread. Frequent attempts have been made to calculate
the vitamin B-6 balance of B-6 vitamers and 4PA in bioassays
employing human subjects (41).
Similar experiments in which the use of data concerning
fecal vitamin B-6 have shown little validity. Unabsorbed
vitamin B-6 from dietary sources as well as microbiologically
synthesized B-6 vitamers are contained in fecal material. The
composition of the diet will influence the microbial
contribution to fecal vitamin B-6 which may affect the
apparent vitamin B-6 nutriture of the subject. The fact that
the intestinal microorganisms produce vitamin B-6 is well
established (41-46). However, the availability of vitamin B-6

11
to the host organism, especially in the absence of coprophagy,
has been unclear (45,47). Since several animals can develop
a vitamin B-6 deficiency even when coprophagy is allowed,
metabolic utilization of microbial vitamin B-6 must be miminal
(48). In animals exposed to conditions of nutritional
deficiency, coprophagy may increase sufficiently to become a
significant source of nutrients. This may explain the results
of Ikeda et al. (49-50), who observed that germfree rats tend
to be more susceptible to vitamin B-6 deficiency than rats
with normal populations of intestinal microorganisms. Hughes
et al. (51) reported that in addition to the intestinal
microflora, nutritionally deprived animals may also benefit
from other environmental sources of microbial vitamins such as
microbial activity on the floor of the pen. As indicated by
these studies, there exists no compelling evidence that
vitamin B-6 can be absorbed from the large intestine in
amounts sufficient to make a detectable contribution to the
daily intake.
A study (52) involving human bioassay in measuring
bioavailability of vitamin B-6 in beef compared with soybeans
suggested the vitamin B-6 in soybeans to be less available
compared to that contained in beef. It has been suggested that
the presence of nondigestible polysaccharides and lignin
components of dietary fiber of the plant derived foods is
responsible for differences in the availability of vitamin B-6
in plant foods compared with animal products (40). A variety

12
of physical and chemical properties of dietary fiber suggest
the possibility of binding or entrapment of the B-6 vitamers,
which may influence intestinal absorption. Schultz and Leklem
(53) did a study comparing vegetarian versus nonvegetarian
women. They observed that although the vegetarian women
consumed more crude fiber than the nonvegetarian women, there
was no significant differences between these two groups for
plasma PLP, urinary 4PA, and urinary vitamin B-6. Therefore,
it was concluded that there appeared to be no adverse effect
of fiber on the bioavailability of vitamin B-6 between these
groups (53). It has been reported that human intestinal
microflora produce vitamin B-6 (54) Several bioavailability
studies have indicated that ingestion of diets high in dietary
fiber or carbohydrate leads to significant increases in
microbial synthesis of vitamin B-6 (55-56).
Tarr et al. (57) evaluated a typical American mixed diet
to its bioavailability of vitamin B-6. They reported 70%
bioavailability of vitamin B-6 based on urinary 4PA and plasma
PLP concentration relative to PN in a formula diet. The
authors speculated that thermal processing was potentially
responsible for the incomplete bioavailability of vitamin B-6
because canned goods, including both animal and plant foods,
composed much of the mixed diet. A study by Nguyen and
Gregory (36) employed rat bioassays to examine the
bioavailability of vitamin B-6 in selected foods as influenced
by thermal processing. The effects of food composition and

13
thermal processing on the relative bioavailability of vitamin
B-6 in beef, spinach, potato and cornmeal were assessed. The
results further demonstrated that diet composition and food
processing must be considered in evaluating the
bioavailability of vitamin B-6.
The instability of certain B-6 vitamers during food
processing and storage may contribute to losses of the
nutritional quality of foods with respect to vitamin B-6.
Processing and storage of foods have been found to have
various effects on vitamin B-6 with losses decreasing the
adequacy of food products as sources of dietary vitamin B-6.
A deficiency of vitamin B-6 in infants who consumed a non-
fortified, heat sterilized canned infant formula, led to
extensive research, during the 1950's, concerning the effects
of food processing on the bioavailability of vitamin B-6 (58) .
Hassinen et al. (59) demonstrated that the two main
naturally occurring B-6 vitamers in milk, PM and PL, were much
less stable than added synthetic PN. The research of
Tomarelli et al. (60) reported that the retorting of milk and
infant formula induced large losses in the availability of
naturally occurring vitamin B-6 in these products.
Very few studies have evaluated the thermal stability of
vitamin B-6 in low moisture food systems. Vitamin
fortification is common practice in the food industry and will
more than likely continue into the future with more and more
types of products being fortified. Vitamin B-6, in the form

14
of PN-HC1, is added to many breakfast cereals at levels of
25% to 100% of the United States Recommended Daily Allowance
(USRDA) per ounce. Gregory et al. (61-62), using a dehydrated
model food system, simulating breakfast cereals, indicated
that the roasting and storage of low moisture food systems
resulted in losses of 50% to 70% of the added PN,PM,and PLP.
The remaining vitamin B-6 was found to be fully available as
determined by rat bioassay using growth, feed efficiency,
erythrocyte aspartate aminotranferase activity (AspAT) and in
vitro (AspAT) coenzyme stimulation. Gregory (63) also
examined vitamin B-6 bioavailability in rice-based, PN-
fortified cereal and non-fat dry milk. The non-fat milk and
rice base breakfast cereal samples were analyzed for vitamin
B-6 by microbiological, HPLC, and rat bioassay procedures.
The results indicated that the vitamin B-6 of the non-fat dry
milk was fully available, while the vitamin B-6 availability
in the cereal product was comparatively low. The apparent
losses of the PN in the fortified cereal was explained by a
first order kinetic model described by Evans et al. (64).
These studies (61,63-64) indicated that food fortified with PN
is susceptible to significant degradation under certain
processing and storage conditions and that the bioavailability
of the remaining vitamin B-6 may not be complete. The above
results suggest that thermal processing of foods does not
induce nutritionally important losses in the bioavailability
of vitamin B-6. At the very least, research to date indicates

15
that any possible adverse effects of thermal processing and
storage on the bioavailability of vitamin B-6 would not be of
sufficient magnitude to explain the incomplete utilization of
the vitamin observed in human subjects (57,65).
Pvridoxine-fl-D-Glucoside
A conjugated form of pyridoxine, 5'-O-(0-D-
glucopyranosyl) pyridoxine (PN-glucoside) has been found to be
a major naturally occurring form of vitamin B-6 in many fruits
and vegetables of human diets. Scudi et al. (66) first
reported the presence of conjugated forms of the vitamin B-6
in 1942 to occur in rice bran. Some thirty-five years later,
Yasumoto et al. (67) reported that the PN component of the
conjugated form of vitamin B-6 in rice bran was in a 1:1 ratio
with glucose and identified this glycosylate as a 5'-
glucopyranosyl-derivative of PN. Several other studies (68-
69) have indicated the synthesis of this glucoside in pea
seedlings as well as the formation of other PN conjugates in
lesser concentrations than the 5'-pyridoxine-glucoside
derivative. Following the first isolation and structural
identification of 5 -0- (/3-D-glucopyranosyl) pyridoxine (PN-
glucoside) from rice bran (67), the possible nutritional
significance of glycosylated forms of vitamin B-6 has been a
subject of intense study. Evidence for the presence of
conjugated forms of vitamin B-6 existing as a variable
proportion of the total vitamin B-6 of plant-derived foods has

16
been observed as well as evidence supporting the widespread
existence of /3-glucoside conjugates of B-6 vitamers in plant
tissues (68). PN-glucoside appears to be formed in plants by
enzymatic transglycosylation (70-73). The glucose moiety of
PN-glucoside occurs at the 5' and 4' positions of the PN ring;
however, the presence of the 5'-isomer as the primary
naturally occurring form of PN-glucoside suggests specificity
for this position by the transglycosylation reaction in most
plant tissue (67). These studies provide evidence that
conjugates of vitamin B-6 exist as a substantial proportion of
the total vitamin B-6 in plant-derived foods (66-69). The
results of Kabir et. al (65) indicated the existance of /3-
glucopyranosyl conjugates in a wide variety of plant products.
The importance of PN-glucoside as a bioavailable source
of vitamin B-6 in humans was first studied by Kabir et al.
(74). The investigators determined the urinary and fecal
excretion patterns of glycosylated vitamin B-6 when different
foods that contained the naturally occurring PN-glucoside were
fed to human subjects. It was observed that an inverse
relationship existed between plant derived PN-glucoside (as
percent of total B-6) and the net B-6 bioavailability in
humans. The investigators suggested that the proportion of
the glucosylated vitamin B-6 in the diet may be useful as an
index of vitamin B-6 bioavailability.
Kabir et al. (65) developed a microbiological assay
procedure for quantitating glycosidic conjugated forms of PN

17
as well as total vitamin B-6 in food. The PN-glucoside, when
measured indirectly by hydrolysis, was reported to occur
widely in plant foods, ranging from 0% to 82% of the total
amount of vitamin B-6 present (74-76).
Studies by Iwami and Yasumoto (76) indicated that
intestinal absorption of PN-glucoside was similar to that of
free PN. However, Ink et al. (77) administered a single, oral
dose of radiolabeled PN and PN-glucoside in an alginate gel
diet using a combination of intrinsic and extrinsic labeling
in the intact rat. They reported that the absorption of the
glucoside was similar to that of unconjugated PN, but the
absorbed PN-glucoside was metabolically utilized less
effectively as vitamin B-6 than was PN. Iwami et al. (76)
measured the relative disappearance from the intestinal
perfusate of PN and a purified 4'-derivative (rather than the
naturally occurring 51-derivative). In the rat, Ink et al.
(77) observed the relative metabolism of the intrinsically and
extrinsically labeled glucoside to have less than 40% of the
bioavailability of PN.
The nutritional properties of these conjugated forms of
vitamin B-6 are not fully understood at the present time.
Although PN-glucoside comprises 25% to 75% of the total
vitamin B-6 in many plant tissues (67,75) its bioavailability
in humans continues to be studied. PN-glucoside has been
found to undergo intestinal absorption in the intact form.
Initial studies employing bioassays with human subjects (40)

18
indicated that the proportion of the total vitamin B-6 which
was present as PN-glucoside correlated closely with the
bioavailability of the vitamin in a variety of foods examined,
including tuna, peanut butter and whole wheat bread. These
results suggested that PN-glucoside would not be available to
humans for intestinal absorption and metabolic utilization.
However, further studies by Bills et al. (78) indicated that
this observed correlation was inconsistent when additional
foods were examined. As discussed earlier, Nelson et al. (38)
reported that the vitamin B-6 from orange juice, a large
portion of which is in the conjugated form, exhibited a slower
rate of absorption during intestinal perfusion of humans than
PN from control solutions. Tsuji et al. (79) observed that
synthetic PN-glucoside exhibited vitamin B-6 activity which
was approximately equivalent to that of PN in a rat bioassay.
In contrast, Trumbo et al. (80) utilizing a rat bioassay
reported that PN-glucoside isolated from alfalfa sprouts
exhibited only 10% to 30% bioavailability relative to PN in
the rat. Studies in rats using radiolabeled PN-glucoside
indicate effective absorption of this B-6 compound and that
the portion of PN-glucoside not hydrolyzed to metabolically
active PN, is rapidly excreted in the urine in the intact form
(77,81). Ink et al. (77) reported the extent of absorption of
radiolabeled PN-glucoside from diets containing intrinsically
enriched alfalfa sprouts was similar, but quantitatively less

19
than that of purified labeled PN-glucoside, suggesting an
inhibitory effect of plant tissues.
Stable-isotopic techniques and methodologies evaluate the
bioavailability of purified deuterium-labeled (d2) PN-
glucoside in human subjects (82) In a study that used stable
isotopic techniques to analyze the bioavailability of orally
administered extrinsically enriched oatmeal containing labeled
PN-glucoside, it was reported that the labeled PN-glucoside
was approximately 58% of that of labeled-PN fed under
identical conditions (82) Although the ability metabolically
to utilize PN-glucoside appears to have a substantial level of
individual variability among human subjects, the
bioavailability of PN-glucoside in human subjects has been
observed to be substantially greater than that of the rat
(82). Further development of the methodology and application
of stable-isotopic techniques will continue to provide greater
clarification of the factors affecting bioavailability of
vitamin B-6.
The above section is a review of the current knowledge
concerning the bioavailability of the vitamin B-6 in foods.
There appears to be a general agreement between the rat
bioassay and the Saccharomvces uvarum assay method for many
foods, although the range of apparent bioavailability relative
to PN was quite large for the rat. These comparative data
provided the first indication that the bioavailability in
certain plant foods may be less than complete. The factors

20
responsible for the differences in apparent bioavailability
of vitamin B-6 are not clear.
Human subjects have also been used in researching the
bioavailability of vitamin B-6 in specific foods and mixed
diets. Bioassays with humans indicated that the overall
bioavailability of vitamin B-6 in a typical American diet is
reasonably high although incomplete. A study evaluating free-
living vegetarian and non-vegetarian women who consumed
quantities of total vitamin B-6 equivalent to those in a mixed
diet found no significant differences between groups with
respect to vitamin B-6 status when comparing plasma PLP levels
(53). This suggests that the bioavailability of vitamin B-6
in vegetarian diets was similar to that in a mixed diet.
Andn et al. (84) reported that a mean of 2.5% of the total
vitamin B-6 present in breast milk of omnivorous lactating
women was in the form of PN-glucoside. Although a mean of 15%
of the total vitamin B-6 in the diet was present as PN-
glucoside, there appeared to be no correlation between the
amount of glycosylated B-6 in the diet and the percent PN-
glucoside in breast milk over the range examined. In
constrast, a study by Reynolds et al. (85) reported that the
percentage of dietary vitamin B-6 present as PN-glucoside was
equivalent to the amount of PN-glucoside found in the breast
milk of lactating Nepalese vegetarian women. Gregory and Ink
(75) did not observe any PN-glucoside in the breast milk of
six lactating American females including three

21
lactoovovegetarians. Differences in the vitamin B-6
composition of the breast milk reported in these studies are
unclear. The factors responsible for the incomplete
bioavailability of vitamin B-6 have not been determined.
Factors Affecting the Bioavailabilitv of Vitamin B-6
Several factors are known to affect the bioavailability
of vitamin B-6 in food products. These include the formation
of certain reaction products during food processing and
storage, fiber type and the quantity present in the food
source, and the presence of PN-glucoside in plant foods. The
results of previously discussed studies suggested that thermal
processing and storage of foods may adversely affect
bioavailability of vitamin B-6 through the formation of
reaction products with amino acid residues of the
proteinatious portion of the food (36,56-57).
The bioavailability of vitamin B-6 from animal products
approaches 100% for most foods. The biological activity of
vitamin B-6 from plant derived products is generally lower.
A series of studies concerning the nutritional quality of
military rations provided the initial impetus to consider the
adverse effect of thermal preservation of canned foods.
Register et al. (83) reported that rats which subsisted
wholely on either of two homogenized combat rations required
supplemental vitamin B-6 to sustain normal growth even though
the diets contained enough vitamin B-6 for adequate rat

22
growth. Harding et al. (86) fed human subjects canned rations
which had been stored for twenty months at 100 F. They
observed marginal vitamin B-6 deficiency in the subjects even
though the rations provided apparently adequate amounts of the
vitamin. Inadequate vitamin B-6 nutriture was most likely a
result of an effect of thermal processing. It is reasonable to
consider that marginal limitations in essential amino acid
content of the ration may have contributed to nutrient
insufficiency. The addition of vitamin B-6 to diets deficient
in certain essential amino acids has been shown to produce a
growth response (87). However, the reason for this apparent
incomplete bioavailability of vitamin B-6 in these canned
rations is unclear. Several infants who consumed a non-
fortified, heat sterilized canned infant formula were found to
be severely vitamin B-6 deficient and suffer with
neurochemically induced convulsions. Extensive research was
subsequently initiated to study the chemical behavior of
vitamin B-6 in milk products (58).
The differing stability properties of the various vitamin
B-6 compounds have been of interest since Hassinen et al. (59)
observed that PM and PL, the naturally occurring forms of
vitamin B-6 in milk, were much less stable than added PN.
Employing microbiological assays, Hassinen (59) demonstrated
that PM and PL added to milk were degraded at the same rate as
naturally occurring B-6 vitamers. Added PN exhibited greater
stability than either PM or PL during thermal processing of

23
canned milk and formula products. Through the use of rat
bioassays and microbiological analyses, Tomarelli et al. (60)
evaluated the bioavailability of vitamin B-6 in milk, spray-
dried and heat sterilized infant formula products. They
concluded that approximately half of the vitamin B-6 present
in retorted milk or infant formula was biologically
unavailable. Davies et al. (88) used rat and chick growth
bioassay methods to estimate the bioavailability of vitamin B-
6 in raw and canned milk. In contrast to the study of
Tomarelli (60), Davies et al. (88) reported the relative loss
of vitamin B-6 in heat processing to be equal as determined by
Saccharomvces uvarum, rat and chick bioassay methods. This
suggests that thermal sterilization in the production of
canned evaporated milk does not adversely affect the
bioavailability of vitamin B-6. Further evaluation of the
data reported by Tomarelli (60) indicated that the
bioavailability of the naturally occurring vitamin B-6 in
thermally sterilized unfortified milk was only 15% less than
that of PN fortified milk samples (37), which further
substantiated the results of Davies et al. (88).
Lushbough et al. (89) examined the bioavailability of
vitamin B-6 in meats and the effect of cooking using
Saccharomvces uvarum and rat growth bioassay methods. The
relative losses of vitamin B-6 during the cooking process were
nearly identical for every product tested. These data suggest
little effect of cooking on the bioavailability of the vitamin

24
B-6 in meats. Studies conducted by Ink et al. (90) used a
combination of intrinsic and extrinsic labeling of liver and
muscle tissue with radiolabeled forms of vitamin B-6. These
data indicated that thermal processing caused partial
destruction of vitamin B-6 in liver and muscle, but had little
or no effect on the bioavailability of the remaining vitamin
B-6.
The bioavailability of vitamin B-6 in foods from plant
origin has been proposed to be adversely influenced by dietary
fiber (36-37). The effect of various forms of dietary fiber
on vitamin B-6 bioavailability has been examined in both
animals and humans.
The effect of dietary fiber on the bioavailability of B-6
vitamers was first studied by Leklem et al. (40) who employed
a human bioassay to examine the utilization of vitamin B-6 in
white verse whole wheat bread. Their data indicated 10% lower
availability of the vitamin B-6 which was attributed to
reduced intestinal absorption by the fiber components. The
results of these studies suggested that the vitamin B-6 of
soybeans was approximately 5% to 10% less available than beef
vitamin B-6. These data have suggested that the presence of
nondigestible components of dietary fiber, may reduce the
bioavailability of the vitamin B-6. Physical properties
inherent of dietary fiber may influence the absorption of the
vitamin through entrapment or alteration of the viscosity of
the intestinal contents (16).

25
The presence and type of purified dietary fiber have only
a minor effect on the bioavailability of the B-6 vitamers (91-
93). Nuygen et al. (91) evaluated the potential for physical
binding of vitamin B-6 using a variety of native and modified
polysaccharides and lignin under conditions similar to the
human intestine. In vitro binding of the B-6 vitamers by
these fiber components did not occur. Differences in
viscosity may have influenced the rate of diffusion, although
studies by Machida and Nagai (93) indicated that a reduced
rate of absorption does not induce a reduction in net vitamin
B-6 absorption. Rat and chick bioassays have been employed to
evaluate the effects of selected dietary fiber components on
the absorption of vitamin B-6 (91,93). The results of these
bioassays indicated no inhibitory effects of these dietary
fiber components on the bioavailability of vitamin B-6. The
results of these studies along with the results of studies
with human subjects (39,56) suggest that dietary fiber has
little effect on the bioavailability of vitamin B-6 in foods.
The slight effect of dietary fiber components on the
bioavailability of vitamin B-6 does not fully account for the
lower bioavailability of B-6 vitamers from plant food products
relative to animal sources.
Many vitamin B-6 antimetabolites have been identified
and determined to be chemical reaction products formed from
vitamin B-6 during the thermal processing or storage of foods.
The potential conversion of various B-6 vitamers to these

26
antagonistic compounds has been proposed to explain certain
observed effects of food processing or storage on the
bioavailability of vitamin B-6 in certain foods (94-96). In
studying the stability of the B-6 vitamers in food products,
PN was found to be the most stable vitamer, followed by PL,
PM, PMP, and the least stable B-6 vitamer being PLP (88) .
Much research has focused on the potential for PL and PLP to
react with such food components as proteins, amino acids and
reducing sugars resulting in the formation of degradation
products of limited bioavailability.
Several studies have reported the characterization of the
by-products resulting from the interaction of PL and PLP with
amino groups of proteins in food products (96-102). The
chemical reaction of PL or PLP with cysteine or other
sulfhydryl amino groups has been proposed as a mechanism
responsible for the lowered bioavailability of vitamin B-6
reported in thermally sterilized milk products (96-98). One
such product produced from the heat induced reaction of
cysteine and PL has been identified as bis-4-pyridoxyl
disulfide (97). Later studies by Srncova et al. (99)
supported the reactivity of milk protein sulfhydryl groups
with PL when high concentrations of PL were present. The
spontaneous reaction of PL or PLP with various thiols and
aminothiols results in the formation of thiohemiacetal and
thiazolidine complexes (94) These chemical complexes are
readily dissociable and therefore would not impair the

27
bioavailability of these B-6 vitamers. Gregory et al. (100-
102) conducted a series of experiments to examine further the
possible influence of thermal processing and storage on the
net bioavailability of vitamin B-6 in foods. When milk
containing radiolabeled PL or PLP was subjected to heat
sterilization, analysis using HPLC revealed no formation of
bis-4-pyridoxal disulfide (103). The losses of the B-6
vitamers PL and PLP were due to the reductive binding of the
aldehydes of these vitamers to food protein as e-
pyridoxyllysyl residues (103). The formation of e-
pyridoxyllysine has been identified as a mechanism of the loss
of vitamin B-6 during thermal processing or low moisture
storage of proteinaceous model food systems (100-101) and
various meat and dairy food products (16). Similar results
were reported in intrinsically enriched chicken liver and
muscle tissues (103). The phosphopyridoxyllysyl complex
accelerated the onset and enhanced the severity of vitamin B-6
deficiency symptoms. The effect of e-pyridoxyllysine on the
bioavailability of vitamin B-6 in foods is a function of the
ratio of
total vitamin
B-6
content of
the diet and
the
addition
product to
the
other B-6
vitamers.
The
bioavailability of the vitamin of diets which contain e-
pyridoxyllysine but which are adequate in vitamin B-6 content
would most likely be high (102). Whether naturally occurring
amounts of vitamin B-6 are low or conditions present during
food processing produce e-pyridoxyllysine, the antimetabolite

28
effects of e-pyridoxyllysine generally decreases the apparent
bioavailability of vitamin B-6 by competitive inhibition of PL
kinase (103) It is reasonable to surmise from the above
studies that a similar antagonistic effect of e-
pyridoxyllysine may have been responsible for the severe
deficiencies observed in infants with compromised vitamin B-6
nutriture and fed diets composed entirely of non fortified
canned infant formulas (60). The mechanisms of action for an
antimetabolite, such as e-pyridoxyllysine, includes inhibition
of any of the enzymes in the metabolic pathway to interconvert
the B-6 vitamers. Other mechanisms of vitamin B-6
antimetabolites are structural analogues of the vitamin and
inhibition of the active coenzyme function of PLP (96).
The complex chemical identity and nutritional properties
of other B-6 vitamer derivatives found in foods have not been
fully determined. The formation of the degradation product,
6-hydroxy-pyridine from the relatively stable B-6 vitamer PN,
has been reported in thermal treatment of various fruits and
vegetables (104) The generation of hydroxyl radicals
apparently mediated the hydroxylation of PN which was
associated with the oxidative degradation of ascorbic acid
(105). Gregory and Leatham (106) reported that 6-hydroxy-PN
has no vitamin B-6 or antivitamin B-6 activity. It appears
that hydroxylation of PN at the 6 position is not a
significant mechanism responsible for the loss of vitamin B-6
activity in foods.

29
A number of derivatives of PN-glucoside have been
isolated and identified. Tadera et al. (73,107) reported
several additional PN-glycosides to be minor components of the
total vitamin B-6 in pea seedlings and rice bran (108).
Assuming that PN-glucoside esters occur naturally and in
great enough quantities, it is unclear whether these glucosyl
compounds contribute significantly to the total vitamin B-6 in
foods and, if so how they may impact vitamin B-6 nutriture
when consumed. However, it is worth mentioning that PN-
glucoside has been the only glycosylated form of vitamin B-6
detected in a variety of foods utilizing HPLC methodology
(71,77). Tadera et al. (108) have observed the presence of a
vitamin B-6 conjugate termed "B-6X". The authors described
this compound as PN-glucoside esterified to an organic acid
based on its response in microbiological assays following
sequential alkaline treatment and hydrolysis with beta-
glucosidase (108) .
The nutritional properties of these conjugated forms of
vitamin B-6 are not fully understood at the present time. The
nutritional significance of these glycosylated vitamin B-6
compounds is related to their potentially limited
bioavialability. As earlier stated, PN-glucoside comprises 5%
to 80% of the total vitamin B-6 in many plant tissues (67,75).
Its bioavailability for humans is approximately 58% relative
to that of PN when administered orally (82) PN-glucoside has
been found to undergo intestinal absorption via hydrolysis of

30
the beta-glycosidic bond of PN-glucoside by mucosal beta-
glucosidase releasing free PN. Initial studies employing
bioassays with human subjects (40) indicated that the
proportion of the total vitamin B-6 which was present as PN-
glucoside correlated inversely with the bioavailability of the
vitamin in the foods examined (tuna, peanut butter, and whole
wheat bread). These results suggested that PN-glucoside would
not be biologically available to humans. However, studies by
Bills et al. (78) indicated that this observed correlation was
inconsistent when additional foods were examined and that the
quantity of PN-glucoside present in a food product could
function as a predictor of the bioavailability of vitamin B-6
from that food source. Stable-isotopic techniques have been
used to evaluate the bioavailability of purified deuterium-
labeled (d2) PN-glucoside in human subjects (82) In these
studies, the bioavailability of orally administered
extrinsically enriched oatmeal containing labeled PN-glucoside
was similar to that of labeled-PN fed under identical
conditions (82) Although metabolic utilization of PN-
glucoside appears to have a substantial level of individual
variability among human subjects, the bioavailability of PN-
glucoside in human subjects has been observed to be
substantially greater than that of the rat (82) Further
application of stable-isotopic techniques provide greater
clarification of the factors affecting bioavailability of
vitamin B-6.

31
The bioavailability of vitamin B-6 compounds is a
function of their extent of absorption and metabolic
utilization, as reflected by conversion to active coenzymatic
forms. It is important to assess the bioavailability of
vitamin B-6 as consumed in a typical mixed diet. The
assessment of vitamin B-6 status in populations is contingent
on identifying and guantifying the different forms of the
vitamin occurring naturally in foods consumed, determining the
potential interactions between these forms and the relative
bioavailability of the different vitamers, and their possible
interactions.
Our present knowledge of the essential role of vitamin
B-6 in preserving health emphasizes the importance of
establishing an accurate intake for vitamin B-6. The adeguacy
of intake is difficult to assess critically without accurate
information regarding the amount of the vitamin in foods
consumed, the effect of various vitamin B-6 compounds on the
enzymes involved in the metabolism of various forms of vitamin
B-6, as well as information regarding vitamin B-6
bioavailability.
Enzymatic Interconversion of B-6 Vitamers
There are several enzymatic reactions responsible for the
interconversion of the B-6 vitamers. The liver serves as the
center of vitamin B-6 metabolism and contains the enzymes
necessary for the metabolic interconversion of the B-6

32
compounds. Following intestinal absorption, the
dephosphorylated B-6 vitamers diffuse into the liver and are
converted to the coenzyme form of the vitamin, PLP (109-110).
A portion of the PLP is released into the circulation and
constitutes the major source of plasma PLP (111).
The dephosphorylated B-6 vitamers, PL, PN, and PM are
phosphorylated by the same kinase enzyme, pyridoxal kinase (EC
2.7.1.35) (PL kinase). The 51-phosphates of PN and PM are
subsequently oxidized by pyridoxamine (pyridoxine) 5'-
phosphate oxidase (EC 1.4.3.5) (PMP (PNP) oxidase) to yield
PLP. This coenzymatic form is bound to apoproteins,
transported to blood or dephosphorylated by phosphatases back
to PL (112). Further metabolism of PL results in
rephosphorylation by PL kinase or oxidation, by aldehyde
oxidase, to 4PA, the catabolic end product of vitamin B-6
metabolism. Vitamin B-6 in its coenzymatic form PLP or as
PLP-dependent enzymes, functions in many metabolic processes.
These include, amino acid metabolism (3), neurochemical
function (113-115), and modulation of hormone action (4-5).
Therefore, it is important that the cellular levels of PLP be
properly regulated.
As stated above, PL kinase is responsible for the
postabsorptive phosphorylation of PL, PN, and PM with
relatively equal efficiency. The result is the formation of
PLP from PL and upon further oxidation of PN and PM by PMP
(PNP) oxidase these two vitamers are also converted to the

33
coenzymatic form (109-110) The PLP that is not released into
circulation is either bound to cellular proteins or
dephosphorylated to PL. The vitamin B-6 taken up by the liver
is predominately maintained in the phosphorylated form (109-
110) .
There exists a cycle of phosphorylation/dephosphorylation
between PL and PLP which is catalyzed by PL kinase and
phosphatase enzymes (109). Under physiological conditions PL
kinase has approximately a 10 fold greater activity than
phosphatases (112). The degradative enzyme aldehyde oxidase
has an activity greater than the kinase and effectively
competes for PL. In order to maintain a steady state of
intracellular PLP a form of metabolic trapping occurs. The
coenzyme is bound to protein which protects it from hydrolysis
and hence, determines the relative amounts of PLP and PL and
creates a shift of equillibrum toward a balanced state (112).
Considering that PLP is required as a conenzyme for a
number of enzymes, it is not surprising that PL kinase is
widely distributed in tissues. This enzyme has been detected
in practically all tissues tested, including liver, kidney,
brain, and muscle (30). PL kinase has also been detected in
erythrocytes, however, the enzyme differs in its properties
compared with other tissue PL kinase (116). Since the liver
is the major site of vitamin B-6 metabolism it is generally
the tissue selected for the study of the enzymes in the B-6
vitamer interconversion pathway (116). It is important to

34
note that in comparative assessment of either of the enzymes
PL kinase or PMP (PNP) oxidase from different sources, the
levels and forms (apo or holo) of the enzyme is markedly
influenced by factors such as species, age, sex and dietary
status of the animal (110). Therefore, the following
discussion regarding the properties of these two enzymes will
be limited to rat liver as the source for the enzymes, unless
otherwise noted.
Hepatic PL kinase has a molecular weight of approximately
60,000 and a pH optimum between 5.5 and 6.0 (117). PL kinase
from all sources studied requires a divalent cation for
activity, with the greatest activity in liver PL kinase seen
with Zn2* (30) The enzyme also requires monovalent cations
with activation greatest in the presence of K* (117).
Although similar to PL kinase in tissue distribution and
molecular weight, PMP (PNP) oxidase differs significantly in
other physical and kinetic properties. The molecular weight
of oxidase is approximately 54,000, each having two identical
subunits of 27,000. PMP (PNP) oxidase occurs widely
distributed in both tissues and cells (118-119). Pogell (32)
was the first to describe PMP (PNP) oxidase as having a
requirement for a flavin cofactor in the oxygen-dependent
conversion of PMP to PLP. Later Wada and Snell (31) observed
that the same oxidase was responsible for the conversion of
both PNP and PMP to PLP. The investigators reported that the
partially purified enzyme required FMN as the specific

35
flavocoenzyme and demonstrated product inhibition as well as
inhibition by some phosphorylated analogs of vitamin B-6
compounds (31).
The relative reactivity of PMP(PNP) oxidase with varying
substrates is pH dependent (118). Within the physiological pH
range both PMP and PNP are suitable substrates for liver PMP
(PNP) oxidase. Although at slightly alkaline pH the V for
PNP is higher, the observed velocity is decreased due to
greater substrate inhibition by PNP even at low substrate
concentrations (112). The values demonstrate that at pH
7.0 PMP and PNP bind with similar affinity, whereas PMP binds
more tightly at the usual assay pH of 8.0 compared to PNP
(118) .
Many compounds have been indicated as inhibitors of the
metabolism of vitamin B-6. When assessing the site of action
of these antagonists it is reasonable to suspect that such
compounds may act on those enzymes responsible for the
formation of the active coenzyme form PLP. It is well
documented that liver PL kinase is inhibited by analogues
containing a 4-formyl group which resemble PL in their
affinity for PL kinase, condensation products formed from PL
and hydroxylamine, O-substituted hydroxylamines, hydrazine and
substituted hydrazines (120). PL kinase has also been shown
to be sensitive to inhibition by degradation products such as
pyridoxyllysine (121).

36
It is well known that the product PLP, inhibits the
activity of PMP(PNP) oxidase (112) As a result, other
phospho-B-6 analogues have been evaluated (118). It was
discovered that a dianionic charge on the 5'-position is
necessary for binding of the substrate analogue and
subsequently inhibition of the enzyme (122). Inhibition of
PMP (PNP) oxidase also occurs when alterations are introduced
in the structure of the flavin coenzyme, FMN (118). In as
much as both PL kinase and PMP (PNP) oxidase have substrate
reactivity with structural analogues, it is not unlikely that
conjugated forms of vitamin B-6 would inhibit these enzymes.
Conflicting data have been reported concerning the extent
of bioavailability of PN-glucoside in vitamin B-6 metabolism.
Most experimental results in rats have indicated 20-30% net
bioavailability of PN-glucoside relative to PN. The
bioavailability of PN-glucoside, although incomplete, is
substantially greater in humans than in rats. While the
incomplete utilization of PN-glucoside has been clearly
established in rats and humans, its potential interactive
effects in vitamin B-6 metabolism have not been examined
previously.
The purpose of this research was to demonstrate the
effects PN-glucoside has on the metabolic utilization of
vitamin B-6. The effort to evaluate these effects is an
important one, when considering the overall nutritional status
of individuals with respect to vitamin B-6. Being aware of

37
the multitude of roles vitamin B-6 plays in the body and its
impact on the overall health of individuals, it is imperative
that information be collected concerning how naturally
occurring forms of vitamin B-6 potentially interact with one
another.

CHAPTER 3
PYRIDOXINE-5 '-/3-D-GLUCOSIDE AFFECTS THE
METABOLIC UTILIZATION OF PYRIDOXINE IN RATS
Introduction
A conjugated form of pyridoxine (PN) was first isolated
from rice bran and identified as 5'-0-(/3-D-
glucopyranosyl)pyridoxine (PN-glucoside) (67). Analysis of a
variety of plant-derived foods by HPLC has shown that PN-
glucoside comprised 5-80 % of the total vitamin B-6 in many
fruits and vegetables (75). Similar results were obtained by
the use of a microbiological assay procedure for the
glycosidic conjugate of pyridoxine (65).
Conflicting data have been reported concerning the
bioavailability of PN-glucoside in vitamin B-6 metabolism.
Studies in rats have indicated that purified PN-glucoside is
relatively well absorbed but undergoes little metabolic
utilization and is rapidly excreted in intact form (77,80).
Most experimental results in rats have indicated 20-30 % net
bioavailability of PN-glucoside relative to PN (77,80-81). In
constrast, Tsuji et al. (79) observed nearly 100 %
bioavailability of PN-glucoside relative to PN in the rat. The
bioavailability of PN-glucoside, although incomplete, is
substantially greater in humans than in rats (82). The mean
38

39
bioavailability of orally administered PN-glucoside is
approximately 58% relative to PN. While the incomplete
utilization of PN-glucoside has been clearly established in
rats and humans, its potential interactive effects in vitamin
B-6 metabolism have not been examined previously.
Materials and Methods
Protocol
Two studies were conducted to evaluate the in vivo
utilization of [UC]PN in the presence of PN-glucoside. In
each of the studies eighteen male Sprague-Dawley rats (Crl:CD
(SD) BR) from Charles River Breeding Laboratories (Wilmington,
MA) weighing 200-300 g, were housed individually in stainless
steel metabolism cages in animal quarters maintained at 241C
with a 12-h light/dark cycle. All procedures for the care and
treatment of the experimental animals were in accordance with
the National Institutes of Health Guidelines.
The rats were fed ad libitum a casein-sucrose (20%:60%)
based diet, adequate in all micronutrients including 7 mg PN
HCl/kg*d for eight days (80) At the end of a seven day
acclimation period the rats were randomly assigned to one of
three treatment groups. On day eight of studies 1 and 2,
following a twelve hour fast, the rats were administered a
dose of 166.5 MBq (4.5 nCi) [UC]PN, equivalent to 240 nmol of
PN. In study 1 the rats received a simultaneous dose of
either 0, 36, or 72 nmol of unlabeled PN-glucoside (treatment

40
groups 1,2, and 3 respectively). In study 2 the rats received
a simultaneous dose of either 0 or 72 nmol unlabeled PN-
glucoside or 21.6 nmol unlabeled PN (treatment groups 1,2, and
3 respectively). The 21.6 nmol dose of PN was selected to
provide 30% of the PN-glucoside treatment, based on an assumed
30% net bioavailability of PN-glucoside relative to PN (4-6).
Doses were administered in 1.5 mL sterile H20 by gavage.
Urine was collected for the ensuing 24 h into foil-covered
flasks to avoid photochemical degradation of vitamin B-6
compounds. The urine collection funnels and flasks were
rinsed with water and rinses pooled with collected urine;
after dilution to 50 mL, urine was stored at -20 C until
analysis. The rats were killed by decapitation following
brief anesthesia, and livers were rapidly excised. The tissue
was divided into 2 g samples, then stored at -20 C until
analysis. All procedures were performed under GEF40G0 gold
lights that emit a wavelength between 500 nm and 750 nm to
prevent photochemical degradation of B-6 vitamers.
Forms of vitamin B-6
PN-glucoside was prepared by biological synthesis using
the propagation of alfalfa sprouts germinated in the presence
of PN-HC1 obtained commercially (Sigma Chemical Company, St.
Louis, MO) (80). The [4,5-uC]pyridoxine hydrochloride 684.5
KBq/nmol (18.5 Ci/mol) was a gift from Hoffmann-LaRoche

41
(Nutley, NJ) with a purity of greater than 98%, as determined
by ion-pair reverse phase HPLC.
Sample preparation
Urine samples were deproteinated by centrifugal
ultrafiltration with micropartition tubes (YMT membrane
filters; Amicon, Danvers, MA) (80) Urinary 4PA analysis was
by reverse-phase HPLC, as described below. Fractions (0.5 mL)
were collected by using an ISCO Cygnet Fraction Collector
(ISCO, Lincoln, NB) Each filtered sample was decolorized
(81) and an aliquot (100 /iL) was then counted for total
radioactivity. Liver tissue was minced and homogenized in 6
mL of 4.3 mol/L trichloroacetic acid with a Polytron
Homogenizer (Brinkman Instruments, Westburg, NY) and then
centrifuged for 20 min. at 12,000 x g (6). The supernatant
was partitioned against an equal volume of diethyl ether to
extract the trichloroacetic acid and filtered with 0.45 un
membrane filters (Gelman Sciences, Inc., Ann Arbor, MI). The
sample was aerated with nitrogen gas, then analyzed for
vitamin B-6 by HPLC as described below, and an aliquot of the
supernatant from each sample was analyzed for radioactivity.
Decolorization of the extract was performed as above prior to
liquid scintillation spectrometry.

42
HPLC equipment and analysis
The separation of radiolabeled B-6 compounds in the urine
and liver tissue was accomplished by ion-pair reverse-phase
HPLC (82). Chromatographic analysis was performed with a
Rainin HP/HPX Drive Module (Rainin Instrument, Woburn, MA).
All of the above analysis utilized a loop injection valve
(Model 904-2,Altex), a fluorometric detector (Model LS-5,
Perkin-Elmer, Norwalk, CT) and an electronic integrator (Model
3388A, Hewlett-Packard, Avondale, PA). Excitation and
emission wavelengths were 295nm (5nm slit width) and 405nm
(5nm slit width) respectively. Two mobile phases were employed
in a gradient elution procedure using an Ultrasphere IP 5 /m
C-18 4.6 mm x 25 cm column (Beckman Instruments, San Ramon,
CA) Mobile phase A contained 0.03 3 mol/L phosphoric acid and
8 mmol/L 1-octanesulfonic acid, adjusted to pH 2.2 with 6
mol/L KOH. Mobile phase B contained 0.033 mol/L phosphoric
acid and 3.4 mol/L acetonitrile, adjusted to pH 2.2 with 6
mol/L KOH and no ion-pairing reagent. Fractions (0.5 mL) were
collected using an ISCO Cygnet Fraction Collector (ISCO,
Lincoln, NB) The identity of the PLP and PNP peaks was
confirmed by monitoring the formation of pyridoxal (PL) and
PN, respectively, by HPLC following incubation of the PLP and
PNP fractions in 0.1 mol/L HC1 at 80 C for 3 h.

43
Measurement of radioactivity
HPLC fractions and extracts of tissue were measured for
radioactivity using a commercial scintillation fluid
(ScintiVerse LC, Fisher Scientific, Orlando, FL) and a liquid
scintillation spectrophotometer (Beckman LS 2800 Beckman
Instruments, San Ramon, CA) A quench curve for UC was used
for conversion of cpm to dpm.
Statistical analysis
In experiment 1 and 2 the distribution of vitamin B-6
compounds and total radioactivity between groups was compared
by the method of least squares analysis of variance using
general linear model procedures of SAS (123) following a log
tranformation of the data to normalize variance (123). Data
are reported as least squares mean the pooled standard error
of the least squares mean (SEM). In experiment 1, orthogonal
contrasts were made to examine the linear and quadratic
effects of different doses of PN-glucoside on dependent
variables. The quadratic effects were not significant for
any dependent variable. Values with p<0.05 were considered
statistically significant and were all for linear effects of
PN-glucoside unless otherwise stated. In expreriment 2 two
orthogonal contrasts were used to compare treatment means.
Contrast one compared group 2 with group 1 and 3; contrast two
compared group 1 with group 3.

44
Table 3.1 Liver B-6 vitamer distribution and total liver 14C
in rats administered varying levels of PN-glucoside in the
dose (Experiment 1).*
LEVEL OF PN-GLUCOSIDE IN THE DOSE
Vitamers
0 nmol
36 nmol 72 nmol
SEM
p<0
PLP
17.5
% liver
37.3
radioactivitv
51.6 7.5
#
PNP
0.0
3.1
16.4
3.4
#
PMP
46.7
36.1
32.0
8.0
#
PL
16.9
14.5
0.0
4.8
+
PN
15.2
7.8
0.0
3.5
+
PM
3.6
1.2
0.0
0.6
+
Total UC
20.0
%
17.0
of dose
12.3
4.0
+
Values based on n=6 are means and pooled standard error
of the mean (SEM). Log transformations were performed on
data prior to analysis of variance. The quadratic effects
were not significant for any dependant variable.
# Significant linear effect, p<0.05.
+ Significant decreased linear effect, p<0.05.
Results
These experiments were designed to investigate the effect
of unlabeled PN-glucoside on the metabolism, in vivo
retention, and excretion processes of simultaneously
administered [UC]PN. The distribution of B-6 vitamers in
liver tissue, total urinary [UC]4PA, and total urinary UC
were determined to evaluate the utilization of [UC]PN in the
presence of PN-glucoside. Data were expressed as a percentage
of total radioactivity administered in the dose or as a
percentage of total [UC] B-6 in the tissue.

45
Table 3.2 Urinary 4 PA and total urinary UC in rats
administered varying levels of PN-glucoside (Experiment 1)*
0 nmol 36 nmol 72 nmol
PN-glucoside PN-glucoside PN-glucoside SEM p<0.05
% urinary radioactivity
4PA 37.6 12.7 10.9 5.1 +
% of dose
14C 26.9 35.1 37.3 12.3 #
Values based on n=6 are means and pooled standard error
of the mean (SEM) Log transformation of the data was
performed prior to the analysis of variance. The
quadratic effects were not significant for any dependant
variable.
* Significant linear effect, p<0.05.
+ Significant decreased linear effect, p<0.05.
The results of study 1 indicated statistically
significant differences (p<0.05) in the quantity and
distribution of labeled vitamin B-6 compounds in liver tissue
and urine among groups fed varying levels of PN-glucoside
(Tables 3.1 and 3.2). A significant linear relationship was
observed between phosphorylated vitamin B-6 compounds
pyridoxal 5'-phosphate (PLP) and pyridoxine 5'-phosphate (PNP)
in liver tissue and the amount of administered PN-glucoside.
Total hepatic 14C decreased linearly (p<0.05) with varying
amounts of PN-glucoside (Table 3.1). The proportion of
hepatic nonphosphorylated 14C-labeled PN, PL and pyridoxamine
(PM) decreased linearly with the amount of PN-glucoside (Table
3.1) .
Urinary excretion of total 14C was showed a significant
linear effect (p<0.05) among groups receiving varying doses of

46
Table 3.3 Liver B-6 vitamer distribution and total liver UC
in rats administered varying levels of PN or PN-glucoside.
(Experiment 2)*
LEVELS OF
PN-GLUCOSIDE
LEVELS
OF PN
Vitamers
0 nmol
Group 1
72 nmol
Group 2
21.6 nmol
Group 3
SEM
PLP
%
31.5
liver radioactivitv
58.4a 31.5
3.4
PNP
0.0
14.4a
0.0
0.1
PMP
41.4
27.2a
40.1
1.0
PL
13.3
0.0a
14.3
0.4
PN
6.3
0.0a
6.0
0.2
PM
7.5
0.0a
8.1
0.4
Total UC
15.1
% of dose
10.0b
15.0
0.0
Values based on n=6 are means and pooled standard error
of the means (SEM) Log transformation of the data was
performed prior to analysis of variance. The quadratic
effects were not significant for any dependant variable.
a Significantly different at p<0.05 for linear effects of
PN-glucoside in the dose.
b Significantly different at p<0.05 for decreased linear
effect.
PN-glucoside (Table 3.2). In addition, the concentration of
urinary [UC]4PA, relative to total urinary 14C, and total
urinary [UC]4PA decreased linearly with the amount of
administered PN-glucoside (Table 3.2).
Experiment 2 was conducted to determine if the results of
experiment 1 were due to a specific effect of PN-glucoside or
simply to an increase in the total amount of available PN
(derived from PN-glucoside). The experimental design was the
same, except on day eight, following a twelve hour fast, the
rats were administered a simultaneous dose of 166.5 MBq (4.5

47
iCi) [UC]PN, while treatment groups 1 and 2 received 0 nmol
and 72 nmol unlabeled PN-glucoside, respectively. Treatment
group 3 received 21.6 nmol unlabeled PN based on an assumed
30% net bioavailability of PN-glucoside relative to PN (77,80-
81) Statistical analysis by orthogonal contrasts of the data
from this study indicated no statistically significant linear
effects between the quantity or distribution of labeled
vitamin B-6 compounds in liver tissue and urine and treatment
groups one and three (Table 3.3). However, there was a
significant linear effect (p<0.05) in the relative
concentration of phosphorylated vitamin B-6 compounds PLP,
Table 3.4 Urinary 4 PA and total urinary UC in rats
administered varying levels of PN or PN-glucoside.
(Experiment 2)*
LEVELS OF PN-GLUCOSIDE LEVELS OF PN
0 nmol
72 nmol
21.6 nmol
SEM
% urinarv
radioactivitv
4 PA
35.5
14.2b
32.7
12.3
UC
%
of dose
30.0
46.0a
27.0
4.0
Values based on n=6 are means and pooled standard error
of the mean (SEM) Log transformation of the data was
performed prior to analysis of variance. The quadratic
effects were not significant for any dependant variable.
a Significantly different at p<0.05 for linear effects of
PN-glucoside in the dose.
b Significantly different at p<0.05 for decreased linear
effects.

48
PNP, and pyridoxamine 51-phosphate (PMP) in liver tissue,
while hepatic 14C decreased linearly (p<0.05) (Table 3.3). The
proportion of hepatic nonphosphorylated forms of [14C]PN, PL
and PM was significantly less among treatment group 2
(p<0.05) than groups 1 and 3. Urinary excretion of total 14C
also was significantly greater (p<0.05) in group two relative
to groups one and three (Table 3.4). In addition, the
concentration of urinary 4PA, relative to total urinary 14C,
was significantly lower (p<0.05) in group two as compared to
groups 1 and 3 (Table 3.4). The results of experiment 2
indicate that the effects of PN-glucoside observed in
experiment 1 were specific for this vitamin B-6 derivative and
were not due to isotope dilution effects of the unlabeled
compound administered.
Discussion
The focus of this research was to determine the direct
effects of PN-glucoside on [14C]PN administered in a
simultaneous dose, similar to what would occur when consuming
a mixed diet. The ratios of PN-glucoside to nonglycosylated
vitamin B-6 (14C-labeled plus dietary) were in the range of 8%
to 16%, which is consistent with observations of typical human
diets (82) Since our interest was in determining the fate of
labeled-vitamin B-6 compounds in the presence of PN-glucoside
the data were expressed as percent of the total radioactivity
in the dose. This enabled the determination of the direct

49
effect of PN-glucoside on the metabolic utilization of [UC]PN
and distribution of the 14C-labeled B-6 metabolites in the
liver 24 h post-dose when the two vitamers are consumed
together. Expression of the data in this manner reflects the
direct effect of PN-glucoside on the distribution of B-6
vitamers and urinary excretion of 4PA as well as interactions
between different naturally occurring forms of vitamin B-6
found in foods consumed simultaneously in a typical mixed
diet. The bioavailability of vitamin B-6 compounds is a
function of their extent of absorption and metabolic
conversion to active coenzymatic forms. Pyridoxine that is
absorbed from the intestinal tract is concentrated initially
in the liver (124) and is the sole source of plasma PLP (125).
Thus, liver plays a central role in the overall metabolism of
vitamin B-6 (126).
The major transformations in hepatic vitamin B-6
metabolism involve phosphorylation catalyzed by pyridoxal
kinase, oxidation of PNP and PMP by pyridoxine (pyridoxamine)
5'-phosphate oxidase (PNP oxidase), along with interconversion
of PLP and PMP through transamination reactions (109-110,127-
129). The principal forms of vitamin B-6 in liver are PLP and
PMP (129-131). PNP is usually present in only trace
quantities because of its rapid oxidation to PLP (130,132-
133). The non-phosphorylated B-6 vitamers constitute less
than 10% of the total vitamin B-6 content in the liver (131).
In this study, the concentration of [UC]PNP increased linearly

50
with the quantity of unlabeled PN-glucoside administered
(p<0.01). This suggests a mechanism involving a possible
inhibition of PNP oxidase. Investigations of possible direct
inhibitory effects of PN-glucoside on PNP oxidase are in
progress. Comparatively high levels (36%) of the non-
phosphorylated B-6 vitamers were observed in the livers of the
rats in group 1, which may have been due to several factors.
The vitamin B-6 requirement for the rat is 6-7 mg PN/kg diet
(135). The experimental diet contained 7 mg PN/kg diet with
the rats consuming, on the average, 70 vitamin B-6 per day.
The dose of labeled-PN contained 240 nmol (49.3 /xg) [UC]PN.
Hence, during the 24 h post dose period, rats received 49.3 fig
vitamin B-6 above their requirement. In addition, a single
observation of tracer distribution at a time point 24 h post
dose, would not necessary reflect steady state concentrations
of endogenous vitamin B-6 in the liver. These factors would
account for greater than normal levels of B-6 metabolites in
liver.
The compensating changes observed in the proportions of
hepatic [UC]PMP and [UC]PLP (Table 3.1) cannot be readily
explained. Unexpectedly, these changes are consistent with
those observed when comparing the metabolism of [UC]PN in
vitamin B-6 adequate verses deficient rats (81). These
results indicate that the administration of unlabeled PN-
glucoside in this study caused changes in metabolic patterns

51
which paralleled, in one respect, those seen in vitamin B-6
deficiency.
The metabolic pathway for the degradation of PLP involves
enzymatic hydrolysis of the phosphate ester bond, and the
oxidation of PL to 4PA (132-133). As a terminal product of
vitamin B-6 metabolism, urinary 4PA reflects the in vivo
metabolic utilization of the vitamin. In these studies, the
extent of conversion of administered [UC]PN to [UC]4PA is
indicative of reduced utilization of [14C]PN which is inversly
related to the dose of PN-glucoside. This study also shows a
negative correlation between percent radioactivity in the
urine as [UC]4PA and the quantity of PN-glucoside
administered. The data demonstrate that total liver 14C was
inversly related to the PN-glucoside dose while total urinary
14C was directly proportional to the amount of PN-glucoside
administered.
The intracellular concentration of PLP in liver tissue is
tightly regulated which does not permit the excess
accumulation of this coenzyme (133). In the liver, newly
synthesized PLP and PMP are contained in compartments that
have a rapid rate of turnover (132). These small and rapidly
mobilized pools are poorly miscible with the endogenous
coenzyme pools (132). The newly synthesized [UC]PLP and
[14C]PMP in the liver would be rapidly metabolized by
hepatocytes and degraded to [14C]4PA and excreted in the urine,
or transported to the circulation as [14C]PL, and [14C]PLP.

52
These B-6 vitamers could then be taken-up by erythrocytes
and/or tissues. This would account for the observed results
of liver UC and urinary UC in response to PN-glucoside dose
seen in this study. When rats were given a quantity of
unlabeled PN equivalent of 30% of the PN-glucoside dose, the
effect on the metabolic utilization and distribution of other
forms of vitamin B-6 differed significantly from the group
administered the PN-glucoside dose.
In previous studies in this laboratory, the absorption of
PN-glucoside was 50% relative to PN in rats fed a diet
adequate in vitamin B-6 when administered as an alginate gel
(77) but was nearly equivalent to PN when given in solution
(81). Analysis of urine showed that most of the absorbed PN-
glucoside was excreted in the intact form, suggesting low
bioavailability (4,6). Previous findings by Trumbo et al.
(80) showed poor utilization of PN-glucoside relative to PN on
the basis of growth and plasma PLP concentration, which
indicated that PN-glucoside has a low biological availability
as vitamin B-6 in the rat. However, PN derived from
hydrolysis of PN-glucoside, can enter into vitamin B-6
metabolic pathways of the liver to produce other forms of
vitamin B-6 (77,81).
Kabir et al. (65) observed with human subjects, an
inverse relationship between the percentage PN-glucoside (of
total vitamin B-6) and overall vitamiln B-6 bioavailability in
food. The investigators suggested that the proportion of the

53
glycosylated vitamin B-6 in the diet may be useful as an index
of vitamin B-6 bioavailability. The nutritional significance
of the effect of PN-glucoside on the metabolic utilization of
other forms of vitamin B-6 is based on the fact that PN-
glucoside is a major naturally occurring form of vitamin B-6
in many fruits and vegetables of human diets (65,75). The
results of the present studies suggest that PN-glucoside is
nutritionally significant in two ways. As shown previously,
PN-glucoside can act as a source of partially available
vitamin B-6 (77,81-82). In addition, this study has shown
that PN-glucoside can act as a weak antagonist that may hinder
the utilization of PN and possibly other forms of the vitamin.
These results indicate that PN-glucoside alters the
metabolism and in vivo retention of [UC]PN in the rat and that
PN-glucoside may retard the utilization of nonglycosylated
forms of vitamin B-6. PN-glucoside represents a significant
proportion of the total vitamin B-6 in many plant derived
foods (75) Therefore, the results of this study are
important in assessing the bioavailability of vitamin B-6 in
a typical mixed diet. The assessment of vitamin B-6 status in
human populations is contingent on identifying and quantifying
the content and distribution of the different forms of the
vitamin found in foods, determining the potential interactions
between these forms and the relative bioavailability of the
different vitamers. If it is found that PN-glucoside affects
the metabolic utilization of PN similarly in humans, it is

54
conceivable that the vitamin B-6 status in humans would not be
accurately reflected by current food consumption data.
Studies evaluating the metabolic interactions of PN-glucoside
on PN in humans are currently in progess.

CHAPTER 4
EFFECTS OF PYRIDOXINE-5' -j3-D-GLUCOSIDE ON THE
METABOLIC UTILIZATION OF PYRIDOXINE IN HUMANS
Introduction
A conjugated form of vitamin B-6, first isolated from
rice bran was identified as 5'-0-(j8-D-
glucopyranosyl)pyridoxine (PN-glucoside) (67). Analysis of a
variety of plant-derived foods indicated that PN-glucoside is
abundant in many fruits and vegetables and the only
significant glycosylated form of vitamin B-6 present in the
foods examined (65,71,75). Assessing the nutritional status
of a person with respect to vitamin B-6 is a function of the
amount and form of vitamin B-6 in food, the bioavailability,
which is the extent of intestinal absorption and metabolic
utilization of the B-6 compounds in the diet and the specific
requirement of the individual.
Previous studies concerning the bioavailability of
vitamin B-6 in rats indicated that purified PN-glucoside is
relatively well absorbed compared to PN, undergoes little
metabolic utilization, and is rapidly excreted (2,77,80-81).
These results indicated 20%-30% net bioavailability of PN-
glucoside relative to PN. The extent of in vivo hydrolysis of
the glycosidic bond, rather than intestinal absorption, was
55

56
the limiting step in the utilization of PN-glucoside as
vitamin B-6 in the rat (2,77,80). A recent study in our
laboratory has shown that PN-glucoside alters the metabolism
and in vivo retention of simultaneously administered PN in the
rat (136).
A stable-isotopic study in humans found the
bioavailability of PN-glucoside to be 58% of that of free PN,
which is greater than that found in the rat (82) In an
investigation of lactating women, Andn et al. (84) reported
that dietary PN-glucoside contributed little to vitamin B-6
nutriture. These findings indicate the need for additional
data concerning factors affecting the utilization of PN-
glucoside and the potential effect of PN-glucoside on the
metabolic utilization of vitamin B-6 by humans consuming a
mixed diet. The study reported here was conducted to
determine the effect of PN-glucoside on the metabolic
utilization of PN in humans through the use of stable-isotopic
methods.
Materials and Methods
Synthesis of forms of vitamin B-6
The deuterium-labeled form of vitamin B-6 used in this
study was [51-C2H2OH]pyridoxine (d2PN) which was prepared in
our laboratory as described by Coburn et al. (108). Mass
spectral analysis indicated that the d2PN species comprised
66% of the d2PN preparation, with a majority of the remainder

57
in the monodeutero (d^ form. All procedures were performed
under yellow-light to prevent photochemical degradation of B-6
vitamers.
PN-glucoside was prepared from PN-HCl using biological
synthesis as previously described (80). Twelve grams of
alfalfa seeds were germinated in the presence of 62 mg of PN-
HCl (297 /mol) obtained commercially (Sigma Chemical Company,
St. Louis, MO) followed by purification by cation-exchange
(80) and gel filtration chromatography on Sephadex G-10
(Pharmacia). The yield of purified PN-glucoside was 237 /mol
(80%). Chromatographic analysis of the purified PN-glucoside
preparation confirmed the absence of free (nonglycosylated) PN
and other forms of vitamin B-6.
Protocols of trials with human subjects
The study was conducted using the same adult men (n=6;
22-32 y) in each of three trials (Table 4.1). All subjects
were in good health and exhibited normal blood chemistry and
hematological values. The subjects were in normal vitamin B-6
status as determined by plasma pyridoxal phosphate (PLP) and
urinary 4-pyridoxic acid (4PA) concentration, erythrocyte
aspartate aminotransferase (AspAT) activity, and erythrocyte
AspAT stimulation by in vitro addition of PLP, as judged by
previously published criteria (137-138). The procedures for
selection of subjects and the experimental protocol were

58
approved by the University of Florida Institutional Review
Board and an informed consent was obtained from each subject.
Three trials were conducted to compare the in vivo
metabolic utilization of d2PN using the same group of six
subjects. The experimental period lasted seven weeks
consisting of three trials, each lasting one week with a two
week wash out period between each trial. A self-selected diet
was consumed for the duration of the study.
On the first day of each trial the subjects collected a
24-h urine sample into foil-covered polyethylene containers
and kept it refrigerated during the collection period. After
an overnight fast, a five ml blood sample was collected into
a Vacutainer brand tube (Becton Dickinson, Rutherford, NJ)
containing ethylenediaminetetraacetic acid (EDTA). Following
these sample collections the subjects consumed a single oral
dose of 5 /mol d2PN and either 2/mol PN or 1/mol PN + 1/mol
PN-glucoside or 2/mol PN-glucoside (trials 1, 2, and 3
respectively) (Table 4.2) contained in 250 ml of apple juice.
The doses of unlabeled PN and PN-glucoside were selected to
provide and equal molar amount (2/mol) of either PN, PN+PN-
glucoside, or PN-glucoside. Urine collection was continued
for the following 48-h period. Urine was analyzed for
creatinine, total 4PA, and d24PA. The blood samples were
centrifuged and the plasma collected was analyzed for plasma
PLP and the erythrocytes were used to determine hemoglobin
concentration and AspAT activity.

59
Analytical methods
Urinary 4PA and plasma PLP were determined by reverse-
phase HPLC procedures with fluorometric detection (48,109).
Erythrocyte AspAT activity was measured by a
spectrophotometric assay procedure (138) with commercially
available reagents (Sigma Chemical, St. Louis, MO). Urinary
creatinine was determined based on the method of Heinegard and
Tiderstrom (139), while hemoglobin in erythrocyte hemolysates
was determined spectrophotometrically as the
cyanomethemoglobin derivative (140).
Mass spectral analysis of deuterium-labeled 4PA
Urinary d24PA was determined by gas chromatography-mass
spectrometry (GCMS) following isolation of 4PA from urine
samples by cation exchange chromatography (Bio-Rad AG 50W-X8,
100-200 mesh, H+ form) and reverse-phase HPLC (Whatman,
Partisil 10 ODS-3 Magnum 9 column, 9mm i.d. x 25 cm) as
described by Gregory et al.(82). GCMS was performed in the
electron capture negative ionization mode (Model 4500 GCMS
system, Finnigan MAT, San Jose, CA) with a DB-5 capillary gas
chromatographic column (J&W Scientific, Folsom, CA) and
methane as a reagent gas. The derivatization (141) was
performed by dissolving the dried 4PA sample in 0.5 ml of 1:1
(v/v) solution of pyridine:acetic anhydride, heated at 100 C
for 90 min to form the 3-acetyl-4PA-lactone which is
evaporated to dryness under a stream of nitrogen. The 3-

60
acetyl-4PA-lactone was dissolved in 50 /I ethyl acetate and 2
IjlI of the resulting solution injected into the GCMS injection
port. As this 4PA derivative eluted from the GC column,
electron capture negative ionization yielded two anions at m/z
207 and 209 for the dQ and d2 4 PA species, respectively.
Selected-ion monitoring at m/z 207 and 209 permitted
simultaneous measurement of the dQ and d2 4PA species during
GCMS analysis. A series of 4PA standards of known molar
ratios of dQ and d2 species were prepared and analyzed to
facilitate quantitative analysis. Calibration curves were
constructed to relate the ratios of observed GCMS peak areas
to actual molar isotope ratios of the d2/dQ 4PA species. The
urinary excretion of d24PA was calculated using the isotope
ratio determined by GCMS, the concentration of total urinary
4PA as determined by reverse-phase HPLC, and total urine
volume.
Statistical analysis
The urinary excretion of d24PA was expressed as a
percentage of the administered dose of labeled PN. The
excretion of d24PA (as a percentage of administered dose),
total urinary 4PA, and vitamin B-6 nutritional status
parameters (see below) was compared by the method of least
squares analysis of variance using general linear model
procedures of SAS (123), following a log transformation of the
data to normalize variance when necessary. Data for d24PA and

61
total urinary 4PA are reported as least squares mean the
pooled standard error of the least squares mean (SEM).
Orthogonal contrasts were made to examine the linear and
quadratic effects of different doses of PN-glucoside on
dependent variables. The quadratic effects were not
significant for any dependent variable. Values with p<0.05,
unless otherwise stated, were considered statistically
significant and were all for linear effects of the amount of
PN-glucoside (123).
Results
Vitamin B-6 nutritional status of subjects
The principal criteria used for assessing the vitamin B-6
status of the subjects were plasma PLP, urinary 4PA excretion
per 24-h, and erythrocyte aspartate aminotransferase
stimulation by in vitro addition of PLP. As shown in Table
4.1, mean values for these subjects were within the proposed
guidelines for vitamin B-6 nutritional adequacy (137-138).
These data indicate the adequate vitamin B-6 status of these
subjects prior to the administration of labeled vitamin B-6 in
each trial, which reflects the adequacy of their self-selected
diets prior to and between experiments. Subjects demonstrated
no statistical significant changes between trials for these
vitamin B-6 nutritional status parameters.

62
Table 4.1 Indicator of vitamin B-6 nutritional status of
human subjects 24 h prior to each trial1.
TRIAL URINARY 4PA PLASMA PLP ERYTHROCYTE
/xmol/24 H nmol/L AspAT
STIMULATION
BY PLP
%
TRIAL 1 10.81.9 86.97.4 35.84.5
(5/xmol d2PN
+ 2/xmol PN)
(n=6)
TRIAL 2 8.01.8 82.410.3 36.46.3
(5/xmol d2PN +
1/xmol PN
+ 1/xmol PN-GLUCOSIDE)
(n=6)
TRIAL 3 7.410.7 83.919.8 16.715.9
(5/xmol d2PN +
2/xmol PN-GLUCOSIDE)
(n=6)
1Mean and SEM, n=6; Blood and urine collections were made
24 h prior to administration of labeled forms of vitamin
B6. Proposed guidelines for adequacy of vitamin B6 status
are: urinary 4PA excretion, >5/xmol/24 h (ref. 139) ; Plasma
PLP, >50 nmol/1 (ref. 139) ; And in vitro stimulation of
erythrocyte aspartate aminotransferase (AspAT) by added
PLP, < 50% (ref. 140).
Stable-isotopic trials
This study was conducted to determine the effect of PN-
glucoside on the metabolic utilization of d2PN by evaluating
the excretion of d24PA following the ingestion of an oral dose
of d2PN while increasing the dose of PN-glucoside across the
three trials. The detection of urinary d24PA was conclusive

63
Table 4.2 Molar isotopic ratio of urinary d24PA.1
EXPERIMENTAL TRIALS
SUBJECT
TRIAL 1
(5/mol d2PN +
2/xmol PN)
(n=6)
TRIAL 2
(5/mol d2PN +
1/mol PN + 1/mol
PN-GLUCOSIDE)
(n=6)
TRIAL 3
(5/mol d2PN
2/mol PN-
GLUCOSIDE)
(n=6)
1
0.2 0.001
d2/d0
0-24 hr
0.091.002
0.071.010
2
0.07.001
0.101.001
0.121.003
3
0.07.003
0.081.003
0.091.001
4
0.17.002
0.151.010
0.101.001
5
0.131.001
0.011.001
0.071.001
6
0.101.001
0.081.001
0.121.002
mean
0.121.021
0.091.018
0.091.009
1
0.051.000
d2/d0
24-48 hr
0.101.001
0.061.000
2
0.091.001
0.061.000
0.031.000
3
0.101.000
0.101.000
0.031.000
4
0.071.001
0.101.000
0.041.001
5
0.091.000
0.101.000
0.061.000
6
0.051.000
0.191.001
0.071.000
mean
0.071.007
0.111.017
0.051.006
1Mean and SEM, n=6.
Values are average of three injections per sample.
evidence of the utilization of d2PN in vitamin B-6 metabolism.
Deuterium-labeled 4PA was detected in urine over 48-h post
dose. The d24PA comprised a small portion of the total
urinary 4PA using this experimental design. However, the
isotope ratio observed was measurable by the GCMS method
employed. Twenty-four hours following the administration of

64
Table 4.3 Percentage of d2PN (5/mol) dose excreted as urinary
d24PA.
EXPERIMENTAL TRIALS
TIME TRIAL 1 TRIAL 2
TRIAL 3
(H) (5/mol d2PN (5/mol d2PN + (5/mol d2PN
+ 2/mol PN) l/mol PN + l/mol 2/mol PN-
(n=6) PN-GLUCOSIDE) GLUCOSIDE)
(n=6) (n=6)
(/mol)
(/mol)
(/mol)
(SEM)
0-24
46.5
27.2
24.3
12.7
24-48
37.1
28.6
17.0
12.6
0-48
83.6
55.8
41.3
17.7
1Values based on n=6 are means and pooled standard error of
the mean (SEM) The quadratic effects were not
significant for any dependent variables. The values in
table are significantly different at p<0.05 for linear
effects of the amount of PN-glucoside.
the isotopically labeled d2PN, the mean d2/dQ molar isotopic
ratio of labeled 4PA was 0.12 0.02, 0.09 0.02, and 0.09
0.01 (mean SEM) for trial 1, 2, and 3 respectively (Table
4.2). These were not significantly different. The 24-48 hr
molar isotopic ratios are also not significantly different
(Table 4.2).
The group mean values for excretion of d24PA as a
percentage of the administered d2PN dose between trials 1 and
3 for the 24-h post-dose period decreased linearly with the
amount of PN-glucoside (Table 4.3). Total 4PA excreted over
a 24-h and 48-h post-dose period showed group mean values for

65
Table 4.4 Total urinary 4PA excretion.1
EXPERIMENTAL TRIALS
TIME
(H)
TRIAL 1
(5/mol d2PN
+ 2/mol PN)
(n=6)
TRIAL 2
(5/mol d2PN +
1/zmol PN + 1/tmol
PN-GLUCOSIDE)
(n=6)
TRIAL 3
(5/mol d2PN
2/mol PN-
GLUCOSIDE)
(n=6)
(/mol)
(/mol)
(/mol)
(SEM)
0-24
26.0
13.5
10.3
3.9
24-48
15.5
10.5
11.5
4.82
0-48
41.5
24.0
21.8
7.63
Values based on n=6 are means and pooled standard error of
the mean (SEM) The quadratic effects were not
significant for any dependent variables. The values in
table are significantly different at p<0.05 for linear
effects of the amount of PN-glucoside.
2 p<0.5
3 p<0.08
trials 1 and 3 to decrease linearly with the amount of PN-
glucoside (Table 4.4). These results indicate that the
percentage of d2PN dose excreted as d24PA showed an inverse
relationship in proportion to the PN-glucoside dose and that
total 4PA decreased linearly with the amount of PN-glucoside.
Discussion
Because PN-glucoside is a major form of vitamin B-6 in
plant derived foods, its influence on the utilization of

66
nonglycosylated species of the vitamin is nutritionally
significant. These effects, which may be minor in a mixed
diet containing both animal and plant derived foods, may be
much more pronounced in vegetarian diets. Kabir et al. (65)
showed an inverse relationship between the content of PN-
glucoside in a food and the net bioavailability in humans.
Our present knowledge of the essential role of vitamin B-
6 in preserving health emphasizes the importance of
establishing an accurate Recommended Dietary Allowance (RDA)
for vitamin B-6. A vitamin B-6 deficiency with apparent
clinical symptoms is rare in the general population, but the
vitamin B-6 requirement has not been clearly determined. It
is significant that over 50% of those evaluated in the 1977-78
USDA Food Consumption Survey consumed only 70% of the 1980 RDA
for vitamin B-6. Females and the elderly were especially
prevalent in the group consuming only 70% of the RDA (2) The
1985 Continuing Survey of Food Intakes of Individuals
indicated that only 27% of women consumed 70% or more of the
RDA for vitamin B-6 (143-144). Although the range of intakes
is large, substantial portions of the population consume
apparently marginally adequate amounts of vitamin B-6.
Adequate vitamin B-6 nutriture is essential to health through
the multiplicity of roles of its active coenzymatic form
pyridoxal 5-phosphate (PLP).
The catabolic pathway for PLP involves enzymatic
hydrolysis of the phosphate ester bond, and the subsequent

67
oxidation of pyridoxal (PL) to 4PA (130-132). As a terminal
metabolite of vitamin B-6 metabolism, urinary 4PA reflects the
in vivo metabolic utilization of the vitamin. The deuterated
form of PN used in this study is labeled at a metabolically
stable position (i.e., 5'-CH2OH) and, therefore, retains the
label throughout absorption, metabolism, transport, and
excretion (82).
The results of this study demonstrate a negative
correlation between the quantity of PN-glucoside administered
and the amount of urinary d24PA derived from oral d2PN. The
data also showed that an inverse relationship between the
amount of total urinary 4PA excreted and the amount of PN-
glucoside in the dose. These findings are consistent with
those observed when comparing the metabolism of [UC]PN in the
presence of increasing doses of PN-glucoside in the rat (136).
The PN derived from hydrolysis of PN-glucoside enters into
vitamin B-6 metabolism to produce other forms of vitamin B-6,
including 4PA (77,81). However, the PN-glucoside that is not
hydrolyzed to PN may not be totally inert but instead appears
to affect the metabolism and interconversion of the B-6
vitamers. The possible inhibition by PN-glucoside of PL
kinase and/or PNP-oxidase, two enzymes in the metabolic
interconversion of B-6 vitamers, is discussed in chapter 5.
The results of these isotopic studies provide evidence that
PN-glucoside alters the metabolism of PN in humans and rats by

68
partially retarding the utilization of nonglycosylated B-6
compounds.
The appropriateness of the RDA is difficult, if not
impossible to assess without accurate information concerning
the content and forms of B-6 compounds in foods consumed, and
accurate data regarding vitamin B-6 bioavailability. This
research provides evidence of interactions amoung various B-6
compounds found in a typical mixed diet through an influence
of PN-glucoside on the utilization of PN. Such an interaction
represents another factor involved in the bioavailability of
vitamin B-6 and must be considered in fully understanding the
adequacy of vitamin B-6 intake. Future research will address
the nutritional significances of these interactions amoung the
vitamin B-6 compounds and the importance of these compounds in
assessing the nutritional status of the population for this
vitamin.

CHAPTER 5
EFFECTS OF PYRIDOXINE-5' -/3-D-GLUCOSIDE ON THE
IN VITRO KINETICS OF PYRIDOXAL KINASE AND
PYRIDOXAMINE (PYRIDOXINE)-5'PHOSPHATE OXIDASE
IN RAT LIVER
Introduction
There are a number of enzymatic reactions in the metabolic
pathway responsible for the interconversion of vitamin B-6
compounds. In particular, PL kinase and PMP (PNP) oxidase are
considered the two most important catalytic enzymes in the
formation of the coenzymatic form of vitamin B-6, PLP (144).
Although other tissues contribute to vitamin B-6 metabolism,
the liver is thought to serve as the center of interconversion
of B-6 vitamers to PLP (109-110).
The B-6 vitamers are postabsorptively taken-up by the
liver and converted to PLP. The non-phosphorylated vitamers
are converted to their 5'-phosphorylated forms by a single
kinase (144). Following phosphorylation, PMP and PNP are
oxidized by PMP (PNP) oxidase to form PLP. This active
coenzymatic form of vitamin B-6 can then either be bound by
cellular apoproteins, released into circulation as PLP, or
dephosphorylated by alkaline phosphatase to PL (144) The PL
69

70
can then itself be released by the cell, oxidized to 4PA, the
metabolic end product of B-6 catabolism, or rephosphorylated
to form PLP.
Previously discussed studies in chapters 3 and 4 provided
evidence that PN-glucoside quantitatively alters the
metabolism and retention of administered PN (136). These
studies indicated that unlabeled PN-glucoside affects the
metabolic utilization of a simultaneous dose of labeled PN
(136). These data showed a linear relationship between the
dose of PN-glucoside and hepatic radiolabeled PNP and PLP
(136) Hepatic UC PN, PL, and PM were inversely proportional
to the dose of PN-glucoside.
The data from these studies led us to postulate that the
effect on vitamin B-6 metabolism exhibited by PN-glucoside may
be caused by a direct affect of the glycosylated vitamer on
the action of either PL kinase, PMP (PNP) oxidase, or both.
The purpose of this study was to determine the effect of PN-
glucoside on the in vitro catabolic activity of PL kinase and
PMP (PNP) oxidase partially from purified rat liver.
Materials and Methods
Protocol
Two studies were conducted to evaluate the effect of PN-
glucoside on the reaction rates of the enzymes PL kinase and
PMP (PNP) oxidase. In each of the studies six male Sprague
Dawley rats (Crl:CD(SD)BR) from Charles River Breeding

71
Laboratories (Wilmington, MA) weighing 200-300 g, were housed
individually in stainless steel cages in animal quarters
maintained at 241 C with a 12 h light/dark cycle. All
procedures for the care and treatment of the experimental
animals were in accordance with the National Institutes of
Health Guidelines.
The rats were fed ad libitum a casein-sucrose (20%:60%)
based diet, adequate in all micronutrients including 7 mg PN
HCl/kg diet for seven days (80) At the end of the seven day
acclamation period the rats were killed by lethal injection of
sodium pentobarbital (W.A. Butler Co., Columbus, OH). The
livers were rapidly excised and stored at -20 C until
analysis. The enzyme assay was performed at the temperature
and reaction time determined from preliminary experiments and
pH as shown in method of Merrill and Wang (116) Four varying
concentrations of PN-glucoside ranging from 0 ¡jlK to 150 xM and
six varying substrate concentrations were used. Preliminary
experiments were also performed to determine the linear
ranges of the assays with respect to enzyme concentration and
reaction time.
Forms of vitamin B-6
PN-glucoside was prepared by biological synthesis using
the propagation of alfalfa sprouts germinated in the presence
of PN-HC1 (Sigma Chemical Co., St. Louis, MO) (80). The [4,5-
3H]PN-HC1 684.5 KBq/nmol (Hoffmann-LaRoche, Nutley, NJ) with

72
a purity of greater than 98%, as determined by ion-pair
reverse phase HPLC. The pyridoxamine-phosphate-HCl and
pyridoxal-51-phosphate were obtained from (Sigma Chemical Co.,
St. Louis, MO).
Pvridoxal kinase
Extraction. Pyridoxal kinase was extracted from the
liver tissue of six rats using the procedure of Merrill et al.
(112). The tissue was stored at -20 C and processed within
72 h following removal from the rat. To each 1 g of liver, 4
ml of 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
(Hepes) (Sigma Chemical, Co., St. Louis, MO) at pH 7.4 and 25
C, was added. The mixture was homogenized 3 0-45 s with a
Polytron PT 20 (Brinkman Instruments, Westbury, NY).> The
resulting homogenates were centrifuged (Model L, type 40
rotor, Beckman Instruments, Pala Alto, CA) at 105,000 x g for
1 h at 4 C. The supernatant was collected and stored at -20
C.
Purification. The extracted PL kinase was purified
using the column chromatography method employed by Cash et al.
(145). Preparation of the 4-pyridoxyl-Sepharose as an
affinity support was adapted from the method used by Kwok and
Churchich (146). Seven grams of aminohexyl-Sepharose
(Sepharose AH, Pharmacia, Ltd. Piscataway, NJ) was pretreated
and washed with water. The Sepharose AH slurry was added to
a solution of 50 ml water containing 1 g PL-HC1 (Sigma
Chemical Co., St. Louis, MO) adjusted to pH 7.0 by addition of

73
NaOH. The slurry was shaken 16 h in the dark at ambient
temperature and then placed in an ice bath with the dropwise
addition of ice cold borohydride until disappearance of the
yellow color. During this period the slurry was shaken and 7%
acetic acid was added to prevent the pH from rising above 9.0.
The slurry was adjusted to pH 6.0 with 7% acetic acid and
packed into a 1.6 x 15 cm chromatography column. The column
was rinsed with 3M KC1 followed by water prior to use. All
rinsing and eluting buffers contained 2mM potassium phophate
(Fisher Scientific, Fairlawn, NJ) at pH 7.0 and lOmM
glutathione.
The 105,000 x g supernatant was applied to the PL-
Sepharose column and washed through with 50 ml of 100 mM KC1
buffer. The column was then rinsed with 500 ml 400mM KC1
followed by 50 ml of lOOmM KCl buffer. The PL kinase was then
eluted and adsorbed directly onto a Bio-Gel HTP
hydroxylapatite column (Bio Rad Laboratories, Richmond, CA),
previously equilibrated with buffer containing no KCl, with
500 ml 10 mM PN adjusted to pH 7.0 in buffer and rinsed with
25 ml lOOmM KCl buffer. The hydroxyapatite column was eluted
with 100 ml of a linear gradient of 2-300mM potassium
phosphate at pH 7.0. Approximately 3 ml fractions were
collected and the active fractions pooled. The enzyme
preparation was then tested for glucosidase activity using the
method of Daniels et al. (147) and for phosphatase activity
employing the PL kinase assay (discussed below) using PLP as

74
substrate followed by an inorganic phosphate assay kit (Sigma
Chemical Co., St. Louis, MO) and measuring the percent
conversion of PLP to PL.
Kinase assay. The assay procedure used was that of
Merrill and Wang (116). All reagents and buffers were at pH
5.75 and the procedure was conducted in GEF40G0 gold lights
that emit at a wavelength between 500 nm and 750 nm to prevent
photodegradation of pyridoxyl compounds. The assay mixture
contained 10 /xL each of 0.2M potassium phosphate, lOmM ATP (
Sigma Chemical Co., St. Louis, MO), 0.8mM ZnCl2, 0.6mM KCl,
0.3mM (20/iCi/mmol) [3H]PN diluted with unlabeled PN to yield
the appropriate specific activity, and 100 g of enzyme
extract with enough water to make a final volume of 100 /L.
The mixture was incubated for 30 min at 37C and the reaction
halted with 0.5 ml ice cold 10 mM ammonium formate. The
samples were transfered to columns which contained 0.2 ml of
DEAE-cellulose (Sigma Chemical Co., St. Louis, MO). The
columns were rinsed with 12 ml of lOmM ammonium formate,
followed by 2 ml water and eluted with 2 ml of 0.5 M KCl. A
2 ml fraction of the final elution was collected and added to
15 ml ScintiVerse II (Fisher Scientific, Fair Lawn, NJ) and
counted for total radioactivity. The reaction rates were
calculated using the following equation of Merrill and Wang
(116):
. ^P^Veaction CPmtime 0
Enzyme units = X 2 X quench factor
Specific Activity
(enzyme preparation)

75
Pyridoxamine (pyridoxine) 51 phosphate oxidase
Extraction. The rat liver PMP (PNP) oxidase was
extracted by the method of Kazarinoff and McCormick (119).
Liver tissue was homogenized in 0.02M potassium phosphate
buffer containing O.lmM mercaptoethanol, pH 7.0 and a ratio of
50 g tissue to 2 00 ml buffer. The homogenate was centrifuged
at 18,000 x g for 30 min and the supernatant adjusted to pH
5.0 with 2N acetic acid while stirring constantly. After 10
min the precipitate was removed by centrifugation at 18,000 x
g for 15 min. Potassium chloride (3.6 g) was added to the
supernatant and the volume brought up to 400 ml with water.
The pH was readjusted to 5.0 with 2N acetic acid.
Purification. The PMP (PNP) oxidase was extracted from
6 rat livers using the method of Kazarinoff and McCormick
(119). The clear supernatant resulting from the extraction
procedure was heated to 50 C and held for 8 min then plunged
into ice and cooled to 20 C and centrifuged at 18,000 x g for
30 min. The supernatant was applied to a DEAE-A50 column
(Sigma Chemical Co., St. Louis, MO) which had been prepared
by swelling overnight in 0.1 M potassium phosphate (pH 8.0).
A linear gradient established between 0.1 and 0.2 M potassium
phosphate at pH 8.0 was used to elute the column. Fractions
were collected and read at A28Q. The active fractions were
pooled. Solid ammonium sulfate (Sigma Chemical Co., St.
Louis, MO) (22.8 g/100 ml, 30% saturation) was added. After
30 min at 4C the precipitate was removed by centrifugation at

76
18,000 x g for 30 min and addition of ammonium sulfate (15.2
g/100 ml, 50% saturation). After 60 min the solution was
centrifuged and the active precipitate was dissolved in one
tenth the initial volume of water. This solution was dialyzed
against 0.02 M potassium phosphate buffer for 24 h and
centrifuged to remove the precipitate. The clear yellow
supernatant was applied to a column of Sephadex G-100 (Sigma
Chemical Co., St. Louis, MO) equilibrated with 0.02 M
phosphate buffer and eluted with the same buffer. Fractions
were collected and read at an A280 and the active fractions
pooled.
Oxidase assay. PMP (PNP) oxidase activity was measured
by the method of Wada and Snell (31) using phenylhydrazine as
the color development agent. The assay mixture contained 1 ml
of ImM PMP solution in 0.2M Tris-HCl, pH 8.0, 10-200 units of
enzyme (one unit of oxidase enzyme catalyzes the formation of
lnmol PLP/mL/min at 37 C), and enough Tris buffer to bring
the total volume to 3.5 ml. The mixture was gently shaken for
30 min at 37 C. The reaction was stopped by the addition of
0.3 mL of 100% (w/v) trichloroacetic acid. There was no
formation of a precipitate therefore, the entire supernatant
was used for the color reaction. The 3 mL supernatant was
placed in a clean test tube, 0.2 mL phenylhydazine reagent (2
g dissolved in 100 mL 10N H2S04) was added and the mixture was
heated for 20 min at 60 C. The reaction mixture was then
cooled to room temperature and the absorbance read at 410 nm

77
verses a reagent blank. The activity was expressed in nmol
PLP/mL enzyme/h.
Statistical analysis
The kinetic parameters Km and V||)ax were calculated by non
linear regression using the EZ-FIT program (149) Confidence
intervals for each and V(nax were calculated by the equation:
b, t 975 X SE, where b1 = or V|nax values (123). Overlapping
confidence intervals are not significantly different at
p<0.05.
Results
These studies were conducted to determine the effects of
PN-glucoside on two enzymes, PL kinase and PMP (PNP) oxidase,
in the metabolic pathway responsible for the interconversion
of vitamin B-6 compounds. These catalytic enzymes were
investigated, in vitro, by evaluating the rates of reaction in
the presence and absence of PN-glucoside. The purification
methods used for rat liver PL kinase and PMP (PNP) oxidase
yielded an enzyme preparation acceptable for kinetic analysis.
PL kinase
Partial purification of the enzyme by the above technique
resulted in a 243-fold increase in specific activity relative
to the crude homogenate (Table 5.1). Preliminary experiments
indicated that the rate of formation of [14C]PNP was linear
over varying enzyme concentrations and the rate of

78
Table 5.1 Purification of pyridoxal kinase from rat liver.
Fraction
Purification
Vol
(ml)
Total
Units
Protein
(mg/ml)
Units/mg
protein
(fold)
Supernatant
80
509
0.25
2.54
PL-Sepharose
6
188
0.05
628
247
/Hydroxyapatite
Units are nmol PNP/min/ml at 37C in phosphate buffer pH
5.75.
50 g rat liver was used.
phosphorylation was constant over at least a 60 min period
under the assay conditions outlined above. The rate of
phosphorylation of PN in this study indicated a of 21 /xM
which is in general agreement with the previously reported
value of 25 /xM measured under similar conditions (9) In the
presence of varying concentrations of PN-gluoside the value
at all levels was not significantly different from that
observed with 0 /xM PN-glucoside (Table 5.2) The Vmax was 16.8
nmol of PNP/h/mg of protein and did not differ significantly
over the range of PN-glucoside concentrations (Table 5.2).
The kinetic parameters (Figure 5.1) were calculated from
analysis using the EZ-FIT Program (149).

79
Table 5.2 Kinetic parameters of pyridoxal kinase with varying
concentrations of PN-glucoside.
PN-glucoside K,,, V
concentration (/iM) (nmol PNP/h/mg protein)
(MM)
0
2 32.0
50
222.2
100
212.4
150
203.0
0.29.01
0.28.01
0.28.01
0.27.01
PMP (PNP^ oxidase
The purification scheme yielded a preparation of 24-fold
increase in enzyme purity compared to the crude homogenate
(Table 5.3). The conversion of the substrate, PMP, to PLP has
previously been shown to exhibit a value of 25/iM (8) The
present study indicated a similar closely related value of
23/xM. Values for Km of the substrate in the presence of all
concentrations of PN-glucoside were not significantly
different (Table 5.4). The maximum velocity values in this
study did not differ signifcantly from the control value of
1.65 nmol PLP/min/mg protein (Table 5.4). All kinetic
parameters (Figure 5.2) were calculated using the EZ-FIT
Program (149).

80
Table 5.3 Purification of pyridoxamine (pyridoxine) phosphate
oxidase from rat liver.
Fraction
Purification
Vol
(ml)
Total
Units
Protein
(mg/ml)
Units/
mg protein
(fold)
Supernatant
200
11,300
15.0
3.77
Acid ppt.
Ammonium
250
7,100
4.5
6.25
1.65
Sulfate
25
3,267
12.1
10.8
2.86
DEAE Sephadex
30
10,100
4.2
78.3
20.77
Sephadex G-100
45
6,961
1.7
91.1
24.16
Units are nmol PNP/min/ml at 37C in phosphate buffer pH
5.75.
50 g of rat liver was used.
Table 5.4 Kinetic parameters of pyridoxamine (pyridoxine)
phosphate oxidase with varying concentrations of PN-glucoside.
PN-glucoside V|nax
concentration (/liM) (nmol PLP/h/mg
(juM) protein)
0
233.1
1.650.10
50
271.0
1.690.09
100
242.2
1.630.11
150
242.0
1.7010.06

Figure 5.1 Double reciprocal plot for the phosphorylation of PN by PL kinase from rat
liver in the presence of various levels of PN-glucoside.

3.00 j
2.75--
2.50 --
cn
£
2.25--
c
2.00-
E
\
1.75--
a.
_j
1.50--
CL
1.25-
O
£
1.00-
0.75 -
0.50
>
0.25 -
0.00-
O uM PN-glucoside
50 uM PN-glucoside
V 100 uM PNglucsido
150 uM PNglucsido
+
4-
+
+
+
14 28 42 56 70 84 98 112
-1 -1
(PYRID0XAMINE-PH0SPHATE) uM
126 140
Figure 5.2 Double reciprocal plot for the conversion of PMP to PLP by PMP (PNP) oxidase
from rat liver in the presence of various levels of PN-glucoside. n>

83
Discussion
An earlier study from this laboratory demonstrated that
the conjugated vitamin B-6 compound, PN-glucoside, affected
the metabolic utilization of PN when administered in a
simultaneous dose, in rats (136). The concentration of
phosphorylated forms of vitamin B-6 increased in proportion to
the quantity of PN-glucoside. These data further indicated a
strong possibility that the effect PN-glucoside exhibited on
vitamin B-6 metabolism may be a direct effect on one or both
of the two enzymes responsible for the interconversion of B-6
vitamers to the active coenzyme form, PLP. Hence, the present
study was undertaken to determine the in vitro effects of PN-
glucoside on the reaction rates of PL kinase and PNP (PNP)
oxidase.
The results showed that neither PL kinase nor PMP (PNP)
oxidase were effected by the concentrations of PN-glucoside
used in the study. The kinetic parameters, and V^, were
within the expected ranges for rat liver and demonstated no
significant differences when measured under conditions of
varying substrate and PN-glucoside concentrations.
When conducting in vitro kinetic studies consideration
should be given to any potential differences that may occur
under in vivo conditions. In the rat, vitamin B-6
postabsorptively diffuses into the liver where it is

84
metabolically trapped via phosphorylation. The vitamin is
maintained predominantly in phosphorylated forms due mainly to
protein binding and a ten fold greater activity of PL kinase
than phosphatase. It has been shown that hydrolysis of PLP is
inhibited by substrate and PMP (112).
Rat liver PMP (PNP) oxidase is inhibited by its product,
PLP, and the presence of PLP increases both the and
(148). The increase in Vmax may be a result of the action of
the more active rat liver phosphatases or an inability to
accurately measure initial velocity due to product inhibition
at early reaction times (148). Phosphatase more rapidly
hydrolyzes PLP when this product is free rather than protein
bound (144). The partially purified enzyme preparation used
in this study showed minimal phosphatase activity.
Approximately 20% of the PLP in vivo is in the free or loosely
bound form and the remaining PLP is bound to proteins.
Therefore, in vivo product inhibition may differ significantly
from in vitro inhibition. However, studies have shown that
crude rat liver homogenate compared with pure enzyme is
inhibited to a similar extent (148). Whether or not PLP
inhibits PMP (PNP) oxidase is dependent on the concentrations
of available substrate and product formed. Greater substrate
inhibition has been reported with PNP, in concentrations as
small as 5 im, due to a reduced enzyme-substrate complex that
reacts less efficently with oxygen (112). This has
significant consideration, since the presence of increasing

85
concentration of PN-glucoside during the metabolism of PN
causes proportional increases in the concentration of liver
PNP (136).
In summary, the results from this study indicate that, in
vitro, PN-glucoside has no direct effect on PL kinase or
PMP(PNP) oxidase. These data, provide evidence that these
biosynthetic enzymes are not directly involved in the changes
in distribution of B-6 vitamers caused by ingested PN-
glucoside, observed in the previous study (136).

CHAPTER 6
SUMMARY AND CONCLUSIONS
Since the isolation and identification of PN-glucoside
from rice bran (67) much focus has been on the comparative
bioavailability of this glycosylated form and other vitamin B-
6 compounds. A number of studies spanning the last decade has
provided a great deal of information regarding the biological
availability of vitamin B-6 relative to its multiple forms.
As discussed in previous chapters, between 5%-80% of the
vitamin B-6 found in a variety of plant derived foods was in
the glycosylated form and both food processing and diet
composition are important considerations when evaluating the
bioavailability of vitamin B-6. Also the net bioavailability
of PN-glucoside relative to PN is less than 40% in rats
(75,77,90). Enzymatic hydroxylation of PN does not exhibit
antivitamin B-6 activity, however other degradative products
such as pyridoxallysine have been shown to be vitamin B-6
analogues and act as competitive inhibitors to enzymes in the
vitamin B-6 metabolic pathway.
The use of stable iotopic methodology showed PN-glucoside
to be 58% as biologically active as PN in humans (41). This
study indicated that although PN-glucoside bioavailability was
86

87
incomplete, it is approximately two fold greater than
previously found in rats.
The utilization of stable isotopic methodology in humans
continues to contribute information regarding the absorption,
metabolism and utilization of vitamin B-6 in humans. In an
effort to advance our understanding of vitamin B-6 nutriture
in humans it is necessary to gather information regarding the
dietary forms of vitamin B-6, their bioavailabililty and the
enzymes responsible for the interconversion of B-6 vitamers.
It is equally important to determine any potential interaction
between multiple forms of B-6 compounds and the effect these
may have on metabolic utilization of B-6 vitamers when
consumed together.
The purpose of the present research was to determine the
effect of PN-glucoside on the metabolic utilization of PN when
administered in a simultaneous dose in rats and humans. The
adventive of stable isotopic techniques allowed for the
evaluation in human subjects. The effect of PN-glucoside on
enzymes responsible for control of the conversion of vitamin
B-6 compounds to the active coenzyme form, PLP was also
determined.
The research discussed in chapters 3-5 was divided into
three separate but interconnected studies. The experimental
design of each study was contingent on the findings of the
prior study or studies. Initially, an animal study was
performed to determine if any interaction between PN-glucoside

88
and PN occurred, which affected metabolic utilization. The
premise of these studies was to simulate ingestion of a mixed
diet by determining the effect of a simultaneous dose of the
two B-6 compounds. Since the interest was in evaluating a
direct affect of PN-glucoside on PN metabolic utilization, the
rats received labeled-PN and varying concentrations of
unlabeled-PN-glucoside. The results indicated an effect of
PN-glucoside on the hepatic metabolism of PN. There existed
a linear relationship between the quantity of phosphorylated
B-6 vitamers and increasing concentrations of PN-glucoside.
It was also noted that the concentration of PNP, a B-6 vitamer
not detected in control rat livers, was directly proportional
to the concentration of PN-glucoside in the dose. The reason
for this finding is presently unclear. These data indicated
a negative correlation between the amount of 4PA detected in
urine and the concentration of PN-glucoside administered in
the dose. A similar correlation to that of 4PA existed for
non-phosphorylated B-6 vitamers in relation to PN-glucoside
concentration in the dose. These results indicated that the
PN-glucoside which was not hydolyzed and therefore
metabolically active as PN, is not biologically inert.
Instead this intact glycosylated B-6 compound exhibits an
effect on the utilization of vitamin B-6 in the rat.
As previously discussed, studies showed that the
bioavailability of PN-glucoside, although incomplete, was
greater in humans than rats. In consideration of this

89
observation, the effect of PN-glucoside on vitamin B-6
metabolism seen in rats would be significant, but of a lesser
magnitude in humans, assuming that intact PN-glucoside is the
active compound.
A second study was designed using stable isotopic methods
to determine the effect of PN-glucoside on metabolism of PN.
Deuterium-labeled [5'-C2H2OH]pyridoxine (d2PN) was used in the
study. The d2PN is metabolically stable and remains attached
to the PN molecule throughout catabolism. Therefore,
metabolic utilization of d2PN may be followed and ultimately
quantified as the urinary catabolite d24PA. The results of
this study indicated a negative correlation between the
quantity of total 4PA excreted and the concentration of PN-
glucoside in the dose.
Review of the results of these two studies showed that
PN-glucoside, when administered simultaneously with PN, has an
effect on the metabolic utilization of vitamin B-6 and that
PN-glucoside which is not hydrolyzed by intestinal mucosal
glucosidase, exhibits some biological activity regarding
vitamin B-6 metabolism. To determine which vitamin B-6
metabolic process was affected by PN-glucoside, study three
was proposed. Data from chapter 3 indicated increased
quantities of phosphorylated B-6 vitamers, specifically PLP
and PNP. These results indicated that either of the enzymes
responsible for controlling the level of phosphorylated B-6
compounds may be affected by PN-glucoside. Study three was

90
designed to determine the in vitro effects of PN-glucoside on
PL kinase and PMP (PNP) oxidase. The results indicated no
direct effect of PN-glucoside on either enzyme as determined
by no significant differences observed between the kinetic
parameters, and V|nax with varying concentrations of the
glycosylated vitamin B-6 compound. Further studies are needed
to assess potential effects of PN-glucoside on other enzymes
in the interconversion pathway of vitamin B-6 metabolism.
These experimenal results provide evidence that PN-
glucoside exhibits a quantitative effect on the metabolic
utilization of PN. Although the biochemical mechanism
responsible for this effect are presently uncertain, it is
evident that consideration in the study of vitamin B-6
bioavailability must incorporate the potential interactions
among B-6 compounds present in foods generally consumed in a
mixed diet.
The information from these studies is consequential in
relation to current USDA Dietary Guidelines, which are used to
convey recommendations to the general population. Present
nutrition education is promoting the consumption of greater
quantities of plant derived foods. Food consumption data have
shown that there are several sub-populations at nutritional
risk for vitamin B-6, specifically pregnant females,
adolescents, and the elderly.those individuals undergoing
growth or aging.

91
The predominant form of vitamin B-6 in many fruits and
vegetables is PN-glucoside and research has indicated that
not only is this vitamin B-6 compound biologically less
available as a usable source of vitamin B-6 it also effects
the metabolic utilization of other B-6 vitamers. In
development of recommendations for dietary allowances and
dietary food intake patterns it is necessary to consider not
only vitamin B-6 availability and biological activity, but
interactions that occur between different vitameric forms and
how these interactions effect bioavialability of the nutrient.

LITERATURE CITED
1. Canhaxn J.E., Baker E.M., Harding R.S., Sauberlich H.E.,
Plough I.C. (1969). Dietary Protein Its
Relationship to Vitamin B-6 Reguirements and Function.
Ann NY Acad Sci 166: 16-29.
2. Food and Nutrition Board (1989). Recommended Dietary
Allowances. Natl Acad Sci., Washington DC.
3. Meister A. (1965). The Function of Vitamin B-6 in
Amino Acid Metabolism. In: Biochemistry of the Amino
Acids, 2nd Ed., Vol. 1 Academic Press, New York. 375-
412.
4. Bender D.A. (1987). Oestrogens and Vitamin B-6 Actions
and Interactions. World Rev Nutr Diet 51: 140.
5. Bunce G.E., Vessal M. (1987). Effect of Zinc and/or
Pyridoxine Deficiency upon Oestrogen Retention and
Oestrogen Receptor Distribution in the Rat Uterus. J
Steroid Biochem 26: 303.
6. Rose C.S., Gyorgy P., Butler M. (1976). Age
Differences in Vitamin B-6 Status in 617 Men. Am J
Clin Nutr 29: 847-853.
7. Schuster K., Bailey L.B., Mahan C.S. (1984). Effect
of Maternal Pyridoxine HCL Supplementation on the
Vitamin B-6 Status of Mother and Infant and on
Pregnancy Outcome. J Nutr 114: 977-988.
8. Brophy M.H., Suteri P.K. (1975). Pyridoxal Phosphate
and Hypertensive Disorders of Pregnancy. Am J Obstet
Gynecol 121:1075-1079.
9. Ranke E., Tauber A., Horonick B. (1960). Vitamin B-6
Deficiency in the Aged. J Gerontol 15: 41-44.
10. Driskell J.A. (1978). Vitamin B-6 Status of the
Elderly. In: Human Vitamin B-6 Requirements. Natl
Acad Sci., Washington DC. 252-256.
92

93
11. Kirksey A., Keaton K., Abernathy R.P. (1978). Vitamin
B-6 Nutritional Status of a group of Female
Adolescents. Am J Clin Nutr 31: 946-954.
12. U.S. Department of Agriculture. (1986) Nationwide
Food Consumption Survey Continuing Survey of Food
Intakes by Individuals, Low Income Women 19-25 Years
and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII
Report No. 85-2. U.S. Department of Agriculture,
Hyattsville, MD.
13. Shane B., Contractor S.F. (1975). Assessment of
Vitamin B-6 Status. Studies on Pregnant Women and Oral
Contraceptive Users. Am J Clin Nutr 28: 739-747.
14. Roepke J.L.B., Kirksey A. (1979). Vitamin B-6
Nutriture During Pregnancy and Lactation. Am J Clin
Nutr 28: 146-156.
15. Pao E.M., Mickle S.J. (1981). Problem Nutrients in
the United States. Food Technol 35(9): 58-69.
16. Gregory J.F., Kirk J.R. (1981). The Bioavailability
of Vitamin B-6 in Foods. Nutri Rev 39(1): 1-8.
17. Turner J.M. (1961). Pyridoxal Phosphate Breakdown by
an Alkaline-Phosphatase Preparation. Biochem J 80:
663-668.
18. Leklem J.E. (1988). Vitamin B-6: Of Reservoirs,
Receptors and Requirements. Nutr Today Sept/Oct: 4-
10.
19. Booth C.C., Brain M.C. (1962). The Absorption of
Tritium Labelled Pyridoxine Hydrochloride in the Rat.
J Physiol 164: 282-294.
20. Scudi J.V., Unna K., Antopol W. (1940). A Study of
Urinary Excretion of Vitamin B-6 by a Colorimetric
Method. J Biol Chem 135: 371-376.
21. Scudi J.V., Koones H.F., Keresztesy J.C. (1940).
Urinary Excretion of Vitamin B-6 in the Rat. Proc Soc
Exptl Biol Med 43: 118-122.
22. Yamada K., Tsuji M. (1968). Transport of Vitamin B-6
in Human Erythrocytes. J Vitaminol (Kyoto) 14: 282-
294.
23. Middleton H.M. (1979). In Vivo Absorption and
Phosphorylation of Pyridoxine-HCl in Rat Jejunum.
Gastroenterol 76: 46-49.

94
24. Tsuji T., Yamada R., Nose Y. (1973). Intestinal
Absorption of Vitamin B-6. I. Pyridoxal Uptake by Rat
Intestinal Tissue. J Nutr Sci Vitaminol 19: 401-417.
25. Yamada R.H., Tsuji T., and Nose Y. (1977). Uptake and
Utilization of Vitamin B-6 and its Phosphate Esters by
Escherichia coli. J Nutr Sci Viltaminol (Tokyo)
23(1):7-17.
26. Middleton H.M. (1977). Uptake of Pyridoxine
Hydrochloride by the Rat Jejunal Mucosa in vitro. J
Nutr 107: 126-131.
27. Middleton H.M. (1979). Intestimal Absorption of
Pyridoxal-5' Phosphate: Disappearance from Perfused
Segments of Rat Jejunum in Vivo. J Nutr 109: 975-981.
28. Hamm M.W., Mehansho H. Henderson L.M. (1979).
Transport and Metabolism of Pyridoxamine and
Pyridoxamine Phosphate in the Small Intestine of the
Rat. J Nutr 109: 1552-1559.
29. Mehansho H., Hamm M.W., Henderson L.M. (1979).
Transport and Metabolism of Pyridoxal and Pyridoxal
Phosphate in the Small Intestine of the Rat. J Nutr
109: 1542-1551.
30. McCormick D.B., Gregory M.E., Snell E.E. (1966).
Pyridoxal Phosphokinases. I. Assay, Distribution,
Purification, and Properties. J Biol Chem 236: 2076-
2084.
31. Wada H., Snell E.E. (1961). The Enzymatic Oxidation
of Pyridoxine and Pyridoxamine Phosphate. J Biol Chem
236: 2089-2095.
32. Pogell B.M. (1958). Enzymatic oxidation of
Pyridoxamine Phosphate to Pyridoxal Phosphate in Rabbit
Liver. J Biol Chem 232: 761-776.
33. Morisue T., Morino Y., Sakamoto Y., Ichihara K.
(1960). Enzymatic Studies on Pyridoxine Metabolism.
III. Pyridoxine Phosphate Oxidase. J Biochem 48: 28-
36.
34. Samra P.S., Snell E.E., Elvehjem C.A. (1947). The
Bioassay of Vitamin B-6 in Natural Materials. J Nutr
33:121-128.

95
35. Toepfer E.W., Polansky M.M., Richardson L.R., Wilkes S.
(1963). Comparison of Vitamin B-6 Values of Selected
Food Samples by Bioassay and Microbiological Assay. J
Ag Food Chem 11: 523-525.
36. Nguyen L.B., Gregory J.F. (1983). Effects of Food
Composition on the Bioavailability of Vitamiln B-6 in
the Rat. J Nutr 113: 1550-1560.
37. Gregory J.F., Ink S.L. (1985). The Bioavailability of
Vitamin B-6. In: Vitamin B-6: Its Role in Health and
Disease. Alan R. Liss, Inc., New York. 3-23.
38. Nelson E.W., Lane H., Cerda J.J. (1976). Comparative
Human Intestinal Bioavailability of Vitamin B-6 from a
Synthetic and a Natural Source. J Nutr 106: 1433-
1437.
39. Nelson E.W., Burgin C.W., Cerda J.J. (1977).
Characterization of Food Binding of Vitamin B-6 in
Orange Juice. J Nutr 107: 2128-2134.
40. Leklem J.E., Miller L.T., Perera A.D., Peffers D.E.
(1980). Bioavailability of Vitamin B-6 from Wheat Bread
in Humans. J Nutr 110: 1819-1828.
41. Gregory J.F. (1990). The Bioavailability of Vitamin
B-6 Recent Findings. Ann NY Acad Sci 585: 86-95.
42. Coburn S.P., Mahuren J.D. (1979). Major Urinary
Metabolites of 4' and 5'-Deoxypyridoxines in Various
Species. IRCS Med Sci 7: 556.
43. Rodwell V.W., Volcani B.E., Ikawa M., Snell E.E.
(1958). Bacterial Oxidation of Vitamin B-6.
Isopyridoxal and 5-Pyridoxic Acid. J Biol Chem 233:
1548-1554.
44. Coburn S.P., Hahuren J.D., Erbelding W.F., Tokwnsend
D.W., Hachey D.L., Klein P.D. (1984). Measurements of
Vitamin B-6 Kinetics in Vivo Using Chronic
Administration of Labelled Pyridoxine. In: Chemical
and Biological Aspects of Vitamin B-6 Catalysis, Part A
(Evangelopoulos A. E., ed.) Alan R. Liss, New York.
43-54.

96
45. Coburn S.P., Mahuren J.D., Szadkowska Z., Schaltenbrand
W.E., Townsend D.W. (1987). Kinetics of Vitamin B-6
Metabolism Examined in Miniture Swine by Continuous
Administration of Labelled Pyridoxine. In:
Mathematical Models in Experimental Nutrition (Canolty
N.L., Cain T.P., eds.) University of Georgia, Athens,
Georgia. 99-111.
46. McElroy L.W., Goss H. (1939). Report on Four Members
of the Vitamin B complex Synthesized in the Rumen of
Sheep. J Biol Chem 130: 437-438.
47. Coates M.E., Ford J.E., Harrison G.F. (1968).
Intestinal Synthesis of Vitamins of the B complex in
Chicks. Br J Nutr 22: 493-500.
48. Gregory J.F., Litherland S.A. (1986). Efficacy of the
Rat Bioassay for the Determination of Biologically
Available Vitamin B-6. J Nutr 116: 87-97.
49. Ikeda M., Hosotani T., Kurimoto K., Mori T., Ueda T.,
Kotake Y., Sakakibara B. (1979). The Differences of
the Metabolism Related to Vitamin B-6 Dependent Enzymes
among Vitamin B-6 Deficient Germfree and Conventional
Rats. J Nutr Sci Vitaminol 25: 131-139.
50. Ikeda M., Hosotani T., Ueda T., Kotake Y., Sakakibara
B. (1979). Effect of Vitamin B-6 Deficiency on the
Levels of Several Water Soluble Vitamins in Tissues of
Germfree and Conventional Rats. J Nutr Sci Vitaminol
25: 141-149.
51. Hughes E.H., Squibb R.L. (1942). Vitamin B-6 in the
Nutrition of the Pig. J Animal Sci 1: 320-325.
52. Wozenski J.R., Leklem J.E., and Miller L.T. (1980).
The Metabolism of Small Doses of Vitamin B-6 in Men.
J Nutr 110 (2): 275-285.
53. Schultz T.D., Leklem J.E. (1987). Vitamin B-6 Status
and Bioavailability in Vegetarian Women. Am J Clin
Nutr 46: 647-51.
54. Sumi, Y., Miyadawa, M., Kansaki, M., and Kotake, Y.
(1977). Vitamin B-6 Deficiency in Germ Free Rats. J
Nutr 107: 1707-1714.
55. Kabir, H., Leklem, J.E., and Miller, L.T. (1983).
Comparative Vitamin B-6 Bioavailability from Tuna,
Whole Wheat Bread and Peanut Butter in Humans. J Nutr
113: 2412-2420.

97
56. Lindberg A.S., Leklem J.E., Miller L.T. (1983). The
Effect of Wheat Bran on the Bioavailability of Vitamin
B-6 in Young Men. J Nutr 113: 2578-2586.
57. Tarr J.B., Tamura T., Stokstad L.R. (1981).
Availability of Vitamin B-6 and Pantothenate in an
Average American Diet in Man. Am J Clin Nutr 34:
1328-1337.
58. Coursin D.B. (1954). Convulsive Seizures in Infants
with Pyridoxine Deficient Diet. J Am Med Assoc 154:
406-408.
59. Hassinen J.B., Durbin G.T., Bernhart F.W. (1954). The
Vitamin B-6 Content of Milk. J Nutr 53:249-257.
60. Tomarelli R.M., Spence E.R., Bernhart F.W. (1955).
Biological Availability of Vitamin B-6 in Heated Milk.
J Agrie Food Chem 3: 338-341.
61. Gregory J.F., Kirk J.R. (1978). Assessment of
Roasting Effects on Vitamin B-6 Stability and
Bioavailability in Dehydrated Food Systems. J Food Sci
42: 1585-1589.
62. Gregory J.F., Kirk J.R. (1978). Assessment of Storage
Effects on Vitamin B-6 Bioavailability in Dehydrated
Food Systems. J Food Sci 43: 1801-1808.
63. Gregory J.F. (1980). Bioavailability of Vitamin B-6
in Nonfat Dry Milk and a Fortified Rice Breakfast
Cereal Product. J Food Sci 45:84-86.
64. Evans S.R., Gregory J.F., Kirk J.R. (1981). Thermal
Degradation Kinetics of Pyridoxine Hydrochloride in
Dehydrated Model Food Systems. J Food Sci 46: 555-
558, 563.
65. Kabir H., Leklem J., Miller L.T. (1983). Measurement
of Glycosylated Vitamin B-6 in Foods. J Food Sci 48:
1422-1425.
66. Scudi J.V. (1942). Conjugated Pyridoxine in Rice Bran
Concentrates. J Biol Hem 145: 637-639.
67. Yasumoto K., Tsuji H., Iwami K., Mitsuda H. (1977).
Isolation from Rice Bran of a Bound Form of Vitamin B-6
and its Identification as 5'-0-(beta-D-
Glucopyranosyl)Pyridoxine. Agrie Biol Chem 41: 1061-
1067.

98
68. Tadera K., Mori E., Yagi F., Kobayashi K., Imada K.,
Imabeppu M. (1985). Isolation and Structure of a Minor
Metabolite of Pyridoxine in Seedlings of Pisum sativum
L. J Nutr Sci Vitaminol 31: 403-408.
69. Suzuki Y., Ishii K. Suga K., Uchida K. (1986).
Formation of beta-Glucosylpyridoxines in Soybean and
Rice Callus. Phytochem 25: 1331-1332.
70. Tadera K., Makamura M., Yagi F., Kobayashi A. (1979).
A Particulate Glucosyltransferase Catalyzing the
Formation of 5'-0-(Beta-D-Glucopyranosyl) Pyridoxine
from Pyridoxine: The Occurrence in the Seedlings of
Pisum sativum L. J Nutr Sci Vitaminol 25: 76-82.
71. Gregory J.F., Sartain D.B. (1991). Improved
Chromatographic Determination of Free and Glycosylated
Forms of Vitamin B-6 in Foods. J Agri and Food Chem
39: 899-905.
72. Suzuki Y., Uchida K. Tsuboi A. (1986). Enzymatic
Formation of Pyridoxine Beta-Glucosides by Wheat Bran
Beta-Glucosidase. Nippon Nogeikagaku Kaishi 53: 189-
196.
73. Tadera K., Yagi F., Dobayashi A. (1982). Specificity
of a Particulte Glucosyltransferase in Seedlings of
Pisum sativum L. which Catalyzes the Formation of 5'-0-
(Beta-D-Glucopyranosyl)Pyridoxine. J Nutr Sci
Vitaminol 28: 359-366.
74. Kabir H., Leklem J.E., Miller L.T. (1983).
Relationship of the Glycosylated Vitamin B-6 Content of
Foods to Vitamin B-6 Bioavailability in Humans. Nutr
Rept Int 28: 709-715.
75. Gregory J.F, Ink S.L. (1987). Identification and
Quantification of Pyridoxine-beta-Glucoside as a Major
Form of Vitamin B-6 in Plant Derived Foods. J Agrie
Food Chem 35: 76-82.
76. Iwami K., Yasumoto K. (1986). Synthesis of
Pyridoxine-beta Glucoside by Rice Bran beta-Glucosidase
and its In Situ Absorption in Rat Small Intestine.
Nutr Res 6: 407-414.
77. Ink S.L., Gregory J.F., Sartain D.B. (1986).
Determination of Pyridoxine beta-Glucoside
Bioavailability using Intrinsic and Extrinsic Labeling
in the Rat. J Agrie Food Chem 34: 857-862.

99
78. Bills N.D., Leklem J.E., Miller L.T. (1987). Vitamin
B-6 Bioavailability in Plant Foods is Inversily
Correlated with Percent Glycosylated Vitamin B-6. Fed
Proc 46: 1487.
79. Tsuji H., Okada J., Iwami K., Yasumoto K., Mitsuda H.
(1977). Availability of Vitamin B-6 and Small
Intestinal Absorption of Pyridoxine-Beta-D-Glucoside in
Rats. Vitamins 51: 153-159.
80. Trumbo P.R., Gregory J.F., Sartain D.B. (1988).
Incomplete Utilization of Pyridoxine-beta-Glucoside as
Vitamin B-6 in the Rat. J Nutr 118: 170-175.
81. Trumbo P.R., Gregory J.F. (1988). Metabolic
Utilization of Pyridoxine-beta-Glucoside in Rats:
Influence of Vitamin B-6 Status and Route of
Administration. J Nutr 118: 1336-1342.
82. Gregory J.F., Trumbo P.R, Bailey L.B, Toth J.P,
Baumgartner T.G., Cerda J.J. (1991). Bioavailability
of Pyridoxine-51 -/3-D-Glucoside Determined in Humans by
Stable-Isotopic Methods. J Nutr 121: 177-186.
83. Register U.D., Lewis U.J., Ruegamer W.R., Elvehuem C.A.
(1950). Studies on the Nutritional Adequacy of Army
Combt Rations. J Nutr 40: 281-294.
84. Andn M.B., Reynolds R.D., Moser-Veillon P.B., and
Howard M.P. (1989). Dietary Intake of Total and
Glycosylated Vitamin B-6 and the Vitamin B-6
Nutritional Status of Unsupplimented Lactating Women
and Their Infants. Am J Clin Nutr 50: 1050-1058.
85. Reynolds R.D. (1988). Bioavailability of Vitamin B-6
from Plant Foods. Am J Clin Nutr 48: 863-867.
86. Harding R.S., Plough I.C., Friedemann T.E. (1959).
The Effect of Storage on the Vitamin B-6 Content of a
Packaged Army Ration, with a Note on the Human
Requirement for the Vitamin. J Nutr 68: 328-331.
87. Sauberlich H.E. (1961). Studies on the Toxicity and
Antagonistism of Amino Acid for Weaning Rats. J Nutr
74: 289-297.
88. Davies M.K., Gregory M.E., Henry K.G. (1959). The
Effect of Heat on the Vitamin B-6 of Milk. J Dairy Res
26: 215-220.

100
89. Lushbough C.H., Weichmann J.M., Schweigert B.S.
(1959). Retention of Vitamin B-6 in Meat During
Cooking. J Nutr 67: 451-459.
90. Ink S.L., Gregory J.F., Sartain D.B. (1986).
Determination of Vitamin B-6 Bioavailability in Animal
Tissues Using Intrinsic and Extrinsic Labeling in the
Rat. J Agrie Food Chem 34: 998-1004.
91. Nguyen L.B., Gregory J.F., Burgin C.W., Cerda J.J.
(1981). In Vitro Binding of Vitamin B-6 by Selected
Polysaccharides, Lignin and Wheat Bran. J Food Sci
46: 1860-1862.
92. Miller L.T., Schultz T.D., Leklem J.E. (198010.
Influence of Citrus Pectin on the Bioavailability of
Vitamin B-6 in Men. Fed Proc 39: 797.
93. Machida Y., Nagai T. (1980). In Vitro Dissolution
Test and Absorption Study of a Per Oral Controlled
Release Dosage Form Containing Pyridoxine Hydrochloride
or Sodium Riboflavin Phosphate with Hydroxypropyl
Cellulose. Chem Pharm Bull 28: 1082-1089.
94. Klosterman H.J. (1974). Vitamin B-6 Antagonists of
Natural Origin. J Agic Food Chem 22: 13-16.
95. Bauerfeind J.C, Miller O.N. (1978). Vitamin B-6:
Nutritional and Pharmaceutical Usage, Stability,
Bioavailability, Antagonists, and Safety. In: Human
Vitamin B-6 Requirements. National Academy of
Sciences, Washington, DC. 78-110.
96. Klosterman H.J. (1981). Vitamin B-6 Antagonists in
Natural Products. In: Antinurients and Natural
Toxicants in Foods, (Ory R.L., ed.), Food and
Nutrition Press, Westport CT. 295-317.
97. Bernhart F.W., D'Amato E., Tomarelli R.M. (1960). The
Vitamin B-6 Activity of Heat Sterilized Milk. Arch
Biochem Biophys 88: 267-269.
98. Wendt G., Bernhart F.W. (1960). The Structure of a
Sulfur-Containing Compound with Vitamin B-6 Activity.
Arch Biochem Biophys 88: 270-272.
Srncova V., Davidek J. (1972). Reaction of Pyridoxal
and Pyridoxal-5-Phosphate with Proteins. J Food Sci
37: 310-312.
99.

101
100. Gregory J.F., Kirk J.R. (1977). Interaction of
Pyridoxal and Pyridoxal Phosphate with Peptides in a
Liquid Model Food System During Thermal Processing. J
Food Sci 42: 1554-1556.
101. Gregory J.F., Kirk J.R. (1978). Vitamin B-6 Activity
for Rats of Epsilon-Pyridoxyllysine Bound to Dietary
Protein. J Nutr 108: 1192-1199.
102. Gregory J.F., Ink S.L., Sartain D.B. (1986).
Degradation and Binding to Food Proteins of Vitamin B-6
Compounds During Thermal Processing. J Food Sci 51:
1345-1351.
103. Gregory J.F. (1980). Effects of Epsilon-
Pyridoxyllysine Bound to Dietary Protein on the Vitamin
B-6 Status of Rats. J Nutr 110: 995-1005.
104. Tadera M., Arima K., Yoshino S., Yagi F., Kobayashi A.
(1986). Conversion of Pyridoxine into 6-
Hydroxypyridoxine by Food Components, Especially
Ascorbic Acid. J Nutr Sci Vitamilnol 32: 267-277.
105. Tadera K., Arima M., Yagi F. (1988). Participation of
Hydroxyl Radical in Hydroxylation of Pyridoxine by
Ascorbic Acid. Agrie Biol Chem 52: 2359-2360.
106. Gregory J.F., Leatham K. (1990). Lack of Vitamin B-6
Activity of 6-Hydroxypyridoxine. J Food Sci 55(4):
1143-1146.
107. Tadera K., Nagano K., Yagi F., Kobayashi A., Imada K.,
Tanaka S. (1988). Isolation and Structure of a New
Metabolite of Pyridoxine in Seedlings of Pisum sativum
L. Agrie Biol Chem 47: 1357-1359.
108. Tadera K., Kaneko T., Yagi F. (1986). Evidence for
the Occurrence and Distribution of a New Type of
Vitamin B-6 Conjugate in Plant Foods. Agrie Biol Chem
50: 2933-2934.
109. Johansson S., Lindstedt S., Tiselius, H.G. (1974).
Metabolic Interconversions of Different Forms of
Vitamin B-6. J Biol Chem 249: 6040-6046.
110. Colombini C.E., McCoy E.E. (1970). Vitamin B-6
Metabolism: The Utilization of [14C]pyridoxine by the
Normal Mouse. Biochem 9: 533-538.
111. Lumeng L., Brashear R.E., Li T.K. (1974). Pyridoxal
5' Phosphate: Source, Protein-binding, and Cellular
Transport. J Lab Clin Med 84: 334-343.

102
112. Merrill A.H., Henderson J.M., Wang E., McDonald B.W.,
Millikan, W.J. (1984). Metabolism of Vitamin B-6 by
Human Liver. J Nutr 114: 1664-1674.
113. Ebadi M. (1978). Vitamin B-6 and Biogenic Amines in
Brain Metabolism. In: Human Vitamin B-6 Requirements.
Nat Acad Sci, Washington, D.C. 129-161.
114. Sourkes T.L. (1966). Dopa Decarboxylase: Substrates,
Coenzymes, Inhibitors. Pharmacol Rev 18: 53-60.
115. Molinoff P.B, Axlerod J. (1971). Biochemistry of
Catecholamines. Ann Rev Biochem 40: 465-500.
116. Merrill A.H., Wang E. (1986). Highly sensitive
Methods of assaying the Enzymes of Vitamin B-6
Metabolism. Methods in Enzymol 122: 110-116.
117. Karawya E., Fonda M.L. (1982). Physical and Kinetic
Properties of Sheep Liver Pyridoxine Kinase. Arch
Biochem and Biophy 216: 170-177.
118. McCormick D.B., Merrill A.H. (1980). Pyridoxamine
(Pyridoxine)5'-phosphate Oxidase. In: Vitamin B-6
Metabolism and Role in Growth. (Tryfiates G.P., ed.).
Food and Nutrion Press, Inc., Westport, Conn. 1-26.
119. Kazarinoff M.N., McCormick D.B. (1975). Rabbit Liver
Pyridoxamine(Pyridoxine)5'-phosphate Oxidase:
Purification and Properties. J Biol Chem 250: 3436-
3442.
120. McCormick D.B., Snell E.E. (1961). Pyridoxal
Phosphokinase: Effects of Inhibitors. J Biol Chem
236: 2085-2088.
121. Gregory J.F. (1980). Effects of e-pyridoxyllysine and
Related Compounds on Liver and Brain Pyridoxal Kinase
and Liver Pyridoxamine(Pyridoxine)5'-phophate Oxidase.
J Biol Chem 255: 2355-2359.
122. McCormick D.B., Kazarinoff M.N., Tsuge H. (1976).
FMN-Dependent Pyridoxamine Phosphate Oxidase from
Rabbit Liver. In: Flavins and Flavoproteins, Fifth
Intern. Symp. (Singer T.P., ed.) Elsvier Publishing
Co., Amsterdam.
123. SAS Inst. Inc. (1990). SAS Users Guide: Statistics
Version 5 Edition. Cary, NC.

103
124. Serebro H.A., Solomon H.M., Johnson J.H., Hendrix, T.R.
(1966) The Intestinal Absorption of Vitamin B-6
Compounds by the Rat and Hamster. Bull Johns Hopkins
Hosp 119: 166-171.
125. Lumeng L., Li T.K. (1974) Vitamin B-6 Metabolism in
Chronic Alcohol Abuse: Pyridoxal Phosphate Levels in
Plasma and the Effects of Acetaldehyde on Pyridoxal
Phosphate Synthesis and Degradation in Human
Erythrocytes. J Clin Invest 53: 693-704.
126. Lumeng L., Li T.K., Lui A. (1985) The Interorgan
Transport and Metabolism of Vitamin B-6. In: Vitamin
B-6, Its role in Health and Disease. (Reynolds R.D.,
Leklem J.E., eds.) Alan R. Liss Inc., New York, on
Dietary Allowances, Food and Nutrition Board, National
Research Council. Natl Acad Sci, Washington, D.C. 35-
54.
127. Johansson S., Lindstedt S., Tiselius H.G. (1968)
Metabolism of [3H8]pyridoxine in Mice. Biochemistry
7: 2327-2332.
128. Contractor S.F., Shane B. (1971) Metabolism of [14C]
Pyridoxol in the Pregnant Rat. Biochim Biophys Acta
230: 127-136.
129. Tryfiates G.P., Saus F.L. (1976) Metabolism of
Pyridoxine in the Liver of Vitamin B-6-Deficient Rats.
Biochim Biophys Acta 451: 333-341.
130. Bain J.A., Williams H.L. (1960) Concentrations of B-6
Vitamers in Tissues and Tissue Fluids. In: Inhibition
in the Nervous System and Gamma-aminobutyric Acid,
(Roberts E., Baxter C.F., Harreveld A.V., Wiersma
C.A.G., Adey W.R., Killiam K.F., eds.) Pergamon
Press, New York.
131. Lyon J.B., Porter J. (1960) The Distribution of
Vitamin B-6 in the Tissues of two inbred Strains of
Mice fed Complete and Vitamin B-6-Deficient Rations. J
Biol Chem 237: 1989-1991.
132. Lumeng L., Li T.K. (1978) Validation of the Diagnostic
Value of Plasma Pyridoxal 5'phosphate Measurements in
Vitamin B-6 Nutrition of the Rat. J Nutr 108: 545-
553.

104
133. Snell E.E., Haskell B.E. (1971) The Metabolism of
Vitamin B-6. In: Comprehensive Biochemistry, Chapter
I Section C. (Florkin M., Stotz E.H., eds.) Vol. 21
Elsevier, Amsterdam. 47-71.
134. Merrill A.H. Jr., Henderson J.M. (1990) Vitamin B-6
Metabolism by Human Liver. In: Vitamin B-6
(Dakshinamurti K., ed.) Vol. 585 Annals of the New
York Academy of Sciences, New York. 110-117.
135. Beaton G.H., Cheney M.C. (1965) Vitamin B-6
Requirements of the Male Albino Rat. J Nutr 87: 125-
134.
136. Gilbert J.A., Gregory J.F. (1991) The Effect of
Pyridoxine-5 '-/3-D-glucoside on the Metabolic
Utilization of Pyridoxine in the Rat. FASEB J 1:A586
(abs. 1250).
137. Sauberlich H.E., Dowdy R.P., Skala J.H. (1974) Vitamin
B-6. In: Laboratory Tests for the Assessment of
Nutritional Status. CRC Press, Cleveland, OH. 37-49.
138. Committee on Enzymes of the Scandinavian Society for
Clinical Chemistry and Clinical Physiology (1974)
Recommended Methods for the Determination of Four
Enzymes in Blood. Scand J Clin Lab Invest 33: 291-
306.
139. Heinegard D., Tiderstrom G. (1973) Determination of
Serum Creatinine by a Direct Colorimetric Method. Clin
Chim Acta 43: 305-310.
140. Crosby W., Munn J.I., Furth F.W. (1954)
Standardization in a Method for Clinical
Hemoglobinometry. U.S. Armed Forces Med J 5: 693-703.
141. Hachey D.L., Coburn S.P., Brown L.T., Erbelding W.F.,
DeMark B., Klein P.D. (1985) Quantitation of Vitamin
B-6 in Biological Samples by Isotope Dilution Mass
Spectrometry. Anal Biochem 151: 159-168.
142. U.S. Department of Agriculture. (1986) Nationwide
Food Consumption Survey, Continuing Survey of Food
Intakes by Individuals, Low-Income Women 19-50 Years
and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII
Report No. 85-2. U.S. Department of Agriculture,
Hyattsville, Md.

105
143. U.S. Department of Agriculture. (1987) Nationwide
Food Consumption Survey, Continuing Survey of Food
Intakes by Individuals, Low-Income Women 19-50 Years
and Their Children 1-5 Years, 1 Day, 1986. NFCS, CSFII
Report No. 85-2. U.S. Department of Agriculture,
Hyattsville, Md.
144. Snell E.E., Haskell B.E. (1971) The Metabolism of
Vitamin B-6 In: Comprehensive Biochemistry. (Florkin
M., Stotz E.H., eds.) Vol. 21 Elsevier/North Holland,
Amsterdam. 41-71.
145. Cash C.D., Maitre M., Rumigny J.F., Mandel P. (1980)
Rapid Purification by Affinity Chromatography of Rat
Brain Pyridoxal Kinase and Pyridoxamine-5-phosphate
Oxidase. Biochem Biophys Res Comm 96: 1755-1760.
146. Kwok F., Churchich J.E. (1980) Interaction between
Pyridoxal Kinase and Pyridoxine-5-P Oxidase, Two
Enzymes Involved in the Metabolism of Vitamin B-6. J
Biol Chem 235: 882-887.
147. Daniels L.B., Coyle P.J., Chaio Y-B, Glew R.H., Labow
R.S. (1981) Purification and Characterization of a
Cytosolic broad Specificity beta-Glucosidase from Human
Liver. J Biol Chem 256: 13004-13013.
148. Merill A.H., Horiike K., McCormick D.B. (1978)
Evidence for the Regulation of Pyridoxal 5'-phosphate
formation in Liver by Pyridoxamine(pyridoxine)-5'-
phosphate Oxidase. Biochem Biophys Res Comm 83: 984-
990.
149. Perrella F.W. (1988) EZ-FIT: A Practical Curve-
Fitting Microcomputer Program for the Analysis of
Enzyme Kinetic Data on IBM-PC Compatible Computers.
Anal Biochem 174: 437-447.

106
Joyce Ann Gilbert was born in Washington, D.C. and
raised in Woodbridge, VA. Upon graduation from high school,
Joyce attended the University of South Carolina where she
received a Bachelor of Science degree in biology. Following
a brief respite from academia, Joyce returned to Clemson
University where she received a Master in Nutritional
Science and became a registered dietitian. After working in
clinical and research dietetics as well as in food
management, Joyce accepted an instructor position in the
Department of Food Science and Human Nutrition at the
University of Florida. Joyce relinquished her faculty
position to pursue a Ph.D. in human nutrition. Joyce has
accepted an assistant professor position at the Pennsylvania
State University upon completion of her dissertation. When
not engaged in the sciences, Joyce enjoys athletics, both as
a participant and fan, art and riding her Harley Davidson.

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Professor of Food Science
and Human Nutrition
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Lynri B. Bailey
Professor of Food Science
and Human Nutriton
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Peggy R. Borum
Professor of Food Science
and Human Nutrition
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Robert J. Cousins
Boston Family Professor of
Human Nutrition
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
a dissertation for the degree of Doctor of Philosophy.
Professor of Medicine
as

This dissertation was submitted to the Graduate Faculty
of the College of Agriculture and to the Graduate School and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy. I *
of tjfr
December 1991
/vt'>
Dean, College of
Agriculture
Dean, Graduate School



43
Measurement of radioactivity
HPLC fractions and extracts of tissue were measured for
radioactivity using a commercial scintillation fluid
(ScintiVerse LC, Fisher Scientific, Orlando, FL) and a liquid
scintillation spectrophotometer (Beckman LS 2800 Beckman
Instruments, San Ramon, CA) A quench curve for UC was used
for conversion of cpm to dpm.
Statistical analysis
In experiment 1 and 2 the distribution of vitamin B-6
compounds and total radioactivity between groups was compared
by the method of least squares analysis of variance using
general linear model procedures of SAS (123) following a log
tranformation of the data to normalize variance (123). Data
are reported as least squares mean the pooled standard error
of the least squares mean (SEM). In experiment 1, orthogonal
contrasts were made to examine the linear and quadratic
effects of different doses of PN-glucoside on dependent
variables. The quadratic effects were not significant for
any dependent variable. Values with p<0.05 were considered
statistically significant and were all for linear effects of
PN-glucoside unless otherwise stated. In expreriment 2 two
orthogonal contrasts were used to compare treatment means.
Contrast one compared group 2 with group 1 and 3; contrast two
compared group 1 with group 3.


50
with the quantity of unlabeled PN-glucoside administered
(p<0.01). This suggests a mechanism involving a possible
inhibition of PNP oxidase. Investigations of possible direct
inhibitory effects of PN-glucoside on PNP oxidase are in
progress. Comparatively high levels (36%) of the non-
phosphorylated B-6 vitamers were observed in the livers of the
rats in group 1, which may have been due to several factors.
The vitamin B-6 requirement for the rat is 6-7 mg PN/kg diet
(135). The experimental diet contained 7 mg PN/kg diet with
the rats consuming, on the average, 70 vitamin B-6 per day.
The dose of labeled-PN contained 240 nmol (49.3 /xg) [UC]PN.
Hence, during the 24 h post dose period, rats received 49.3 fig
vitamin B-6 above their requirement. In addition, a single
observation of tracer distribution at a time point 24 h post
dose, would not necessary reflect steady state concentrations
of endogenous vitamin B-6 in the liver. These factors would
account for greater than normal levels of B-6 metabolites in
liver.
The compensating changes observed in the proportions of
hepatic [UC]PMP and [UC]PLP (Table 3.1) cannot be readily
explained. Unexpectedly, these changes are consistent with
those observed when comparing the metabolism of [UC]PN in
vitamin B-6 adequate verses deficient rats (81). These
results indicate that the administration of unlabeled PN-
glucoside in this study caused changes in metabolic patterns


98
68. Tadera K., Mori E., Yagi F., Kobayashi K., Imada K.,
Imabeppu M. (1985). Isolation and Structure of a Minor
Metabolite of Pyridoxine in Seedlings of Pisum sativum
L. J Nutr Sci Vitaminol 31: 403-408.
69. Suzuki Y., Ishii K. Suga K., Uchida K. (1986).
Formation of beta-Glucosylpyridoxines in Soybean and
Rice Callus. Phytochem 25: 1331-1332.
70. Tadera K., Makamura M., Yagi F., Kobayashi A. (1979).
A Particulate Glucosyltransferase Catalyzing the
Formation of 5'-0-(Beta-D-Glucopyranosyl) Pyridoxine
from Pyridoxine: The Occurrence in the Seedlings of
Pisum sativum L. J Nutr Sci Vitaminol 25: 76-82.
71. Gregory J.F., Sartain D.B. (1991). Improved
Chromatographic Determination of Free and Glycosylated
Forms of Vitamin B-6 in Foods. J Agri and Food Chem
39: 899-905.
72. Suzuki Y., Uchida K. Tsuboi A. (1986). Enzymatic
Formation of Pyridoxine Beta-Glucosides by Wheat Bran
Beta-Glucosidase. Nippon Nogeikagaku Kaishi 53: 189-
196.
73. Tadera K., Yagi F., Dobayashi A. (1982). Specificity
of a Particulte Glucosyltransferase in Seedlings of
Pisum sativum L. which Catalyzes the Formation of 5'-0-
(Beta-D-Glucopyranosyl)Pyridoxine. J Nutr Sci
Vitaminol 28: 359-366.
74. Kabir H., Leklem J.E., Miller L.T. (1983).
Relationship of the Glycosylated Vitamin B-6 Content of
Foods to Vitamin B-6 Bioavailability in Humans. Nutr
Rept Int 28: 709-715.
75. Gregory J.F, Ink S.L. (1987). Identification and
Quantification of Pyridoxine-beta-Glucoside as a Major
Form of Vitamin B-6 in Plant Derived Foods. J Agrie
Food Chem 35: 76-82.
76. Iwami K., Yasumoto K. (1986). Synthesis of
Pyridoxine-beta Glucoside by Rice Bran beta-Glucosidase
and its In Situ Absorption in Rat Small Intestine.
Nutr Res 6: 407-414.
77. Ink S.L., Gregory J.F., Sartain D.B. (1986).
Determination of Pyridoxine beta-Glucoside
Bioavailability using Intrinsic and Extrinsic Labeling
in the Rat. J Agrie Food Chem 34: 857-862.


17
as well as total vitamin B-6 in food. The PN-glucoside, when
measured indirectly by hydrolysis, was reported to occur
widely in plant foods, ranging from 0% to 82% of the total
amount of vitamin B-6 present (74-76).
Studies by Iwami and Yasumoto (76) indicated that
intestinal absorption of PN-glucoside was similar to that of
free PN. However, Ink et al. (77) administered a single, oral
dose of radiolabeled PN and PN-glucoside in an alginate gel
diet using a combination of intrinsic and extrinsic labeling
in the intact rat. They reported that the absorption of the
glucoside was similar to that of unconjugated PN, but the
absorbed PN-glucoside was metabolically utilized less
effectively as vitamin B-6 than was PN. Iwami et al. (76)
measured the relative disappearance from the intestinal
perfusate of PN and a purified 4'-derivative (rather than the
naturally occurring 51-derivative). In the rat, Ink et al.
(77) observed the relative metabolism of the intrinsically and
extrinsically labeled glucoside to have less than 40% of the
bioavailability of PN.
The nutritional properties of these conjugated forms of
vitamin B-6 are not fully understood at the present time.
Although PN-glucoside comprises 25% to 75% of the total
vitamin B-6 in many plant tissues (67,75) its bioavailability
in humans continues to be studied. PN-glucoside has been
found to undergo intestinal absorption in the intact form.
Initial studies employing bioassays with human subjects (40)


96
45. Coburn S.P., Mahuren J.D., Szadkowska Z., Schaltenbrand
W.E., Townsend D.W. (1987). Kinetics of Vitamin B-6
Metabolism Examined in Miniture Swine by Continuous
Administration of Labelled Pyridoxine. In:
Mathematical Models in Experimental Nutrition (Canolty
N.L., Cain T.P., eds.) University of Georgia, Athens,
Georgia. 99-111.
46. McElroy L.W., Goss H. (1939). Report on Four Members
of the Vitamin B complex Synthesized in the Rumen of
Sheep. J Biol Chem 130: 437-438.
47. Coates M.E., Ford J.E., Harrison G.F. (1968).
Intestinal Synthesis of Vitamins of the B complex in
Chicks. Br J Nutr 22: 493-500.
48. Gregory J.F., Litherland S.A. (1986). Efficacy of the
Rat Bioassay for the Determination of Biologically
Available Vitamin B-6. J Nutr 116: 87-97.
49. Ikeda M., Hosotani T., Kurimoto K., Mori T., Ueda T.,
Kotake Y., Sakakibara B. (1979). The Differences of
the Metabolism Related to Vitamin B-6 Dependent Enzymes
among Vitamin B-6 Deficient Germfree and Conventional
Rats. J Nutr Sci Vitaminol 25: 131-139.
50. Ikeda M., Hosotani T., Ueda T., Kotake Y., Sakakibara
B. (1979). Effect of Vitamin B-6 Deficiency on the
Levels of Several Water Soluble Vitamins in Tissues of
Germfree and Conventional Rats. J Nutr Sci Vitaminol
25: 141-149.
51. Hughes E.H., Squibb R.L. (1942). Vitamin B-6 in the
Nutrition of the Pig. J Animal Sci 1: 320-325.
52. Wozenski J.R., Leklem J.E., and Miller L.T. (1980).
The Metabolism of Small Doses of Vitamin B-6 in Men.
J Nutr 110 (2): 275-285.
53. Schultz T.D., Leklem J.E. (1987). Vitamin B-6 Status
and Bioavailability in Vegetarian Women. Am J Clin
Nutr 46: 647-51.
54. Sumi, Y., Miyadawa, M., Kansaki, M., and Kotake, Y.
(1977). Vitamin B-6 Deficiency in Germ Free Rats. J
Nutr 107: 1707-1714.
55. Kabir, H., Leklem, J.E., and Miller, L.T. (1983).
Comparative Vitamin B-6 Bioavailability from Tuna,
Whole Wheat Bread and Peanut Butter in Humans. J Nutr
113: 2412-2420.


79
Table 5.2 Kinetic parameters of pyridoxal kinase with varying
concentrations of PN-glucoside.
PN-glucoside K,,, V
concentration (/iM) (nmol PNP/h/mg protein)
(MM)
0
2 32.0
50
222.2
100
212.4
150
203.0
0.29.01
0.28.01
0.28.01
0.27.01
PMP (PNP^ oxidase
The purification scheme yielded a preparation of 24-fold
increase in enzyme purity compared to the crude homogenate
(Table 5.3). The conversion of the substrate, PMP, to PLP has
previously been shown to exhibit a value of 25/iM (8) The
present study indicated a similar closely related value of
23/xM. Values for Km of the substrate in the presence of all
concentrations of PN-glucoside were not significantly
different (Table 5.4). The maximum velocity values in this
study did not differ signifcantly from the control value of
1.65 nmol PLP/min/mg protein (Table 5.4). All kinetic
parameters (Figure 5.2) were calculated using the EZ-FIT
Program (149).


Mass spectral analysis of deuterium-labeled
4 PA 59
Statistical analysis 60
Results 61
Vitamin B-6 nutritional status of subjects 61
Stable-isotopic trials 61
Discussion 65
CHAPTER 5
EFFECTS OF PYRIDOXINE-5' -/3-D-GLUCOSIDE ON THE
IN VITRO KINETICS OF PYRIDOXAL KINASE AND PYRIDOXAMINE
(PYRIDOXINE)-5'PHOSPHATE OXIDASE IN RAT LIVER 69
Introduction 69
Materials and Methods 70
Protocol 70
Forms of vitamin B-6 71
Pyridoxal kinase 72
Pyridoxamine (pyridoxine) 5' phosphate
oxidase 75
Statistical analysis 77
Results 77
PL kinase 77
PMP (PNP) oxidase 79
Discussion 83
CHAPTER 6
SUMMARY AND CONCLUSIONS 86
LITERATURE CITED 92
BIOGRAPHICAL SKETCH 106
V


95
35. Toepfer E.W., Polansky M.M., Richardson L.R., Wilkes S.
(1963). Comparison of Vitamin B-6 Values of Selected
Food Samples by Bioassay and Microbiological Assay. J
Ag Food Chem 11: 523-525.
36. Nguyen L.B., Gregory J.F. (1983). Effects of Food
Composition on the Bioavailability of Vitamiln B-6 in
the Rat. J Nutr 113: 1550-1560.
37. Gregory J.F., Ink S.L. (1985). The Bioavailability of
Vitamin B-6. In: Vitamin B-6: Its Role in Health and
Disease. Alan R. Liss, Inc., New York. 3-23.
38. Nelson E.W., Lane H., Cerda J.J. (1976). Comparative
Human Intestinal Bioavailability of Vitamin B-6 from a
Synthetic and a Natural Source. J Nutr 106: 1433-
1437.
39. Nelson E.W., Burgin C.W., Cerda J.J. (1977).
Characterization of Food Binding of Vitamin B-6 in
Orange Juice. J Nutr 107: 2128-2134.
40. Leklem J.E., Miller L.T., Perera A.D., Peffers D.E.
(1980). Bioavailability of Vitamin B-6 from Wheat Bread
in Humans. J Nutr 110: 1819-1828.
41. Gregory J.F. (1990). The Bioavailability of Vitamin
B-6 Recent Findings. Ann NY Acad Sci 585: 86-95.
42. Coburn S.P., Mahuren J.D. (1979). Major Urinary
Metabolites of 4' and 5'-Deoxypyridoxines in Various
Species. IRCS Med Sci 7: 556.
43. Rodwell V.W., Volcani B.E., Ikawa M., Snell E.E.
(1958). Bacterial Oxidation of Vitamin B-6.
Isopyridoxal and 5-Pyridoxic Acid. J Biol Chem 233:
1548-1554.
44. Coburn S.P., Hahuren J.D., Erbelding W.F., Tokwnsend
D.W., Hachey D.L., Klein P.D. (1984). Measurements of
Vitamin B-6 Kinetics in Vivo Using Chronic
Administration of Labelled Pyridoxine. In: Chemical
and Biological Aspects of Vitamin B-6 Catalysis, Part A
(Evangelopoulos A. E., ed.) Alan R. Liss, New York.
43-54.


91
The predominant form of vitamin B-6 in many fruits and
vegetables is PN-glucoside and research has indicated that
not only is this vitamin B-6 compound biologically less
available as a usable source of vitamin B-6 it also effects
the metabolic utilization of other B-6 vitamers. In
development of recommendations for dietary allowances and
dietary food intake patterns it is necessary to consider not
only vitamin B-6 availability and biological activity, but
interactions that occur between different vitameric forms and
how these interactions effect bioavialability of the nutrient.


84
metabolically trapped via phosphorylation. The vitamin is
maintained predominantly in phosphorylated forms due mainly to
protein binding and a ten fold greater activity of PL kinase
than phosphatase. It has been shown that hydrolysis of PLP is
inhibited by substrate and PMP (112).
Rat liver PMP (PNP) oxidase is inhibited by its product,
PLP, and the presence of PLP increases both the and
(148). The increase in Vmax may be a result of the action of
the more active rat liver phosphatases or an inability to
accurately measure initial velocity due to product inhibition
at early reaction times (148). Phosphatase more rapidly
hydrolyzes PLP when this product is free rather than protein
bound (144). The partially purified enzyme preparation used
in this study showed minimal phosphatase activity.
Approximately 20% of the PLP in vivo is in the free or loosely
bound form and the remaining PLP is bound to proteins.
Therefore, in vivo product inhibition may differ significantly
from in vitro inhibition. However, studies have shown that
crude rat liver homogenate compared with pure enzyme is
inhibited to a similar extent (148). Whether or not PLP
inhibits PMP (PNP) oxidase is dependent on the concentrations
of available substrate and product formed. Greater substrate
inhibition has been reported with PNP, in concentrations as
small as 5 im, due to a reduced enzyme-substrate complex that
reacts less efficently with oxygen (112). This has
significant consideration, since the presence of increasing


24
B-6 in meats. Studies conducted by Ink et al. (90) used a
combination of intrinsic and extrinsic labeling of liver and
muscle tissue with radiolabeled forms of vitamin B-6. These
data indicated that thermal processing caused partial
destruction of vitamin B-6 in liver and muscle, but had little
or no effect on the bioavailability of the remaining vitamin
B-6.
The bioavailability of vitamin B-6 in foods from plant
origin has been proposed to be adversely influenced by dietary
fiber (36-37). The effect of various forms of dietary fiber
on vitamin B-6 bioavailability has been examined in both
animals and humans.
The effect of dietary fiber on the bioavailability of B-6
vitamers was first studied by Leklem et al. (40) who employed
a human bioassay to examine the utilization of vitamin B-6 in
white verse whole wheat bread. Their data indicated 10% lower
availability of the vitamin B-6 which was attributed to
reduced intestinal absorption by the fiber components. The
results of these studies suggested that the vitamin B-6 of
soybeans was approximately 5% to 10% less available than beef
vitamin B-6. These data have suggested that the presence of
nondigestible components of dietary fiber, may reduce the
bioavailability of the vitamin B-6. Physical properties
inherent of dietary fiber may influence the absorption of the
vitamin through entrapment or alteration of the viscosity of
the intestinal contents (16).


74
substrate followed by an inorganic phosphate assay kit (Sigma
Chemical Co., St. Louis, MO) and measuring the percent
conversion of PLP to PL.
Kinase assay. The assay procedure used was that of
Merrill and Wang (116). All reagents and buffers were at pH
5.75 and the procedure was conducted in GEF40G0 gold lights
that emit at a wavelength between 500 nm and 750 nm to prevent
photodegradation of pyridoxyl compounds. The assay mixture
contained 10 /xL each of 0.2M potassium phosphate, lOmM ATP (
Sigma Chemical Co., St. Louis, MO), 0.8mM ZnCl2, 0.6mM KCl,
0.3mM (20/iCi/mmol) [3H]PN diluted with unlabeled PN to yield
the appropriate specific activity, and 100 g of enzyme
extract with enough water to make a final volume of 100 /L.
The mixture was incubated for 30 min at 37C and the reaction
halted with 0.5 ml ice cold 10 mM ammonium formate. The
samples were transfered to columns which contained 0.2 ml of
DEAE-cellulose (Sigma Chemical Co., St. Louis, MO). The
columns were rinsed with 12 ml of lOmM ammonium formate,
followed by 2 ml water and eluted with 2 ml of 0.5 M KCl. A
2 ml fraction of the final elution was collected and added to
15 ml ScintiVerse II (Fisher Scientific, Fair Lawn, NJ) and
counted for total radioactivity. The reaction rates were
calculated using the following equation of Merrill and Wang
(116):
. ^P^Veaction CPmtime 0
Enzyme units = X 2 X quench factor
Specific Activity
(enzyme preparation)


53
glycosylated vitamin B-6 in the diet may be useful as an index
of vitamin B-6 bioavailability. The nutritional significance
of the effect of PN-glucoside on the metabolic utilization of
other forms of vitamin B-6 is based on the fact that PN-
glucoside is a major naturally occurring form of vitamin B-6
in many fruits and vegetables of human diets (65,75). The
results of the present studies suggest that PN-glucoside is
nutritionally significant in two ways. As shown previously,
PN-glucoside can act as a source of partially available
vitamin B-6 (77,81-82). In addition, this study has shown
that PN-glucoside can act as a weak antagonist that may hinder
the utilization of PN and possibly other forms of the vitamin.
These results indicate that PN-glucoside alters the
metabolism and in vivo retention of [UC]PN in the rat and that
PN-glucoside may retard the utilization of nonglycosylated
forms of vitamin B-6. PN-glucoside represents a significant
proportion of the total vitamin B-6 in many plant derived
foods (75) Therefore, the results of this study are
important in assessing the bioavailability of vitamin B-6 in
a typical mixed diet. The assessment of vitamin B-6 status in
human populations is contingent on identifying and quantifying
the content and distribution of the different forms of the
vitamin found in foods, determining the potential interactions
between these forms and the relative bioavailability of the
different vitamers. If it is found that PN-glucoside affects
the metabolic utilization of PN similarly in humans, it is



PAGE 1

THE EFFECTS OF PYRIDOXINE 5 1 -/3-D-GLUCOSIDE ON THE METABOLIC UTILIZATION OF PYRIDOXINE IN RATS AND HUMANS By JOYCE ANN GILBERT A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1991 ufJiosiryjERORiDA libraries

PAGE 2

This work is dedicated to my parents, Helen and Reginald, who showed me life has no boundaries, only opportunities.

PAGE 3

ACKNOWLEDGEMENTS I thank Drs. Lynn Bailey, Peggy Borum, Robert Cousins and James Cerda for allowing me the previlage of studying with the best scientists in their respective fields. Special thanks go to my advisor and research mentor, Dr. Jesse Gregory, for his example of integrity and dedication in all endeavors. I also thank Doris Sartain and my collegues in the "Yellow Lab" for their endless patience and sensitivity. iii

PAGE 4

TABLE OF CONTENTS ACKNOWLEDGEMENTS iii LIST OF TABLES vi LIST OF FIGURES vii Abstract viii CHAPTER 1 INTRODUCTION 1 CHAPTER 2 LITERATURE REVIEW 4 Bioavailability of Vitamin B-6 4 Pyridoxine-/3-D-Glucoside 15 Factors Affecting the Bioavailability of Vitamin B-6 21 Enzymatic Interconversion of B-6 Vitamers .... 31 CHAPTER 3 PYRIDOXINE-5'-0-D-GLUCOSIDE AFFECTS THE METABOLIC UTILIZATION OF PYRIDOXINE IN RATS 38 Introduction 38 Materials and Methods 39 Protocol 39 Forms of vitamin B-6 40 Sample preparation 41 HPLC equipment and analysis 42 Measurement of radioactivity 43 Statistical analysis 43 Results 44 Discussion 48 CHAPTER 4 EFFECTS OF PYRIDOXINE-5 1 -/3-D-GLUCOSIDE ON THE METABOLIC UTILIZATION OF PYRIDOXINE IN HUMANS .... 55 Introduction 55 Materials and Methods 56 Synthesis of forms of vitamin B-6 56 Protocols of trials with human subjects ... 57 Analytical methods 59 iv

PAGE 5

Mass spectral analysis of deuterium-labeled 4 PA 59 Statistical analysis 60 Results 61 Vitamin B-6 nutritional status of subjects . 61 Stable-isotopic trials 61 Discussion 65 CHAPTER 5 EFFECTS OF PYRIDOXINE-5 ' -0-D-GLUCOSIDE ON THE IN VITRO KINETICS OF PYRIDOXAL KINASE AND PYRIDOXAMINE (PYRIDOXINE) -5' PHOSPHATE OXIDASE IN RAT LIVER 69 Introduction 69 Materials and Methods 70 Protocol 70 Forms of vitamin B-6 71 Pyridoxal kinase 72 Pyridoxamine (pyridoxine) 5 1 phosphate oxidase 75 Statistical analysis 77 Results 77 PL kinase 77 PMP (PNP) oxidase 79 Discussion 83 CHAPTER 6 SUMMARY AND CONCLUSIONS 86 LITERATURE CITED 92 BIOGRAPHICAL SKETCH 106 v

PAGE 6

LIST OF TABLES Table 3.1 Liver B-6 vitamer distribution and total liver U C in rats administered varying levels of PNglucoside in the dose (Experiment 1) .* 44 Table 3.2 Urinary 4 PA and total urinary 14 C in rats administered varying levels of PN-glucoside (Experiment 1)* 45 Table 3.3 Liver B-6 vitamer distribution and total liver H C in rats administered varying levels of PN or PN-glucoside. (Experiment 2)* 46 Table 3.4 Urinary 4 PA and total urinary 14 C in rats administered varying levels of PN or PN-glucoside. (Experiment 2)* 47 Table 4.1 Indicator of vitamin B-6 nutritional status of human subjects 24 h prior to each trial 1 . ... 62 Table 4.2 Molar isotopic ratio of urinary d 2 4PA. 1 . . 63 Table 4.3 Percentage of d 2 PN (5/xmol) dose excreted as urinary d 2 4PA 64 Table 4.4 Total urinary 4 PA excretion. 1 65 Table 5.1 Purification of pyridoxal kinase from rat liver 78 Table 5.2 Kinetic parameters of pyridoxal kinase with varying concentrations of PN-glucoside 79 Table 5.3 Purification of pyridoxamine (pyridoxine) phosphate oxidase from rat liver 80 Table 5.4 Kinetic parameters of pyridoxamine (pyridoxine) phosphate oxidase with varying concentrations of PN-glucoside 80 vi

PAGE 7

LIST OF FIGURES Figure 5.1 Double reciprocal plot for the phosphorylation of PN by PL kinase from rat liver in the presence of various levels of PNglucoside 81 Figure 5.2 Double reciprocal plot for the conversion of PMP to PLP by PMP (PNP) oxidase from rat liver in the presence of various levels of PN-glucoside. . 82 vii

PAGE 8

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE EFFECTS OF PYRIDOXINE 5 1 -0-D-GLUCOSIDE ON THE METABOLIC UTILIZATION OF PYRIDOXINE IN RATS AND HUMANS By Joyce Ann Gilbert December, 1991 Chairperson: Dr. Jesse F. Gregory III Major Department: Food Science and Human Nutrition A major form of vitamin B-6 in plant-derived foods is 5'0-(/3-D-glucopyranosyl) pyridoxine (PN-glucoside) . Previous studies have shown that PN-glucoside is poorly available as a source of vitamin B-6 in rats and undergoes incomplete utilization in humans. The present research was conducted to determine whether unlabeled PN-glucoside affects the metabolic utilization of simultaneously administered isotopically labeled pyridoxine (PN) in rats and humans. In addition, the in vitro effect of PN-glucoside on the activity of enzymes in the vitamin B-6 metabolic pathway, specifically pyridoxalkinase and pyridoxamine (pyridoxine) -phosphate oxidase was determined. Experimental results with rats given [ 14 C]PN indicated that urinary excretion of U C increased significantly with viii

PAGE 9

increasing dose of PN-glucoside, while hepatic U C decreased significantly as the PN-glucoside dose increased. The proportion of hepatic 14 C-labeled pyridoxal, PN, and pyridoxamine decreased whereas hepatic pyridoxine phosphate and pyridoxal phosphate increased in proportion to the PNglucoside dose. In addition, the concentration of urinary [ U C] 4-pyridoxic acid (4PA) , relative to total urinary U C, decreased as the dose of PN-glucoside increased. Stable isotopic methodology was employed to determine whether PNglucoside affected the metabolic utilization of simultaneously administered deuterium-labeled PN (d 2 PN) in humans. Experimental results showed that twenty-four hour urinary excretion of 4 PA was decreased significantly with increasing dose of PN-glucoside. The percentage of ingested d 2 PN excreted as d 2 4PA showed an inverse relationship, that was statistically significant, in proportion to the PN-glucoside dose. In vitro enzyme assays indicated that PN-glucoside had no significant effect on the activity of partially purified pyridoxal-kinase and pyridoxamine (pyridoxine) phosphate oxidase. These results provide evidence that PN-glucoside weakly retards the metabolic utilization of nonglycosylated forms of vitamin B-6. However, the effect of PN-glucoside on PN is not due to the direct effect of PN-glucoside on the enzymes PL kinase and PMP (PNP) oxidase. ix

PAGE 10

CHAPTER 1 INTRODUCTION Vitamin B-6 has often been referred to as the protein vitamin because of its association with amino acid metabolism and the influence of dietary intakes of protein on vitamin B-6 requirements (1-3). Adequate vitamin B-6 nutriture is essential to health through its numerous roles in the body as the active coenzyme form, pyridoxal 5' -phosphate (PLP) . A large majority of these roles involves metabolism of amino acids. These functions include, among others, nonoxidative decarboxylation of amino acids, transamination, desulf hydration, and enzymes affecting reactions of amino acid side chains. Such PLP dependent enzymes are important functional complexes in the biosynthesis and catabolism of essential and non-essential amino acids and some provide a connection between the amino acid and intermediates of carbohydrate metabolism. Recent research in animal models has suggested a role for PLP in the modulation of hormone actions (4-5). Nutritional status with respect to vitamin B-6 is influenced by the amount of the vitamin ingested, the extent 1

PAGE 11

2 of absorption and metabolic utilization of the B-6 compounds in the diet, and the specific requirements of the individual. Although vitamin B-6 is widely distributed in nature, the amount in food is relatively small. The American diet often contains 1 2 mg per day (2) . Marginal vitamin B-6 nutritional status has been well documented in several segments of the American population including adolescents, pregnant women and the elderly (6-15) . However, vitamin B-6 deficiency with apparent clinical symptoms is not widespread in the general population. It is significant that over 50% of the subjects of the USDA 1977-78 Nationwide Food Consumption Survey consumed less than 70% of the Recommended Dietary Allowance (RDA) for vitamin B-6 (2) . The 1985 Continuing Survey of Food Intakes of Individuals indicated that only 27% of women consumed 70% or more of the RDA for vitamin B-6 (12) . Women of all ages and the elderly have been especially prevalent in the group consuming less than the RDA for vitamin B-6. Vitamin B-6 is required in amino acid metabolism and therefore, the RDA for this vitamin is related to protein intake. Twice the RDA for protein, which is considered the upper boundary of acceptable level of protein intake, was used to establish the RDA for vitamin B-6. The RDA for vitamin B-6 was revised using a dietary ratio of 0.016 mg vitamin B-6/g protein. Our present knowledge of the essential role of vitamin B6 in preserving health emphasizes the importance of

PAGE 12

3 establishing an accurate RDA for vitamin B-6. Establishing an adeguate RDA for vitamin B-6 is a complex task. The appropriateness of the RDA for vitamin B-6 is dependent on accurate information on the content of the vitamin in foods consumed, the enzymes involved in the metabolism of the various forms of vitamin B-6, as well as information regarding vitamin B-6 bioavailability.

PAGE 13

CHAPTER 2 LITERATURE REVIEW Bioavailability of Vitamin B-6 The concept of bioavailability is most appropriate when considering the extent of intestinal absorption and metabolic utilization and any potential antagonistic effects of all naturally occurring dietary forms of the vitamin. Hence, net bioavailability is defined as the portion of total dietary vitamin B-6 that is biologically active following intestinal absorption (16) . Vitamin B-6 as it occurs in mammalian tissues and fluids encompasses three interconvertible 3hydroxy-2-methylpyridine compounds including pyridoxine (PN) , pyridoxal (PL), pyridoxamine (PM) , their corresponding 5 1 phosphoryl derivatives (PNP, PLP, and PMP respectively) and the excretory catabolite 4-pyridoxic acid (4 PA) (17) . Vitamin B-6 in nature is found in phosphorylated forms and closely associated with proteins. In foods, PL, PN, and PM are found as both phosphorylated and nonphosphorylated forms. PL, PM, PLP, and PMP are the predominant forms in animal products, while PN, in its free vitaminic form and as the glycoside (discussed in a later section) are the major forms in most plant foods (18) . 4

PAGE 14

5 To quantitate total vitamin B-6 in most food, test microorganisms are generally utilized. The most common organism used is Saccharomyces uvarum . Test microorganisms demonstrate low biological activity of vitamin B-6 in its bound (pyridoxyl-amino acid and amines) glycosylated and phosphorylated forms until they are released by acid hydrolysis. However, this acid treatment of food is not entirely representative of the digestive process in the gastrointestinal tract. Results of early studies of intestinal absorption of vitamin B-6 indicated that absorption occurs by a simple passive diffusion mechanism (19-22) . A number of studies indicate that the rate of uptake by intestinal tissue of the B-6 vitamers, PN, PM, and PL increased proportionally to the dose over a wide range of intralumenal concentrations (19,2331) . Transport of PN across the intestinal wall occurred by diffusion, independent of accumulation within the tissue. There is no evidence of a saturable transport system either in vivo (19,23-24) or in vitro (23,26-31). These data also indicate that transport across the intestinal wall from the mucosal to the serosal side of the non-phosphorylated B-6 vitamers to be quantitatively similar. PN in the free and conjugated forms is rapidly absorbed by the intestine directly or following hydrolysis by intestinal enzymes. PL and PM are present in food primarily as the 5 1 -phosphates and require hydrolysis by lumenal phophatases. The phosphorylated B-6

PAGE 15

vitamers, PLP and PMP are well absorbed following enzymatic dephosphorylation, although these vitamers are slowly absorbed as the intact 5'-phosphate esters (28-30). Studies have indicated that the transport of the B-6 vitamers into the intestinal mucosal cell involves a saturable enzymatic process, identified as intracellular phosphorylation by PL kinase (EC 2.7.1.35) (24, 28-29). Most ingested vitamin B-6 is absorbed by the jejunum in the nonphosphorylated forms, PL, PN, and PM. Tissues, especially liver, rapidly take up circulating vitamin B-6, where the phophorylated B-6 vitamers are hydrolyzed by plasma membrane phosphatases and enter the cells by a facilitated process and diffussion followed by metabolic trapping (28-29) . PN is phosphorylated to PNP by PL kinase which has been detected in all mammalian tissues investigated (32) . PNP is then converted to PLP by flavindependent PNP oxidase (EC 1.4.3.5) (33) which, in contrast to the widely distributed PL kinase, is found in few tissues, mainly liver, erythrocytes, kidney, and brain (34-35) . PLP can be transformed to PMP by transamination or hydrolyzed to PL by phosphatases (17) . PL is then converted to pyridoxic acid (4 PA) by aldehyde oxidase (18) or again to PLP by PL kinase. PLP is bound by cellular protein or released into plasma, by the cell, as PLP, PL or 4 PA. The analysis of kinase and phosphatase enzymes in liver tissue indicates similar activities. The relative activities of these enzymes account for the accumulation of the 5 '-phosphate compounds;

PAGE 16

7 however, it also allows for the dephosphorylation of PLP that is not protein bound and its release from the liver as PL or oxidized to 4 PA (31) . Regulation of vitamin B-6 metabolism may also occur through the conversion of PNP and PMP to PLP, which is highly sensitive to product inhibition (30) . Recent advances in analytical techniques have enabled collection of more precise data concerning the quantity and various forms of vitamin B-6 compounds present in the diet. However, present knowledge of the bioavailability of vitamin B-6 is still not sufficient enough to determine an accurate assessment of adequate dietary intake. The bioavailability of a nutrient such as vitamin B-6 is primarily determined experimentally by comparing the concentration of vitamin B-6 that is biologically active to total vitamin B-6 ingested (16) . Traditionally biologically available vitamin B-6 has been determined through animal bioassays (36) . Human bioassays, to determine biologically active vitamin B-6 in food, have been performed, but are generally lengthy procedures with limited precision. Significant advances and recent applications in isotopic methodology offers a useful alternative to bioassay for the sutdy of bioavailability of vitamin B-6 (36) . Results of studies utilizing rat bioassays to compare animal products with plant-derived products have indicated that the bioavailability of vitamin B-6 in animal products is greater than that of plant-derived products (36). In general, a poor

PAGE 17

8 correlation was observed in these samples when comparing rat growth and plasma PLP concentration. These results were difficult to interpret because of the potential effect of diet composition on the synthesis of vitamin B-6 by intestinal microflora. The lack of agreement between rat growth and plasma PLP, suggests that data from bioassays to determine biologically available vitamin B-6 in food sources, specifically those in which the test diets may differ greatly from the reference diet in the type of carbohydrate, are somewhat eguivocal (37) . The bioavailability of vitamin B-6 has been studied in somewhat more detail than many of the other vitamins. However, few generalizations can be stated regarding the overall bioavailability of the B-6 vitamers or the factors influencing it. Essential to measuring the bioavailability of vitamin B-6 in a food source is the determination of the total vitamin B-6 content in the food. This is complicated by the interaction of the vitamin with food components. These interactive compounds may or may not be biologically available. The interactions of vitamin B-6 with food components and with other vitamin B-6 compounds are important considerations since they may represent sources of available vitamin B-6 in the diet. Experimental data regarding the bioavailability of naturally occurring vitamin B-6 in foods were initially reported by Sarma et al. (34). The authors utilized a comparison of rat bioassay and Saccharomvces uvarum assay

PAGE 18

9 results for a variety of plant and animal products. Liver fractions, whole wheat and yellow corn exhibited bioavailability values of between 65% and 70% compared with an apparent value of approximately 100%. These results were the first indication of a wide variation and potentially incomplete bioavailability of vitamin B-6 in cereal grains and other food products (34) . A comparison of rat growth bioassays and Saccharomyces uvarum data for dried beef, lima beans, non-fat dry milk and whole wheat flour yielded a good correlation between the assays with small differences obtained for the flour and milk products (35). Nelson et al. (38) examined the rate of intestinal absorption of the vitamin B-6 in orange juice in human subjects. A triple lumen perfusion technigue was employed to determine the relative absorption of naturally occurring vitamin B-6 in orange juice. Their results demonstrated a significant decrease in rate of absorption of vitamin B-6 in orange juice (42%) than that of synthetic B-6 vitamers in saline (67%) and saline-glucose solutions (79%) . The lower rate of absorption of the naturally occurring vitamin B-6 in orange juice was thought to be a result of interactions of the vitamin with naturally occurring food components in the orange juice (38) . A follow-up study conducted by Nelson et al. (39) investigated the nature and extent of binding of different forms of vitamin B-6 in orange juice. The study suggested extensive and egual binding of

PAGE 19

10 both PL and PN in orange juice to a small dialyzable molecule which is heat stable and non-protein in nature. The interaction of vitamin B-6 with this component of orange juice was suggested as being responsible for the lower rate of absorption of naturally occurring vitamin B-6 in orange juice. Leklem and coworkers (40) were the first to employ a bioassay with human subjects to examine the effect of dietary fiber on the bioavailability of the vitamin. The study investigated the bioavailability of vitamin B-6 in white and whole wheat bread. Significant differences in the bioavailability of B-6 vitamers were not apparent when plasma PLP and erythrocyte aminotransferase data were examined. However, fecal vitamin B-6 and urinary 4 PA excretion data suggested incomplete utilization of the vitamin in the whole wheat bread. Frequent attempts have been made to calculate the vitamin B-6 balance of B-6 vitamers and 4 PA in bioassays employing human subjects (41) . Similar experiments in which the use of data concerning fecal vitamin B-6 have shown little validity. Unabsorbed vitamin B-6 from dietary sources as well as microbiologically synthesized B-6 vitamers are contained in fecal material. The composition of the diet will influence the microbial contribution to fecal vitamin B-6 which may affect the apparent vitamin B-6 nutriture of the subject. The fact that the intestinal microorganisms produce vitamin B-6 is well established (41-46) . However, the availability of vitamin B-6

PAGE 20

11 to the host organism, especially in the absence of coprophagy, has been unclear (45,47). Since several animals can develop a vitamin B-6 deficiency even when coprophagy is allowed, metabolic utilization of microbial vitamin B-6 must be miminal (48) . In animals exposed to conditions of nutritional deficiency, coprophagy may increase sufficiently to become a significant source of nutrients. This may explain the results of Ikeda et al. (49-50), who observed that germfree rats tend to be more susceptible to vitamin B-6 deficiency than rats with normal populations of intestinal microorganisms. Hughes et al. (51) reported that in addition to the intestinal microflora, nutritionally deprived animals may also benefit from other environmental sources of microbial vitamins such as microbial activity on the floor of the pen. As indicated by these studies, there exists no compelling evidence that vitamin B-6 can be absorbed from the large intestine in amounts sufficient to make a detectable contribution to the daily intake. A study (52) involving human bioassay in measuring bioavailability of vitamin B-6 in beef compared with soybeans suggested the vitamin B-6 in soybeans to be less available compared to that contained in beef. It has been suggested that the presence of nondigestible polysaccharides and lignin components of dietary fiber of the plant derived foods is responsible for differences in the availability of vitamin B-6 in plant foods compared with animal products (40) . A variety

PAGE 21

12 of physical and chemical properties of dietary fiber suggest the possibility of binding or entrapment of the B-6 vitamers, which may influence intestinal absorption. Schultz and Leklem (53) did a study comparing vegetarian versus nonvegetarian women. They observed that although the vegetarian women consumed more crude fiber than the nonvegetarian women, there was no significant differences between these two groups for plasma PLP, urinary 4PA, and urinary vitamin B-6. Therefore, it was concluded that there appeared to be no adverse effect of fiber on the bioavailability of vitamin B-6 between these groups (53) . It has been reported that human intestinal microflora produce vitamin B-6 (54) . Several bioavailability studies have indicated that ingestion of diets high in dietary fiber or carbohydrate leads to significant increases in microbial synthesis of vitamin B-6 (55-56) . Tarr et al. (57) , evaluated a typical American mixed diet to its bioavailability of vitamin B-6. They reported 70% bioavailability of vitamin B-6 based on urinary 4 PA and plasma PLP concentration relative to PN in a formula diet. The authors speculated that thermal processing was potentially responsible for the incomplete bioavailability of vitamin B-6 because canned goods, including both animal and plant foods, composed much of the mixed diet. A study by Nguyen and Gregory (36) employed rat bioassays to examine the bioavailability of vitamin B-6 in selected foods as influenced by thermal processing. The effects of food composition and

PAGE 22

13 thermal processing on the relative bioavailability of vitamin B-6 in beef, spinach, potato and cornmeal were assessed. The results further demonstrated that diet composition and food processing must be considered in evaluating the bioavailability of vitamin B-6. The instability of certain B-6 vitamers during food processing and storage may contribute to losses of the nutritional guality of foods with respect to vitamin B-6. Processing and storage of foods have been found to have various effects on vitamin B-6 with losses decreasing the adeguacy of food products as sources of dietary vitamin B-6. A deficiency of vitamin B-6 in infants who consumed a nonfortified, heat sterilized canned infant formula, led to extensive research, during the 1950 's, concerning the effects of food processing on the bioavailability of vitamin B-6 (58) . Hassinen et al. (59) demonstrated that the two main naturally occurring B-6 vitamers in milk, PM and PL, were much less stable than added synthetic PN. The research of Tomarelli et al. (60) reported that the retorting of milk and infant formula induced large losses in the availability of naturally occurring vitamin B-6 in these products. Very few studies have evaluated the thermal stability of vitamin B-6 in low moisture food systems. Vitamin fortification is common practice in the food industry and will more than likely continue into the future with more and more types of products being fortified. Vitamin B-6, in the form

PAGE 23

14 of PN-HC1, is added to many breakfast cereals at levels of 25% to 100% of the United States Recommended Daily Allowance (USRDA) per ounce. Gregory et al. (61-62), using a dehydrated model food system, simulating breakfast cereals, indicated that the roasting and storage of low moisture food systems resulted in losses of 50% to 70% of the added PN,PM,and PLP. The remaining vitamin B-6 was found to be fully available as determined by rat bioassay using growth, feed efficiency, erythrocyte aspartate aminotranferase activity (AspAT) and in vitro (AspAT) coenzyme stimulation. Gregory (63) also examined vitamin B-6 bioavailability in rice-based, PNfortified cereal and non-fat dry milk. The non-fat milk and rice base breakfast cereal samples were analyzed for vitamin B-6 by microbiological, HPLC, and rat bioassay procedures. The results indicated that the vitamin B-6 of the non-fat dry milk was fully available, while the vitamin B-6 availability in the cereal product was comparatively low. The apparent losses of the PN in the fortified cereal was explained by a first order kinetic model described by Evans et al. (64). These studies (61,63-64) indicated that food fortified with PN is susceptible to significant degradation under certain processing and storage conditions and that the bioavailability of the remaining vitamin B-6 may not be complete. The above results suggest that thermal processing of foods does not induce nutritionally important losses in the bioavailability of vitamin B-6. At the very least, research to date indicates

PAGE 24

15 that any possible adverse effects of thermal processing and storage on the bioavailability of vitamin B-6 would not be of sufficient magnitude to explain the incomplete utilization of the vitamin observed in human subjects (57,65). Pyridoxine-ff-D-Glucoside A conjugated form of pyridoxine, 5'-0-(/9-Dglucopyranosyl) pyridoxine (PN-glucoside) has been found to be a major naturally occurring form of vitamin B-6 in many fruits and vegetables of human diets. Scudi et al. (66) first reported the presence of conjugated forms of the vitamin B-6 in 1942 to occur in rice bran. Some thirty-five years later, Yasumoto et al. (67) reported that the PN component of the conjugated form of vitamin B-6 in rice bran was in a 1:1 ratio with glucose and identified this glycosylate as a 5'glucopyranosyl-derivative of PN. Several other studies (6869) have indicated the synthesis of this glucoside in pea seedlings as well as the formation of other PN conjugates in lesser concentrations than the 5 1 -pyridoxine-glucoside derivative. Following the first isolation and structural identification of 5 1 -0(j3-D-glucopyranosyl) pyridoxine (PNglucoside) from rice bran (67), the possible nutritional significance of glycosylated forms of vitamin B-6 has been a subject of intense study. Evidence for the presence of conjugated forms of vitamin B-6 existing as a variable proportion of the total vitamin B-6 of plant-derived foods has

PAGE 25

16 been observed as well as evidence supporting the widespread existence of /3-glucoside conjugates of B-6 vitamers in plant tissues (68) . PN-glucoside appears to be formed in plants by enzymatic transglycosylation (70-73) . The glucose moiety of PN-glucoside occurs at the 5' and 4 1 positions of the PN ring; however, the presence of the 5 '-isomer as the primary naturally occurring form of PN-glucoside suggests specificity for this position by the transglycosylation reaction in most plant tissue (67) . These studies provide evidence that conjugates of vitamin B-6 exist as a substantial proportion of the total vitamin B-6 in plant-derived foods (66-69) . The results of Kabir et. al (65) indicated the existance of /9glucopyranosyl conjugates in a wide variety of plant products. The importance of PN-glucoside as a bioavailable source of vitamin B-6 in humans was first studied by Kabir et al. (74) . The investigators determined the urinary and fecal excretion patterns of glycosylated vitamin B-6 when different foods that contained the naturally occurring PN-glucoside were fed to human subjects. It was observed that an inverse relationship existed between plant derived PN-glucoside (as percent of total B-6) and the net B-6 bioavailability in humans. The investigators suggested that the proportion of the glucosylated vitamin B-6 in the diet may be useful as an index of vitamin B-6 bioavailability. Kabir et al. (65) developed a microbiological assay procedure for guantitating glycosidic conjugated forms of PN

PAGE 26

17 as well as total vitamin B-6 in food. The PN-glucoside, when measured indirectly by hydrolysis, was reported to occur widely in plant foods, ranging from 0% to 82% of the total amount of vitamin B-6 present (74-76) . Studies by Iwami and Yasumoto (76) indicated that intestinal absorption of PN-glucoside was similar to that of free PN. However, Ink et al. (77) administered a single, oral dose of radiolabeled PN and PN-glucoside in an alginate gel diet using a combination of intrinsic and extrinsic labeling in the intact rat. They reported that the absorption of the glucoside was similar to that of unconjugated PN, but the absorbed PN-glucoside was metabolically utilized less effectively as vitamin B-6 than was PN. Iwami et al. (76) measured the relative disappearance from the intestinal perfusate of PN and a purified 4 1 -derivative (rather than the naturally occurring 5 1 -derivative) . In the rat, Ink et al. (77) observed the relative metabolism of the intrinsically and extrinsically labeled glucoside to have less than 40% of the bioavailability of PN. The nutritional properties of these conjugated forms of vitamin B-6 are not fully understood at the present time. Although PN-glucoside comprises 25% to 75% of the total vitamin B-6 in many plant tissues (67,75) its bioavailability in humans continues to be studied. PN-glucoside has been found to undergo intestinal absorption in the intact form. Initial studies employing bioassays with human subjects (40)

PAGE 27

18 indicated that the proportion of the total vitamin B-6 which was present as PN-glucoside correlated closely with the bioavailability of the vitamin in a variety of foods examined, including tuna, peanut butter and whole wheat bread. These results suggested that PN-glucoside would not be available to humans for intestinal absorption and metabolic utilization. However, further studies by Bills et al. (78) indicated that this observed correlation was inconsistent when additional foods were examined. As discussed earlier, Nelson et al. (38) reported that the vitamin B-6 from orange juice, a large portion of which is in the conjugated form, exhibited a slower rate of absorption during intestinal perfusion of humans than PN from control solutions. Tsuji et al. (79) observed that synthetic PN-glucoside exhibited vitamin B-6 activity which was approximately equivalent to that of PN in a rat bioassay. In contrast, Trumbo et al. (80) utilizing a rat bioassay reported that PN-glucoside isolated from alfalfa sprouts exhibited only 10% to 30% bioavailability relative to PN in the rat. Studies in rats using radiolabeled PN-glucoside indicate effective absorption of this B-6 compound and that the portion of PN-glucoside not hydrolyzed to metabolically active PN, is rapidly excreted in the urine in the intact form (77,81). Ink et al. (77) reported the extent of absorption of radiolabeled PN-glucoside from diets containing intrinsically enriched alfalfa sprouts was similar, but quantitatively less

PAGE 28

19 than that of purified labeled PN-glucoside, suggesting an inhibitory effect of plant tissues. Stable-isotopic techniques and methodologies evaluate the bioavailability of purified deuterium-labeled (d2) PNglucoside in human subjects (82) . In a study that used stable isotopic techniques to analyze the bioavailability of orally administered extrinsically enriched oatmeal containing labeled PN-glucoside, it was reported that the labeled PN-glucoside was approximately 58% of that of labeled-PN fed under identical conditions (82) . Although the ability metabolically to utilize PN-glucoside appears to have a substantial level of individual variability among human subjects, the bioavailability of PN-glucoside in human subjects has been observed to be substantially greater than that of the rat (82) . Further development of the methodology and application of stable-isotopic techniques will continue to provide greater clarification of the factors affecting bioavailability of vitamin B-6. The above section is a review of the current knowledge concerning the bioavailability of the vitamin B-6 in foods. There appears to be a general agreement between the rat bioassay and the Saccharomyces uvarum assay method for many foods, although the range of apparent bioavailability relative to PN was quite large for the rat. These comparative data provided the first indication that the bioavailability in certain plant foods may be less than complete. The factors

PAGE 29

20 responsible for the differences in apparent bioavailability of vitamin B-6 are not clear. Human subjects have also been used in researching the bioavailability of vitamin B-6 in specific foods and mixed diets. Bioassays with humans indicated that the overall bioavailability of vitamin B-6 in a typical American diet is reasonably high although incomplete. A study evaluating freeliving vegetarian and non-vegetarian women who consumed quantities of total vitamin B-6 equivalent to those in a mixed diet found no significant differences between groups with respect to vitamin B-6 status when comparing plasma PLP levels (53) . This suggests that the bioavailability of vitamin B-6 in vegetarian diets was similar to that in a mixed diet. Andon et al. (84) reported that a mean of 2.5% of the total vitamin B-6 present in breast milk of omnivorous lactating women was in the form of PN-glucoside. Although a mean of 15% of the total vitamin B-6 in the diet was present as PNglucoside, there appeared to be no correlation between the amount of glycosylated B-6 in the diet and the percent PNglucoside in breast milk over the range examined. In constrast, a study by Reynolds et al. (85) reported that the percentage of dietary vitamin B-6 present as PN-glucoside was equivalent to the amount of PN-glucoside found in the breast milk of lactating Nepalese vegetarian women. Gregory and Ink (75) did not observe any PN-glucoside in the breast milk of six lactating American females including three

PAGE 30

21 lactoovovegetarians. Differences in the vitamin B-6 composition of the breast milk reported in these studies are unclear. The factors responsible for the incomplete bioavailability of vitamin B-6 have not been determined. Factors Affecting the Bioavailability of Vitamin B-6 Several factors are known to affect the bioavailability of vitamin B-6 in food products. These include the formation of certain reaction products during food processing and storage, fiber type and the guantity present in the food source, and the presence of PN-glucoside in plant foods. The results of previously discussed studies suggested that thermal processing and storage of foods may adversely affect bioavailability of vitamin B-6 through the formation of reaction products with amino acid residues of the proteinatious portion of the food (36,56-57). The bioavailability of vitamin B-6 from animal products approaches 100% for most foods. The biological activity of vitamin B-6 from plant derived products is generally lower. A series of studies concerning the nutritional guality of military rations provided the initial impetus to consider the adverse effect of thermal preservation of canned foods. Register et al. (83) reported that rats which subsisted wholely on either of two homogenized combat rations reguired supplemental vitamin B-6 to sustain normal growth even though the diets contained enough vitamin B-6 for adeguate rat

PAGE 31

22 growth. Harding et al. (86) fed human subjects canned rations which had been stored for twenty months at 100 °F. They observed marginal vitamin B-6 deficiency in the subjects even though the rations provided apparently adequate amounts of the vitamin. Inadequate vitamin B-6 nutriture was most likely a result of an effect of thermal processing. It is reasonable to consider that marginal limitations in essential amino acid content of the ration may have contributed to nutrient insufficiency. The addition of vitamin B-6 to diets deficient in certain essential amino acids has been shown to produce a growth response (87) . However, the reason for this apparent incomplete bioavailability of vitamin B-6 in these canned rations is unclear. Several infants who consumed a nonfortified, heat sterilized canned infant formula were found to be severely vitamin B-6 deficient and suffer with neurochemically induced convulsions. Extensive research was subsequently initiated to study the chemical behavior of vitamin B-6 in milk products (58) . The differing stability properties of the various vitamin B-6 compounds have been of interest since Hassinen et al. (59) observed that PM and PL, the naturally occurring forms of vitamin B-6 in milk, were much less stable than added PN. Employing microbiological assays, Hassinen (59) demonstrated that PM and PL added to milk were degraded at the same rate as naturally occurring B-6 vitamers. Added PN exhibited greater stability than either PM or PL during thermal processing of

PAGE 32

23 canned milk and formula products. Through the use of rat bioassays and microbiological analyses, Tomarelli et al. (60) evaluated the bioavailability of vitamin B-6 in milk, spraydried and heat sterilized infant formula products. They concluded that approximately half of the vitamin B-6 present in retorted milk or infant formula was biologically unavailable. Davies et al. (88) used rat and chick growth bioassay methods to estimate the bioavailability of vitamin B6 in raw and canned milk. In contrast to the study of Tomarelli (60), Davies et al. (88) reported the relative loss of vitamin B-6 in heat processing to be equal as determined by Saccharomyces uvarum . rat and chick bioassay methods. This suggests that thermal sterilization in the production of canned evaporated milk does not adversely affect the bioavailability of vitamin B-6. Further evaluation of the data reported by Tomarelli (60) indicated that the bioavailability of the naturally occurring vitamin B-6 in thermally sterilized unfortified milk was only 15% less than that of PN fortified milk samples (37) , which further substantiated the results of Davies et al. (88). Lushbough et al. (89) examined the bioavailability of vitamin B-6 in meats and the effect of cooking using Saccharomyces uvarum and rat growth bioassay methods. The relative losses of vitamin B-6 during the cooking process were nearly identical for every product tested. These data suggest little effect of cooking on the bioavailability of the vitamin

PAGE 33

24 B-6 in meats. Studies conducted by Ink et al. (90) used a combination of intrinsic and extrinsic labeling of liver and muscle tissue with radiolabeled forms of vitamin B-6. These data indicated that thermal processing caused partial destruction of vitamin B-6 in liver and muscle, but had little or no effect on the bioavailability of the remaining vitamin B-6. The bioavailability of vitamin B-6 in foods from plant origin has been proposed to be adversely influenced by dietary fiber (36-37) . The effect of various forms of dietary fiber on vitamin B-6 bioavailability has been examined in both animals and humans. The effect of dietary fiber on the bioavailability of B-6 vitamers was first studied by Leklem et al. (40) who employed a human bioassay to examine the utilization of vitamin B-6 in white verse whole wheat bread. Their data indicated 10% lower availability of the vitamin B-6 which was attributed to reduced intestinal absorption by the fiber components. The results of these studies suggested that the vitamin B-6 of soybeans was approximately 5% to 10% less available than beef vitamin B-6. These data have suggested that the presence of nondigestible components of dietary fiber, may reduce the bioavailability of the vitamin B-6. Physical properties inherent of dietary fiber may influence the absorption of the vitamin through entrapment or alteration of the viscosity of the intestinal contents (16) .

PAGE 34

25 The presence and type of purified dietary fiber have only a minor effect on the bioavailability of the B-6 vitamers (9193). Nuygen et al. (91) evaluated the potential for physical binding of vitamin B-6 using a variety of native and modified polysaccharides and lignin under conditions similar to the human intestine. In vitro binding of the B-6 vitamers by these fiber components did not occur. Differences in viscosity may have influenced the rate of diffusion, although studies by Machida and Nagai (93) indicated that a reduced rate of absorption does not induce a reduction in net vitamin B-6 absorption. Rat and chick bioassays have been employed to evaluate the effects of selected dietary fiber components on the absorption of vitamin B-6 (91,93). The results of these bioassays indicated no inhibitory effects of these dietary fiber components on the bioavailability of vitamin B-6. The results of these studies along with the results of studies with human subjects (39,56) suggest that dietary fiber has little effect on the bioavailability of vitamin B-6 in foods. The slight effect of dietary fiber components on the bioavailability of vitamin B-6 does not fully account for the lower bioavailability of B-6 vitamers from plant food products relative to animal sources. Many vitamin B-6 antimetabolites have been identified and determined to be chemical reaction products formed from vitamin B-6 during the thermal processing or storage of foods. The potential conversion of various B-6 vitamers to these

PAGE 35

26 antagonistic compounds has been proposed to explain certain observed effects of food processing or storage on the bioavailability of vitamin B-6 in certain foods (94-96) . In studying the stability of the B-6 vitamers in food products, PN was found to be the most stable vitamer, followed by PL, PM, PMP, and the least stable B-6 vitamer being PLP (88) . Much research has focused on the potential for PL and PLP to react with such food components as proteins, amino acids and reducing sugars resulting in the formation of degradation products of limited bioavailability. Several studies have reported the characterization of the by-products resulting from the interaction of PL and PLP with amino groups of proteins in food products (96-102) . The chemical reaction of PL or PLP with cysteine or other sulfhydryl amino groups has been proposed as a mechanism responsible for the lowered bioavailability of vitamin B-6 reported in thermally sterilized milk products (96-98) . One such product produced from the heat induced reaction of cysteine and PL has been identified as bis-4-pyridoxyl disulfide (97). Later studies by Srncova et al. (99) supported the reactivity of milk protein sulfhydryl groups with PL when high concentrations of PL were present. The spontaneous reaction of PL or PLP with various thiols and aminothiols results in the formation of thiohemiacetal and thiazolidine complexes (94) . These chemical complexes are readily dissociable and therefore would not impair the

PAGE 36

27 bioavailability of these B-6 vitamers. Gregory et al. (100102) conducted a series of experiments to examine further the possible influence of thermal processing and storage on the net bioavailability of vitamin B-6 in foods. When milk containing radiolabeled PL or PLP was subjected to heat sterilization, analysis using HPLC revealed no formation of bis-4-pyridoxal disulfide (103). The losses of the B-6 vitamers PL and PLP were due to the reductive binding of the aldehydes of these vitamers to food protein as epyridoxyllysyl residues (103) . The formation of epyridoxyllysine has been identified as a mechanism of the loss of vitamin B-6 during thermal processing or low moisture storage of proteinaceous model food systems (100-101) and various meat and dairy food products (16) . Similar results were reported in intrinsically enriched chicken liver and muscle tissues (103) . The phosphopyridoxyllysyl complex accelerated the onset and enhanced the severity of vitamin B-6 deficiency symptoms. The effect of e-pyridoxyllysine on the bioavailability of vitamin B-6 in foods is a function of the ratio of total vitamin B-6 content of the diet and the addition product to the other B-6 vitamers. The bioavailability of the vitamin of diets which contain epyridoxyllysine but which are adeguate in vitamin B-6 content would most likely be high (102). Whether naturally occurring amounts of vitamin B-6 are low or conditions present during food processing produce e-pyridoxyllysine, the antimetabolite

PAGE 37

28 effects of e-pyridoxyl lysine generally decreases the apparent bioavailability of vitamin B-6 by competitive inhibition of PL kinase (103) . It is reasonable to surmise from the above studies that a similar antagonistic effect of epyridoxyllysine may have been responsible for the severe deficiencies observed in infants with compromised vitamin B-6 nutriture and fed diets composed entirely of non fortified canned infant formulas (60) . The mechanisms of action for an antimetabolite, such as e-pyridoxyllysine, includes inhibition of any of the enzymes in the metabolic pathway to interconvert the B-6 vitamers. Other mechanisms of vitamin B-6 antimetabolites are structural analogues of the vitamin and inhibition of the active coenzyme function of PLP (96) . The complex chemical identity and nutritional properties of other B-6 vitamer derivatives found in foods have not been fully determined. The formation of the degradation product, 6-hydroxy-pyridine from the relatively stable B-6 vitamer PN, has been reported in thermal treatment of various fruits and vegetables (104) . The generation of hydroxyl radicals apparently mediated the hydroxylation of PN which was associated with the oxidative degradation of ascorbic acid (105) . Gregory and Leatham (106) reported that 6-hydroxy-PN has no vitamin B-6 or antivitamin B-6 activity. It appears that hydroxylation of PN at the 6 position is not a significant mechanism responsible for the loss of vitamin B-6 activity in foods.

PAGE 38

29 A number of derivatives of PN-glucoside have been isolated and identified. Tadera et al. (73,107) reported several additional PN-glycosides to be minor components of the total vitamin B-6 in pea seedlings and rice bran (108) . Assuming that PN-glucoside esters occur naturally and in great enough quantities, it is unclear whether these glucosyl compounds contribute significantly to the total vitamin B-6 in foods and, if so how they may impact vitamin B-6 nutriture when consumed. However, it is worth mentioning that PNglucoside has been the only glycosylated form of vitamin B-6 detected in a variety of foods utilizing HPLC methodology (71,77). Tadera et al. (108) have observed the presence of a vitamin B-6 conjugate termed "B-6X". The authors described this compound as PN-glucoside esterified to an organic acid based on its response in microbiological assays following sequential alkaline treatment and hydrolysis with betaglucosidase (108) . The nutritional properties of these conjugated forms of vitamin B-6 are not fully understood at the present time. The nutritional significance of these glycosylated vitamin B-6 compounds is related to their potentially limited bioavialability . As earlier stated, PN-glucoside comprises 5% to 80% of the total vitamin B-6 in many plant tissues (67,75) . Its bioavailability for humans is approximately 58% relative to that of PN when administered orally (82) . PN-glucoside has been found to undergo intestinal absorption via hydrolysis of

PAGE 39

30 the beta-glycosidic bond of PN-glucoside by mucosal betaglucosidase releasing free PN. Initial studies employing bioassays with human subjects (40) indicated that the proportion of the total vitamin B-6 which was present as PNglucoside correlated inversely with the bioavailability of the vitamin in the foods examined (tuna, peanut butter, and whole wheat bread) . These results suggested that PN-glucoside would not be biologically available to humans. However, studies by Bills et al. (78) indicated that this observed correlation was inconsistent when additional foods were examined and that the quantity of PN-glucoside present in a food product could function as a predictor of the bioavailability of vitamin B-6 from that food source. Stable-isotopic techniques have been used to evaluate the bioavailability of purified deuteriumlabeled (d 2 ) PN-glucoside in human subjects (82) . In these studies, the bioavailability of orally administered extrinsically enriched oatmeal containing labeled PN-glucoside was similar to that of labeled-PN fed under identical conditions (82) . Although metabolic utilization of PNglucoside appears to have a substantial level of individual variability among human subjects, the bioavailability of PNglucoside in human subjects has been observed to be substantially greater than that of the rat (82). Further application of stable-isotopic techniques provide greater clarification of the factors affecting bioavailability of vitamin B-6.

PAGE 40

31 The bioavailability of vitamin B-6 compounds is a function of their extent of absorption and metabolic utilization, as reflected by conversion to active coenzymatic forms. It is important to assess the bioavailability of vitamin B-6 as consumed in a typical mixed diet. The assessment of vitamin B-6 status in populations is contingent on identifying and guantifying the different forms of the vitamin occurring naturally in foods consumed, determining the potential interactions between these forms and the relative bioavailability of the different vitamers, and their possible interactions. Our present knowledge of the essential role of vitamin B-6 in preserving health emphasizes the importance of establishing an accurate intake for vitamin B-6. The adeguacy of intake is difficult to assess critically without accurate information regarding the amount of the vitamin in foods consumed, the effect of various vitamin B-6 compounds on the enzymes involved in the metabolism of various forms of vitamin B-6, as well as information regarding vitamin B-6 bioavailability. Enzymatic Interconversion of B-6 Vitamers There are several enzymatic reactions responsible for the interconversion of the B-6 vitamers. The liver serves as the center of vitamin B-6 metabolism and contains the enzymes necessary for the metabolic interconversion of the B-6

PAGE 41

32 compounds. Following intestinal absorption, the dephosphorylated B-6 vitamers diffuse into the liver and are converted to the coenzyme form of the vitamin, PLP (109-110) . A portion of the PLP is released into the circulation and constitutes the major source of plasma PLP (111) . The dephosphorylated B-6 vitamers, PL, PN, and PM are phosphorylated by the same kinase enzyme, pyridoxal kinase (EC 2.7.1.35) (PL kinase). The 5 ' -phosphates of PN and PM are subsequently oxidized by pyridoxamine (pyridoxine) 5'phosphate oxidase (EC 1.4.3.5) (PMP (PNP) oxidase) to yield PLP. This coenzymatic form is bound to apoproteins, transported to blood or dephosphorylated by phosphatases back to PL (112). Further metabolism of PL results in rephosphorylation by PL kinase or oxidation, by aldehyde oxidase, to 4 PA, the catabolic end product of vitamin B-6 metabolism. Vitamin B-6 in its coenzymatic form PLP or as PLP-dependent enzymes, functions in many metabolic processes. These include, amino acid metabolism (3) , neurochemical function (113-115) , and modulation of hormone action (4-5) . Therefore, it is important that the cellular levels of PLP be properly regulated. As stated above, PL kinase is responsible for the postabsorptive phosphorylation of PL, PN, and PM with relatively equal efficiency. The result is the formation of PLP from PL and upon further oxidation of PN and PM by PMP (PNP) oxidase these two vitamers are also converted to the

PAGE 42

33 coenzymatic form (109-110) . The PLP that is not released into circulation is either bound to cellular proteins or dephosphorylated to PL. The vitamin B-6 taken up by the liver is predominately maintained in the phosphorylated form (109110) . There exists a cycle of phosphorylation/dephosphorylation between PL and PLP which is catalyzed by PL kinase and phosphatase enzymes (109) . Under physiological conditions PL kinase has approximately a 10 fold greater activity than phosphatases (112) . The degradative enzyme aldehyde oxidase has an activity greater than the kinase and effectively competes for PL. In order to maintain a steady state of intracellular PLP a form of metabolic trapping occurs. The coenzyme is bound to protein which protects it from hydrolysis and hence, determines the relative amounts of PLP and PL and creates a shift of equillibrum toward a balanced state (112). Considering that PLP is required as a conenzyme for a number of enzymes, it is not surprising that PL kinase is widely distributed in tissues. This enzyme has been detected in practically all tissues tested, including liver, kidney, brain, and muscle (30) . PL kinase has also been detected in erythrocytes, however, the enzyme differs in its properties compared with other tissue PL kinase (116) . Since the liver is the major site of vitamin B-6 metabolism it is generally the tissue selected for the study of the enzymes in the B-6 vitamer interconversion pathway (116) . It is important to

PAGE 43

34 note that in comparative assessment of either of the enzymes PL kinase or PMP (PNP) oxidase from different sources, the levels and forms (apo or holo) of the enzyme is markedly influenced by factors such as species, age, sex and dietary status of the animal (110) . Therefore, the following discussion regarding the properties of these two enzymes will be limited to rat liver as the source for the enzymes, unless otherwise noted. Hepatic PL kinase has a molecular weight of approximately 60,000 and a pH optimum between 5.5 and 6.0 (117). PL kinase from all sources studied requires a divalent cation for activity, with the greatest activity in liver PL kinase seen with Zn 2+ (30) . The enzyme also requires monovalent cations with activation greatest in the presence of K* (117) . Although similar to PL kinase in tissue distribution and molecular weight, PMP (PNP) oxidase differs significantly in other physical and kinetic properties. The molecular weight of oxidase is approximately 54,000, each having two identical subunits of 27,000. PMP (PNP) oxidase occurs widely distributed in both tissues and cells (118-119) . Pogell (32) was the first to describe PMP (PNP) oxidase as having a requirement for a flavin cofactor in the oxygen-dependent conversion of PMP to PLP. Later Wada and Snell (31) observed that the same oxidase was responsible for the conversion of both PNP and PMP to PLP. The investigators reported that the partially purified enzyme required FMN as the specific

PAGE 44

35 f lavocoenzyme and demonstrated product inhibition as well as inhibition by some phosphorylated analogs of vitamin B-6 compounds (31) . The relative reactivity of PMP(PNP) oxidase with varying substrates is pH dependent (118) . Within the physiological pH range both PMP and PNP are suitable substrates for liver PMP (PNP) oxidase. Although at slightly alkaline pH the V for PNP is higher, the observed velocity is decreased due to greater substrate inhibition by PNP even at low substrate concentrations (112) . The values demonstrate that at pH 7.0 PMP and PNP bind with similar affinity, whereas PMP binds more tightly at the usual assay pH of 8.0 compared to PNP (118) . Many compounds have been indicated as inhibitors of the metabolism of vitamin B-6. When assessing the site of action of these antagonists it is reasonable to suspect that such compounds may act on those enzymes responsible for the formation of the active coenzyme form PLP. It is well documented that liver PL kinase is inhibited by analogues containing a 4-formyl group which resemble PL in their affinity for PL kinase, condensation products formed from PL and hydroxy 1 amine, O-substituted hydroxylamines, hydrazine and substituted hydrazines (120) . PL kinase has also been shown to be sensitive to inhibition by degradation products such as pyridoxyl lysine (121) .

PAGE 45

36 It is well known that the product PLP, inhibits the activity of PMP(PNP) oxidase (112) . As a result, other phospho-B-6 analogues have been evaluated (118) . It was discovered that a dianionic charge on the 5* -position is necessary for binding of the substrate analogue and subsequently inhibition of the enzyme (122) . Inhibition of PMP (PNP) oxidase also occurs when alterations are introduced in the structure of the flavin coenzyme, FMN (118) . In as much as both PL kinase and PMP (PNP) oxidase have substrate reactivity with structural analogues, it is not unlikely that conjugated forms of vitamin B-6 would inhibit these enzymes. Conflicting data have been reported concerning the extent of bioavailability of PN-glucoside in vitamin B-6 metabolism. Most experimental results in rats have indicated 20-30% net bioavailability of PN-glucoside relative to PN. The bioavailability of PN-glucoside, although incomplete, is substantially greater in humans than in rats. While the incomplete utilization of PN-glucoside has been clearly established in rats and humans, its potential interactive effects in vitamin B-6 metabolism have not been examined previously. The purpose of this research was to demonstrate the effects PN-glucoside has on the metabolic utilization of vitamin B-6. The effort to evaluate these effects is an important one, when considering the overall nutritional status of individuals with respect to vitamin B-6. Being aware of

PAGE 46

37 the multitude of roles vitamin B-6 plays in the body and its impact on the overall health of individuals, it is imperative that information be collected concerning how naturally occurring forms of vitamin B-6 potentially interact with one another.

PAGE 47

CHAPTER 3 PYRIDOXINE-5'-0-D-GLUCOSIDE AFFECTS THE METABOLIC UTILIZATION OF PYRIDOXINE IN RATS Introduction A conjugated form of pyridoxine (PN) was first isolated from rice bran and identified as 5'-0-(/3-Dglucopyranosyl) pyridoxine (PN-glucoside) (67) . Analysis of a variety of plant-derived foods by HPLC has shown that PNglucoside comprised 5-80 % of the total vitamin B-6 in many fruits and vegetables (75) . Similar results were obtained by the use of a microbiological assay procedure for the glycosidic conjugate of pyridoxine (65) . Conflicting data have been reported concerning the bioavailability of PN-glucoside in vitamin B-6 metabolism. Studies in rats have indicated that purified PN-glucoside is relatively well absorbed but undergoes little metabolic utilization and is rapidly excreted in intact form (77,80). Most experimental results in rats have indicated 20-30 % net bioavailability of PN-glucoside relative to PN (77,80-81) . In constrast, Tsuji et al. (79) observed nearly 100 % bioavailability of PN-glucoside relative to PN in the rat. The bioavailability of PN-glucoside, although incomplete, is substantially greater in humans than in rats (82) . The mean 38

PAGE 48

39 bioavailability of orally administered PN-glucoside is approximately 58% relative to PN. While the incomplete utilization of PN-glucoside has been clearly established in rats and humans, its potential interactive effects in vitamin B-6 metabolism have not been examined previously. Materials and Methods Protocol Two studies were conducted to evaluate the in vivo utilization of [ U C]PN in the presence of PN-glucoside. In each of the studies eighteen male Sprague-Dawley rats (Crl:CD (SD) BR) from Charles River Breeding Laboratories (Wilmington, MA) , weighing 200-300 g, were housed individually in stainless steel metabolism cages in animal quarters maintained at 24±1"C with a 12-h light/ dark cycle. All procedures for the care and treatment of the experimental animals were in accordance with the National Institutes of Health Guidelines. The rats were fed ad libitum a casein-sucrose (20%: 60%) based diet, adequate in all micronutrients including 7 mg PN HCl/kg«d for eight days (80) . At the end of a seven day acclimation period the rats were randomly assigned to one of three treatment groups. On day eight of studies 1 and 2, following a twelve hour fast, the rats were administered a dose of 166.5 MBq (4.5 juCi) [ U C]PN, equivalent to 240 nmol of PN. In study 1 the rats received a simultaneous dose of either 0, 36, or 72 nmol of unlabeled PN-glucoside (treatment

PAGE 49

40 groups 1,2, and 3 respectively). In study 2 the rats received a simultaneous dose of either 0 or 72 nmol unlabeled PNglucoside or 21.6 nmol unlabeled PN (treatment groups 1,2, and 3 respectively). The 21.6 nmol dose of PN was selected to provide 30% of the PN-glucoside treatment, based on an assumed 30% net bioavailability of PN-glucoside relative to PN (4-6) . Doses were administered in 1.5 mL sterile H 2 0 by gavage. Urine was collected for the ensuing 24 h into foil-covered flasks to avoid photochemical degradation of vitamin B-6 compounds. The urine collection funnels and flasks were rinsed with water and rinses pooled with collected urine; after dilution to 50 mL, urine was stored at -20° C until analysis. The rats were killed by decapitation following brief anesthesia, and livers were rapidly excised. The tissue was divided into 2 g samples, then stored at -20° C until analysis. All procedures were performed under GEF40G0 gold lights that emit a wavelength between 500 nm and 750 ran to prevent photochemical degradation of B-6 vitamers. Forms of vitamin B-6 PN-glucoside was prepared by biological synthesis using the propagation of alfalfa sprouts germinated in the presence of PN-HC1 obtained commercially (Sigma Chemical Company, St. Louis, MO) (80). The [4 , 514 C]pyridoxine hydrochloride 684.5 KBq/nmol (18.5 Ci/mol) was a gift from Hof fmann-LaRoche

PAGE 50

41 (Nutley, NJ) with a purity of greater than 98%, as determined by ion-pair reverse phase HPLC. Sample preparation Urine samples were deproteinated by centrifugal ultrafiltration with micropartition tubes (YMT membrane filters; Amicon, Danvers, MA) (80) . Urinary 4 PA analysis was by reverse-phase HPLC, as described below. Fractions (0.5 mL) were collected by using an ISCO Cygnet Fraction Collector (ISCO, Lincoln, NB) . Each filtered sample was decolorized (81) and an aliquot (100 /xL) was then counted for total radioactivity. Liver tissue was minced and homogenized in 6 mL of 4.3 mol/L trichloroacetic acid with a Polytron Homogenizer (Brinkman Instruments, Westburg, NY) , and then centrifuged for 20 min. at 12,000 x g (6). The supernatant was partitioned against an equal volume of diethyl ether to extract the trichloroacetic acid and filtered with 0.45 membrane filters (Gelman Sciences, Inc., Ann Arbor, MI). The sample was aerated with nitrogen gas, then analyzed for vitamin B-6 by HPLC as described below, and an aliquot of the supernatant from each sample was analyzed for radioactivity. Decolorization of the extract was performed as above prior to liquid scintillation spectrometry.

PAGE 51

42 HPLC equipment and analysis The separation of radiolabeled B-6 compounds in the urine and liver tissue was accomplished by ion-pair reverse-phase HPLC (82) . Chromatographic analysis was performed with a Rainin HP/HPX Drive Module (Rainin Instrument, Woburn, MA) . All of the above analysis utilized a loop injection valve (Model 904-2, Altex) , a fluorometric detector (Model LS-5, Perkin-Elmer, Norwalk, CT) and an electronic integrator (Model 3388A, Hewlett-Packard, Avondale, PA) . Excitation and emission wavelengths were 295nm (5nm slit width) and 405nm (5nm slit width) respectively. Two mobile phases were employed in a gradient elution procedure using an Ultrasphere IP 5 urn C-18 4.6 mm x 25 cm column (Beckman Instruments, San Ramon, CA) . Mobile phase A contained 0.033 mol/L phosphoric acid and 8 mmol/L 1-octanesulfonic acid, adjusted to pH 2.2 with 6 mol/L KOH. Mobile phase B contained 0.033 mol/L phosphoric acid and 3.4 mol/L acetonitrile, adjusted to pH 2.2 with 6 mol/L KOH and no ion-pairing reagent. Fractions (0.5 mL) were collected using an ISCO Cygnet Fraction Collector (ISCO, Lincoln, NB) . The identity of the PLP and PNP peaks was confirmed by monitoring the formation of pyridoxal (PL) and PN, respectively, by HPLC following incubation of the PLP and PNP fractions in 0.1 mol/L HC1 at 80° C for 3 h.

PAGE 52

43 Measurement of radioactivity HPLC fractions and extracts of tissue were measured for radioactivity using a commercial scintillation fluid (ScintiVerse LC, Fisher Scientific, Orlando, FL) and a liquid scintillation spectrophotometer (Beckman LS 2800 Beckman Instruments, San Ramon, CA) . A quench curve for U C was used for conversion of cpm to dpm. Statistical analysis In experiment 1 and 2 the distribution of vitamin B-6 compounds and total radioactivity between groups was compared by the method of least squares analysis of variance using general linear model procedures of SAS (123) following a log tranformation of the data to normalize variance (123) . Data are reported as least squares mean ± the pooled standard error of the least squares mean (SEM) . In experiment 1, orthogonal contrasts were made to examine the linear and quadratic effects of different doses of PN-glucoside on dependent variables. The quadratic effects were not significant for any dependent variable. Values with p<0.05 were considered statistically significant and were all for linear effects of PN-glucoside unless otherwise stated. In expreriment 2 two orthogonal contrasts were used to compare treatment means. Contrast one compared group 2 with group 1 and 3; contrast two compared group 1 with group 3 .

PAGE 53

44 Table 3.1 Liver B-6 vitamer distribution and total liver 14 C in rats administered varying levels of PN-glucoside in the dose (Experiment 1) .* LEVEL OF PN-GLUCOSIDE IN THE DOSE Vitamers 0 nmol 36 nmol 72 nmol SEM p<0.05 % liver radioactivi tY PLP 17.5 37.3 51.6 7.5 # PNP 0.0 3.1 16.4 3.4 # PMP 46.7 36.1 32.0 8.0 # PL 16.9 14.5 0.0 4.8 + PN 15.2 7.8 0.0 3.5 + PM 3.6 1.2 0.0 0.6 + Total 14 C % of dose 20.0 17.0 12.3 4.0 + Values based on n=6 are means and pooled standard error of the mean (SEM) . Log transformations were performed on data prior to analysis of variance. The quadratic effects were not significant for any dependant variable. * Significant linear effect, p<0.05. * Significant decreased linear effect, p<0.05. Results These experiments were designed to investigate the effect of unlabeled PN-glucoside on the metabolism, in vivo retention, and excretion processes of simultaneously administered [ U C]PN. The distribution of B-6 vitamers in liver tissue, total urinary [ 14 C]4PA, and total urinary 14 C were determined to evaluate the utilization of [ 14 C]PN in the presence of PN-glucoside. Data were expressed as a percentage of total radioactivity administered in the dose or as a percentage of total [ 14 C] B-6 in the tissue.

PAGE 54

45 Table 3.2 Urinary 4 PA and total urinary 14 C in rats administered varying levels of PN-glucoside (Experiment 1)* 0 nmol 36 nmol 72 nmol PN-glucoside PN-glucoside PN-glucoside SEM p<0.05 % urinary radioactivity 4 PA 37.6 12.7 10.9 5.1 + % of dose U C 26.9 35.1 37.3 12.3 # * Values based on n=6 are means and pooled standard error of the mean (SEM) . Log transformation of the data was performed prior to the analysis of variance. The quadratic effects were not significant for any dependant variable. * Significant linear effect, p<0.05. * Significant decreased linear effect, p<0.05. The results of study 1 indicated statistically significant differences (p<0.05) in the quantity and distribution of labeled vitamin B-6 compounds in liver tissue and urine among groups fed varying levels of PN-glucoside (Tables 3.1 and 3.2). A significant linear relationship was observed between phosphorylated vitamin B-6 compounds pyridoxal 5' -phosphate (PLP) and pyridoxine 5' -phosphate (PNP) in liver tissue and the amount of administered PN-glucoside. Total hepatic 14 C decreased linearly (p<0.05) with varying amounts of PN-glucoside (Table 3.1). The proportion of hepatic nonphosphorylated 14 C-labeled PN, PL and pyridoxamine (PM) decreased linearly with the amount of PN-glucoside (Table 3.1) . Urinary excretion of total U C was showed a significant linear effect (p<0.05) among groups receiving varying doses of

PAGE 55

46 Table 3.3 Liver B-6 vitamer distribution and total liver 14 C in rats administered varying levels of PN or PN-glucoside. (Experiment 2)* LEVELS OF PN-GLUCOSIDE LEVELS OF PN Vitamers 0 nmol Group 1 72 nmol Group 2 21.6 nmol Group 3 SEM * liver radioactivity PLP 31.5 58. 4 a 31.5 3.4 PNP 0.0 14. 4 a 0.0 0.1 PMP 41.4 27. 2 a 40.1 1.0 PL 13.3 0.0 a 14.3 0.4 PN 6.3 0.0 a 6.0 0.2 PM 7.5 0.0 a 8.1 0.4 Total U C % of dose 15.1 10. 0 b 15.0 0.0 * Values based on n=6 are means and pooled standard error of the means (SEM) . Log transformation of the data was performed prior to analysis of variance. The quadratic effects were not significant for any dependant variable. 8 Significantly different at p<0.05 for linear effects of PN-glucoside in the dose. b Significantly different at p<0.05 for decreased linear effect. PN-glucoside (Table 3.2). In addition, the concentration of urinary [ 14 C]4PA, relative to total urinary 14 C, and total urinary [ U C]4PA decreased linearly with the amount of administered PN-glucoside (Table 3.2). Experiment 2 was conducted to determine if the results of experiment 1 were due to a specific effect of PN-glucoside or simply to an increase in the total amount of available PN (derived from PN-glucoside) . The experimental design was the same, except on day eight, following a twelve hour fast, the rats were administered a simultaneous dose of 166.5 MBq (4.5

PAGE 56

47 /iCi) [ U C]PN, while treatment groups 1 and 2 received 0 nmol and 72 nmol unlabeled PN-glucoside, respectively. Treatment group 3 received 21.6 nmol unlabeled PN based on an assumed 30% net bioavailability of PN-glucoside relative to PN (77,8081) . Statistical analysis by orthogonal contrasts of the data from this study indicated no statistically significant linear effects between the quantity or distribution of labeled vitamin B-6 compounds in liver tissue and urine and treatment groups one and three (Table 3.3). However, there was a significant linear effect (p<0.05) in the relative concentration of phosphorylated vitamin B-6 compounds PLP, Table 3.4 Urinary 4 PA and total urinary 14 C in rats administered varying levels of PN or PN-glucoside. (Experiment 2)* LEVELS OF PN-GLUCOSIDE LEVELS OF PN 0 nmol 72 nmol 21.6 nmol SEM % urinary radioactivity 4 PA 35.5 14. 2 b 32.7 12.3 % of dose 14 C 30.0 46. 0 a 27.0 4.0 * Values based on n=6 are means and pooled standard error of the mean (SEM) . Log transformation of the data was performed prior to analysis of variance. The quadratic effects were not significant for any dependant variable. a Significantly different at p<0.05 for linear effects of PN-glucoside in the dose. b Significantly different at p<0.05 for decreased linear effects.

PAGE 57

48 PNP, and pyridoxamine 5" -phosphate (PMP) in liver tissue, while hepatic U C decreased linearly (p<0.05) (Table 3.3) . The proportion of hepatic nonphosphorylated forms of [ 14 C]PN, PL and PM was significantly less among treatment group 2 (p<0.05) than groups 1 and 3. Urinary excretion of total U C also was significantly greater (p<0.05) in group two relative to groups one and three (Table 3.4). In addition, the concentration of urinary 4 PA, relative to total urinary 14 C, was significantly lower (p<0.05) in group two as compared to groups 1 and 3 (Table 3.4). The results of experiment 2 indicate that the effects of PN-glucoside observed in experiment 1 were specific for this vitamin B-6 derivative and were not due to isotope dilution effects of the unlabeled compound administered. Discussion The focus of this research was to determine the direct effects of PN-glucoside on [ 14 C]PN administered in a simultaneous dose, similar to what would occur when consuming a mixed diet. The ratios of PN-glucoside to nonglycosylated vitamin B-6 ( 14 C-labeled plus dietary) were in the range of 8% to 16%, which is consistent with observations of typical human diets (82) . Since our interest was in determining the fate of labeled-vitamin B-6 compounds in the presence of PN-glucoside the data were expressed as percent of the total radioactivity in the dose. This enabled the determination of the direct

PAGE 58

49 effect of PN-glucoside on the metabolic utilization of [ U C]PN and distribution of the u C-labeled B-6 metabolites in the liver 24 h post-dose when the two vitamers are consumed together. Expression of the data in this manner reflects the direct effect of PN-glucoside on the distribution of B-6 vitamers and urinary excretion of 4 PA as well as interactions between different naturally occurring forms of vitamin B-6 found in foods consumed simultaneously in a typical mixed diet. The bioavailability of vitamin B-6 compounds is a function of their extent of absorption and metabolic conversion to active coenzymatic forms. Pyridoxine that is absorbed from the intestinal tract is concentrated initially in the liver (124) and is the sole source of plasma PLP (125) . Thus, liver plays a central role in the overall metabolism of vitamin B-6 (126) . The major transformations in hepatic vitamin B-6 metabolism involve phosphorylation catalyzed by pyridoxal kinase, oxidation of PNP and PMP by pyridoxine (pyridoxamine) 5* -phosphate oxidase (PNP oxidase) , along with interconversion of PLP and PMP through transamination reactions (109-110,127129) . The principal forms of vitamin B-6 in liver are PLP and PMP (129-131). PNP is usually present in only trace quantities because of its rapid oxidation to PLP (130,132133) . The non-phosphorylated B-6 vitamers constitute less than 10% of the total vitamin B-6 content in the liver (131) . In this study, the concentration of [ U C]PNP increased linearly

PAGE 59

50 with the quantity of unlabeled PN-glucoside administered (p<0.01). This suggests a mechanism involving a possible inhibition of PNP oxidase. Investigations of possible direct inhibitory effects of PN-glucoside on PNP oxidase are in progress. Comparatively high levels (36%) of the nonphosphorylated B-6 vitamers were observed in the livers of the rats in group 1, which may have been due to several factors. The vitamin B-6 requirement for the rat is 6-7 mg PN/kg diet (135) . The experimental diet contained 7 mg PN/kg diet with the rats consuming, on the average, 70 /xg vitamin B-6 per day. The dose of labeled-PN contained 240 nmol (49.3 ng) [ U C]PN. Hence, during the 24 h post dose period, rats received 49.3 jug vitamin B-6 above their requirement. In addition, a single observation of tracer distribution at a time point 24 h postdose, would not necessary reflect steady state concentrations of endogenous vitamin B-6 in the liver. These factors would account for greater than normal levels of B-6 metabolites in liver. The compensating changes observed in the proportions of hepatic [ U C]PMP and [ U C]PLP (Table 3.1) cannot be readily explained. Unexpectedly, these changes are consistent with those observed when comparing the metabolism of [ U C]PN in vitamin B-6 adequate verses deficient rats (81) . These results indicate that the administration of unlabeled PNglucoside in this study caused changes in metabolic patterns

PAGE 60

51 which paralleled, in one respect, those seen in vitamin B-6 deficiency. The metabolic pathway for the degradation of PLP involves enzymatic hydrolysis of the phosphate ester bond, and the oxidation of PL to 4 PA (132-133) . As a terminal product of vitamin B-6 metabolism, urinary 4 PA reflects the in vivo metabolic utilization of the vitamin. In these studies, the extent of conversion of administered [ U C]PN to [ U C]4PA is indicative of reduced utilization of [ U C]PN which is inversly related to the dose of PN-glucoside. This study also shows a negative correlation between percent radioactivity in the urine as [ 14 C]4PA and the guantity of PN-glucoside administered. The data demonstrate that total liver U C was inversly related to the PN-glucoside dose while total urinary U C was directly proportional to the amount of PN-glucoside administered. The intracellular concentration of PLP in liver tissue is tightly regulated which does not permit the excess accumulation of this coenzyme (133) . In the liver, newly synthesized PLP and PMP are contained in compartments that have a rapid rate of turnover (132). These small and rapidly mobilized pools are poorly miscible with the endogenous coenzyme pools (132). The newly synthesized [ U C]PLP and [ U C]PMP in the liver would be rapidly metabolized by hepatocytes and degraded to [ 14 C]4PA and excreted in the urine, or transported to the circulation as [ U C]PL, and [ U C]PLP.

PAGE 61

52 These B-6 vitamers could then be taken-up by erythrocytes and/or tissues. This would account for the observed results of liver U C and urinary 14 C in response to PN-glucoside dose seen in this study. When rats were given a quantity of unlabeled PN equivalent of 30% of the PN-glucoside dose, the effect on the metabolic utilization and distribution of other forms of vitamin B-6 differed significantly from the group administered the PN-glucoside dose. In previous studies in this laboratory, the absorption of PN-glucoside was 50% relative to PN in rats fed a diet adequate in vitamin B-6 when administered as an alginate gel (77) but was nearly equivalent to PN when given in solution (81) . Analysis of urine showed that most of the absorbed PNglucoside was excreted in the intact form, suggesting low bioavailability (4,6). Previous findings by Trumbo et al. (80) showed poor utilization of PN-glucoside relative to PN on the basis of growth and plasma PLP concentration, which indicated that PN-glucoside has a low biological availability as vitamin B-6 in the rat. However, PN derived from hydrolysis of PN-glucoside, can enter into vitamin B-6 metabolic pathways of the liver to produce other forms of vitamin B-6 (77,81). Kabir et al. (65) observed with human subjects, an inverse relationship between the percentage PN-glucoside (of total vitamin B-6) and overall vitamiln B-6 bioavailability in food. The investigators suggested that the proportion of the

PAGE 62

53 glycosylated vitamin B-6 in the diet may be useful as an index of vitamin B-6 bioavailability. The nutritional significance of the effect of PN-glucoside on the metabolic utilization of other forms of vitamin B-6 is based on the fact that PNglucoside is a major naturally occurring form of vitamin B-6 in many fruits and vegetables of human diets (65,75). The results of the present studies suggest that PN-glucoside is nutritionally significant in two ways. As shown previously, PN-glucoside can act as a source of partially available vitamin B-6 (77,81-82). In addition, this study has shown that PN-glucoside can act as a weak antagonist that may hinder the utilization of PN and possibly other forms of the vitamin. These results indicate that PN-glucoside alters the metabolism and in vivo retention of [ 14 C]PN in the rat and that PN-glucoside may retard the utilization of nonglycosylated forms of vitamin B-6. PN-glucoside represents a significant proportion of the total vitamin B-6 in many plant derived foods (75) . Therefore, the results of this study are important in assessing the bioavailability of vitamin B-6 in a typical mixed diet. The assessment of vitamin B-6 status in human populations is contingent on identifying and quantifying the content and distribution of the different forms of the vitamin found in foods, determining the potential interactions between these forms and the relative bioavailability of the different vitamers. If it is found that PN-glucoside affects the metabolic utilization of PN similarly in humans, it is

PAGE 63

54 conceivable that the vitamin B-6 status in humans would not be accurately reflected by current food consumption data. Studies evaluating the metabolic interactions of PN-glucoside on PN in humans are currently in progess.

PAGE 64

CHAPTER 4 EFFECTS OF PYRIDOXINE-5 1 -/3-D-GLUCOSIDE ON THE METABOLIC UTILIZATION OF PYRIDOXINE IN HUMANS Introduction A conjugated form of vitamin B-6, first isolated from rice bran was identified as 5'-0-()9-Dglucopyranosyl) pyridoxine (PN-glucoside) (67). Analysis of a variety of plant-derived foods indicated that PN-glucoside is abundant in many fruits and vegetables and the only significant glycosylated form of vitamin B-6 present in the foods examined (65,71,75). Assessing the nutritional status of a person with respect to vitamin B-6 is a function of the amount and form of vitamin B-6 in food, the bioavailability, which is the extent of intestinal absorption and metabolic utilization of the B-6 compounds in the diet and the specific reguirement of the individual. Previous studies concerning the bioavailability of vitamin B-6 in rats indicated that purified PN-glucoside is relatively well absorbed compared to PN, undergoes little metabolic utilization, and is rapidly excreted (2,77,80-81). These results indicated 20%-30% net bioavailability of PNglucoside relative to PN. The extent of in vivo hydrolysis of the glycosidic bond, rather than intestinal absorption, was 55

PAGE 65

56 the limiting step in the utilization of PN-glucoside as vitamin B-6 in the rat (2,77,80). A recent study in our laboratory has shown that PN-glucoside alters the metabolism and in vivo retention of simultaneously administered PN in the rat (136) . A stable-isotopic study in humans found the bioavailability of PN-glucoside to be 58% of that of free PN, which is greater than that found in the rat (82) . In an investigation of lactating women, Andon et al. (84) reported that dietary PN-glucoside contributed little to vitamin B-6 nutriture. These findings indicate the need for additional data concerning factors affecting the utilization of PNglucoside and the potential effect of PN-glucoside on the metabolic utilization of vitamin B-6 by humans consuming a mixed diet. The study reported here was conducted to determine the effect of PN-glucoside on the metabolic utilization of PN in humans through the use of stable-isotopic methods . Materials and Methods Synthesis of forms of vitamin B-6 The deuterium-labeled form of vitamin B-6 used in this study was [5 1 -C 2 H 2 OH]pyridoxine (d 2 PN) which was prepared in our laboratory as described by Coburn et al. (108). Mass spectral analysis indicated that the d 2 PN species comprised 66% of the d 2 PN preparation, with a majority of the remainder

PAGE 66

57 in the monodeutero (d^ form. All procedures were performed under yellow-light to prevent photochemical degradation of B-6 vitamers. PN-glucoside was prepared from PN-HC1 using biological synthesis as previously described (80) . Twelve grams of alfalfa seeds were germinated in the presence of 62 mg of PNHC1 (297 /xmol) obtained commercially (Sigma Chemical Company, St. Louis, MO) followed by purification by cation-exchange (80) and gel filtration chromatography on Sephadex G-10 (Pharmacia) . The yield of purified PN-glucoside was 237 nmol (80%) . Chromatographic analysis of the purified PN-glucoside preparation confirmed the absence of free (nonglycosylated) PN and other forms of vitamin B-6. Protocols of trials with human subjects The study was conducted using the same adult men (n=6; 22-32 y) in each of three trials (Table 4.1). All subjects were in good health and exhibited normal blood chemistry and hematological values. The subjects were in normal vitamin B-6 status as determined by plasma pyridoxal phosphate (PLP) and urinary 4-pyridoxic acid (4 PA) concentration, erythrocyte aspartate aminotransferase (AspAT) activity, and erythrocyte AspAT stimulation by in vitro addition of PLP, as judged by previously published criteria (137-138) . The procedures for selection of subjects and the experimental protocol were

PAGE 67

58 approved by the University of Florida Institutional Review Board and an informed consent was obtained from each subject. Three trials were conducted to compare the in vivo metabolic utilization of d 2 PN using the same group of six subjects. The experimental period lasted seven weeks consisting of three trials, each lasting one week with a two week wash out period between each trial. A self-selected diet was consumed for the duration of the study. On the first day of each trial the subjects collected a 24-h urine sample into foil-covered polyethylene containers and kept it refrigerated during the collection period. After an overnight fast, a five ml blood sample was collected into a Vacutainer brand tube (Becton Dickinson, Rutherford, NJ) containing ethylenediaminetetraacetic acid (EDTA) . Following these sample collections the subjects consumed a single oral dose of 5 jimol d 2 PN and either 2jumol PN or l/xmol PN + 1/xmol PN-glucoside or 2/xmol PN-glucoside (trials 1, 2, and 3 respectively) (Table 4.2) contained in 250 ml of apple juice. The doses of unlabeled PN and PN-glucoside were selected to provide and equal molar amount (2/xmol) of either PN, PN+ PNglucoside, or PN-glucoside. Urine collection was continued for the following 48-h period. Urine was analyzed for creatinine, total 4 PA, and d 2 4PA. The blood samples were centrifuged and the plasma collected was analyzed for plasma PLP and the erythrocytes were used to determine hemoglobin concentration and AspAT activity.

PAGE 68

59 Analytical methods Urinary 4 PA and plasma PLP were determined by reversephase HPLC procedures with fluorometric detection (48,109). Erythrocyte AspAT activity was measured by a spectrophotometric assay procedure (138) with commercially available reagents (Sigma Chemical, St. Louis, MO) . Urinary creatinine was determined based on the method of Heinegard and Tiderstrom (139) , while hemoglobin in erythrocyte hemolysates was determined spectrophotometrically as the cyanomethemoglobin derivative (140) . Mass spectral analysis of deuterium-labeled 4 PA Urinary d 2 4PA was determined by gas chromatography-mass spectrometry (GCMS) following isolation of 4 PA from urine samples by cation exchange chromatography (Bio-Rad AG 50W-X8, 100-200 mesh, H + form) and reverse-phase HPLC (Whatman, Partisil 10 ODS-3 Magnum 9 column, 9mm i.d. x 25 cm) as described by Gregory et al. (82) . GCMS was performed in the electron capture negative ionization mode (Model 4500 GCMS system, Finnigan MAT, San Jose, CA) with a DB-5 capillary gas chromatographic column (J&W Scientific, Folsom, CA) and methane as a reagent gas. The derivatization (141) was performed by dissolving the dried 4 PA sample in 0.5 ml of 1:1 (v/v) solution of pyridine: acetic anhydride, heated at 100° C for 90 min to form the 3-acetyl-4PA-lactone which is evaporated to dryness under a stream of nitrogen. The 3-

PAGE 69

60 acetyl-4PA-lactone was dissolved in 50 nl ethyl acetate and 2 Ml of the resulting solution injected into the GCMS injection port. As this 4 PA derivative eluted from the GC column, electron capture negative ionization yielded two anions at m/z 207 and 209 for the d 0 and d 2 4 PA species, respectively. Selected-ion monitoring at m/z 207 and 209 permitted simultaneous measurement of the d 0 and d 2 4 PA species during GCMS analysis. A series of 4 PA standards of known molar ratios of d 0 and d 2 species were prepared and analyzed to facilitate quantitative analysis. Calibration curves were constructed to relate the ratios of observed GCMS peak areas to actual molar isotope ratios of the d 2 /d 0 4 PA species. The urinary excretion of d 2 4PA was calculated using the isotope ratio determined by GCMS, the concentration of total urinary 4 PA as determined by reverse-phase HPLC, and total urine volume . Statistical analysis The urinary excretion of d 2 4PA was expressed as a percentage of the administered dose of labeled PN. The excretion of d 2 4PA (as a percentage of administered dose) , total urinary 4 PA, and vitamin B-6 nutritional status parameters (see below) was compared by the method of least squares analysis of variance using general linear model procedures of SAS (123), following a log transformation of the data to normalize variance when necessary. Data for d 2 4PA and

PAGE 70

61 total urinary 4 PA are reported as least squares mean ± the pooled standard error of the least squares mean (SEM) . Orthoqonal contrasts were made to examine the linear and quadratic effects of different doses of PN-qlucoside on dependent variables. The quadratic effects were not significant for any dependent variable. Values with p<0.05, unless otherwise stated, were considered statistically significant and were all for linear effects of the amount of PN-glucoside (123). Results Vitamin B-6 nutritional status of subjects The principal criteria used for assessing the vitamin B-6 status of the subjects were plasma PLP, urinary 4 PA excretion per 24-h, and erythrocyte aspartate aminotransferase stimulation by in vitro addition of PLP. As shown in Table 4.1, mean values for these subjects were within the proposed guidelines for vitamin B-6 nutritional adequacy (137-138) . These data indicate the adequate vitamin B-6 status of these subjects prior to the administration of labeled vitamin B-6 in each trial, which reflects the adequacy of their self-selected diets prior to and between experiments. Subjects demonstrated no statistical significant changes between trials for these vitamin B-6 nutritional status parameters.

PAGE 71

62 Table 4.1 Indicator of vitamin B-6 nutritional status of human subjects 24 h prior to each trial 1 . TRIAL URINARY 4 PA PLASMA PLP ERYTHROCYTE /xmol/24 H nmol/L AspAT STIMULATION BY PLP % TRIAL 1 10.8±1.9 86.9±7.4 35.8±4.5 (5/imol d 2 PN + 2/xmol PN) (n=6) TRIAL 2 8.0±1.8 82.4±10.3 36.4±6.3 (5/imol d 2 PN + 1/Limol PN + 1/imol PN-GLUCOSIDE) (n=6) TRIAL 3 7.410.7 83.9±9.8 16.7±5.9 (5/imol d 2 PN + 2jUmol PN-GLUCOSIDE) (n=6) Wean and SEM, n=6; Blood and urine collections were made 24 h prior to administration of labeled forms of vitamin B6. Proposed guidelines for adequacy of vitamin B6 status are: urinary 4 PA excretion, >5/xmol/24 h (ref . 139) ; Plasma PLP, >50 nmol/1 (ref. 139) ; And in vitro stimulation of erythrocyte aspartate aminotransferase (AspAT) by added PLP, < 50% (ref. 140) . Stable-isotopic trials This study was conducted to determine the effect of PNglucoside on the metabolic utilization of d 2 PN by evaluating the excretion of d 2 4PA following the ingestion of an oral dose of d 2 PN while increasing the dose of PN-glucoside across the three trials. The detection of urinary d 2 4PA was conclusive

PAGE 72

Table 4.2 Molar isotopic ratio of urinary d 2 4PA. 63 EXPERIMENTAL TRIALS SUBJECT TRIAL 1 TRIAL 2 TRIAL 3 (5/xmol d 2 PN + (5/imol d 2 PN + (5/imol d 2 PN 2/imol PN) l/nmol PN + 1/imol 2/imol PN(n=6) PN-GLUCOSIDE) GLUCOSIDE) (n=6) (n=6) d2/d0 0-24 hr 1 0.20±.001 0.091.002 0.071.010 2 0.071.001 0.101.001 0.121.003 3 0.071.003 0.081.003 0.091.001 4 0.171.002 0.151.010 0.101.001 5 0.131.001 0.011.001 0.071.001 6 0.101.001 0.081.001 0.121.002 mean 0.121.021 0.091.018 0.091.009 d2/d0 24-48 hr 1 0.051.000 0.101.001 0.061.000 2 0.091.001 0.061.000 0.031.000 3 0.101.000 0.101.000 0.031.000 4 0.071.001 0.101.000 0.041.001 5 0.091.000 0.101.000 0.061.000 6 0.051.000 0.191.001 0.071.000 mean 0.071.007 0.111.017 0.051.006 1 Mean and SEM, n=6. Values are average of three injections per sample. evidence of the utilization of d 2 PN in vitamin B-6 metabolism. Deuterium-labeled 4 PA was detected in urine over 48-h postdose. The d 2 4PA comprised a small portion of the total urinary 4 PA using this experimental design. However, the isotope ratio observed was measurable by the GCMS method employed. Twenty-four hours following the administration of

PAGE 73

64 Table 4.3 Percentage of d 2 PN (5/xmol) dose excreted as urinary d 2 4PA. EXPERIMENTAL TRIALS TIME TRIAL 1 TRIAL 2 TRIAL 3 (H) (5/xmol d 2 PN (5/imol d 2 PN + (5jimol d 2 PN + 2/imol PN) 1/imol PN + l/mol 2/xmol PN(n=6) PN-GLUCOSIDE) GLUCOSIDE) (n=6) (n=6) (/xmol) (/imol) (/imol) (SEM) 0-24 46.5 27.2 24.3 12.7 24-48 37.1 28.6 17.0 12.6 0-48 83.6 55.8 41.3 17.7 1 Values based on n=6 are means and pooled standard error of the mean (SEM) . The quadratic effects were not significant for any dependent variables. The values in table are significantly different at p<0.05 for linear effects of the amount of PN-glucoside. the isotopically labeled d 2 PN, the mean d 2 /d Q molar isotopic ratio of labeled 4 PA was 0.12 ± 0.02, 0.09 ± 0.02, and 0.09 ± 0.01 (mean ± SEM) for trial 1, 2, and 3 respectively (Table 4.2). These were not significantly different. The 24-48 hr molar isotopic ratios are also not significantly different (Table 4.2) . The group mean values for excretion of d 2 4PA as a percentage of the administered d 2 PN dose between trials 1 and 3 for the 24-h post-dose period decreased linearly with the amount of PN-glucoside (Table 4.3). Total 4 PA excreted over a 24-h and 48-h post-dose period showed group mean values for

PAGE 74

65 Table 4.4 Total urinary 4 PA excretion. EXPERIMENTAL TRIALS TIME TRIAL 1 TRIAL 2 TRIAL 3 (H) (5Mmol d 2 PN (5/xmol d 2 PN + (5/imol d 2 PN + 2/xmol PN) 1/xmol PN + 1/imol 2/Ltmol PN(n=6) PN-GLUCOSIDE) GLUCOSIDE) (n=6) (n=6) (/imol) (/imol) (/umol) (SEM) 0-24 26.0 13.5 10.3 3.9 24-48 15.5 10.5 11.5 4.8 2 0-48 41.5 24.0 21.8 7.6 3 Values based on n=6 are means and pooled standard error of the mean (SEM) . The quadratic effects were not significant for any dependent variables. The values in table are significantly different at p<0.05 for linear effects of the amount of PN-glucoside. 2 p<0.5 3 p<0.08 trials 1 and 3 to decrease linearly with the amount of PNglucoside (Table 4.4). These results indicate that the percentage of d 2 PN dose excreted as d 2 4PA showed an inverse relationship in proportion to the PN-glucoside dose and that total 4 PA decreased linearly with the amount of PN-glucoside. Discussion Because PN-glucoside is a major form of vitamin B-6 in plant derived foods, its influence on the utilization of

PAGE 75

66 nonglycosylated species of the vitamin is nutritionally significant. These effects, which may be minor in a mixed diet containing both animal and plant derived foods, may be much more pronounced in vegetarian diets. Kabir et al. (65) showed an inverse relationship between the content of PNglucoside in a food and the net bioavailability in humans. Our present knowledge of the essential role of vitamin B6 in preserving health emphasizes the importance of establishing an accurate Recommended Dietary Allowance (RDA) for vitamin B-6. A vitamin B-6 deficiency with apparent clinical symptoms is rare in the general population, but the vitamin B-6 requirement has not been clearly determined. It is significant that over 50% of those evaluated in the 1977-78 USDA Food Consumption Survey consumed only 70% of the 1980 RDA for vitamin B-6. Females and the elderly were especially prevalent in the group consuming only 70% of the RDA (2) . The 1985 Continuing Survey of Food Intakes of Individuals indicated that only 27% of women consumed 70% or more of the RDA for vitamin B-6 (143-144) . Although the range of intakes is large, substantial portions of the population consume apparently marginally adequate amounts of vitamin B-6. Adequate vitamin B-6 nutriture is essential to health through the multiplicity of roles of its active coenzymatic form pyridoxal 5' -phosphate (PLP) . The catabolic pathway for PLP involves enzymatic hydrolysis of the phosphate ester bond, and the subsequent

PAGE 76

67 oxidation of pyridoxal (PL) to 4 PA (130-132). As a terminal metabolite of vitamin B-6 metabolism, urinary 4 PA reflects the in vivo metabolic utilization of the vitamin. The deuterated form of PN used in this study is labeled at a metabolically stable position (i.e., 5'-CH 2 OH) and, therefore, retains the label throughout absorption, metabolism, transport, and excretion (82) . The results of this study demonstrate a negative correlation between the quantity of PN-glucoside administered and the amount of urinary d 2 4PA derived from oral d 2 PN. The data also showed that an inverse relationship between the amount of total urinary 4 PA excreted and the amount of PNglucoside in the dose. These findings are consistent with those observed when comparing the metabolism of [ 14 C]PN in the presence of increasing doses of PN-glucoside in the rat (136) . The PN derived from hydrolysis of PN-glucoside enters into vitamin B-6 metabolism to produce other forms of vitamin B-6, including 4 PA (77,81). However, the PN-glucoside that is not hydrolyzed to PN may not be totally inert but instead appears to affect the metabolism and interconversion of the B-6 vitamers. The possible inhibition by PN-glucoside of PL kinase and/or PNP-oxidase, two enzymes in the metabolic interconversion of B-6 vitamers, is discussed in chapter 5. The results of these isotopic studies provide evidence that PN-glucoside alters the metabolism of PN in humans and rats by

PAGE 77

68 partially retarding the utilization of nonglycosylated B-6 compounds . The appropriateness of the RDA is difficult, if not impossible to assess without accurate information concerning the content and forms of B-6 compounds in foods consumed, and accurate data regarding vitamin B-6 bioavailability. This research provides evidence of interactions amoung various B-6 compounds found in a typical mixed diet through an influence of PN-glucoside on the utilization of PN. Such an interaction represents another factor involved in the bioavailability of vitamin B-6 and must be considered in fully understanding the adequacy of vitamin B-6 intake. Future research will address the nutritional significances of these interactions amoung the vitamin B-6 compounds and the importance of these compounds in assessing the nutritional status of the population for this vitamin.

PAGE 78

CHAPTER 5 EFFECTS OF PYRIDOXINE-5 1 -0-D-GLUCOSIDE ON THE IN VITRO KINETICS OF PYRIDOXAL KINASE AND PYRIDOXAMINE (PYRIDOXINE) -5 • PHOSPHATE OXIDASE IN RAT LIVER Introduction There are a number of enzymatic reactions in the metabolic pathway responsible for the interconversion of vitamin B-6 compounds. In particular, PL kinase and PMP (PNP) oxidase are considered the two most important catalytic enzymes in the formation of the coenzymatic form of vitamin B-6, PLP (144) . Although other tissues contribute to vitamin B-6 metabolism, the liver is thought to serve as the center of interconversion of B-6 vitamers to PLP (109-110). The B-6 vitamers are postabsorptively taken-up by the liver and converted to PLP. The non-phosphorylated vitamers are converted to their 5 1 -phosphorylated forms by a single kinase (144) . Following phosphorylation, PMP and PNP are oxidized by PMP (PNP) oxidase to form PLP. This active coenzymatic form of vitamin B-6 can then either be bound by cellular apoproteins, released into circulation as PLP, or dephosphorylated by alkaline phosphatase to PL (144). The PL 69

PAGE 79

70 can then itself be released by the cell, oxidized to 4PA, the metabolic end product of B-6 catabolism, or rephosphorylated to form PLP. Previously discussed studies in chapters 3 and 4 provided evidence that PN-glucoside quantitatively alters the metabolism and retention of administered PN (136) . These studies indicated that unlabeled PN-glucoside affects the metabolic utilization of a simultaneous dose of labeled PN (136) . These data showed a linear relationship between the dose of PN-glucoside and hepatic radiolabeled PNP and PLP (136) . Hepatic U C PN, PL, and PM were inversely proportional to the dose of PN-glucoside. The data from these studies led us to postulate that the effect on vitamin B-6 metabolism exhibited by PN-glucoside may be caused by a direct affect of the glycosylated vitamer on the action of either PL kinase, PMP (PNP) oxidase, or both. The purpose of this study was to determine the effect of PNglucoside on the in vitro catabolic activity of PL kinase and PMP (PNP) oxidase partially from purified rat liver. Materials and Methods Protocol Two studies were conducted to evaluate the effect of PNglucoside on the reaction rates of the enzymes PL kinase and PMP (PNP) oxidase. In each of the studies six male Sprague Dawley rats (Crl:CD(SD) BR) from Charles River Breeding

PAGE 80

71 Laboratories (Wilmington, MA) , weighing 200-300 g, were housed individually in stainless steel cages in animal quarters maintained at 24±1° C with a 12 h light/dark cycle. All procedures for the care and treatment of the experimental animals were in accordance with the National Institutes of Health Guidelines. The rats were fed ad libitum a casein-sucrose (20%: 60%) based diet, adequate in all micronutrients including 7 mg PN HCl/kg diet for seven days (80) . At the end of the seven day acclamation period the rats were killed by lethal injection of sodium pentobarbital (W.A. Butler Co., Columbus, OH). The livers were rapidly excised and stored at -20° C until analysis. The enzyme assay was performed at the temperature and reaction time determined from preliminary experiments and pH as shown in method of Merrill and Wang (116) . Four varying concentrations of PN-glucoside ranging from 0 juM to 150 jiM and six varying substrate concentrations were used. Preliminary experiments were also performed to determine the linear ranges of the assays with respect to enzyme concentration and reaction time. Forms of vitamin B-6 PN-glucoside was prepared by biological synthesis using the propagation of alfalfa sprouts germinated in the presence of PN-HC1 (Sigma Chemical Co., St. Louis, MO) (80). The [4,53 H]PN-HC1 684.5 KBq/nmol (Hof fmann-LaRoche, Nutley, NJ) with

PAGE 81

72 a purity of greater than 98%, as determined by ion-pair reverse phase HPLC. The pyridoxamine-phosphate-HCl and pyridoxal -5 1 -phosphate were obtained from (Sigma Chemical Co., St. Louis, MO) . Pyridoxal kinase Extraction . Pyridoxal kinase was extracted from the liver tissue of six rats using the procedure of Merrill et al. (112) . The tissue was stored at -20° C and processed within 72 h following removal from the rat. To each 1 g of liver, 4 ml of 50 mM 4(2-hydroxyethyl) -1-piperazineethanesulfonic acid (Hepes) (Sigma Chemical, Co., St. Louis, MO) at pH 7.4 and 25° C, was added. The mixture was homogenized 30-45 s with a Polytron PT 20 (Brinkman Instruments, Westbury, NY).> The resulting homogenates were centrifuged (Model L, type 40 rotor, Beckman Instruments, Pala Alto, CA) at 105,000 x g for 1 h at 4° C. The supernatant was collected and stored at -20" C Purification . The extracted PL kinase was purified using the column chromatography method employed by Cash et al. (145) . Preparation of the 4-pyridoxyl-Sepharose as an affinity support was adapted from the method used by Kwok and Churchich (146) . Seven grams of aminohexyl-Sepharose (Sepharose AH, Pharmacia, Ltd. Piscataway, NJ) was pretreated and washed with water. The Sepharose AH slurry was added to a solution of 50 ml water containing 1 g PL-HC1 (Sigma Chemical Co., St. Louis, MO) adjusted to pH 7.0 by addition of

PAGE 82

73 NaOH. The slurry was shaken 16 h in the dark at ambient temperature and then placed in an ice bath with the dropwise addition of ice cold borohydride until disappearance of the yellow color. During this period the slurry was shaken and 7% acetic acid was added to prevent the pH from rising above 9.0. The slurry was adjusted to pH 6.0 with 7% acetic acid and packed into a 1.6 x 15 cm chromatography column. The column was rinsed with 3M KC1 followed by water prior to use. All rinsing and eluting buffers contained 2mM potassium phophate (Fisher Scientific, Fairlawn, NJ) at pH 7.0 and lOmM glutathione. The 105,000 x g supernatant was applied to the PLSepharose column and washed through with 50 ml of 100 mM KC1 buffer. The column was then rinsed with 500 ml 400mM KC1 followed by 50 ml of lOOmM KCl buffer. The PL kinase was then eluted and adsorbed directly onto a Bio-Gel HTP hydroxy 1 apatite column (Bio Rad Laboratories, Richmond, CA) , previously equilibrated with buffer containing no KCl, with 500 ml 10 mM PN adjusted to pH 7.0 in buffer and rinsed with 25 ml lOOmM KCl buffer. The hydroxyapatite column was eluted with 100 ml of a linear gradient of 2-300mM potassium phosphate at pH 7.0. Approximately 3 ml fractions were collected and the active fractions pooled. The enzyme preparation was then tested for glucosidase activity using the method of Daniels et al. (147) and for phosphatase activity employing the PL kinase assay (discussed below) using PLP as

PAGE 83

74 substrate followed by an inorganic phosphate assay kit (Sigma Chemical Co., St. Louis, MO) and measuring the percent conversion of PLP to PL. Kinase assay . The assay procedure used was that of Merrill and Wang (116) . All reagents and buffers were at pH 5.75 and the procedure was conducted in GEF40G0 gold lights that emit at a wavelength between 500 nm and 750 nm to prevent photodegradation of pyridoxyl compounds. The assay mixture contained 10 /iL each of 0.2M potassium phosphate, lOmM ATP ( Sigma Chemical Co., St. Louis, MO), 0.8mM ZnCl 2 , 0.6mM KCl, 0.3mM (20MCi/mmol) [ 3 H]PN diluted with unlabeled PN to yield the appropriate specific activity, and 100 nq of enzyme extract with enough water to make a final volume of 100 tiL. The mixture was incubated for 30 min at 37 °C and the reaction halted with 0.5 ml ice cold 10 mM ammonium formate. The samples were transfered to columns which contained 0.2 ml of DEAE-cellulose (Sigma Chemical Co., St. Louis, MO). The columns were rinsed with 12 ml of lOmM ammonium formate, followed by 2 ml water and eluted with 2 ml of 0.5 M KCl. A 2 ml fraction of the final elution was collected and added to 15 ml ScintiVerse II (Fisher Scientific, Fair Lawn, NJ) and counted for total radioactivity. The reaction rates were calculated using the following equation of Merrill and Wang (116) : C P m reaction ~ C P m t ime 0 Enzyme units = X 2 X quench factor Specific Activity (enzyme preparation)

PAGE 84

75 Pyridoxamine (pyridoxine) 5' phosphate oxidase Extraction . The rat liver PMP (PNP) oxidase was extracted by the method of Kazarinoff and McCormick (119) . Liver tissue was homogenized in 0.02M potassium phosphate buffer containing O.lmM mercaptoethanol , pH 7.0 and a ratio of 50 g tissue to 200 ml buffer. The homogenate was centrifuged at 18,000 x g for 30 min and the supernatant adjusted to pH 5.0 with 2N acetic acid while stirring constantly. After 10 min the precipitate was removed by centrifugation at 18,000 x g for 15 min. Potassium chloride (3.6 g) was added to the supernatant and the volume brought up to 400 ml with water. The pH was readjusted to 5.0 with 2N acetic acid. Purification . The PMP (PNP) oxidase was extracted from 6 rat livers using the method of Kazarinoff and McCormick (119) . The clear supernatant resulting from the extraction procedure was heated to 50 "C and held for 8 min then plunged into ice and cooled to 20 "C and centrifuged at 18,000 x g for 30 min. The supernatant was applied to a DEAE-A50 column (Sigma Chemical Co., St. Louis, MO) which had been prepared by swelling overnight in 0.1 M potassium phosphate (pH 8.0). A linear gradient established between 0.1 and 0.2 M potassium phosphate at pH 8.0 was used to elute the column. Fractions were collected and read at A 280 . The active fractions were pooled. Solid ammonium sulfate (Sigma Chemical Co., St. Louis, MO) (22.8 g/100 ml, 30% saturation) was added. After 30 min at 4°C the precipitate was removed by centrifugation at

PAGE 85

76 18,000 x g for 30 min and addition of ammonium sulfate (15.2 g/100 ml, 50% saturation) . After 60 min the solution was centrifuged and the active precipitate was dissolved in one tenth the initial volume of water. This solution was dialyzed against 0.02 M potassium phosphate buffer for 24 h and centrifuged to remove the precipitate. The clear yellow supernatant was applied to a column of Sephadex G-100 (Sigma Chemical Co., St. Louis, MO) eguilibrated with 0.02 M phosphate buffer and eluted with the same buffer. Fractions were collected and read at an A 280 and the active fractions pooled. Oxidase assay . PMP (PNP) oxidase activity was measured by the method of Wada and Snell (31) using phenylhydrazine as the color development agent. The assay mixture contained 1 ml of ImM PMP solution in 0.2M Tris-HCl, pH 8.0, 10-200 units of enzyme (one unit of oxidase enzyme catalyzes the formation of lnmol PLP/mL/min at 37 °C) , and enough Tris buffer to bring the total volume to 3.5 ml . The mixture was gently shaken for 30 min at 37 °C. The reaction was stopped by the addition of 0.3 mL of 100% (w/v) trichloroacetic acid. There was no formation of a precipitate therefore, the entire supernatant was used for the color reaction. The 3 mL supernatant was placed in a clean test tube, 0.2 mL phenylhydazine reagent (2 g dissolved in 100 mL 10N H 2 S0 4 ) was added and the mixture was heated for 20 min at 60 °C. The reaction mixture was then cooled to room temperature and the absorbance read at 410 nm

PAGE 86

77 verses a reagent blank. The activity was expressed in nmol PLP/mL enzyme/h. Statistical analysis The kinetic parameters 1^ and were calculated by non linear regression using the EZ-FIT program (149) . Confidence intervals for each K,,, and were calculated by the equation: b, ± t 975 X SE, where b, = 1^ or values (123) . Overlapping confidence intervals are not significantly different at p<0. 05. Results These studies were conducted to determine the effects of PN-glucoside on two enzymes, PL kinase and PMP (PNP) oxidase, in the metabolic pathway responsible for the interconversion of vitamin B-6 compounds. These catalytic enzymes were investigated, in vitro, by evaluating the rates of reaction in the presence and absence of PN-glucoside. The purification methods used for rat liver PL kinase and PMP (PNP) oxidase yielded an enzyme preparation acceptable for kinetic analysis. PL kinase Partial purification of the enzyme by the above technique resulted in a 243-fold increase in specific activity relative to the crude homogenate (Table 5.1). Preliminary experiments indicated that the rate of formation of [ U C]PNP was linear over varying enzyme concentrations and the rate of

PAGE 87

78 Table 5.1 Purification of pyridoxal kinase from rat liver. Fraction Vol Total Protein Units/mg Purification (ml) Units (mg/ml) protein (fold) Supernatant 80 509 0.25 2.54 PL-Sepharose 6 188 0.05 628 247 /Hydroxyapatite Units are nmol PNP/min/ml at 37 °C in phosphate buffer pH 5.75. 50 g rat liver was used. phosphorylation was constant over at least a 60 min period under the assay conditions outlined above. The rate of phosphorylation of PN in this study indicated a of 21 /tiM which is in general agreement with the previously reported value of 25 /xM measured under similar conditions (9) . In the presence of varying concentrations of PN-gluoside the 1^ value at all levels was not significantly different from that observed with 0 jzM PN-glucoside (Table 5.2) . The V max was 16.8 nmol of PNP/h/mg of protein and did not differ significantly over the range of PN-glucoside concentrations (Table 5.2). The kinetic parameters (Figure 5.1) were calculated from analysis using the EZ-FIT Program (149) .

PAGE 88

79 Table 5.2 Kinetic parameters of pyridoxal kinase with varying concentrations of PN-glucoside. PN-glucoside concentration (MM) (MM) V. max (nmol PNP/h/mg protein) 0 50 100 150 23±2.0 22±2.2 21±2.4 20±3.0 0.29±.01 0.28±.01 0.281.01 0.27±.01 PMP (PNP) oxidase The purification scheme yielded a preparation of 24-fold increase in enzyme purity compared to the crude homogenate (Table 5.3) . The conversion of the substrate, PMP, to PLP has previously been shown to exhibit a 1^ value of 25/xM (8) . The present study indicated a similar 1^ closely related value of 23/iM. Values for 1^ of the substrate in the presence of all concentrations of PN-glucoside were not significantly different (Table 5.4). The maximum velocity values in this study did not differ signifcantly from the control value of 1.65 nmol PLP/min/mg protein (Table 5.4). All kinetic parameters (Figure 5.2) were calculated using the EZ-FIT Program (149) .

PAGE 89

80 Table 5.3 Purification of pyridoxamine (pyridoxine) phosphate oxidase from rat liver. Fraction Vol Total Protein Units/ Purification (ml) Units (mg/ml) mg protein (fold) Supernatant 200 11,300 15. 0 3.77 Acid ppt. 250 7, 100 4.5 6.25 1.65 Ammonium Sulfate 25 3,267 12.1 10.8 2.86 DEAE Sephadex 30 10, 100 4.2 78.3 20.77 Sephadex G-100 45 6,961 1.7 91.1 24.16 Units are nmol PNP/min/ml at 37 °C in phosphate buffer pH 5.75. 50 g of rat liver was used. Table 5.4 Kinetic parameters of pyridoxamine (pyridoxine) phosphate oxidase with varying concentrations of PN-glucoside. PN-glucoside concentration (MM) (MM) V max (nmol PLP/h/mg protein) 0 50 100 150 23±3.1 27±1.0 24±2.2 24±2.0 1.65±0.10 1.69±0.09 1.63±0.11 1.70±0.06

PAGE 90

4J81 0» > •H -H fa H

PAGE 91

romomoinomo fO CN N N W r-' r-' r-' o CO CN CM ^ T 00 00 O CD CN 00 CN 2 3 Ul 5 Q. CO O X Q. I LU Z < X o g K >0. m o in o rin cn q dodo S 82 d •H X o (6LU/UILU/dld |OUJU) A CM H H &» O •rl ^ fa
PAGE 92

83 Discussion An earlier study from this laboratory demonstrated that the conjugated vitamin B-6 compound, PN-glucoside, affected the metabolic utilization of PN when administered in a simultaneous dose, in rats (136) . The concentration of phosphorylated forms of vitamin B-6 increased in proportion to the quantity of PN-glucoside. These data further indicated a strong possibility that the effect PN-glucoside exhibited on vitamin B-6 metabolism may be a direct effect on one or both of the two enzymes responsible for the interconversion of B-6 vitamers to the active coenzyme form, PLP. Hence, the present study was undertaken to determine the in vitro effects of PNglucoside on the reaction rates of PL kinase and PNP (PNP) oxidase. The results showed that neither PL kinase nor PMP (PNP) oxidase were effected by the concentrations of PN-glucoside used in the study. The kinetic parameters, and V^, were within the expected ranges for rat liver and demonstated no significant differences when measured under conditions of varying substrate and PN-glucoside concentrations. When conducting in vitro kinetic studies consideration should be given to any potential differences that may occur under in vivo conditions. In the rat, vitamin B-6 postabsorptively diffuses into the liver where it is

PAGE 93

84 metabolically trapped via phosphorylation. The vitamin is maintained predominantly in phosphorylated forms due mainly to protein binding and a ten fold greater activity of PL kinase than phosphatase. It has been shown that hydrolysis of PLP is inhibited by substrate and PMP (112) . Rat liver PMP (PNP) oxidase is inhibited by its product, PLP, and the presence of PLP increases both the K,,, and (148) . The increase in V MX may be a result of the action of the more active rat liver phosphatases or an inability to accurately measure initial velocity due to product inhibition at early reaction times (148) . Phosphatase more rapidly hydrolyzes PLP when this product is free rather than protein bound (144) . The partially purified enzyme preparation used in this study showed minimal phosphatase activity. Approximately 20% of the PLP in vivo is in the free or loosely bound form and the remaining PLP is bound to proteins. Therefore, in vivo product inhibition may differ significantly from in vitro inhibition. However, studies have shown that crude rat liver homogenate compared with pure enzyme is inhibited to a similar extent (148) . Whether or not PLP inhibits PMP (PNP) oxidase is dependent on the concentrations of available substrate and product formed. Greater substrate inhibition has been reported with PNP, in concentrations as small as 5 jxm, due to a reduced enzyme-substrate complex that reacts less efficently with oxygen (112) . This has significant consideration, since the presence of increasing

PAGE 94

85 concentration of PN-glucoside during the metabolism of PN causes proportional increases in the concentration of liver PNP (136) . In summary, the results from this study indicate that, in vitro, PN-glucoside has no direct effect on PL kinase or PMP(PNP) oxidase. These data, provide evidence that these biosynthetic enzymes are not directly involved in the changes in distribution of B-6 vitamers caused by ingested PNglucoside, observed in the previous study (136) .

PAGE 95

CHAPTER 6 SUMMARY AND CONCLUSIONS Since the isolation and identification of PN-glucoside from rice bran (67) much focus has been on the comparative bioavailability of this glycosylated form and other vitamin B6 compounds. A number of studies spanning the last decade has provided a great deal of information regarding the biological availability of vitamin B-6 relative to its multiple forms. As discussed in previous chapters, between 5%-80% of the vitamin B-6 found in a variety of plant derived foods was in the glycosylated form and both food processing and diet composition are important considerations when evaluating the bioavailability of vitamin B-6. Also the net bioavailability of PN-glucoside relative to PN is less than 40% in rats (75,77,90). Enzymatic hydroxylation of PN does not exhibit antivitamin B-6 activity, however other degradative products such as pyridoxallysine have been shown to be vitamin B-6 analogues and act as competitive inhibitors to enzymes in the vitamin B-6 metabolic pathway. The use of stable iotopic methodology showed PN-glucoside to be 58% as biologically active as PN in humans (41) . This study indicated that although PN-glucoside bioavailability was 86

PAGE 96

87 incomplete, it is approximately two fold greater than previously found in rats. The utilization of stable isotopic methodology in humans continues to contribute information regarding the absorption, metabolism and utilization of vitamin B-6 in humans. In an effort to advance our understanding of vitamin B-6 nutriture in humans it is necessary to gather information regarding the dietary forms of vitamin B-6, their bioavailability and the enzymes responsible for the interconversion of B-6 vitamers. It is equally important to determine any potential interaction between multiple forms of B-6 compounds and the effect these may have on metabolic utilization of B-6 vitamers when consumed together. The purpose of the present research was to determine the effect of PN-glucoside on the metabolic utilization of PN when administered in a simultaneous dose in rats and humans. The adventive of stable isotopic techniques allowed for the evaluation in human subjects. The effect of PN-glucoside on enzymes responsible for control of the conversion of vitamin B-6 compounds to the active coenzyme form, PLP was also determined. The research discussed in chapters 3-5 was divided into three separate but interconnected studies. The experimental design of each study was contingent on the findings of the prior study or studies. Initially, an animal study was performed to determine if any interaction between PN-glucoside

PAGE 97

88 and PN occurred, which affected metabolic utilization. The premise of these studies was to simulate ingestion of a mixed diet by determining the effect of a simultaneous dose of the two B-6 compounds. Since the interest was in evaluating a direct affect of PN-glucoside on PN metabolic utilization, the rats received labeled-PN and varying concentrations of unlabeled-PN-glucoside. The results indicated an effect of PN-glucoside on the hepatic metabolism of PN. There existed a linear relationship between the guantity of phosphorylated B-6 vitamers and increasing concentrations of PN-glucoside. It was also noted that the concentration of PNP, a B-6 vitamer not detected in control rat livers, was directly proportional to the concentration of PN-glucoside in the dose. The reason for this finding is presently unclear. These data indicated a negative correlation between the amount of 4 PA detected in urine and the concentration of PN-glucoside administered in the dose. A similar correlation to that of 4 PA existed for non-phosphorylated B-6 vitamers in relation to PN-glucoside concentration in the dose. These results indicated that the PN-glucoside which was not hydolyzed and therefore metabolically active as PN, is not biologically inert. Instead this intact glycosylated B-6 compound exhibits an effect on the utilization of vitamin B-6 in the rat. As previously discussed, studies showed that the bioavailability of PN-glucoside, although incomplete, was greater in humans than rats. In consideration of this

PAGE 98

89 observation, the effect of PN-glucoside on vitamin B-6 metabolism seen in rats would be significant, but of a lesser magnitude in humans, assuming that intact PN-glucoside is the active compound. A second study was designed using stable isotopic methods to determine the effect of PN-glucoside on metabolism of PN. Deuterium-labeled [5 1 -C 2 H 2 OH]pyridoxine (d 2 PN) was used in the study. The d 2 PN is metabolically stable and remains attached to the PN molecule throughout catabolism. Therefore, metabolic utilization of d 2 PN may be followed and ultimately quantified as the urinary catabolite d 2 4PA. The results of this study indicated a negative correlation between the quantity of total 4 PA excreted and the concentration of PNglucoside in the dose. Review of the results of these two studies showed that PN-glucoside, when administered simultaneously with PN, has an effect on the metabolic utilization of vitamin B-6 and that PN-glucoside which is not hydrolyzed by intestinal mucosal glucosidase, exhibits some biological activity regarding vitamin B-6 metabolism. To determine which vitamin B-6 metabolic process was affected by PN-glucoside, study three was proposed. Data from chapter 3 indicated increased quantities of phosphorylated B-6 vitamers, specifically PLP and PNP. These results indicated that either of the enzymes responsible for controlling the level of phosphorylated B-6 compounds may be affected by PN-glucoside. Study three was

PAGE 99

90 designed to determine the in vitro effects of PN-glucoside on PL kinase and PMP (PNP) oxidase. The results indicated no direct effect of PN-glucoside on either enzyme as determined by no significant differences observed between the kinetic parameters, K^, and V MX with varying concentrations of the glycosylated vitamin B-6 compound. Further studies are needed to assess potential effects of PN-glucoside on other enzymes in the interconversion pathway of vitamin B-6 metabolism. These experimenal results provide evidence that PNglucoside exhibits a quantitative effect on the metabolic utilization of PN. Although the biochemical mechanism responsible for this effect are presently uncertain, it is evident that consideration in the study of vitamin B-6 bioavailability must incorporate the potential interactions among B-6 compounds present in foods generally consumed in a mixed diet. The information from these studies is consequential in relation to current USDA Dietary Guidelines, which are used to convey recommendations to the qeneral population. Present nutrition education is promoting the consumption of greater quantities of plant derived foods. Food consumption data have shown that there are several sub-populations at nutritional risk for vitamin B-6, specifically pregnant females, adolescents, and the elderly .those individuals undergoing growth or aging.

PAGE 100

91 The predominant form of vitamin B-6 in many fruits and vegetables is PN-glucoside and research has indicated that not only is this vitamin B-6 compound biologically less available as a usable source of vitamin B-6 it also effects the metabolic utilization of other B-6 vitamers. In development of recommendations for dietary allowances and dietary food intake patterns it is necessary to consider not only vitamin B-6 availability and biological activity, but interactions that occur between different vitameric forms and how these interactions effect bioavialability of the nutrient.

PAGE 101

LITERATURE CITED 1. Canham J.E., Baker E.M., Harding R.S., Sauberlich H.E., Plough I.e. (1969). Dietary Protein Its Relationship to Vitamin B-6 Reguirements and Function. Ann NY Acad Sci 166: 16-29. 2. Food and Nutrition Board (1989). Recommended Dietary Allowances. Natl Acad Sci., Washington DC. 3. Meister A. (1965). The Function of Vitamin B-6 in Amino Acid Metabolism. In: Biochemistry of the Amino Acids, 2nd Ed., Vol. 1 Academic Press, New York. 375412. 4. Bender D.A. (1987). Oestrogens and Vitamin B-6 Actions and Interactions. World Rev Nutr Diet 51: 140. 5. Bunce G.E., Vessal M. (1987). Effect of Zinc and/or Pyridoxine Deficiency upon Oestrogen Retention and Oestrogen Receptor Distribution in the Rat Uterus. J Steroid Biochem 26: 303. 6. Rose C.S., Gyorgy P., Butler M. (1976). Age Differences in Vitamin B-6 Status in 617 Men. Am J Clin Nutr 29: 847-853. 7. Schuster K. , Bailey L.B., Mahan C.S. (1984). Effect of Maternal Pyridoxine HCL Supplementation on the Vitamin B-6 Status of Mother and Infant and on Pregnancy Outcome. J Nutr 114: 977-988. 8. Brophy M.H., Suteri P.K. (1975). Pyridoxal Phosphate and Hypertensive Disorders of Pregnancy. Am J Obstet Gynecol 121:1075-1079. 9. Ranke E. , Tauber A., Horonick B. (1960). Vitamin B-6 Deficiency in the Aged. J Gerontol 15: 41-44. 10. Driskell J. A. (1978). Vitamin B-6 Status of the Elderly. In: Human Vitamin B-6 Reguirements. Natl Acad Sci., Washington DC. 252-256. 92

PAGE 102

93 11. Kirksey A., Keaton K. , Abernathy R.P. (1978). Vitamin B-6 Nutritional Status of a group of Female Adolescents. Am J Clin Nutr 31: 946-954. 12. U.S. Department of Agriculture. (1986) Nationwide Food Consumption Survey Continuing Survey of Food Intakes by Individuals, Low Income Women 19-25 Years and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII Report No. 85-2. U.S. Department of Agriculture, Hyattsville, MD. 13. Shane B. , Contractor S.F. (1975). Assessment of Vitamin B-6 Status. Studies on Pregnant Women and Oral Contraceptive Users. Am J Clin Nutr 28: 739-747. 14. Roepke J.L.B., Kirksey A. (1979). Vitamin B-6 Nutriture During Pregnancy and Lactation. Am J Clin Nutr 28: 146-156. 15. Pao E.M., Mickle S.J. (1981). Problem Nutrients in the United States. Food Technol 35(9): 58-69. 16. Gregory J.F., Kirk J.R. (1981). The Bioavailability of Vitamin B-6 in Foods. Nutri Rev 39(1): 1-8. 17. Turner J.M. (1961). Pyridoxal Phosphate Breakdown by an Alkaline-Phosphatase Preparation. Biochem J 80: 663-668. 18. Leklem J.E. (1988) . Vitamin B-6: Of Reservoirs, Receptors and Requirements . Nutr Today Sept/Oct: 410. 19. Booth C.C., Brain M.C. (1962). The Absorption of Tritium Labelled Pyridoxine Hydrochloride in the Rat. J Physiol 164: 282-294. 20. Scudi J. V., Unna K. , Antopol W. (1940). A Study of Urinary Excretion of Vitamin B-6 by a Colorimetric Method. J Biol Chem 135: 371-376. 21. Scudi J. V., Koones H.F., Keresztesy J.C. (1940). Urinary Excretion of Vitamin B-6 in the Rat. Proc Soc Exptl Biol Med 43: 118-122. 22. Yamada K. , Tsuji M. (1968). Transport of Vitamin B-6 in Human Erythrocytes. J Vitaminol (Kyoto) 14: 282294. 23. Middleton H.M. (1979). In Vivo Absorption and Phosphorylation of Pyridoxine-HCl in Rat Jejunum. Gastroenterol 76: 46-49.

PAGE 103

94 24. Tsuji T., Yamada R. , Nose Y. (1973). Intestinal Absorption of Vitamin B-6. I. Pyridoxal Uptake by Rat Intestinal Tissue. J Nutr Sci Vitaminol 19: 401-417. 25. Yamada R.H. , Tsuji T. , and Nose Y. (1977). Uptake and Utilization of Vitamin B-6 and its Phosphate Esters by Escherichia coli. J Nutr Sci Viltaminol (Tokyo) 23(1) '.1-11 . 26. Middleton H.M. (1977). Uptake of Pyridoxine Hydrochloride by the Rat Jejunal Mucosa in vitro. J Nutr 107: 126-131. 27. Middleton H.M. (1979). Intestimal Absorption of Pyridoxal-5' Phosphate: Disappearance from Perfused Segments of Rat Jejunum in Vivo. J Nutr 109: 975-981. 28. Hamm M.W. , Mehansho H., Henderson L.M. (1979). Transport and Metabolism of Pyridoxamine and Pyridoxamine Phosphate in the Small Intestine of the Rat. J Nutr 109: 1552-1559. 29. Mehansho H. , Hamm M.W. , Henderson L.M. (1979). Transport and Metabolism of Pyridoxal and Pyridoxal Phosphate in the Small Intestine of the Rat. J Nutr 109: 1542-1551. 30. McCormick D.B., Gregory M.E., Snell E.E. (1966). Pyridoxal Phosphokinases . I. Assay, Distribution, Purification, and Properties. J Biol Chem 236: 20762084. 31. Wada H. , Snell E.E. (1961). The Enzymatic Oxidation of Pyridoxine and Pyridoxamine Phosphate. J Biol Chem 236: 2089-2095. 32. Pogell B.M. (1958). Enzymatic oxidation of Pyridoxamine Phosphate to Pyridoxal Phosphate in Rabbit Liver. J Biol Chem 232: 761-776. 33. Morisue T. , Morino Y., Sakamoto Y., Ichihara K. (1960) . Enzymatic Studies on Pyridoxine Metabolism. III. Pyridoxine Phosphate Oxidase. J Biochem 48: 2836. 34. Samra P.S., Snell E.E., Elvehjem C.A. (1947). The Bioassay of Vitamin B-6 in Natural Materials. J Nutr 33:121-128.

PAGE 104

95 35. Toepfer E.W. , Polansky M.M. , Richardson L.R., Wilkes S. (1963) . Comparison of Vitamin B-6 Values of Selected Food Samples by Bioassay and Microbiological Assay. J Ag Food Chem 11: 523-525. 36. Nguyen L.B., Gregory J.F. (1983). Effects of Food Composition on the Bioavailability of Vitamiln B-6 in the Rat. J Nutr 113: 1550-1560. 37. Gregory J.F., Ink S.L. (1985). The Bioavailability of Vitamin B-6. In: Vitamin B-6: Its Role in Health and Disease. Alan R. Liss, Inc., New York. 3-23. 38. Nelson E.W., Lane H. , Cerda J.J. (1976). Comparative Human Intestinal Bioavailability of Vitamin B-6 from a Synthetic and a Natural Source. J Nutr 106: 14331437. 39. Nelson E.W., Burgin C.W. , Cerda J.J. (1977). Characterization of Food Binding of Vitamin B-6 in Orange Juice. J Nutr 107: 2128-2134. 40. Leklem J.E., Miller L.T., Perera A.D., Peffers D.E. (1980) . Bioavailability of Vitamin B-6 from Wheat Bread in Humans. J Nutr 110: 1819-1828. 41. Gregory J.F. (1990). The Bioavailability of Vitamin B-6 — Recent Findings. Ann NY Acad Sci 585: 86-95. 42. Coburn S.P., Mahuren J.D. (1979). Major Urinary Metabolites of 4' and 5 ' -Deoxypyridoxines in Various Species. IRCS Med Sci 7: 556. 43. Rodwell V.W., Volcani B.E., Ikawa M. , Snell E.E. (1958) . Bacterial Oxidation of Vitamin B-6. Isopyridoxal and 5-Pyridoxic Acid. J Biol Chem 233: 1548-1554. 44. Coburn S.P., Hahuren J.D., Erbelding W.F., Tokwnsend D.W., Hachey D.L., Klein P.D. (1984). Measurements of Vitamin B-6 Kinetics in Vivo Using Chronic Administration of Labelled Pyridoxine. In: Chemical and Biological Aspects of Vitamin B-6 Catalysis, Part A (Evangelopoulos A. E. , ed.) Alan R. Liss, New York. 43-54.

PAGE 105

96 45. Coburn S.P., Mahuren J.D., Szadkowska Z., Schaltenbrand W.E., Townsend D.W. (1987). Kinetics of Vitamin B-6 Metabolism Examined in Miniture Swine by Continuous Administration of Labelled Pyridoxine. In: Mathematical Models in Experimental Nutrition (Canolty N.L. , Cain T.P., eds.) University of Georgia, Athens, Georgia. 99-111. 46. McElroy L.W., Goss H. (1939). Report on Four Members of the Vitamin B complex Synthesized in the Rumen of Sheep. J Biol Chem 130: 437-438. 47. Coates M.E., Ford J.E., Harrison G.F. (1968). Intestinal Synthesis of Vitamins of the B complex in Chicks. Br J Nutr 22: 493-500. 48. Gregory J.F., Litherland S.A. (1986). Efficacy of the Rat Bioassay for the Determination of Biologically Available Vitamin B-6. J Nutr 116: 87-97. 49. Ikeda M. , Hosotani T., Kurimoto K. , Mori T. , Ueda T. , Kotake Y., Sakakibara B. (1979). The Differences of the Metabolism Related to Vitamin B-6 Dependent Enzymes among Vitamin B-6 Deficient Germfree and Conventional Rats. J Nutr Sci Vitaminol 25: 131-139. 50. Ikeda M. , Hosotani T., Ueda T. , Kotake Y . , Sakakibara B. (1979). Effect of Vitamin B-6 Deficiency on the Levels of Several Water Soluble Vitamins in Tissues of Germfree and Conventional Rats. J Nutr Sci Vitaminol 25: 141-149. 51. Hughes E.H., Squibb R.L. (1942). Vitamin B-6 in the Nutrition of the Pig. J Animal Sci 1: 320-325. 52. Wozenski J.R., Leklem J.E., and Miller L.T. (1980). The Metabolism of Small Doses of Vitamin B-6 in Men. J Nutr 110 (2): 275-285. 53. Schultz T.D., Leklem J.E. (1987). Vitamin B-6 Status and Bioavailability in Vegetarian Women. Am J Clin Nutr 46: 647-51. 54. Sumi, Y., Miyadawa, M. , Kansaki, M. , and Kotake, Y. (1977) . Vitamin B-6 Deficiency in Germ Free Rats. J Nutr 107: 1707-1714. 55. Kabir, H. , Leklem, J.E., and Miller, L.T. (1983). Comparative Vitamin B-6 Bioavailability from Tuna, Whole Wheat Bread and Peanut Butter in Humans. J Nutr 113: 2412-2420.

PAGE 106

97 56. Lindberg A.S. , LeklemJ.E., Miller L.T. (1983). The Effect of Wheat Bran on the Bioavailability of Vitamin B-6 in Young Men. J Nutr 113: 2578-2586. 57. Tarr J.B., Tamura T. , Stokstad L.R. (1981). Availability of Vitamin B-6 and Pantothenate in an Average American Diet in Man. Am J Clin Nutr 34: 1328-1337. 58. Coursin D.B. (1954). Convulsive Seizures in Infants with Pyridoxine Deficient Diet. J Am Med Assoc 154: 406-408. 59. Hassinen J.B., Durbin G.T., Bernhart F.W. (1954). The Vitamin B-6 Content of Milk. J Nutr 53:249-257. 60. Tomarelli R.M. , Spence E.R., Bernhart F.W. (1955). Biological Availability of Vitamin B-6 in Heated Milk. J Agric Food Chem 3: 338-341. 61. Gregory J.F., Kirk J.R. (1978). Assessment of Roasting Effects on Vitamin B-6 Stability and Bioavailability in Dehydrated Food Systems. J Food Sci 42: 1585-1589. 62. Gregory J.F., Kirk J.R. (1978). Assessment of Storage Effects on Vitamin B-6 Bioavailability in Dehydrated Food Systems. J Food Sci 43: 1801-1808. 63. Gregory J.F. (1980). Bioavailability of Vitamin B-6 in Nonfat Dry Milk and a Fortified Rice Breakfast Cereal Product. J Food Sci 45:84-86. 64. Evans S.R., Gregory J.F., Kirk J.R. (1981). Thermal Degradation Kinetics of Pyridoxine Hydrochloride in Dehydrated Model Food Systems. J Food Sci 46: 555558, 563. 65. Kabir H. , Leklem J., Miller L.T. (1983). Measurement of Glycosylated Vitamin B-6 in Foods. J Food Sci 48: 1422-1425. 66. Scudi J.V. (1942). Conjugated Pyridoxine in Rice Bran Concentrates. J Biol Hem 145: 637-639. 67. Yasumoto K. , Tsuji H. , Iwami K. , Mitsuda H. (1977). Isolation from Rice Bran of a Bound Form of Vitamin B-6 and its Identification as 5 1 -O(beta-DGlucopyranosyl) Pyridoxine. Agric Biol Chem 41: 10611067.

PAGE 107

98 68. Tadera K. , Mori E. , Yagi F. , Kobayashi K. , Imada K. , Imabeppu M. (1985) . Isolation and Structure of a Minor Metabolite of Pyridoxine in Seedlings of Pi sum sativum £. J Nutr Sci Vitaminol 31: 403-408. 69. Suzuki Y. , Ishii K. , Suga K. , Uchida K. (1986). Formation of beta-Glucosylpyridoxines in Soybean and Rice Callus. Phytochem 25: 1331-1332. 70. Tadera K. , Makamura M. , Yagi F., Kobayashi A. (1979). A Particulate Glucosyltransf erase Catalyzing the Formation of 5 ' -0(Beta-D-Glucopyranosyl) Pyridoxine from Pyridoxine: The Occurrence in the Seedlings of Pi sum sativum L. J Nutr Sci Vitaminol 25: 76-82. 71. Gregory J.F., Sartain D.B. (1991). Improved Chromatographic Determination of Free and Glycosylated Forms of Vitamin B-6 in Foods. J Agri and Food Chem 39: 899-905. 72. Suzuki Y., Uchida K. , Tsuboi A. (1986). Enzymatic Formation of Pyridoxine Beta-Glucosides by Wheat Bran Beta-Glucosidase. Nippon Nogeikagaku Kaishi 53: 189196. 73. Tadera K. , Yagi F., Dobayashi A. (1982). Specificity of a Particulte Glucosyltransf erase in Seedlings of Pisuro sativum L. which Catalyzes the Formation of 5*-0(Beta-D-Glucopyranosyl) Pyridoxine. J Nutr Sci Vitaminol 28: 359-366. 74. Kabir H. , Leklem J.E., Miller L.T. (1983). Relationship of the Glycosylated Vitamin B-6 Content of Foods to Vitamin B-6 Bioavailability in Humans. Nutr Rept Int 28: 709-715. 75. Gregory J.F, Ink S.L. (1987). Identification and Quantification of Pyridoxine-beta-Glucoside as a Major Form of Vitamin B-6 in Plant Derived Foods. J Agric Food Chem 35: 76-82. 76. Iwami K. , Yasumoto K. (1986). Synthesis of Pyridoxine-beta Glucoside by Rice Bran beta-Glucosidase and its In Situ Absorption in Rat Small Intestine. Nutr Res 6: 407-414. 77. Ink S.L., Gregory J.F., Sartain D.B. (1986). Determination of Pyridoxine beta-Glucoside Bioavailability using Intrinsic and Extrinsic Labeling in the Rat. J Agric Food Chem 34: 857-862.

PAGE 108

99 78. Bills N.D., Leklem J.E., Miller L.T. (1987). Vitamin B-6 Bioavailability in Plant Foods is Inversily Correlated with Percent Glycosylated Vitamin B-6. Fed Proc 46: 1487. 79. Tsuji H., Okada J., Iwami K. , Yasumoto K. , Mitsuda H. (1977). Availability of Vitamin B-6 and Small Intestinal Absorption of Pyridoxine-Beta-D-Glucoside in Rats. Vitamins 51: 153-159. 80. Trumbo P.R., Gregory J.F., Sartain D.B. (1988). Incomplete Utilization of Pyridoxine-beta-Glucoside as Vitamin B-6 in the Rat. J Nutr 118: 170-175. 81. Trumbo P.R., Gregory J.F. (1988). Metabolic Utilization of Pyridoxine-beta-Glucoside in Rats: Influence of Vitamin B-6 Status and Route of Administration. J Nutr 118: 1336-1342. 82. Gregory J.F., Trumbo P.R, Bailey L.B, Toth J.P, Baumgartner T.G., Cerda J.J. (1991). Bioavailability of Pyridoxine-5 1 -0-D-Glucoside Determined in Humans by Stable-Isotopic Methods. J Nutr 121: 177-186. 83. Register U.D., Lewis U.J., Ruegamer W.R., Elvehuem C.A. (1950) . Studies on the Nutritional Adeguacy of Army Combt Rations. J Nutr 40: 281-294. 84. Andon M.B., Reynolds R.D., Moser-Veillon P.B., and Howard M.P. (1989). Dietary Intake of Total and Glycosylated Vitamin B-6 and the Vitamin B-6 Nutritional Status of Unsupplimented Lactating Women and Their Infants. Am J Clin Nutr 50: 1050-1058. 85. Reynolds R.D. (1988). Bioavailability of Vitamin B-6 from Plant Foods. Am J Clin Nutr 48: 863-867. 86. Harding R.S., Plough I.C., Friedemann T.E. (1959). The Effect of Storage on the Vitamin B-6 Content of a Packaged Army Ration, with a Note on the Human Reguirement for the Vitamin. J Nutr 68: 328-331. 87. Sauberlich H.E. (1961). Studies on the Toxicity and Antagonistism of Amino Acid for Weaning Rats. J Nutr 74: 289-297. 88. Davies M.K., Gregory M.E., Henry K.G. (1959). The Effect of Heat on the Vitamin B-6 of Milk. J Dairy Res 26: 215-220.

PAGE 109

100 89. Lushbough C.H., Weichmann J.M., Schweigert B.S. (1959) . Retention of Vitamin B-6 in Meat During Cooking. J Nutr 67: 451-459. 90. Ink S.L., Gregory J.F., Sartain D.B. (1986). Determination of Vitamin B-6 Bioavailability in Animal Tissues Using Intrinsic and Extrinsic Labeling in the Rat. J Agric Food Chem 34: 998-1004. 91. Nguyen L.B., Gregory J.F., Burgin C.W. , Cerda J.J. (1981). In Vitro Binding of Vitamin B-6 by Selected Polysaccharides, Lignin and Wheat Bran. J Food Sci 46: 1860-1862. 92. Miller L.T., Schultz T.D., Leklem J.E. (198010. Influence of Citrus Pectin on the Bioavailability of Vitamin B-6 in Men. Fed Proc 39: 797. 93. Machida Y. , Nagai T. (1980). In Vitro Dissolution Test and Absorption Study of a Per Oral Controlled Release Dosage Form Containing Pyridoxine Hydrochloride or Sodium Riboflavin Phosphate with Hydroxypropyl Cellulose. Chem Pharm Bull 28: 1082-1089. 94. Klosterman H.J. (1974). Vitamin B-6 Antagonists of Natural Origin. J Agic Food Chem 22: 13-16. 95. Bauerfeind J.C, Miller O.N. (1978). Vitamin B-6: Nutritional and Pharmaceutical Usage, Stability, Bioavailability, Antagonists, and Safety. In: Human Vitamin B-6 Requirements. National Academy of Sciences, Washington, DC. 78-110. 96. Klosterman H.J. (1981) . Vitamin B-6 Antagonists in Natural Products. In: Antinurients and Natural Toxicants in Foods, (Ory R.L., ed.), Food and Nutrition Press, Westport CT. 295-317. 97. Bernhart F.W. , D'Amato E. , Tomarelli R.M. (1960). The Vitamin B-6 Activity of Heat Sterilized Milk. Arch Biochem Biophys 88: 267-269. 98. Wendt G. , Bernhart F.W. (1960). The Structure of a Sulfur-Containing Compound with Vitamin B-6 Activity. Arch Biochem Biophys 88: 270-272. 99. Srncova V., Davidek J. (1972). Reaction of Pyridoxal and Pyridoxal-5-Phosphate with Proteins. J Food Sci 37: 310-312.

PAGE 110

101 100. Gregory J.F., Kirk J.R. (1977). Interaction of Pyridoxal and Pyridoxal Phosphate with Peptides in a Liquid Model Food System During Thermal Processing. J Food Sci 42: 1554-1556. 101. Gregory J.F., Kirk J.R. (1978). Vitamin B-6 Activity for Rats of Epsilon-Pyridoxyllysine Bound to Dietary Protein. J Nutr 108: 1192-1199. 102. Gregory J.F., Ink S.L. , Sartain D.B. (1986). Degradation and Binding to Food Proteins of Vitamin B-6 Compounds During Thermal Processing. J Food Sci 51: 1345-1351. 103. Gregory J.F. (1980). Effects of EpsilonPyridoxyllysine Bound to Dietary Protein on the Vitamin B-6 Status of Rats. J Nutr 110: 995-1005. 104. Tadera M. , Arima K. , Yoshino S., Yagi F. , Kobayashi A. (1986) . Conversion of Pyridoxine into 6Hydroxypyridoxine by Food Components, Especially Ascorbic Acid. J Nutr Sci Vitamilnol 32: 267-277. 105. Tadera K. , Arima M. , Yagi F. (1988). Participation of Hydroxyl Radical in Hydroxylation of Pyridoxine by Ascorbic Acid. Agric Biol Chem 52: 2359-2360. 106. Gregory J.F., Leatham K. (1990). Lack of Vitamin B-6 Activity of 6-Hydroxypyridoxine. J Food Sci 55(4): 1143-1146. 107. Tadera K. , Nagano K. , Yagi F., Kobayashi A., Imada K. , Tanaka S. (1988) . Isolation and Structure of a New Metabolite of Pyridoxine in Seedlings of Pisum sativum L. Agric Biol Chem 47: 1357-1359. 108. Tadera K. , Kaneko T. , Yagi F. (1986). Evidence for the Occurrence and Distribution of a New Type of Vitamin B-6 Conjugate in Plant Foods. Agric Biol Chem 50: 2933-2934. 109. Johansson S., Lindstedt S., Tiselius, H.G. (1974). Metabolic Interconversions of Different Forms of Vitamin B-6. J Biol Chem 249: 6040-6046. 110. Colombini C.E., McCoy E.E. (1970). Vitamin B-6 Metabolism: The Utilization of [ 14 C] pyridoxine by the Normal Mouse. Biochem 9: 533-538. 111. Lumeng L. , Brashear R.E., Li T.K. (1974). Pyridoxal 5 1 Phosphate: Source, Protein-binding, and Cellular Transport. J Lab Clin Med 84: 334-343.

PAGE 111

102 112. Merrill A.H., Henderson J.M., Wang E. , McDonald B.W., Millikan, W.J. (1984). Metabolism of Vitamin B-6 by Human Liver. J Nutr 114: 1664-1674. 113. Ebadi M. (1978). Vitamin B-6 and Biogenic Amines in Brain Metabolism. In: Human Vitamin B-6 Requirements. Nat Acad Sci, Washington, D.C. 129-161. 114. Sourkes T.L. (1966). Dopa Decarboxylase: Substrates, Coenzymes, Inhibitors. Pharmacol Rev 18: 53-60. 115. Molinoff P.B, Axlerod J. (1971). Biochemistry of Catecholamines. Ann Rev Biochem 40: 465-500. 116. Merrill A.H., Wang E. (1986). Highly sensitive Methods of assaying the Enzymes of Vitamin B-6 Metabolism. Methods in Enzymol 122: 110-116. 117. Karawya E., Fonda M.L. (1982). Physical and Kinetic Properties of Sheep Liver Pyridoxine Kinase. Arch Biochem and Biophy 216: 170-177. 118. McCormick D.B., Merrill A.H. (1980). Pyridoxamine (Pyridoxine) 5 ' -phosphate Oxidase. In: Vitamin B-6 Metabolism and Role in Growth. (Tryfiates G.P., ed.). Food and Nutr ion Press, Inc., Westport, Conn. 1-26. 119. Kazarinoff M.N., McCormick D.B. (1975). Rabbit Liver Pyridoxamine (Pyridoxine) 5' -phosphate Oxidase: Purification and Properties. J Biol Chem 250: 34363442. 120. McCormick D.B., Snell E.E. (1961). Pyridoxal Phosphokinase: Effects of Inhibitors. J Biol Chem 236: 2085-2088. 121. Gregory J.F. (1980). Effects of e-pyridoxyllysine and Related Compounds on Liver and Brain Pyridoxal Kinase and Liver Pyridoxamine (Pyridoxine) 5 1 -phophate Oxidase. J Biol Chem 255: 2355-2359. 122. McCormick D.B., Kazarinoff M.N., Tsuge H. (1976). FMN-Dependent Pyridoxamine Phosphate Oxidase from Rabbit Liver. In: Flavins and Flavoproteins, Fifth Intern. Symp. (Singer T.P., ed.) Elsvier Publishing Co., Amsterdam. 123. SAS Inst. Inc. (1990). SAS Users Guide: Statistics Version 5 Edition. Cary, NC.

PAGE 112

103 124. Serebro H.A. , Solomon H.M. , Johnson J.H., Hendrix, T.R. (1966) The Intestinal Absorption of Vitamin B-6 Compounds by the Rat and Hamster. Bull Johns Hopkins Hosp 119: 166-171. 125. Lumeng L. , Li T.K. (1974) Vitamin B-6 Metabolism in Chronic Alcohol Abuse: Pyridoxal Phosphate Levels in Plasma and the Effects of Acetaldehyde on Pyridoxal Phosphate Synthesis and Degradation in Human Erythrocytes. J Clin Invest 53: 693-704. 126. Lumeng L. , Li T.K., Lui A. (1985) The Interorgan Transport and Metabolism of Vitamin B-6. In: Vitamin B-6, Its role in Health and Disease. (Reynolds R.D., Leklem J.E., eds.) Alan R. Liss Inc., New York, on Dietary Allowances, Food and Nutrition Board, National Research Council. Natl Acad Sci, Washington, D.C. 3554. 127. Johansson S., Lindstedt S., Tiselius H.G. (1968) Metabolism of [ 3 H 8 ]pyridoxine in Mice. Biochemistry 7: 2327-2332. 128. Contractor S.F., Shane B. (1971) Metabolism of [ 14 C] Pyridoxol in the Pregnant Rat. Biochim Biophys Acta 230: 127-136. 129. Tryfiates G.P., Saus F.L. (1976) Metabolism of Pyridoxine in the Liver of Vitamin B-6-Def icient Rats. Biochim Biophys Acta 451: 333-341. 130. Bain J. A., Williams H.L. (1960) Concentrations of B-6 Vitamers in Tissues and Tissue Fluids. In: Inhibition in the Nervous System and Gamma-aminobutyric Acid, (Roberts E., Baxter C.F., Harreveld A.V. , Wiersma C.A.G., Adey W.R., Killiam K.F., eds.) Pergamon Press, New York. 131. Lyon J.B., Porter J. (1960) The Distribution of Vitamin B-6 in the Tissues of two inbred Strains of Mice fed Complete and Vitamin B-6-Def icient Rations. J Biol Chem 237: 1989-1991. 132. Lumeng L. , Li T.K. (1978) Validation of the Diagnostic Value of Plasma Pyridoxal 5' phosphate Measurements in Vitamin B-6 Nutrition of the Rat. J Nutr 108: 545553.

PAGE 113

104 133. Snell E.E., Haskell B.E. (1971) The Metabolism of Vitamin B-6. In: Comprehensive Biochemistry, Chapter I Section C. (Florkin M. , Stotz E.H., eds.) Vol. 21 Elsevier, Amsterdam. 47-71. 134. Merrill A.H. Jr., Henderson J.M. (1990) Vitamin B-6 Metabolism by Human Liver. In: Vitamin B-6 (Dakshinamurti K. , ed.) Vol. 585 Annals of the New York Academy of Sciences, New York. 110-117. 135. Beaton G.H., Cheney M.C. (1965) Vitamin B-6 Requirements of the Male Albino Rat. J Nutr 87: 125134. 136. Gilbert J. A., Gregory J.F. (1991) The Effect of Pyridoxine-5 1 -/J-D-glucoside on the Metabolic Utilization of Pyridoxine in the Rat. FASEB J 1:A586 (abs. 1250). 137. Sauberlich H.E., Dowdy R.P., Skala J.H. (1974) Vitamin B-6. In: Laboratory Tests for the Assessment of Nutritional Status. CRC Press, Cleveland, OH. 37-49. 138. Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology (1974) Recommended Methods for the Determination of Four Enzymes in Blood. Scand J Clin Lab Invest 33: 291306. 139. Heinegard D. , Tiderstrom G. (1973) Determination of Serum Creatinine by a Direct Color imetric Method. Clin Chim Acta 43: 305-310. 140. Crosby W. , Munn J.I., Furth F.W. (1954) Standardization in a Method for Clinical Hemoglobinometry. U.S. Armed Forces Med J 5: 693-703. 141. Hachey D.L. , Coburn S.P., Brown L.T., Erbelding W.F., DeMark B. , Klein P.D. (1985) Quantitation of Vitamin B-6 in Biological Samples by Isotope Dilution Mass Spectrometry. Anal Biochem 151: 159-168. 142. U.S. Department of Agriculture. (1986) Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Low-Income Women 19-50 Years and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII Report No. 85-2. U.S. Department of Agriculture, Hyattsville, Md.

PAGE 114

105 143. U.S. Department of Agriculture. (1987) Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Low-Income Women 19-50 Years and Their Children 1-5 Years, 1 Day, 1986. NFCS, CSFII Report No. 85-2. U.S. Department of Agriculture, Hyattsville, Md. 144. Snell E.E., Haskell B.E. (1971) The Metabolism of Vitamin B-6 In: Comprehensive Biochemistry. (Florkin M. , Stotz E.H., eds.) Vol. 21 Elsevier/North Holland, Amsterdam. 41-71. 145. Cash CD., Maitre M. , Rumigny J.F., Mandel P. (1980) Rapid Purification by Affinity Chromatography of Rat Brain Pyridoxal Kinase and Pyridoxamine-5-phosphate Oxidase. Biochem Biophys Res Comm 96: 1755-1760. 146. Kwok F., Churchich J.E. (1980) Interaction between Pyridoxal Kinase and Pyridoxine-5-P Oxidase, Two Enzymes Involved in the Metabolism of Vitamin B-6. J Biol Chem 235: 882-887. 147. Daniels L.B., Coyle P.J., Chaio Y-B, Glew R.H., Labow R.S. (1981) Purification and Characterization of a Cytosolic broad Specificity beta-Glucosidase from Human Liver. J Biol Chem 256: 13004-13013. 148. Merill A.H., Horiike K. , McCormick D.B. (1978) Evidence for the Regulation of Pyridoxal 5' -phosphate formation in Liver by Pyridoxamine (pyridoxine) -5 • phosphate Oxidase. Biochem Biophys Res Comm 83: 984990. 149. Perrella F.W. (1988) EZ-FIT: A Practical CurveFitting Microcomputer Program for the Analysis of Enzyme Kinetic Data on IBM-PC Compatible Computers. Anal Biochem 174: 437-447.

PAGE 115

106 Joyce Ann Gilbert was born in Washington, D.C. and raised in Woodbridge, VA. Upon graduation from high school, Joyce attended the University of South Carolina where she received a Bachelor of Science degree in biology. Following a brief respite from academia, Joyce returned to Clemson University where she received a Master in Nutritional Science and became a registered dietitian. After working in clinical and research dietetics as well as in food management, Joyce accepted an instructor position in the Department of Food Science and Human Nutrition at the University of Florida. Joyce relinquished her faculty position to pursue a Ph.D. in human nutrition. Joyce has accepted an assistant professor position at the Pennsylvania State University upon completion of her dissertation. When not engaged in the sciences, Joyce enjoys athletics, both as a participant and fan, art and riding her Harley Davidson.

PAGE 116

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. JfSsse F. Gregory, In, '-ehair Professor of Food Science and Human Nutrition I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Lynri>B. Bailey [_) Professor of Food Science and Human Nutriton I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Peggy R. Borum Professor of Food Science and Human Nutrition I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. gt--c--«-r < . Robert J. Cousins Boston Family Professor of Human Nutrition I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Professor of Medicine

PAGE 117

This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. / December 1991 / X Dean, College of Agriculture Dean, Graduate School


52
These B-6 vitamers could then be taken-up by erythrocytes
and/or tissues. This would account for the observed results
of liver UC and urinary UC in response to PN-glucoside dose
seen in this study. When rats were given a quantity of
unlabeled PN equivalent of 30% of the PN-glucoside dose, the
effect on the metabolic utilization and distribution of other
forms of vitamin B-6 differed significantly from the group
administered the PN-glucoside dose.
In previous studies in this laboratory, the absorption of
PN-glucoside was 50% relative to PN in rats fed a diet
adequate in vitamin B-6 when administered as an alginate gel
(77) but was nearly equivalent to PN when given in solution
(81). Analysis of urine showed that most of the absorbed PN-
glucoside was excreted in the intact form, suggesting low
bioavailability (4,6). Previous findings by Trumbo et al.
(80) showed poor utilization of PN-glucoside relative to PN on
the basis of growth and plasma PLP concentration, which
indicated that PN-glucoside has a low biological availability
as vitamin B-6 in the rat. However, PN derived from
hydrolysis of PN-glucoside, can enter into vitamin B-6
metabolic pathways of the liver to produce other forms of
vitamin B-6 (77,81).
Kabir et al. (65) observed with human subjects, an
inverse relationship between the percentage PN-glucoside (of
total vitamin B-6) and overall vitamiln B-6 bioavailability in
food. The investigators suggested that the proportion of the


CHAPTER 4
EFFECTS OF PYRIDOXINE-5' -j3-D-GLUCOSIDE ON THE
METABOLIC UTILIZATION OF PYRIDOXINE IN HUMANS
Introduction
A conjugated form of vitamin B-6, first isolated from
rice bran was identified as 5'-0-(j8-D-
glucopyranosyl)pyridoxine (PN-glucoside) (67). Analysis of a
variety of plant-derived foods indicated that PN-glucoside is
abundant in many fruits and vegetables and the only
significant glycosylated form of vitamin B-6 present in the
foods examined (65,71,75). Assessing the nutritional status
of a person with respect to vitamin B-6 is a function of the
amount and form of vitamin B-6 in food, the bioavailability,
which is the extent of intestinal absorption and metabolic
utilization of the B-6 compounds in the diet and the specific
requirement of the individual.
Previous studies concerning the bioavailability of
vitamin B-6 in rats indicated that purified PN-glucoside is
relatively well absorbed compared to PN, undergoes little
metabolic utilization, and is rapidly excreted (2,77,80-81).
These results indicated 20%-30% net bioavailability of PN-
glucoside relative to PN. The extent of in vivo hydrolysis of
the glycosidic bond, rather than intestinal absorption, was
55


CHAPTER 1
INTRODUCTION
Vitamin B-6 has often been referred to as the protein
vitamin because of its association with amino acid metabolism
and the influence of dietary intakes of protein on vitamin B-6
requirements (1-3). Adequate vitamin B-6 nutriture is
essential to health through its numerous roles in the body as
the active coenzyme form, pyridoxal 5'-phosphate (PLP). A
large majority of these roles involves metabolism of amino
acids. These functions include, among others, nonoxidative
decarboxylation of amino acids, transamination,
desulfhydration, and enzymes affecting reactions of amino acid
side chains. Such PLP dependent enzymes are important
functional complexes in the biosynthesis and catabolism of
essential and non-essential amino acids and some provide a
connection between the amino acid and intermediates of
carbohydrate metabolism. Recent research in animal models has
suggested a role for PLP in the modulation of hormone actions
(4-5).
Nutritional status with respect to vitamin B-6 is
influenced by the amount of the vitamin ingested, the extent
1


45
Table 3.2 Urinary 4 PA and total urinary UC in rats
administered varying levels of PN-glucoside (Experiment 1)*
0 nmol 36 nmol 72 nmol
PN-glucoside PN-glucoside PN-glucoside SEM p<0.05
% urinary radioactivity
4PA 37.6 12.7 10.9 5.1 +
% of dose
14C 26.9 35.1 37.3 12.3 #
Values based on n=6 are means and pooled standard error
of the mean (SEM) Log transformation of the data was
performed prior to the analysis of variance. The
quadratic effects were not significant for any dependant
variable.
* Significant linear effect, p<0.05.
+ Significant decreased linear effect, p<0.05.
The results of study 1 indicated statistically
significant differences (p<0.05) in the quantity and
distribution of labeled vitamin B-6 compounds in liver tissue
and urine among groups fed varying levels of PN-glucoside
(Tables 3.1 and 3.2). A significant linear relationship was
observed between phosphorylated vitamin B-6 compounds
pyridoxal 5'-phosphate (PLP) and pyridoxine 5'-phosphate (PNP)
in liver tissue and the amount of administered PN-glucoside.
Total hepatic 14C decreased linearly (p<0.05) with varying
amounts of PN-glucoside (Table 3.1). The proportion of
hepatic nonphosphorylated 14C-labeled PN, PL and pyridoxamine
(PM) decreased linearly with the amount of PN-glucoside (Table
3.1) .
Urinary excretion of total 14C was showed a significant
linear effect (p<0.05) among groups receiving varying doses of


32
compounds. Following intestinal absorption, the
dephosphorylated B-6 vitamers diffuse into the liver and are
converted to the coenzyme form of the vitamin, PLP (109-110).
A portion of the PLP is released into the circulation and
constitutes the major source of plasma PLP (111).
The dephosphorylated B-6 vitamers, PL, PN, and PM are
phosphorylated by the same kinase enzyme, pyridoxal kinase (EC
2.7.1.35) (PL kinase). The 51-phosphates of PN and PM are
subsequently oxidized by pyridoxamine (pyridoxine) 5'-
phosphate oxidase (EC 1.4.3.5) (PMP (PNP) oxidase) to yield
PLP. This coenzymatic form is bound to apoproteins,
transported to blood or dephosphorylated by phosphatases back
to PL (112). Further metabolism of PL results in
rephosphorylation by PL kinase or oxidation, by aldehyde
oxidase, to 4PA, the catabolic end product of vitamin B-6
metabolism. Vitamin B-6 in its coenzymatic form PLP or as
PLP-dependent enzymes, functions in many metabolic processes.
These include, amino acid metabolism (3), neurochemical
function (113-115), and modulation of hormone action (4-5).
Therefore, it is important that the cellular levels of PLP be
properly regulated.
As stated above, PL kinase is responsible for the
postabsorptive phosphorylation of PL, PN, and PM with
relatively equal efficiency. The result is the formation of
PLP from PL and upon further oxidation of PN and PM by PMP
(PNP) oxidase these two vitamers are also converted to the


94
24. Tsuji T., Yamada R., Nose Y. (1973). Intestinal
Absorption of Vitamin B-6. I. Pyridoxal Uptake by Rat
Intestinal Tissue. J Nutr Sci Vitaminol 19: 401-417.
25. Yamada R.H., Tsuji T., and Nose Y. (1977). Uptake and
Utilization of Vitamin B-6 and its Phosphate Esters by
Escherichia coli. J Nutr Sci Viltaminol (Tokyo)
23(1):7-17.
26. Middleton H.M. (1977). Uptake of Pyridoxine
Hydrochloride by the Rat Jejunal Mucosa in vitro. J
Nutr 107: 126-131.
27. Middleton H.M. (1979). Intestimal Absorption of
Pyridoxal-5' Phosphate: Disappearance from Perfused
Segments of Rat Jejunum in Vivo. J Nutr 109: 975-981.
28. Hamm M.W., Mehansho H. Henderson L.M. (1979).
Transport and Metabolism of Pyridoxamine and
Pyridoxamine Phosphate in the Small Intestine of the
Rat. J Nutr 109: 1552-1559.
29. Mehansho H., Hamm M.W., Henderson L.M. (1979).
Transport and Metabolism of Pyridoxal and Pyridoxal
Phosphate in the Small Intestine of the Rat. J Nutr
109: 1542-1551.
30. McCormick D.B., Gregory M.E., Snell E.E. (1966).
Pyridoxal Phosphokinases. I. Assay, Distribution,
Purification, and Properties. J Biol Chem 236: 2076-
2084.
31. Wada H., Snell E.E. (1961). The Enzymatic Oxidation
of Pyridoxine and Pyridoxamine Phosphate. J Biol Chem
236: 2089-2095.
32. Pogell B.M. (1958). Enzymatic oxidation of
Pyridoxamine Phosphate to Pyridoxal Phosphate in Rabbit
Liver. J Biol Chem 232: 761-776.
33. Morisue T., Morino Y., Sakamoto Y., Ichihara K.
(1960). Enzymatic Studies on Pyridoxine Metabolism.
III. Pyridoxine Phosphate Oxidase. J Biochem 48: 28-
36.
34. Samra P.S., Snell E.E., Elvehjem C.A. (1947). The
Bioassay of Vitamin B-6 in Natural Materials. J Nutr
33:121-128.


7
however, it also allows for the dephosphorylation of PLP that
is not protein bound and its release from the liver as PL or
oxidized to 4PA (31). Regulation of vitamin B-6 metabolism
may also occur through the conversion of PNP and PMP to PLP,
which is highly sensitive to product inhibition (30).
Recent advances in analytical techniques have enabled
collection of more precise data concerning the quantity and
various forms of vitamin B-6 compounds present in the diet.
However, present knowledge of the bioavailability of vitamin
B-6 is still not sufficient enough to determine an accurate
assessment of adequate dietary intake.
The bioavailability of a nutrient such as vitamin B-6 is
primarily determined experimentally by comparing the
concentration of vitamin B-6 that is biologically active to
total vitamin B-6 ingested (16). Traditionally biologically
available vitamin B-6 has been determined through animal
bioassays (36). Human bioassays, to determine biologically
active vitamin B-6 in food, have been performed, but are
generally lengthy procedures with limited precision.
Significant advances and recent applications in isotopic
methodology offers a useful alternative to bioassay for the
sutdy of bioavailability of vitamin B-6 (36). Results of
studies utilizing rat bioassays to compare animal products
with plant-derived products have indicated that the
bioavailability of vitamin B-6 in animal products is greater
than that of plant-derived products (36). In general, a poor


12
of physical and chemical properties of dietary fiber suggest
the possibility of binding or entrapment of the B-6 vitamers,
which may influence intestinal absorption. Schultz and Leklem
(53) did a study comparing vegetarian versus nonvegetarian
women. They observed that although the vegetarian women
consumed more crude fiber than the nonvegetarian women, there
was no significant differences between these two groups for
plasma PLP, urinary 4PA, and urinary vitamin B-6. Therefore,
it was concluded that there appeared to be no adverse effect
of fiber on the bioavailability of vitamin B-6 between these
groups (53). It has been reported that human intestinal
microflora produce vitamin B-6 (54) Several bioavailability
studies have indicated that ingestion of diets high in dietary
fiber or carbohydrate leads to significant increases in
microbial synthesis of vitamin B-6 (55-56).
Tarr et al. (57) evaluated a typical American mixed diet
to its bioavailability of vitamin B-6. They reported 70%
bioavailability of vitamin B-6 based on urinary 4PA and plasma
PLP concentration relative to PN in a formula diet. The
authors speculated that thermal processing was potentially
responsible for the incomplete bioavailability of vitamin B-6
because canned goods, including both animal and plant foods,
composed much of the mixed diet. A study by Nguyen and
Gregory (36) employed rat bioassays to examine the
bioavailability of vitamin B-6 in selected foods as influenced
by thermal processing. The effects of food composition and


increasing dose of PN-glucoside, while hepatic 14C decreased
significantly as the PN-glucoside dose increased. The
proportion of hepatic 14C-labeled pyridoxal, PN, and
pyridoxamine decreased whereas hepatic pyridoxine phosphate
and pyridoxal phosphate increased in proportion to the PN-
glucoside dose. In addition, the concentration of urinary
[14C] 4-pyridoxic acid (4PA) relative to total urinary 14C,
decreased as the dose of PN-glucoside increased. Stable
isotopic methodology was employed to determine whether PN-
glucoside affected the metabolic utilization of simultaneously
administered deuterium-labeled PN (d2PN) in humans.
Experimental results showed that twenty-four hour urinary
excretion of 4PA was decreased significantly with increasing
dose of PN-glucoside. The percentage of ingested d2PN
excreted as d24PA showed an inverse relationship, that was
statistically significant, in proportion to the PN-glucoside
dose. In vitro enzyme assays indicated that PN-glucoside had
no significant effect on the activity of partially purified
pyridoxal-kinase and pyridoxamine (pyridoxine) phosphate
oxidase. These results provide evidence that PN-glucoside
weakly retards the metabolic utilization of nonglycosylated
forms of vitamin B-6. However, the effect of PN-glucoside on
PN is not due to the direct effect of PN-glucoside on the
enzymes PL kinase and PMP (PNP) oxidase.
ix


100
89. Lushbough C.H., Weichmann J.M., Schweigert B.S.
(1959). Retention of Vitamin B-6 in Meat During
Cooking. J Nutr 67: 451-459.
90. Ink S.L., Gregory J.F., Sartain D.B. (1986).
Determination of Vitamin B-6 Bioavailability in Animal
Tissues Using Intrinsic and Extrinsic Labeling in the
Rat. J Agrie Food Chem 34: 998-1004.
91. Nguyen L.B., Gregory J.F., Burgin C.W., Cerda J.J.
(1981). In Vitro Binding of Vitamin B-6 by Selected
Polysaccharides, Lignin and Wheat Bran. J Food Sci
46: 1860-1862.
92. Miller L.T., Schultz T.D., Leklem J.E. (198010.
Influence of Citrus Pectin on the Bioavailability of
Vitamin B-6 in Men. Fed Proc 39: 797.
93. Machida Y., Nagai T. (1980). In Vitro Dissolution
Test and Absorption Study of a Per Oral Controlled
Release Dosage Form Containing Pyridoxine Hydrochloride
or Sodium Riboflavin Phosphate with Hydroxypropyl
Cellulose. Chem Pharm Bull 28: 1082-1089.
94. Klosterman H.J. (1974). Vitamin B-6 Antagonists of
Natural Origin. J Agic Food Chem 22: 13-16.
95. Bauerfeind J.C, Miller O.N. (1978). Vitamin B-6:
Nutritional and Pharmaceutical Usage, Stability,
Bioavailability, Antagonists, and Safety. In: Human
Vitamin B-6 Requirements. National Academy of
Sciences, Washington, DC. 78-110.
96. Klosterman H.J. (1981). Vitamin B-6 Antagonists in
Natural Products. In: Antinurients and Natural
Toxicants in Foods, (Ory R.L., ed.), Food and
Nutrition Press, Westport CT. 295-317.
97. Bernhart F.W., D'Amato E., Tomarelli R.M. (1960). The
Vitamin B-6 Activity of Heat Sterilized Milk. Arch
Biochem Biophys 88: 267-269.
98. Wendt G., Bernhart F.W. (1960). The Structure of a
Sulfur-Containing Compound with Vitamin B-6 Activity.
Arch Biochem Biophys 88: 270-272.
Srncova V., Davidek J. (1972). Reaction of Pyridoxal
and Pyridoxal-5-Phosphate with Proteins. J Food Sci
37: 310-312.
99.


101
100. Gregory J.F., Kirk J.R. (1977). Interaction of
Pyridoxal and Pyridoxal Phosphate with Peptides in a
Liquid Model Food System During Thermal Processing. J
Food Sci 42: 1554-1556.
101. Gregory J.F., Kirk J.R. (1978). Vitamin B-6 Activity
for Rats of Epsilon-Pyridoxyllysine Bound to Dietary
Protein. J Nutr 108: 1192-1199.
102. Gregory J.F., Ink S.L., Sartain D.B. (1986).
Degradation and Binding to Food Proteins of Vitamin B-6
Compounds During Thermal Processing. J Food Sci 51:
1345-1351.
103. Gregory J.F. (1980). Effects of Epsilon-
Pyridoxyllysine Bound to Dietary Protein on the Vitamin
B-6 Status of Rats. J Nutr 110: 995-1005.
104. Tadera M., Arima K., Yoshino S., Yagi F., Kobayashi A.
(1986). Conversion of Pyridoxine into 6-
Hydroxypyridoxine by Food Components, Especially
Ascorbic Acid. J Nutr Sci Vitamilnol 32: 267-277.
105. Tadera K., Arima M., Yagi F. (1988). Participation of
Hydroxyl Radical in Hydroxylation of Pyridoxine by
Ascorbic Acid. Agrie Biol Chem 52: 2359-2360.
106. Gregory J.F., Leatham K. (1990). Lack of Vitamin B-6
Activity of 6-Hydroxypyridoxine. J Food Sci 55(4):
1143-1146.
107. Tadera K., Nagano K., Yagi F., Kobayashi A., Imada K.,
Tanaka S. (1988). Isolation and Structure of a New
Metabolite of Pyridoxine in Seedlings of Pisum sativum
L. Agrie Biol Chem 47: 1357-1359.
108. Tadera K., Kaneko T., Yagi F. (1986). Evidence for
the Occurrence and Distribution of a New Type of
Vitamin B-6 Conjugate in Plant Foods. Agrie Biol Chem
50: 2933-2934.
109. Johansson S., Lindstedt S., Tiselius, H.G. (1974).
Metabolic Interconversions of Different Forms of
Vitamin B-6. J Biol Chem 249: 6040-6046.
110. Colombini C.E., McCoy E.E. (1970). Vitamin B-6
Metabolism: The Utilization of [14C]pyridoxine by the
Normal Mouse. Biochem 9: 533-538.
111. Lumeng L., Brashear R.E., Li T.K. (1974). Pyridoxal
5' Phosphate: Source, Protein-binding, and Cellular
Transport. J Lab Clin Med 84: 334-343.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EGD5YVOYC_XMRVND INGEST_TIME 2015-02-19T21:42:00Z PACKAGE AA00028760_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


5
To quantitate total vitamin B-6 in most food, test
microorganisms are generally utilized. The most common
organism used is Saccharomvces uvarum. Test microorganisms
demonstrate low biological activity of vitamin B-6 in its
bound (pyridoxyl-amino acid and amines) glycosylated and
phosphorylated forms until they are released by acid
hydrolysis. However, this acid treatment of food is not
entirely representative of the digestive process in the
gastrointestinal tract.
Results of early studies of intestinal absorption of
vitamin B-6 indicated that absorption occurs by a simple
passive diffusion mechanism (19-22) A number of studies
indicate that the rate of uptake by intestinal tissue of the
B-6 vitamers, PN, PM, and PL increased proportionally to the
dose over a wide range of intralumenal concentrations (19,23-
31). Transport of PN across the intestinal wall occurred by
diffusion, independent of accumulation within the tissue.
There is no evidence of a saturable transport system either in
vivo (19,23-24) or in vitro (23,26-31). These data also
indicate that transport across the intestinal wall from the
mucosal to the serosal side of the non-phosphorylated B-6
vitamers to be quantitatively similar. PN in the free and
conjugated forms is rapidly absorbed by the intestine directly
or following hydrolysis by intestinal enzymes. PL and PM are
present in food primarily as the 5'-phosphates and require
hydrolysis by lumenal phophatases. The phosphorylated B-6


This work is dedicated to my parents, Helen and Reginald,
who showed me life has no boundaries, only opportunities.


59
Analytical methods
Urinary 4PA and plasma PLP were determined by reverse-
phase HPLC procedures with fluorometric detection (48,109).
Erythrocyte AspAT activity was measured by a
spectrophotometric assay procedure (138) with commercially
available reagents (Sigma Chemical, St. Louis, MO). Urinary
creatinine was determined based on the method of Heinegard and
Tiderstrom (139), while hemoglobin in erythrocyte hemolysates
was determined spectrophotometrically as the
cyanomethemoglobin derivative (140).
Mass spectral analysis of deuterium-labeled 4PA
Urinary d24PA was determined by gas chromatography-mass
spectrometry (GCMS) following isolation of 4PA from urine
samples by cation exchange chromatography (Bio-Rad AG 50W-X8,
100-200 mesh, H+ form) and reverse-phase HPLC (Whatman,
Partisil 10 ODS-3 Magnum 9 column, 9mm i.d. x 25 cm) as
described by Gregory et al.(82). GCMS was performed in the
electron capture negative ionization mode (Model 4500 GCMS
system, Finnigan MAT, San Jose, CA) with a DB-5 capillary gas
chromatographic column (J&W Scientific, Folsom, CA) and
methane as a reagent gas. The derivatization (141) was
performed by dissolving the dried 4PA sample in 0.5 ml of 1:1
(v/v) solution of pyridine:acetic anhydride, heated at 100 C
for 90 min to form the 3-acetyl-4PA-lactone which is
evaporated to dryness under a stream of nitrogen. The 3-


68
partially retarding the utilization of nonglycosylated B-6
compounds.
The appropriateness of the RDA is difficult, if not
impossible to assess without accurate information concerning
the content and forms of B-6 compounds in foods consumed, and
accurate data regarding vitamin B-6 bioavailability. This
research provides evidence of interactions amoung various B-6
compounds found in a typical mixed diet through an influence
of PN-glucoside on the utilization of PN. Such an interaction
represents another factor involved in the bioavailability of
vitamin B-6 and must be considered in fully understanding the
adequacy of vitamin B-6 intake. Future research will address
the nutritional significances of these interactions amoung the
vitamin B-6 compounds and the importance of these compounds in
assessing the nutritional status of the population for this
vitamin.


78
Table 5.1 Purification of pyridoxal kinase from rat liver.
Fraction
Purification
Vol
(ml)
Total
Units
Protein
(mg/ml)
Units/mg
protein
(fold)
Supernatant
80
509
0.25
2.54
PL-Sepharose
6
188
0.05
628
247
/Hydroxyapatite
Units are nmol PNP/min/ml at 37C in phosphate buffer pH
5.75.
50 g rat liver was used.
phosphorylation was constant over at least a 60 min period
under the assay conditions outlined above. The rate of
phosphorylation of PN in this study indicated a of 21 /xM
which is in general agreement with the previously reported
value of 25 /xM measured under similar conditions (9) In the
presence of varying concentrations of PN-gluoside the value
at all levels was not significantly different from that
observed with 0 /xM PN-glucoside (Table 5.2) The Vmax was 16.8
nmol of PNP/h/mg of protein and did not differ significantly
over the range of PN-glucoside concentrations (Table 5.2).
The kinetic parameters (Figure 5.1) were calculated from
analysis using the EZ-FIT Program (149).


77
verses a reagent blank. The activity was expressed in nmol
PLP/mL enzyme/h.
Statistical analysis
The kinetic parameters Km and V||)ax were calculated by non
linear regression using the EZ-FIT program (149) Confidence
intervals for each and V(nax were calculated by the equation:
b, t 975 X SE, where b1 = or V|nax values (123). Overlapping
confidence intervals are not significantly different at
p<0.05.
Results
These studies were conducted to determine the effects of
PN-glucoside on two enzymes, PL kinase and PMP (PNP) oxidase,
in the metabolic pathway responsible for the interconversion
of vitamin B-6 compounds. These catalytic enzymes were
investigated, in vitro, by evaluating the rates of reaction in
the presence and absence of PN-glucoside. The purification
methods used for rat liver PL kinase and PMP (PNP) oxidase
yielded an enzyme preparation acceptable for kinetic analysis.
PL kinase
Partial purification of the enzyme by the above technique
resulted in a 243-fold increase in specific activity relative
to the crude homogenate (Table 5.1). Preliminary experiments
indicated that the rate of formation of [14C]PNP was linear
over varying enzyme concentrations and the rate of


CHAPTER 6
SUMMARY AND CONCLUSIONS
Since the isolation and identification of PN-glucoside
from rice bran (67) much focus has been on the comparative
bioavailability of this glycosylated form and other vitamin B-
6 compounds. A number of studies spanning the last decade has
provided a great deal of information regarding the biological
availability of vitamin B-6 relative to its multiple forms.
As discussed in previous chapters, between 5%-80% of the
vitamin B-6 found in a variety of plant derived foods was in
the glycosylated form and both food processing and diet
composition are important considerations when evaluating the
bioavailability of vitamin B-6. Also the net bioavailability
of PN-glucoside relative to PN is less than 40% in rats
(75,77,90). Enzymatic hydroxylation of PN does not exhibit
antivitamin B-6 activity, however other degradative products
such as pyridoxallysine have been shown to be vitamin B-6
analogues and act as competitive inhibitors to enzymes in the
vitamin B-6 metabolic pathway.
The use of stable iotopic methodology showed PN-glucoside
to be 58% as biologically active as PN in humans (41). This
study indicated that although PN-glucoside bioavailability was
86


102
112. Merrill A.H., Henderson J.M., Wang E., McDonald B.W.,
Millikan, W.J. (1984). Metabolism of Vitamin B-6 by
Human Liver. J Nutr 114: 1664-1674.
113. Ebadi M. (1978). Vitamin B-6 and Biogenic Amines in
Brain Metabolism. In: Human Vitamin B-6 Requirements.
Nat Acad Sci, Washington, D.C. 129-161.
114. Sourkes T.L. (1966). Dopa Decarboxylase: Substrates,
Coenzymes, Inhibitors. Pharmacol Rev 18: 53-60.
115. Molinoff P.B, Axlerod J. (1971). Biochemistry of
Catecholamines. Ann Rev Biochem 40: 465-500.
116. Merrill A.H., Wang E. (1986). Highly sensitive
Methods of assaying the Enzymes of Vitamin B-6
Metabolism. Methods in Enzymol 122: 110-116.
117. Karawya E., Fonda M.L. (1982). Physical and Kinetic
Properties of Sheep Liver Pyridoxine Kinase. Arch
Biochem and Biophy 216: 170-177.
118. McCormick D.B., Merrill A.H. (1980). Pyridoxamine
(Pyridoxine)5'-phosphate Oxidase. In: Vitamin B-6
Metabolism and Role in Growth. (Tryfiates G.P., ed.).
Food and Nutrion Press, Inc., Westport, Conn. 1-26.
119. Kazarinoff M.N., McCormick D.B. (1975). Rabbit Liver
Pyridoxamine(Pyridoxine)5'-phosphate Oxidase:
Purification and Properties. J Biol Chem 250: 3436-
3442.
120. McCormick D.B., Snell E.E. (1961). Pyridoxal
Phosphokinase: Effects of Inhibitors. J Biol Chem
236: 2085-2088.
121. Gregory J.F. (1980). Effects of e-pyridoxyllysine and
Related Compounds on Liver and Brain Pyridoxal Kinase
and Liver Pyridoxamine(Pyridoxine)5'-phophate Oxidase.
J Biol Chem 255: 2355-2359.
122. McCormick D.B., Kazarinoff M.N., Tsuge H. (1976).
FMN-Dependent Pyridoxamine Phosphate Oxidase from
Rabbit Liver. In: Flavins and Flavoproteins, Fifth
Intern. Symp. (Singer T.P., ed.) Elsvier Publishing
Co., Amsterdam.
123. SAS Inst. Inc. (1990). SAS Users Guide: Statistics
Version 5 Edition. Cary, NC.


90
designed to determine the in vitro effects of PN-glucoside on
PL kinase and PMP (PNP) oxidase. The results indicated no
direct effect of PN-glucoside on either enzyme as determined
by no significant differences observed between the kinetic
parameters, and V|nax with varying concentrations of the
glycosylated vitamin B-6 compound. Further studies are needed
to assess potential effects of PN-glucoside on other enzymes
in the interconversion pathway of vitamin B-6 metabolism.
These experimenal results provide evidence that PN-
glucoside exhibits a quantitative effect on the metabolic
utilization of PN. Although the biochemical mechanism
responsible for this effect are presently uncertain, it is
evident that consideration in the study of vitamin B-6
bioavailability must incorporate the potential interactions
among B-6 compounds present in foods generally consumed in a
mixed diet.
The information from these studies is consequential in
relation to current USDA Dietary Guidelines, which are used to
convey recommendations to the general population. Present
nutrition education is promoting the consumption of greater
quantities of plant derived foods. Food consumption data have
shown that there are several sub-populations at nutritional
risk for vitamin B-6, specifically pregnant females,
adolescents, and the elderly.those individuals undergoing
growth or aging.


TABLE OF CONTENTS
ACKNOWLEDGEMENTS iii
LIST OF TABLES vi
LIST OF FIGURES vii
Abstract viii
CHAPTER 1
INTRODUCTION 1
CHAPTER 2
LITERATURE REVIEW 4
Bioavailability of Vitamin B-6 4
Pyridoxine-/3-D-Glucoside 15
Factors Affecting the Bioavailability of Vitamin
B-6 21
Enzymatic Interconversion of B-6 Vitamers .... 31
CHAPTER 3
PYRIDOXINE-5 -/3-D-GLUCOSIDE AFFECTS THE METABOLIC
UTILIZATION OF PYRIDOXINE IN RATS 38
Introduction 38
Materials and Methods 39
Protocol 39
Forms of vitamin B-6 40
Sample preparation 41
HPLC equipment and analysis 42
Measurement of radioactivity 43
Statistical analysis 43
Results 44
Discussion 48
CHAPTER 4
EFFECTS OF PYRIDOXINE-5 '-/3-D-GLUCOSIDE ON THE
METABOLIC UTILIZATION OF PYRIDOXINE IN HUMANS .... 55
Introduction 55
Materials and Methods 56
Synthesis of forms of vitamin B-6 56
Protocols of trials with human subjects ... 57
Analytical methods 59
iv


18
indicated that the proportion of the total vitamin B-6 which
was present as PN-glucoside correlated closely with the
bioavailability of the vitamin in a variety of foods examined,
including tuna, peanut butter and whole wheat bread. These
results suggested that PN-glucoside would not be available to
humans for intestinal absorption and metabolic utilization.
However, further studies by Bills et al. (78) indicated that
this observed correlation was inconsistent when additional
foods were examined. As discussed earlier, Nelson et al. (38)
reported that the vitamin B-6 from orange juice, a large
portion of which is in the conjugated form, exhibited a slower
rate of absorption during intestinal perfusion of humans than
PN from control solutions. Tsuji et al. (79) observed that
synthetic PN-glucoside exhibited vitamin B-6 activity which
was approximately equivalent to that of PN in a rat bioassay.
In contrast, Trumbo et al. (80) utilizing a rat bioassay
reported that PN-glucoside isolated from alfalfa sprouts
exhibited only 10% to 30% bioavailability relative to PN in
the rat. Studies in rats using radiolabeled PN-glucoside
indicate effective absorption of this B-6 compound and that
the portion of PN-glucoside not hydrolyzed to metabolically
active PN, is rapidly excreted in the urine in the intact form
(77,81). Ink et al. (77) reported the extent of absorption of
radiolabeled PN-glucoside from diets containing intrinsically
enriched alfalfa sprouts was similar, but quantitatively less


CHAPTER 2
LITERATURE REVIEW
Bioavailabilitv of Vitamin B-6
The concept of bioavailability is most appropriate when
considering the extent of intestinal absorption and metabolic
utilization and any potential antagonistic effects of all
naturally occurring dietary forms of the vitamin. Hence, net
bioavailability is defined as the portion of total dietary
vitamin B-6 that is biologically active following intestinal
absorption (16). Vitamin B-6 as it occurs in mammalian
tissues and fluids encompasses three interconvertible 3-
hydroxy-2-methylpyridine compounds including pyridoxine (PN) ,
pyridoxal (PL), pyridoxamine (PM), their corresponding 5'-
phosphoryl derivatives (PNP, PLP, and PMP respectively) and
the excretory catabolite 4-pyridoxic acid (4PA) (17). Vitamin
B-6 in nature is found in phosphorylated forms and closely
associated with proteins. In foods, PL, PN, and PM are found
as both phosphorylated and nonphosphorylated forms. PL, PM,
PLP, and PMP are the predominant forms in animal products,
while PN, in its free vitaminic form and as the glycoside
(discussed in a later section) are the major forms in most
plant foods (18).
4


44
Table 3.1 Liver B-6 vitamer distribution and total liver 14C
in rats administered varying levels of PN-glucoside in the
dose (Experiment 1).*
LEVEL OF PN-GLUCOSIDE IN THE DOSE
Vitamers
0 nmol
36 nmol 72 nmol
SEM
p<0
PLP
17.5
% liver
37.3
radioactivitv
51.6 7.5
#
PNP
0.0
3.1
16.4
3.4
#
PMP
46.7
36.1
32.0
8.0
#
PL
16.9
14.5
0.0
4.8
+
PN
15.2
7.8
0.0
3.5
+
PM
3.6
1.2
0.0
0.6
+
Total UC
20.0
%
17.0
of dose
12.3
4.0
+
Values based on n=6 are means and pooled standard error
of the mean (SEM). Log transformations were performed on
data prior to analysis of variance. The quadratic effects
were not significant for any dependant variable.
# Significant linear effect, p<0.05.
+ Significant decreased linear effect, p<0.05.
Results
These experiments were designed to investigate the effect
of unlabeled PN-glucoside on the metabolism, in vivo
retention, and excretion processes of simultaneously
administered [UC]PN. The distribution of B-6 vitamers in
liver tissue, total urinary [UC]4PA, and total urinary UC
were determined to evaluate the utilization of [UC]PN in the
presence of PN-glucoside. Data were expressed as a percentage
of total radioactivity administered in the dose or as a
percentage of total [UC] B-6 in the tissue.


Figure 5.1 Double reciprocal plot for the phosphorylation of PN by PL kinase from rat
liver in the presence of various levels of PN-glucoside.


105
143. U.S. Department of Agriculture. (1987) Nationwide
Food Consumption Survey, Continuing Survey of Food
Intakes by Individuals, Low-Income Women 19-50 Years
and Their Children 1-5 Years, 1 Day, 1986. NFCS, CSFII
Report No. 85-2. U.S. Department of Agriculture,
Hyattsville, Md.
144. Snell E.E., Haskell B.E. (1971) The Metabolism of
Vitamin B-6 In: Comprehensive Biochemistry. (Florkin
M., Stotz E.H., eds.) Vol. 21 Elsevier/North Holland,
Amsterdam. 41-71.
145. Cash C.D., Maitre M., Rumigny J.F., Mandel P. (1980)
Rapid Purification by Affinity Chromatography of Rat
Brain Pyridoxal Kinase and Pyridoxamine-5-phosphate
Oxidase. Biochem Biophys Res Comm 96: 1755-1760.
146. Kwok F., Churchich J.E. (1980) Interaction between
Pyridoxal Kinase and Pyridoxine-5-P Oxidase, Two
Enzymes Involved in the Metabolism of Vitamin B-6. J
Biol Chem 235: 882-887.
147. Daniels L.B., Coyle P.J., Chaio Y-B, Glew R.H., Labow
R.S. (1981) Purification and Characterization of a
Cytosolic broad Specificity beta-Glucosidase from Human
Liver. J Biol Chem 256: 13004-13013.
148. Merill A.H., Horiike K., McCormick D.B. (1978)
Evidence for the Regulation of Pyridoxal 5'-phosphate
formation in Liver by Pyridoxamine(pyridoxine)-5'-
phosphate Oxidase. Biochem Biophys Res Comm 83: 984-
990.
149. Perrella F.W. (1988) EZ-FIT: A Practical Curve-
Fitting Microcomputer Program for the Analysis of
Enzyme Kinetic Data on IBM-PC Compatible Computers.
Anal Biochem 174: 437-447.


8
correlation was observed in these samples when comparing rat
growth and plasma PLP concentration. These results were
difficult to interpret because of the potential effect of diet
composition on the synthesis of vitamin B-6 by intestinal
microflora. The lack of agreement between rat growth and
plasma PLP, suggests that data from bioassays to determine
biologically available vitamin B-6 in food sources,
specifically those in which the test diets may differ greatly
from the reference diet in the type of carbohydrate, are
somewhat equivocal (37). The bioavailability of vitamin B-6
has been studied in somewhat more detail than many of the
other vitamins. However, few generalizations can be stated
regarding the overall bioavailability of the B-6 vitamers or
the factors influencing it.
Essential to measuring the bioavailability of vitamin B-6
in a food source is the determination of the total vitamin B-6
content in the food. This is complicated by the interaction
of the vitamin with food components. These interactive
compounds may or may not be biologically available. The
interactions of vitamin B-6 with food components and with
other vitamin B-6 compounds are important considerations since
they may represent sources of available vitamin B-6 in the
diet. Experimental data regarding the bioavailability of
naturally occurring vitamin B-6 in foods were initially
reported by Sarma et al. (34). The authors utilized a
comparison of rat bioassay and Saccharomvces uvarum assay


41
(Nutley, NJ) with a purity of greater than 98%, as determined
by ion-pair reverse phase HPLC.
Sample preparation
Urine samples were deproteinated by centrifugal
ultrafiltration with micropartition tubes (YMT membrane
filters; Amicon, Danvers, MA) (80) Urinary 4PA analysis was
by reverse-phase HPLC, as described below. Fractions (0.5 mL)
were collected by using an ISCO Cygnet Fraction Collector
(ISCO, Lincoln, NB) Each filtered sample was decolorized
(81) and an aliquot (100 /iL) was then counted for total
radioactivity. Liver tissue was minced and homogenized in 6
mL of 4.3 mol/L trichloroacetic acid with a Polytron
Homogenizer (Brinkman Instruments, Westburg, NY) and then
centrifuged for 20 min. at 12,000 x g (6). The supernatant
was partitioned against an equal volume of diethyl ether to
extract the trichloroacetic acid and filtered with 0.45 un
membrane filters (Gelman Sciences, Inc., Ann Arbor, MI). The
sample was aerated with nitrogen gas, then analyzed for
vitamin B-6 by HPLC as described below, and an aliquot of the
supernatant from each sample was analyzed for radioactivity.
Decolorization of the extract was performed as above prior to
liquid scintillation spectrometry.


54
conceivable that the vitamin B-6 status in humans would not be
accurately reflected by current food consumption data.
Studies evaluating the metabolic interactions of PN-glucoside
on PN in humans are currently in progess.


16
been observed as well as evidence supporting the widespread
existence of /3-glucoside conjugates of B-6 vitamers in plant
tissues (68). PN-glucoside appears to be formed in plants by
enzymatic transglycosylation (70-73). The glucose moiety of
PN-glucoside occurs at the 5' and 4' positions of the PN ring;
however, the presence of the 5'-isomer as the primary
naturally occurring form of PN-glucoside suggests specificity
for this position by the transglycosylation reaction in most
plant tissue (67). These studies provide evidence that
conjugates of vitamin B-6 exist as a substantial proportion of
the total vitamin B-6 in plant-derived foods (66-69). The
results of Kabir et. al (65) indicated the existance of /3-
glucopyranosyl conjugates in a wide variety of plant products.
The importance of PN-glucoside as a bioavailable source
of vitamin B-6 in humans was first studied by Kabir et al.
(74). The investigators determined the urinary and fecal
excretion patterns of glycosylated vitamin B-6 when different
foods that contained the naturally occurring PN-glucoside were
fed to human subjects. It was observed that an inverse
relationship existed between plant derived PN-glucoside (as
percent of total B-6) and the net B-6 bioavailability in
humans. The investigators suggested that the proportion of
the glucosylated vitamin B-6 in the diet may be useful as an
index of vitamin B-6 bioavailability.
Kabir et al. (65) developed a microbiological assay
procedure for quantitating glycosidic conjugated forms of PN


11
to the host organism, especially in the absence of coprophagy,
has been unclear (45,47). Since several animals can develop
a vitamin B-6 deficiency even when coprophagy is allowed,
metabolic utilization of microbial vitamin B-6 must be miminal
(48). In animals exposed to conditions of nutritional
deficiency, coprophagy may increase sufficiently to become a
significant source of nutrients. This may explain the results
of Ikeda et al. (49-50), who observed that germfree rats tend
to be more susceptible to vitamin B-6 deficiency than rats
with normal populations of intestinal microorganisms. Hughes
et al. (51) reported that in addition to the intestinal
microflora, nutritionally deprived animals may also benefit
from other environmental sources of microbial vitamins such as
microbial activity on the floor of the pen. As indicated by
these studies, there exists no compelling evidence that
vitamin B-6 can be absorbed from the large intestine in
amounts sufficient to make a detectable contribution to the
daily intake.
A study (52) involving human bioassay in measuring
bioavailability of vitamin B-6 in beef compared with soybeans
suggested the vitamin B-6 in soybeans to be less available
compared to that contained in beef. It has been suggested that
the presence of nondigestible polysaccharides and lignin
components of dietary fiber of the plant derived foods is
responsible for differences in the availability of vitamin B-6
in plant foods compared with animal products (40). A variety


33
coenzymatic form (109-110) The PLP that is not released into
circulation is either bound to cellular proteins or
dephosphorylated to PL. The vitamin B-6 taken up by the liver
is predominately maintained in the phosphorylated form (109-
110) .
There exists a cycle of phosphorylation/dephosphorylation
between PL and PLP which is catalyzed by PL kinase and
phosphatase enzymes (109). Under physiological conditions PL
kinase has approximately a 10 fold greater activity than
phosphatases (112). The degradative enzyme aldehyde oxidase
has an activity greater than the kinase and effectively
competes for PL. In order to maintain a steady state of
intracellular PLP a form of metabolic trapping occurs. The
coenzyme is bound to protein which protects it from hydrolysis
and hence, determines the relative amounts of PLP and PL and
creates a shift of equillibrum toward a balanced state (112).
Considering that PLP is required as a conenzyme for a
number of enzymes, it is not surprising that PL kinase is
widely distributed in tissues. This enzyme has been detected
in practically all tissues tested, including liver, kidney,
brain, and muscle (30). PL kinase has also been detected in
erythrocytes, however, the enzyme differs in its properties
compared with other tissue PL kinase (116). Since the liver
is the major site of vitamin B-6 metabolism it is generally
the tissue selected for the study of the enzymes in the B-6
vitamer interconversion pathway (116). It is important to


70
can then itself be released by the cell, oxidized to 4PA, the
metabolic end product of B-6 catabolism, or rephosphorylated
to form PLP.
Previously discussed studies in chapters 3 and 4 provided
evidence that PN-glucoside quantitatively alters the
metabolism and retention of administered PN (136). These
studies indicated that unlabeled PN-glucoside affects the
metabolic utilization of a simultaneous dose of labeled PN
(136). These data showed a linear relationship between the
dose of PN-glucoside and hepatic radiolabeled PNP and PLP
(136) Hepatic UC PN, PL, and PM were inversely proportional
to the dose of PN-glucoside.
The data from these studies led us to postulate that the
effect on vitamin B-6 metabolism exhibited by PN-glucoside may
be caused by a direct affect of the glycosylated vitamer on
the action of either PL kinase, PMP (PNP) oxidase, or both.
The purpose of this study was to determine the effect of PN-
glucoside on the in vitro catabolic activity of PL kinase and
PMP (PNP) oxidase partially from purified rat liver.
Materials and Methods
Protocol
Two studies were conducted to evaluate the effect of PN-
glucoside on the reaction rates of the enzymes PL kinase and
PMP (PNP) oxidase. In each of the studies six male Sprague
Dawley rats (Crl:CD(SD)BR) from Charles River Breeding


6
vitamers, PLP and PMP are well absorbed following enzymatic
dephosphorylation, although these vitamers are slowly absorbed
as the intact 5'-phosphate esters (28-30). Studies have
indicated that the transport of the B-6 vitamers into the
intestinal mucosal cell involves a saturable enzymatic
process, identified as intracellular phosphorylation by PL
kinase (EC 2.7.1.35) (24, 28-29). Most ingested vitamin B-6
is absorbed by the jejunum in the nonphosphorylated forms, PL,
PN, and PM. Tissues, especially liver, rapidly take up
circulating vitamin B-6, where the phophorylated B-6 vitamers
are hydrolyzed by plasma membrane phosphatases and enter the
cells by a facilitated process and diffussion followed by
metabolic trapping (28-29). PN is phosphorylated to PNP by PL
kinase which has been detected in all mammalian tissues
investigated (32). PNP is then converted to PLP by flavin-
dependent PNP oxidase (EC 1.4.3.5) (33) which, in contrast to
the widely distributed PL kinase, is found in few tissues,
mainly liver, erythrocytes, kidney, and brain (34-35). PLP
can be transformed to PMP by transamination or hydrolyzed to
PL by phosphatases (17). PL is then converted to pyridoxic
acid (4PA) by aldehyde oxidase (18) or again to PLP by PL
kinase. PLP is bound by cellular protein or released into
plasma, by the cell, as PLP, PL or 4PA. The analysis of
kinase and phosphatase enzymes in liver tissue indicates
similar activities. The relative activities of these enzymes
account for the accumulation of the 5'-phosphate compounds;


62
Table 4.1 Indicator of vitamin B-6 nutritional status of
human subjects 24 h prior to each trial1.
TRIAL URINARY 4PA PLASMA PLP ERYTHROCYTE
/xmol/24 H nmol/L AspAT
STIMULATION
BY PLP
%
TRIAL 1 10.81.9 86.97.4 35.84.5
(5/xmol d2PN
+ 2/xmol PN)
(n=6)
TRIAL 2 8.01.8 82.410.3 36.46.3
(5/xmol d2PN +
1/xmol PN
+ 1/xmol PN-GLUCOSIDE)
(n=6)
TRIAL 3 7.410.7 83.919.8 16.715.9
(5/xmol d2PN +
2/xmol PN-GLUCOSIDE)
(n=6)
1Mean and SEM, n=6; Blood and urine collections were made
24 h prior to administration of labeled forms of vitamin
B6. Proposed guidelines for adequacy of vitamin B6 status
are: urinary 4PA excretion, >5/xmol/24 h (ref. 139) ; Plasma
PLP, >50 nmol/1 (ref. 139) ; And in vitro stimulation of
erythrocyte aspartate aminotransferase (AspAT) by added
PLP, < 50% (ref. 140).
Stable-isotopic trials
This study was conducted to determine the effect of PN-
glucoside on the metabolic utilization of d2PN by evaluating
the excretion of d24PA following the ingestion of an oral dose
of d2PN while increasing the dose of PN-glucoside across the
three trials. The detection of urinary d24PA was conclusive


64
Table 4.3 Percentage of d2PN (5/mol) dose excreted as urinary
d24PA.
EXPERIMENTAL TRIALS
TIME TRIAL 1 TRIAL 2
TRIAL 3
(H) (5/mol d2PN (5/mol d2PN + (5/mol d2PN
+ 2/mol PN) l/mol PN + l/mol 2/mol PN-
(n=6) PN-GLUCOSIDE) GLUCOSIDE)
(n=6) (n=6)
(/mol)
(/mol)
(/mol)
(SEM)
0-24
46.5
27.2
24.3
12.7
24-48
37.1
28.6
17.0
12.6
0-48
83.6
55.8
41.3
17.7
1Values based on n=6 are means and pooled standard error of
the mean (SEM) The quadratic effects were not
significant for any dependent variables. The values in
table are significantly different at p<0.05 for linear
effects of the amount of PN-glucoside.
the isotopically labeled d2PN, the mean d2/dQ molar isotopic
ratio of labeled 4PA was 0.12 0.02, 0.09 0.02, and 0.09
0.01 (mean SEM) for trial 1, 2, and 3 respectively (Table
4.2). These were not significantly different. The 24-48 hr
molar isotopic ratios are also not significantly different
(Table 4.2).
The group mean values for excretion of d24PA as a
percentage of the administered d2PN dose between trials 1 and
3 for the 24-h post-dose period decreased linearly with the
amount of PN-glucoside (Table 4.3). Total 4PA excreted over
a 24-h and 48-h post-dose period showed group mean values for


37
the multitude of roles vitamin B-6 plays in the body and its
impact on the overall health of individuals, it is imperative
that information be collected concerning how naturally
occurring forms of vitamin B-6 potentially interact with one
another.


CHAPTER 3
PYRIDOXINE-5 '-/3-D-GLUCOSIDE AFFECTS THE
METABOLIC UTILIZATION OF PYRIDOXINE IN RATS
Introduction
A conjugated form of pyridoxine (PN) was first isolated
from rice bran and identified as 5'-0-(/3-D-
glucopyranosyl)pyridoxine (PN-glucoside) (67). Analysis of a
variety of plant-derived foods by HPLC has shown that PN-
glucoside comprised 5-80 % of the total vitamin B-6 in many
fruits and vegetables (75). Similar results were obtained by
the use of a microbiological assay procedure for the
glycosidic conjugate of pyridoxine (65).
Conflicting data have been reported concerning the
bioavailability of PN-glucoside in vitamin B-6 metabolism.
Studies in rats have indicated that purified PN-glucoside is
relatively well absorbed but undergoes little metabolic
utilization and is rapidly excreted in intact form (77,80).
Most experimental results in rats have indicated 20-30 % net
bioavailability of PN-glucoside relative to PN (77,80-81). In
constrast, Tsuji et al. (79) observed nearly 100 %
bioavailability of PN-glucoside relative to PN in the rat. The
bioavailability of PN-glucoside, although incomplete, is
substantially greater in humans than in rats (82). The mean
38


15
that any possible adverse effects of thermal processing and
storage on the bioavailability of vitamin B-6 would not be of
sufficient magnitude to explain the incomplete utilization of
the vitamin observed in human subjects (57,65).
Pvridoxine-fl-D-Glucoside
A conjugated form of pyridoxine, 5'-O-(0-D-
glucopyranosyl) pyridoxine (PN-glucoside) has been found to be
a major naturally occurring form of vitamin B-6 in many fruits
and vegetables of human diets. Scudi et al. (66) first
reported the presence of conjugated forms of the vitamin B-6
in 1942 to occur in rice bran. Some thirty-five years later,
Yasumoto et al. (67) reported that the PN component of the
conjugated form of vitamin B-6 in rice bran was in a 1:1 ratio
with glucose and identified this glycosylate as a 5'-
glucopyranosyl-derivative of PN. Several other studies (68-
69) have indicated the synthesis of this glucoside in pea
seedlings as well as the formation of other PN conjugates in
lesser concentrations than the 5'-pyridoxine-glucoside
derivative. Following the first isolation and structural
identification of 5 -0- (/3-D-glucopyranosyl) pyridoxine (PN-
glucoside) from rice bran (67), the possible nutritional
significance of glycosylated forms of vitamin B-6 has been a
subject of intense study. Evidence for the presence of
conjugated forms of vitamin B-6 existing as a variable
proportion of the total vitamin B-6 of plant-derived foods has


87
incomplete, it is approximately two fold greater than
previously found in rats.
The utilization of stable isotopic methodology in humans
continues to contribute information regarding the absorption,
metabolism and utilization of vitamin B-6 in humans. In an
effort to advance our understanding of vitamin B-6 nutriture
in humans it is necessary to gather information regarding the
dietary forms of vitamin B-6, their bioavailabililty and the
enzymes responsible for the interconversion of B-6 vitamers.
It is equally important to determine any potential interaction
between multiple forms of B-6 compounds and the effect these
may have on metabolic utilization of B-6 vitamers when
consumed together.
The purpose of the present research was to determine the
effect of PN-glucoside on the metabolic utilization of PN when
administered in a simultaneous dose in rats and humans. The
adventive of stable isotopic techniques allowed for the
evaluation in human subjects. The effect of PN-glucoside on
enzymes responsible for control of the conversion of vitamin
B-6 compounds to the active coenzyme form, PLP was also
determined.
The research discussed in chapters 3-5 was divided into
three separate but interconnected studies. The experimental
design of each study was contingent on the findings of the
prior study or studies. Initially, an animal study was
performed to determine if any interaction between PN-glucoside


34
note that in comparative assessment of either of the enzymes
PL kinase or PMP (PNP) oxidase from different sources, the
levels and forms (apo or holo) of the enzyme is markedly
influenced by factors such as species, age, sex and dietary
status of the animal (110). Therefore, the following
discussion regarding the properties of these two enzymes will
be limited to rat liver as the source for the enzymes, unless
otherwise noted.
Hepatic PL kinase has a molecular weight of approximately
60,000 and a pH optimum between 5.5 and 6.0 (117). PL kinase
from all sources studied requires a divalent cation for
activity, with the greatest activity in liver PL kinase seen
with Zn2* (30) The enzyme also requires monovalent cations
with activation greatest in the presence of K* (117).
Although similar to PL kinase in tissue distribution and
molecular weight, PMP (PNP) oxidase differs significantly in
other physical and kinetic properties. The molecular weight
of oxidase is approximately 54,000, each having two identical
subunits of 27,000. PMP (PNP) oxidase occurs widely
distributed in both tissues and cells (118-119). Pogell (32)
was the first to describe PMP (PNP) oxidase as having a
requirement for a flavin cofactor in the oxygen-dependent
conversion of PMP to PLP. Later Wada and Snell (31) observed
that the same oxidase was responsible for the conversion of
both PNP and PMP to PLP. The investigators reported that the
partially purified enzyme required FMN as the specific


26
antagonistic compounds has been proposed to explain certain
observed effects of food processing or storage on the
bioavailability of vitamin B-6 in certain foods (94-96). In
studying the stability of the B-6 vitamers in food products,
PN was found to be the most stable vitamer, followed by PL,
PM, PMP, and the least stable B-6 vitamer being PLP (88) .
Much research has focused on the potential for PL and PLP to
react with such food components as proteins, amino acids and
reducing sugars resulting in the formation of degradation
products of limited bioavailability.
Several studies have reported the characterization of the
by-products resulting from the interaction of PL and PLP with
amino groups of proteins in food products (96-102). The
chemical reaction of PL or PLP with cysteine or other
sulfhydryl amino groups has been proposed as a mechanism
responsible for the lowered bioavailability of vitamin B-6
reported in thermally sterilized milk products (96-98). One
such product produced from the heat induced reaction of
cysteine and PL has been identified as bis-4-pyridoxyl
disulfide (97). Later studies by Srncova et al. (99)
supported the reactivity of milk protein sulfhydryl groups
with PL when high concentrations of PL were present. The
spontaneous reaction of PL or PLP with various thiols and
aminothiols results in the formation of thiohemiacetal and
thiazolidine complexes (94) These chemical complexes are
readily dissociable and therefore would not impair the


66
nonglycosylated species of the vitamin is nutritionally
significant. These effects, which may be minor in a mixed
diet containing both animal and plant derived foods, may be
much more pronounced in vegetarian diets. Kabir et al. (65)
showed an inverse relationship between the content of PN-
glucoside in a food and the net bioavailability in humans.
Our present knowledge of the essential role of vitamin B-
6 in preserving health emphasizes the importance of
establishing an accurate Recommended Dietary Allowance (RDA)
for vitamin B-6. A vitamin B-6 deficiency with apparent
clinical symptoms is rare in the general population, but the
vitamin B-6 requirement has not been clearly determined. It
is significant that over 50% of those evaluated in the 1977-78
USDA Food Consumption Survey consumed only 70% of the 1980 RDA
for vitamin B-6. Females and the elderly were especially
prevalent in the group consuming only 70% of the RDA (2) The
1985 Continuing Survey of Food Intakes of Individuals
indicated that only 27% of women consumed 70% or more of the
RDA for vitamin B-6 (143-144). Although the range of intakes
is large, substantial portions of the population consume
apparently marginally adequate amounts of vitamin B-6.
Adequate vitamin B-6 nutriture is essential to health through
the multiplicity of roles of its active coenzymatic form
pyridoxal 5-phosphate (PLP).
The catabolic pathway for PLP involves enzymatic
hydrolysis of the phosphate ester bond, and the subsequent


85
concentration of PN-glucoside during the metabolism of PN
causes proportional increases in the concentration of liver
PNP (136).
In summary, the results from this study indicate that, in
vitro, PN-glucoside has no direct effect on PL kinase or
PMP(PNP) oxidase. These data, provide evidence that these
biosynthetic enzymes are not directly involved in the changes
in distribution of B-6 vitamers caused by ingested PN-
glucoside, observed in the previous study (136).


46
Table 3.3 Liver B-6 vitamer distribution and total liver UC
in rats administered varying levels of PN or PN-glucoside.
(Experiment 2)*
LEVELS OF
PN-GLUCOSIDE
LEVELS
OF PN
Vitamers
0 nmol
Group 1
72 nmol
Group 2
21.6 nmol
Group 3
SEM
PLP
%
31.5
liver radioactivitv
58.4a 31.5
3.4
PNP
0.0
14.4a
0.0
0.1
PMP
41.4
27.2a
40.1
1.0
PL
13.3
0.0a
14.3
0.4
PN
6.3
0.0a
6.0
0.2
PM
7.5
0.0a
8.1
0.4
Total UC
15.1
% of dose
10.0b
15.0
0.0
Values based on n=6 are means and pooled standard error
of the means (SEM) Log transformation of the data was
performed prior to analysis of variance. The quadratic
effects were not significant for any dependant variable.
a Significantly different at p<0.05 for linear effects of
PN-glucoside in the dose.
b Significantly different at p<0.05 for decreased linear
effect.
PN-glucoside (Table 3.2). In addition, the concentration of
urinary [UC]4PA, relative to total urinary 14C, and total
urinary [UC]4PA decreased linearly with the amount of
administered PN-glucoside (Table 3.2).
Experiment 2 was conducted to determine if the results of
experiment 1 were due to a specific effect of PN-glucoside or
simply to an increase in the total amount of available PN
(derived from PN-glucoside). The experimental design was the
same, except on day eight, following a twelve hour fast, the
rats were administered a simultaneous dose of 166.5 MBq (4.5