Folate status response of pregnant women and nonpregnant controls to controlled folate intakes

MISSING IMAGE

Material Information

Title:
Folate status response of pregnant women and nonpregnant controls to controlled folate intakes
Physical Description:
Unknown
Creator:
Caudill, Marie Anne, 1967-
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 28007625
oclc - 37823657
System ID:
AA00026796:00001


This item is only available as the following downloads:


Full Text










FOLATE STATUS RESPONSE OF
PREGNANT WOMEN AND NONPREGNANT CONTROLS
TO CONTROLLED FOLATE INTAKES











By

MARIE ANNE CAUDILL











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

1997




























This dissertation is dedicated to my mom and dad who from the beginning taught

me that along with hard work comes opportunity. I thank them for their

unconditional love and support.











ACKNOWLEDGMENTS


The author wishes to thank Dr. Lynn B. Bailey, her major professor, for

the opportunity to work on a meaningful and exciting research project. Her

guidance and support will always be appreciated and her dedication and

commitment to contributing scientific knowledge admired. She has been a true

inspiration in every way. The author extends much appreciation to the members

of her graduate committee, Dr. Peggy Borum, Dr. Jesse Gregory, II, Dr. Gail

Kauwell and Dr. Sharleen Simpson for their support, assistance, and advice.

Special thanks are expressed to Rita Harris, Kami Pratt, Hanh and Loan

Bui and Linda Armada for their contribution and dedication to research goals and

to Vicki Sullivan for her friendship and support. Additional thanks are extended

to Jeff Opalko, Kelli Garcia, Jerry Williamson, Christine Pfeiffer, Lisa Rogers and

Richard Bonnette for their laboratory assistance and advice. Recognition is also

given to the faculty, administrative staff and student volunteers for their help and

assistance.

The author expresses her gratitude and appreciation to her parents, sister,

brother and other family members including Rich Dolby, Diane Woods, Sue and

Sy Caudill and Pat Sickles for their love and encouragement along the way.

Special recognition is extended to Tony Caudill, her best friend and husband, for

his continued love, motivation, and understanding.


iii













TABLE OF CONTENTS

ACKNOWLEDGMENTS......................................................... iii

LIST OF TABLES............................................................ ...... vii

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

LIST OF ABBREVIATIONS............ ............. ............................ x

ABSTRACT ....................................................................... xii

CHAPTERS

1 INTRODUCTION .................................... ........................
Hypotheses........................................................................ 6
Specific Aims.................................................................... 6

2 LITERATURE REVIEW ........................................................ 9

History...................................... ....... ...................... 9
Folate Biochemistry ............................ .............................. 9
Food Distribution and Folate Bioavailability........................... .... 11
Transport and Tissue Distribution........................... .................. 15
Folate Functions ...................... ......................................... 19
Turnover and Excretion of Folate............................................. 23
Folate Requirements and Recommended Dietary Allowances ............. 28
Folate Deficiency and Pregnancy.............................................. .... 39
Folate Supplementation During Pregnancy.................................. 41
Neural Tube Defects ........................... ................................ 43
Assessment of Folate Status...................................................... 47
Stable Isotopes................................................................ 54

3 STUDY DESIGN AND METHODS....................................... 57

Study Design ....................................................................... 57
Subjects .............................................................................. 59
Diet.............................. ....... .......................... 59
Folate Extraction from Menus................................................... 62

iv









Synthesis, Purification and Identification of Deuterated Folic Acid......... 63
Supplemental Folic Acid................................... .............. 66
Purification and Administration of ['3C]pABG................................ 68
Blood Collection and Processing ........ ........... ........................ 69
Urine Collection and Processing................................................. 72
Analyses Performed by Microbiological Assay................................ 72
Serum Folate............ .... ................................... ............... 72
Red Cell Folate............................................................... 74
U rinary Folate............... .............. ...................... ............... 75
Urine and Serum Creatinine Determinations (GFR)............................ 75
HPLC Quantitation of Urinary Intact Folates, Urinary pABG and apABG
And Preparation for GCMS .............................. ............. 76
Overview ........................................ .................. ............. 76
Procedures ..................................................................... 77
A. Purification of Folate Binding Protein from Whey.................... 77
B. Preparation of Folate Binding Protein Affinity Columns........... 78
C. Binding Capacity Determination of Affigel-10 FBP Columns...... 79
D. Purification of Urine Samples using Affinity Chromatography....... 80
E. HPLC Quantitation of Urinary 5-Methyl-THF and Folic Acid........ 81
F. Chemical Cleavage of Intact Folates to pABG........................ 82
G. HPLC Isolation of pABG and Derivatization.......................... 83
H. Synthesis of [3H]pABG...................................... ................ 84
I. Purification of [3H]pABG by HPLC................................... 85
J. Synthesis of [3H]apABG ........................................... 86
K. Purification of [3H]apABG by HPLC............................... 87
L. Preparation of Ion Exchange Columns and Fractionation of Urine.. 88
M. Deacetylation of Urinary apABG and Rechromatography............ 88
N. Diazotization and Solid-phase extraction.............................. 89
0. Regeneration of pABG ................................... .. ........... ... 90
P. HPLC Quantitation of Urinary pABG................................ ... 90
Q. Recovery of [3H]pABG and [3H]apABG added to urine samples... 91
R. Preparation of Catabolite Samples for GCMS Analysis............... 92
Statistical Analyses................................. ....................... 92
Interpretation of the Data...... ......... ............................... ...... 94
Serum Folate......................................................... .......... 95
Red Cell Folate............................................................. 95
Urinary Intact Folate.............................................. ........ 97
Urinary Catabolites (apABG and pABG) ................................. 97

4 RESULTS..................................... ................ .................. 99

Folate Content of Menus ............................................. .... ... 99
Food Record Analyses ............................................................ 100

V









Serum Folate ........................................................................ 100
Red Cell Folate..................................................................... 103
Hematocrit......... ............... ........................................... 106
Urinary 5-Methyl-THF............ .......................................... 108
Urinary Folic Acid....................................................... ..... .. 111
Urinary Total Folate................................................................ 112
Glomerular Filtration Rate...................................................... 112
Urinary apABG....................................................................... 113
Urinary pABG..................................... .......................... 116

5 DISCUSSION AND CONCLUSIONS........................................ .. 119

6 SUMMARY............................................................... ..... 131

APPENDIX........................................................... ............. 133
Microbiological Assay Procedures ................................................ 133
Microplate Media Preparation for L. casei ................................... 133
Folic Acid Standard Preparation................................. ......... 133


LITERATURE CITED............................................................. 135

BIOGRAPHICAL SKETCH ... .................................................... 151
























vi














LIST OF TABLES

Table page

1. Five-day cycle menus .................... ...................................... 60

2. Nutrient composition of five-day cycle menus............................... 61

3. Folate content of menus............................................... ...... 99

4. Comparisons of initial, steady state and final total urinary folate
excretion by microbiological assay and HPLC...... ................... 112

5 Glomerular filtration rates at wk 0, 4, 8 and 12............................ 113






























vii













LIST OF FIGURES

Figure page

1. Structure of folic acid (CI9H19N7O6) ........................................... 10

2. Functions and interconversions of coenzyme forms of THF............. 20

3. Conversion of homocysteine to methionine and regeneration of THF..... 22

4. Overview of protocol............................................................. 58

5. Overview of catabolite isolation and purification .......................... 89

6. Serum folate concentration throughout 12 wk study........................ .. 101

7. Group comparisons of mean initial and final serum folate concentration... 102

8. Serum folate concentration at steady state..................................... 103

9. Red cell folate concentration throughout 12 wk study....................... 103

10. Group comparisons of mean initial and final red cell folate concentration. 105

11. Red cell folate concentration at steady state.................................. 106

12. Hematocrit throughout 12 wk study..................................... ...... 107

13. Group comparisons of mean initial and final hematocrit...................... 107

14. Urinary 5-methyl-THF excretion throughout 12 wk study................... 108

15. Group comparisons of mean initial and final 5-methyl-THF excretion.... 109

16. Urinary 5-methyl-THF excretion at steady state............................ .... 110





viii









Figure age

17. Comparisons in final group urinary folic acid excretion....................... 111

18. Urinary apABG excretion throughout 12 wk study.......................... 113

19. Group comparisons of mean initial and final urinary apABG excretion... 114

20. Comparisons of mean overall urinary apABG excretion.................... 115

21. Urinary pABG excretion throughout the 12 wk study....................... 116

22. Group comparisons of mean initial and final urinary pABG excretion..... 117

23. Urinary pABG excretion at steady state........................................ 118
































ix













LIST OF ABBREVIATIONS


Abbreviation Meaning

apABG acetamidobenzoylglutamate
ADP adenosine 5'-diphosphate
AICAR 5-formiminoimidazole 4-carboxamide ribonucleotide
ATP adenosine 5'triphosphate
BSA bovine serum albumin
CDC Centers for Disease Control
d day
DCl deuterium chloride
DEAE diethylaminoethyl
DHF dihydrofolate
DHHS Department of Health and Human Services
DNA deoxyribonucleic acids
dTMP deoxythymidylic acid
dUMP deoxyuridylic acid
dUST deoxyuridine suppression test
EDTA ethylenediamine teraacetic acid
FBP folate binding protein
GCMS gas chromatography mass spectrometry
GAR glycinamide ribotide
glu glutamate
h hour
HPLC high performance liquid chromatography
HCI hydrochloric acid
mg milligram
MeCN acetonitrile
MRC Medical Research Council
MTHFR methylenetetrahydrofolate reductase
NaCI sodium chloride
NaOD sodium deuteroxide
NaOH sodium hydroxide
ng nanogram
nm nanometer



x










Abbreviation Meaning

NHANES II Second National Health and Nutrition Examination
Survey
NMR nuclear magnetic resonance
pABA p-aminobenzoic acid
pABG p-aminobenzoylglutamic acid
Pd palladium
PHS Public Health Service
PLP pyridoxal 5'-phosphate
RDA Recommended Dietary Allowance
RNA ribonucleic acids
SAH s-adenosylhomocysteine
SAM s-adenosylmethionine
THF tetrahydrofolate
Lg microgram
US United States
wk week
y year



























xi












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

FOLATE STATUS RESPONSE OF
PREGNANT WOMEN AND NONPREGNANT CONTROLS
TO CONROLLED FOLATE INTAKES

By

Marie Anne Caudill

August, 1997

Chairperson: Dr. Lynn B. Bailey
Major Department: Food Science and Human Nutrition

Folate is a B vitamin that plays a major coenzymatic role in one-carbon

metabolism and is a key participant in the metabolism of certain amino acids and

biosynthesis of pyrimidines and purines. Folate requirements are higher during

pregnancy, although the increment above that required by nonpregnant women has

not been ascertained. The 1989 RDA for both pregnant and nonpregnant women

was reduced by approximately one-half. The rationale for the reduction in

pregnant women from 800 to 400 pig/d was based primarily on studies in which

folate intake was not controlled.

The purpose of this 84 d study was to investigate folate status response of

pregnant subjects (n= 12) during their second trimester and nonpregnant controls

(n=12) to defined folate intakes comparable to the current (400 gg/d) and former

(800 tg/d) RDA for pregnant women. Subjects consumed either 450 or 850 pg/d
xii









provided as a combination of dietary folate and synthetic folic acid. The outcome

variables used to assess folate status response included serum, red cell and urinary

folate and the folate catabolites pABG and apABG. Comparisons between and

within supplementation groups were made at steady state or the time at which the

subjects had acclimatized to their assigned folate intake.

No differences (P>0.05) were detected in serum and red cell folate

concentrations or in urinary 5-methyl-THF excretion between pregnant and

nonpregnant women consuming either 450 or 850 p.g/d. Serum folate and 5-

methyl-THF excretion were higher (P<0.05) in subjects consuming 850 compared

to 450 glg/d. Urinary folic acid (14.5 gg/d) was detected only in nonpregnant

controls consuming 850 pg/d folate. No differences (P>0.05) were detected in

pABG or apABG concentrations between pregnant and nonpregnant women

consuming either 450 or 850 ptg/d. Pregnant women consuming 450 compared to

850 pg/d excreted less (P<0.05) of both catabolites by the end of the study.

In summary, no differences were detected in the outcome variables

measured between nonpregnant and pregnant women within the same

supplementation group. Further, normal folate status was maintained by all

groups. These data suggest that 450 tg/d of folate derived from both dietary and

synthetic sources is adequate for pregnant women during their second trimester.

This level of intake equates to approximately 600 gg/d dietary equivalents

assuming 50% and 75% availability of food folate and synthetic folic acid

consumed with meals, respectively.
xiii














CHAPTER 1
INTRODUCTION


Folate is a generic descriptor for compounds that have nutritional

properties and chemical structures similar to those of folic acid (pteroylglutamic

acid). Thus, it includes folic acid, the oxidized, monoglutamate form of the

vitamin provided in supplements and fortified foods, and the reduced, naturally

occurring, polyglutamyl folates, which are the predominant dietary forms. De

novo biosynthesis of folates occurs only in bacteria and plants so humans must

obtain this vitamin from their diet.

Folate in its fully reduced state and polyglutamyl form participates in a

variety of coenzyme reactions involving one-carbon transfers required for amino

acid metabolism and nucleic acid synthesis, as well as the biosynthesis of specific

lipids, hormones and neurotransmitters. Deficiency of the vitamin leads to

impaired cell division and to abnormalities in numerous synthetic pathways

dependent on one-carbon transfer reactions. A wide array of clinical problems

may result from a deficiency including megaloblastic anemia, growth failure,

malabsorption syndromes and increased risk for certain types of cancer and

atherosclerosis (Bailey 1990a, Giovannucci et al. 1993, Selhub et al. 1995).

Pregnancy is associated with increased requirements for folate, although

controversy exists regarding the amount of folate required to meet the demands of
I








2
the pregnant woman and her developing fetus (NRC 1989, Subcommittee on

Dietary Intake and Nutrient Supplement During Pregnancy 1990, Herbert 1987a,

Bailey 1995). Adequate folate intake is essential during pregnancy since this time

represents a state of increased cellular division due to the rapid growth of the

fetus, the placenta and the maternal organs such as the uterus (Cunningham et al.

1993). In addition, there is an increase in erythropoiesis, a folate requiring

process, and plasma resulting in the expansion of maternal blood volume

(Cunningham et al. 1993). The pregnancy induced hypervolemia which starts

during the first trimester serves to meet the demands of the enlarged organs

(Cunningham et al. 1993). Due to the severe demands that are placed on the

supply of folate during pregnancy for the synthesis of DNA, pregnant women are

at increased risk of developing a folate deficiency (Bailey et al. 1980, O'Connor

1994). Sauberlich (1990) summarized numerous studies both within the United

States and world wide that illustrated the widespread prevalence of poor folate

nutriture during pregnancy. Bailey et al. (1980) reported that among a group of

low income pregnant women, folate deficiency was significantly more prevalent

than iron deficiency at the initiation of prenatal care. These data clearly illustrate

that many women are unable to meet the increased demand for folate during

pregnancy through diet alone which has resulted in the acceptance of routine

prenatal folate supplementation (Sauberlich 1995).

Unfortunately, due to lack of data regarding the size of the folate body

pool and equilibrium of the vitamin during pregnancy, the National Academy of








3
Sciences Committee on Dietary Intake and Nutrient Supplements During

Pregnancy (1990) could not establish a specific supplementation recommendation

for folic acid. Despite the widespread acceptance of folic acid supplementation

for the vast majority of pregnant women, the Committee recommended that folic

acid supplementation be reserved for specific "high risk" population groups such

as pregnant adolescents but not for the pregnant population as a whole.

However, this recommendation preceded conclusive research regarding NTDs

and therefore was not addressed by this committee..

The 1989 RDA for folate was lowered from 800 pg/d (NRC 1980) to 400

ig/d (NRC 1989) for pregnant women. The reduction in the RDA for pregnant

women was largely based on the findings of two studies (NRC 1989). Chanarin

et al. (1968b) reported that 100 lg/d of synthetic folic acid, in addition to dietary

folate, was able to maintain red cell folate concentrations in pregnant women

throughout gestation. Bates et al. (1982) subsequently reported an estimated

dietary folate intake of 190 Lg/d for this population group. This estimate (190

pg/d) in conjunction with Chanarin's report that 100 pg/d plus diet maintained red

cell folate concentrations was used by the 10h edition RDA committee to support

the 400 pg/d RDA (NRC 1989). Colman et al. (1975) found that maize fortified

with 300 p.g/d folic acid consumed with a constant diet maintained normal serum

and red cell folate concentrations during the last thirty days of pregnancy in a

rural, poorly nourished, African population. Interpretation of these studies is

complex, and illustrates the importance of investigating folate requirements in








4
healthy pregnant women under controlled metabolic conditions during which time

the response of pregnant women to known levels of folate intake can be assessed.

Currently, relatively little is known about the requirement for folate during

pregnancy, the specific tissue needs for folate, or how to assess folate status

during gestation (i.e. the range of acceptable values) (O'Connor 1994).

In recent years, studies have been conducted to assess the appropriateness

of the decrease in the RDA for nonpregnant women of child-bearing age and adult

men. O'Keefe et al. (1995) reported that the 1989 RDA for nonpregnant women

of 180 Lg/d was insufficient to maintain initial folate status in healthy,

unsupplemented nonpregnant women and that 400 lpg/d more adequately met the

criteria of an RDA. Similarly, Jacob et al. (1994) reported that 200 plg of folate

per day was insufficient to achieve normal plasma homocysteine concentrations

after a moderate depletion period in a depletion-repletion protocol involving ten

healthy male adults. These data suggested that both males and nonpregnant

females require more than the current RDA to maintain normal folate nutriture.

McPartlin et al. (1993) used an alternative approach in estimating folate

requirements in pregnant women at one time point during each trimester. Their

approach was based on the quantitation of the urinary folate catabolites, p-

aminobenzoylglutamate (pABG) and the more predominant form,

acetamidobenzolyglutamate (apABG). These catabolites were hypothesized to

represent folate turn over by McPartlin and associates (1993) and therefore a more

direct way to estimate requirements. McPartlin et al. (1993) reported that based








5
on the urinary excretion of pABG and apABG, the current RDA for pregnant

women was inadequate particularly in the second trimester. Limitations of this

study included the fact that folate intake was not controlled and folate status was

not assessed.

We conducted a controlled metabolic trial to investigate the response of

pregnant women in their second trimester to folate intakes comparable to the

current RDA (400 lg/d) and the former RDA (800 pg/d). Two nonpregnant

groups consisting of women similar in age, lifestyle and demographics were

included as a control. The decision to select the second trimester was based on

the findings of McPartlin et al. (1993) which indicated that the second trimester

represented the highest rate of folate catabolism and thus the greatest requirement

for folate. The overall goal of the study was to provide metabolic data which

could be used to assess the adequacy of the current RDA for pregnant women.

One objective of this study was to determine if differences existed in folate status

and catabolite excretion between pregnant subjects and nonpregnant controls

consuming equivalent folate intakes. A second objective was to determine if

differences existed in folate status and catabolite excretion between pregnant

subjects consuming folate intakes approximating the current RDA (400 Jpg/d)

compared to the former RDA (800 ug/d).








6
Hypotheses

The major hypotheses and specific aims of this research were as follows.

Hypothesis 1. The response of pregnant women and nonpregnant controls

to two defined levels of folate intake (450 or 850 pg) would indicate a greater

folate requirement for pregnant women as determined by changes in folate status

indices and catabolite excretion.

Hypothesis 2. Folate response indices would reflect a decline in folate

nutriture approaching negative folate balance among pregnant women consuming

450 tg/d of folate and maintenance of folate status in pregnant women consuming

850 ug/d.

Hypothesis 3. Folate response indices would reflect maintenance of

folate status in nonpregnant women consuming 450 pg/d of folate and an excess

exceeding the body's ability to metabolize or store folate in nonpregnant women

consuming 850 pg/d of folate.

Specific Aims

Aim 1. To determine weekly serum and red cell folate concentrations via

microbiological assay during the study period and monthly concentrations during

the follow-up period. Differences between and within supplementation groups

would help assess the adequacy of folate intake.

Aim 2. To determine weekly urinary excretion of intact urinary folate via

microbiological assay and to determine the specific forms of excreted urinary

intact folate by high performance liquid chromatography. Data would be







7
obtained related to the amount of excreted, unmetabolized urinary intact folate

(folic acid) and the amount of excreted, metabolized urinary intact folate (5-

methyl-tetrahydrofolate) and comparisons would be made between and within

supplementation groups.

Aim 3. To quantitate the urinary excretion of the two major folate

catabolites, pABG and apABG. Information on folate turnover and possibly

requirements during pregnancy would be obtained based on the premise that

catabolized folate reflects utilization and that which needs to be replaced daily.

Folate was provided as a combination of dietary and synthetic sources.

During the first half of the study, a small percentage of the total daily intake of

folate came from folate labeled with deuterium followed by provision of 100%

unlabeled folate. Labeling of synthetic folic acid with deuterium, a staple isotope,

can be used safely in humans (Gregory et al. 1990). Information on folate

kinetics, including turnover rates and pool sizes, can be obtained by comparing

the ratio of labeled folate to unlabeled folate in biological samples collected from

humans consuming deuterated folic acid. A small subsample of subjects

consisting of one subject from each group (n=4) was also given [13C]pABG

midway through the 12 wk period so that the bioavailability of exogenous pABG

could be evaluated and reported separately by a co-worker.

Several parameters were used to assess folate nutriture including serum

folate, red cell folate, urinary intact folate and the urinary breakdown products,

pABG and apABG. Kinetic data will also be obtained based on the isotopic








8
enrichment of biological samples including urine, red cell and plasma and

reported separately. Plasma homocysteine, a functional index of folate status, was

measured by a co-worker.













CHAPTER 2
LITERATURE REVIEW


History

In 1931, Dr. Lucy Wills described a macrocytic anemia in pregnant Hindu

women that responded to treatment with yeast extract. Later, Wills and Bilimoria

(1932) described a similar macrocytic anemia in monkeys that responded to a

factor present in crude liver extracts. The active principle in liver and yeast

extracts described by Wills had many properties in common with bacterial

"growth factors" present in liver (Snell and Peterson 1940) and spinach leaves

(Mitchell et al. 1941). The term folic acid, derived from the Latin word "folium"

meaning "leaf," was coined by Mitchell et al. (1941) after isolating this substance

from spinach leaves. Folic acid was synthesized by Angier et al. (1945), and it

was shown to be the common factor in the preceding observations.

Folate Biochemistry

The folic acid molecule consists of three parts: the pteridine moiety (with

an amino group in position 2 and hydroxyl group in position 4); the para-amino-

benzoic acid; and the L-glutamic acid (Figure 1). The pteridine component is

linked by a methylene bridge to para-aminobenzoic acid which is joined by

peptide linkage to glutamic acid. The first two fractions compose pteroic acid

which leads to the classification of folic acid as pteroyl-glutamic acid. Folic acid

9








10
(CI9H19N706) is named in the chemical literature as a derivative of L-glutamic

acid: N-[4-[{(2-amino- ,4-dihydro-4-oxo-6-pteridinyl)methyl]-amino}benzoyl]-

L-glutamic acid. Crystalline folic acid is yellow with a molecular weight of

441.41 g/mol. Folic acid, itself, is not biochemically active but becomes so after

it has undergone reduction. Naturally occurring folates include a very large

number of compounds because (a) the pirazine ring (atoms 4a,5,6,7,8 and 8a) of

the pteridine moiety can occur in three states of reduction with (b) six different

one-carbon substituents present at N-5 and/or N-10 and

(c) the polyglutamyl chain may have as many as 12 glutamyl residues (Krumdieck

1990).


N N
2N 1 8 0 COOH
3 6 II I
"4 5 CHK-NH- C-NH- CH-CH -CH.-COOH

OH
I II II I
pteridine p-aminobenzoic acid glutamic acid

pteroic acid


pteroylglutamic acid







Figure 1. Structure of Folic Acid (Ci9H19N706)


In nature, folate exists mainly in the conjugated forms. Pentaglutamates

and hexaglutamates predominate as the intracellular folates in animal tissue, while








11
extracellular folates (i.e. in plasma, spinal fluid, bile and urine) are

monoglutamate derivatives (Krumdieck 1990). Most natural folates are unstable

and are degraded by heat, oxidation and ultraviolet light. Some tetrahydrofolate

(THF) derivatives including 5-formyl-THF and 5-methyl-THF are relatively heat

stable. Acid will destroy THF and 5,10-methylene-THF. The addition of

ascorbic acid and other reducing agents has helped tremendously in preventing

the oxidation and subsequent degradation of folates in biological samples.

Folates have limited solubility in water and in most organic solvents such

as methanol. Folates are minimally soluble in the mildly acidic range (pH 2 to 4),

with solubility increasing in proportion to the pH above this range (Gregory

1989). Folates are also highly soluble at very low pH.

Food Distribution and Folate Bioavailability

Folate is found in a wide variety of foods originating from both animal

and plant sources. Liver, mushrooms, orange juice and green leafy vegetables are

particularly good sources (Combs 1992). White bread, rolls and crackers,

although not rich sources of folate, represent major contributors of folate in the

U.S. diet according to the Second National Health and Nutrition Examination

Survey (NHANES II) data, 1976-1980 (Subar et al. 1989). These sources are

likely to remain major contributors secondary to the FDA mandating the

fortification of bread products with folic acid (FDA 1996). Dietary folate exists

primarily in the reduced form as polyglutamyl derivatives of THF with 5-methyl-

THF and 10-formyl-THF predominating. Very little free folate (folyl

monoglutamate) is found in foods. Food storage and preparation can greatly








12
affect the stability of folate. Exposure of the folate molecule to heat, ultraviolet

light, air and metal ions will result in oxidation and subsequent cleavage

rendering it inactive (Herbert 1987a, Gregory 1989), although a small portion of

the oxidized folates may be converted back to biologically active forms (Gregory

1989, Shane 1995). Further, because folate is a water soluble vitamin, it can be

leached out of foods during periods of boiling.

Before folates can be absorbed from the gut, glutamyl residues in excess

of one must be removed because only monoglutamyl forms can enter the

enterocyte. This cleavage is accomplished by an exocarboxypeptidase folyl y

glutamyl hydrolase, more commonly referred to as folate conjugase (EC

3.4.12.10) (Halsted 1979) which functions at a pH optimum of 6.5 (Gregory

1995). Medications or conditions that increase this pH will decrease the activity

of this enzyme possibly leading to folate malabsorption. Natural conjugase

inhibitors found in certain foods including yeast, beans and lentils may also

reduce folate availability (Bailey 1988, Butterworth et al. 1974). However, Wei

et al. (1996) reported high relative bioavailability of both monoglutamyl and

hexaglutamyl folic acid when added to lima beans and tomatoes. Chronic

exposure to alcohol and possibly zinc deficiency may also reduce folate

bioavailability (Bailey 1988, Naughton et al. 1988, Tamura et al. 1978).

The monoglutamate form of folate is actively transported across the

proximal small intestine by a saturable, pH-dependent process (Gregory 1995,

Halsted 1990, Rose 1978). Two folate-binding proteins (FBPs) exist in the








13
jejunal brush border and are thought to be involved in folate binding and transport

(Shoda et al. 1993). Folic acid can also be absorbed by a nonsaturable

mechanism involving passive diffusion when high concentrations are consumed

(Gregory 1995).

The term "folate bioavailability" refers to the efficiency of utilization of

this vitamin, including physiological and biochemical processes involved in

intestinal absorption, transport, metabolism and excretion (Gregory 1995). The

bioavailability of dietary folate is estimated to be approximately 50% (Sauberlich

et al. 1987). In contrast, synthetic folic acid consumed under fasting conditions

is approximately 100% available (Gregory 1995). Foods fortified with synthetic

folic acid were initially considered to be only 50% available (Colman et al. 1975).

More recently, however, studies have found that foods fortified with synthetic

folic acid are highly available (Cuskelly et al. 1996, Pfeiffer et al. 1997). Pfeiffer

et al. (1997) determined the bioavailability of ['3Cs]folic acid in various fortified

cereal-grain products and in water as a control, relative to the intravenously

injected [2H2]folic acid. Measurement of isotope excretion ratios of urinary folates

(% of [ C5]folate dose / % of [2H2]folate dose) indicated no differences among

treatment groups or the control suggesting that folic acid fortified cereal-grain

foods were highly available. Pfeiffer and associates (1997) also indicated that

consumption of ['3Cs]folic acid after a light breakfast meal did not significantly

effect its bioavailability. The influence of food on the bioavailability of

simultaneously ingested folic acid supplements has not been firmly established.








14
However, based on recently reported data related to fortified foods, one could

hypothesize that it is more available than endogenous folate present in food and

less available than folic acid consumed under fasting conditions.

Once folate enters the enterocyte, it usually undergoes reduction to THF

and either methylation or formylation (Perry 1973, Selhub et al. 1983). However,

folate metabolism within the enterocyte is not required for entry into portal blood

or for transport (Shane 1995). When high doses of various folates are given, little

metabolism occurs, and the majority of folate appears in portal circulation

unchanged (Herbert and Das 1994). The degree of metabolism in the intestinal

mucosa is a function of the amount of folate administered (Gregory 1995).

Because the reduction process is readily saturable (Gregory 1995), supplemental

folic acid may be more efficiently reduced and methylated if divided into several

smaller doses (Lucock et al. 1989b). Gregory (1995) suggested that this may

increase the efficiency of its metabolic utilization and improve its bioavailability.

In 1966, Giles hypothesized that the high incidence of folate deficiency

during pregnancy was a consequence of reduced folate absorption. Iyengar and

Buba (1975) subsequently conducted a study to compare folic acid absorption in

pregnant women to that of nonpregnant women. Two milligrams of folic acid

along with 10 tCi of tritiated folic acid were administered to 9 pregnant women

between 10 to 26 wk gestation undergoing therapeutic termination of pregnancy

and 6 nonpregnant women undergoing sterilization. Absorption varied between

65 and 95 percent, with a mean absorption of 80 percent in both pregnant and








15
nonpregnant subjects. The results indicated that there was no alteration in

absorption of a large dose of folic acid during the first two trimesters of

pregnancy. A limitation to this study was lack of comparison between pregnant

and nonpregnant women in their ability to absorb polyglutamyl folates. Studies

(Bailey et al. 1988, Gregory et al. 1991) conducted in nonpregnant subjects to

determine the bioavailability of mono- and polyglutamyl folate suggested partial

availability of polyglutamyl folates compared to monoglutamyl forms. However,

Wei et al. (1996). reported high availability of both polyglutamyl and

monoglutamyl forms when added to lima beans and tomatoes.

Transport and Tissue Distribution

Monoglutamates, mainly 5-methyl-THF, and to a lesser extent folic acid,

dihydrofolate (DHF) and formylated forms are present in portal circulation.

Approximately two-thirds of the folate in plasma is bound to protein (Groff et al.

1995). The majority of these proteins bind folate with relatively low affinity and

are non-saturable (Ratnam and Freisheim, 1990). Examples include albumin and

a-2 macroglobin (Groff et al. 1995). In contrast, folate binding proteins present

in low amounts in plasma and blood cells bind folate with high affinity (Wagner

1995). Free folate accounts for the remainder of folate found in the plasma.

Folate can then be taken up by the liver or transported to extrahepatic tissues.

Within the liver, folate is metabolized to polyglutamate derivatives by folyl

polyglutamate synthetase and retained or released into blood or bile. The folate

derivative, 5-methyl-THF, is a poor substrate for folyl polyglutamate synthetase







16
and in order to be conjugated must first have its methyl group removed once it

enters the cell. This can be accomplished by methionine synthase, a vitamin B12-

dependent enzyme. Polyglutamates cannot cross biological membranes and

deconjugation to the monoglutamate form (removal of glutamyl residues in excess

of one) must occur prior to releasing folate into circulation (Combs 1992).

Tetrahydrofolate, 5-methyl-THF and 10-formyl-THF are the predominate

folate derivatives found in the liver and are typically bound to glutamates varying

in length from 4 to 7 (Groff et al. 1995). Most of the 5-methyl and 10-formyl

derivatives are secreted into bile which are then reabsorbed via enterohepatic

circulation or excreted via the feces. Approximately half of the total folate that

reaches peripheral tissues occurs by this recirculation process (Steinberg 1984).

Folate secreted directly into circulation is transported to other tissues bound to

either low or high affinity folate binders as previously described.

Recent studies from numerous laboratories suggest the existence of

several distinct mechanisms for transport of folate across membranes (Henderson

1990). Folate transport into tissue cells can occur by a carrier-mediated process

that may or may not require ATP (Groff et al. 1995) or by folate binding proteins

that are associated with the plasma membrane of cells (Da Costa and Rothenberg

1996). These membrane-associated folate binding proteins act as folate receptors

and thereby facilitate cellular uptake of folate (Da Costa and Rothenberg 1996).

These receptors are attached to the plasma membrane via a

glycosylphosphatidylinositol anchor and can also be released from the membrane








17
as a soluble protein upon hydrolysis by a glycosyl phosphatidylinositol-specific

phospholipase C (Lee et al. 1992). Although many cells do not express this

membrane-associated FBP, high levels are found in placental trophoblastic cells,

ovaries, fallopian tubes and mammary glands (Weitman et al. 1992). It was

hypothesized that the membrane-associated folate receptor's of placental

trophoblastic cells provides the following two advantages: 1) ensures adequate

folate supply even under conditions of deficiency secondary to the receptors high

affinity for folates; and 2) prevents potentially harmful folate analogues from

entering the placental trophoblastic cells due to the high substrate specificity of

the receptor (Selhub 1994). The folate receptors found in placental cells thereby

appear to ensure adequate folate supply to the fetus even when maternal stores are

depleted (Selhub 1994). Recently, an ex vivo placental cotyledon perfusion

model was used by Henderson et al. (1995) to test the hypothesis that under

physiologic conditions, placental folate receptors played a major functional role in

transplacental folate transport. Their results indicated that maternal-to-fetal folate

transfer is a two-step process. The first step concentrates 5-methyl-THF (or folic

acid) by binding it to placental folate receptors on the maternally facing chorionic

surface. Gradual release of 5-methyl-THF from this pool adds to incoming

circulating folates to generate an intervillous blood concentration (placental

compartment) three times that of maternal blood. The second step involves

transfer of folates to fetal circulation along a downhill concentration gradient.

Thus, folate bound to placental folate receptors are predestined for transplacental








18
folate transport, since incoming (dietary) folates displace placental folate receptor

bound folates that then passively diffuse to the fetus (Antony 1996). This study

provided a mechanism for the observation that the developing fetus drains

maternal folate supplies as evidenced by the significantly higher plasma and red

cell folate concentrations of newborns compared to maternal concentrations (Ek

and Magnus 1981).

Folate receptors are also involved in folate transport across proximal

renal tubular cells (Selhub 1994). Data suggest that after glomerular filtration, the

luminal folate binds folate receptors in the renal tubular cells and is internalized

rapidly via folate receptor mediated endocytosis. Dissociation of folate from the

receptor occurs in the low pH of endocytotic vesicles allowing the return of folate

to the blood, where it can be taken up by other cells, and the recycling of apo-

folate receptor back to the luminal brush-border membrane.

Once within cells, 5-methyl-THF (predominate folate derivative in plasma

or urine) is demethylated and converted back to a polyglutamyl form by

folylpoly-y-glutamate synthetase in order to become a functional coenzyme,

accepting and transferring one-carbon fragments (Wagner 1995). The addition of

glutamyl residues also represents a way of keeping the folate in the cell. The

mitochondrial folate pool is distinct from the cytosolic pool in that 5-methyl-THF

is located in the cytosol, while 10-formyl-THF is in the mitochondria (Shane

1995). Tissues are limited in their ability to store folate beyond their normal

cellular requirements. The total body content of humans is estimated to be








19
approximately 10 mg, of which one-half resides in the liver (Herbert 1987, Combs

1992).

Folate Functions

Folate coenzymes are involved in numerous reactions that involve the

transfer of one-carbon units as depicted by Figure 2. These one-carbon units exist

at various levels of oxidation ranging from methane, the most reduced, to carbon

dioxide, the most oxidized (Wagner 1995) and are carried by folate when the

coenzyme is in the fully reduced THF form. As depicted by the diagram, folate

is reduced in two steps to THF by dihydrofolate reductase which utilizes NADPH

+ IH. The majority of folate derivatives are required for amino acid metabolism

and deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) synthesis which is

exemplified by the following reactions.

Tetrahydrofolate can accept a one-carbon unit from serine forming glycine

and 5,10-methylene-THF (A). This reaction is reversible via the donation of a

hydroxymethyl group from 5,10-methylene-THF to glycine producing serine and

THF. In mammalian systems, the 13-carbon of serine provides the majority of

carbon units with THF serving as the carrier of these activated carbon atoms

(Shane and Stokstad 1985). For example, the folate derivative, 5,10-methylene-

THF, produced from the one-carbon donation by serine to THF, can either be

reduced in a nonreversible reaction by NADPH and FADH2 to 5-methyl-THF

(B), oxidized to 5,10-methenyl-THF in a reversible reaction (F) or function as a

one-carbon donor in a reaction required for DNA synthesis (G).









20
Enzymes involved in interconversions
of coenzymes forms of THF
(A) Serine hydroxymethyltransferase Folate
(coenzyme-PLP)
(B) Methylene-THF reductase NADPH+ H*
(C) Methionine synthetase
(coenzyme-B 2)
(D) Formiminotransferase
(E) Cyclodeaminase NADP*
(F) Methenyl-THF reductase
(G) Thymidylate synthetase
(H) Formate-activating enzyme Dihydrofolate
(I) Phosphoribosylglycinamide transformylase (DBF)
(J) Aminocarboxamide ribotide transformylase
(K) Cyclohydrolase NADPH+ H'
Methotrexate Reductase dTMP

Homocysteine Methionine NADP*

5-methyl-THF Tetrahydrofolate (THUM


FAD Serine FIU Formate I J GAR
B ) PLP FIGLU ATP SAICAR
FADH Glycine ADP
NADPH u' ADPl "

5,10-methylene-THF 5-formimino-THF 0-formyl-THF




S0 H20
NADP K

NADPH

5.10- methenyl-THF







Figure 2. Functions and interconversions of coenzyme forms of THF.
Modified from Groff JL, Gropper SS and Hunt, SM. Advanced Nutrition and
Human Metabolism, New York: West Publishing Co. 1995; p 266.




The folate derivative, 5-methyl-THF, can donate its methyl group to

homocysteine to form methionine and regenerate THF (C). This reaction (C),


catalyzed by methionine synthase, requires vitamin B12 in addition to folate







21
(Figure 3). Vitamin B12 is bound tightly to methionine synthase and serves to

acquire the methyl group from 5-methyl-THF and transfer it to homocysteine

thereby producing methionine and regenerating THF. Vitamin B12 deficiency

induces a secondary folate deficiency which is best explained by the "methyl

trap" hypothesis (Shane and Stokstad 1985). When the activity of methionine

synthase is reduced due to lack of vitamin B12, 5-methyl-THF accumulates and

prevents the regeneration of THF ultimately inhibiting the generation of other

folate one-carbon forms. The megaloblastic anemia that may result can be

explained by a lack of folate coenzymes for DNA precursor synthesis in blood

cells (Shane 1995). Hence, megaloblastic anemia may develop not only as the

result of a primary folate deficiency but also from a primary vitamin B12

deficiency which causes a secondary folate deficiency.

Histidine metabolism also relies on the availability of folate (D). Urocanic

acid is generated by the deamination of histidine which can undergo further

metabolism to yield formiminoglutamate (FIGLU). At this point, THF accepts

the formimino from FIGLU producing glutamic acid and 5-formimino-THF, the

latter can be converted to 5,10-methenyl-THF (E) and then reduced to 5,10-

methylene-THF (F). Hence, 5,10-methylene-THF, an important coenzyme in

DNA synthesis, can be produced by two separate reactions (A) and (F).

Thymidylate synthase requires 5,10-methylene-THF and deoxyuridylate (dUMP)

to generate the pyrimidine, thymidylic acid and DHF (G). Tetrahydrofolate is

regenerated from DHF by dihydrofolate reductase. Because both dihydrofolate








22
reductase and thymidylate synthase are very active during cell division, they have

been the target of chemotherapeutic drugs. For example, methotrexate binds the

active site of dihydrofolate reductase which prevents the regeneration of THF

thereby reducing the availability of 5,10-methylene-THF needed for cell division.



5-methyl- Cobalamin Methionine
THF
ATP
\Methionine adenosyl transferase
P P + PP

SMethionine S-adenosylmethionine
Synthetase (SAM)
or
Homocysteine
5,10-methylene Homocysteine I
THF methyl Acceptor of methyl group
transferase
CH3 acceptor

S-adenosyl homocysteine
(SAH)

H,O

Adenosine
THF Methyl
Cobalamin Homocysteine
Figure 3. The conversion of homocysteine to methionine and the
regeneration of THF. Reproduced from Groff JL, Gropper SS and Hunt, SM.
Advanced Nutrition and Human Metabolism, New York: West Publishing Co.
1995; p 266.


Formate is required for the conversion of THF to 10-formyl-THF (H)

which is the folate coenzyme needed for purine ring formation and consequently

both RNA and DNA synthesis. Carbon 8 of the purine atom is generated by the

formylation of glycinamide ribotide (GAR), a reaction involving 10-formyl-THF,

to form formylglycinamidine ribotide (FGAR) and THF (I). Purine ring atom C2








23
is acquired by formylation of 5-aminoimidazole 4-carboxamide ribonucleotide

(AICAR) involving 10-formyl-THF to yield 5-formiminoimidazole 4-

carboxamide ribotide (FAICAR) and THF (J). The folate derivative, 10-formyl-

THF, can also be converted to 5,10- methenyl-THF (K) and then reduced to 5,10-

methylene-THF as dictated by the needs of the cell.



Turnover and Excretion of Folate

Although studies conducted in both rat and humans confirm that

folylpolyglutamates turn over in mammalian cells and tissues, the mechanism of

turn over has not been well characterized (Shane, 1995). Since only a small

proportion of supplemental or dietary folate is recovered in the urine when

consumed in physiological amounts (Jukes et al. 1947), it has been speculated that

the tissues alter these compounds in metabolism. In the late 1950s, Dinning and

co-workers reported that upon incubating folic acid with liver slices, a

"diazotizable amine" was formed. He predicted that the cleavage of folate to a

"diazotizable amine" was one pathway for the metabolic disposal of folic acid.

His experiments with rats supported his prediction when a "diazotizable amine"

was found to be a major urinary excretory product following administration of

either folic acid or 5-formyl-THF. Approximately half of the excreted metabolite

was in free form, while the other half was N-acetylated. The investigators

reported that cleavage to a diazotizable amine was a pathway of major importance

for the metabolic disposal of folate based upon the magnitude of the amine







24
excretion. Two decades later, administration of [3H] and/or [14C] labeled folic

acid to rats revealed that folate catabolism takes place by cleavage of the C9-N10

bond, and not by excretion of inactive forms of the vitamin (Barford et al. 1978,

Connor et al. 1979, Murphy et al. 1976, 1978 and 1979). The principle

catabolites of a tracer dose of [3H]pteroylglutamate (labeled on the benzene ring)

or [14C] pteroylglutamate (labeled on the pterin moiety) were apABG or

derivatives of the pterin component, respectively (Barford et al. 1978, Connor et

al. 1979, Murphy et al. 1976, 1978 and 1979). This mode of catabolism was

subsequently confirmed in humans (Krumdieck et al.1978, Pheasant et al. 1979,

Saleh et al. 1980). Krumdieck et al. (1978) administered [2-14C] folic acid to a

human volunteer and determined both urinary and fecal radioactivity, as well as

urinary excretion of folate over a 129 day period of observation. Urinary pterin

and isoxanthopterin were found to be radioactive which supported the mode of

catabolism described in rats in which folate is cleaved in half producing pterin

and pABG moieties. A substantial amount of [4C]folate and derivatives were

also detected in the feces indicating that this is an important avenue of elimination

of folate and derivatives (i.e. catabolites) (Krumdieck, 1978). Pheasant et al.

(1979) gave an oral dose of 2-[14C], 3',5',9-[3H]folic acid (60 pg or 5 mg folic

acid) to five subjects and obtained urine collections over the following 24 h

period. Urinary scission products including apABG were identified as the major

urinary species following the low dose. Saleh et al. (1980) administered an oral

dose of a mixture of [2-14C]- and [3',5',7,9-3H]-Pte Glu to 14 subjects and








25
identified the metabolites present in the urine collected 0-24 h after administration

of the radioactive tracers. Their data clearly illustrated that orally administered

folic acid is incorporated into the reduced folate pool in man and undergoes

scission at the C9-N10 bond to yield metabolites qualitatively similar to those

produced in rats. Both free pABG and acetylated pABG were detected in urine in

these studies.

It was hypothesized by Murphy et al. (1976) that both the oxidative

cleavage and N-acetylation occur intracellularly and that free pABG may be the

result of labile folates emerging from the cell and degrading in circulation or

during passage into the bladder. Murphy et al. (1976) speculated that within the

cell, folate is cleaved between the pteridine and the pABG moieties resulting in a

mixture of pteridines and p-aminobenzoylpolyglutamate. This latter compound is

recognized as an undesirable catabolite by the cell and is subsequently taken up

by the lysosomes and converted to pABG. Acetylation of pABG follows and this

compound is excreted into the urine. It was also hypothesized by Murphy et al.

(1976) that folate deficiency so often associated with pregnancy may result from

an increased catabolism of folate due to the presence of the fetus and not just from

the increased folate requirement of rapidly growing tissue (Murphy et al.1976 and

1978) or the increased excretion of the intact folate molecule (Fleming 1972,

Landon and Hytten, 1971).

In 1993, McNulty et al. conducted a study in rats to examine the possible

role of increased folate catabolism as a contributory factor to the folate deficiency








26
of pregnancy in rats. They utilized a newly developed high performance liquid

chromatography (HPLC) method allowing for the quantitative determination of

endogenous folate catabolites in rat urine. Three groups of female rats including

one nonpregnant and two pregnant groups were investigated. One of the pregnant

groups was allowed to feed freely whereas the other was pair-fed to the

nonpregnant control group. Sources of exogenous (dietary) folate were kept to a

minimum in order to limit excess serum folate concentrations, which could

potentially contribute to the folate catabolites formed exogenously (i.e. folate

break down in urine after excretion). The authors reported that pregnancy induces

an elevation in the rate of folate catabolism in rats, even in the presence of low

dietary folate. Pregnancy in rats caused a significant increase in urinary levels of

the folate catabolite apABG compared with nonpregnant animals. Between the

two pregnant groups, urinary apABG concentrations were consistently higher in

the ad lib fed group which suggested that a certain amount of the observed

elevation in folate break down was due to a greater food intake. However,

because apABG concentration peaked prior to a reduction in weight at parturition

for both pregnant groups (i.e. day 18 versus day 21, respectively), the observed

increase in apABG was determined to be a real effect of pregnancy. The peak

apABG concentration at day 18 was thought to reflect a change in placental

growth from hyperplasia to hypertrophy. DNA synthesis, a folate requiring

process, takes place during cell division (hyperplasia) and not during the increase

in cell size. The authors suggested that increased folate catabolism during








27
pregnancy was a consequence of increased generation of DHF, a labile folate

coenzyme, resulting from participation in DNA synthesis (i.e. deoxyuridylic acid

accepts one-carbon unit from 5,10-methylene producing thymidylic acid and

DHF) (McNulty et al. 1993). Later, this hypothesis was expanded to include the

increased generation of THF which is also a labile folate coenzyme resulting from

the increased participation of 10-formyl-THF in de novo purine synthesis

(McNulty et al. 1995). In contrast to apABG, urinary pABG concentrations were

not detectable in this investigation and appeared to be dependent on dietary folate

intake which was minimal. Again, it was reiterated that unacetylated pABG in

urine may arise from the cleavage of intact folates during collection or analysis or

may be the result of either incomplete acetylation of the catabolite in vivo or

deacetylation of apABG generated in vivo (McNulty et al. 1993). The authors

noted that pABG does not contribute a substantial amount to normal folate

catabolite excretion, nor does it vary during pregnancy. No significant

differences in urinary folate concentrations were found between the pregnant and

nonpregnant rats indicating that increased losses of intact folates through the urine

would not be contributing factors to folate deficiency. Overall, it was concluded

that pregnancy induces an elevation in the rate of folate catabolism and that this

may impose a drain on available folates, thus offering an explanation for folate

deficiency associated with pregnancy.

Following the rat study of McNulty et al. (1993), a similar approach

involving measurement of urinary catabolite excretion was used by McPartlin et








28
al. (1993) in human pregnant women. Six normal pregnant women provided 24 h

urine samples at one time point during each trimester and postpartum. Six

nonpregnant women of the same age were used as controls. The women entered a

metabolic ward for 42 h where they were provided with a defined diet containing

no exogenous source of pABG and a known folate concentration. The first 18 h

was considered a wash out period for the folate breakdown products arising from

normal diet (prior to entering metabolic ward) and urine was not collected.

Throughout the final 24 h, urine was collected and subsequently analyzed for

pABG and apABG by HPLC. Mean total 24 h excretion of apABG and pABG in

micrograms were converted to folate equivalents on the basis that the molecular

weight of either catabolite is approximately one-half intact folate. Urinary

apABG, the predominant catabolite, expressed as folate equivalents was

significantly greater in the second trimester compared to the first or third

trimester, postpartum or in the non-pregnant women. The change in the nature of

growth from hyperplasia to hypertrophy was provided as an explanation for these

findings. McPartlin et al. (1993) concluded that 24 h urinary concentrations of

folate catabolites reflected daily metabolic turnover of folate and should thus

indicate dietary requirement.

Folate Requirements and Recommended Dietary Allowances (RDAs)

Since 1941, RDAs have been prepared by the Food and Nutrition Board

and are intended to reflect the best scientific evaluation on nutrient allowances for

the maintenance of good health and to serve as the basis for evaluating the







29
adequacy of diets of groups of people (NRC 1989). Recommended dietary

allowances are defined as the levels of intake of essential nutrients that, on the

basis of scientific knowledge, are judged by the Food and Nutrition Board to be

adequate to meet the known nutrient needs of practically all healthy persons

(NRC 1989). Once the average minimal requirement for a nutrient has been

estimated, adjustments are made to account for bioavailability, individual

variation and the need for adequate reserves. Hence, a safety factor is

incorporated into the RDAs which, in theory, ensures that the requirements of

98% of the population will be covered. For many people, the RDA may exceed

their actual requirement by approximately 50%.

RDAs are based upon a variety of study designs. Depletion-repletion

studies determine nutrient requirements by measuring the amount of the vitamin

needed to bring a deficient individual to a normal status. Nutrient balance studies

measure nutrient status in relation to intake, and requirements are based upon the

level of the vitamin needed to maintain normal status. Biochemical indices are

often utilized in these study types. Requirements have also been estimated by

quantitating how much of a specific nutrient is consumed by free living healthy

populations who are maintaining normal nutrient status. The information derived

from such studies dictate how the RDAs are revised approximately every five

years.

The folate RDA was decreased by 50% or more for all the age-sex groups

in the 10th edition (NRC 1989). One rationale for reducing the RDA in








30
nonpregnant women was the observation that folate intake estimated by

population surveys including the Second National Health and Nutrition

Examination Survey (NHANES II) (LSRO 1984) was approximately 50% lower

than the RDA reported in the 9th edition of the RDAs (NRC 1980) and evidence

of widespread folate inadequacy was lacking (Senti and Pilch 1984). There are

limitations inherent in basing RDAs on estimated dietary folate intakes by normal

populations. First, underreporting is a recognized limitation and would result in

underestimation of requirements especially since actual intake may be twice as

high as reported intake. Second, existing food composition tables underestimate

the amount of folate contained in the food item. Recently modified folate

extraction procedures from food including double extractions, pH optimizations

and trienzyme treatments (Gregory et al. 1990, Martin et al. 1990, Tamura et al.

1997) improved folate recovery from the food matrix and yielded higher folate

values when assayed microbiologically. For example, an extraction buffer with a

pH of 7.85 was found to yield higher folate values than more acidic buffers

because it prevents the partial interconversion and degradation of folates

(Gregory et al. 1990, Wilson and Home 1984). Trienzyme treatment facilitates

retrieval of folates trapped in the food matrix and increases measurable folate

levels by 20% in whole wheat flour compared with conjugase digestion alone

(Martin et al. 1990).

The work of Sauberlich et al. (1987) was also used by the committee to

support a reduction in the folate RDA, although it appears that these data were








31
misinterpreted by the committee. Ten adult nonpregnant women followed a folate

depletion regimen for 28 d followed by a repletion period. Plasma folate levels

decreased throughout the depletion period and continued to fall until 200 glg/d

dietary folate was provided. The RDA committee concluded from these data that

the 200 J.g/d level satisfied the criteria for establishing an RDA. However, 200

pg/d represented the minimal dietary requirement for nonpregnant women and

had not been adjusted to account for individual variation and allowance for

storage. Sauberlich et al. (1987) suggested an intake of 300 pg/d was necessary

in order to provide an allowance for storage. In addition, the study was not

carried out long enough to determine the amount of folate required to achieve

acceptable serum or red cell folate concentrations.

In 1992, Bailey suggested that the new RDAs for folate may not provide

an adequate safety allowance for population groups at risk and should be

reevaluated as new information evolves. Recently, two studies have been

conducted which suggest that the current RDA for nonpregnant women is

inadequate. O'Keefe et al. (1995) investigated the response of nonpregnant

women of child-bearing age to three different levels of folate intake. Total

intakes of 200, 300 or 400 lg folate per day were provided as a combination of

dietary sources (30 jig) and either 170, 270 or 370 ag/d synthetic folic acid. The

consumption of 200 and 300 lg/d resulted in significantly lower mean serum

folate concentrations than the 400 pg/d. The authors reported that women

consuming folate intakes of 200 and 300 aLg/d were unable to maintain serum








32
folate concentrations in the acceptable range (>3 ng/mL) throughout the duration

of the 10 week study. In addition, plasma homocysteine concentrations changed

inversely to serum and erythrocyte folate levels with evidence of elevated levels

(>16 unol/L) in the subjects consuming the 200 pg/d level of intake. Based upon

these data, the investigators concluded that 400 CJg/d appears to more closely

fulfill the criteria of a dietary allowance.

Based on the urinary excretion of folate catabolites, apABG and pABG,

McPartlin et al. (1993) also demonstrated that 180 tg/d for nonpregnant women

was inadequate. Assay of 24 h urine collections for pABG and apABG revealed

that upon conversion of the catabolites to folate equivalents and addition of

urinary intact folate, the requirement to replace the catabolites and urinary

excretion of intact folate was 99 + 22 gg/d. Following adjustments for

bioavailability and the addition of two standard deviations, the RDA was

computed as 280 gg for nonpregnant women. Since McPartlin and associates

(1993) did not consider the other avenues of folate loss such as fecal excretion,

folate requirements were underestimated resulting in an underestimation of the

folate RDA. It is important to remember that urinary excretion of folate

catabolites is just one component of folate requirements.

Perhaps the most compelling reason to revisit the current RDA for folate

in nonpregnant women is the recent Public Health Service (PHS) recommendation

that all women of child bearing age consume 400 g.g/d of folate in order to reduce

their risk of having a pregnancy affected with spina bifida or other neural tube








33
defects (NTDs) (CDC 1992). This recommendation was based upon numerous

studies which convincingly illustrated the efficacy of 400 tg or more of folate in

reducing the risk of NTDs (Czeizel and Dudas 1992, MRC 1991, Werler et al.

1993). The protective role of folate in reducing both the occurrence and the

recurrence of NTDs has prompted the Department of Health and Human Services

and the Food and Drug Administration (FDA) to require the fortification of all

enriched grain products with folic acid as of January 1, 1998 (FDA 1996). It is

estimated that the consumption of enriched grain products will increase folate

intake by an average of 100 pg/d. Considering that the average diet provides an

estimated 200 jLg/day, improvement of food choices or supplementation may still

be warranted.

The current RDA of 200 pg/d for adult males also appears to be

inadequate based on the results of a recently conducted metabolic "depletion-

repletion" study by Jacobs et al. (1994). Plasma homocysteine concentrations

rose during folate depletion and correlated inversely with plasma folate

concentrations. During folate repletion, the current RDA for adult men, 200 tig

/d, was ineffective in normalizing high homocysteine concentrations resulting

from folate depletion. The investigators concluded that the current RDA for adult

men may not provide the expected margin of protection normally incorporated

upon derivatization.

Since the publication of the 1989 10th edition of the RDA for folate,

information has been compiled from effectively designed and controlled studies








34
that strongly suggests a need for revision for both the adult male and nonpregnant

female RDA. When the extra demands of pregnancy for folate are considered, the

adequacy of the current RDA of 400 ig/d is questionable. At this time,

metabolically controlled studies have not been conducted that investigate the

response of pregnant women to defined levels of folate intake. The basis for the

change in the RDA for folate from 800 lg to 400 pg/d during pregnancy stems

from investigations that provided either (a) known amounts of supplemental folic

acid to an uncontrolled diet or (b) provided folic acid to malnourished pregnant

women. Chanarin et al. (1968b) investigated folate requirements in 206 women

from approximately wk 15 to 38. The women were randomly assigned to receive

either an iron supplement or an iron supplement with 100 gg of folic acid. Red

cell and serum folates were determined by microbiological assay using

Lactobacillus casei (L. casei) as the test organism. At baseline, no differences in

serum or red cell folate concentrations existed between the two groups. However,

as the study progressed, red cell folate concentrations declined over time in those

not receiving 100 pig/d supplemental folic acid compared to maintenance of red

cell folate concentrations in those women receiving supplemental folic acid.

Dietary folate intake was estimated to be 676 tig/d based on the microbiological

quantitation of food folate from one hundred eleven separate 24 h food collections

which were duplicates of the actual meals consumed over 7 consecutive days

(range 198 pg to 1,615 Lg) (Chanarin et al. 1968a). The investigators concluded

that folate requirements in pregnancy were met by a total folate intake of 776








35
pg/d which included the supplement of 100 pg of folic acid. This study supports

the 1980 RDA (9th edition) for folic acid of 800 gg/d and was cited as a reference

by this RDA committee (NRC 1980). Bates et al. (1982) subsequently estimated

dietary intakes of 190 gg/d by this population group (United Kingdom). This

estimation in conjunction with Chanarin's report that 100 lg/d supplemental folic

acid plus dietary folate maintained red cell folate concentrations was cited by the

10th RDA committee in support of the 400 pig/d RDA.

Another study cited by the RDA committee for supporting the 400 Lg/d

level for pregnant women was the investigation conducted by Colman et al.

(1975) in which African women during their last 30 days of pregnancy consumed

maize meal fortified with folic acid for daily intakes of either 300 or 500 pg folic

acid. Changes in folate nutritional status (serum and red cell folate) were

compared to the changes observed in pregnant women who were receiving folic

acid as: (1) 300 pg in tablet form; (2) maize meal without folic acid; or (3) maize

meal containing 1000 pg. The last two comparison groups were from a previous

study (Colman et al. 1974) which indicated that folic acid-fortified maize meal

prevents the progression of folate depletion during pregnancy. In the current

investigation, daily intakes of maize meal containing 300 pg added folic acid

were effective in preventing the progression of folate depletion in late pregnancy.

The RDA committee assumed that the effective dose was 168 plg of synthetic

folic acid since it is generally agreed that food folate is approximately 50%

available. Based partly on the results of Colman's study, the Committee








36
contended that 400 pig/d during pregnancy would be sufficient to build or

maintain maternal folate stores and to keep pace with the increased folate needed

to support rapidly growing tissue. This study, however, was not designed to

assess folate requirements in pregnant women. It was conducted solely to

determine the smallest amount of folic acid that, when added to maize meal,

prevented folate deficiency in pregnancy. The authors did not state how much

endogenous dietary folate was consumed on a daily basis (i.e. folate content of

unfortified maize meal or other foods). This amount would have to be known in

order to use this study as a basis for establishing an RDA for pregnant women.

For example, if these women consumed 100 pg/d dietary folate from unfortified

maize meal then their actual intake increases from 300 to 400 pg/d. Individual

variation and allowance for storage would also have to be considered. Another

reason that this study is not suitable for basing an RDA for pregnant women is

that the population group studied was nutritionally compromised and not

representative of a normal healthy population. Although, maize meal fortified

with 300 pg folic acid plus low intakes of endogenous folate may be effective in

preventing the progression of folate depletion in late pregnancy in this population

group, it is unknown whether it would promote adequate folate nutriture in well-

nourished pregnant populations. The adequacy of maternal folate stores

throughout gestation is an important consideration due to increased transfer of

folate from maternal to fetal stores during the last weeks of pregnancy (Ek 1980,

Ek and Magnus 1981). Women with inadequate folate stores are more likely to







37
become deficient which may not become evident until the post-partum period.

Martinez (1980) found that the best predictor of a woman's red blood cell folate

concentration was the age of her youngest child.

Since the publication of the 10th edition of the RDAs for pregnant women,

the study conducted by McPartlin et al. (1993) not only refuted the current RDA

for nonpregnant women but also for pregnant women. They hypothesized that the

increased requirement for folate during pregnancy may relate to its accelerated

breakdown to apABG and its deacetylated derivative pABG due to folate's

participation in DNA biosynthesis. The results of their study supported their

hypothesis that folate catabolism is accelerated in pregnancy particularly during

the second trimester. The investigators were also able to clearly demonstrate that

the increased excretion of catabolites noted during the second trimester was

independent of the increase in glomerular filtration rate which increased around

the 8" wk and remained elevated throughout the entire pregnancy.

Recommended dietary allowances were computed by McPartlin et al. (1993) by

converting both catabolites to folate equivalents, adding the small quantity of

urinary intact folate and adjusting the total for bioavailability and population

variance. The estimated RDAs based on the urinary excretion of folate

catabolites and intact folates corresponded to 280 Lg/d (nonpregnant and 1s

trimester), 660 pg/d (2nd trimester) and 470 pg/d (3d trimester). Again the RDAs

have been underestimated as urinary excretion of folate catabolites is only one

component of folate requirements. Limitations to this study included the fact that








38
dietary intake was not controlled between hospital admissions and the 18 h "wash

out" period may not have been sufficient for eliminating exogenous pABG arising

from previous uncontrolled dietary folate intake. It is possible that the breakdown

products are retained by the body for a longer duration than that allotted by the

investigators, and if so, the catabolites found in the urine may not represent an

endogenous source. The absorption and fate of folate catabolites derived from

food remains to be determined. Furthermore, McPartlin and his coworkers

assessed urinary excretion of catabolites at one time point only for each trimester,

and data on other indices of folate status were not reported. Another possible

limitation of this study is the inclusion of unacetylated pABG along with the

major catabolite, apABG. However, until the origin of pABG is known, it seems

prudent to include this catabolite in the calculations. McPartlin and coworkers

analyzed each catabolite separately, so as information becomes available,

appropriate adjustments can be made to their interpretation of the data if

necessary.

The importance of adequate folate nutriture and appropriate folate intake

during pregnancy is widely recognized, yet the data we have for basing

recommendations regarding folate requirement during pregnancy, is plagued with

limitations. A controlled metabolic study which investigates the response of

pregnant women to defined levels of folate intake utilizing a variety of folate

indices to assess folate status response is long overdue.








39
Folate Deficiency and Pregnancy

Since folate governs the synthesis of the precursors of DNA and RNA, the

formation and development of every human cell is dependent upon an adequate

supply of this vitamin. Pregnancy represents a state of rapid cell division

involving both maternal and fetal tissue. In a relatively short period, a two celled

embryo develops into a two billion celled fetus (Hibbard 1964). This occurs

simultaneously with the rapid proliferation of various maternal tissues including

the uterus and the hematopoietic system (Hibbard 1964). The increased demand

that pregnancy imposes on folate nutriture is reflected in the doubling of the RDA

from nonpregnant conditions. Cooper et al. (1970) concluded that 500-600 ig/d

of dietary folate was required to prevent all evidence of folate deficiency in

pregnancy and many women can not meet the increased requirement for folate

through diet alone. Poor nutrition has been recognized as a major factor in the

development of folic acid deficiency since the 1960s (Stone et al. 1967). The

decline in folate status (both serum and red cell folate) during pregnancy has been

well-documented in both developed (Bailey et al. 1980, Cooper et al. 1970, Ek et

al. 1981, Lowenstein et al. 1966, Qvist et al. 1986) and underdeveloped countries

(Colman et al. 1975). Bailey et al. (1980) conducted a study to investigate folate

status at the initiation of prenatal care in 269 pregnant low-income subjects. They

found that serum folate was low (3 to 6 ng/mL) and deficient (< 3 ng/mL) in 48%

and 15% of the subjects, respectively and 29% of the subjects had subnormal

(<140 ng/mL) red cell folate concentrations. Qvist et al. (1986) reported








40
decreasing plasma folate levels with advancing pregnancy so that, at term, almost

50% of the women had subnormal plasma folate concentrations which persisted

over the following 2 months. The investigators also found that 10% of the

women had subnormal red cell folate concentrations at term which increased to

33% during the post-partum period. The authors contended that this observation

could be explained by the delayed reflection of folate stores by the red cell folate

values in periods of increased demand (Qvist et al. 1986). Ek and Magnus

(1981) observed similar folate status changes throughout gestation in 43 women.

They found a positive correlation between plasma folate and red cell folate

concentrations at different stages of pregnancy supporting the delayed reflection

of folate stores by red cell folate concentrations (Ek and Magnus 1981).

The problems associated with folic acid deficiency have been recognized

since the late 1950s. The most overt sign of folate deficiency is megaloblastic

anemia which is most commonly observed in the late stages of pregnancy. Earlier

indications (preanemia) include serum folate concentrations below 3 ng/mL and

red cell folate concentrations below 140 ng/mL. Folate deficiency has been

associated with low birth weight, cessation of pregnancy, delayed maturation of

the nervous system, growth retardation, megaloblastic anemia and congenital

anomalies (Hibbard 1964, Shojania 1964, Strelling et al. 1966, Baumslag et al.

1970, Iyengar and Rajalakshmi 1975, Smithells et al. 1976). Recently, O'Scholl

et al. (1996) conducted a prospective study examining the influence of folate

intake on two pregnancy outcomes associated with poverty and disadvantage:








41
preterm delivery and infant low birth weight (LBW). The investigators reported

that women with a low mean daily folate intake (5 240 gg/d) had an

approximately twofold greater risk of preterm delivery and incidence of infant

low birth weight after controlling for maternal characteristics, energy intake and

other related nutrients. In addition, lower concentrations of serum folate at wk 28

were also associated with a greater risk of preterm delivery and low birth weight.

Although overt megaloblastic anemia is infrequent in the United States, it seems

desirable to consume enough folate to maintain maternal stores to keep pace with

the increased demand for folate during pregnancy.

Folate Supplementation During Pregnancy

Several studies have been conducted to determine the amount of

supplemental folic acid required to maintain normal folate status during

pregnancy (Cooper et al. 1970, Dawson 1966, Hansen and Rybo, 1960,

Lowenstein et al. 1966, Willoughby and Jewel 1966). Unfortunately, dietary

folate intake was not controlled and interpretation of these data is difficult.

Willoughby (1967) and Willoughby and Jewel (1968) reported that 300-350 gg/d

synthetic folic acid in addition to diet was able to maintain normal folate status in

pregnant women throughout gestation, whereas supplemental amounts of either

100 or 200 pg/d were inadequate. Dawson (1966) also reported that 150 gg/d

folic acid plus dietary folate was insufficient to maintain normal serum folate

concentrations. In contrast, Chanarin et al. (1968b) reported that 100 pIg/d

supplemental folic acid, in addition to diet, was adequate. Lowenstein et al.








42
(1966) found that consumption of 500 pg/d synthetic folic acid plus diet resulted

in higher serum folate concentrations in pregnant women compared to

nonpregnant controls (11.0, 7.0 ng/mL, respectively) and therefore may be above

the minimal daily requirement. Cooper et al. (1970) concluded that the decline in

folate status during pregnancy could be prevented with 200-300 plg folic acid/day

in addition to diet. Overall, these data support a supplementation level of

approximately 300 pg/d.

Programs are in place to aid pregnant women in obtaining a sufficient

amount of folate. Unfortunately, not enough is known about folate metabolism

and requirements during pregnancy to provide women with the optimal amount,

and problems arising with oversupplementation of folate have been raised. The

biggest reluctance to give prophylactic folic acid during pregnancy is due to the

possible risk of masking a developing B12 deficiency. However, since the

incidence of pernicious anemia is low among women of child-bearing age, the

masking of a B12 deficiency leading to neurological complications would be

extremely rare (Tamura 1995). Another concern is the possibility of interfering

with zinc absorption. In 1984, Mukherjee et al. conducted a study involving 450

pregnant women and reported that high maternal plasma folate concentrations

during pregnancy were associated with an increased rate of pregnancy

complications and fetal distress They reasoned that folic acid supplementation

adversely affected the intestinal absorption of zinc which resulted in the

complications. In contrast, Tamura et al. (1992) found high serum folate








43
concentrations were associated with favorable pregnancy outcome and reported

higher birth weight and Apgar scores of newborns, reduced prevalence of fetal-

growth retardation, and lower incidence of maternal infection around delivery.

Recently, Kauwell et al. (1995) conducted a study to determine the effect of

supplemental folate on zinc nutriture and found that folic acid supplementation

did not adversely affect zinc status.

Overall, the detrimental effects of folic acid supplementation on

pregnancy outcome are speculative considering supplementation has been in use

for the past 40 y with no identifiable adverse effects. However, because the

effects of oversupplementation with folic acid on the developing fetus are

unknown (Scott et al. 1991) and the fetus has demonstrated its ability to sequester

folate from maternal circulation, high supplemental doses of folic acid (> Img)

should be avoided under normal circumstances.

Neural Tube Defects

Neural tube defects (NTDs) are a group of serious malformations which

affect the developing nervous system and include anencephaly, spina bifida and

encephalocele. Approximately 2,500 babies with these abnormalities are born

each year in the US which represents approximately 1 to 2 NTD affected infants

per 1000 births; many more are aborted either spontaneously and electively

(Crandall et al. 1995). Neural tube defects arise within one month of conception

and over 90% are born to women without a prior family history of this disorder

(Crandall et al. 1995). Although the cause of NTDs is unknown, it is now well








44
established that maternal folic acid supplementation during the periconceptual

period significantly reduces the risk of giving birth to a baby with a NTD.

Several studies have confirmed that for women without a prior history of NTDs, a

60 to 100% decrease in risk was obtained if folic acid was taken prior to

conception and throughout the first trimester (Milunsky et al.1989, MRC 1991,

Mulinare et al. 1988). Studies which have investigated the effect of folic acid

supplementation on the recurrence of NTDs have also reported similar reductions

after folic acid supplementation (MRC 1991, Laurence et al. 1981, Smithells et al.

1983, Vergel et al. 1990). The effective dose of folic acid ranged between 0.4 mg

to 5 mg/d. The neural tube closure occurs within 28 d after conception, a time

when many women do not even know they are pregnant. Hence, NTD prevention

is best achieved by adequate folic acid intake throughout the reproductive years

(Crandall et al. 1995). In 1992, the Public Health Service recommended that all

women of childbearing age in the United States who are capable of becoming

pregnant should consume 0.4 mg of folic acid per day for the purpose of reducing

their risk of having a pregnancy affected with an NTD (CDC 1992). Daly et al.

(1995) subsequently demonstrated that red cell folate concentrations of 400

ng/mL or higher were associated with a low risk of folate-responsive NTDs.

Brown et al. (1997) found that red cell folate concentrations > 400 ng/mL could

be achieved by folate intakes of at least 450 .Lg/d (supplement users) or 500 pg/d

(food and folic acid fortified cereals, only). A limitation to the study of Brown et

al. was that dietary estimates of folate consumption were obtained during the








45
study period and not 2-3 months earlier during folate incorporation into

reticulocytes.

The average consumption of dietary folate by women in the United States

is approximately 200 pg/d (National Academy of Sciences 1990). It is possible

for women to increase their consumption of dietary folates by increasing their

consumption of fruits, vegetables and whole grains. Unfortunately, for various

reasons, adherence to dietary guidelines is not widespread. Fortification of staple

foods stuffs with folic acid is an effective and inexpensive approach to assist all

women in increasing their daily intake of folic acid. Recently, the FDA amended

the standards of identity for enriched cereal-grain products and has given

manufacturers two years to comply with its decision requiring the fortification of

enriched grain products with folic acid (FDA 1996). The level of folic acid

fortification permitted (140 Lg per 100 g of cereal grain product) is low enough to

keep daily folate intake below 1 mg yet high enough to enhance folate intake by

an average of 100 jig/day.

As stated before, the cause of NTDs is unknown at this time. Davis et

al.(1995) found that the absorption of folic acid routinely consumed in

supplements and fortified food products was not impaired in women with a

history of a pregnancy with a NTD. Recent evidence suggests impairment of

homocysteine remethylation resulting in a possible methionine shortage at a

crucial stage of development (Mills et al. 1995). The investigators suspect a

defect in methionine synthase, the enzyme responsible for converting








46
homocysteine to methionine. Methionine synthase requires both the presence of

folate (5-methyl-THF) and vitamin B12. It is a key enzyme in many methylation

reactions, including the production of myelin, and it also catalyzes the synthesis

of THF, a compound required for DNA synthesis, following its conversion to

5,10-methylene-THF. Further, homocysteine, itself, is reportedly a putative

embryotoxin (Lucock et al. 1995). An earlier study conducted by Kirke et al.

(1993) found that both plasma folate and plasma B12 were independent risk

factors for NTDs, and they suggested that vitamin B12 should be considered, as

well as folate, in any supplementation program aimed at preventing NTDs. In

lieu of the recent discovery of elevated homocysteine concentrations and the

possibility of impaired homocysteine metabolism, supplementation with B12, in

addition to folic acid, may result in a further reduction of NTDs. However, the

findings of van der Put et al. (1995) of a mutation in the gene coding for

methylenetetrahydrofolate reductase (MTHFR) as a risk factor for spina bifida

would support folic acid supplementation alone. Reduced MTHFR activity would

lead to lower 5-methyl-THF concentrations which would result in elevated

homocysteine concentrations. However, this is a common gene mutation in the

general population and is not a definitive marker for those at risk of delivering a

baby with a NTD.

An association between NTDs and obesity, independent of the effects of

folic acid intake, has also been reported (Werler et al. 1996). These investigators

collected data from 1988 to 1994 in a case-control surveillance program of birth







47
defects and found a positive association between prepregnant weight and NTD

risk which was not alleviated by adequate folate intake among the heavier

women. The authors noted that the mechanism by which increased weight may

affect NTD risk is not clear but speculated that their diets may be deficient in

nutrients (other than folate) that may be necessary for closing of the neural tube or

that obesity affects some metabolic factors which may have a direct effect on

neural tube development (i.e. insulin).

Assessment of Folate Status

Assessment of folate nutriture during pregnancy is complicated by the

physiological changes that occur over this nine month period. A healthy woman

bearing a normal-sized fetus will increase her plasma volume from 25% to 80%

with an average value of 40% (Bartels et al. 1989). Plasma volume begins to rise

during the latter part of the first trimester, reaches a maximum at approximately

25 wk gestation and remains relatively constant until the 34th wk of gestation

(Bartels et al. 1989). As a result of this hemodilution, the concentration of several

nutrients, including folate, decline. Unfortunately, little is known as to what

constitutes normal plasma folate concentrations in pregnancy. As O'Connor

(1994) emphasized in a recent review, it is difficult to discern the meaning of a

low serum folate concentration during pregnancy secondary to the physiological

changes that occur during this time and a reduction in serum folate may be

normal. Hall et al. (1976) found that the fall in serum folate concentration was

due principally to plasma volume expansion and should therefore be regarded as








48
physiological. However, women consuming supplemental folic acid (-300 gg)

plus low dietary folate are able to maintain serum folate concentrations within the

normal range for nonpregnant women (Cooper et al. 1970, Willoughby 1967,

Willoughby and Jewel 1968). Currently we can only evaluate the nutriture of

pregnant women against nonpregnant standards and based on the critical role

folate plays in producing a healthy baby and the link between low serum folate

concentrations and folate deficiency, it seems desirable to maintain serum folate

concentrations in the range deemed appropriate for nonpregnant women.

Altered renal function is another physiological change which makes

assessing folate status during pregnancy a challenge. In the non-pregnant state,

the renal tubules are very effective in reabsorbing water-soluble vitamins like

folate. During pregnancy, however, Landon and Hytten (1972) reported that

urinary folate excretion increases approximately four fold.

Several folate status indices are available to clinicians and researchers to

obtain information on folate nutriture in both pregnant women and the general

population. Importantly, all have certain limitations which must be considered.

Determinations of serum and erythrocyte folate concentrations are commonly

performed and represent the most practical approach for population studies. The

main folate derivative in serum is the reduced, monoglutamate form, 5-methyl-

THF (Gibson 1990). Serum does not contain polyglutamates. Serum folate

concentration is reflective of acute folate status that changes in response to current

dietary intake. For example, serum folate values increase rapidly after the








49
ingestion of folate (dietary or supplemental) and decrease rapidly on a folate

deficient diet. Hence, serum folate concentration is not a valid index of long term

folate status or tissue stores unless it is used in conjunction with controlled dietary

intake over time. Furthermore, normal serum folate concentrations have been

documented in individuals with known folate deficiency and low levels have been

found in individuals with normal status (Lindenbaum and Allen 1995). Several

non-nutritional factors can also affect serum folate concentrations. An acute drop

in serum folate results after alcohol ingestion because reabsorption of folate by

the kidney is impaired (Hillman et al. 1977). Smoking has also been shown to

lower serum folate concentration possibly by enhancing erythropoiesis, which in

turn, increases folate requirements (Gibson 1990). An elevated serum folate

concentration may result from hemolysis of red blood cells occurring during

phlebotomy. Stagnant-loop syndrome, acute renal failure and liver damage may

also result in elevated serum folate levels (Gibson 1990). Serum folate values

below 3.0 ng/mL are considered to reflect deficiency; values between 3.0 6.0

ng/mL reflect borderline deficiency; while values above 6.0 ng/mL are considered

to be within an acceptable range (6.0- 20 ng/mL).

In contrast to serum folate concentration, folate within the erythrocyte is

not influenced by hemodilution (Bailey 1990) and therefore, provides a better

measure of folate nutriture during pregnancy. However, it is important to

remember that the red cell folate concentrations reflect folate available for

incorporation in the red cell at the time of erythropoiesis and, therefore, represents








50
folate status 2 to 3 months prior to the time of blood sampling. Red cell folate

concentrations are less sensitive to short-term fluctuations in folate status than

serum folate levels and are therefore, a good indicator of long term folate status.

Because changes in erythrocyte folate concentrations take several months to

reveal any significant changes in folate nutriture, they cannot be used to assess

recent changes in folate nutriture. Red cell folate concentrations are a more

reliable index of folate status than serum folate concentrations and are the

preferred biochemical measurement for population studies. A major limitation,

however, of erythrocyte folate levels is they cannot be used to distinguish

between megaloblastic anemia due to folate deficiency and that due to a B2

deficiency or a combination deficiency. However, low folate levels found both

in the red cell and serum or plasma is highly suggestive of a folate deficiency

(Sauberlich 1995). Red cell folate concentrations below 140 ng/mL are indicative

of folate deficiency. Values between 140 160 ng/mL are borderline while values

above 160 ng/mL are within acceptable range.

Serum and erythrocyte folate concentrations can be assayed using either

microbiological or radioisotope dilution techniques. The microbiological assay is

considered to be the gold standard and is widely accepted for assaying folate

concentrations in biological samples including blood and urine, as well as folate

concentrations in tissues and food. The test organism most commonly used is L.

casei because it responds to the greatest number of different folate derivatives,

including those with up to three L-glutamic acid residues (Herbert, 1987b). Other








51
microorganisms that have been used include Streptococcusfaecium which does

not respond to 5-methyl-THF and Pedicococcus cerevisiae which does not

respond to either folic acid or 5-methyl-THF (Tamura 1990). A major limitation

to the microbiological assay is that it cannot be used for serum or erythrocyte

samples containing antibiotics or methotrexate because these substances inhibit

bacterial growth. Since the polyglutamate form of folate is present in

erythrocytes, an incubation period is required to ensure complete hydrolysis to the

mono, di or tri glutamate form by the conjugase enzyme. Ascorbic acid is added

to biological samples to act as a reducing agent and prevent the oxidation of the

folate. Pfeiffer and Gregory (1996) recently determined that the pH of the

biological sample is important during the incubation period since the human

serum pteroylpolyglutamate hydrolase (conjugase) enzyme functions optimally at

pH 4.5 (Krungkrai 1987). For this reason ascorbic acid should be used as the

preserving agent in the red blood cell hemolysate as it yields a final pH of

approximately 4.0 rather than the sodium ascorbate which yields a higher pH

(Pfeiffer and Gregory 1996).

The radioisotope dilution technique is not as widely accepted as the

microbiological technique but is used frequently in the clinical setting because it

is fast and is not affected by antibiotics, tranquilizers, and or antimitotic agents

which can interfere with the growth of L. casei (Gibson 1990). This technique

uses a folate binding protein derived from milk and ['2I]-labeled folic acid or

methyl-tetrahydrofolate. The unlabeled folate present in the biological sample








52
competes with the labeled folate for available binding sites on the folate binding

protein. Charcoal is used to adsorb both the unbound labeled and unlabeled folate

which is removed later by centrifugation. A scintillation counter is used to

measure the amount of bound labeled folate existing in the supernatant. The

decrease in the radioactivity is proportional to the folate concentration in the

sample. High performance liquid chromatography techniques have also been

developed for the assessment of folate in biological samples (Gregory et al. 1984,

Selhub 1990).

Formiminoglutamate (FIGLU) excretion is a functional index of folate

nutriture. Normally histidine is converted to glutamic acid via the action of

formimino transferase and its coenzyme tetrahydrofolate. Because folate is

required in this reaction, a folate deficiency will result in the accumulation and

excretion of FIGLU in the urine particularly after a histidine load. Although the

FIGLU test is highly sensitive, it is not specific for folate. Deficiency of vitamin

B12 or the enzyme formimino transferase will also produce high urinary excretion

of FIGLU after a histidine load. The use of the FIGLU excretion test in pregnant

women is controversial. Rothman (1970) reviewed several studies showing

increased FIGLU excretion in early pregnancy with a progressive decrease in the

level toward term and a second increase post partum. Hence the use of the FIGLU

excretion as a functional index of folate status during pregnancy is not

widespread.








53
Polymorphonuclear leukocyte lobe count is also utilized as a functional

index. Neutrophils normally have three to four segments. In megaloblastic

anemia hypersegmentation occurs increasing the number of lobes and is a

characteristic early feature of folate and vitamin B12 deficiency (Colman 1981).

Neutrophil segmentation also occurs under other conditions and is thus not very

specific. However, for persons who may be both iron and folate deficient,

evaluation of neutrophil hypersegmentation is particularly helpful in diagnosing

the folate deficiency that is being masked by the co-existing iron deficiency (Das

et al. 1978).

Deoxyuridine suppression test (dUST) is yet another way to assess

functional folate status. This is an in vitro test performed on short term cultures of

proliferating cells. The test is based on the need for folate in the methylation of

deoxyuridine to thymidine. Bone marrow cells, lymphocytes or whole blood are

preincubated with unlabeled deoxyuridine. Tritiated thymidine is then added to

the cells and if folate deficiency exists, [3H]thymidine will not get incorporated

into DNA. Sauberlich et al. (1987) found that this test did not correlate well with

either plasma or red cell folate concentrations of non-pregnant women.

Furthermore, the use of this test has never been validated using pregnant or

lactating women (O'Connor 1994).

Determining plasma homocysteine concentration is gaining momentum as

a functional index of folate status. Plasma homocysteine concentration is

inversely correlated with serum folate and red cell folate concentrations and has








54
proven to be a sensitive, although not specific, index of folate status because

plasma homocysteine concentrations are also elevated with a vitamin B6 or B12

deficiency.

Stable-Isotope Methods

Isotopes are forms of an element that have the same number of protons

(atomic number), but which differ in the number of neutrons and thus atomic

weight. Because an atom's chemical characteristics are defined by its atomic

number, isotopes behave similarly. Stable isotopes do not undergo nuclear decay

and are therefore, nonradioactive. Both stable and radioisotopes act as tracers in

that they follow or outline the pathway or fate of a tracee. In order to be an

effective tracer, the isotopically labeled compound must act the same as the

nonlabeled species in the system of interest. Gregory et al. (1990) conducted a

study to determine whether stable-isotope-labeled folates could be employed for

studies of folate absorption and metabolism in human subjects. After a saturation

regimen, eleven adult males consumed 677 nmol (300 pg) each of 3',5'-labeled

bideuterofolic acid and glutamate labeled tetradeutero folic acid. The molar ratio

of labeled folates in the urine were compared to the molar ratio in the ingested

dose and found to be similar (not significantly different). In addition, the

percentage of labeled folate in the urine over a forty-eighty hour period

approached the percentage of the ingested labeled folate. These data revealed that

the two labeled forms of folate behaved similarly in vivo when compared to each

other and when compared to unlabeled folate.








55
The interest in utilizing stable isotopes as tracers resurfaced in the 1970s

primarily due to the interfacing of gas chromatography with mass spectrometry.

The power of gas chromatography as a separation technique made it possible to

analyze biological samples in complicated matrixes. In order to analyze a sample

by gas chromatography mass spectrometry (GCMS), the sample must be volatile

and heat stable. For this reason organic compounds are often derivatized. Prior to

derivatization, folate is cleaved and the pABG fraction which contains the label is

isolated. Trifluoroacetic anhydride and trifluoroethanol are added forming a

lactam of (N-trifluoroacetyl-a-trifluoroethyl)-p-aminobenzoylglutamate as the

major product (Toth and Gregory 1988). This compound is well suited for GCMS

analysis because it is volatile and stable. The isotope ratios (i.e., D2 to DO) or the

enrichment of the sample (refers to the abundance of the label in excess of natural

abundance) can then be determined in the biological sample.

There are several advantages to using stable isotopes with safety being the

most important one. The only possible toxicity that may result from substitution

of a stable isotope for its lighter and more abundant isotope is an "isotope effect"

which is a slowing of biochemical reactions because of the greater mass of the

stable isotope (Van Langenhove 1986). Browne (1986) contended that concern

about toxicity from stable-isotope labels should not limit the use of stable-isotope-

labeled drugs, because there is now abundant evidence that the amount of stable-

isotope label (including deuterium) that can be accumulated with repeated

administration of even large amounts of drug does not pose any risk to humans.








56
Importantly, due to the increased sensitivity and specificity that GCMS analysis

provides, only minimal amounts of tracers are required and they can be

chronically administered to the same subject. In addition, a molecule can be

labeled with more than one different isotope and each one can be detected by

GCMS. This concept enables us to study the bioavailability of pABG labeled

with ['3C] at the same time we are investigating [2H2]folic acid turnover.

However, the availability of commercially labeled stable isotopes is limited, so

synthesis is a major consideration (Gregory and Toth 1990).














CHAPTER 3
STUDY DESIGN AND METHODS


Study Design

A two-by-two factorial study design was used in which pregnant women

(n=12) and nonpregnant controls (n= 12) were randomly assigned to consume

folate intakes of either 450 ig/d or 850 lg/d for 84 d. These intakes

approximated the current and former RDA for pregnant women. Four

experimental groups (n=6) were thus established: pregnant subjects fed either 450

or 850 pg/d and nonpregnant subjects fed the same two levels. Subjects

consumed folate as a combination of food folate (120 15 tg/d) and synthetic

folic acid (either 330 or 730 [tg/d) taken with meals. Approximately fifteen

percent of the total folate intake was provided as deuterated folic acid (D2) during

the first one-half of the study then switched to 100% unlabeled folic acid (DO) at

the beginning of wk 7 allowing for future investigation of folate kinetics. In

addition, 4 subjects (one from each group) consumed ['3C]pABG at the beginning

of wk 7 to provide information as to the fate of exogenous pABG. Blood and

urine were collected at baseline and thereafter on a weekly basis for 12 wk (wk

14-25 gestation for pregnant subjects) except during wk 7 and 8 when multiple

collections were obtained. Folate status was assessed by measurement of serum


57








58
folate, red cell folate, urinary 5-methyl-THF, urinary folic acid and urinary pABG

and apABG. After completion of the study, a subsample of our subjects

consumed either 200 p.g/d (4 pregnant; 3 nonpregnant) or 600 pg/d (4 pregnant; 4

nonpregnant) synthetic folic acid plus uncontrolled dietary folate and folate and

over the following 3 months returned to the clinical research center for monthly

blood draws. See Figure 4 for an overview of the protocol.




1 2
CONTROLLED INTAKE Follow-Up
N P I 0 1 2 3 4 5 6 7 8 9 10 11 12 1 17 20 24
P (w k gestation) 13 14 15 16 17 18 19 20 21 22 23 24 25 29 33 37
Daily FA
Daily FA 3 D2 --- DO ------ DO--
Supplement

[3 C ]p A B G
*
B lood
B **Q********* Q ***


24-h Urine C6 0 3 E 0 D C
1:1 M =






Figure 4. Overview of Protocol. 'Controlled dietary folate (120 gg/d);
Uncontrolled dietary intake; Two levels of supplemental folic acid (330 and 730
gg/d) during the 12 wk metabolic study and two levels (200 and 600 Rg/d) during
the follow-up study; 4Administration of 3C-pABG to 4 subjects; 5One blood
draw/wk except following the switch to unlabeled folic acid then six collections
over 2 wk and except during follow-up period then 1 draw/mo; and 6One 24 h
urine collection/wk except following switch to unlabeled folic acid then seven
collections over 2 wk. NP = Non Pregnant; P = Pregnant








59


Subjects

Healthy pregnant female subjects (18-35 y, 14 wk gestation) and

nonpregnant controls (18-35 y) with normal blood chemistry profiles, normal

blood folate concentrations and normal health status as determined by medical

histories were eligible for participation. Exclusion criteria included chronic drug

(including oral contraceptives and folate antagonists), alcohol or tobacco use.

The majority of pregnant subjects (n=10) were consuming prenatal vitamins

containing folic acid (0.4 -1 mg) prior to starting the study. Gestational age in

pregnant subjects was determined by sonogram in conjunction with the date of

last menstrual period. Approval of the study protocol by the Institutional Review

Board of the University of Florida and signed informed consents by participants

were obtained. Compliance to the study protocol was ensured by close daily

contact in a positive environment monitored by the research team who observed

consumption of folic acid with meals. Nonpregnant women maintained their

body weight within 5% of base line throughout the study period and pregnant

women gained approximately 0.45 kg per week.

Diet

A 5-d menu cycle consisting of five dinners and three breakfasts and

lunches was designed as detailed in Table 1. Conventional foods were selected to

provide meals that were varied and palatable. Folate content was minimized by

thrice boiling chicken, ground beef, green beans and white potatoes and by using








60
Table 1. Five-day cycle menus.

Day 1 Day 2 Day 3 Day 4 Day 5

Breakfast

Waffles Shredded Wheat Biscuits Waffles Shredded Wheat
Margarine Milk Margarine Margarine Milk
Syrup Raisins Jelly Syrup Raisins
Cranberry juice Brown sugar Canned pears Cranberry juice Brown sugar
Apple Sauce Cranberry juice Apple Sauce
Cranberry juice Cranberry juice

Lunch

Sandwich, Sandwich, Crackers Sandwich, Sandwich,
Pita Biscuit Tuna Pita Biscuit
Turkey breast Ham Mayonnaise Turkey breast Ham
Cheese Cheese Raisins Cheese Cheese
Mayonnaise Mustard Chocolate chip Mayonnaise Mustard
Canned pears Canned peaches Cookies Canned pears Canned peaches
Granola bar Apple newtons Cola Granola Bar Apple newtons

Dinner

Casserole, Turkey Casserole, Macaroni and Chicken breast
Chicken Rice Ground beef Cheese Barbeque sauce
Rice Gravy Onion Turkey Ham Potato
Cheese soup Green beans Cream of Green beans Margarine
Margarine Margarine mushroom soup Margarine Green Beans
Green beans Icecream Margarine Ginger Snaps Canned pears
Newtons Chocolate syrup Green beans Devil's food
Apple juice Apple juice Frozen juice Cookies
flavored bar Apple juice

Snacks
Oreo cookies Oatmeal cookies Graham crackers Granola bar Rice cakes
Milk Milk Milk Milk Milk
Yogurt Yogurt Yogurt Yogurt Yogurt
Popcorn Popcorn Popcorn Popcorn Popcorn








61
canned fruits and vegetables and refined starches. The menus were analyzed for

their folate content in our laboratory and provided a mean of 120 15 pg/d.

Energy and all other nutrients were analyzed by the Minnesota Nutrient Data

System version 2.7 (Nutrition Coordinating Center 1994). Table 2 includes the

nutrient composition of the three menus. The menu cycle provided z 2500

calories/d of which z 60% came from carbohydrate, 25% from fat and 15% from

protein. A custom formulated supplement (Tischon Corp., Westbury, NJ) was

used to provide the RDA for all essential nutrients not met by the diet except

folate. Loss of water soluble vitamins/electrolytes in the boiled food items was

accounted for in the supplement. Caloric energy was modified to maintain weight

or ensure weight gain in nonpregnant and pregnant subjects, respectively.

Examples of non-nutritive food sources included margarine, candy, Jello, Cool-

Whip and sweetened or unsweetened beverages.

Table 2. Nutrient composition of 5-d cycle menus" 2

Day 1 Day 2 Day 3 Day 4 Day 5

Energy, kcal 2540 2791 2472 2603 2490

Protein, g 83 85 85 78 88

Fat, g 60 65 81 74 50

Folate, pg 114.6 134.7 99.2 118.2 135.1
1Supplementation for pregnant subjects was provided by Tischon Corporation (Westbury, New
York) -0.4 mg thiamin mononitrate, 18 mg iron fumarate, 4.5 mg zinc sulfate, 230 RE retinyl
palmitate, 3 TE d-a tocopheryl acetate, 120 IU Vitamin D3, 7 mg niacinamide, 1.5 mg pyridoxine
HCI, 16 lg phylloquinone.
2Supplementation for nonpregnant controls was provided by Tischon Corporation (Westbury,
New York)- 3 mg iron fumarate, 1.5 mg zinc sulfate, 230 RE retinyl palmitate, 120 IU Vitamin
D3, 5 mg niacinamide, 0.9 mg pyridoxine HCI, 14 pg phylloquinone.








62
Folate Extraction from Menus

The three day cycle breakfast and lunch menus and the five day cycle

dinner menus were analyzed for folate content three separate times during the

study period using the method described by Gregory et al. (1990) and Martin et al.

(1990). Each menu was prepared as though it was being consumed by the

subjects. The food items were weighed, blenderized and aliquoted into 10 mL

tubes. The blenderized samples were protected from light and stored at -200 C

until analyzed for folate content. Six grams of blenderized menu were extracted

in 20 mL boiling buffer containing 50 mM Hepes/50 mM Ches, pH 7.85 with 2%

ascorbate and 0.2 M 2-mercaptoethanol. The buffer containing the sample was

placed in a boiling water bath for 10 minutes and then cooled on ice for 10

minutes. Following homogenization using a Polytron (Type P100/35, Brinkman

Instruments, Westbury, NY) at a setting of 5 for 1 minute, rat plasma conjugase

(500 p) was added to the homogenate to deconjugate polyglutamyl folates. Rat

plasma conjugase was obtained from Pel-Freeze (Rogers, AR) and dialyzed for 24

h through 12,000-13,000 mw tubing in cold 50 mM Hepes/50 mM Ches, pH 7.85

buffer containing 2% (w/v) sodium ascorbate, 10 mM 2-mercaptoethanol, and 2

teaspoons of acid washed charcoal. The conjugase was stored at -85 OC in 1.5 mL

polypropylene microcentrifuge tubes (Fisher Scientific, Pittsburgh, PA). Alpha-

amylase (20 mg/mL) (Sigma Chemical Co., St. Louis, MO, No. A-0273) was also

added at this time to aid in the release of folate from the carbohydrate matrix. The

samples incubated for 4 h at 37 C, after which protease (2 mg/mL) (Sigma








63
Chemical Co., St. Louis, MO, No. #P-5147) was added to release any folate in the

protein matrix of the food. This was followed by an additional 1 h incubation

period at 37 C. The samples were next placed in a boiling water bath for 5

minutes to inactivate the enzymes and cooled on ice for 10 minutes. Samples

were centrifuged at 15,000 rpm, 5 C, for 25 minutes using a Sorvall RC-5

Superspeed refrigerated centrifuge (DuPont Instruments, Newton, CT). The

supernatant was filtered through Whatman #1 paper and kept on ice in a

graduated cylinder. The pellet was resuspended in 5 mL Hepes-Ches buffer,

homogenized and centrifuged again at 15,000 rpm, 40 C, for 25 minutes. The

supernatant was then filtered through Whatman #1 paper and added to the

previous filtrate. The graduated cylinders were filled to 30 mL with Hepes-Ches

buffer and 1 mL portions were dispensed into five 1.5 mL vials for each menu.

Nitrogen was bubbled through each vial for 10 seconds prior to storage at -20 C.

Extractions were performed in duplicate for each menu for the three different

times it was sampled. The folate extracted from the menus was quantitated in

triplicate by microbiological assay using L. casei (Tamura 1990). A blank was

utilized throughout the extraction procedure so that folate derived from the

addition of the enzymes could be subtracted.



Synthesis, Purification and Identification of Deuterated Folic Acid

Deuterated folic acid was synthesized according to the method described

by Gregory (1990). The first step involved bromination of folic acid with








64
bromine at the 3' and 5' positions of the benzoic acid ring as described by

Cosulich et al. (1951). Six grams of unlabeled folic acid (#F-7876) were

dissolved in 75.6 mL of 6 N hydrochloric acid (HCI), diluted with an equal

volume of distilled water and placed in an ice bath on a stirrer. A large excess of

bromine vapor was used (126 mmol) to brominate the folic acid molecule at the

specified positions via nitrogen gas. The reaction was conducted until all the

bromine was consumed (approximately 20 minutes). Following dilution of

dibromo folic acid with 75 mL distilled H20, the sample was filtered through a

sintered glass funnel under vacuum and refrigeration, washed several times and

dried in a dessicator for several days.

The second step involved synthesis of deuterated folic acid from dibromo

folic acid. A 2.0 g portion of dry crude brominated folic acid was suspended in

60 mL deuterium oxide (D20) and dissolved by the addition of 1.0 mL of 40%

(w/w) sodium deuteroxide (NaOD). The use of D20 and NaOD allowed for the

removal of the exchangeable protons from the 3',5'-dibromofolate. After 5

minutes, 1.6 mL of 35% (w/w) deuterium chloride (DCI) was added to precipitate

the 3'5'-dibromofolic acid. Following centrifugation for 15 minutes at 5,000g (40

C), the pellet was suspended in 30 mL D20, dissolved with 0.5 mL NaOD (40%)

and transferred to a hydrogenation bottle. Deuterium oxide and NaOD were

added to achieve a 200 mL and 2.5 mL total of D20 and NaOD, respectively. The

catalyst, 0.25 g of 10% palladium (Pd) on carbon (Kodak; Rochester, NY), was

then added and nitrogen gas was bubbled into the solution to exclude dissolved








65
oxygen. Deuterium gas was utilized to replace the bromine at the 3' and 5'

positions of the benzene ring with deuterium. This catalytic dehalogenation was

performed at room temperature for approximately 1 h. At the end of the reaction

period, the catalyst was removed by filtering the mixture through Whatman #1

paper under vacuum and washing several times. Crude [2H2]folic acid was

recovered from the filtrate by acidification to pH 2 with HCI followed by

centrifugation. The crude [ H2]folic acid was dried under a dessicator for several

days and weighed before starting the final step of purification. The weighed

crude [2H2]folic acid (1.3 g) was then divided in half and ~ 636 mg was purified

by ion-exchange chromatography on a column containing DEAE-Sephadex A-25

(Sigma Chemical Co., St Louis, MO).

Diethylaminoethyl (DEAE) sephadex A-25 (75g) was hydrated and

suspended in 0.025 M sodium phosphate buffer containing 0.1 M NaCI, pH 7

(Buffer A) and poured into the column (50 cm length, 2.5 cm ID). Approximately

1 L of Buffer A ran through the column for equilibration. Half of the crude

[2H2]folic acid sample (636 mg) was dissolved in 2-3 mL of 0.1 M NaOH and

approximately 10 mL of Buffer A. The sample required sonication to ensure it

was fully in solution. Litmus paper was used to check the pH which was

subsequently adjusted to pH 8. The dissolved sample was applied onto the

column and a fraction collector was assembled to collect 300 drops per tube or

about 15 mL. Once the sample was loaded, a 2-L gradient containing Buffer A

and 0.025 M sodium phosphate buffer with 0.6 M NaCI (Buffer B) was used to








66
elute [2Hz]folic acid. Fractions containing pure [2H2]folic acid were identified

via HPLC using a 0.033 M phosphate buffer containing 15% acetonitrile (MeCN)

as a mobile phase. The detector was set at 280 nm and a folic acid standard was

injected first to determine the retention time of folic acid. The [2H2]folic acid

fractions eluting at the retention time of folic acid were pooled and precipitated by

acidification with HCI. Following centrifugation, the purified [ H2]folic acid was

desiccated over P205 and weighed. The purity of the crude [ H2]folic acid (630

mg) was assessed spectrophotometrically and via HPLC and proton NMR. NMR

analysis indicated complete labeling of the 3',5' positions of the benzoic acid ring

as no signal was detected for 3',5'-protons.





Supplemental Folic Acid

Commercially available folic acid (Sigma Chemical, St. Louis, MO) was

used to prepare the folic acid supplements. Subjects received either 330 or 730

pg/d as synthetic folic acid. Deuterated [3',5'-2H2] folic acid was synthesized

by the method of Gregory (1990) as previously described. In order to increase

absorption of the synthetic folic acid, the day's requirement was divided into two

portions so that half was consumed in the morning and the other half in the

evening. Stock solutions were prepared containing the required amounts of both

the deuterated and the unlabeled folic acid for the first 6 weeks of the study. At a

later date, stock solutions containing only unlabeled folic acid were prepared for








67
the second half of the study period. Food grade phosphoric acid and sodium

phosphate (dibasic) were used to make the buffer in which [2H2]folic acid and

unlabeled folic acid were dissolved. For the 450 pg level, 60 pg of deuterated

folic acid (-15% of total folate) was required per day. Since the subjects were

consuming supplemental folic acid twice per day, only 30 pg was required per

supplement tube. The deuterated folic acid stock solution was prepared for the

450 pg/d groups by weighing out 45 mg deuterated folate, dissolving in 0.1 N

NaOH, and filling to 150 mL for a final concentration of 30 pg/100 piL. The

exact concentration of all stock solutions was determined spectrophotometrically

at 282 nm using a molar absorptivity coefficient at pH 7.0 of 27,600 L (mol -

cm)-1 (Blakley 1969). The volume needed per tube was adjusted according to the

actual concentration. The subjects on the 450 pg/d level also received 270 lg/d

of unlabeled folic acid or 135 pg twice/d. Folic acid (202.5 mg) was weighed and

dissolved with 0.1 N NaOH and filled to 150 mL with buffer for a final

concentration of 135 pg/100 pL. For the 850 pg/d group, 120 ag of the

deuterated folic acid (- 15% of total folate) was required per day or 60 pg per

tube. To make the deuterated stock solution for this level, 90 mg of deuterated

folic acid was weighed out and dissolved with 0.1 N NaOH and filled to 150 mL

with buffer for a final concentration of 60lg/100 LL. The 850 pg groups required

610 pg of unlabeled folic acid or 305 pg per tube. Folic acid (457.5 mg) was

weighed out and dissolved with 0.1 N NaOH and filled to 150 mL with buffer for








68
a final concentration of 305 pg/100 pL. The stock solution for the second half of

the study was prepared in a similar fashion except the folate contribution from the

deuterated folic acid was replaced with unlabeled folic acid. All stock solutions

were stored in a -700C freezer and the concentration was verified

spectrophotometrically before making up a new batch of supplements to ensure

that no degradation of folate had occurred during storage. For human

consumption, stock solutions in appropriate volumes (-100 pL/tube) were

dispensed into 50 mL conical tubes to which approximately 40 mL of apple juice

was added. The tubes were then capped tightly, vortexed and stored at -200C

until used (within 3 months). Supplemental folic acid content was confirmed

periodically by HPLC and found to be fully stable.



Purification and Administration of [13C]pABG

Labeled pABG was provided to a subsample of the subjects (n=4) in order

to obtain information on the absorption and excretion of exogenous pABG which

is found in food. [13C]Folic acid, kindly provided by the lab of Dr. Gregory, was

cleaved at the C9 and N10 bond via zinc dust suspended in water (0.75g/3mL

H20) and 5 N HC1. After removal of the zinc dust by centrifugation, the

supernatant containing both the pteridine and pABG moieties was removed and

adjusted to pH 7. At this point, the supernatant became cloudy due to formation

of a precipitate and required a second centrifugation. The final supernatant

containing both fragments of the molecule was then applied to a size exclusion








69
gel (P2). The two fragments were separated on the basis of size upon the addition

of water to the gel after the sample was loaded. Small 1-2 mL fractions were

collected and analyzed by HPLC to identify the fractions containing the pABG

fragment. A pABG standard (Sigma #A-0879) was used to identify the retention

time of pABG and the appropriate fractions were pooled. The concentration of

the pooled pABG was then determined by HPLC using a pABG standard curve.

One hundred microgram portions of the pABG water mixture was dispensed into

50 mL conical tubes and filled to 40 mL with apple juice. Four subjects (one

from each group) received the one time administration of 100 Ag [13C]pABG at

the half way point of the study (beginning of week 7). This was followed by five

consecutive 24 h urine collections. The enrichment of the urine with [13C]-

labeled pABG will be determined by GCMS and evaluated in terms of

bioavailability. Another aspect of this study involves the quantitation of pABG in

the diet. This method has been worked out by the laboratory of Dr. Gregory and

may be an ancillary project.

Blood Collection and Processing

Weekly fasting blood samples were drawn by a certified phlebotomist

throughout the twelve wk study. Multiple blood collections (i.e. six in a 2 wk

period) were obtained at the switch-over period starting at the beginning of the 7

wk. Blood processing occurred within one h of the blood draw. A total of 17

blood collections per subject were obtained during the 12 wk study period. The

initial blood draw obtained on day one was representative of baseline








70
concentrations of serum folate, red cell folate and plasma homocysteine, while the

last blood draw at the end of wk 12 represented final concentrations. Precautions

were taken to protect all blood samples from light during collection, processing

and storage. As mentioned previously, upon completion of the twelve wk study, a

subgroup of the pregnant (n=8) and nonpregnant women (n=7) returned once a

month for phlebotomy during a three month follow-up period. In addition,

maternal and cord blood and a blood chemistry profile (SMAC 25) were obtained

at the time of delivery for a subgroup of the pregnant women (n=8).

Serum was collected in 13 mL silica-coated tubes (Vacutainer Becton

Dickinson, Rutherford, NJ) and held at room temperature for 30 to 60 minutes to

allow time for clotting. The clotted portion of each sample was removed by

centrifugation at 2,000 rpm for 30 minutes (International Equipment Company

Model HN-S II Centrifuge, Needham Heights, MA). The serum was then

dispensed in duplicate into 1.5 mL vials containing approximately 1.5 mg of

ascorbate (to ensure folate stability during storage) and subsequently frozen at -

200C until analysis of serum folate by microbiological assay.

Plasma and whole blood were collected in 10 mL ethylenediamine

tetraacetic acid (EDTA) tubes (Vacutainer Becton, Dickinson, Rutherford, NJ).

For a small subset of subjects (n=4) an extra 13 mL EDTA tube (Vacutainer ,

Becton, Dickinson, Rutherford, NJ) was used to collect plasma to be analyzed at a

later date by GCMS to obtain information on folate kinetics.








71
Plasma to be assayed for homocysteine was placed immediately on ice and

centrifuged at 2,000 rpm (IEC Model HN-SII Centrifuge, Needham, MA) for 30

minutes in the cold room. After centrifugation, the plasma was pipetted in

duplicate into 5 mL scintillation vials and placed in the freezer at -200 C for

subsequent analysis by a coworker. Plasma to be assayed for deuterated folic acid

was kept at room temperature and centrifuged at 2,000 rpm for 30 minutes

(International Equipment Company Model HN-S II Centrifuge, Needham

Heights, MA). The plasma was then transferred to a 10 mL conical tube. The

plasma obtained after centrifuging the whole blood for the packed red blood cells

was added to the appropriate tube to increase the total plasma obtained (about 8

mL).

Whole blood was mixed at room temperature and 75 pL was dispensed in

duplicate into 1.5 mL vials. A 1:20 dilution of whole blood was then performed

with a 0.1% ascorbate solution made with ascorbate (Na salt) and milli-Q water

(Millipore, Bedford, MA). The whole blood folate diluted with ascorbate and

milli-Q water was incubated at room temperature for at least 30 minutes in order

to allow for deconjugation of the polyglutamates to mono-, di- or tri- glutamates.

From the same sample, duplicate samples of whole blood were drawn into

capillary tubes and centrifuged for 7 minutes at 2,000 rpm. Hematocrit levels

were determined using a microhematocrit tube reading device.

The packed red cells from the homocysteine tube and the blood remaining

in the whole blood tube were combined in a 10 mL conical tube and centrifuged








72
at 2,000 rpm for 20 minutes to obtain approximately 5 mL of packed red cells.

An equal volume (1:1) of ascorbate dissolved in milli-Q water (0.2% solution)

was added to the red cells following the removal of plasma. After mixing, the red

cells were stored at -20 C until subsequent analysis of the deuterated folate by a

coworker.



Urine Collection and Processing

Brown, opaque, 2.5 L bottles containing 3 g of ascorbic acid (Na salt) to

protect the urinary folates from degradation were used to collect the 24 h urine

samples. Urine was collected weekly and for five consecutive days during the

switch-over period. A total of eighteen, 24 h urine collections were obtained per

subject. The subjects were instructed to start the collection after the first void and

to end the collection after their first void the next morning. The urine was kept

cold at all times during the collection process and returned to the investigator after

the 24 h period. The volume of each 24 h urine collection was recorded, and

urine was transferred into 200 mL containers and 1.5 mL vials and stored prior to

analysis at -200 C.

Analyses Performed by Microbiological Assay

Serum Folate Total serum folate concentrations were assessed microbiologically

using the test organism L. casei. A slightly modified version of the method

described by Tamura (1990) utilizing the 96-well microtiter plate and serial

dilutions was used for analyzing the samples. The L. casei (microbe #9640) was








73
grown in ATCC 7830 media (Difco Laboratories, Detroit, MI). Serum contains

only the monoglutamate form (mainly 5-methyl-THF) to which the L. casei

microbe responds. The samples were added to assay medium containing minimal

folic acid and were subsequently inoculated with the microorganism. The

samples incubated over night, and the growth response of the organism was

estimated the next day by measuring the increase in turbidity of the medium

which is proportional to the folate concentration in the serum or red blood cell

hemolysate. Calibration curves for the assay were constructed using 10 ng/mL

folic acid standard.

Falcon (Becton Dickinson, Rutherford, NJ), 96-well, flat bottomed,

sterile, low evaporation tissue culture plates with lids were used for this assay.

For each assay, 20 pil of control, standard and samples were added in duplicate to

selected wells of the plate followed by the addition of 130 pl buffer to these select

wells. One hundred and fifty microliters of buffer was then dispensed into all

wells. A total of five serial dilutions of the samples, standard and pooled control

was made using an 8-channel pipette (Costar8-Pette #04670, Costar Corp.,

Cambridge, MA). All wells were inoculated with 150 gll of the medium

containing the test organism (5 pL/mL media), covered and placed in the

incubator at 370 C for 18 hours.

After incubation, the growth of the microorganism was measured by

reading the turbidity of each well at 650 nm using a microtiter plate reader

(Molecular Devices UV Max, Menlo Park,CA) interfaced with a computer. Data








74
were calculated using Soft Max data analysis software (Molecular Devices). The

turbidity reading of each sample was translated by the computer into a folate

concentration (ng/mL) based on the known amount of folate in the standard. A

standard curve was generated from the standards using a four parameter equation.

The linear portion of the standard curve was used to determine the folate

concentration (ng/mL) of the samples. Procedures for preparing the medium, the

reagents and the standard are detailed in the Appendix.

Within assay and day-to-day variation was assessed by measuring

previously determined folate concentrations of pooled plasma each time the assay

was performed. The inter- and intra- coefficient of variation for this assay were ~

8% and 7%, respectively.

Red cell folate Red cell folate concentrations were also analyzed

microbiologically using the microorganism, L. casei. Prior to analysis, the pH of

each sample was reduced to approximately 4.0 with 50 ld 0.1 M HCI and

incubated for sixty minutes in order to allow the conjugase to cleave the

polyglutamate form of the reduced folate to a mono-, di- or tri- glutamate. Due to

the addition of the acid, the dilution factor was adjusted from 20 to 21. The

acidification and incubation steps were necessary at this point due to the use of

sodium ascorbate instead of free ascorbic acid. All samples were analyzed in

triplicate and the intra- and inter- coefficient of variation were less than 10%.

Red blood cell folate concentrations were determined by multiplying the whole

blood folate (ng/mL) obtained directly from the assay by the dilution factor (21)








75
followed by subtraction of serum folate and hematocrit and division of this value

by the hematocrit as illustrated in the following equation.



[Whole blood folatel [serum folate x (1 hematocrit / 100)]
hematocrit
100


Urinary folate Concentrations of intact urinary folate were determined

microbiologically using L. casei as the test organism (Tamura 1990). Samples

were analyzed in triplicate and the intra- and inter- coefficient of variation were <

10%. Appropriate dilutions were performed on urine samples that were running

higher than what could be determined using the standard curve (> 20 ng/mL)



Urine and Serum Creatinine Determinations (Glomerular Filtration Rate)

Urine and serum creatinine determinations were measured using a Sigma

Diagnostics kit (Sigma Chemical Co., St. Louis, MO. No. 555-A). Three

milliliter of water, 3 mL of creatinine standard and 3 mL of sample were added to

cuvettes labeled as blank, standard and sample(s) (serum or urine), respectively.

Urine samples were diluted 10:1 with Milli-Q water. Alkaline picrate solution

was added to all cuvettes, mixed and allowed to stand at room temperature for 8-

12 min. The absorbance of standard and sample were read against a blank at 500

nm. Acid Reagent (No. 555-2) (0.1 mL) was added to all cuvettes, mixed and

allowed to stand at room temperature. Following 5 minutes, the absorbance of

standard and sample versus a blank were read. Concentrations of unknowns were








76
calculated from the absorbance of a 3 mg/dL creatinine standard ranging in

concentration from 0 to 10 mg/dL.


Calculations: Creatinine (mg/dL)
= 3 x [(Initial Final Absorption Test) / (Initial Final Absorption Std)]

Creatinine Clearance (ml/min) = urine (mg/dL)*urine (mLmin)
serum (mg/dL)


HPLC Quantitation of Urinary Intact Folates, pABG and apABG and
Preparation for GCMS

Overview

Weekly urine samples (30 200 mL) for each subject (n=312) were passed

through affinity chromatography columns in duplicate containing immobilized

folate binding protein. Urinary folates bind tightly to the column material at

neutral pH. The non-folate sample effluent from the affinity columns was

collected and frozen at -20 oC for subsequent pABG and apABG analysis.

Following the removal of contaminants with a high salt neutralizer solution, folate

was eluted in 5 mL 0.1 N HCI and collected separately. Of the 5 mL folate

effluent, 1 mL was saved for quantitation of urinary 5-methyl-THF and folic acid

at weeks 0, 4, 8 and 12. The remaining 4 mL was prepared for GCMS analysis in

order to determine isotopic enrichment of urine with deuterium.

Urinary pABG and apABG were measured in duplicate on a biweekly

basis in pregnant women (weeks 0, 2, 4, 6, 8, 10 and 12) and on a monthly basis

in nonpregnant controls (weeks 0, 4, 8 and 12) by the method of McPartlin et al.

(1992) with minor modifications. The eluant containing both pABG and apABG








77
(from above) was loaded onto ion exchange chromatography columns to separate

pABG and apABG. Tritiated pABG and [3H]apABG served as internal standards

and were added to urine samples prior to application on ion exchange

chromatography columns. After separation, the catabolites remained isolated

throughout the remainder of the analysis. Acetamidobenzoylglutamate was

deacetylated to form pABG and reapplied to the ion exchange columns. The

pABG fractions were diazotized and loaded onto sep-pak cartridges for further

purification and concentration. Azo-pABG was eluted in methanol and pABG

was regenerated. High performance liquid chromatography was used to

quantitate both endogenous pABG and pABG derived from apABG. High

performance liquid chromatography was also used to isolate and collect the

pABG fractions which were then prepared for GCMS so information can be

obtained on folate kinetics. See Figure 5 for overview of catabolite purification

and quantitation process.

Procedures

A. Purification of the Folate Binding Protein from Dried Whey An affinity

chromatography sorbent coupled to purified bovine milk folate-binding protein

(FBP) was prepared by a minor modification of the method of Selhub et al. (1980)

as described by Gregory and Toth (1988). Folate-sepharose column material

(Sigma Chemical Co., St. Louis, MO, No. F-7889) was gently poured into a glass

column for the purpose of binding the FBP in dried whey (Meloskim, Kerry

Ingredients). The column was washed with 500 mL 0.2 M acetic acid and 1 L of








78
water followed by equilibration with 1 L of 1.0 M KH2PO4. Dried whey (825g)

was suspended in 6 L of water (best to do in batches). The pH of the mixture was

adjusted to 7.0 with 5 M NaOH and subsequently centrifuged for 10 minutes at

8,000 rpm. The supernatant was filtered through glass wool to remove excess

lipid and then loaded onto the folate sepharose column at room temperature.

After loading, the column was washed with 1 M NaCI (1 L) containing 0.05 M

Tris HC1, pH 7.4 for the purpose of removing compounds other than the FBP.

Following a wash with H20 (1L), adsorbed FBP was eluted in 0.2 M acetic acid

(approximately 500 mL) and collected in 5 to 7 mL portions by a fraction

collector. A spectrophotometer (X = 280 nm) was used to determine which

fractions contained the protein. The protein containing fractions were pooled for a

total of approximately 300 mL. Dialysis tubing was used to remove the acetic

acid. Following placement of the sample in the dialysis tubing, it was placed in

11 L of refrigerated milli Q-water and allowed to sit on a stirrer overnight. The

next day, the 11 L refrigerated water was exchanged for fresh refrigerated Milli

Q-water and allowed to sit overnight on a stirrer for the second time. The

dialyzed FBP solution was freeze dried and stored at -20 OC until coupled to

Affigel 10.

B. Preparation of Folate Binding Protein Affinity Columns A 25 mL bottle of

Affi-gel 10 Gel (Bio-Rad Laboratories, Richmond, CA; No. #153-6046) was

gently shaken and transferred to a glass fritted funnel in order to drain the excess

liquid without letting the bed go dry. The bed was washed with 75-100 mL ice








79
cold distilled water and the moist gel transferred to a small (125 mL) erlenmeyer

flask. Purified folate binding protein (100 mg) from step A was suspended in 20

mL of 0.1 M KH2PO4, pH 7.0 at 40 C and mixed with gel in erlenmeyer flask.

The mixture was agitated gently at 40 C for 4 hours in a shaking ice water bath.

After agitation, the gel was filtered on a fritted glass filter and the effluent from

each wash was collected separately. The gel was washed with 3 x 20 mL cold

0.05 potassium acetate, pH 6.6 buffer and 2 x 20 mL cold 0.05 potassium acetate

containing 1.0 M NaCI, pH 6.6 buffer. The Bio-Rad method was used to analyze

the collected fractions for protein that did not couple to the affinity columns.

Bovine serum albumin (BSA) was utilized to make a five point standard curve

ranging from 0 to 25 pg/mL. The protein concentration of the fractions was

determined spectrophotometrically at 595 nm.

C. Binding Capacity Determination of Affigel-10 FBP Columns Affigel-lO FBP

material containing immobilized FBP (section B) was loaded into 12 glass

columns (7 mm i.d. x 20 cm) with a long pasteur pipette. The columns were filled

to a height of 6 cm which was approximately 2.5 mL of material and equilibrated

with 10-20 mL 0.1 M KH2PO4, pH 7.0. Affinity chromatography columns

containing immobilized FBP were previously found to retain folic acid, 10-

formyl-folic acid, THF, 5-methyl-THF, 10-formyl-THF, and 5,10,methenyl-THF

by Gregory and Toth (1988a). The amount of folate required to saturate the

affinity columns was determined by applying ten 1 mL fractions of 10 pg folic

acid /mL 0.1 M potassium phosphate buffer, pH 7.0 (determined








80
spectrophotometrically). After discarding the first 2 mL, the 10 mL eluates were

collected separately and the concentrations determined spectrophotometrically at

pH 7.0 using 0.1 M potassium phosphate buffer, pH 7.0 as a blank. The columns

were subsequently washed with 5.0 mL of 0.025 M KH2PO4 containing 1.0 M

NaCI (pH 7.0), followed by 5.0 mL of 0.025 M KH2PO4 (pH 7.0). Both effluents

were discarded. The binding capacity of the columns was determined by elution

of the bound folates with 0.1 N HC1. The first 2 mL effluent was discarded and

the following 5 mL of 0.1 N HCI collected in a 10 mL volumetric flask which was

filled to 10 mL with 0.1 N HC1. The concentration of the eluted folate was

measured spectrophotometrically at 296 nm using a molar absorptivity coefficient

at an acidic pH of 20,000 L (mol cm)-1 (Blakley 1969). Hydrochloric acid

(0.1 N) was used as the blank. The sum of the folate found in the load and the

folate found in the acidic effluent should equal the amount of folate initially

applied to the columns. The columns were then washed with 5 mL 0.1 N HCI and

re-equilibrated with approximately 15 mL of 0.1 M KH2PO4, pH 7.0. The total

binding capacity of the columns was approximately 25 pg (56 nmol) of folate.

D. Purification of Urine Samples using Affinity Chromatoraphy The volumes

of urine that were loaded on the affinity columns were calculated so that one third

of the total binding capacity (C) was not exceeded. This determination was based

on the urinary folate concentrations predetermined microbiologically. Following

adjustment of urine to pH 7.0, up to 200 mL of urine (volume varied depending

on urinary folate concentration) was filtered through Whatman #1 filter paper.








81
The filtered urine samples were loaded onto affinity columns pre-equilibrated

with 0.1 M KH2PO4, pH 7.0. A peristalic pump was used to maintain a flow rate

of 0.3 mL/min. The non-folate effluent from the load was saved for analysis of

pABG and apABG, as described later. The columns were washed with 5 mL of

0.025 M KH2PO4 containing 1.0 M NaCL, pH 7.0 and 5 mL of 0.025 M

KH2PO4, pH 7.0. The folate was eluted with 0.1 N HCI. The first 2 mL

contained no folate and was discarded. The following 5 mL were collected in a

volumetric flask. Four milliliter were saved for cleavage reactions and

subsequent GCMS analysis. No ascorbate was added to these samples. Ascorbate

(50 pl of a 10% mixture) was added to the remaining 1 mL solution and stored at

-20 oC for subsequent quantitation of urinary 5-methyl-THF and folic acid by

HPLC (E). The columns were washed with an additional 5 mL of 0.1 N HC1 to

ensure all compounds had eluted followed by re-equilibration with 15-20 mL of

0.1 M KH2PO4, pH 7.0 buffer.

E. HPLC Ouantitation of Urinary 5-Methyl-THF and Folic Acid The

concentrations of 5-methyl-THF and folic acid in the affinity column eluate were

determined in duplicate for wk 0, 4, 8 and 12 by a reverse phase HPLC procedure

described by Gregory et al. (1984) with minor modifications. A fluorescence

detector (Spectrovision FD-300 Dual-Monochromator) connected to a U120

Universal Interface was used to monitor the native fluorescence of 5-methyl-THF.

The detector was set at 295 nm excitation and 356 nm emission wavelengths. An

ultraviolet (UV) absorption detector monitoring at 280 nm was used for detection








82
of folic acid. Both metabolites were quantitated under the following conditions:

50--l injection volume, octadecylsilica column (Ultramex C18, 5 .rm particle

size, 4.6 mm i.d. x 250 mm; Phenomenex, Torrance, CA USA), flow rate of 1.0

mlJmin, isocratic mobile phase composed of 12% (v/v) acetonitrile and 0.033 M

phosphoric acid, pH 2.3. Quantitation of folic acid and 5-methyl-THF was

accomplished relative to standards ranging in concentration from 30 -1000

ng/mL. Standards were prepared from commercially available 5-methyl-THF and

folic acid (Sigma Chemical, St. Louis, MO) using published molar absorptivities

(Blakley 1969) to determine concentrations of stock solutions.

F. Chemical Cleavage of Intact Folates to pABG The 4 mL affinity column

eluate (D) containing 5-methyl-THF, folic acid and possibly THF was subjected

to chemical cleavage of the C9-N 10 bond forming pteridines and pABG (Toth

and Gregory 1988). Para-aminobenzoylglutamate, containing the deuterium, was

then isolated by HPLC and subsequently analyzed by GCMS. Cleavage of 5-

methyl-THF was accomplished by adding 2 M potassium acetate (150 Wl) and IN

NaOH (150 pl) to the 4 mL eluate yielding a pH of 5.7- 6.0. Five percent (v/v)

hydrogen peroxide (220 pl) was added to oxidize 5-methyl-THF to 5-methyl-

DHF. After exactly 30 seconds, 450 pl of 0.1% bovine hepatic catalase (Sigma

Chemical Co., No. C-40) was added to decompose the remaining hydrogen

peroxide and stop the reaction. To cleave 5-methyl-dihydrofolate, samples were

acidified by the addition of 500 pl of 5 N HCI and incubated at room temperature

for 60 minutes.








83
Folic acid was reductively cleaved in the acid mixture by the addition of

270 pl of a suspension of zinc dust (1 g zinc in 3 mL milli-Q water) with periodic

agitation for 15 minutes. The zinc was subsequently removed by centrifugation in

a clinical centrifuge at 2,000 rpm for 5 minutes. Supernatant was removed with a

Pasteur pipette and stored at -20 OC until HPLC isolation of the pABG. These

procedures yielded complete conversion of intact folates to pABG as monitored

by HPLC.

G. HPLC Isolation of pABG and derivatization HPLC isolation and

derivatization of pABG was performed by a co-worker. The pABG fragment

from step F was isolated and collected by HPLC using a Prodigy preparative C18

column (10 pLm particle size; 250 x 10 mm; Phenomenex, Torrance, CA USA).

Formic acid (0.1 M) containing 6% (v/v) acetonitrile was used as the mobile

phase at a flow rate of 4.0 mL/min. When collecting peaks for GCMS analyses,

the pH of the mobile phase should not be adjusted. The retention time of pABG

was predetermined with a pABG standard (Sigma #A-0879) dissolved in 0.1 N

NaOH. Detection was achieved with an absorbance detector monitoring at 280

nm. The pABG peak (~ 4 mL) was manually collected and evaporated to dryness

at 90 C under a constant flow of nitrogen gas, capped tightly and kept

refrigerated overnight. The dried pABG fractions were brought to room

temperature, flushed with nitrogen gas for 30 seconds and derivatized by addition

of 200 pl trifluoroacetic anhydride and 100 lil trifluoroethanol. Samples were

then capped and placed in a heating block at 900 C for one h. After cooling to








84
room temperature, excess reagent was evaporated under nitrogen gas at room

temperature and the residue dissolved in 50 l1 dry ethyl acetate. Samples were

transferred to autosampler vials with polyspring inserts and stored at -20 C

before GCMS analysis. The end product of the above reaction was a lactam of

NI -trifluoroacetyl-p-aminobenzoyl-glutamate-a-trifluoroethyl ester that retains

all isotopic labeling on the pABG component of the folates (Gregory and Toth

1988a). This lactam is well suited for GCMS analysis for determination of

isotopic ratios.

H. Synthesis of [3H1pABG Tritiated pABG was prepared from 250 microcuries

[3H]folic acid (Amersham International, No. TRK 212) which is labeled at

positions C7, C9 and 3' and 5' of the pABG moiety. To 200 .1 of the [3H]folic

acid, 100 Wil 5 M HCI and 100 pl of Zn dust suspended in water (1.0 g per 3 mL

milli-Q water) was added in order to cleave the intact molecule. The mixture was

shaken intermittently for 15 minutes at room temperature followed by

centrifugation for 10 minutes at 2,000 rpm to pellet the zinc. The supernatant was

removed and stored in the refrigerator. Unlabeled folic acid (100 pIg/mL) was

diluted with a 1% sodium ascorbate solution to a final concentration of 4 gg/mL

in order to resemble the conditions of the commercially available [3H]folic acid

and cleaved as described above. Unlabeled pABG standards were prepared from

commercially available pABG (Sigma Chemical Co., St. Louis, MO No. A-0879).

Following adjustment of 100 ll pABG stock solution (100 ug/mL) to pH 13 by








85
diluting 1:10 with 0.1N NaOH (10g/mL), unlabeled pABG concentration was

verified spectrophotometrically at 272 nm using 15,500 as the extinction

coefficient (Blakley 1969).

I. Purification of [3H1pABG by HPLC Tritiated pABG was purified under the

following conditions: 100 pl injection volume; octadecylsilica column (Microsorb

C18, 3 ptm particle size, 4.6 x 100 mm; Rainin; Woburn, MA), mobile phase

consisting of 0.1 M formic acid, pH 4.0, 6 % acetonitrile. Before running the

radioactive pABG-pteridine mixture, unlabeled pABG was injected to determine

the retention time of pABG. Cleaved, unlabeled folate containing both the

pteridine moiety and pABG was also run to assure the pteridine and pABG

moieties were eluting separately. Once the retention time of 3.55 minutes for

pABG was determined and lack of interfering compounds assured, the tritiated

sample was injected four separate times and pABG was collected and counted via

a scintillation counter. In order to assure the isolation and collection of

[3H]pABG, unlabeled pABG was spiked with 20 ptl of [3H]pABG, injected into

the HPLC, collected in one milliliter fractions for 10 minutes, and counted. The

isolation of [3H]pABG was verified since the majority of counts matched the

retention time of unlabeled pABG. To further confirm [3H]pABG had been

isolated, a small portion (20 ll) was applied to equilibrated ion exchange columns

and after washing with 0.1 N HCI was eluted with 0.6 N HC1. Both effluents

were collected and counted. Recovery of 95-100% of [3H]pABG in the 0.6 N







86
HC1 eluate provided additional evidence that [3H]pABG had been isolated. One-

half of the tritiated pABG was used for the synthesis of [3H]apABG and the other

half (-3 mL) was diluted to 500 mL with 0.1% sodium ascorbate solution and

mixed well. Five hundred microliters of the diluted [3H]pABG were transferred

to 1.5 mL polypropylene microcentrifuge tubes (Fisher Scientific, Pittsburgh, PA)

and stored at -20 OC. Each tube (500 pl) contained ~ 58,000 dpm.

J. Synthesis of [3HlapABG The [3H]pABG starting material (2.5 mL) was

concentrated by solid phase adsorption (C 18 Sep Pak, Waters). The [3H]pABG

fraction was converted to its diazotized derivative by the method of Bratton and

Marshall (1939). Two hundred microliters of 5N HCI and 200 p1 of 1% sodium

nitrite (lg/100 mL) were added to 2.5 mL of the isolated tritiated pABG (J) and

allowed to sit five minutes. Para-aminobenzoylglutamate, a primary amine, reacts

in acid conditions with the nitrite to form a diazonium salt. Next, 200 pl of

ammonium sulfamate (5%) was added to destroy excess HNO2. After five

minutes, 200 lil napthylethylene diamine (1%) was added to couple with pABG.

A minimum incubation period of 30 minutes was allowed for complete formation

of the purple colored azo compound. The diazotized radiolabeled derivative was

applied to and retained on a preactivated (3 mL methanol followed by 5 mL

water) Cl8 Sep Pak cartridge (Waters). The cartridge was washed with 10 mL

HCI (0.05 M) and azo-pABG eluted with 3 mL methanol. Tritiated para-

aminobenzoylglutamate was regenerated by the addition of 50 pl 5M HCI and 50








87
ll zinc dust suspended in water. The mixture was vortexed intermittently for 15

minutes followed by centrifugation at 2,000 rpm for 10 minutes. The acetamido

derivative was prepared by transferring the supernatant to a conical glass tube

and evaporating it to dryness under a stream of nitrogen gas in a reactitherm set at

400 C. Thirty-four microliters of 50% (v/v) acetic acid and 5 ill of acetic

anhydride was added to the dried extract. The tube was plugged with cotton,

vortexed gently and incubated at room temperature for 24 h. A nonradioactive

apABG standard was prepared by the method of Baker et al. (1964). One gram of

pABG was dissolved in 10 mL of 50% acetic acid and following addition of 1.5

mL acetic anhydride, the mixture sat overnight at ambient temperature. Whatman

paper (#1) was used to filter the mixture followed by washing with milli-Q water.

The residue was dessicated to remove water and stored at -20 C.

K. Purification of [3 HapABG by HPLC The same conditions as described in (I)

were used to isolate the radioactive apABG. Nonradioactive apABG standard

synthesized previously (J) was dissolved in water and injected to determine the

retention time of this fragment (~ 8 minutes). The tritiated sample was injected

and the peak corresponding to the retention time of unlabeled apABG manually

collected. To verify isolation of [3H]apABG, unlabeled apABG was spiked with

10 Wl [3H]apABG, injected into the HPLC, manually collected in 1 mL fractions

for the following 30 minutes and then counted. Since the majority of counts

matched the retention time of unlabeled apABG, it was assumed that [3 H]apABG




Full Text

PAGE 1

FOLATE STATUS RESPONSE OF PREGNANT WOMEN AND NONPREGNANT CONTROLS TO CONTROLLED FOLATE INTAKES By MARIE ANNE CAUDILL 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 1997

PAGE 2

This dissertation is dedicated to my mom and dad who from the beginning taught me that along with hard work comes opportunity. I thank them for their unconditional love and support.

PAGE 3

ACKNOWLEDGMENTS The author wishes to thank Dr. Lynn B. Bailey, her major professor, for the opportunity to work on a meaningful and exciting research project. Her guidance and support will always be appreciated and her dedication and commitment to contributing scientific knowledge admired. She has been a true inspiration in every way. The author extends much appreciation to the members of her graduate committee, Dr. Peggy Borum, Dr. Jesse Gregory, JU, Dr. Gail Kauwell and Dr. Sharleen Simpson for their support, assistance, and advice. Special thanks are expressed to Rita Harris, Kami Pratt, Hanh and Loan Bui and Linda Armada for their contribution and dedication to research goals and to Vicki Sullivan for her friendship and support. Additional thanks are extended to Jeff Opalko, Kelli Garcia, Jerry Williamson, Christine Pfeiffer, Lisa Rogers and Richard Bonnette for their laboratory assistance and advice. Recognition is also given to the faculty, administrative staff and student volunteers for their help and assistance. The author expresses her gratitude and appreciation to her parents, sister, brother and other family members including Rich Dolby, Diane Woods, Sue and Sy Caudill and Pat Sickles for their love and encouragement along the way. Special recognition is extended to Tony Caudill, her best friend and husband, for his continued love, motivation, and understanding. iii

PAGE 4

TABLE OF CONTENTS ACKNOWLEDGMENTS iii LIST OF TABLES vii LIST OF FIGURES vii LIST OF ABBREVIATIONS x ABSTRACT xii CHAPTERS 1 INTRODUCTION 1 Hypotheses 6 Specific Aims 6 2 LITERATURE REVIEW 9 History 9 Folate Biochemistry 9 Food Distribution and Folate Bioavailability 11 Transport and Tissue Distribution 15 Folate Functions 19 Turnover and Excretion of Folate 23 Folate Requirements and Recommended Dietary Allowances 28 Folate Deficiency and Pregnancy 39 Folate Supplementation During Pregnancy 41 Neural Tube Defects 43 Assessment of Folate Status 47 Stable Isotopes 54 3 STUDY DESIGN AND METHODS 57 Study Design 57 59 59 62 Subjects Diet Folate Extraction from Menus iv

PAGE 5

Synthesis, Purification and Identification of Deuterated Folic Acid 63 Supplemental Folic Acid 66 Purification and Administration of [ 13 C]pABG 68 Blood Collection and Processing 69 Urine Collection and Processing 72 Analyses Performed by Microbiological Assay 72 Serum Folate 72 Red Cell Folate 74 Urinary Folate 75 Urine and Serum Creatinine Determinations (GFR) 75 HPLC Quantitation of Urinary Intact Folates, Urinary pABG and apABG And Preparation for GCMS 76 Overview 76 Procedures 77 A. Purification of Folate Binding Protein from Whey 77 B. Preparation of Folate Binding Protein Affinity Columns 78 C. Binding Capacity Determination of Affigel10 FBP Columns 79 D. Purification of Urine Samples using Affinity Chromatography 80 E. HPLC Quantitation of Urinary 5-Methyl-THF and Folic Acid 81 F. Chemical Cleavage of Intact Folates to pABG 82 G. HPLC Isolation of pABG and Derivatization 83 H. Synthesis of [ 3 H]pABG 84 L Purification of [ 3 H]pABG by HPLC 85 J. Synthesis of [ 3 H]apABG 86 K. Purification of [ 3 H]apABG by HPLC 87 L. Preparation of Ion Exchange Columns and Fractionation of Urine.. 88 M. Deacetylation of Urinary apABG and Rechromatography 88 N. Diazotization and Solid-phase extraction 89 O. Regeneration of pABG 90 P. HPLC Quantitation of Urinary pABG 90 Q. Recovery of [ 3 H]pABG and [ 3 H]apABG added to urine samples. 91 R. Preparation of Catabolite Samples for GCMS Analysis 92 Statistical Analyses 92 Interpretation of the Data 94 Serum Folate 95 Red Cell Folate 95 Urinary Intact Folate 97 Urinary Catabolites (apABG and pABG) 97 4 RESULTS 99 Folate Content of Menus 99 Food Record Analyses 100 V

PAGE 6

Serum Folate 100 Red Cell Folate 103 Hematocrit 106 Urinary 5-Methyl-THF 108 Urinary Folic Acid Ill Urinary Total Fol ate 112 Glomerular Filtration Rate 112 Urinary apABG 113 Urinary pABG 116 5 DISCUSSION AND CONCLUSIONS 119 6 SUMMARY 131 APPENDIX 133 Microbiological Assay Procedures 133 Microplate Media Preparation for L. casei 133 Folic Acid Standard Preparation 133 LITERATURE CITED 135 BIOGRAPHICAL SKETCH 151 vi

PAGE 7

LIST OF TABLES Table page 1 Five-day cycle menus 60 2. Nutrient composition of five-day cycle menus 61 3. Folate content of menus 99 4. Comparisons of initial, steady state and final total urinary folate excretion by microbiological assay and HPLC 1 12 5 Glomerular filtration rates at wk 0, 4, 8 and 12 113 vii

PAGE 8

LIST OF FIGURES Figure page 1. Structure of folic acid (C 19 Hi9N 7 06) 10 2. Functions and interconversions of coenzyme forms of THF 20 3. Conversion of homocysteine to methionine and regeneration of THF 22 4. Overview of protocol 58 5. Overview of catabolite isolation and purification 89 6. Serum folate concentration throughout 12 wk study 101 7. Group comparisons of mean initial and final serum folate concentration... 102 8. Serum folate concentration at steady state 103 9. Red cell folate concentration throughout 12 wk study 103 10. Group comparisons of mean initial and final red cell folate concentration. 105 1 1 Red cell folate concentration at steady state 106 12. Hematocrit throughout 12 wk study 107 13. Group comparisons of mean initial and final hematocrit 107 14. Urinary 5-methyl-THF excretion throughout 12 wk study 108 15. Group comparisons of mean initial and final 5-methyl-THF excretion.... 109 16. Urinary 5-methyl-THF excretion at steady state 1 10 viii

PAGE 9

Figure page 17. Comparisons in final group urinary folic acid excretion 1 1 1 18. Urinary apABG excretion throughout 12 wk study 1 13 19. Group comparisons of mean initial and final urinary apABG excretion. 114 20. Comparisons of mean overall urinary apABG excretion 1 15 21. Urinary pABG excretion throughout the 12 wk study 1 16 22. Group comparisons of mean initial and final urinary pABG excretion 1 17 23. Urinary pABG excretion at steady state 118 ix

PAGE 10

LIST OF ABBREVIATIONS ADDreviauon Meaning acetamidobenzoylglutamate ATYD adenosine 5 '-diphosphate 5-formiminoimidazole 4-carboxamide ribonucleotide ATP Air adenosine 5 'triphosphate DO A bovine serum albumin Centers for Disease Control (J day deuterium chloride diethylaminoethyl Drir dihydrofolate Department of Health and Human Services deoxyribonucleic acids OlMr deoxythymidylic acid JT T\ A T~% dUMr deoxyundyhc acid dUb 1 deoxyundine suppression test T7TYT A ethylenediamine teraacetic acid rrJr folate binding protein uLMi> gas chromatography mass spectrometry GAR giycinamiae riDonoe giu glutamate h hour HPLC high performance liquid chromatography HC1 hydrochloric acid mg milligram MeCN acetonitrile MRC Medical Research Council MTHFR methylenetetrahydrofolate reductase NaCl sodium chloride NaOD sodium deuteroxide NaOH sodium hydroxide ng nanogram nm nanometer

PAGE 11

Abbreviation Meaning NHANES H Second National Health and Nutrition Examination Survey NMR nuclear magnetic resonance pABA p-aminobenzoic acid pABG p-aminobenzoylglutamic acid Pd palladium PHS Public Health Service PLP pyridoxal 5 '-phosphate RDA Recommended Dietary Allowance RNA ribonucleic acids SAH s-adenosylhomocysteine SAM s-adenosylmethionine THF tetrahydrofolate M8 microgram US United States wk week y year xi

PAGE 12

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 FOLATE STATUS RESPONSE OF PREGNANT WOMEN AND NONPREGNANT CONTROLS TO CONROLLED FOLATE INTAKES By Marie Anne Caudill August, 1997 Chairperson: Dr. Lynn B. Bailey Major Department: Food Science and Human Nutrition Folate is a B vitamin that plays a major coenzymatic role in one-carbon metabolism and is a key participant in the metabolism of certain amino acids and biosynthesis of pyrimidines and purines. Folate requirements are higher during pregnancy, although the increment above that required by nonpregnant women has not been ascertained. The 1989 RDA for both pregnant and nonpregnant women was reduced by approximately one-half. The rationale for the reduction in pregnant women from 800 to 400 pg/d was based primarily on studies in which folate intake was not controlled. The purpose of this 84 d study was to investigate folate status response of pregnant subjects (n=12) during their second trimester and nonpregnant controls (n=12) to defined folate intakes comparable to the current (400 pg/d) and former (800 pg/d) RDA for pregnant women. Subjects consumed either 450 or 850 pg/d xii

PAGE 13

provided as a combination of dietary folate and synthetic folic acid. The outcome variables used to assess folate status response included serum, red cell and urinary folate and the folate catabolites pABG and apABG. Comparisons between and within supplementation groups were made at steady state or the time at which the subjects had acclimatized to their assigned folate intake. No differences (P>0.05) were detected in serum and red cell folate concentrations or in urinary 5-methyl-THF excretion between pregnant and nonpregnant women consuming either 450 or 850 ug/d. Serum folate and 5methyl-THF excretion were higher (P<0.05) in subjects consuming 850 compared to 450 ug/d. Urinary folic acid (14.5 ug/d) was detected only in nonpregnant controls consuming 850 ug/d folate. No differences (P>0.05) were detected in pABG or apABG concentrations between pregnant and nonpregnant women consuming either 450 or 850 ug/d. Pregnant women consuming 450 compared to 850 ug/d excreted less (P<0.05) of both catabolites by the end of the study. In summary, no differences were detected in the outcome variables measured between nonpregnant and pregnant women within the same supplementation group. Further, normal folate status was maintained by all groups. These data suggest that 450 ug/d of folate derived from both dietary and synthetic sources is adequate for pregnant women during their second trimester. This level of intake equates to approximately 600 ug/d dietary equivalents assuming 50% and 75% availability of food folate and synthetic folic acid consumed with meals, respectively. xiii

PAGE 14

CHAPTER 1 INTRODUCTION Folate is a generic descriptor for compounds that have nutritional properties and chemical structures similar to those of folic acid (pteroylglutamic acid). Thus, it includes folic acid, the oxidized, monoglutamate form of the vitamin provided in supplements and fortified foods, and the reduced, naturally occurring, polyglutamyl folates, which are the predominant dietary forms. De novo biosynthesis of folates occurs only in bacteria and plants so humans must obtain this vitamin from their diet. Folate in its fully reduced state and polyglutamyl form participates in a variety of coenzyme reactions involving one-carbon transfers required for amino acid metabolism and nucleic acid synthesis, as well as the biosynthesis of specific lipids, hormones and neurotransmitters. Deficiency of the vitamin leads to impaired cell division and to abnormalities in numerous synthetic pathways dependent on one-carbon transfer reactions. A wide array of clinical problems may result from a deficiency including megaloblastic anemia, growth failure, malabsorption syndromes and increased risk for certain types of cancer and atherosclerosis (Bailey 1990a, Giovannucci et al. 1993, Selhub et al. 1995). Pregnancy is associated with increased requirements for folate, although controversy exists regarding the amount of folate required to meet the demands of l

PAGE 15

2 the pregnant woman and her developing fetus (NRC 1989, Subcommittee on Dietary Intake and Nutrient Supplement During Pregnancy 1990, Herbert 1987a, Bailey 1995). Adequate folate intake is essential during pregnancy since this time represents a state of increased cellular division due to the rapid growth of the fetus, the placenta and the maternal organs such as the uterus (Cunningham et al. 1993) In addition, there is an increase in erythropoiesis, a folate requiring process, and plasma resulting in the expansion of maternal blood volume (Cunningham et al. 1993). The pregnancy induced hypervolemia which starts during the first trimester serves to meet the demands of the enlarged organs (Cunningham et al. 1993). Due to the severe demands that are placed on the supply of folate during pregnancy for the synthesis of DNA, pregnant women are at increased risk of developing a folate deficiency (Bailey et al. 1980, O'Connor 1994) Sauberlich (1990) summarized numerous studies both within the United States and world wide that illustrated the widespread prevalence of poor folate nutriture during pregnancy. Bailey et al. (1980) reported that among a group of low income pregnant women, folate deficiency was significantly more prevalent than iron deficiency at the initiation of prenatal care. These data clearly illustrate that many women are unable to meet the increased demand for folate during pregnancy through diet alone which has resulted in the acceptance of routine prenatal folate supplementation (Sauberlich 1995). Unfortunately, due to lack of data regarding the size of the folate body pool and equilibrium of the vitamin during pregnancy, the National Academy of

PAGE 16

3 Sciences Committee on Dietary Intake and Nutrient Supplements During Pregnancy (1990) could not establish a specific supplementation recommendation for folic acid. Despite the widespread acceptance of folic acid supplementation for the vast majority of pregnant women, the Committee recommended that folic acid supplementation be reserved for specific "high risk" population groups such as pregnant adolescents but not for the pregnant population as a whole. However, this recommendation preceded conclusive research regarding NTDs and therefore was not addressed by this committee.. The 1989 RDA for folate was lowered from 800 ug/d (NRC 1980) to 400 p.g/d (NRC 1989) for pregnant women. The reduction in the RDA for pregnant women was largely based on the findings of two studies (NRC 1989). Chanarin et al. (1968b) reported that 100 ug/d of synthetic folic acid, in addition to dietary folate, was able to maintain red cell folate concentrations in pregnant women throughout gestation. Bates et al. (1982) subsequently reported an estimated dietary folate intake of 190 |ag/d for this population group. This estimate (190 |ag/d) in conjunction with Chanarin's report that 100 ug/d plus diet maintained red cell folate concentrations was used by the 10 th edition RDA committee to support the 400 ug/d RDA (NRC 1989). Colman et al. (1975) found that maize fortified with 300 ug/d folic acid consumed with a constant diet maintained normal serum and red cell folate concentrations during the last thirty days of pregnancy in a rural, poorly nourished, African population. Interpretation of these studies is complex, and illustrates the importance of investigating folate requirements in

PAGE 17

healthy pregnant women under controlled metabolic conditions during which time the response of pregnant women to known levels of folate intake can be assessed. Currently, relatively little is known about the requirement for folate during pregnancy, the specific tissue needs for folate, or how to assess folate status during gestation (i.e. the range of acceptable values) (O'Connor 1994). In recent years, studies have been conducted to assess the appropriateness of the decrease in the RDA for nonpregnant women of child-bearing age and adult men. O'Keefe et al. (1995) reported that the 1989 RDA for nonpregnant women of 180 ug/d was insufficient to maintain initial folate status in healthy, unsupplemented nonpregnant women and that 400 ug/d more adequately met the criteria of an RDA. Similarly, Jacob et al. (1994) reported that 200 u.g of folate per day was insufficient to achieve normal plasma homocysteine concentrations after a moderate depletion period in a depletion-repletion protocol involving ten healthy male adults. These data suggested that both males and nonpregnant females require more than the current RDA to maintain normal folate nutriture. McPartlin et al. (1993) used an alternative approach in estimating folate requirements in pregnant women at one time point during each trimester. Their approach was based on the quantitation of the urinary folate catabolites, paminobenzoylglutamate (pABG) and the more predominant form, acetamidobenzolyglutamate (apABG). These catabolites were hypothesized to represent folate turn over by McPartlin and associates (1993) and therefore a more direct way to estimate requirements. McPartlin et al. (1993) reported that based

PAGE 18

on the urinary excretion of pABG and apABG, the current RDA for pregnant women was inadequate particularly in the second trimester. Limitations of this study included the fact that folate intake was not controlled and folate status was not assessed. We conducted a controlled metabolic trial to investigate the response of pregnant women in their second trimester to folate intakes comparable to the current RDA (400 ug/d) and the former RDA (800 ug/d). Two nonpregnant groups consisting of women similar in age, lifestyle and demographics were included as a control. The decision to select the second trimester was based on the findings of McPartlin et al. (1993) which indicated that the second trimester represented the highest rate of folate catabolism and thus the greatest requirement for folate. The overall goal of the study was to provide metabolic data which could be used to assess the adequacy of the current RDA for pregnant women. One objective of this study was to determine if differences existed in folate status and catabolite excretion between pregnant subjects and nonpregnant controls consuming equivalent folate intakes. A second objective was to determine if differences existed in folate status and catabolite excretion between pregnant subjects consuming folate intakes approximating the current RDA (400 ug/d) compared to the former RDA (800 ug/d).

PAGE 19

Hypotheses The major hypotheses and specific aims of this research were as follows. Hypothesis 1. The response of pregnant women and nonpregnant controls to two defined levels of folate intake (450 or 850 ug) would indicate a greater folate requirement for pregnant women as determined by changes in folate status indices and catabolite excretion. Hypothesis 2. Folate response indices would reflect a decline in folate nutriture approaching negative folate balance among pregnant women consuming 450 pg/d of folate and maintenance of folate status in pregnant women consuming 850 pg/d. Hypothesis 3. Folate response indices would reflect maintenance of folate status in nonpregnant women consuming 450 pg/d of folate and an excess exceeding the body's ability to metabolize or store folate in nonpregnant women consuming 850 pg/d of folate. Specific Aims Aim 1. To determine weekly serum and red cell folate concentrations via microbiological assay during the study period and monthly concentrations during the follow-up period. Differences between and within supplementation groups would help assess the adequacy of folate intake. Aim 2. To determine weekly urinary excretion of intact urinary folate via microbiological assay and to determine the specific forms of excreted urinary intact folate by high performance liquid chromatography. Data would be

PAGE 20

7 obtained related to the amount of excreted, unmetabolized urinary intact folate (folic acid) and the amount of excreted, metabolized urinary intact folate (5methyl-tetrahydrofolate) and comparisons would be made between and within supplementation groups. Aim 3. To quantitate the urinary excretion of the two major folate catabolites, pABG and apABG. Information on folate turnover and possibly requirements during pregnancy would be obtained based on the premise that catabolized folate reflects utilization and that which needs to be replaced daily. Folate was provided as a combination of dietary and synthetic sources. During the first half of the study, a small percentage of the total daily intake of folate came from folate labeled with deuterium followed by provision of 100% unlabeled folate. Labeling of synthetic folic acid with deuterium, a staple isotope, can be used safely in humans (Gregory et al. 1990). Information on folate kinetics, including turnover rates and pool sizes, can be obtained by comparing the ratio of labeled folate to unlabeled folate in biological samples collected from humans consuming deuterated folic acid. A small subsample of subjects consisting of one subject from each group (n=4) was also given [ l3 C]pABG midway through the 12 wk period so that the bioavailability of exogenous pABG could be evaluated and reported separately by a co-worker. Several parameters were used to assess folate nutriture including serum folate, red cell folate, urinary intact folate and the urinary breakdown products, pABG and apABG. Kinetic data will also be obtained based on the isotopic

PAGE 21

enrichment of biological samples including urine, red cell and plasma and reported separately. Plasma homocysteine, a functional index of folate status, measured by a co-worker.

PAGE 22

CHAPTER 2 LITERATURE REVIEW History In 1931, Dr. Lucy Wills described a macrocytic anemia in pregnant Hindu women that responded to treatment with yeast extract. Later, Wills and Bilimoria (1932) described a similar macrocytic anemia in monkeys that responded to a factor present in crude liver extracts. The active principle in liver and yeast extracts described by Wills had many properties in common with bacterial "growth factors" present in liver (Snell and Peterson 1940) and spinach leaves (Mitchell et al. 1941). The term folic acid, derived from the Latin word "folium" meaning "leaf," was coined by Mitchell et al. (1941) after isolating this substance from spinach leaves. Folic acid was synthesized by Angier et al. (1945), and it was shown to be the common factor in the preceding observations. Folate Biochemistry The folic acid molecule consists of three parts: the pteridine moiety (with an amino group in position 2 and hydroxyl group in position 4); the para-aminobenzoic acid; and the L-glutamic acid (Figure 1). The pteridine component is linked by a methylene bridge to para-aminobenzoic acid which is joined by peptide linkage to glutamic acid. The first two fractions compose pteroic acid which leads to the classification of folic acid as pteroyl-glutamic acid. Folic acid 9

PAGE 23

10 (C19H19N7O6) is named in the chemical literature as a derivative of L-glutamic acid: N-[4-[{(2-amino-l,4-dihydro-4-oxo-6-pteridinyl)methyl]-amino}benzoyl]L-glutamic acid. Crystalline folic acid is yellow with a molecular weight of 441.41 g/mol. Folic acid, itself, is not biochemically active but becomes so after it has undergone reduction. Naturally occurring folates include a very large number of compounds because (a) the pirazine ring (atoms 4a,5,6,7,8 and 8a) of the pteridine moiety can occur in three states of reduction with (b) six different one-carbon substituents present at N-5 and/or N-10 and (c) the polyglutamyl chain may have as many as 12 glutamyl residues (Krumdieck 1990). N N 3 OH 6 CH, NH 0 COOH II I C-NHCH-CH, -CH, COOH pteridine p-aminobenzoic acid glutamic acid pteroic acid pteroylglutamic acid Figure 1. Structure of Folic Acid (C 19 H 19 N 7 0 6 ) In nature, folate exists mainly in the conjugated forms. Pentaglutamates and hexaglutamates predominate as the intracellular folates in animal tissue, while

PAGE 24

11 extracellular folates (i.e. in plasma, spinal fluid, bile and urine) are monoglutamate derivatives (Krumdieck 1990). Most natural folates are unstable and are degraded by heat, oxidation and ultraviolet light. Some tetrahydrofolate (THF) derivatives including 5-formyl-THF and 5-methyl-THF are relatively heat stable. Acid will destroy THF and 5, 10-methylene-THF. The addition of ascorbic acid and other reducing agents has helped tremendously in preventing the oxidation and subsequent degradation of folates in biological samples. Folates have limited solubility in water and in most organic solvents such as methanol. Folates are minimally soluble in the mildly acidic range (pH 2 to 4), with solubility increasing in proportion to the pH above this range (Gregory 1989). Folates are also highly soluble at very low pH. Food Distribution and Folate Bioavailability Folate is found in a wide variety of foods originating from both animal and plant sources. Liver, mushrooms, orange juice and green leafy vegetables are particularly good sources (Combs 1992). White bread, rolls and crackers, although not rich sources of folate, represent major contributors of folate in the U.S. diet according to the Second National Health and Nutrition Examination Survey (NHANES H) data, 1976-1980 (Subar et al. 1989). These sources are likely to remain major contributors secondary to the FDA mandating the fortification of bread products with folic acid (FDA 1996). Dietary folate exists primarily in the reduced form as polyglutamyl derivatives of THF with 5-methylTHF and 10-formyl-THF predominating. Very little free folate (folyl monoglutamate) is found in foods. Food storage and preparation can greatly

PAGE 25

affect the stability of folate. Exposure of the folate molecule to heat, ultraviolet light, air and metal ions will result in oxidation and subsequent cleavage rendering it inactive (Herbert 1987a, Gregory 1989), although a small portion of the oxidized folates may be converted back to biologically active forms (Gregory 1989, Shane 1995). Further, because folate is a water soluble vitamin, it can be leached out of foods during periods of boiling. Before folates can be absorbed from the gut, glutamyl residues in excess of one must be removed because only monoglutamyl forms can enter the enterocyte. This cleavage is accomplished by an exocarboxypeptidase folyl y glutamyl hydrolase, more commonly referred to as folate conjugase (EC 3.4.12.10) (Halsted 1979) which functions at a pH optimum of 6.5 (Gregory 1995). Medications or conditions that increase this pH will decrease the activity of this enzyme possibly leading to folate malabsorption. Natural conjugase inhibitors found in certain foods including yeast, beans and lentils may also reduce folate availability (Bailey 1988, Butterworth et al. 1974). However, Wei et al. (1996) reported high relative bioavailability of both monoglutamyl and hexaglutamyl folic acid when added to lima beans and tomatoes. Chronic exposure to alcohol and possibly zinc deficiency may also reduce folate bioavailability (Bailey 1988, Naughton et al. 1988, Tamura et al. 1978). The monoglutamate form of folate is actively transported across the proximal small intestine by a saturable, pH-dependent process (Gregory 1995, Halsted 1990, Rose 1978). Two folate-binding proteins (FBPs) exist in the

PAGE 26

13 jejunal brush border and are thought to be involved in folate binding and transport (Shoda et al. 1993). Folic acid can also be absorbed by a nonsaturable mechanism involving passive diffusion when high concentrations are consumed (Gregory 1995). The term "folate bioavailability" refers to the efficiency of utilization of this vitamin, including physiological and biochemical processes involved in intestinal absorption, transport, metabolism and excretion (Gregory 1995). The bioavailability of dietary folate is estimated to be approximately 50% (Sauberlich et al. 1987). In contrast, synthetic folic acid consumed under fasting conditions is approximately 100% available (Gregory 1995). Foods fortified with synthetic folic acid were initially considered to be only 50% available (Colman et al. 1975). More recently, however, studies have found that foods fortified with synthetic folic acid are highly available (Cuskelly et al. 1996, Pfeiffer et al. 1997). Pfeiffer et al. (1997) determined the bioavailability of [ 13 C 5 ]folic acid in various fortified cereal-grain products and in water as a control, relative to the intravenously injected [ H 2 ]folic acid. Measurement of isotope excretion ratios of urinary folates (% of [ 13 C 5 ]folate dose / % of [ 2 H 2 ]folate dose) indicated no differences among treatment groups or the control suggesting that folic acid fortified cereal-grain foods were highly available. Pfeiffer and associates (1997) also indicated that consumption of [ 13 C 5 ]folic acid after a light breakfast meal did not significantly effect its bioavailability. The influence of food on the bioavailability of simultaneously ingested folic acid supplements has not been firmly established.

PAGE 27

14 However, based on recently reported data related to fortified foods, one could hypothesize that it is more available than endogenous folate present in food and less available than folic acid consumed under fasting conditions. Once folate enters the enterocyte, it usually undergoes reduction to THF and either methylation or formylation (Perry 1973, Selhub et al. 1983). However, folate metabolism within the enterocyte is not required for entry into portal blood or for transport (Shane 1995). When high doses of various folates are given, little metabolism occurs, and the majority of folate appears in portal circulation unchanged (Herbert and Das 1994). The degree of metabolism in the intestinal mucosa is a function of the amount of folate administered (Gregory 1995). Because the reduction process is readily saturable (Gregory 1995), supplemental folic acid may be more efficiently reduced and methylated if divided into several smaller doses (Lucock et al. 1989b). Gregory (1995) suggested that this may increase the efficiency of its metabolic utilization and improve its bioavailability. In 1966, Giles hypothesized that the high incidence of folate deficiency during pregnancy was a consequence of reduced folate absorption. Iyengar and Buba (1975) subsequently conducted a study to compare folic acid absorption in pregnant women to that of nonpregnant women. Two milligrams of folic acid along with 10 f^Ci of tritiated folic acid were administered to 9 pregnant women between 10 to 26 wk gestation undergoing therapeutic termination of pregnancy and 6 nonpregnant women undergoing sterilization. Absorption varied between 65 and 95 percent, with a mean absorption of 80 percent in both pregnant and

PAGE 28

nonpregnant subjects. The results indicated that there was no alteration in absorption of a large dose of folic acid during the first two trimesters of pregnancy. A limitation to this study was lack of comparison between pregnant and nonpregnant women in their ability to absorb polyglutamyl folates. Studies (Bailey et al. 1988, Gregory et al. 1991) conducted in nonpregnant subjects to determine the bioavailability of monoand polyglutamyl folate suggested partial availability of polyglutamyl folates compared to monoglutamyl forms. However, Wei et al. (1996). reported high availability of both polyglutamyl and monoglutamyl forms when added to lima beans and tomatoes. Transport and Tissue Distribution Monoglutamates, mainly 5-methyl-THF, and to a lesser extent folic acid, dihydrofolate (DHF) and formylated forms are present in portal circulation. Approximately two-thirds of the folate in plasma is bound to protein (Groff et al. 1995). The majority of these proteins bind folate with relatively low affinity and are non-saturable (Ratnam and Freisheim, 1990). Examples include albumin and a-2 macroglobin (Groff et al. 1995). In contrast, folate binding proteins present in low amounts in plasma and blood cells bind folate with high affinity (Wagner 1995). Free folate accounts for the remainder of folate found in the plasma. Folate can then be taken up by the liver or transported to extrahepatic tissues. Within the liver, folate is metabolized to polyglutamate derivatives by folyl polyglutamate synthetase and retained or released into blood or bile. The folate derivative, 5-methyl-THF, is a poor substrate for folyl polyglutamate synthetase

PAGE 29

and in order to be conjugated must first have its methyl group removed once it enters the cell. This can be accomplished by methionine synthase, a vitamin B 12 dependent enzyme. Polyglutamates cannot cross biological membranes and deconjugation to the monoglutamate form (removal of glutamyl residues in excess of one) must occur prior to releasing folate into circulation (Combs 1992). Tetrahydrofolate, 5-methyl-THF and 10-formyl-THF are the predominate folate derivatives found in the liver and are typically bound to glutamates varying in length from 4 to 7 (Groff et al. 1995). Most of the 5-methyl and 10-formyl derivatives are secreted into bile which are then reabsorbed via enterohepatic circulation or excreted via the feces. Approximately half of the total folate that reaches peripheral tissues occurs by this recirculation process (Steinberg 1984). Folate secreted directly into circulation is transported to other tissues bound to either low or high affinity folate binders as previously described. Recent studies from numerous laboratories suggest the existence of several distinct mechanisms for transport of folate across membranes (Henderson 1990). Folate transport into tissue cells can occur by a carrier-mediated process that may or may not require ATP (Groff et al. 1995) or by folate binding proteins that are associated with the plasma membrane of cells (Da Costa and Rothenberg 1996). These membrane-associated folate binding proteins act as folate receptors and thereby facilitate cellular uptake of folate (Da Costa and Rothenberg 1996). These receptors are attached to the plasma membrane via a glycosylphosphatidylinositol anchor and can also be released from the membrane

PAGE 30

17 as a soluble protein upon hydrolysis by a glycosyl phosphatidylinositol-specific phospholipase C (Lee et al. 1992). Although many cells do not express this membrane-associated FBP, high levels are found in placental trophoblastic cells, ovaries, fallopian tubes and mammary glands (Weitman et al. 1992). It was hypothesized that the membrane-associated folate receptor's of placental trophoblastic cells provides the following two advantages: 1) ensures adequate folate supply even under conditions of deficiency secondary to the receptors high affinity for folates; and 2) prevents potentially harmful folate analogues from entering the placental trophoblastic cells due to the high substrate specificity of the receptor (Selhub 1994). The folate receptors found in placental cells thereby appear to ensure adequate folate supply to the fetus even when maternal stores are depleted (Selhub 1994). Recently, an ex vivo placental cotyledon perfusion model was used by Henderson et al. (1995) to test the hypothesis that under physiologic conditions, placental folate receptors played a major functional role in transplacental folate transport. Their results indicated that maternal-to-fetal folate transfer is a two-step process. The first step concentrates 5-methyl-THF (or folic acid) by binding it to placental folate receptors on the maternally facing chorionic surface. Gradual release of 5-methyl-THF from this pool adds to incoming circulating folates to generate an intervillous blood concentration (placental compartment) three times that of maternal blood. The second step involves transfer of folates to fetal circulation along a downhill concentration gradient. Thus, folate bound to placental folate receptors are predestined for transplacental

PAGE 31

18 folate transport, since incoming (dietary) folates displace placental folate receptor bound folates that then passively diffuse to the fetus (Antony 1996). This study provided a mechanism for the observation that the developing fetus drains maternal folate supplies as evidenced by the significantly higher plasma and red cell folate concentrations of newborns compared to maternal concentrations (Ek and Magnus 1981). Folate receptors are also involved in folate transport across proximal renal tubular cells (Selhub 1994). Data suggest that after glomerular filtration, the luminal folate binds folate receptors in the renal tubular cells and is internalized rapidly via folate receptor mediated endocytosis. Dissociation of folate from the receptor occurs in the low pH of endocytotic vesicles allowing the return of folate to the blood, where it can be taken up by other cells, and the recycling of apofolate receptor back to the luminal brush-border membrane. Once within cells, 5-methyl-THF (predominate folate derivative in plasma or urine) is demethylated and converted back to a polyglutamyl form by folylpoly-y-glutamate synthetase in order to become a functional coenzyme, accepting and transferring one-carbon fragments (Wagner 1995). The addition of glutamyl residues also represents a way of keeping the folate in the cell. The mitochondrial folate pool is distinct from the cytosolic pool in that 5-methyl-THF is located in the cytosol, while 10-formyl-THF is in the mitochondria (Shane 1995). Tissues are limited in their ability to store folate beyond their normal cellular requirements. The total body content of humans is estimated to be

PAGE 32

19 approximately 10 mg, of which one-half resides in the liver (Herbert 1987, Combs 1992). Folate Functions Folate coenzymes are involved in numerous reactions that involve the transfer of one-carbon units as depicted by Figure 2. These one-carbon units exist at various levels of oxidation ranging from methane, the most reduced, to carbon dioxide, the most oxidized (Wagner 1995) and are carried by folate when the coenzyme is in the fully reduced THF form. As depicted by the diagram, folate is reduced in two steps to THF by dihydrofolate reductase which utilizes NADPH + H + The majority of folate derivatives are required for amino acid metabolism and deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) synthesis which is exemplified by the following reactions. Tetrahydrofolate can accept a one-carbon unit from serine forming glycine and 5,10-methylene-THF (A). This reaction is reversible via the donation of a hydroxymethyl group from 5,10-methylene-THF to glycine producing serine and THF. In mammalian systems, the 0-carbon of serine provides the majority of carbon units with THF serving as the carrier of these activated carbon atoms (Shane and Stokstad 1985). For example, the folate derivative, 5,10-methyleneTHF, produced from the one-carbon donation by serine to THF, can either be reduced in a nonreversible reaction by NADPH and FADH2 to 5-methyl-THF (B), oxidized to 5,10-methenyl-THF in a reversible reaction (F) or function as a one-carbon donor in a reaction required for DNA synthesis (G).

PAGE 33

20 Enzymes involved in interconversions of coenzymes forms of THF (A) Serine hydroxymethyltransferase (coenzyme-PLP) (B) Methylene-THF reductase (C) Methionine synthetase (coenzyme-B 12 ) (D) Formiminotransferase (E) Cyclodeaminase (F) Methenyl-THF reductase (G) Thymidylate synthetase (H) Formate-activating enzyme (I) Phosphoribosylglycinamide transform v [ase (J) Aminocarboxamide ribotide transform ylase (K) Cyclohydrolase NADPH+ H Homocysteine Methionine 5-methyl-THF Methotrexate Tetrahydrofolate ( THF) Q dUMP 5-formimino-THF Q 10-formyI-THF 10-methenyl-THF Figure 2. Functions and interconversions of coenzyme forms of THF. Modified from Groff JL, Gropper SS and Hunt, SM. Advanced Nutrition and Human Metabolism, New York: West Publishing Co. 1995; p 266. The folate derivative, 5-methyl-THF, can donate its methyl group to homocysteine to form methionine and regenerate THF (C). This reaction (C), catalyzed by methionine synthase, requires vitamin B 12 in addition to folate

PAGE 34

(Figure 3). Vitamin B 12 is bound tightly to methionine synthase and serves to acquire the methyl group from 5-methyl-THF and transfer it to homocysteine thereby producing methionine and regenerating THF. Vitamin B 12 deficiency induces a secondary folate deficiency which is best explained by the "methyl trap" hypothesis (Shane and Stokstad 1985). When the activity of methionine synthase is reduced due to lack of vitamin Bi 2 5-methyl-THF accumulates and prevents the regeneration of THF ultimately inhibiting the generation of other folate one-carbon forms. The megaloblastic anemia that may result can be explained by a lack of folate coenzymes for DNA precursor synthesis in blood cells (Shane 1995). Hence, megaloblastic anemia may develop not only as the result of a primary folate deficiency but also from a primary vitamin B 12 deficiency which causes a secondary folate deficiency. Histidine metabolism also relies on the availability of folate (D). Urocanic acid is generated by the deamination of histidine which can undergo further metabolism to yield formiminoglutamate (FIGLU). At this point, THF accepts the formimino from FIGLU producing glutamic acid and 5-formimino-THF, the latter can be converted to 5,10-methenyl-THF (E) and then reduced to 5,10methylene-THF (F). Hence, 5,10-methylene-THF, an important coenzyme in DNA synthesis, can be produced by two separate reactions (A) and (F). Thymidylate synthase requires 5,10-methylene-THF and deoxyuridylate (dUMP) to generate the pyrimidine, thymidylic acid and DHF (G). Tetrahydrofolate is regenerated from DHF by dihydrofolate reductase. Because both dihydrofolate

PAGE 35

22 reductase and thymidylate synthase are very active during cell division, they have been the target of chemotherapeutic drugs. For example, methotrexate binds the active site of dihydrofolate reductase which prevents the regeneration of THF thereby reducing the availability of 5,10-methylene-THF needed for cell division. Figure 3. The conversion of homocysteine to methionine and the regeneration of THF. Reproduced from Groff JL, Gropper SS and Hunt, SM. Advanced Nutrition and Human Metabolism, New York: West Publishing Co. 1995; p 266. Formate is required for the conversion of THF to 10-formyl-THF (H) which is the folate coenzyme needed for purine ring formation and consequently both RNA and DNA synthesis. Carbon 8 of the purine atom is generated by the formylation of glycinamide ribotide (GAR), a reaction involving 10-formyl-THF, to form formylglycinamidine ribotide (FGAR) and THF (I). Purine ring atom C2

PAGE 36

23 is acquired by formylation of 5-aminoimidazole 4-carboxamide ribonucleotide (AICAR) involving 10-formyl-THF to yield 5-formiminoimidazole 4carboxamide ribotide (FAICAR) and THF (J). The folate derivative, 10-formylTHF, can also be converted to 5,10methenyl-THF (K) and then reduced to 5,10methylene-THF as dictated by the needs of the cell. Turnover and Excretion of Folate Although studies conducted in both rat and humans confirm that folylpolyglutamates turn over in mammalian cells and tissues, the mechanism of turn over has not been well characterized (Shane, 1995). Since only a small proportion of supplemental or dietary folate is recovered in the urine when consumed in physiological amounts (Jukes et al. 1947), it has been speculated that the tissues alter these compounds in metabolism. In the late 1950s, Dinning and co-workers reported that upon incubating folic acid with liver slices, a "diazotizable amine" was formed. He predicted that the cleavage of folate to a "diazotizable amine" was one pathway for the metabolic disposal of folic acid. His experiments with rats supported his prediction when a "diazotizable amine" was found to be a major urinary excretory product following administration of either folic acid or 5-formyl-THF. Approximately half of the excreted metabolite was in free form, while the other half was N-acetylated. The investigators reported that cleavage to a diazotizable amine was a pathway of major importance for the metabolic disposal of folate based upon the magnitude of the amine

PAGE 37

24 excretion. Two decades later, administration of [ 3 H] and/or [ 14 C] labeled folic acid to rats revealed that folate catabolism takes place by cleavage of the C9-N10 bond, and not by excretion of inactive forms of the vitamin (Barford et al. 1978, Connor et al. 1979, Murphy et al. 1976, 1978 and 1979). The principle catabolites of a tracer dose of [ 3 H]pteroylglutamate (labeled on the benzene ring) or [ 14 C] pteroylglutamate (labeled on the pterin moiety) were apABG or derivatives of the pterin component, respectively (Barford et al. 1978, Connor et al. 1979, Murphy et al. 1976, 1978 and 1979). This mode of catabolism was subsequently confirmed in humans (Krumdieck et al.1978, Pheasant et al. 1979, Saleh et al. 1980). Krumdieck et al. (1978) administered [214 C] folic acid to a human volunteer and determined both urinary and fecal radioactivity, as well as urinary excretion of folate over a 129 day period of observation. Urinary pterin and isoxanthopterin were found to be radioactive which supported the mode of catabolism described in rats in which folate is cleaved in half producing pterin and pABG moieties. A substantial amount of [ 14 C]folate and derivatives were also detected in the feces indicating that this is an important avenue of elimination of folate and derivatives (i.e. catabolites) (Krumdieck, 1978). Pheasant et al. (1979) gave an oral dose of 2-[ 14 C], 3',5',9-[ 3 H]folic acid (60 ug or 5 mg folic acid) to five subjects and obtained urine collections over the following 24 h period. Urinary scission products including apABG were identified as the major urinary species following the low dose. Saleh et al. (1980) administered an oral dose of a mixture of [214 C]and [3',5',7,93 H]-Pte Glu to 14 subjects and

PAGE 38

25 identified the metabolites present in the urine collected 0-24 h after administration of the radioactive tracers. Their data clearly illustrated that orally administered folic acid is incorporated into the reduced folate pool in man and undergoes scission at the C9-N10 bond to yield metabolites qualitatively similar to those produced in rats. Both free pABG and acetylated pABG were detected in urine in these studies. It was hypothesized by Murphy et al. (1976) that both the oxidative cleavage and N-acetylation occur intracellularly and that free pABG may be the result of labile folates emerging from the cell and degrading in circulation or during passage into the bladder. Murphy et al. (1976) speculated that within the cell, folate is cleaved between the pteridine and the pABG moieties resulting in a mixture of pteridines and p-aminobenzoylpolyglutamate. This latter compound is recognized as an undesirable catabolite by the cell and is subsequently taken up by the lysosomes and converted to pABG. Acetylation of pABG follows and this compound is excreted into the urine. It was also hypothesized by Murphy et al. (1976) that folate deficiency so often associated with pregnancy may result from an increased catabolism of folate due to the presence of the fetus and not just from the increased folate requirement of rapidly growing tissue (Murphy et all 976 and 1978) or the increased excretion of the intact folate molecule (Fleming 1972, Landon and Hytten, 1971). In 1993, McNulty et al. conducted a study in rats to examine the possible role of increased folate catabolism as a contributory factor to the folate deficiency

PAGE 39

of pregnancy in rats. They utilized a newly developed high performance liquid chromatography (HPLC) method allowing for the quantitative determination of endogenous folate catabolites in rat urine. Three groups of female rats including one nonpregnant and two pregnant groups were investigated. One of the pregnant groups was allowed to feed freely whereas the other was pair-fed to the nonpregnant control group. Sources of exogenous (dietary) folate were kept to a minimum in order to limit excess serum folate concentrations, which could potentially contribute to the folate catabolites formed exogenously (i.e. folate break down in urine after excretion). The authors reported that pregnancy induces an elevation in the rate of folate catabolism in rats, even in the presence of low dietary folate. Pregnancy in rats caused a significant increase in urinary levels of the folate catabolite apABG compared with nonpregnant animals. Between the two pregnant groups, urinary apABG concentrations were consistently higher in the ad lib fed group which suggested that a certain amount of the observed elevation in folate break down was due to a greater food intake. However, because apABG concentration peaked prior to a reduction in weight at parturition for both pregnant groups (i.e. day 18 versus day 21, respectively), the observed increase in apABG was determined to be a real effect of pregnancy. The peak apABG concentration at day 18 was thought to reflect a change in placental growth from hyperplasia to hypertrophy. DNA synthesis, a folate requiring process, takes place during cell division (hyperplasia) and not during the increase in cell size. The authors suggested that increased folate catabolism during

PAGE 40

27 pregnancy was a consequence of increased generation of DHF, a labile folate coenzyme, resulting from participation in DNA synthesis (i.e. deoxyuridylic acid accepts one-carbon unit from 5,10-methylene producing thymidylic acid and DHF) (McNulty et al. 1993). Later, this hypothesis was expanded to include the increased generation of THF which is also a labile folate coenzyme resulting from the increased participation of 10-formyl-THF in de novo purine synthesis (McNulty et al. 1995). In contrast to apABG, urinary pABG concentrations were not detectable in this investigation and appeared to be dependent on dietary folate intake which was minimal. Again, it was reiterated that unacetylated pABG in urine may arise from the cleavage of intact folates during collection or analysis or may be the result of either incomplete acetylation of the catabolite in vivo or deacetylation of apABG generated in vivo (McNulty et al. 1993). The authors noted that pABG does not contribute a substantial amount to normal folate catabolite excretion, nor does it vary during pregnancy. No significant differences in urinary folate concentrations were found between the pregnant and nonpregnant rats indicating that increased losses of intact folates through the urine would not be contributing factors to folate deficiency. Overall, it was concluded that pregnancy induces an elevation in the rate of folate catabolism and that this may impose a drain on available folates, thus offering an explanation for folate deficiency associated with pregnancy. Following the rat study of McNulty et al. (1993), a similar approach involving measurement of urinary catabolite excretion was used by McPartlin et

PAGE 41

28 al. (1993) in human pregnant women. Six normal pregnant women provided 24 h urine samples at one time point during each trimester and postpartum. Six nonpregnant women of the same age were used as controls. The women entered a metabolic ward for 42 h where they were provided with a defined diet containing no exogenous source of pABG and a known folate concentration. The first 18 h was considered a wash out period for the folate breakdown products arising from normal diet (prior to entering metabolic ward) and urine was not collected. Throughout the final 24 h, urine was collected and subsequently analyzed for pABG and apABG by HPLC. Mean total 24 h excretion of apABG and pABG in micrograms were converted to folate equivalents on the basis that the molecular weight of either catabolite is approximately one-half intact folate. Urinary apABG, the predominant catabolite, expressed as folate equivalents was significantly greater in the second trimester compared to the first or third trimester, postpartum or in the non-pregnant women. The change in the nature of growth from hyperplasia to hypertrophy was provided as an explanation for these findings. McPartlin et al. (1993) concluded that 24 h urinary concentrations of folate catabolites reflected daily metabolic turnover of folate and should thus indicate dietary requirement. Folate Requirements and Recommended Dietary Allowances (RDAs) Since 1941, RDAs have been prepared by the Food and Nutrition Board and are intended to reflect the best scientific evaluation on nutrient allowances for the maintenance of good health and to serve as the basis for evaluating the

PAGE 42

29 adequacy of diets of groups of people (NRC 1989). Recommended dietary allowances are defined as the levels of intake of essential nutrients that, on the basis of scientific knowledge, are judged by the Food and Nutrition Board to be adequate to meet the known nutrient needs of practically all healthy persons (NRC 1989). Once the average minimal requirement for a nutrient has been estimated, adjustments are made to account for bioavailability, individual variation and the need for adequate reserves. Hence, a safety factor is incorporated into the RDAs which, in theory, ensures that the requirements of 98% of the population will be covered. For many people, the RDA may exceed their actual requirement by approximately 50%. RDAs are based upon a variety of study designs. Depletion-repletion studies determine nutrient requirements by measuring the amount of the vitamin needed to bring a deficient individual to a normal status. Nutrient balance studies measure nutrient status in relation to intake, and requirements are based upon the level of the vitamin needed to maintain normal status. Biochemical indices are often utilized in these study types. Requirements have also been estimated by quantitating how much of a specific nutrient is consumed by free living healthy populations who are maintaining normal nutrient status. The information derived from such studies dictate how the RDAs are revised approximately every five years. The folate RDA was decreased by 50% or more for all the age-sex groups in the 10 th edition (NRC 1989). One rationale for reducing the RDA in

PAGE 43

nonpregnant women was the observation that folate intake estimated by population surveys including the Second National Health and Nutrition Examination Survey (NHANES II) (LSRO 1984) was approximately 50% lower than the RDA reported in the 9 th edition of the RDAs (NRC 1980) and evidence of widespread folate inadequacy was lacking (Senti and Pilch 1984). There are limitations inherent in basing RDAs on estimated dietary folate intakes by normal populations. First, underreporting is a recognized limitation and would result in underestimation of requirements especially since actual intake may be twice as high as reported intake. Second, existing food composition tables underestimate the amount of folate contained in the food item. Recently modified folate extraction procedures from food including double extractions, pH optimizations and trienzyme treatments (Gregory et al. 1990, Martin et al. 1990, Tamura et al. 1997) improved folate recovery from the food matrix and yielded higher folate values when assayed microbiologically. For example, an extraction buffer with a pH of 7.85 was found to yield higher folate values than more acidic buffers because it prevents the partial interconversion and degradation of folates (Gregory et al. 1990, Wilson and Home 1984). Trienzyme treatment facilitates retrieval of folates trapped in the food matrix and increases measurable folate levels by 20% in whole wheat flour compared with conjugase digestion alone (Martin etal. 1990). The work of Sauberlich et al. (1987) was also used by the committee to support a reduction in the folate RDA, although it appears that these data were

PAGE 44

31 misinterpreted by the committee. Ten adult nonpregnant women followed a folate depletion regimen for 28 d followed by a repletion period. Plasma folate levels decreased throughout the depletion period and continued to fall until 200 p.g/d dietary folate was provided. The RDA committee concluded from these data that the 200 p.g/d level satisfied the criteria for establishing an RDA. However, 200 p.g/d represented the minimal dietary requirement for nonpregnant women and had not been adjusted to account for individual variation and allowance for storage. Sauberlich et al. (1987) suggested an intake of 300 u,g/d was necessary in order to provide an allowance for storage. In addition, the study was not carried out long enough to determine the amount of folate required to achieve acceptable serum or red cell folate concentrations. In 1992, Bailey suggested that the new RDAs for folate may not provide an adequate safety allowance for population groups at risk and should be reevaluated as new information evolves. Recently, two studies have been conducted which suggest that the current RDA for nonpregnant women is inadequate. O'Keefe et al. (1995) investigated the response of nonpregnant women of child-bearing age to three different levels of folate intake. Total intakes of 200, 300 or 400 jag folate per day were provided as a combination of dietary sources (30 ug) and either 170, 270 or 370 p.g/d synthetic folic acid. The consumption of 200 and 300 p.g/d resulted in significantly lower mean serum folate concentrations than the 400 p.g/d. The authors reported that women consuming folate intakes of 200 and 300 fxg/d were unable to maintain serum

PAGE 45

32 folate concentrations in the acceptable range (>3 ng/mL) throughout the duration of the 10 week study. In addition, plasma homocysteine concentrations changed inversely to serum and erythrocyte folate levels with evidence of elevated levels (>16 umol/L) in the subjects consuming the 200 ug/d level of intake. Based upon these data, the investigators concluded that 400 ug/d appears to more closely fulfill the criteria of a dietary allowance. Based on the urinary excretion of folate catabolites, apABG and pABG, McPartlin et al. (1993) also demonstrated that 180 ug/d for nonpregnant women was inadequate. Assay of 24 h urine collections for pABG and apABG revealed that upon conversion of the catabolites to folate equivalents and addition of urinary intact folate, the requirement to replace the catabolites and urinary excretion of intact folate was 99 22 ug/d. Following adjustments for bioavailability and the addition of two standard deviations, the RDA was computed as 280 ug for nonpregnant women. Since McPartlin and associates (1993) did not consider the other avenues of folate loss such as fecal excretion, folate requirements were underestimated resulting in an underestimation of the folate RDA. It is important to remember that urinary excretion of folate catabolites is just one component of folate requirements. Perhaps the most compelling reason to revisit the current RDA for folate in nonpregnant women is the recent Public Health Service (PHS) recommendation that all women of child bearing age consume 400 ug/d of folate in order to reduce their risk of having a pregnancy affected with spina bifida or other neural tube

PAGE 46

defects (NTDs) (CDC 1992). This recommendation was based upon numerous studies which convincingly illustrated the efficacy of 400 ug or more of folate in reducing the risk of NTDs (Czeizel and Dudas 1992, MRC 1991, Werler et al. 1993). The protective role of folate in reducing both the occurrence and the recurrence of NTDs has prompted the Department of Health and Human Services and the Food and Drug Administration (FDA) to require the fortification of all enriched grain products with folic acid as of January 1, 1998 (FDA 1996). It is estimated that the consumption of enriched grain products will increase folate intake by an average of 100 ug/d. Considering that the average diet provides an estimated 200 p.g/day, improvement of food choices or supplementation may still be warranted. The current RDA of 200 ug/d for adult males also appears to be inadequate based on the results of a recently conducted metabolic "depletionrepletion" study by Jacobs et al. (1994). Plasma homocysteine concentrations rose during folate depletion and correlated inversely with plasma folate concentrations. During folate repletion, the current RDA for adult men, 200 ug /d, was ineffective in normalizing high homocysteine concentrations resulting from folate depletion. The investigators concluded that the current RDA for adult men may not provide the expected margin of protection normally incorporated upon derivatization. Since the publication of the 1989 10 th edition of the RDA for folate, information has been compiled from effectively designed and controlled studies

PAGE 47

34 that strongly suggests a need for revision for both the adult male and nonpregnant female RDA. When the extra demands of pregnancy for folate are considered, the adequacy of the current RDA of 400 ug/d is questionable. At this time, metabolically controlled studies have not been conducted that investigate the response of pregnant women to defined levels of folate intake. The basis for the change in the RDA for folate from 800 ug to 400 ug/d during pregnancy stems from investigations that provided either (a) known amounts of supplemental folic acid to an uncontrolled diet or (b) provided folic acid to malnourished pregnant women. Chanarin et al. (1968b) investigated folate requirements in 206 women from approximately wk 15 to 38. The women were randomly assigned to receive either an iron supplement or an iron supplement with 100 ug of folic acid. Red cell and serum folates were determined by microbiological assay using Lactobacillus casei (L. casei) as the test organism. At baseline, no differences in serum or red cell folate concentrations existed between the two groups. However, as the study progressed, red cell folate concentrations declined over time in those not receiving 100 ug/d supplemental folic acid compared to maintenance of red cell folate concentrations in those women receiving supplemental folic acid. Dietary folate intake was estimated to be 676 ug/d based on the microbiological quantitation of food folate from one hundred eleven separate 24 h food collections which were duplicates of the actual meals consumed over 7 consecutive days (range 198 ug to 1,615 ug) (Chanarin et al. 1968a). The investigators concluded that folate requirements in pregnancy were met by a total folate intake of 776

PAGE 48

[ig/d which included the supplement of 100 ug of folic acid. This study supports the 1980 RDA (9 th edition) for folic acid of 800 ug/d and was cited as a reference by this RDA committee (NRC 1980). Bates et al. (1982) subsequently estimated dietary intakes of 190 jug/d by this population group (United Kingdom). This estimation in conjunction with Chanarin's report that 100 ug/d supplemental folic acid plus dietary folate maintained red cell folate concentrations was cited by the 10 th RDA committee in support of the 400 ug/d RDA. Another study cited by the RDA committee for supporting the 400 (ag/d level for pregnant women was the investigation conducted by Colman et al. (1975) in which African women during their last 30 days of pregnancy consumed maize meal fortified with folic acid for daily intakes of either 300 or 500 ug folic acid. Changes in folate nutritional status (serum and red cell folate) were compared to the changes observed in pregnant women who were receiving folic acid as: (1) 300 ug in tablet form; (2) maize meal without folic acid; or (3) maize meal containing 1000 ug. The last two comparison groups were from a previous study (Colman et al. 1974) which indicated that folic acid-fortified maize meal prevents the progression of folate depletion during pregnancy. In the current investigation, daily intakes of maize meal containing 300 ug added folic acid were effective in preventing the progression of folate depletion in late pregnancy. The RDA committee assumed that the effective dose was 168 ug of synthetic folic acid since it is generally agreed that food folate is approximately 50% available. Based partly on the results of Colman's study, the Committee

PAGE 49

36 contended that 400 |ig/d during pregnancy would be sufficient to build or maintain maternal folate stores and to keep pace with the increased folate needed to support rapidly growing tissue. This study, however, was not designed to assess folate requirements in pregnant women. It was conducted solely to determine the smallest amount of folic acid that, when added to maize meal, prevented folate deficiency in pregnancy. The authors did not state how much endogenous dietary folate was consumed on a daily basis (i.e. folate content of unfortified maize meal or other foods). This amount would have to be known in order to use this study as a basis for establishing an RDA for pregnant women. For example, if these women consumed 100 ug/d dietary folate from unfortified maize meal then their actual intake increases from 300 to 400 u,g/d. Individual variation and allowance for storage would also have to be considered. Another reason that this study is not suitable for basing an RDA for pregnant women is that the population group studied was nutritionally compromised and not representative of a normal healthy population. Although, maize meal fortified with 300 ug folic acid plus low intakes of endogenous folate may be effective in preventing the progression of folate depletion in late pregnancy in this population group, it is unknown whether it would promote adequate folate nutriture in wellnourished pregnant populations. The adequacy of maternal folate stores throughout gestation is an important consideration due to increased transfer of folate from maternal to fetal stores during the last weeks of pregnancy (Ek 1980, Ek and Magnus 1981). Women with inadequate folate stores are more likely to

PAGE 50

37 become deficient which may not become evident until the post-partum period. Martinez (1980) found that the best predictor of a woman's red blood cell folate concentration was the age of her youngest child. Since the publication of the 10 th edition of the RDAs for pregnant women, the study conducted by McPartlin et al. (1993) not only refuted the current RDA for nonpregnant women but also for pregnant women. They hypothesized that the increased requirement for folate during pregnancy may relate to its accelerated breakdown to apABG and its deacetylated derivative pABG due to folate's participation in DNA biosynthesis. The results of their study supported their hypothesis that folate catabolism is accelerated in pregnancy particularly during the second trimester. The investigators were also able to clearly demonstrate that the increased excretion of catabolites noted during the second trimester was independent of the increase in glomerular filtration rate which increased around the 8 th wk and remained elevated throughout the entire pregnancy. Recommended dietary allowances were computed by McPartlin et al. (1993) by converting both catabolites to folate equivalents, adding the small quantity of urinary intact folate and adjusting the total for bioavailability and population variance. The estimated RDAs based on the urinary excretion of folate catabolites and intact folates corresponded to 280 ug/d (nonpregnant and 1 st trimester), 660 ng/d (2 nd trimester) and 470 p.g/d (3 rd trimester). Again the RDAs have been underestimated as urinary excretion of folate catabolites is only one component of folate requirements. Limitations to this study included the fact that

PAGE 51

38 dietary intake was not controlled between hospital admissions and the 18 h "wash out" period may not have been sufficient for eliminating exogenous pABG arising from previous uncontrolled dietary folate intake. It is possible that the breakdown products are retained by the body for a longer duration than that allotted by the investigators, and if so, the catabolites found in the urine may not represent an endogenous source. The absorption and fate of folate catabolites derived from food remains to be determined. Furthermore, McPartlin and his coworkers assessed urinary excretion of catabolites at one time point only for each trimester, and data on other indices of folate status were not reported. Another possible limitation of this study is the inclusion of unacetylated pABG along with the major catabolite, apABG. However, until the origin of pABG is known, it seems prudent to include this catabolite in the calculations. McPartlin and coworkers analyzed each catabolite separately, so as information becomes available, appropriate adjustments can be made to their interpretation of the data if necessary. The importance of adequate folate nutriture and appropriate folate intake during pregnancy is widely recognized, yet the data we have for basing recommendations regarding folate requirement during pregnancy, is plagued with limitations. A controlled metabolic study which investigates the response of pregnant women to defined levels of folate intake utilizing a variety of folate indices to assess folate status response is long overdue.

PAGE 52

39 Folate Deficiency and Pregnancy Since folate governs the synthesis of the precursors of DNA and RNA, the formation and development of every human cell is dependent upon an adequate supply of this vitamin. Pregnancy represents a state of rapid cell division involving both maternal and fetal tissue. In a relatively short period, a two celled embryo develops into a two billion celled fetus (Hibbard 1964). This occurs simultaneously with the rapid proliferation of various maternal tissues including the uterus and the hematopoietic system (Hibbard 1964). The increased demand that pregnancy imposes on folate nutriture is reflected in the doubling of the RDA from nonpregnant conditions. Cooper et al. (1970) concluded that 500-600 (ag/d of dietary folate was required to prevent all evidence of folate deficiency in pregnancy and many women can not meet the increased requirement for folate through diet alone. Poor nutrition has been recognized as a major factor in the development of folic acid deficiency since the 1960s (Stone et al. 1967). The decline in folate status (both serum and red cell folate) during pregnancy has been well-documented in both developed (Bailey et al. 1980, Cooper et al. 1970, Ek et al. 1981, Lowenstein et al. 1966, Qvist et al. 1986) and underdeveloped countries (Colman et al. 1975). Bailey et al. (1980) conducted a study to investigate folate status at the initiation of prenatal care in 269 pregnant low-income subjects. They found that serum folate was low (3 to 6 ng/mL) and deficient (< 3 ng/mL) in 48% and 15% of the subjects, respectively and 29% of the subjects had subnormal (<140 ng/mL) red cell folate concentrations. Qvist et al. (1986) reported

PAGE 53

decreasing plasma folate levels with advancing pregnancy so that, at term, almost 50% of the women had subnormal plasma folate concentrations which persisted over the following 2 months. The investigators also found that 10% of the women had subnormal red cell folate concentrations at term which increased to 33% during the post-partum period. The authors contended that this observation could be explained by the delayed reflection of folate stores by the red cell folate values in periods of increased demand (Qvist et al. 1986). Ek and Magnus (1981) observed similar folate status changes throughout gestation in 43 women. They found a positive correlation between plasma folate and red cell folate concentrations at different stages of pregnancy supporting the delayed reflection of folate stores by red cell folate concentrations (Ek and Magnus 1981). The problems associated with folic acid deficiency have been recognized since the late 1950s. The most overt sign of folate deficiency is megaloblastic anemia which is most commonly observed in the late stages of pregnancy. Earlier indications (preanemia) include serum folate concentrations below 3 ng/mL and red cell folate concentrations below 140 ng/mL. Folate deficiency has been associated with low birth weight, cessation of pregnancy, delayed maturation of the nervous system, growth retardation, megaloblastic anemia and congenital anomalies (Hibbard 1964, Shojania 1964, Strelling et al. 1966, Baumslag et al. 1970, Iyengar and Rajalakshmi 1975, Smithells et al. 1976). Recently, O'Scholl et al. (1996) conducted a prospective study examining the influence of folate intake on two pregnancy outcomes associated with poverty and disadvantage:

PAGE 54

preterm delivery and infant low birth weight (LBW). The investigators reported that women with a low mean daily folate intake (< 240 ug/d) had an approximately twofold greater risk of preterm delivery and incidence of infant low birth weight after controlling for maternal characteristics, energy intake and other related nutrients. In addition, lower concentrations of serum folate at wk 28 were also associated with a greater risk of preterm delivery and low birth weight. Although overt megaloblastic anemia is infrequent in the United States, it seems desirable to consume enough folate to maintain maternal stores to keep pace with the increased demand for folate during pregnancy. Folate Supplementation During Pregnancy Several studies have been conducted to determine the amount of supplemental folic acid required to maintain normal folate status during pregnancy (Cooper et al. 1970, Dawson 1966, Hansen and Rybo, 1960, Lowenstein et al. 1966, Willoughby and Jewel 1966). Unfortunately, dietary folate intake was not controlled and interpretation of these data is difficult. Willoughby (1967) and Willoughby and Jewel (1968) reported that 300-350 ug/d synthetic folic acid in addition to diet was able to maintain normal folate status in pregnant women throughout gestation, whereas supplemental amounts of either 100 or 200 (ag/d were inadequate. Dawson (1966) also reported that 150 |ag/d folic acid plus dietary folate was insufficient to maintain normal serum folate concentrations. In contrast, Chanarin et al. (1968b) reported that 100 p.g/d supplemental folic acid, in addition to diet, was adequate. Lowenstein et al.

PAGE 55

(1966) found that consumption of 500 fig/d synthetic folic acid plus diet resulted in higher serum folate concentrations in pregnant women compared to nonpregnant controls (1 1.0, 7.0 ng/mL, respectively) and therefore may be above the minimal daily requirement. Cooper et al. (1970) concluded that the decline in folate status during pregnancy could be prevented with 200-300 p.g folic acid/day in addition to diet. Overall, these data support a supplementation level of approximately 300 ug/d. Programs are in place to aid pregnant women in obtaining a sufficient amount of folate. Unfortunately, not enough is known about folate metabolism and requirements during pregnancy to provide women with the optimal amount, and problems arising with oversupplementation of folate have been raised. The biggest reluctance to give prophylactic folic acid during pregnancy is due to the possible risk of masking a developing B 12 deficiency. However, since the incidence of pernicious anemia is low among women of child-bearing age, the masking of a B 12 deficiency leading to neurological complications would be extremely rare (Tamura 1995). Another concern is the possibility of interfering with zinc absorption. In 1984, Mukherjee et al. conducted a study involving 450 pregnant women and reported that high maternal plasma folate concentrations during pregnancy were associated with an increased rate of pregnancy complications and fetal distress They reasoned that folic acid supplementation adversely affected the intestinal absorption of zinc which resulted in the complications. In contrast, Tamura et al. (1992) found high serum folate

PAGE 56

43 concentrations were associated with favorable pregnancy outcome and reported higher birth weight and Apgar scores of newborns, reduced prevalence of fetalgrowth retardation, and lower incidence of maternal infection around delivery. Recently, Kauwell et al. (1995) conducted a study to determine the effect of supplemental folate on zinc nutriture and found that folic acid supplementation did not adversely affect zinc status. Overall, the detrimental effects of folic acid supplementation on pregnancy outcome are speculative considering supplementation has been in use for the past 40 y with no identifiable adverse effects. However, because the effects of oversupplementation with folic acid on the developing fetus are unknown (Scott et al. 1991) and the fetus has demonstrated its ability to sequester folate from maternal circulation, high supplemental doses of folic acid (> lmg) should be avoided under normal circumstances. Neural Tube Defects Neural tube defects (NTDs) are a group of serious malformations which affect the developing nervous system and include anencephaly, spina bifida and encephalocele. Approximately 2,500 babies with these abnormalities are born each year in the US which represents approximately 1 to 2 NTD affected infants per 1000 births; many more are aborted either spontaneously and electively (Crandall et al. 1995). Neural tube defects arise within one month of conception and over 90% are born to women without a prior family history of this disorder (Crandall et al. 1995). Although the cause of NTDs is unknown, it is now well

PAGE 57

44 established that maternal folic acid supplementation during the periconceptual period significantly reduces the risk of giving birth to a baby with a NTD. Several studies have confirmed that for women without a prior history of NTDs, a 60 to 100% decrease in risk was obtained if folic acid was taken prior to conception and throughout the first trimester (Milunsky et al.1989, MRC 1991, Mulinare et al. 1988). Studies which have investigated the effect of folic acid supplementation on the recurrence of NTDs have also reported similar reductions after folic acid supplementation (MRC 1991, Laurence et al. 1981, Smithells et al. 1983, Vergel et al. 1990). The effective dose of folic acid ranged between 0.4 mg to 5 mg/d. The neural tube closure occurs within 28 d after conception, a time when many women do not even know they are pregnant. Hence, NTD prevention is best achieved by adequate folic acid intake throughout the reproductive years (Crandall et al. 1995). In 1992, the Public Health Service recommended that all women of childbearing age in the United States who are capable of becoming pregnant should consume 0.4 mg of folic acid per day for the purpose of reducing their risk of having a pregnancy affected with an NTD (CDC 1992). Daly et al. (1995) subsequently demonstrated that red cell folate concentrations of 400 ng/mL or higher were associated with a low risk of folate-responsive NTDs. Brown et al. (1997) found that red cell folate concentrations > 400 ng/mL could be achieved by folate intakes of at least 450 ug/d (supplement users) or 500 ug/d (food and folic acid fortified cereals, only). A limitation to the study of Brown et al. was that dietary estimates of folate consumption were obtained during the

PAGE 58

study period and not 2-3 months earlier during folate incorporation into reticulocytes. The average consumption of dietary folate by women in the United States is approximately 200 ug/d (National Academy of Sciences 1990). It is possible for women to increase their consumption of dietary folates by increasing their consumption of fruits, vegetables and whole grains. Unfortunately, for various reasons, adherence to dietary guidelines is not widespread. Fortification of staple foods stuffs with folic acid is an effective and inexpensive approach to assist all women in increasing their daily intake of folic acid. Recently, the FDA amended the standards of identity for enriched cereal-grain products and has given manufacturers two years to comply with its decision requiring the fortification of enriched grain products with folic acid (FDA 1996). The level of folic acid fortification permitted (140 (ag per 100 g of cereal grain product) is low enough to keep daily folate intake below 1 mg yet high enough to enhance folate intake by an average of 100 ng/day. As stated before, the cause of NTDs is unknown at this time. Davis et al.(1995) found that the absorption of folic acid routinely consumed in supplements and fortified food products was not impaired in women with a history of a pregnancy with a NTD. Recent evidence suggests impairment of homocysteine remethylation resulting in a possible methionine shortage at a crucial stage of development (Mills et al. 1995). The investigators suspect a defect in methionine synthase, the enzyme responsible for converting

PAGE 59

46 homocysteine to methionine. Methionine synthase requires both the presence of folate (5-methyl-THF) and vitamin B 12 It is a key enzyme in many methylation reactions, including the production of myelin, and it also catalyzes the synthesis of THF, a compound required for DNA synthesis, following its conversion to 5,10-methylene-THF. Further, homocysteine, itself, is reportedly a putative embryotoxin (Lucock et al. 1995). An earlier study conducted by Kirke et al. (1993) found that both plasma folate and plasma B 12 were independent risk factors for NTDs, and they suggested that vitamin Bi 2 should be considered, as well as folate, in any supplementation program aimed at preventing NTDs. In lieu of the recent discovery of elevated homocysteine concentrations and the possibility of impaired homocysteine metabolism, supplementation with B 12 in addition to folic acid, may result in a further reduction of NTDs. However, the findings of van der Put et al. (1995) of a mutation in the gene coding for methylenetetrahydrofolate reductase (MTHFR) as a risk factor for spina bifida would support folic acid supplementation alone. Reduced MTHFR activity would lead to lower 5-methyl-THF concentrations which would result in elevated homocysteine concentrations. However, this is a common gene mutation in the general population and is not a definitive marker for those at risk of delivering a baby with a NTD. An association between NTDs and obesity, independent of the effects of folic acid intake, has also been reported (Werler et al. 1996). These investigators collected data from 1988 to 1994 in a case-control surveillance program of birth

PAGE 60

defects and found a positive association between prepregnant weight and NTD risk which was not alleviated by adequate folate intake among the heavier women. The authors noted that the mechanism by which increased weight may affect NTD risk is not clear but speculated that their diets may be deficient in nutrients (other than folate) that may be necessary for closing of the neural tube or that obesity affects some metabolic factors which may have a direct effect on neural tube development (i.e. insulin). Assessment of Folate Status Assessment of folate nutriture during pregnancy is complicated by the physiological changes that occur over this nine month period. A healthy woman bearing a normal-sized fetus will increase her plasma volume from 25% to 80% with an average value of 40% (Bartels et al. 1989). Plasma volume begins to rise during the latter part of the first trimester, reaches a maximum at approximately 25 wk gestation and remains relatively constant until the 34 th wk of gestation (Bartels et al. 1989). As a result of this hemodilution, the concentration of several nutrients, including folate, decline. Unfortunately, little is known as to what constitutes normal plasma folate concentrations in pregnancy. As O'Connor (1994) emphasized in a recent review, it is difficult to discern the meaning of a low serum folate concentration during pregnancy secondary to the physiological changes that occur during this time and a reduction in serum folate may be normal. Hall et al. (1976) found that the fall in serum folate concentration was due principally to plasma volume expansion and should therefore be regarded as

PAGE 61

physiological. However, women consuming supplemental folic acid (-300 ug) plus low dietary folate are able to maintain serum folate concentrations within the normal range for nonpregnant women (Cooper et al. 1970, Willoughby 1967, Willoughby and Jewel 1968). Currently we can only evaluate the nutriture of pregnant women against nonpregnant standards and based on the critical role folate plays in producing a healthy baby and the link between low serum folate concentrations and folate deficiency, it seems desirable to maintain serum folate concentrations in the range deemed appropriate for nonpregnant women. Altered renal function is another physiological change which makes assessing folate status during pregnancy a challenge. In the non-pregnant state, the renal tubules are very effective in reabsorbing water-soluble vitamins like folate. During pregnancy, however, Landon and Hytten (1972) reported that urinary folate excretion increases approximately four fold. Several folate status indices are available to clinicians and researchers to obtain information on folate nutriture in both pregnant women and the general population. Importantly, all have certain limitations which must be considered. Determinations of serum and erythrocyte folate concentrations are commonly performed and represent the most practical approach for population studies. The main folate derivative in serum is the reduced, monoglutamate form, 5-methylTHF (Gibson 1990). Serum does not contain polyglutamates. Serum folate concentration is reflective of acute folate status that changes in response to current dietary intake. For example, serum folate values increase rapidly after the

PAGE 62

49 ingestion of folate (dietary or supplemental) and decrease rapidly on a folate deficient diet. Hence, serum folate concentration is not a valid index of long term folate status or tissue stores unless it is used in conjunction with controlled dietary intake over time. Furthermore, normal serum folate concentrations have been documented in individuals with known folate deficiency and low levels have been found in individuals with normal status (Lindenbaum and Allen 1995). Several non-nutritional factors can also affect serum folate concentrations. An acute drop in serum folate results after alcohol ingestion because reabsorption of folate by the kidney is impaired (Hillman et al. 1977). Smoking has also been shown to lower serum folate concentration possibly by enhancing erythropoiesis, which in turn, increases folate requirements (Gibson 1990). An elevated serum folate concentration may result from hemolysis of red blood cells occurring during phlebotomy. Stagnant-loop syndrome, acute renal failure and liver damage may also result in elevated serum folate levels (Gibson 1990). Serum folate values below 3.0 ng/mL are considered to reflect deficiency; values between 3.0 6.0 ng/mL reflect borderline deficiency; while values above 6.0 ng/mL are considered to be within an acceptable range (6.020 ng/mL). In contrast to serum folate concentration, folate within the erythrocyte is not influenced by hemodilution (Bailey 1990) and therefore, provides a better measure of folate nutriture during pregnancy. However, it is important to remember that the red cell folate concentrations reflect folate available for incorporation in the red cell at the time of erythropoiesis and, therefore, represents

PAGE 63

50 folate status 2 to 3 months prior to the time of blood sampling. Red cell folate concentrations are less sensitive to short-term fluctuations in folate status than serum folate levels and are therefore, a good indicator of long term folate status. Because changes in erythrocyte folate concentrations take several months to reveal any significant changes in folate nutriture, they cannot be used to assess recent changes in folate nutriture. Red cell folate concentrations are a more reliable index of folate status than serum folate concentrations and are the preferred biochemical measurement for population studies. A major limitation, however, of erythrocyte folate levels is they cannot be used to distinguish between megaloblastic anemia due to folate deficiency and that due to a B 12 deficiency or a combination deficiency. However, low folate levels found both in the red cell and serum or plasma is highly suggestive of a folate deficiency (Sauberlich 1995). Red cell folate concentrations below 140 ng/mL are indicative of folate deficiency. Values between 140 160 ng/mL are borderline while values above 1 60 ng/mL are within acceptable range. Serum and erythrocyte folate concentrations can be assayed using either microbiological or radioisotope dilution techniques. The microbiological assay is considered to be the gold standard and is widely accepted for assaying folate concentrations in biological samples including blood and urine, as well as folate concentrations in tissues and food. The test organism most commonly used is L. casei because it responds to the greatest number of different folate derivatives, including those with up to three L-glutamic acid residues (Herbert, 1987b). Other

PAGE 64

51 microorganisms that have been used include Streptococcus faecium which does not respond to 5-methyl-THF and Pedicococcus cerevisiae which does not respond to either folic acid or 5-methyl-THF (Tamura 1990). A major limitation to the microbiological assay is that it cannot be used for serum or erythrocyte samples containing antibiotics or methotrexate because these substances inhibit bacterial growth. Since the polyglutamate form of folate is present in erythrocytes, an incubation period is required to ensure complete hydrolysis to the mono, di or tri glutamate form by the conjugase enzyme. Ascorbic acid is added to biological samples to act as a reducing agent and prevent the oxidation of the folate. Pfeiffer and Gregory (1996) recently determined that the pH of the biological sample is important during the incubation period since the human serum pteroylpolyglutamate hydrolase (conjugase) enzyme functions optimally at pH 4.5 (Krungkrai 1987). For this reason ascorbic acid should be used as the preserving agent in the red blood cell hemolysate as it yields a final pH of approximately 4.0 rather than the sodium ascorbate which yields a higher pH (Pfeiffer and Gregory 1996). The radioisotope dilution technique is not as widely accepted as the microbiological technique but is used frequently in the clinical setting because it is fast and is not affected by antibiotics, tranquilizers, and or antimitotic agents which can interfere with the growth of L. casei (Gibson 1990). This technique uses a folate binding protein derived from milk and [ 125 I]-labeled folic acid or methyl-tetrahydrofolate. The unlabeled folate present in the biological sample

PAGE 65

52 competes with the labeled folate for available binding sites on the folate binding protein. Charcoal is used to adsorb both the unbound labeled and unlabeled folate which is removed later by centrifugation. A scintillation counter is used to measure the amount of bound labeled folate existing in the supernatant. The decrease in the radioactivity is proportional to the folate concentration in the sample. High performance liquid chromatography techniques have also been developed for the assessment of folate in biological samples (Gregory et al. 1984, Selhub 1990). Formiminoglutamate (FIGLU) excretion is a functional index of folate nutriture. Normally histidine is converted to glutamic acid via the action of formimino transferase and its coenzyme tetrahydrofolate. Because folate is required in this reaction, a folate deficiency will result in the accumulation and excretion of FIGLU in the urine particularly after a histidine load. Although the FIGLU test is highly sensitive, it is not specific for folate. Deficiency of vitamin B ]2 or the enzyme formimino transferase will also produce high urinary excretion of FIGLU after a histidine load. The use of the FIGLU excretion test in pregnant women is controversial. Rothman (1970) reviewed several studies showing increased FIGLU excretion in early pregnancy with a progressive decrease in the level toward term and a second increase post partum. Hence the use of the FIGLU excretion as a functional index of folate status during pregnancy is not widespread.

PAGE 66

Polymorphonuclear leukocyte lobe count is also utilized as a functional index. Neutrophils normally have three to four segments. In megaloblastic anemia hypersegmentation occurs increasing the number of lobes and is a characteristic early feature of folate and vitamin B I2 deficiency (Colman 1981). Neutrophil segmentation also occurs under other conditions and is thus not very specific. However, for persons who may be both iron and folate deficient, evaluation of neutrophil hypersegmentation is particularly helpful in diagnosing the folate deficiency that is being masked by the co-existing iron deficiency (Das etal. 1978). Deoxyuridine suppression test (dUST) is yet another way to assess functional folate status. This is an in vitro test performed on short term cultures of proliferating cells. The test is based on the need for folate in the methylation of deoxyuridine to thymidine. Bone marrow cells, lymphocytes or whole blood are preincubated with unlabeled deoxyuridine. Tritiated thymidine is then added to the cells and if folate deficiency exists, [ 3 H]thymidine will not get incorporated into DNA. Sauberlich et al. (1987) found that this test did not correlate well with either plasma or red cell folate concentrations of non-pregnant women. Furthermore, the use of this test has never been validated using pregnant or lactating women (O'Connor 1994). Determining plasma homocysteine concentration is gaining momentum as a functional index of folate status. Plasma homocysteine concentration is inversely correlated with serum folate and red cell folate concentrations and has

PAGE 67

proven to be a sensitive, although not specific, index of folate status because plasma homocysteine concentrations are also elevated with a vitamin B 6 or B 12 deficiency. Stable-Isotope Methods Isotopes are forms of an element that have the same number of protons (atomic number), but which differ in the number of neutrons and thus atomic weight. Because an atom's chemical characteristics are defined by its atomic number, isotopes behave similarly. Stable isotopes do not undergo nuclear decay and are therefore, nonradioactive. Both stable and radioisotopes act as tracers in that they follow or outline the pathway or fate of a tracee. In order to be an effective tracer, the isotopically labeled compound must act the same as the nonlabeled species in the system of interest. Gregory et al. (1990) conducted a study to determine whether stable-isotope-labeled folates could be employed for studies of folate absorption and metabolism in human subjects. After a saturation regimen, eleven adult males consumed 677 nmol (300 u.g) each of 3',5'-labeled bideuterofolic acid and glutamate labeled tetradeutero folic acid. The molar ratio of labeled folates in the urine were compared to the molar ratio in the ingested dose and found to be similar (not significantly different). In addition, the percentage of labeled folate in the urine over a forty-eighty hour period approached the percentage of the ingested labeled folate. These data revealed that the two labeled forms of folate behaved similarly in vivo when compared to each other and when compared to unlabeled folate.

PAGE 68

55 The interest in utilizing stable isotopes as tracers resurfaced in the 1970s primarily due to the interfacing of gas chromatography with mass spectrometry. The power of gas chromatography as a separation technique made it possible to analyze biological samples in complicated matrixes. In order to analyze a sample by gas chromatography mass spectrometry (GCMS), the sample must be volatile and heat stable. For this reason organic compounds are often derivatized. Prior to derivatization, folate is cleaved and the pABG fraction which contains the label is isolated. Trifluoroacetic anhydride and trifluoroethanol are added forming a lactam of (N-trifluoroacetyl-a-trifluoroethyl)-p-aminobenzoylglutamate as the major product (Toth and Gregory 1988). This compound is well suited for GCMS analysis because it is volatile and stable. The isotope ratios (i.e., D2 to DO) or the enrichment of the sample (refers to the abundance of the label in excess of natural abundance) can then be determined in the biological sample. There are several advantages to using stable isotopes with safety being the most important one. The only possible toxicity that may result from substitution of a stable isotope for its lighter and more abundant isotope is an "isotope effect" which is a slowing of biochemical reactions because of the greater mass of the stable isotope (Van Langenhove 1986). Browne (1986) contended that concern about toxicity from stable-isotope labels should not limit the use of stable-isotopelabeled drugs, because there is now abundant evidence that the amount of stableisotope label (including deuterium) that can be accumulated with repeated administration of even large amounts of drug does not pose any risk to humans.

PAGE 69

56 Importantly, due to the increased sensitivity and specificity that GCMS analysis provides, only minimal amounts of tracers are required and they can be chronically administered to the same subject. In addition, a molecule can be labeled with more than one different isotope and each one can be detected by GCMS. This concept enables us to study the bioavailability of pABG labeled 13 1 with [ C] at the same time we are investigating [ H2]folic acid turnover. However, the availability of commercially labeled stable isotopes is limited, so synthesis is a major consideration (Gregory and Toth 1990).

PAGE 70

CHAPTER 3 STUDY DESIGN AND METHODS Study Design A two-by-two factorial study design was used in which pregnant women (n=12) and nonpregnant controls (n=12) were randomly assigned to consume folate intakes of either 450 ug/d or 850 ug/d for 84 d. These intakes approximated the current and former RDA for pregnant women. Four experimental groups (n=6) were thus established: pregnant subjects fed either 450 or 850 ug/d and nonpregnant subjects fed the same two levels. Subjects consumed folate as a combination of food folate (120 15 ug/d) and synthetic folic acid (either 330 or 730 ug/d) taken with meals. Approximately fifteen percent of the total folate intake was provided as deuterated folic acid (D2) during the first one-half of the study then switched to 100% unlabeled folic acid (DO) at the beginning of wk 7 allowing for future investigation of folate kinetics. In addition, 4 subjects (one from each group) consumed [ 13 C]pABG at the beginning of wk 7 to provide information as to the fate of exogenous pABG. Blood and urine were collected at baseline and thereafter on a weekly basis for 12 wk (wk 14-25 gestation for pregnant subjects) except during wk 7 and 8 when multiple collections were obtained. Folate status was assessed by measurement of serum 57

PAGE 71

58 folate, red cell folate, urinary 5-methyl-THF, urinary folic acid and urinary pABG and apABG. After completion of the study, a subsample of our subjects consumed either 200 p.g/d (4 pregnant; 3 nonpregnant) or 600 ug/d (4 pregnant; 4 nonpregnant) synthetic folic acid plus uncontrolled dietary folate and folate and over the following 3 months returned to the clinical research center for monthly blood draws. See Figure 4 for an overview of the protocol. N p P(wk gestation) CONTROLLED INTAKE Follow U p 0 12 3 13 14 15 16 4 5 17 18 6 7 8 9 10 1112 19 20 21 22 23 24 25 1 r 17 2 0 2 4 2 9 3 3 3 7 1 Daily FA 3 Supplem ent [ 1 3 C ] p A B G B lo o d 5 D 2 — DO DO 2 4 -h Urine Figure 4. Overview of Protocol. 'Controlled dietary folate (120 jag/d); Uncontrolled dietary intake; 3 Two levels of supplemental folic acid (330 and 730 p.g/d) during the 12 wk metabolic study and two levels (200 and 600 p.g/d) during the follow-up study; 4 Administration of 13 C-pABG to 4 subjects; 5 One blood draw/wk except following the switch to unlabeled folic acid then six collections over 2 wk and except during follow-up period then 1 draw/mo; and 6 One 24 h urine collection/wk except following switch to unlabeled folic acid then seven collections over 2 wk. NP = Non Pregnant; P = Pregnant

PAGE 72

59 Subjects Healthy pregnant female subjects (18-35 y, 14 wk gestation) and nonpregnant controls (18-35 y) with normal blood chemistry profiles, normal blood folate concentrations and normal health status as determined by medical histories were eligible for participation. Exclusion criteria included chronic drug (including oral contraceptives and folate antagonists), alcohol or tobacco use. The majority of pregnant subjects (n=10) were consuming prenatal vitamins containing folic acid (0.4 -1 mg) prior to starting the study. Gestational age in pregnant subjects was determined by sonogram in conjunction with the date of last menstrual period. Approval of the study protocol by the Institutional Review Board of the University of Florida and signed informed consents by participants were obtained. Compliance to the study protocol was ensured by close daily contact in a positive environment monitored by the research team who observed consumption of folic acid with meals. Nonpregnant women maintained their body weight within 5% of base line throughout the study period and pregnant women gained approximately 0.45 kg per week. Diet A 5-d menu cycle consisting of five dinners and three breakfasts and lunches was designed as detailed in Table 1. Conventional foods were selected to provide meals that were varied and palatable. Folate content was minimized by thrice boiling chicken, ground beef, green beans and white potatoes and by using

PAGE 73

Table 1 Five-day cycle menus. Day 1 Day 2 Day 3 Waffles Margarine Syrup Cranberry juice Shredded Wheat Milk Raisins Brown sugar Apple Sauce Cranberry juice Breakfast Biscuits Margarine Jelly Canned pears Cranberry juice Lunch Day 4 Waffles Margarine Syrup Cranberry juice Day 5 Shredded Wheat Milk Raisins Brown sugar Apple Sauce Cranberry juice Sandwich, Pita Turkey breast Cheese Mayonnaise Canned pears Granola bar Sandwich, Biscuit Ham Cheese Mustard Canned peaches Apple newtons Crackers Tuna Mayonnaise Raisins Chocolate chip Cookies Cola Dinner Sandwich, Pita Turkey breast Cheese Mayonnaise Canned pears Granola Bar Sandwich, Biscuit Ham Cheese Mustard Canned peaches Apple newtons Casserole, Chicken Rice Cheese soup Margarine Green beans Newtons Apple juice Turkey Rice Gravy Green beans Margarine Icecream Chocolate syrup Apple juice Casserole, Ground beef Onion Cream of mushroom soup Margarine Green beans Frozen juice flavored bar Macaroni and Cheese Turkey Ham Green beans Margarine Ginger Snaps Chicken breast Barbeque sauce Potato Margarine Green Beans Canned pears Devil's food Cookies Apple juice Oreo cookies Milk Yogurt Popcorn Oatmeal cookies Milk Yogurt Popcorn Snacks Graham crackers Milk Yogurt Popcorn Granola bar Milk Yogurt Popcorn Rice cakes Milk Yogurt Popcorn

PAGE 74

61 canned fruits and vegetables and refined starches. The menus were analyzed for their folate content in our laboratory and provided a mean of 120 15 u.g/d. Energy and all other nutrients were analyzed by the Minnesota Nutrient Data System version 2.7 (Nutrition Coordinating Center 1994). Table 2 includes the nutrient composition of the three menus. The menu cycle provided 2500 calories/d of which 60% came from carbohydrate, 25% from fat and 15% from protein. A custom formulated supplement (Tischon Corp., Westbury, NJ) was used to provide the RDA for all essential nutrients not met by the diet except folate. Loss of water soluble vitamins/electrolytes in the boiled food items was accounted for in the supplement. Caloric energy was modified to maintain weight or ensure weight gain in nonpregnant and pregnant subjects, respectively. Examples of non-nutritive food sources included margarine, candy, Jello, CoolWhip and sweetened or unsweetened beverages. Table 2. Nutrient composition of 5-d cycle menus 1 2 Day 1 Day 2 Day 3 Day 4 Day 5 Energy, kcal 2540 2791 2472 2603 2490 Protein, g 83 85 85 78 88 Fat,g 60 65 81 74 50 Folate, u.g 114.6 134.7 99.2 118.2 135.1 Supplementation for pregnant subjects was provided by Tischon Corporation (Westbury, New York) -0.4 mg thiamin mononitrate, 18 mg iron fumarate, 4.5 mg zinc sulfate, 230 RE retinyl palmitate, 3 TE d-a tocopheryl acetate, 120 IU Vitamin D3, 7 mg niacinamide, 1.5 mg pyridoxine HC1, 16 ug phylloquinone. 2 Supplementation for nonpregnant controls was provided by Tischon Corporation (Westbury, New York)3 mg iron fumarate, 1.5 mg zinc sulfate, 230 RE retinyl palmitate, 120 IU Vitamin D3, 5 mg niacinamide, 0.9 mg pyridoxine HC1, 14 ug phylloquinone.

PAGE 75

62 Folate Extraction from Menus The three day cycle breakfast and lunch menus and the five day cycle dinner menus were analyzed for folate content three separate times during the study period using the method described by Gregory et al. (1990) and Martin et al. (1990). Each menu was prepared as though it was being consumed by the subjects. The food items were weighed, blenderized and aliquoted into 10 mL tubes. The blenderized samples were protected from light and stored at -20 C until analyzed for folate content. Six grams of blenderized menu were extracted in 20 mL boiling buffer containing 50 mM Hepes/50 mM Ches, pH 7.85 with 2% ascorbate and 0.2 M 2-mercaptoethanol. The buffer containing the sample was placed in a boiling water bath for 10 minutes and then cooled on ice for 10 minutes. Following homogenization using a Polytron (Type PI 00/35, Brinkman Instruments, Westbury, NY) at a setting of 5 for 1 minute, rat plasma conjugase (500 ul) was added to the homogenate to deconjugate polyglutamyl folates. Rat plasma conjugase was obtained from Pel-Freeze (Rogers, AR) and dialyzed for 24 h through 12,000-13,000 mw tubing in cold 50 mM Hepes/50 mM Ches, pH 7.85 buffer containing 2% (w/v) sodium ascorbate, 10 mM 2-mercaptoethanol, and 2 teaspoons of acid washed charcoal. The conjugase was stored at -85 C in 1.5 mL polypropylene microcentrifuge tubes (Fisher Scientific, Pittsburgh, PA). Alphaamylase (20 mg/mL) (Sigma Chemical Co., St. Louis, MO, No. A-0273) was also added at this time to aid in the release of folate from the carbohydrate matrix. The samples incubated for 4 h at 37 C, after which protease (2 mg/mL) (Sigma

PAGE 76

Chemical Co., St. Louis, MO, No. #P-5147) was added to release any folate in the protein matrix of the food. This was followed by an additional 1 h incubation period at 37 C. The samples were next placed in a boiling water bath for 5 minutes to inactivate the enzymes and cooled on ice for 10 minutes. Samples were centrifuged at 15,000 rpm, 5 C, for 25 minutes using a Sorvall RC-5 Superspeed refrigerated centrifuge (DuPont Instruments, Newton, CT). The supernatant was filtered through Whatman #1 paper and kept on ice in a graduated cylinder. The pellet was resuspended in 5 mL Hepes-Ches buffer, homogenized and centrifuged again at 15,000 rpm, 4 C, for 25 minutes. The supernatant was then filtered through Whatman #1 paper and added to the previous filtrate. The graduated cylinders were filled to 30 mL with Hepes-Ches buffer and 1 mL portions were dispensed into five 1.5 mL vials for each menu. Nitrogen was bubbled through each vial for 10 seconds prior to storage at -20 C. Extractions were performed in duplicate for each menu for the three different times it was sampled. The folate extracted from the menus was quantitated in triplicate by microbiological assay using L. casei (Tamura 1990). A blank was utilized throughout the extraction procedure so that folate derived from the addition of the enzymes could be subtracted. Synthesis, Purification and Identification of Deuterated Folic Acid Deuterated folic acid was synthesized according to the method described by Gregory (1990). The first step involved bromination of folic acid with

PAGE 77

64 bromine at the 3' and 5' positions of the benzoic acid ring as described by Cosulich et al. (1951). Six grams of unlabeled folic acid (#F-7876) were dissolved in 75.6 mL of 6 N hydrochloric acid (HQ), diluted with an equal volume of distilled water and placed in an ice bath on a stirrer. A large excess of bromine vapor was used (126 mmol) to brominate the folic acid molecule at the specified positions via nitrogen gas. The reaction was conducted until all the bromine was consumed (approximately 20 minutes). Following dilution of dibromo folic acid with 75 mL distilled H2O, the sample was filtered through a sintered glass funnel under vacuum and refrigeration, washed several times and dried in a dessicator for several days. The second step involved synthesis of deuterated folic acid from dibromo folic acid. A 2.0 g portion of dry crude brominated folic acid was suspended in 60 mL deuterium oxide (D 2 0) and dissolved by the addition of 1.0 mL of 40% (w/w) sodium deuteroxide (NaOD). The use of D 2 0 and NaOD allowed for the removal of the exchangeable protons from the 3',5'-dibromofolate. After 5 minutes, 1.6 mL of 35% (w/w) deuterium chloride (DC1) was added to precipitate the 3'5'-dibromofolic acid. Following centrifugation for 15 minutes at 5,000g (4 C), the pellet was suspended in 30 mL D 2 0, dissolved with 0.5 mL NaOD (40%) and transferred to a hydrogenation bottle. Deuterium oxide and NaOD were added to achieve a 200 mL and 2.5 mL total of D 2 0 and NaOD, respectively. The catalyst, 0.25 g of 10% palladium (Pd) on carbon (Kodak; Rochester, NY), was then added and nitrogen gas was bubbled into the solution to exclude dissolved

PAGE 78

65 oxygen. Deuterium gas was utilized to replace the bromine at the 3' and 5' positions of the benzene ring with deuterium. This catalytic dehalogenation was performed at room temperature for approximately 1 h. At the end of the reaction period, the catalyst was removed by filtering the mixture through Whatman #1 paper under vacuum and washing several times. Crude [ 2 H 2 ]folic acid was recovered from the filtrate by acidification to pH 2 with HC1 followed by centrifugation. The crude [ 2 H 2 ]folic acid was dried under a dessicator for several days and weighed before starting the final step of purification. The weighed crude [ 2 H 2 ]folic acid (1.3 g) was then divided in half and ~ 636 mg was purified by ion-exchange chromatography on a column containing DEAE-Sephadex A-25 (Sigma Chemical Co., St Louis, MO). Diethylaminoethyl (DEAE) sephadex A-25 (75g) was hydrated and suspended in 0.025 M sodium phosphate buffer containing 0. 1 M NaCl, pH 7 (Buffer A) and poured into the column (50 cm length, 2.5 cm ID). Approximately 1 L of Buffer A ran through the column for equilibration. Half of the crude [ 2 H 2 ]folic acid sample (636 mg) was dissolved in 2-3 mL of 0.1 M NaOH and approximately 10 mL of Buffer A. The sample required sonication to ensure it was fully in solution. Litmus paper was used to check the pH which was subsequently adjusted to pH 8. The dissolved sample was applied onto the column and a fraction collector was assembled to collect 300 drops per tube or about 15 mL. Once the sample was loaded, a 2-L gradient containing Buffer A and 0.025 M sodium phosphate buffer with 0.6 M NaCl (Buffer B) was used to

PAGE 79

elute [ 2 H 2 ]folic acid. Fractions containing pure [ 2 H 2 ]folic acid were identified via HPLC using a 0.033 M phosphate buffer containing 15% acetonitrile (MeCN) as a mobile phase. The detector was set at 280 nm and a folic acid standard was injected first to determine the retention time of folic acid. The [ H 2 ]folic acid fractions eluting at the retention time of folic acid were pooled and precipitated by acidification with HC1. Following centrifugation, the purified [ 2 H 2 ]folic acid was desiccated over P 2 0 5 and weighed. The purity of the crude [ 2 H 2 ]folic acid (630 mg) was assessed spectrophotometrically and via HPLC and proton NMR. NMR analysis indicated complete labeling of the 3 ',5' positions of the benzoic acid ring as no signal was detected for 3 ',5 '-protons. Supplemental Folic Acid Commercially available folic acid (Sigma Chemical, St. Louis, MO) was used to prepare the folic acid supplements. Subjects received either 330 or 730 7 ug/d as synthetic folic acid. Deuterated [3',5'H2] folic acid was synthesized by the method of Gregory (1990) as previously described. In order to increase absorption of the synthetic folic acid, the day's requirement was divided into two portions so that half was consumed in the morning and the other half in the evening. Stock solutions were prepared containing the required amounts of both the deuterated and the unlabeled folic acid for the first 6 weeks of the study. At a later date, stock solutions containing only unlabeled folic acid were prepared for

PAGE 80

the second half of the study period. Food grade phosphoric acid and sodium phosphate (dibasic) were used to make the buffer in which [ H2]folic acid and unlabeled folic acid were dissolved. For the 450 jj.g level, 60 ug of deuterated folic acid (-15% of total folate) was required per day. Since the subjects were consuming supplemental folic acid twice per day, only 30 ug was required per supplement tube. The deuterated folic acid stock solution was prepared for the 450 ug/d groups by weighing out 45 mg deuterated folate, dissolving in 0. 1 N NaOH, and filling to 150 mL for a final concentration of 30 ug/100 uL. The exact concentration of all stock solutions was determined spectrophotometrically at 282 nm using a molar absorptivity coefficient at pH 7.0 of 27,600 L • (mol • cm) ^ (Blakley 1969). The volume needed per tube was adjusted according to the actual concentration. The subjects on the 450 ug/d level also received 270 ug/d of unlabeled folic acid or 135 jag twice/d. Folic acid (202.5 mg) was weighed and dissolved with 0. 1 N NaOH and filled to 150 mL with buffer for a final concentration of 135 ug/ 100 uL. For the 850 ug/d group, 120 jig of the deuterated folic acid (~ 15% of total folate) was required per day or 60 ug per tube. To make the deuterated stock solution for this level, 90 mg of deuterated folic acid was weighed out and dissolved with 0.1 N NaOH and filled to 150 mL with buffer for a final concentration of 60ug/100 uL. The 850 ug groups required 610 ug of unlabeled folic acid or 305 ug per tube. Folic acid (457.5 mg) was weighed out and dissolved with 0. 1 N NaOH and filled to 150 mL with buffer for

PAGE 81

68 a final concentration of 305 ug/100 uL. The stock solution for the second half of the study was prepared in a similar fashion except the folate contribution from the deuterated folic acid was replaced with unlabeled folic acid. All stock solutions were stored in a -70C freezer and the concentration was verified spectrophotometrically before making up a new batch of supplements to ensure that no degradation of folate had occurred during storage. For human consumption, stock solutions in appropriate volumes (-100 uL/tube) were dispensed into 50 mL conical tubes to which approximately 40 mL of apple juice was added. The tubes were then capped tightly, vortexed and stored at -20C until used (within 3 months). Supplemental folic acid content was confirmed periodically by HPLC and found to be fully stable. 13 Purification and Administration of [ C]pABG Labeled pABG was provided to a subsample of the subjects (n=4) in order to obtain information on the absorption and excretion of exogenous pABG which 13 is found in food. [ C]Folic acid, kindly provided by the lab of Dr. Gregory, was cleaved at the C9 and N10 bond via zinc dust suspended in water (0.75g/3mL H2O) and 5 N HC1. After removal of the zinc dust by centrifugation, the supernatant containing both the pteridine and pABG moieties was removed and adjusted to pH 7. At this point, the supernatant became cloudy due to formation of a precipitate and required a second centrifugation. The final supernatant containing both fragments of the molecule was then applied to a size exclusion

PAGE 82

gel (P2). The two fragments were separated on the basis of size upon the addition of water to the gel after the sample was loaded. Small 1-2 mL fractions were collected and analyzed by HPLC to identify the fractions containing the pABG fragment. A pABG standard (Sigma #A-0879) was used to identify the retention time of pABG and the appropriate fractions were pooled. The concentration of the pooled pABG was then determined by HPLC using a pABG standard curve. One hundred microgram portions of the pABG water mixture was dispensed into 50 mL conical tubes and filled to 40 mL with apple juice. Four subjects (one from each group) received the one time administration of 100 ug [ 13 C]pABG at the half way point of the study (beginning of week 7). This was followed by five consecutive 24 h urine collections. The enrichment of the urine with [ 13 C]labeled pABG will be determined by GCMS and evaluated in terms of bioavailability. Another aspect of this study involves the quantitation of pABG in the diet. This method has been worked out by the laboratory of Dr. Gregory and may be an ancillary project. Blood Collection and Processing Weekly fasting blood samples were drawn by a certified phlebotomist throughout the twelve wk study. Multiple blood collections (i.e. six in a 2 wk period) were obtained at the switch-over period starting at the beginning of the 7 wk. Blood processing occurred within one h of the blood draw. A total of 17 blood collections per subject were obtained during the 12 wk study period. The initial blood draw obtained on day one was representative of baseline

PAGE 83

70 concentrations of serum folate, red cell folate and plasma homocysteine, while the last blood draw at the end of wk 12 represented final concentrations. Precautions were taken to protect all blood samples from light during collection, processing and storage. As mentioned previously, upon completion of the twelve wk study, a subgroup of the pregnant (n=8) and nonpregnant women (n=7) returned once a month for phlebotomy during a three month follow-up period. In addition, maternal and cord blood and a blood chemistry profile (SMAC 25) were obtained at the time of delivery for a subgroup of the pregnant women (n=8). Serum was collected in 13 mL silica-coated tubes (Vacutainer Becton Dickinson, Rutherford, NJ) and held at room temperature for 30 to 60 minutes to allow time for clotting. The clotted portion of each sample was removed by centrifugation at 2,000 rpm for 30 minutes (International Equipment Company Model HN-S II Centrifuge, Needham Heights, MA). The serum was then dispensed in duplicate into 1 .5 mL vials containing approximately 1 .5 mg of ascorbate (to ensure folate stability during storage) and subsequently frozen at 20C until analysis of serum folate by microbiological assay. Plasma and whole blood were collected in 10 mL ethylenediamine tetraacetic acid (EDTA) tubes (Vacutainer Becton, Dickinson, Rutherford, NJ). For a small subset of subjects (n=4) an extra 13 mL EDTA tube (Vacutainer Becton, Dickinson, Rutherford, NJ) was used to collect plasma to be analyzed at a later date by GCMS to obtain information on folate kinetics.

PAGE 84

71 Plasma to be assayed for homocysteine was placed immediately on ice and centrifuged at 2,000 rpm (DEC Model HN-SII Centrifuge, Needham, MA) for 30 minutes in the cold room. After centrifugation, the plasma was pipetted in duplicate into 5 mL scintillation vials and placed in the freezer at -20 C for subsequent analysis by a coworker. Plasma to be assayed for deuterated folic acid was kept at room temperature and centrifuged at 2,000 rpm for 30 minutes (International Equipment Company Model HN-S II Centrifuge, Needham Heights, MA). The plasma was then transferred to a 10 mL conical tube. The plasma obtained after centrifuging the whole blood for the packed red blood cells was added to the appropriate tube to increase the total plasma obtained (about 8 mL). Whole blood was mixed at room temperature and 75 uL was dispensed in duplicate into 1.5 mL vials. A 1:20 dilution of whole blood was then performed with a 0. 1 % ascorbate solution made with ascorbate (Na salt) and milli-Q water (Millipore, Bedford, MA). The whole blood folate diluted with ascorbate and milli-Q water was incubated at room temperature for at least 30 minutes in order to allow for deconjugation of the polyglutamates to mono-, dior triglutamates. From the same sample, duplicate samples of whole blood were drawn into capillary tubes and centrifuged for 7 minutes at 2,000 rpm. Hematocrit levels were determined using a microhematocrit tube reading device. The packed red cells from the homocysteine tube and the blood remaining in the whole blood tube were combined in a 10 mL conical tube and centrifuged

PAGE 85

72 at 2,000 rpm for 20 minutes to obtain approximately 5 mL of packed red cells. An equal volume (1:1) of ascorbate dissolved in milli-Q water (0.2% solution) was added to the red cells following the removal of plasma. After mixing, the red cells were stored at -20 C until subsequent analysis of the deuterated folate by a coworker. Urine Collection and Processing Brown, opaque, 2.5 L bottles containing 3 g of ascorbic acid (Na salt) to protect the urinary folates from degradation were used to collect the 24 h urine samples. Urine was collected weekly and for five consecutive days during the switch-over period. A total of eighteen, 24 h urine collections were obtained per subject. The subjects were instructed to start the collection after the first void and to end the collection after their first void the next morning. The urine was kept cold at all times during the collection process and returned to the investigator after the 24 h period. The volume of each 24 h urine collection was recorded, and urine was transferred into 200 mL containers and 1 .5 mL vials and stored prior to analysis at -20 C. Analyses Performed by Microbiological Assay Serum Folate Total serum folate concentrations were assessed microbiologically using the test organism L. casei. A slightly modified version of the method described by Tamura (1990) utilizing the 96well microtiter plate and serial dilutions was used for analyzing the samples. The L. casei (microbe #9640) was

PAGE 86

73 grown in ATCC 7830 media (Difco Laboratories, Detroit, MI). Serum contains only the monoglutamate form (mainly 5-methyl-THF) to which the L. casei microbe responds. The samples were added to assay medium containing minimal folic acid and were subsequently inoculated with the microorganism. The samples incubated over night, and the growth response of the organism was estimated the next day by measuring the increase in turbidity of the medium which is proportional to the folate concentration in the serum or red blood cell hemolysate. Calibration curves for the assay were constructed using 10 ng/mL folic acid standard. Falcon (Becton Dickinson, Rutherford, NJ), 96-well, flat bottomed, sterile, low evaporation tissue culture plates with lids were used for this assay. For each assay, 20 u.1 of control, standard and samples were added in duplicate to selected wells of the plate followed by the addition of 130 ^1 buffer to these select wells. One hundred and fifty microliters of buffer was then dispensed into all wells. A total of five serial dilutions of the samples, standard and pooled control was made using an 8-channel pipette (Costar8-Pette #04670, Costar Corp., Cambridge, MA). All wells were inoculated with 150 ul of the medium containing the test organism (5 uL/mL media), covered and placed in the incubator at 37 C for 1 8 hours. After incubation, the growth of the microorganism was measured by reading the turbidity of each well at 650 nm using a microtiter plate reader (Molecular Devices UV Max, Menlo Park,CA) interfaced with a computer. Data

PAGE 87

74 were calculated using Soft Max data analysis software (Molecular Devices). The turbidity reading of each sample was translated by the computer into a folate concentration (ng/mL) based on the known amount of folate in the standard. A standard curve was generated from the standards using a four parameter equation. The linear portion of the standard curve was used to determine the folate concentration (ng/mL) of the samples. Procedures for preparing the medium, the reagents and the standard are detailed in the Appendix. Within assay and day-to-day variation was assessed by measuring previously determined folate concentrations of pooled plasma each time the assay was performed. The interand intracoefficient of variation for this assay were ~ 8% and 7%, respectively. Red cell folate Red cell folate concentrations were also analyzed microbiologically using the microorganism, L. casei. Prior to analysis, the pH of each sample was reduced to approximately 4.0 with 50 0.1 M HC1 and incubated for sixty minutes in order to allow the conjugase to cleave the polyglutamate form of the reduced folate to a mono-, dior triglutamate. Due to the addition of the acid, the dilution factor was adjusted from 20 to 21 The acidification and incubation steps were necessary at this point due to the use of sodium ascorbate instead of free ascorbic acid. All samples were analyzed in triplicate and the intraand intercoefficient of variation were less than 10%. Red blood cell folate concentrations were determined by multiplying the whole blood folate (ng/mL) obtained directly from the assay by the dilution factor (21)

PAGE 88

followed by subtraction of serum folate and hematocrit and division of this value by the hematocrit as illustrated in the following equation. [Whole blood folatel [serum folate x (1 hematocrit / 100)1 hematocrit 100 Urinary folate Concentrations of intact urinary folate were determined microbiologically using L. casei as the test organism (Tamura 1990). Samples were analyzed in triplicate and the intraand intercoefficient of variation were < 10%. Appropriate dilutions were performed on urine samples that were running higher than what could be determined using the standard curve (> 20 ng/mL) Urine and Serum Creatinine Determinations (Glomerular Filtration Rate) Urine and serum creatinine determinations were measured using a Sigma Diagnostics kit (Sigma Chemical Co., St. Louis, MO. No. 555-A). Three milliliter of water, 3 mL of creatinine standard and 3 mL of sample were added to cuvettes labeled as blank, standard and sample(s) (serum or urine), respectively. Urine samples were diluted 10:1 with Milli-Q water. Alkaline picrate solution was added to all cuvettes, mixed and allowed to stand at room temperature for 812 min. The absorbance of standard and sample were read against a blank at 500 nm. Acid Reagent (No. 555-2) (0. 1 mL) was added to all cuvettes, mixed and allowed to stand at room temperature. Following 5 minutes, the absorbance of standard and sample versus a blank were read. Concentrations of unknowns were

PAGE 89

76 calculated from the absorbance of a 3 mg/dL creatinine standard ranging in concentration from 0 to 10 mg/dL. Calculations: Creatinine (mg/dL) = 3 x [(Initial Final Absorption Test) / (Initial Final Absorption Std)] Creatinine Clearance (ml/min) = urine (mg/dL)*urine (mL/min) serum (mg/dL) HPLC Quantitation of Urinary Intact Folates, pABG and apABG and Preparation for GCMS Overview Weekly urine samples (30 200 mL) for each subject (n=312) were passed through affinity chromatography columns in duplicate containing immobilized folate binding protein. Urinary folates bind tightly to the column material at neutral pH. The non-folate sample effluent from the affinity columns was collected and frozen at -20 C for subsequent pABG and apABG analysis. Following the removal of contaminants with a high salt neutralizer solution, folate was eluted in 5 mL 0. 1 N HC1 and collected separately. Of the 5 mL folate effluent, 1 mL was saved for quantitation of urinary 5-methyl-THF and folic acid at weeks 0, 4, 8 and 12. The remaining 4 mL was prepared for GCMS analysis in order to determine isotopic enrichment of urine with deuterium. Urinary pABG and apABG were measured in duplicate on a biweekly basis in pregnant women (weeks 0, 2, 4, 6, 8, 10 and 12) and on a monthly basis in nonpregnant controls (weeks 0, 4, 8 and 12) by the method of McPartlin et al. (1992) with minor modifications. The eluant containing both pABG and apABG

PAGE 90

(from above) was loaded onto ion exchange chromatography columns to separate pABG and apABG. Tritiated pABG and [ 3 H]apABG served as internal standards and were added to urine samples prior to application on ion exchange chromatography columns. After separation, the catabolites remained isolated throughout the remainder of the analysis. Acetamidobenzoylglutamate was deacetylated to form pABG and reapplied to the ion exchange columns. The pABG fractions were diazotized and loaded onto sep-pak cartridges for further purification and concentration. Azo-pABG was eluted in methanol and pABG was regenerated. High performance liquid chromatography was used to quantitate both endogenous pABG and pABG derived from apABG. High performance liquid chromatography was also used to isolate and collect the pABG fractions which were then prepared for GCMS so information can be obtained on folate kinetics. See Figure 5 for overview of catabolite purification and quantitation process. Procedures A. Purification of the Folate Binding Protein from Dried Whey An affinity chromatography sorbent coupled to purified bovine milk folate-binding protein (FBP) was prepared by a minor modification of the method of Selhub et al. (1980) as described by Gregory and Toth (1988). Folate-sepharose column material (Sigma Chemical Co., St. Louis, MO, No. F-7889) was gently poured into a glass column for the purpose of binding the FBP in dried whey (Meloskim, Kerry Ingredients). The column was washed with 500 mL 0.2 M acetic acid and 1 L of

PAGE 91

water followed by equilibration with 1 L of 1 .0 M KH2PO4. Dried whey (825g) was suspended in 6 L of water (best to do in batches). The pH of the mixture was adjusted to 7.0 with 5 M NaOH and subsequently centrifuged for 10 minutes at 8,000 rpm. The supernatant was filtered through glass wool to remove excess lipid and then loaded onto the folate sepharose column at room temperature. After loading, the column was washed with 1 M NaCl (1 L) containing 0.05 M Tris HC1, pH 7.4 for the purpose of removing compounds other than the FBP. Following a wash with H2O (1L), adsorbed FBP was eluted in 0.2 M acetic acid (approximately 500 mL) and collected in 5 to 7 mL portions by a fraction collector. A spectrophotometer (X = 280 nm) was used to determine which fractions contained the protein. The protein containing fractions were pooled for a total of approximately 300 mL. Dialysis tubing was used to remove the acetic acid. Following placement of the sample in the dialysis tubing, it was placed in 1 1 L of refrigerated milli Q-water and allowed to sit on a stirrer overnight. The next day, the 1 1 L refrigerated water was exchanged for fresh refrigerated Milli Q-water and allowed to sit overnight on a stirrer for the second time. The dialyzed FBP solution was freeze dried and stored at -20 C until coupled to Affigel 10. B. Preparation of Folate Binding Protein Affinity Columns A 25 mL bottle of Affi-gel 10 Gel (Bio-Rad Laboratories, Richmond, CA; No. #153-6046) was gently shaken and transferred to a glass fritted funnel in order to drain the excess liquid without letting the bed go dry. The bed was washed with 75-100 mL ice

PAGE 92

cold distilled water and the moist gel transferred to a small (125 mL) erlenmeyer flask. Purified folate binding protein (100 mg) from step A was suspended in 20 mL of 0. 1 M KH2PO4, pH 7.0 at 4 C and mixed with gel in erlenmeyer flask. The mixture was agitated gently at 4 C for 4 hours in a shaking ice water bath. After agitation, the gel was filtered on a fritted glass filter and the effluent from each wash was collected separately. The gel was washed with 3 x 20 mL cold 0.05 potassium acetate, pH 6.6 buffer and 2 x 20 mL cold 0.05 potassium acetate containing 1.0 M NaCl, pH 6.6 buffer. The Bio-Rad method was used to analyze the collected fractions for protein that did not couple to the affinity columns. Bovine serum albumin (BSA) was utilized to make a five point standard curve ranging from 0 to 25 ug/mL. The protein concentration of the fractions was determined spectrophotometrically at 595 nm. C. Binding Capacity Determination of Affigel-10 FBP Columns Affigel-10 FBP material containing immobilized FBP (section B) was loaded into 12 glass columns (7 mm i.d. x 20 cm) with a long pasteur pipette. The columns were filled to a height of 6 cm which was approximately 2.5 mL of material and equilibrated with 10-20 mL 0. 1 M KH2PO4, pH 7.0. Affinity chromatography columns containing immobilized FBP were previously found to retain folic acid, 10formyl-folic acid, THF, 5-methyl-THF, 10-formyl-THF, and 5,10,methenyl-THF by Gregory and Toth (1988a). The amount of folate required to saturate the affinity columns was determined by applying ten 1 mL fractions of 10 ug folic acid /mL 0.1 M potassium phosphate buffer, pH 7.0 (determined

PAGE 93

80 spectrophotometrically). After discarding the first 2 mL, the 10 mL eluates were collected separately and the concentrations determined spectrophotometrically at pH 7.0 using 0. 1 M potassium phosphate buffer, pH 7.0 as a blank. The columns were subsequently washed with 5.0 mL of 0.025 M KH2PO4 containing 1.0 M NaCl (pH 7.0), followed by 5.0 mL of 0.025 M KH2PO4 (pH 7.0). Both effluents were discarded. The binding capacity of the columns was determined by elution of the bound folates with 0. 1 N HC1. The first 2 mL effluent was discarded and the following 5 mL of 0.1 N HC1 collected in a 10 mL volumetric flask which was filled to 10 mL with 0.1 N HC1. The concentration of the eluted folate was measured spectrophotometrically at 296 nm using a molar absorptivity coefficient at an acidic pH of 20,000 L • (mol • cm)" 1 (Blakley 1969). Hydrochloric acid (0. 1 N) was used as the blank. The sum of the folate found in the load and the folate found in the acidic effluent should equal the amount of folate initially applied to the columns. The columns were then washed with 5 mL 0. 1 N HC1 and re-equilibrated with approximately 15 mL of 0.1 M KH2PO4, pH 7.0. The total binding capacity of the columns was approximately 25 ng (56 nmol) of folate. D. Purification of Urine Samples using Affinity Chromatography The volumes of urine that were loaded on the affinity columns were calculated so that one third of the total binding capacity (C) was not exceeded. This determination was based on the urinary folate concentrations predetermined microbiologically. Following adjustment of urine to pH 7.0, up to 200 mL of urine (volume varied depending on urinary folate concentration) was filtered through Whatman #1 filter paper.

PAGE 94

81 The filtered urine samples were loaded onto affinity columns pre-equilibrated with 0.1 M KH2PO4, pH 7.0. A peristalic pump was used to maintain a flow rate of 0.3 mL/min. The non-folate effluent from the load was saved for analysis of pABG and apABG, as described later. The columns were washed with 5 mL of 0.025 M KH2PO4 containing 1 .0 M NaCL, pH 7.0 and 5 mL of 0.025 M KH2PO4, pH 7.0. The folate was eluted with 0.1 N HC1. The first 2 mL contained no folate and was discarded. The following 5 mL were collected in a volumetric flask. Four milliliter were saved for cleavage reactions and subsequent GCMS analysis. No ascorbate was added to these samples. Ascorbate (50 ul of a 10% mixture) was added to the remaining 1 mL solution and stored at -20 C for subsequent quantitation of urinary 5-methyl-THF and folic acid by HPLC (E). The columns were washed with an additional 5 mL of 0. 1 N HC1 to ensure all compounds had eluted followed by re-equilibration with 15-20 mL of 0.1 M KH2PO4, pH 7.0 buffer. E. HPLC Quantitation of Urinary 5-Methyl-THF and Folic Acid The concentrations of 5-methyl-THF and folic acid in the affinity column eluate were determined in duplicate for wk 0, 4, 8 and 12 by a reverse phase HPLC procedure described by Gregory et al. (1984) with minor modifications. A fluorescence detector (Spectrovision FD-300 Dual-Monochromator) connected to a U 120 Universal Interface was used to monitor the native fluorescence of 5-methyl-THF. The detector was set at 295 nm excitation and 356 nm emission wavelengths. An ultraviolet (UV) absorption detector monitoring at 280 nm was used for detection

PAGE 95

82 of folic acid. Both metabolites were quantitated under the following conditions: 50-ul injection volume, octadecylsilica column (Ultramex Ci8, 5 |j.m particle size, 4.6 mm i.d. x 250 mm; Phenomenex, Torrance, CA USA), flow rate of 1.0 mL/min, isocratic mobile phase composed of 12% (v/v) acetonitrile and 0.033 M phosphoric acid, pH 2.3. Quantitation of folic acid and 5-methyl-THF was accomplished relative to standards ranging in concentration from 30 -1000 ng/mL. Standards were prepared from commercially available 5-methyl-THF and folic acid (Sigma Chemical, St. Louis, MO) using published molar absorptivities (Blakley 1969) to determine concentrations of stock solutions. F. Chemical Cleavage of Intact Folates to pABG The 4 mL affinity column eluate (D) containing 5-methyl-THF, folic acid and possibly THF was subjected to chemical cleavage of the C9-N10 bond forming pteridines and pABG (Toth and Gregory 1988). Para-aminobenzoylglutamate, containing the deuterium, was then isolated by HPLC and subsequently analyzed by GCMS. Cleavage of 5methyl-THF was accomplished by adding 2 M potassium acetate (150 ul) and IN NaOH ( 1 50 ul) to the 4 mL eluate yielding a pH of 5.76.0. Five percent (v/v) hydrogen peroxide (220 (al) was added to oxidize 5-methyl-THF to 5-methylDHF. After exactly 30 seconds, 450 (il of 0. 1 % bovine hepatic catalase (Sigma Chemical Co., No. C-40) was added to decompose the remaining hydrogen peroxide and stop the reaction. To cleave 5-methyl-dihydrofolate, samples were acidified by the addition of 500 (al of 5 N HC1 and incubated at room temperature for 60 minutes.

PAGE 96

83 Folic acid was reductively cleaved in the acid mixture by the addition of 270 ul of a suspension of zinc dust (1 g zinc in 3 mL milli-Q water) with periodic agitation for 15 minutes. The zinc was subsequently removed by centrifugation in a clinical centrifuge at 2,000 rpm for 5 minutes. Supernatant was removed with a Pasteur pipette and stored at -20 C until HPLC isolation of the pABG. These procedures yielded complete conversion of intact folates to pABG as monitored by HPLC. G. HPLC Isolation of pABG and derivatization HPLC isolation and derivatization of pABG was performed by a co-worker. The pABG fragment from step F was isolated and collected by HPLC using a Prodigy preparative Ci8 column (10 um particle size; 250 x 10 mm; Phenomenex, Torrance, CA USA). Formic acid (0. 1 M) containing 6% (v/v) acetonitrile was used as the mobile phase at a flow rate of 4.0 mL/min. When collecting peaks for GCMS analyses, the pH of the mobile phase should not be adjusted. The retention time of pABG was predetermined with a pABG standard (Sigma #A-0879) dissolved in 0. 1 N NaOH. Detection was achieved with an absorbance detector monitoring at 280 nm. The pABG peak (~ 4 mL) was manually collected and evaporated to dryness at 90 C under a constant flow of nitrogen gas, capped tightly and kept refrigerated overnight. The dried pABG fractions were brought to room temperature, flushed with nitrogen gas for 30 seconds and derivatized by addition of 200 ul trifluoroacetic anhydride and 100 ul trifluoroethanol. Samples were then capped and placed in a heating block at 90 C for one h. After cooling to

PAGE 97

84 room temperature, excess reagent was evaporated under nitrogen gas at room temperature and the residue dissolved in 50 ul dry ethyl acetate. Samples were transferred to autosampler vials with polyspring inserts and stored at -20 C before GCMS analysis. The end product of the above reaction was a lactam of Nl-trifluoroacetyl-p-aminobenzoyl-glutamate-a-trifluoroethyl ester that retains all isotopic labeling on the pABG component of the folates (Gregory and Toth 1988a). This lactam is well suited for GCMS analysis for determination of isotopic ratios. 3 H. Synthesis of [ HIpABG Tritiated pABG was prepared from 250 microcuries 3 [ H]folic acid (Amersham International, No. TRK 212) which is labeled at positions CI, C9 and 3' and 5' of the pABG moiety. To 200 ul of the [ 3 H]folic acid, 100 ul 5 M HC1 and 100 ul of Zn dust suspended in water (1.0 g per 3 mL milli-Q water) was added in order to cleave the intact molecule. The mixture was shaken intermittently for 15 minutes at room temperature followed by centrifugation for 10 minutes at 2,000 rpm to pellet the zinc. The supernatant was removed and stored in the refrigerator. Unlabeled folic acid (100 ug/mL) was diluted with a 1 % sodium ascorbate solution to a final concentration of 4 ug/mL in order to resemble the conditions of the commercially available [ 3 H]folic acid and cleaved as described above. Unlabeled pABG standards were prepared from commercially available pABG (Sigma Chemical Co., St. Louis, MO No. A-0879). Following adjustment of 100 ul pABG stock solution (100 ug/mL) to pH 13 by

PAGE 98

diluting 1:10 with 0.1N NaOH (lOug/mL), unlabeled pABG concentration was verified spectrophotometrically at 272 nm using 15,500 as the extinction coefficient (Blakley 1969). I. Purification of T 3 HlpABG by HPLC Tritiated pABG was purified under the following conditions: 100 ul injection volume; octadecylsilica column (Microsorb Ci8, 3 um particle size, 4.6 x 100 mm; Rainin; Woburn, MA), mobile phase consisting of 0. 1 M formic acid, pH 4.0, 6 % acetonitrile. Before running the radioactive pABG-pteridine mixture, unlabeled pABG was injected to determine the retention time of pABG. Cleaved, unlabeled folate containing both the pteridine moiety and pABG was also run to assure the pteridine and pABG moieties were eluting separately. Once the retention time of 3.55 minutes for pABG was determined and lack of interfering compounds assured, the tritiated sample was injected four separate times and pABG was collected and counted via a scintillation counter. In order to assure the isolation and collection of [ H]pABG, unlabeled pABG was spiked with 20 ul of [ H]pABG, injected into the HPLC, collected in one milliliter fractions for 10 minutes, and counted. The 3 isolation of [ HjpABG was verified since the majority of counts matched the 3 retention time of unlabeled pABG. To further confirm [ H]pABG had been isolated, a small portion (20 ul) was applied to equilibrated ion exchange columns and after washing with 0. 1 N HC1 was eluted with 0.6 N HC1. Both effluents were collected and counted. Recovery of 95-100% of [ 3 H]pABG in the 0.6 N

PAGE 99

86 3 HC1 eluate provided additional evidence that [ H]pABG had been isolated. One3 half of the tritiated pABG was used for the synthesis of [ HJapABG and the other half (~3 mL) was diluted to 500 mL with 0. 1 % sodium ascorbate solution and 3 mixed well. Five hundred microliters of the diluted [ H]pABG were transferred to 1.5 mL polypropylene microcentrifuge tubes (Fisher Scientific, Pittsburgh, PA) and stored at -20 C. Each tube (500 ul) contained ~ 58,000 dpm. J. Synthesis of r 3 HlapABG The [ 3 H]pABG starting material (2.5 mL) was 3 concentrated by solid phase adsorption (C 18 Sep Pak, Waters). The [ H]pABG fraction was converted to its diazotized derivative by the method of Bratton and Marshall (1939). Two hundred microliters of 5N HC1 and 200 ul of 1% sodium nitrite (lg/100 mL) were added to 2.5 mL of the isolated tritiated pABG (J) and allowed to sit five minutes. Para-aminobenzoylglutamate, a primary amine, reacts in acid conditions with the nitrite to form a diazonium salt. Next, 200 ul of ammonium sulfamate (5%) was added to destroy excess HN02After five minutes, 200 ul napthylethylene diamine (1%) was added to couple with pABG. A minimum incubation period of 30 minutes was allowed for complete formation of the purple colored azo compound. The diazotized radiolabeled derivative was applied to and retained on a preactivated (3 mL methanol followed by 5 mL water) Ci8 Sep Pak cartridge (Waters). The cartridge was washed with 10 mL HC1 (0.05 M) and azo-pABG eluted with 3 mL methanol. Tritiated paraaminobenzoylglutamate was regenerated by the addition of 50 ul 5M HC1 and 50

PAGE 100

87 ul zinc dust suspended in water. The mixture was vortexed intermittently for 15 minutes followed by centrifugation at 2,000 rpm for 10 minutes. The acetamido derivative was prepared by transferring the supernatant to a conical glass tube and evaporating it to dryness under a stream of nitrogen gas in a reactitherm set at 40 C. Thirty-four microliters of 50% (v/v) acetic acid and 5 ul of acetic anhydride was added to the dried extract. The tube was plugged with cotton, vortexed gently and incubated at room temperature for 24 h. A nonradioactive apABG standard was prepared by the method of Baker et al. (1964). One gram of pABG was dissolved in 10 mL of 50% acetic acid and following addition of 1.5 mL acetic anhydride, the mixture sat overnight at ambient temperature. Whatman paper (#1) was used to filter the mixture followed by washing with milli-Q water. The residue was dessicated to remove water and stored at -20 C. 3 K. Purification of [ HlapABG by HPLC The same conditions as described in (I) were used to isolate the radioactive apABG. Nonradioactive apABG standard synthesized previously (J) was dissolved in water and injected to determine the retention time of this fragment (~ 8 minutes). The tritiated sample was injected and the peak corresponding to the retention time of unlabeled apABG manually 3 collected. To verify isolation of [ H]apABG, unlabeled apABG was spiked with 3 10 ul [ H]apABG, injected into the HPLC, manually collected in 1 mL fractions for the following 30 minutes and then counted. Since the majority of counts matched the retention time of unlabeled apABG, it was assumed that [ H]apABG

PAGE 101

88 had been isolated. The pooled tritiated apABG was diluted with 0. 1 % sodium ascorbate solution up to a final volume of 160 mL and 1 mL portions containing 23,000 dpm were transferred into 1.5 mL polypropylene microcentrifuge tubes (Fisher Scientific, Pittsburgh, PA) and stored at -20 C. L. Preparation of Ion-Exchange Columns and Fractionation of Urine One column per 20 mL urine sample was prepared by filling a glass column (15 mm i.d. x 200 mm) with a slurry of Dowex 50W (Sigma Chemical Co. St. Louis, MO) cation exchange resin (7 g/ 10 mL H 2 0) to a height of approximately 5.0 cm. The column was activated with 50 mL HC1 (6 M), washed with 50 mL H2O and equilibrated with 10 mL HC1 (0. 1M). The sample eluate from the affinity columns (step D) was divided into two 20 mL portions and acidified to a final concentration of 0.1 M with 5 M HC1. At this point, tritiated internal standards were added to samples analyzed over the first 3 wk (-110 samples). Following addition of tritiated internal standards, samples were applied to the ion exchange chromatography columns. Acetamidobenzoylglutamate eluted in the sample volume and the following 50 mL 0. 1 mol/L HC1 effluent, which were pooled and retained. A second 0. 1 mol/L HC1 wash was applied (higher concentrations resulted in pABG elution) and discarded. Para-aminobenzolyglutamate was eluted and collected in 100 mL of 0.6 mol/L HC1. The columns were reactivated, washed and re-equilibrated as described for reuse in step N. M. Deacetylation of Urinary apABG and Rechromatography The fraction containing apABG (M) was acidified to 0.2 mol/L HC1 and heated at 100 C for

PAGE 102

89 60 minutes to deacetylate apABG and then cooled at ambient temperature. Following adjustment to 0. 1 mol/L HC1 by a 1 : 1 dilution with H2O, the samples were reapplied to equilibrated ion exchange columns and washed with 100 mL 0.1 M HC1. Para-aminobenzoylglutamate representing apABG was eluted and collected in 100 mL 0.6 mol/L HC1. The two pABG fractions were kept separate throughout the remainder of the purification and quantification process. Urine I apABG acidify boil t pABG Affinity Columns Ion Exchange Columns Diazotization Solid Phase Extraction Evaporation & Regeneration HPLC Quantitation [ 3 H]stds pABG Figure 5. Overview of catabolite isolation and purification process. N. Diazotization and Solid-Phase Extraction The 100 mL fractions containing the native pABG (step L) and the deacetylated pABG (step M) were derivatized by the method of Bratton and Marshall (1939). One milliliter of 5.0 mol/L HC1 and 1 mL sodium nitrite solution (1%) were added to each sample and allowed to react for five minutes at room temperature. One milliliter ammonium sulfamate

PAGE 103

solution (5%) was then added to destroy excess HN02After five minutes, 1 mL N-(l-napthyl)ethylenediamine solution was added to couple with pABG. The samples sat overnight at room temperature to allow complete formation of the purple-colored azo-N-(l-naphthyl)-ethylenediamine derivative of pABG (azopABG). Each sample (-104 mL) was applied to a preactivated Cj8 Sep Pak cartridge (Waters) under negative pressure for further purification and concentration. Occasionally, samples turned a dark purple color and the volume loaded was reduced to prevent overloading the Ci8 Sep Pak cartridge. Following a 10 mL 0.05 mol/L HC1 wash, azo-pABG was eluted in 4 mL 100% methanol. The solution was evaporated to dryness at 40 C under a stream of nitrogen. O. Regeneration of pABG The residue (N) was reconstituted in 250 ul H2O and pABG was regenerated upon the addition of 25 ul HC1 (5M) and 25 ul Zn dust suspension (1 g /3 mL milli-Q water). After 15 minutes of intermittent mixing, the reaction mixture was transferred to a 1 .5 mL microcentrifuge tube, and centrifuged at 2,000 rpm for ten minutes. The supernatant was removed, filtered to remove zinc particles, and stored at -20 C prior to HPLC analysis. Recovery 3 3 of [ H]pABG and [ H]apABG was ~ 80 and 60%, respectively, and were the values used to correct final estimates of catabolites. P. HPLC Quantitation of Urinary pABG The amount of pABG (representing both the endogenous pABG and apABG) in a sample was determined by reversephase HPLC under the following conditions: 100-uL injection volume, octadecylsilica column (Ultramex Cj8, 5 um particle size; 3.5 mm i.d. x 250 mm;

PAGE 104

Phenomenex, Torrance, CA USA), flow rate 1 .0 mL/min, isocratic mobile phase composed of 0. 1 M formic acid, pH 4.0 and acetonitrile (2%). An ultraviolet (UV) absorption detector (Dionex AD20) monitoring at 280 nm was used for detection of pABG. Quantification of pABG was accomplished relative to standards ranging in concentration from 100 1500 ng/mL prepared from commercial pABG (Sigma Chemical Co., St. Louis, MO) using published molar absorptivities (Blakley 1969) to determine the concentration of the pABG in stock solution. The computed value for each urine extract was adjusted by the final percentage recovery of the radioactive standards, either tritiated pABG or tritiated apABG that were added to the initial aliquot of urine. This value was then adjusted for total urinary volume to give a daily output value of catabolite. 3 3 Q. Recovery of f HIdABG and r HlapABG added to Urine Samples Prior to catabolite quantitation, the method of McPartlin et al. (1992) was optimized through utilization of tritiated standards. Although McPartlin and associates (1992) also utilized tritiated internal standards, the author of this dissertation was able to identify steps in which significant amounts of either [ 3 H]pABG or [ H]apABG were lost. Identifying these losses was accomplished by removing and counting 500 uL sample after each step. Appropriate adjustments were made such as substituting a 0.05 M HC1 wash for the 25% methanol wash during the solid phase extraction enabling greater recovery of the tritiated standards at the end of the procedure. Following optimization of the method, recovery experiments were conducted on ten samples on ten different days. After

PAGE 105

establishing that inter and intra coefficients of variation were < 10%, thereby verifying the reproducibility and accuracy of the method, tritiated standards were only added to the first 1 10 urine samples which were analyzed within the first four weeks of analysis. It was noted that after 4 wk, the standards started decomposing yielding lower recovery percentages due to both breakdown of pABG and loss of tritium and therefore were not used after this 4 wk period. No attempts were made to repurify the standards. Recovery of radiolabeled material prior to HPLC injection was expected to vary between 30 to 40% based on the 3 findings of McPartlin et al. (1992). In the current study, recovery of [ H]pABG 3 and [ H]apABG were ~ 807c and 60%, respectively, and were the values used to adjust our final catabolite concentrations. R. Preparation of Catabolite Samples for GCMS Analysis The catabolites were isolated and manually collected from the HPLC and prepared for GCMS as described in sections F and G. However, the pH of the mobile phase add been adjusted to 4.0 and these collections may need to be repeated for GCMS analyses. Statistical Analyses Our study design was a two-by-two factorial with repeated measures. Two groups of women (pregnant and nonpregnant) were assigned to consume one of two supplement levels (450 or 850 ug/d) over a 12 wk period. One objective of this study was to determine if differences existed in folate status and catabolite excretion between pregnant women assigned to consume folate intakes of 450

PAGE 106

93 ug/d compared to 850 \xg/d. A second objective was to determine if differences existed between nonpregnant controls and pregnant subjects within the same supplementation group. Data were obtained at baseline (i.e. immediately prior to consuming defined diet) and then on a weekly basis. Prior to starting the study, the majority of pregnant women were consuming prenatal vitamins containing ~ 1 mg folic acid. Most folate status indices were affected by this prior folate supplementation and differences (P < 0.05) were detected at baseline between pregnant and nonpregnant women. Thus, acclimatization to assigned folate intake was necessary to compare folate status indices between groups. After trying a variety of monotone models, the data were modeled using a negative exponential model: |a + p 0 (0 exp (-Pi x week)) as it was the best fit for the data and allowed for estimation of steady state. Two key assumptions underlying this model included the following: (1) steady state was achievable within 12 wk; and 2) the data were approximately monotone for each group (always increasing, decreasing or maintaining). Steady state means for each group were assessed essentially by comparing the area under the curve at one time interval (i.e., week 0 to week 1) to the mean area under the curve of the following time intervals. When the ratio of these areas was > 0.95 or < 1 .05 (the slope approximated zero), steady state was presumed (Gibaldi and Perrier 1982). Comparisons were made for differences in initial, final and steady state group means. All tests of initial, final and steady state means were performed at alpha = 0.05 using standard ANOVA techniques with appropriate contrasts (SAS/STAT Version 6, SAS

PAGE 107

Institute, Cary, NC). A standard repeated measures analysis (MANOVA) was used to compare overall urinary apABG excretion among the groups under the assumption that prior folate supplementation did not affect urinary apABG excretion over time (i.e., the trend over time remained fairly constant for each group). Initial and final group means were compared using ANOVA with appropriate contrasts. Comparisons were made at alpha = 0.05 between groups for differences in initial, overall and final group means. Pearson correlations were used to describe associations between folate status indices, catabolite excretion and other pertinent parameters. Interpretation of the Data A similar approach to that taken in the controlled metabolic studies conducted by O'Keefe et al. (1995) and Jacob et al. (1994) was used to interpretate the folate status data. Interpretation of urinary folate catabolite excretion was based on the study conducted by McPartlin et al. (1993). The treatment response variables evaluated were as follows: (a) serum folate; (b) red cell folate; (c) urinary 5-methyl-THF; (d) urinary folic acid; and (e) urinary catabolites (apABG and pABG). Kinetic data and plasma homocysteine will be addressed in ancillary projects. The significance of changes in each of these response variables was determined statistically as described above. Each response variable was evaluated

PAGE 108

95 independently and in conjunction with other variables. The data interpretation, as discussed below, was tailored to address the following hypotheses: Hypothesis 1 The response of pregnant women and nonpregnant controls to two defined levels of folate intake (450 or 850 ug/d) will indicate a greater folate requirement for pregnant women as determined by changes in folate status indices and catabolite excretion. Hypothesis 2 Folate response indices will reflect a decline in folate nutriture approaching negative folate balance among pregnant women consuming 450 ug/d of folate and maintenance of folate status in pregnant women consuming 850 ug/d. Hypothesis 3 Folate response indices will reflect maintenance of folate status in nonpregnant women consuming 450 ug/d of folate and an excess exceeding the body's ability to metabolize or store folate in nonpregnant women consuming 850 ug/d of folate. Serum Folate In a controlled metabolic study, serum folate concentration should provide a sensitive measure of the overall folate status of the individual. As the requirement for folate increases, serum folate was expected to decrease under conditions of constant folate intake. Rapidly declining serum folate concentrations may be suggestive of inadequate folate consumption, whereas maintenance of serum folate concentrations in the acceptable range may be reflective of sufficient folate consumption. In contrast, serum folate concentrations that have increased significantly and subsequently leveled off may

PAGE 109

96 be reflective of folate consumption beyond the need for metabolism and storage. The results of the other variables aided in interpretation of the above mentioned scenarios. The effect of hemodilution on serum folate concentrations in pregnant women as a result of an increase in plasma volume was considered. Similar declines were expected in pregnant women consuming either 450 or 850 ug/d. Initial concentrations of serum folate were also considered, and comparisons were made between groups after acclimatization to the prescribed folate intake. Red Cell Folate Red cell folate concentrations reflect liver stores and are considered to be a measure of long-term status (Herbert 1990). Folate is taken up and retained only by the reticulocyte and not by the mature cell as the latter possesses negligible folylpolyglutamate synthetase activity (Shane 1995). Since the life span of the red cell is 120 days, red cell folate concentrations are reflective of folate status 2 to 3 months prior to analysis and significant changes in red cell folate concentrations may not occur within the time constraints of this study. However, changes in red cell concentrations (perhaps not significant) are expected to be observed if folate status changes and should parallel those observed in serum folate concentrations. Red cell folate concentrations that gradually decline over time may be suggestive of inadequate folate intake and declining folate stores. Maintenance of red cell folate concentrations in the normal range would be indicative of adequate folate consumption whereas an increase in red cell folate concentrations followed by a leveling off may reflect excess folate consumption. Red cell folate concentrations obtained during the

PAGE 110

follow-up period were very informative as these values were reflective of folate status during the study period when subjects were consuming defined folate intakes. Other indices were used in conjunction with red cell folate concentrations to most appropriately interpretate the response. Urinary Folate Urinary folate is predominately comprised of both unmetabolized folic acid and 5-methyl THF. Dramatic increases over time in the urinary excretion of unmetabolized folic acid may be reflective of folate intake above that necessary to maintain metabolism, whereas significant decreases over time may be indicative of inadequate folate consumption. Pregnant women were expected to excrete more intact folate during pregnancy secondary to the increase in glomerular filtration rate. Therefore, creatinine clearance was monitored throughout the study period in both the pregnant subjects and their nonpregnant controls so that urinary excretion of unmetabolized intact folate could be evaluated independent of the increase in glomerular filtration rate. Urinary Catabolites (apABG and pABG) Excretion of the catabolites was expected to remain constant throughout the study period in nonpregnant women, regardless of folate intake. In the pregnant women, catabolite excretion was expected to vary secondary to changes in utilization rates during the second trimester also regardless of the level of folate intake. We expected to see similar variations in all pregnant women since they were gestationally, the same age. It was hypothesized that the pregnant women on either supplementation level would be excreting greater quantities of the breakdown products throughout the entire

PAGE 111

study period (12 weeks) secondary to their greater requirement for folate compared to the nonpregnant controls. Importantly, data on glomerular filtration rate (serum creatinine and urinary creatinine) were utilized to determine if the potential increase was independent of glomerular filtration rate. The combined pABG and apABG fractions were used as representation of folate utilization and as a possible guideline for requirements when assessed with the other parameters discussed above.

PAGE 112

CHAPTER 4 RESULTS Folate Content of Menus The folate content of each menu is shown in Table 3. All menus were extracted in duplicate and measured in triplicate at three separate time points Table 3. Folate content of menus. Menu 1 2 3 MEAN (SD) INTRA/INTER (m) (MS) (ml CV% Bl 12.3 13.2 11.0 11.8 (0.8) 5.0/9.1 1 1 8 1 l.o 1 1 A 1 1 i j i.j B2 31.8 22.7 34.5 30.6 (4) 7.0/13.7 30.8 29.5 34.3 B3 12.7 16.6 15.8 14.6 (2.2) 8.5 / 14.4 13.5 17.2 12.0 LI 18.2 19.6 18.5 18.8(1.3) 7.5/3.8 17.8 17.1 20.1 L2 19.2 13.8 18.6 17.1 (2.0) 6.0/13.6 19.1 15.4 16.3 L3 12.2 16.6 15.8 14.7 (2.4) 13.7/13.2 15.0 17.2 11.2 SNACK (AM+AFT) 41.3 33.7 34.9 37.8 (4.0) 6.9/8.5 39.0 34.5 43.2 Dl + PM SNACK 46.4 40.3 45.8 45.6 (3.1) 6.5 / 2.0 44.4 49.6 47.3 D2 + PM SNACK 45.8 48.9 53.2 47.5 (6.1) 7.7/12.4 40.7 41.2 55.4 D3 + PM SNACK 29.6 33.0 44.0 36.6 (5.8) 5.3/16.1 34.6 35.0 43.4 D4 + PM SNACK 52.1 52.3 42.9 48.7 (4.1) 6.5 / 6.7 51.8 44.8 48.0 D5 + PM SNACK 53.8 56.7 53.7 53.1 (3.3) 6.0/4.4 49.0 53.8 47.7 99

PAGE 113

100 throughout the study. The diet provided a mean folate intake of 120 15 pg/d. The intraassay and interassay coefficient of variation (CV) were 7.0 % and 12 %, respectively. Food Record Analyses During the 3 month follow-up study, the mean dietary folate intake for pregnant and nonpregnant subjects was 293 39 and 379 78 pg/d, respectively in the 450 pg/d group and 278 18 and 197 75 ug/d, respectively in the 850 pg/d group. In addition to dietary folate, subjects previously consuming 450 or 850 pg/d during the study consumed either 200 or 600 pg/d supplemental folic acid, respectively. Total folate intake (dietary folate + supplemental folic acid) during the follow-up period was estimated to be 493 39 and 579 78 pg/d for pregnant and nonpregnant women in the 450 pg/d group, respectively, and 878 18 and 797 75 pg/d for pregnant and nonpregnant women in the 850 pg/d group, respectively. Serum Folate Weekly serum folate concentrations were measured microbiologically in triplicate. Serum folate response for each experimental group throughout the 12 wk study is illustrated in Figure 6. At baseline, differences in serum folate concentrations (P < 0.05) existed between pregnant and nonpregnant women assigned to consume 450 pg/d (22.32 8.5, 1 1.63 7.4 ng/mL, respectively; mean S.D.) or 850 pg/d (20.33 9.9, 8.70 4.6 ng/mL, respectively) (Figure 7). No differences (P > 0.05) existed among pregnant women assigned to

PAGE 114

101 consume either 450 or 850 |ig/d (22.32 8.5, 20.33 9.9 ng/mL, respectively) or among nonpregnant women assigned to consume these levels (1 1.63 7.4, 8.70 4.6 ng/mL, respectively) (Figure 7). At steady state, no differences (P > 0.05) were detected between pregnant and nonpregnant women within the 450 |ig/d (1 1.78 4.1, 1 1.61 5.0 ng/mL, respectively) or within the 850 (ig/d groups (19.68 5.9, 19.16 4.0 ng/mL, respectively) (Figure 8). Differences (P<0.05) were found between pregnant subjects (1 1.76 4.1, 19.98 5.9 ng/mL) and nonpregnant subjects (11 .53 5.0, 19.16 + 4.0 ng/mL) consuming 450 fig/d compared to 850 ug/d, respectively. 25 20 1 15 E 10 2 — 4— i A X 1 A .A. I A I A A 6 Weeks 10 12 450 NP 850 NP -A• 450 P 850 P Figure 6. Serum folate concentration (group means) throughout the 12 wk study. Steady state was achieved at wk 1 and 9 for the 450 and 850 ^ig/d nonpregnant groups, respectively, and at wk 8 and 1 for the 450 and 850 (ig/d pregnant groups, respectively.

PAGE 115

102 No differences were detected in final mean serum folate concentrations between pregnant and nonpregnant women consuming 450 ug/d (1 1.93 4.7, 12.22 5.7 ng/mL, respectively) or 850 ug/d (18.98 6.8, 19.85 4.2 ng/mL, respectively) (Figure 7). Differences (P<0.05) were found between pregnant subjects (1 1.93 4.7, 18.98 6.8 ng/mL) and nonpregnant subjects (12.21 5.7, 19.85 4.2 ng/mL) consuming 450 or 850 ug/d, respectively (Figure 7). All subjects maintained acceptable serum folate concentrations (> 6.0 ng/mL) (Sauberlich et al. 1974) throughout the study period. Acceptable serum folate concentrations were also maintained throughout the follow-up period in pregnant and nonpregnant women consuming 450 ug/d (15.09 + 6.3, 1 1.39 4.0 ng/mL, respectively) and 850 ug/d (21.00 6.0, 18.49 9.2 ng/mL, respectively). 0 12 Week Figure 7. Group comparisons of initial and final mean serum folate concentration (mean SD) analyzed separately. Bars designated by the same small letter were not significantly different (P>0.05).

PAGE 116

103 450 P 450 NP 850 P 850 NP Figure 8. Serum folate concentration (mean SD) at steady state. Bars designated by the same small number were not significantly different (P>0.05). Red Cell Folate Weekly red cell folate concentrations were measured microbiologically in triplicate. Red cell folate response throughout the 12 wk study is illustrated in Figure 9. Figure 9. Red cell folate concentration (group means) throughout the 12 wk study. Steady state was achieved at wk 1 for the 450 ug/d nonpregnant group and at wk 1 and 7 for the 450 and 850 ug/d pregnant groups. Steady state was not achieved within the 12 wk period by the 850 ug/d nonpregnant group.

PAGE 117

104 At baseline, differences (P<0.05) were not observed in red cell folate concentrations between pregnant and nonpregnant subjects assigned to consume either 450 ug/d (610.43 69.8, 491.72 175.1 ng/mL, respectively) or 850 ug/d (518.05 155.1, 338.80 85.6 ng/mL, respectively) (Figure 10). No differences (P>0.05) were detected at baseline between pregnant women (610.43 69.8, 518.05 155.1 ng/mL ) or nonpregnant controls (491.71 175.1, 338.80 85.5 ng/mL ) assigned to the consume 450 or 850 ug/d, respectively (Figure 10). At steady state, no differences (P>0.05) existed between pregnant and nonpregnant women consuming 450 ug/d (640.94 1 1 1.1, 480.87 170.8 ng/mL, respectively) or between pregnant women consuming 450 or 850 ug/d (640.94 111.1, 764.21 91.6 ng/mL, respectively) (Figure 11). Steady state was not achieved by the nonpregnant group consuming 850 ug/d within the 12 week study period and comparisons to this group could not be made. Differences (P<0.05) in final red cell folate concentrations were not detected between pregnant and nonpregnant women consuming 450 ug/d (537.37 111.2, 459.73 167.0 ng/mL, respectively) or 850 ug/d (700.50 139.7, 543.6 161.9 ng/mL, respectively) (Figure 10). No differences in final mean red cell folate concentrations were detected between pregnant women (537.37 111.2, 700.50 139.7 ng/mL) or nonpregnant controls (459.73 167.0, 543.58 161.9 ng/mL) consuming either 450 or 850 ug/d, respectively (Figure 10). Red cell folate correlated positively (r =0.45; P = 0.03 ) with serum folate. Acceptable red cell

PAGE 118

105 folate concentrations (>160 ng/mL) (Sauberlich et al. 1974) were maintained throughout the study period. In the subjects who participated in the follow-up study, red cell folate was also maintained within normal concentrations by pregnant and nonpregnant women consuming 450 ug/d (576.21 120.0, 436.98 105.0 ng/mL, respectively) and 850 ug/d (753.00 97.1, 550.01 136.2 ng/mL, respectively). 1000r 800 600 3 400 200 450 P 450 NP 850P 850 NP od 12 Week Figure 10. Group comparisons of initial and final mean red cell folate concentration (mean SD) analyzed separately. Bars designated by the same small letter were not significantly different (P>0.05).

PAGE 119

106 900 b 800 E 700 ab ab "D 01 tr 0 450 P 450 NP 850 P 850 NP Figure 11. Red cell folate concentration (mean SD) at steady state. Final mean red cell folate concentration was used for the 850 \xg/d nonpregnant group. Bars designated by the same small number were not significantly different (P>0.05). Since folate intake did not affect hematocrit, data from pregnant and nonpregnant subjects were pooled, respectively. Acclimatization to the diet was not necessary so comparisons were made between initial and final mean hematocrits. Hematocrit changes over time are illustrated in Figure 12. At baseline, no differences (P>0.05) were detected between pregnant and nonpregnant women (37.08 2.3, 38.83 1.7 %, respectively) (Figure 13). Differences (P<0.05) were detected in final mean hematocrit between pregnant and nonpregnant women (34.46 2.3, 38.83 1.6 %, respectively) (Figure 12). These differences would have been greater if additional iron had not been consumed by several pregnant subjects to prevent further decline in hematocrit. Hematocrit.

PAGE 120

107 40 39 38 0 2 4 6 8 10 12 Weeks Figure 12. Hematocrit (group means) throughout the 12 wk study. Week Figure 13. Group comparisons of initial and final hematocrit (mean SD). Bars designated by the same small letter were not significantly different (P>0.05).

PAGE 121

108 Urinary 5-Methyl-Tetrahydrofolate Urinary 5-methyl-THF excretion was measured by HPLC in duplicate at baseline and thereafter on a monthly basis (i.e. wk 0, 4, 8 and 12) for all subjects. The urinary excretion of 5-methyl-THF for each experimental group is shown in Figure 14. 0 2 4 6 8 10 Weeks Figure 14. Urinary 5-methyl-THF excretion (group means) throughout the 12 wk study. Steady state was achieved at wk 1 and 6 for the 450 and 850 u.g/d nonpregnant groups, respectively and at wk 6 and 2 for the 450 and 850 u.g/d pregnant groups.

PAGE 122

109 Differences (P<0.05) in baseline urinary 5-methyl-THF excretion were observed between pregnant and nonpregnant women assigned to consume 450 u.g/d (66. 1 1 106.1, 13.66 17.8 ug/d, respectively) or 850 ug/d (158.30 52.9, 19.00 + 40.0 ug/d, respectively). Differences (P<0.05) were not detected among pregnant (66.12 106.1, 158.30 52.9 ug/d) or nonpregnant women (13.66 17.8, 18.95 40.0 u.g/d) assigned to consume 450 compared to 850 ug/d, respectively largely due to the enormous variability in the urinary excretion of this metabolite (Figure 15). Figure 15. Group comparisons of initial and final mean 5-methyl-THF excretion (mean SD) analyzed separately. Bars designated by the same small letter were not significantly different (P>0.05). At steady state, no differences were detected between pregnant and nonpregnant women consuming 450 ug/d (4.20 1 .4, 6.48 4.7 ug/d, respectively) or 850

PAGE 123

110 ug/d (87.38 44.4, 64.59 26.2 ug/d, respectively). (Figure 15). Differences were detected between pregnant women (4.20 1.4, 87.38 44.4 ug/d) and nonpregnant controls (6.48 4.7, 64.59 + 26.2 ug/d) consuming 450 ug/d compared to 850 ug/d, respectively. No differences (P<0.05) were detected in final mean 5-methyl-THF excretion between pregnant and nonpregnant women consuming 450 ug/d (3.75 1.0, 4.23 3.0 ug/d, respectively) or 850 ug/d (79.02 39.8, 63.85 30.6 ug/d, respectively) (Figure 15). Differences (P<0.05) were detected between pregnant women (3.75 1.0, 79.02 39.8 ug/d) and nonpregnant controls (4.23 3.0, 63.85 30.6 ug/d) consuming 450 ug/d compared to 850 ug/d, respectively. Urinary 5-methyl-THF was positively correlated with serum (r =0.74; P = 0.0001) and red cell folate (r =0.27; P = 0.21). 450 P 450 NP 850 P 850 NP Figure 16. Urinary 5-methyl-THF excretion (mean SD) at steady state. Bars designated by the same small number were not significantly different (P>0.05).

PAGE 124

Ill Urinary Folic Acid Urinary folic acid excretion was measured by HPLC in duplicate at baseline and thereafter on a monthly basis (i.e. wk 0, 4, 8 and 12) for all subjects. At baseline, only 2 of the 12 nonpregnant subjects were excreting folic acid in the amounts of 14.02 and 63.04 ug/d. Among the pregnant women, 2 subjects assigned to the 450 |ag/d and 5 subjects assigned to the 850 u.g/d were excreting folic acid (45.35 41.2, 42.12 41.6). At the end of the 84 d protocol, no detectable folic acid was being excreted by either of the 450 p.g/d groups nor the 850 ug/d pregnant group. Final mean urinary folic acid excretion of the nonpregnant women consuming 850 (j.g/d was 14.57 12.6 ug/d (19% of total urinary folate; 2% folate intake) (Figure 17). 450 P 450 NP 850 P 850 NP Figure 17. Comparisons in final group urinary folic acid concentrations (mean SD). Bars designated by a star were significantly different.

PAGE 125

112 Urinary Total Folate Weekly urinary excretion of total intact folate was measured microbiologically. Urinary folate excretion was higher than that measured by HPLC (5-methyl-THF + folic acid). However, the change over time and differences detected between and within supplementation groups were similar (Table 4). Table 4. Comparison of initial, steady state and final total urinary folate excretion (p.g/d) measured by microbiological assay (MA) and HPLC. Baseline (MA) Baseline* (HPLC) Steady State* (MA) Steady State* (HPLC) Final* (MA) Final* (HPLC) 450 NP 18.2 (24.4) 16.0(18.2) 14.8(8.1) 6.5 (5.3) 18.0(16.1) 4.2 (3.1) 450 P 111.5 (172.1) 81.1 (106.1) 14.1 (4.2) 4.2 (2.2) 9.7 (3.0) 3.8(1.4) 850 NP 29.2 (53.1) 29.4 (39.8) 175.6 (50.7) 79.2 (25.7) 131.9 (55.3) 80.9 (31.4) 850 P 266.2 (214.2) 193.4 (60.3) 157.5 (72.7) 87.4 (44.1) 132.0 (62.4) 79 (39.6) Values are mean SD (n=6). Glomerular Filtration Rates Glomerular filtration rates were computed from serum and urine creatinine concentrations at baseline and thereafter on a monthly basis. The results are shown in the following table. Since folate intake should not affect glomerular filtration rates, pregnant subjects were compared to nonpregnant subjects. No

PAGE 126

113 differences in GFR existed between pregnant subjects and nonpregnant controls nor did GFR increase in pregnant women with progression of pregnancy. Table 5. Glomerular filtration rates (mL/min) at weeks 0, 4, 8 and 12. Wk o' Wk4' Wk 8" Wit 12" Nonpregnant 84.5 24 107.3 19 110.8132 103.8 1 36 Pregnant 93.6126 98.4131 98.2137 81.1 126 Values are mean 1 SD (n=12). Urinary apABG Urinary apABG excretion was measured bimonthly (wk 0, 2, 4, 6, 8, 10,12) in pregnant subjects and monthly (wk 0, 4, 8, 12) in nonpregnant subjects. Urinary apABG excretion throughout the 12 wk period is illustrated in Figure 18. 1 450 NP • •*• 850 NP • •*• 450 P 850 P 12 Weeks Figure 18 Urinary apABG excretion (group means) throughout the 12 wk study.

PAGE 127

114 No differences (P>0.05) existed in baseline group means between pregnant and nonpregnant subjects assigned to consume either 450 ug/d (36.18 13.4, 29.40 9.3 ug/d, respectively) or 850 ng/d (35.58 1 1.0, 29.85 8.9 ug/d, respectively) or between pregnant women (36.18 13.4, 35.6 1 1 ug/d) or nonpregnant women (29.40 9.3, 29.85 8.9 ug/d) assigned to consume either 450 or 850 ug/d folate, respectively (Figure 19). 450 P 450 NP 850 P 850 NP Week Figure 19. Group comparisons of mean initial and final apABG excretion (mean SD) analyzed separately. Bars designated by the same small letter were not significantly different (P>0.05). No differences (P>0.05) were detected in overall group means between pregnant and nonpregnant women consuming 450 ug/d (26.53 8.3, 31.04 + 7.2 ug/d, respectively) or 850 ug/d (30.27 7.6, 36.86 8.6 ug/d, respectively). Differences (P<0.05) were not detected between pregnant subjects (26.53 8.3, 30.27 7.6 ug/d) or nonpregnant controls (31.04 7.2, 36.86 8.6 ug/d)

PAGE 128

115 consuming either 450 or 850 \ig/d, respectively (Figure 20). No differences (P>0.05) were detected in final means between pregnant and nonpregnant women consuming 450 ug/d (21.28 + 7.3, 28.86 9.5 [xg/d, respectively) or 850 p.g/d (35.71 6.5, 32.56 9.4 ug/d, respectively). Pregnant women consuming 450 Hg/d compared to 850 |j.g/d excreted less (P<0.05) urinary apABG (21.3 + 7.3, 35.7 6.5 fig/d, respectively) at the end of the study (Figure 19). 50 40 450 P 450 NP 850 P 850 NP Figure 20. Comparison of overall mean urinary apABG excretion (mean SD). Bars designated by the same small number were not different (P>0.05). Little or no change was observed over time within the groups except for the pregnant women consuming 450 [ig/d in which urinary apABG decreased by 40% (Figure 19). Urinary apABG was not significantly correlated with serum folate (r=0.13; P=0.54), red cell folate (r=.017; P=0.94) or urinary 5-methyl-THF (r=0.03; P=0.89).

PAGE 129

116 Urinary pABG Urinary pABG excretion was measured bimonthly (wk 0, 2, 4, 6, 8, 10, 12) in pregnant subjects and monthly (wk 0, 4, 8, 12) in nonpregnant subjects. Urinary pABG excretion throughout the study period is illustrated in Figure 21. No differences (P>0.05) in urinary pABG excretion at baseline were detected between pregnant women (9.07 3.8, 16.75 10.9 ug/d) or nonpregnant controls (6.42 2.5, 5. 1 1 1 .4 ug/d) assigned to consume either 450 or 850 ug/d, respectively (Figure 22). The pregnant women assigned to consume 850 ug/d were excreting more pABG (P<0.05) than either nonpregnant group at baseline. I j 20 15 10 450 NP 850 NP • • 450 P 850 P 4 8 Weeks 12 Figure 21 Urinary pABG excretion (group means) throughout the 12 wk study. Steady state was achieved at wk 1 and 2 for the 450 and 850 ug/d nonpregnant groups, respectively, and at wk 6 and 2 for the 450 and 850 ug/d pregnant groups. At steady state, no differences (P>0.05) were detected in urinary pABG excretion between pregnant and nonpregnant women consuming 450 ug/d (4.46

PAGE 130

117 2.8 vs. 8.23 2.1 p.g/d, respectively) or 850 [ig/d (1 1.36 4.4 vs. 8.14 2.3 ug/d, respectively) (Figure 23). Differences (P<0.05) were detected at steady state between pregnant women (4.46 2.8, 1 1.36 4.4 ug/d) consuming 450 ug/d compared to 850 ng/d, respectively. No differences (P>0.05) were detected at steady state between nonpregnant consuming 450 ug/d compared to 850 u.g/d. No differences (P>0.05) were detected in final group means between pregnant and nonpregnant women consuming 450 (ig/d (4.62 2.3, 5.73 2.1 (ig/d, respectively) or 850 (ig/d (8.53 2.1, 8.57 2.3 (ig/d, respectively) (Figure 21). Differences (P<0.05) were detected in final group means between pregnant women (4.62 2.3, 8.53 2. 1 u.g/d) consuming 450 (ig/d compared to 850 (ig/d, respectively. Urinary pABG was positively correlated with serum folate (r = 0.42; P = 0.04) and urinary 5-methyl-THF (r = 0.52; P = 0.001). Figure 22. Group comparisons of mean initial and final pABG excretion (mean SD) analyzed separately. Bars designated by the same small letter were not significantly different (P>0.05).

PAGE 131

450 P 450 NP 850 P 850 NP Figure 23. Urinary pABG excretion (mean SD) at steady state. Bars designated by the same small number were not significantly different (P>0.05).

PAGE 132

CHAPTER 5 DISCUSSION AND CONCLUSION This 84 d metabolic study was designed to investigate the folate status response of pregnant women consuming defined folate intakes comparable to the current (400 p-g/d) and former (800 ug/d) RDA for folate in pregnant women. This is the first controlled metabolic study conducted to investigate folate status response of pregnant women to defined folate intakes. Pregnant women during their second trimester (wk 14-25) and nonpregnant controls consumed either 450 or 850 ug/d as a combination of synthetic folic acid and dietary folate. Several folate status parameters were affected by prior consumption of prenatal vitamins containing approximately 1 mg of folic acid by pregnant women and differences were observed at baseline between pregnant and nonpregnant women. Therefore, groups were compared at steady state or the time at which the groups had acclimatized to their defined folate intake. Serum folate concentration is considered a sensitive index of recent folate status (Herbert 1987b) as it is highly influenced by recent dietary intake. However, under metabolic conditions, in which folate intake is constant, serum folate concentration should reflect the overall folate status of the individual. The rapid decline in serum folate concentrations in the pregnant women consuming 450 119

PAGE 133

120 fig/d illustrated serum folate response to a lower folate intake. Once acclimatized to this lower intake, serum folate concentrations were maintained within normal limits in the pregnant group consuming 450 ug/d and were equivalent to the nonpregnant controls on the same supplementation level. Hemodilution did not appear to be a factor in the initial decline in serum folate concentration since the 850 ug/d pregnant group did not experience any decline throughout the 12 wk period and steady state serum folate concentrations were equivalent to the 850 ug/d controls. Hemodilution or expansion of blood volume is a known physiological consequence of pregnancy (Cunningham et al. 1993) and a significant decline in hematocrit from baseline was observed in the pregnant women participating in this study (Figures 12 and 13). It can be assumed, therefore, that the pregnant women had higher total amounts of serum folate at steady state than the nonpregnant controls. Red cell folate concentration reflects liver folate concentration and is considered to be an indicator of long term folate status (Herbert 1987). Folate is accumulated only by developing reticulocytes (Shane 1995) and since the life span of red cells is about 120 days, red cell folate concentration more accurately reflects folate status 2-3 months prior to the time of analysis. However, since red cells are being synthesized daily over a 12 wk period, one should be able to detect some change in concentration. This is especially true during pregnancy when red cell production increases by approximately 33% (Blackburn and Loper, 1992) resulting in greater changes in red cell folate concentration if inadequate amounts

PAGE 134

121 of folate are available at the time of incorporation. In our pregnant group consuming 450 ug/d, red cell folate was maintained throughout the study period and was equivalent to the 450 ug/d control group at steady state. Perhaps more importantly, normal red cell folate concentrations were also maintained throughout the three month follow-up period in pregnant subjects who had consumed 450 ug/d during the study. In addition, 450 ug/d (food folate + supplemental folic acid) was sufficient to maintain mean red cell folate concentrations above 400 ng/mL throughout the study and follow-up period thus supporting the findings of Brown et al. (1997) who reported that red cell folate concentrations of 400 ng/mL could be achieved by folate intakes of at least 450 ug/d. This is informative as red cell folate concentrations of > 400 ng/mL are associated with a low risk of folate-responsive NTDs. Red cell folate concentrations were also maintained above 400 ng/mL in both pregnant and nonpregnant women consuming 850 ug/d throughout the study and follow-up period. Although urinary excretion of folate is not considered to be a reliable index of folate status (Sauberlich et al. 1987), one might expect to observe differences between pregnant and nonpregnant women and between supplementation groups. These differences may reflect metabolic differences thereby enhancing the information gained from blood indices. Pregnant women have been found to excrete significantly more folate than nonpregnant women which is hypothesized by some investigators to partially account for the increase

PAGE 135

122 in requirements during pregnancy (Fleming 1972, Landon and Hytten 1971). The 450 ug/d supplementation groups were excreting similar amounts of 5-methylTHF (5 ug/d) at steady state and no detectable folic acid. The pregnant and nonpregnant women in the 850 ug/d group excreted similar amounts of 5-methylTHF ( 80 vs. 60 ug/d, respectively) at steady state, approximately 15 fold higher than the 450 ug/d groups. Only the 850 ug/d nonpregnant group excreted detectable folic acid (19% of total urinary folate). Saleh et al. (1980) reported a lower folic acid to 5-methyl-THF ratio in patients with malignant disease (state of increased cellular proliferation) compared to controls. They suggested that malignant disease increased the demand for folate and led to more rapid metabolism of folic acid to the reduced folate pool as indicated by an increase in 5-methyl-THF relative to folic acid. Their explanation may apply to our finding that only nonpregnant women in the 850 ug/d group excreted unmetabolized folic acid since pregnancy also represents a period of increased cellular proliferation. The higher urinary excretion of 5-methyl-THF (metabolized form) in the 850 ug/d groups may reflect the saturable process of folate reabsorption from glomerular filtrate by proximal tubular cells (Williams and Huang 1982). Overall, these data on urinary folate excretion indicate that pregnant women are not excreting more folate than nonpregnant women and support the blood data in which no differences between pregnant and nonpregnant women within the same supplementation group were detected.

PAGE 136

123 Glomerular filtration rate (GFR) was measured at baseline and thereafter on a monthly basis in order to assess differences in urinary excretion of folate and catabolites between pregnant and nonpregnant women independent of changes in GFR. Cunningham et al. (1993) reported that GFR increases as much as 50 percent by the beginning of the second trimester and has been found to persist to term. Our data indicated no difference in GFR between pregnant and nonpregnant women at baseline or at the end of the study for reasons unexplained. Glomerular filtration rate was thus not considered when assessing urinary excretion of folate derivatives or catabolites. Folate catabolites, pABG and the more predominant form, apABG, were measured at numerous time points throughout the 1 2 wk study. Our results did not confirm the work of McPartlin et al. (1993) who reported a doubling of folate catabolism by pregnant women during the 2 nd trimester. In contrast, we found that the mean urinary excretion of both pABG and apABG was equivalent to that of our nonpregnant controls. No differences (P>0.05) in baseline, overall or final means were detected in urinary apABG excretion between pregnant and nonpregnant women within the same supplementation group which suggests that pregnancy did not affect urinary apABG excretion during this 84 d study. Final mean apABG excretion was lower (P < 0.05) in pregnant women consuming 450 ug/d compared to 850 j^g/d indicating that over the long-term, urinary apABG was affected by folate intake. The 40% decline in urinary apABG excretion from baseline measurements in pregnant women consuming 450 u.g/d also illustrated

PAGE 137

124 the effect of folate intake during pregnancy on apABG excretion since these women were previously consuming approximately twice the amount of folate. These observations may suggest that pregnant women become more efficient at conserving or recycling folate when folate intake is somewhat restricted. Urinary pABG excretion was greater (P<0.05) in pregnant women compared to nonpregnant controls at baseline which may have reflected partial degradation of urinary intact folate which was also higher at baseline in pregnant women due to prior supplementation with folic acid. McNulty et al. (1993) found that urinary excretion of pABG in rats appeared to be dependent on dietary folate level whereas as apABG excretion was not. At steady state, urinary pABG excretion in pregnant women was equivalent to nonpregnant controls within the same supplementation group indicating that pregnancy did not affect urinary pABG excretion. Pregnant women consuming 850 ug/d compared to 450 ug/d were excreting more (P<0.05) urinary pABG. This may be explained by the positive correlation (r=0.53; P<0.01) between urinary pABG and 5-methyl-THF excretion which suggests that intact folate may have degraded in the urine. Differences (P<0.05) were observed in urinary 5-methyl-THF excretion at steady state between pregnant subjects consuming 850 versus 450 pg/d. However, nonpregnant women consuming 850 ug/d versus 450 ug/d were also excreting more urinary 5-methyl-THF but not more pABG at steady state. As the study progressed, however, folate intake appeared to be affecting the urinary excretion of pABG in nonpregnant subjects although significant differences in final group

PAGE 138

125 means were not obtained. The origin of urinary pABG may represent any of the following: (1) endogenous pABG that does not get acetylated prior to excretion; (2) folate breakdown occurring in circulation or in the bladder (Geohegan et al. 1995); (3) dietary pABG; (4) degradation of urinary intact folates; or (5) apABG that gets deacetylated (McNulty et al. 1993). Although our findings of equivalent apABG excretion between nonpregnant and pregnant women differed from McPartlin et al. (1993) who reported a doubling of apABG excretion by pregnant women during the second trimester, agreement existed in nonpregnant catabolite values between this study, McPartlin et al. (1993) and Kownacki-Brown et al. (1995). The mean urinary excretion of apABG and pABG in our study were ~ 34 and 7.2 ug/d, respectively compared to -32 and 13 ug/d, respectively (McPartlin et al. 1993), and 66 u.g/d (apABG + pABG) prior to supplementation (Kownacki-Brown et al. 1995). Our mean urinary pABG excretion is lower than that of McPartlin et al. (1993) and may reflect removal of intact folates by affinity chromatography columns containing folate binding protein prior to catabolite isolation and quantitation. If affinity chromatography columns are not used, unremoved intact folates may break down upon acidification or during elution from the ion exchange columns. The lower urinary pABG concentrations may also be the result of lower dietary pABG intake since our subjects received the majority of folate from folic acid, the most stable form of the vitamin.

PAGE 139

126 The concept of deriving requirements based solely on catabolite excretion may be inappropriate during pregnancy and even in nonpregnant humans since other avenues of folate loss or accretion are not considered. For example red cell production increases by 33% (Blackburn and Loper, 1992) resulting in greater folate uptake by reticulocytes. This extra demand for folate will not be reflected in urinary catabolite concentration. The doubling in uterus size may require more folate coenzymes of which the majority may be recycled upon utilization and not degraded and excreted as urinary catabolites. It is recognized that the developing fetus drains maternal folate supplies as evidenced by the significantly higher plasma and red cell folate concentrations in newborns compared to maternal concentrations (Ek and Magnus 1981, Giles 1 966). Folate accumulated by the placenta and delivered to the developing fetus will not be estimated by quantitation of urinary folate catabolites. In addition, measurement of urinary catabolite excretion alone does not account for endogenous fecal folate loss which may be significant (Krumdieck et al. 1978). Aside from other avenues of folate loss, studies conducted in rats (Connor et al. 1979, Pheasant et al. 1981) and humans (Saleh et al. 1980) have identified p-acetamidobenzoate (pABA) as a folate catabolite whose quantitation would increase folate requirements when based on catabolite excretion. However, these studies compared rates of catabolism only and did not measure true endogenous urinary pABA total excretion, hence the contribution of pABA to folate requirements is unknown.

PAGE 140

127 The differences between our findings and those of McPartlin et al. (1993) may be the result of differences in protocols since, in the present study, folate intake was highly controlled for a 12 wk period during which time 24 h urine collections were obtained. In contrast, folate intake was only controlled for a 42 h period in the study by McPartlin et al. (1993). It is possible that pregnant women consumed more folate containing food items prior to study participation and thereby consumed more pABG. In rats, intravenous administration of radiolabeled pABG resulted in substantial conversion to apABG (Murphy et al. 1976). In addition, Pheasant et al. (1981) reported that urinary pABG and its acetylated derivative were detected following an oral dose of radiolabeled pABG. The fate of dietary pABG in humans is unknown although it has been reported (unpublished data) that pABG cannot enter cells and is rapidly excreted in urine (Johns et al. 1961) suggesting that it is excreted unacetylated. Another potential explanation for the discrepancy is the effect of weight gain on apABG excretion. In rats, urinary apABG excretion was consistently higher in the pregnant rats with free access to food compared to a pregnant pair fed group (pair fed to nonpregnant controls) suggesting that one component of the observed elevation in folate catabolism was due to greater food intake (McNulty et al. 1993). Furthermore, comparisons of urinary apABG excretion in nonpregnant female rats at weights comparable to those attained by the pregnant rats indicated a positive association between apABG excretion and weight gain (McNulty et al. 1993). ). In our protocol, weight gain was strictly controlled (5.0 kg during the 12 wk period)

PAGE 141

128 which provides another possible explanation for differences between protocols since in the study of McPartlin et al.(1993) food intake was not controlled longterm. Although unrestricted calorie intake and perhaps greater weight gain may partially account for the higher apABG excretion observed by McPartlin et al. (1993), it does not fully explain the lack of difference in apABG excretion between our nonpregnant and pregnant subjects. Our definitions of second trimester also differed (personal communication) since urinary catabolite excretion was measured from wk 14 to 26 in this study and at one time point between wk 26 to 32 in the study of McPartlin et al. (1993). The equivalent response of pregnant and nonpregnant to folate intakes of either 450 ug/d or 850 ug/d was not anticipated secondary to the pregnant woman's increased demand for folate coenzymes required for DNA synthesis and greater folate uptake by developing red cells and fetus. In order to detect differences in folate status response between pregnant and nonpregnant women, it is likely that lower folate intakes are necessary. It is also possible that differences may have been detected with the progression of pregnancy. Our finding of equivalent response between pregnant and nonpregnant women does not suggest that pregnant women do not require more folate but does suggest that 450 u.g/d was sufficient in maintaining folate status equivalent to that of nonpregnant controls. Overall, these data suggest that 450 ug/d derived from both food (120 ug) and synthetic folic acid (330 ug) is adequate to maintain normal folate status in

PAGE 142

129 well-nourished pregnant women during their second trimester. However, the question remains as to how to translate this finding into food folate and an RDA. Recent studies (Cuskelly et al. 1996, Pfeiffer et al. 1997) indicate that foods fortified with synthetic folic acid are highly available unlike naturally occurring food folates. Therefore, one may hypothesize that synthetic folic acid consumed with meals is more available than endogenous food folate (50%) (Sauberlich et al. 1987) but less available than the essentially complete absorption of synthetic folic acid (100%) consumed under fasting conditions (Gregory 1995). A reasonable, although somewhat conservative, approach is to assume that synthetic folic acid consumed with meals is approximately 75% available. Based on the above assumptions, our subjects in the 450 ug/d group consumed 307 ug/d available folate which translates into 615 ug/d dietary equivalents. Our dietary equivalent estimate resembles the estimations of Chanarin et al. (1968a, 1968b), Cooper et al. (1970) and McPartlin et al. (1993). It also supports the recommendation of Willoughby et al. (1968) regarding the administration of 300-350 ug/d synthetic folic acid throughout gestation assuming an average folate intake of 1 50 ug/d (low dietary folate). It is questionable whether the majority of pregnant women can consume 600 ug/d of folate from diet alone (LSRO 1980, Brown et al. 1997, Huber et al. 1988) although folic acid enrichment of cereal-grain foods (140 jag/ 1 OOg product) effective January 1,1998 (FDA 1996) is estimated to increase average daily consumption by 80-100 u.g (Brown et al., 1997, FDA 1996). Our findings indicate that prenatal vitamins

PAGE 143

130 containing more than 500 \xg/d are not necessary to maintain adequate folate status in well-nourished pregnant populations and support the findings of Lowenstein et al. (1966) which suggested that provision of 500 u.g/d synthetic folic acid, in addition to low dietary folate intake, may be above the minimal daily requirement since higher serum folates were observed in pregnant women compared to nonpregnant controls (1 1.0, 7.02 ng/mL, respectively). Because the effects of oversupplementation with folic acid on the developing fetus are unknown (Scott et al. 1991), high supplemental doses (> 1000 u.g/d) should be avoided under normal circumstances. In conclusion, 450 ug/d consumed as a combination of synthetic folic acid and dietary folate or approximately 600 ug/d dietary equivalents is sufficient to maintain normal folate status throughout the second trimester. The first and third trimesters or postpartum period were not investigated under controlled conditions in the current study, therefore future areas of research should encompass these time frames under controlled conditions as well as the response of previously unsupplemented pregnant women to defined folate intakes.

PAGE 144

CHAPTER 6 SUMMARY An 84 d metabolic study was conducted in order to assess the folate status response of pregnant women and nonpregnant controls to folate intakes comparable to the current and former RDA for pregnant women, 400 and 800 ug/d, respectively. The overall goal of this investigation was to provide metabolic data that may assist in the interpretation of current dietary recommendations. Subjects consumed either 450 or 850 ug/d of folate derived from both synthetic and dietary sources. A series of folate response parameters were then evaluated in relation to folate intake in both pregnant subjects and nonpregnant controls. Following sample analyses, the data indicated that the response of pregnant women was equivalent to that of nonpregnant controls in either the 450 or 850 ug/d supplementation groups. In addition, normal folate nutriture was maintained by subjects in all groups. These results suggested that 450 ug/d was adequate during the second trimester of pregnancy. Since our overall goal was to assist in the interpretation of current dietary recommendations, it was necessary to consider the bioavailability of the folate consumed by our subjects which was a mixture of synthetic folic acid and dietary folate. It is generally agreed that dietary folate is approximately 50% available and that synthetic folic acid, consumed under fasting conditions, is approximately 100% available. Our subjects consumed the 131

PAGE 145

synthetic folic acid (330 pg/d) along with food thus we conservatively estimated it was 75% available. Following adjustments for bioavailability, subjects in the 450 pg/d consumed approximately 600 pg/d dietary equivalents. In conclusion, our data suggests that 600 pg/d dietary folate is adequate in maintaining normal folate status throughout the second trimester of pregnancy.

PAGE 146

APPENDIX Microbiological Assay Procedures Microplate Medium Preparation for L. casei Folic acid medium-Difco (dehydrated) (230.3 g) was weighed into a tared plastic weigh boat. The dehydrated medium was added to a glass flask to which 2,450 mL of Milli-Q water was added. The flask was covered with foil and the media mixed and brought to a boil. After one to two minutes, the media was removed from heat and cooled to room temperature. The media was then dispensed into disposable glass tubes, capped and placed into a metal rack holder. The tubes containing the media were covered with foil and autoclaved for 5 minutes. After cooling to room temperature, the medium was stored in the refrigerator at 4 C. The microorganism, L. Casei, was added to the media just prior to use. Folic Acid Standard Preparation Ten milligrams of folic acid (Sigma F-7876) was weighed and dissolved in 0. 1 N NaOH. The dissolved folic acid solution was transferred to a 100 mL volumetric flask and filled to 100 mL with Milli-Q water to a final concentration of 100 ug/mL. ALIO dilution was performed by taking 100 uL of folate solution and adding 900 uL of 0. 1 M KH 2 P0 4 (pH 7). The exact 133

PAGE 147

134 concentration of the folate solution was determined spectrophotometrically using 0.1 M KH 2 P0 4 (pH 7) as the blank. Concentration by weight: 0.1 mg/ml x mmol/441 .4mg = 0.2266 x 10" 3 mg/mmol = 226.6 x 10" 3 ug/mmol = 226.6 ug/L = 226.6 uM. Expected Absorbance for folic acid at pH 7.0 and 282 nm with an extinction coefficient of 27,600 is: 226.6 |iM x 27,600 x 10" 6 x 10' 1 = 0.6253 A 1:10 000 dilution of the folate solution was performed after adjusting for actual concentration. Sodium ascorbate (0.1%; w/v) was added to the folate standard (1 L). One milliliter portions were aliquoted while gently mixing to ensure homogeneity. The standards were stored at -20C The remaining folate stock solution with a concentration of 100 p.g/mL was stored at -20 C.

PAGE 148

LITERATURE CITED Angier, R.B., Boothe, J.H., Hutchings, B.L., Mowat, J.H., Semb, J., Stockstad, E.L.R., Subba, R.Y., Waller, C.W., Cosulich, D.B., Fahrenbach, M.J., Hultquist, M.E., Kuh, E., Morthey, E.H., Sceger, D.R., Sickels, J.P.& Smith, J.M. (1945) Synthesis of a compound identical with the L. casei factor isolated from liver. Science. 102: 227. Antony, A.C. (1996) Folate receptors. Annu. Rev. Nutr. 16: 501-521. Baker, B.L., Santi, D.V., Almaula, P. I. & Werkheiser, W. C. (1964) Analogs of tetrahydrofolic acid. X. Synthetic and enzymatic studies on the contribution of the p-aminobezoyl-glutamate moiety of pyrimidyl anlogs to binding to some folic cofactor area enzymes. J. Med. Chem. 7: 24-30. Bailey, L.B. (1988) Factors affecting folate bioavailability. Food. Tech. 42: 206-212. Bailey, L.B. (1990a) The role of folate in human nutrition. Nutrition Today September/October: 12-19. Bailey, L.B. (1990b) Folate status assessment. J. Nutr. 120: 1508-151 1. Bailey, L.B. (1992) Evaluation of a new recommended dietary allowance for folate. J. Am. Diet. Assoc. 92: 463-468,471. Bailey, L.B. (1995) Folate Requirements and dietary recommendations. In: Folate in Health and Disease (L.B. Bailey, ed.), pp. 123-151. Marcel Dekker, Inc., New York, NY. Bailey, L.B., Barton, L.E., Hillier, S.E. & Cerda, J.J. (1988) Bioavailability of mono and polyglutamyl folate in human subjects. Nutr. Rept. Int. 38: 509518. Bailey, LB., Cerda, J.J., Block, B.S., Busby, J.M., Vargas, L, Chandler, C.J. & Halsted C. (1984) Effect of age on poly-and monoglutamyl folacin absorption in human subjects. J. Nutr. 1 14: 1770-1776. 135

PAGE 149

136 Bailey, L.B., Mahan, C.S. & Dimperio, D. (1980) Folacin and iron status in lowincome pregnant adolescents and mature women. Am. J. Clin. Nutr. 33:. 1997-2001. Barford, P. A., Staff, F.J. & Blair, J.A. (1978) The metabolic fate of [2,4 C]-and [3\5',93 H]-folic acid in the rat. Biochem. J. 174: 579-583. Bartels, P.C., Helleman, P.W. & Soons, J.B. (1989) Investigation of red cell sizedistribution histograms related to folate, vitamin B 12 and iron state in the course of pregnancy. Scand. J. Clin. Lab. Invest. 49: 763. Bates, C.J., Black, A.E., Phillips, D.R., Wright, A.F. & Southgate, D.A. (1982) The discrepancy between normal folate intakes and the folate RDA. Human Nutr. Appl. Nutr. 36: 422-429. Baumslag, N., Eldstein, T. & Metz, J. (1970) Reduction of incidence of prematurity by folic acid supplementation in pregnancy. Brit. Med. J. 1: 16. Blackburn, S. & Loper, D.L. (eds) (1992) The hematologic and hemostatic systems. In: Maternal, Fetal, and Neonatal Physiology. A Clinical Perspective, pp. 159-200. WB Saunders Co., Philadelphia, PA. Blakley, R.L. (1969) The Biochemistry of Folic Acid and Related Pteridines. pp. 92-94 Wiley, New York, NY. Blair, J.A. (1957) Isolation of isoxanthopterin from human urine. Biochem. J. 68: 385-387. Bratton, A.C., Marshall, E.K. (1939) A new coupling component for sulfanilamide determination. J. Biol. Chem. 128: 537-550. Brown, J.E., Jacobs, D.R., Hartman, T.J., Barosso, G.M., Stang, J.S., Gross, M.D. & Zueske, MA. (1997) Predictors of red cell folate level in women attempting pregnancy. J. Am. Med. Assoc. 277: 548-552. Browne, T.R. (1986) Stable isotopes in pharmacology studies: present and future. J.Clin. Pharmacol. 26: 485-489. Butterworth, C.E., Newman, A. J. & Krumdieck, C.L. (1974) Tropical sprue: a consideration of possible etiologic mechanism with emphasis on pteroylpolyglutamate metabolism. Trans. Am. Clin. Climatol. Assn. 86: 11.

PAGE 150

137 Centers for Disease Control (1992) Recommendations for the use of folic acid to reduce the number of cases of spina bifida and other neural tube defects. Morb. Mort. Wkly. Rep. 41:1-7. Chanarin, L, Rothman, D., Perry, J. & Stratfull, D. (1968a) Normal dietary folate, iron, and protein intake, with particular reference to pregnancy. Brit. Med. J. 2: 394-397. Chanarin, I., Rothman, D., Ward, A. & Perry, J. (1968b) Folate status and requirement in pregnancy. Br. Med. J. 2: 390-394. Colman, N. (1981) Laboratory assessment of folate status. Clinics in Laboratory Medicine 1 : 755-796. Colman, N., Barker, M., Green, R. & Metz, J. (1974) Prevention of folate deficiency in pregnancy by food fortification. Am. J. Clin. Nutr. 27(4): 339-344. Colman, N., Green, R. & Metz, J. (1975) Prevention of folate deficiency by food fortification. II. Absorption of folic acid from fortified staple foods. Am. J. Clin. Nutr. 28: 459. Colman, N., Larsen, J.V., Barker, M., Barker, A., Green, R. & Metz J. (1975) Prevention of folate deficiency by food fortification. HI. Effect in pregnant subjects of varying amounts of added folic acid. Am. J. Clin. Nutr. 28: 465-470. Combs, GF (1992) Folate. In: The Vitamins — Fundamental Aspects in Nutrition and Health, pp. 357-376. Academic Press, INC., San Diego, CA.. Connor, M.J., Pheasant, A.E., Blair, J. A. (1979) The identification of pacetamidobenzoate as a folate degradation product in rat urine. Biochem. J. 178: 795-797. Cooper, B.A., Cantlie, G.S.D. & Brunton, L. The case for folic acid supplements during pregnancy. Am. J. Clin. Nutr. 23: 848-854. Cosulich, D.B., Seeger, D.R., Fahrenbach, M.J., Roth, B., Moway, J.H., Smith, J.M. Jr. & Hultquist, M.E. (1951) Analogs of Pteroylglutamic acid. VI. 3\5'-Dihalopteroyl derivatives. J.Am. Chem. Soc. 73: 2554-2557. Crandall, B.F., Corson, V.L., Goldberg, J.D., Knight, G. & Salafsky, I.S. (1995) Folic acid and pregnancy. Am. J. Med. Gen. 55: 134-135.

PAGE 151

138 Cunningham, F.G., MacDonald, P.C., Gant, N.F. (1989) Physiology in pregnancy. In: Williams, ed. Obstetrics. Appleton & Lange, San Mateo, CA. Cuskelly, G.J., McNulty, H. & Scott, J.M. (1996) Effect of increasing dietary folate on red cell folate: implications for prevention of neural tube defects. Lancet 347: 657-659. Czeizel, A. & Dudas, L (1992) Prevention of first occurrences of neural tube defects by periconceptional vitamin supplementation. N. Engl. J. Med. 327: 1831-1835. Da Costa, M. & Rothenberg, S.P. (1996) Purification and characterization of folate binding proteins from rat placenta. Biochim. et Biophys. Acta 1292: 23-30. Daly, L.E., Kirke, P.N., Molloy, A., Weir, D.G. & Scott, J.M. (1995) Folate levels and neural tube defects. J. Am. Med. Assoc. 274: 1698-1702. Das, K.C., Herbert, V., Colman, N. & Longo, D.L. (1978) Unmasking covert folate deficiency in iron-deficient subjects with neutrophil hypersegmentatiomdU suppression tests on lymphocytes and bone marrow. Br. J. Haematol. 39: 357-375. Davis, B.A., Bailey, L.B., Gregory, J.F. Ill, Toth, J.P., Dean, J & Stevenson, R.E. (1995) Folic acid absorption in women with a history of pregnancy with neural tube defect. Am. J. Clin. Nutr. 62: 782-784. Dawson, D.W. (1966) Microdoses of folic acid in pregnancy. J. Obstet. Gynaec. Brit. Cwlth. 73: 44-48. Dinning, J.S., Sime, J.T., Work, P.S., Allen, B. & Day, P.L. (1957) The metabolic conversion of folic acid and citrovorum factor to a diazotizable amine. Arch.Biochem. Biophys. 66: 114-119. Ek, J. (1980) Plasma and red cell folate in newborn infants and their mothers in relation to gestation age. J. Pediatr. 97: 288. Ek, J. & Magnus, E.M. (1981) Plasma and red blood cell folate during normal pregnancies. Acta. Obset. Gynecol. Scand. 60: 247-251. Fleming, A.F. (1972) Urinary excretion of folate in pregnancy. Br. J. Obstet. Gynaecol. 72: 916-920.

PAGE 152

139 Food and Drug Administration (March 5, 1996) Food standards: Amendment of standards of identity for enriched grain products to require addition of folic acid. 21. CFR Parts 136, 137, 139. Fed. Regist. 61: 8781-8797. Geoghegan, F.L., McPartlin, J.M., Weir, D.G. & Scott, J.M. (1995) Paraacetamidobenzoylglutamate is a suitable indicator of folate catabolism in rats. J. Nutr. 125: 2563-2570. Gibaldi, M.D. & Perrier, D. (1982) Pharmacokinetics. Marcel-Dekker, Inc., New York, NY. Gibson, R.S. (1990) Assessment of the status of folate and vitamin B 12 In: Principles of Nutritional Assessment, pp. 461-486 Oxford University Press, New York, NY. Giles, C. (1966) An account of 335 cases of megaloblastic anemia of pregnancy and the puerpium. J. Clin. Path. 19: 1-11. Giovannucci, E., Stampfer, M., Colditz, G, Rimm, E., Trichopolous, D., Rosner, B., Speizer, F. & Willet, W. (1993) Folate, methionine, and alcohol intake and risk of colorectal adenoma. J. Natl. Cancer. Inst. 128: 225-236. Goldenberg, R.L., Tamura, M.D., Cliver, S.P., Cutter, G.R., Hoffman, H.D. & Copper, R.L. (1992) Serum folate and fetal growth retardation: a matter of compliance? Obstet. Gynecol. 79: 719-722. Gregory, J.F. III. (1989) Chemical and nutritional aspects of folate research: analytical procedures, methods of folate synthesis, stability, and bioavailability of dietary folates. Adv. Food Nutr. Res. 33: 1-101. Gregory, J.F. III. (1990) Improved synthesis of [3',5'2 H 2 ]folic acid: extent and specificity of deuterium labeling. J. Agric. Food. Chem. 38(4): 1073-1076. Gregory, J.F. III. (1995) Bioavailability of folate. In: Folate in Health and Disease. (L.B.Bailey, ed) pp. 195-235 Marcel Dekker, Inc., New York, NY. Gregory, J.F. Ill, Bailey, L.B., Toth, J.P. & Cerda, J.J. (1990) Stable-isotope methods for assessment of folate bioavailability. Am. J. Clin. Nutr. 5 1 : 212-215. Gregory, J.F. m, Bhandari, S.D., Bailey, L.B., Toth, J.P., Baumgartner, T.G. & Cerda, J.J. (1991) Relative bioavailability of deuterium-labeled monoglutamyl and hexaglutamyl folates in human subjects. Am. J. Clin. Nutr. 53: 736-740.

PAGE 153

140 Gregory, J.F. Ill, Engelhardt, R., Bhandari, S.D., Sartain, D.B., Gustafson, S.K. (1990) Adequacy of extraction techniques for determination of folate in foods and other biological materials. J. Food. Comp. Anal. 3: 1-11. Gregory, J.F. Ill, Sartain, D.B. & Day, B.P.F. (1984) Fluorometric determination of folacin in biological materials using high performance liquid chromatography. J. Nutr. 1 14: 341-353. Gregory, J.F. HI, Toth, J.P. (1990) Stable-isotope methods for in vivo investigation of folate absorption and metabolism. In: Folic Acid Metabolism in Health and Disease. (M.F. Picciano, E.L.R. Stokstad & J.F.Gregory, eds), pp. 151-169. Wiley-Liss Inc., New York, NY. Gregory, J,F, III, Toth, J.P. (1988) Chemical synthesis of deuterated folate monoglutamate and in vivo assessment of urinary excretion of deuterated folates in man. Anal. Biochem. 170: 94-104. Groff, J.L., Gropper, S.S., Hunt, S.M. (1995) The water soluble vitamins. In: Advanced Nutrition and Human Metabolism, pp. 262-270. West Publishing Co., New York, NY. Hall, M.H., Pirani, B.B.K. & Campbell, D. (1977) The cause of the fall in serum folate in normal pregnancy. Brit. J. Obstet. Gynaecol. 83: 132-136. Halsted, C.H. (1990) Intestinal absorption of dietary folates. In: Folic Acid Metabolism in Health and Disease. (M.F. Picciano, E.L.R. Stokstad & J.F.Gregory, eds), pp. 151-169. pp. 23-45. Wiley-Liss Inc., New York, NY. Hansen, H. & Rybo, G. (1967) Folic acid dosage in profylactic treatment during pregnancy. Acta. Obstet. Gynec. Scand. 46(7): 107-112. Henderson, G.I. (1990) Folate-binding proteins. Ann.Rev. Nutr. 10: 319. Henderson, G.I., Perez, T., Schenker, S., Mackins, J. & Antony, A.C. (1995) Maternal-to-fetal transfer of 5-methyltetrahydrofolate by the perfused human placental cotyledon: Evidence for a concentrative role by placental folate receptors in fetal folate delivery. J. Lab.Clin. Med. 126: 184-203. Herbert, V. (1962) Experimental nutritional folate deficiency in man. Trans. Assoc. Am. Physicians 75: 307-320. Herbert, V. (1987a) Recommended dietary intakes (RDI) of folate in humans. Am. J. Clin. Nutr. 45: 661-670.

PAGE 154

141 Herbert, V. (1987b) Making sense of laboratory tests of folate status: Folate requirements to sustain normality. Am. J. Hematol. 26: 199-207. Herbert, V. (1990) Development of human folate deficiency. In: Folic Acid Metabolism in Health and Disease. (M.F. Picciano, E.L.R. Stokstad & J.F.Gregory, eds), pp. 195-210. pp. 23-45. Wiley-Liss Inc., New York, NY. Herbert, V. & Das, K.C. (1994) Folic acid and vitamin B 12 In: Modern Nutrition in Health and Disease (M.E. Shils, J.A. Olson & M. Shike eds). pp. 402425. Lea and Fabiger, Philadelphia, PN. Hibbard, B.M. (1964) The role of folic acid in pregnancy with particular reference to anemia, abruption and abortion. J. Obstet. Gynaecol. Br. Commonw. 71:529-542. Hillman, R.S., McGuffin, R. & Campbell, C. (1977) Alcohol interference with the folate enterohepatic cycle. Trans. Assoc.Am. Physicians 90: 145-156. Huber, A.M., Wallins, L.L. & DeRusso, P. (1988) Folate nutriture in pregnancy. J. Am. Diet. Assoc. 88: 791-794. Iyengar, L & Babu, S. (1975) Folic acid absorption in pregnancy. Br. J. Obstet. Gyn&ol. 82: 20-23. Iyengar, L. & Rajalakshmi, K. (1975) Effect of folic acid supplementation on birth weights of infants. Am. J. Obst. Gyn. 122: 332. Jacob, R.A., Wu, M., Henning, S.M. & Swenseid, M. (1994) Homocysteine increases as folate decreases in plasma of healthy men during short-term dietary folate and methyl group restriction. J. Nutr. 124: 1072-1080. Johns, D.G., Sperti, S. & Burgen, A.S.V. (1961) The disposal of tritiated folic acid injected intravenously in man. Canad. Med. Assoc. J. 84: 77-79. Jukes, T.H., Franklin, A.L., Stokstad, E.L.R. & Boehne, J.W. (1947) The urinary excretion of pteroylglutamic acid and certain related compounds. 13501355. Kauwell, G.P., Bailey, L.B., Gregory, J.F., Bowling, D.W., Cousins, R.J. (1995) Zinc status is not adversely affected by folic acid supplementation and zinc intake does not impair folate utilization in human subjects. J. Nutr. 125: 66-72.

PAGE 155

142 Kirke, P.N., Molloy, A.M., Daly, L.E., Burke, H., Weir, D.G. & Scott, J.M. (1993) Maternal plasma folate and vitamin B 12 are independent risk factors for neural tube defects. Quart. J. Med. 86: 703-708. Kownacki-Brown, P.A., Wang, C, Bailey, L.B., Toth, J.P.& Gregory, J.F. HI. (1993) Urinary excretion of deuterium-labeled folate and the metabolite paminobenzoylglutamate. J. Nutr. 123: 1101-1108. Krumdieck, C.L. (1990) Folic acid. In: Present Knowledge in Nutrition. (M.L. Brown ed), pp. 179-188. International Life Sciences Institute-Nutrition Foundation, Washington, DC. Krumdieck, C.L., Fukushima, K., Fukushima, T., Shiota, T. & Butterworth, C.E. (1978) A long-term study of the excretion of folate and pterins in a human subject after ingestion of C folic acid, with observations on the effect of diphenylhydantoin administration. Am. J. Clin. Nutr. 31: 88-93. Krungkrai, Y. (1987) High-performance liquid chromatographic assay for pteroylpolyglutamate hydrolase. J. Chromatogr. 417: 47-56. Landon, M.J. & Hytten, F.E. (1971) The excretion of folate in pregnancy. J.Obstet. Gynaecol. Br. Comm. 78(9): 769-775. Landon, M.J. & Hytten, F.E. (1972) Plasma folate levels following an oral load of folic acid during pregnancy. J. Obstet. Gynaecol. Br. Commonw. 79(7): 577-583. Laurence, K.M., James, N., Miller, M., Tennant, G.B. & Campbell, H. (1981) Double blind randomized controlled trial of folate treatment before conception to prevent recurrence of neural tube defects. Br. Med. J. 282: 1509-1511. Lee, H.C., Shoda, R., Krall, J.A., Foster, J.D., Selhub, J. & Rosenberry, T.L. (1992) Folate binding protein from kidney brush border membranes contains components of a glycoinositol phospholipid anchor. Biochemistry 31: 3236-3243. Life Sciences Research Office (1980) Assessment of the folate nutritional status of the US population based on data collected in the second national health and nutrition examination survey, 1976-1980. Federation of American Societies for Experimental Biology, Bethesda, MD.

PAGE 156

143 Lindenbaum, J. & Allen, R.H. (1995) Clinical Spectrum and Diagnosis of Folate Deficiency. In: Folate in Health and Disease (L.B. Baileyed), pp. 43-73. Marcel Dekker, Inc., New York, NY. Lowenstein, L., Cantlie, G., Ramos, O. & Brunton, L. (1966) The incidence and prevention of folate deficiency in a pregnant clinic population. Canad Med Ass J. 95: 797-806. Lucock, M.D., Wild, J., Smithells, R.W. & Harley, R. (1989a) In vivo characterization of the absorption and biotransformation of pteroylmonoglutamic acid in man: a model for future studies. Biochem. Med. Metab. Biol. 42: 30-42. Lucock, M.D., Wild, J., Smithell. R,W, & Harely, R. (1989b) Biotransformation of pteroylmonoglutamic acid during absorption: implications of Michaelis-Menten kinetics. Eur. J. Clin. Nutr. 43: 631-635. Martin, J.I., Landen, W.O. & Soliman, A.-G.M. (1990) Application of a trienzyme extraction for total folate determination in foods. J. Assoc. Off. Anal. Chem.73(5): 805-808. Martinez, O.B. (1980) Red cell folate values of a group of non-pregnant mothers. Can. J. Public Health. 71: 163. McNulty, H., McPartlin, J.M., Weir, D.G. & Scott, J.M. (1993) Folate catabolism is increased during pregnancy in rats. J. Nutr. 123: 1089-1093. McNulty, H., McPartlin, J.M., Weir, D.G & Scott, J.M. (1995) Folate catabolism is related to growth rate in weanling rats. J. Nutr. 125(1): 99-103. McPartlin, J.M., Courtney, G., McNulty, H., Weir, D.G. & Scott, J.M. (1992) The quantitative analysis of endogenous folate catabolites in human urine. Anal.Biochem. 206: 256-261. McPartlin, J.M., Halligan, A., Scott, J.M, Darling, M. & Weir, D.G. (1993) Accelerated folate breakdown in pregnancy. Lancet 341: 148-149. Metz, J. (1970) Folate deficiency conditioned by lactation. Am. J. Clin. Nutr. 23: 843-847. Mills, J.L., McPartlin, J.M., Kirke, P.N., Lee, Y.J., Conley, M.R., Weir, D.G. & Scott, J.M. (1995)Homocysteine metabolism in pregnancies complicated by neural tube defects. Lancet 345: 149-151.

PAGE 157

144 Mills, J.L., Rhoads, G.G., Simpson, J.L., Cunningham, G.C., Conly, M.R., Lassman, M.R., Walden, M.E., Depp, O.R., Hoffman, H.F. & NIH NTD study group. (1989) The absence of a relationship between the periconceptional use of vitamins and NTDs. N. Engl. J. Med. 321: 430435. Milunsky, A., Jick, H., Jick, S.S., Bruell, C.L., MacLaughlin, D.S., Rothman, K.J. & Willett, W. (1989) Multivitamin/folic acid supplementation in early pregnancy reduces the prevalences of NTDs. J.A.M.A. 262: 2847-2852. Mitchell, H.K., Snell, E.E. & Williams, R.J. (1941) The concentration of "folic acid". J. Am. Chem. Soc. 63: 2284. MRC Vitamin Study Research Group (1991) Prevention of NTDs: Results of the Medical Research Council Vitamin Study. Lancet 338: 131-137. Mukherjee, M.D., Sandstead, H.H., Ratnaparkhi, M.V., Johnson, L.D., Milne, D.B. & Stelling, H.P. (1984) Maternal zinc, iron, folic acid and protein nutriture and outcome of human pregnancy. Am. J. Clin. Nutr. 40: 496507. Mulinare, J., Cordero, J.F., Erickson, J.D. & Berry, R. (1988) Periconceptional use of multivitamins and the occurrence of NTDs. J.A.M.A. 260: 31413145 Murphy, M., Boyle, P., Weir, D. & Scott, J.M. (1978) The identification of the products of folate catabolism in the rat. Br. J. Haematol. 38: 211-218. Murphy, M., Keating, M., Boyle, P., Weir, D.G. & Scott, J.M. (1976) The elucidation of the mechanism of folate catabolism in the rat. Biochem. Biophys. Res. Comm.71(4): 1017-1024. Murphy, M. & Scott, J.M. (1979) The turnover catabolism and excretion of folate administered at physiological concentrations in the rat. Biochem. Biophys. Acta. 583: 535-539. National Academy of Sciences (1990) Nutrition during Pregnancy. Institute of Medicine, Food and Nutrition Board. National Academy Press: 365 Washington, DC. National Research Council (1980) Recommended Dietary Allowances. 9 ed. National Academy Press, Washington, DC.

PAGE 158

145 National Research Council (1989) Recommended Dietary Allowances. 10 th ed. National Academy Press, Washington, DC. Naughton, C.A., Reisenauer, A. M., Chandler, C. J. & Halsted, C. H. (1988) Intestinal absorption of folates in alcoholic pigs. Fed. Proc. 2 (6). Abstr. 7655. Nutritional Coordinating Center (1994) University of Minnesota, Food Database Version9a: Nutrient Database Version 2.7. Minneapolis, MN. O'Connor, D.L. (1994) Folate status during pregnancy and lactation. In: Nutrient Regulation during Pregnancy, Lactation, and Infant Growth (L. Allen, J. King & Lonnerdal, B. eds), pp. 157-172 Plenum Press, New York, NY. O'Keefe, C.A., Bailey, L.B., Hofler, S., Davis, B., Gregory, J. & Cerda, J.J. (1995) Folate status response to dietary folate in nonpregnant women. J. Nutr. 125: 2717-2725. O'Scholl, T.O., Hediger, M.L., Schall, J.L, Khoo,C.S. & Fischer, R.L. (1996) Dietary and serum folate: their influence on the outcome of pregnancy. Am. J. Clin. Nutr. 63: 520-525. Perry, J. & Chanarin, L (1973) Formylation of folate as a step in physiological folate absorption. Br. Med. J. 2: 58-59. Pheasant, A.E., Blair, J.A., Allan, R.N. (1979) Folic acid metabolism in man. In: Chemistry and Biology of Pteridines. (Kisliuk & Brown, eds). Elsevier North Holland, Inc., New York, NY. Pheasant, A.E., Connor, M.J. & Blair, J.A. (1981) The Metabolism and physiological disposition of radioactively labeled folate derivatives in the rat. Biochem Med 26: 435-450. Pfeiffer, CM. & Gregory, J.F. (1996) Enzymatic deconjugation of erythrocyte polyglutamyl folates during preparation of folate assay: Investigation using reverse-phase liquid chromatography. Clin. Chem. 42 (1 1): 18471854. Pfeiffer, CM., Rogers, L.M., Bailey, L.B. & Gregory, J.F. (1997) Bioavailability of supplemental folic acid consumed with or without food determined using a dual-label isotope protocol. Submitted for publication.

PAGE 159

146 Qvist, L, Mohammed, A., Jagerstad, M. & Svensson, S. (1986) Iron, zinc and folate status during pregnancy and two months after delivery. Acta Obstet. Gynecol. Scand. 65: 15-22. Ratnam, M.& Freisheim, J.H. (1990). Proteins involved in the transport of folates and antifolates by normal and neoplastic cells. In: Folic Acid in Metabolism in Health and Disease (M.J. Picciano, E.L.R. Stokstad & J.F. Gregory, eds.), pp. 91-120. Wiley-Liss, New York, NY. Rhode, B.M., Cooper, B.A. & Framer, F.A. (1983) Effect of orange juice, folic acid and oral contraceptives on serum folate in women taking a folaterestricted diet. J. Am. Coll. Nutr. 2: 221-230. Rose, R.C., Koch, M.J. & Nahrwold, D.L. (1978) Folic acid transport by mammalian small intestine. Am. J. Physiol. 235: E678-E685. Rothman, D. (1970) Folic acid in pregnancy. Am. J. Obstet. Gynecol. 108: 149175. Saleh, A.M., Pheasant, A.E., Blair, J.A. & Allan, R.N. (1980) The effect of malignant disease on metabolism of pterolyglutamic acid in man. Biochem. Soc. Trans. 8: 566-567. Sauberlich, H.E., Dowdy, R.P. & Skala, J.H. (1974) Folacin (Folic Acid, Pteroylglutamic Acid, Folate). In: Laboratory Tests For The Assessment of Nutritional Status, pp. 49-60 CRC Press, Inc., Cleveland, OH. Sauberlich, H.E., Kretsch, M.J., Skala, J.H., Johnson, H.L. & Taylor, P.C. (1987) Folate requirement and metabolism in nonpregnant women. Am. J. Clin. Nutr. 46: 1016-1028. Sauberlich, H.E. (1995) Folate status of U.S. population groups. In: Folate in Health and Disease (L.B. Bailey ed), pp 171-194. Marcel Dekker, Inc., New York, NY. Sauberlich, H.E. (1990) Evaluation of folate nutrition in population groups. In: Folic Acid Metabolism in Health and Disease. (M.F. Picciano, E.L.R. Stokstad, J.F.Gregory eds), pp. 151-169. Wiley-Liss Inc., New York, NY. Scott, J.M., Kirke, P.N., O'Broin, S. & Weir, D.G. (1991) Folic acid to prevent neural tube defects. Lancet 338: 505. Scott, K.C. & Gregory, J.F. III. (1996) The fate of [ 3 H]folic acid in folate adequate rats. J.Nutr. Biochem. 7: 262-269.

PAGE 160

147 Scott, J.M., Weir, D.G. & Kirke, P.N. (1995) Folate and neural tube defects. In: Folate in Health and Disease (L.B. Bailey ed.) pp. 329-360. Marcel Dekker, Inc., New York, NY. Selhub, J. (1990) Analysis of tissue folate composition. In: Folic Acid Metabolism in Health and Disease (M.F. Picciano, E.L.R. Stokstad, J.F. Gregory eds) pp. 171-194 Wiley-Liss, New York, NY. Selhub, J. (1994) Folate binding proteins — Mechanisms for placental and intestinal uptake. In: Nutrient Regulation during Pregnancy, Lactation, and Infant Growth. (L. Allen, J. King, B. Lonnerdal eds), pp. 157-172. Penum Press, New York, Selhub, J., Ahmad, O. & Rosenberg, I.H. (1980) Preparation and use of affinity with bovine milk-folate protein covalently linked to Sepahorose 4B. Methods Enzymol. 66: 686-690. Selhub, J., Jacques, P.F., Bostom, A.G., et al. (1995) Association between plasma homocysteine concentrations and extracranial carotid stenosis. New Engl. J. Med. 332: 286-291. Selhub, J., Khar, G.J. & Rosenberg, I.H. (1983) Gastrointestinal absorption of folate and antifolates. Pharmacol. Ther. 20: 397-418. Senti, F.R. & Pilch, S.M. (1984). Analysis of folate data from the second National Health and Nutrition Examination Survey (NHANES II). J. Nutr. 1 15(1 1): 1398-1402. Shane, B. (1995) Folate chemistry and metabolism. In: Folate in Health and Disease. (L.B. Bailey ed), pp. 123-151. Marcel Dekker, Inc., New York, NY. Shane, B. & Stokstad, E.L.R. (1985) Vitamin B 12 -folate interrelationships. Ann. Rev. Nutr. 5: 115-141. Shoda, R., Mason, J.B., Selhub, J. & Rosenberg, I.H. (1993) Folate binding in intestinal brush border membranes: evidence for the presence of two binding activities. J. Nutr. Biochem. 1: 257-261. Shojania, M.A. (1984) Folic acid and vitamin B 12 deficiency in pregnancy and in the neonatal period. Clin. Perinatol. 11: 433-359.

PAGE 161

148 Shojania, M. & Gross, S. (1964) Folic acid deficiency and prematurity. J. Pediatr. 64(3): 323-29. Smith, A.M., Picciano, M.F. & Deering, R.H. (1985) Folate intake and blood concentrations of term infants. Am. J. Clin. Nutr. 41: 590-598. Smithells, R.W., Shepard, S. & Schorah, C.J. (1976) Vitamin deficiencies and neural tube defects Arch. Dis. Child. 51: 944-950. Smithells, R.W., Nevin, N.C., Seller, M.J., Sheppard, S., Harris, R., Read, A.F., Fielding, D.W., Walter. S., Schorah, C.J. & Wild, J. (1983) Further experience of vitamin supplementation for the prevention of NTD recurrences. Lancet. 1: 1027-1931. Snell, E.E. & Peterson, W.H. (1940) Growth factors for bacteria X. Additional factors required by certain lactic acid bacteria. J. Bacteriol. 39: 273-285. Steinberg, S.E. (1984) Mechanisms of folate homeostasis. Am. J. Physiol. 246: G3 19-324. Stone, M.L., Luhby, A.L., Feldman, R., Gordon, M. & Cooperman, J.M. (1967) Folic acid metabolism in pregnancy. Am. J. Obst.Gynec. 99(5): 638-646. Strelling, M.K., Blackledge, G.D., Godall, H.B. & Walker, C.H.M. (1966) Megalobastic anemia and whole blood folate levels in premature infants. Lancet 1: 898. Subar, A.F., Block, G. & James, D. (1989) Folate intake and food sources in the US population. Am. J. Clin. Nutr. 50: 508-516. Subcommittee on Dietary Intake and Nutrient Supplements During Pregnancy (1990) Water-soluble vitamins. Folate. In: Nutrition During Pregnancy, pp. 361-367. National Academy of Sciences, Washington, DC. Tamura, T. (1990) Microbiological assay of folates. In: Folic Acid Metabolism in Health and Disease (M.F. Picciano, E.L.R Stokstad, J.F. Gregory eds), pp. 121-137. Wiley-Liss, Inc., New York, NY. Tamura, T. (1995) Nutrient Interaction of folate and zinc. In: Folate in Health and Disease (L.B. Bailey ed), pp. 287-312. Marcel Dekker, Inc., New York, NY. Tamura, T. (1997) Food folate assay with protease, a-amylase and folate conjugase Treatments. J. Agric. Food Chem. 45, 135-139.

PAGE 162

149 Tamura, T., Goldenberg, R.L., Freeberg, L.E., Cliver, S.P., Cutter, G.R. & Hoffman, H.J. (1992) Maternal serum folate and zinc concentrations and their relationships to pregnancy outcome. Am. J. Clin. Nutr. 56: 365-370. Tamura, T., Shane, B., Baer, M. T., King, J.C., Margen, S. & Stokstad, E. L. (1978) Absorption of monoand polyglutamyl folates in zinc-depleted man. Am. J. Clin. Nutr. 31(11): 1984-1987. Toth, J.P. & Gregory, J.F. IB.. (1988) Analysis of folacin by GC/MS: Structures and mass spectra of fluorinated derivatives of para-aminobenzoyl glutamic acid. Biomed.Environ. Mass. Spec. 17: 73-79. van der Put, N.M.J, Steegers-Theunissen, R.P.M., Frosst, P., Trijbels, F.J.M., Eskes, R.K.A.B., van den Heuvel, L.P., Mariman, E.C.M., den Heyer, M., Rozen, R. & Blom, H.J. (1995) Mutated methylenetetrahydrofolate reductase as a risk factor for spina bifida. Lancet 346: 1070-1071. Van Langenhove, A. (1968) Isotope effect: Definitions and consequences for pharmacologic studies. In: Browne TR, Baillie TA (eds): Proceedings of the Firth Symposioum on Frontiers of Pharmacology. J. Clin. Pharmacol. 26: 474-480. Vergel, R.G, Sanchez, L.R., Heredero, B.L., Rodriguez, P.L. & Martinez, A.J. (1990) Primary prevention of NTDs with folic acid supplementation: Cuban experience. Prenat. Diagn. 10: 149-152. Wagner, C. (1995) Biochemical role of folate in cellular metabolism. In: Folate in Health and Disease (L.B. Bailey ed), pp. 23-42. Marcel Dekker, Inc., New York, NY. Wei, M.M., Bailey, L.B., Toth, J.P & Gregory, J.F. m. (1996) Bioavailability for humans of deuterium-labeled monoglutamyl and polyglutamyl folates is affected by selected foods. J. Nutr. 126(12): 3100-3108. Weitman, S.D., Weinberg, A.G., Coney, L.R., Zarawske, V.R., Jennings, D.S. & Kamen, B.A. (1992) Cellular localization of the folate receptor: potential role in drug toxicity and folate homeostasis. Cancer Res. 52: 6708-671 1. Werler, M.M., Louik, C, Shapiro, S. & Mitchell, A,A. (1996) Prepregnant weight in relation to risk of neural tube defects. J.A.M.A. 275(14): 1089-1092. Werler, M.M., Shapiro, S. & Mitchell, A.A. (1993) Periconceptional folic acid exposure and risk of occurrence neural tube defects. J.A.M.A. 269: 12571261.

PAGE 163

150 Williams, W.M., Huang, K.C. (1982) Renal tubular transport of folic acid and methotrexate. Am. J. Physiol. (Renal fluid Electrolyte Physiol) 11: F484F490. Willoughby, M.L.N. (1967) An investigation of folic requirements in pregnancy, n. Brit. J. Haemat. 13: 503-509. Willoughby, M.L.N. & Jewell, F.G. (1968) Folate status throughout pregnancy and in postpartum period. Brit. Med. J. 4: 356-360. Wills, L. (1931) Treatment of "pernicious anemia of pregnancy" and "tropical anemia" with special reference to yeast extract as a curative agent. Br. Med. J. 1: 1059-1064. Wills, L. & Bilimoria, H.S. (1932) Studies in pernicious anemia of pregnancy. Part V. Production of a macrocytic anemia in monkeys be deficient feeding. Indian J. Med. Res. 20: 391-404. Wilson, S.D. & Home, D.W. (1984) High-performance liquid chromatographic determination of the distribution of naturally occurring folic acid derivatives in rat liver. Anal.Biochem. 142: 529-535.

PAGE 164

BIOGRAPHICAL SKETCH Marie Anne Caudill was born on November 3 r 1967, in Sarnia, Ontario, Canada. At 1 1 years of age, her goal was to obtain a tennis scholarship at a university located in the southern United States. In 1987, she accepted a tennis scholarship at Lamar University in Beaumont, Texas. Two years later, she transferred to the University of North Florida where she received a Bachelor of Art in Physical Education, a Bachelor of Science in Health Science and a Master of Science in Health Science and also met her husband, Tony. During her fifth year on athletic scholarship, she competed in the NAIA women's marathon for the University of North Florida and became national champion. In May 1994, she completed a dietetic internship at Emory University Hospital in Atlanta, Georgia. Three months later, she entered the doctoral program in the Food Science and Human Nutrition Department at the University of Florida, Gainesville. She graduated with her doctoral degree in August, 1997 and has accepted a faculty position in nutrition at California State Polytechnic University located in Pomona. 151

PAGE 165

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. i Bailey, Chairperson >r 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. ;ggvR. 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. /esse F. Gregory, in 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. Gail P. Kauwell Associate 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. Sharleen H. Simpson ( / Sharleen Associate Professor of Nursing

PAGE 166

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. August, 1997 Dean, College of Agriculture Dean, Graduate School