Isolation and metabolism of radioactive retinol, retinal and retinoic acid by the rat in vivo

MISSING IMAGE

Material Information

Title:
Isolation and metabolism of radioactive retinol, retinal and retinoic acid by the rat in vivo
Physical Description:
viii, 95 leaves : ill. ; 29 cm.
Language:
English
Creator:
Dunagin, Percy Elford, 1934-
Publication Date:

Subjects

Subjects / Keywords:
Retinoids   ( mesh )
Vitamin A   ( mesh )
Rats   ( mesh )
Biochemistry and Molecular Biology Thesis Ph.D   ( mesh )
Dissertations, Academic -- biochemistry and molecular biology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1964.
Bibliography:
Includes bibliographical references (leaves 92-95).
Statement of Responsibility:
by Percy Elford Dunagin, Jr.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000895089
notis - AEK3657
oclc - 25309003
System ID:
AA00009500:00001


This item is only available as the following downloads:


Full Text












ISOLATION AND METABOLISM
OF RADIOACTIVE RETINOL, RETINAL AND
RETINOIC ACID BY THE RAT IN VIVO












By
PERCY ELFORD DUNAGIN, JR.


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











UNIVERSITY OF FLORIDA


December, 1964














ACKNOWLEDGEMENT


Sincere appreciation is expressed to Or. James Allen Olson for

his capable supervision of the author's graduate program and his

helpful guidance in the area of scientific communication.

Acknowledgment is given to the other members of the author's

supervisory committee for their suggestions during the preparation of

the manuscript nd to the members of the Departmnts of Biochemistry

and Physiology for their contributions to the author's graduate

education.

Betty Dunagin deserves special recognition for her help and

encouragement, and especially for her tireless efforts In training

their two children during the course of the author's graduate study.














TABLE OF CONTENTS


Page


ACKNOWLEDGEMENT. . . ,

LIST OF TABLES . .

LIST OF FIGURES. . *. .

INTRODUCTION *. . .

MATERIALS AND METHODS. . . .

Materials . . .


Radioactive retinol derivatives .
Non-radioactive retinol derivatives
Solvents fr column chromatography.


Methods a. .. .. .

Metabolic studies on bil cannulated rats .
Ion exchage chromatography of tissues. .* .
Silicic acid chromatography of fraction II of bile ..
Synthesis of retinol derivatives. . .
Gas liquid chromatography (GLC) . .
Thin layer chromtography (TLC) . .
Ultraviolet spectrophotometry .. .
Methods for measurement of radioactivity. .
Hydrolysis of tissue extracts. .

EXPERIMENTAL RESULTS . . .

Analysis of Six Retinol Derivatives by Gas Chromatography .

Gas chromtography of Individual derivatives on
short columCs. . *
Identification of chromatographcally separated
compounds by ultraviolet spectra *. .
The relationship between the mount Injected and the
detector response. .* .* .
Electron capture detection of retinal and methyl
retinoate. ,* . .*
Effect of vearous factors on the stability of retinal
end methyl retnoate during gas chromatography .








Control Studies on Non-radioactive and Radioactive
Retinolc Acid. ..... .. ..... .. .. 29

Stability of retinoic acid to various conditions of
analysis . 29
Irradiation and Isomerization of retinoic acid. 30
Efficiency of methylation with diazoamthane and
recovery from gas chromatography .. 34
Analysis of purified radioactive retinoic aecid 35

Distribution of Radioactivity in Various Tissues After the
Administration of Labeled Substrates .. 39

Characterization of Free Retinoic Acid in Fraction II of
Bil*, Intestine, nd Liver After the Intravenous
Administration of Retineoc Acid. . 4)3

Characterization of Netaboltes of Retlnoate In Bile After
the Administration of C14-Retinoic Acid. .... 4.

Experimental procedure. ..... ..... 46
The rate of disappearance of labeled retinoef acid
from various tissues and its excretion in the bile 50
Fraction I of bile. . 50
Fraction of bile . . 57
Fraction Itt of bile. .......... 60
Investigation of bile by thin layer chromtography
(TLC). .. . . 64
Purification and characterization of fraction III of
bile by chromatography on siltic acid .. 60

Characterization of Retinal Metabolites . 72

Fraction I of bile. . .......... 72
Fraction II of bile .. . 7
Fraction II of bile. .... ...... 76
DISCUSSION . . . 80

SUMMARY. ....... . 90

REFERENCES . . 92













LIST OF TBLES


Table Page

Comparison of Isomers of methyl retinoats .3
II. Radioactivity recovered from various tissues after
administration of 6,7-C41-retinol derivatives. ... 40
III. Specific activities of retnoic acid from various tissues
4.5 hrs. after the injection of 6,7C14-rettnolc acid.. 47
IV. Specific activities of 6,7C14-retinolc acid found tn
fractions of bile. . . 56
V. Thin layer chromatogray of rat bile 10 hrs. after
administration of C-retnoc acid. . 65

VI. Specific activity of retfnolc acid found in fraction IX
of bile 8 hrs. after injection of 6,7-C4-retinal. 74

VII. Specific activity of retinolc acid found In fraction III
of bile 8 hrs. after injection of 6,7-C1 -retinal. 77














LIST OF FIGURES


Figure Page

1. Apparatus for automatic gradient elution
chromatography . . 12

2. Gas chromatography of Individual retinol derivatives
on Gas Chrom P .. 21

3. Gas chromatography of retinol and retlnyl acetate 22

4. Spectra of retinMl derivatives before and after gas
chromatography .. ..* .. 2)

5. Calibration curves for four retinol derivatives
separated by gas chromatography ... ... 25

6. Comparison of the recorder response with the amunt
of retinal recovered by gas chromatography 26

7. Comparison of the responses of a flane ionization
detector and an electron capture detector to a
mixture of retinal and methyl retinoate. 28

8. Gas chromatography of the methyl esters of an fsomerate
of retinoic acid . . 32

9. Gas chromatography of radioactive retinoic acid
purif ed by ton exchange chromatography 36
10. Gas chromatography of radioactive retlnoic acid
purified by silicic acid chromatography. 38
11. The cumulative excretion rate of radioactivity into
th bile after the administration of 1 to 3 mg of
C14-retnol. Clk-retinal, or CI-retinoic acid 41
12. Separation of bite by Ion exchange chromatography
after administration of Cl4-retinol, C14-retir?'
or CIA-rettnotc acid . 42

13. Separation of a liver extract by ion exchange
chromatography after adminiptration of C 4-
retinol, C -retinal, or C1 -retlnoic acid 44

vt









14. Separation of an Intestinal extract by Ion exchange
chromatogrphy after administration of Cl4-
retlnol, C -retinal, or Cl*-retlanol acid 45

15. Ultraviolet spectra of fraction I of bile, liver,
and intestine 4.5 hrs. after the administration
of retinolc acid . . 48

16. Flow sheet fpr the treatment of bile after injection
of 6,7-Cl4-retinoic acid . 49

17. Effect of time on the recovery of Injected radioactive
retinoic acid n various tissues 51
18. Spectra of fraction I of bile .. . 52

19. Gas chromatography of fraction IX obtained after the
hydrolysis of fraction I of bile collected from
rats injected with Cl'-retinoec acid . 5$

20. Ultraviolet spectra of Icers of retinoate collected
after gas chromatography of the methylated
hydrolysate of fraction I. .. . 5
21. The ultraviolet spectrum of fraction 11 of bile before
and after gas chromatographic analysis 58
22. Gas chromatography of fraction 11 from bile eater the
administration of CI-retinoic acid. 59

2). Ultraviolet spectrum of fraction IXX of bile after
administration of Cl-retinoic acid. 61

24. Gas chromtography of mathylated fraction I1 obtained
by ion exchange chromtography of a basic
hydrolysate of fraction II of bile from a rat
treated with Cli-retinoic acid .. 62
25. Ultraviolet spectrum of retlnoate isomers collected
free gas chromatography of methylated fraction XII
hydrolysate. . . 63
26. Thin layer chromatographic comparison of whole bile
and fractions of bile obtained by ion exchange
chromatography 66

27. Silicic acid chromatography of fraction III of bile
after the administration of cli-retinoic acid. .. 69
28. Thin layer chromatography of components isolated by
stiltcc acid chromatography of fraction III
from bile. . . 70


vil









29. Behavior of derivatives of purified radioactive
fraction III which were parated n thin layer
plates of Silica Get1 by development with
beneneschloroformcmathalne (4I1tsl). .

30. Gas chromatography of fraction II obtained by Ion
exchange chromtogrgphy of bite after the
administration of C"-retnal. . .

31. Gas chromatography of fraction XI obtained by Ion
exchange chromtography of the basic hydrolysate
of fraction IXX after administration of ClK-retinal.


vi l













INTRODUCTION


Substances possessing vitamin A activity were distinguished from

other "accessory factors" in 1922 after a series of careful investi-

gations by McCollum, Davis, Osborne, and Mendel (2-5). Since the early
work with butter fat and cod liver oil, substances with vitamin A

activity have been found in many plants (carotenolds) and antal tissues
(mostly retinol esters) (6). The Ingestion of diets containing

inadequate amounts of these substances produces characteristic
deficiency symptoms (7-11) which include (a) defective dark adaptation

and ultimately complete loss of the visual process, (b) keratinizatlon
of epithelial tissues, (c) nerve degeneration, (d) cessation of growth,

and (e) fa lure of the reproductive processes.

All symptoms of vitamin A deficiency can be prevented or relieved
by adequate intake of carotene, retlnot, retinal, or retlnol ester (12-13).

Retinoic acid is equally as effective as retinol in growth
promotion (14-16) and In preventing corniflcation of vaginal epithelial

tissue (17), but is inactive in promoting normal reproduction (18), or
in preventing visual malfunctions, where retinal has a specific

metabolic function (19, 20). Since retinoic acid apparently is not

In accordance with the definitive rules for the nomenclature of
vitamins which were approved by the Commission on the Nomenclature of
Biological Chemistry of the International Union of Pure and Applied
Chemistry (1), the terms retnol, retinal, and retinoc acid are used
nt place of vitamin A alcohol, vitamin A aldehyde (retnene), and
vitamin A acid, respectively. The term vitamin A is used to refer to
the general properties of retinolt its derivatives, and its provitamins.
1







2

reduced to retlnol or retinal In vivo either the free acid or one of

its metabolites must be biologically active In growth promotion and In
the maintenance of epithelial tissue.

Metabolic interrelationships anng p-carotene, retlnol, retinal,
and retinoic acid have been reviewed by several Investigators (21-23).

Dietary P-carotene is converted to retinol ester in the intestine.
Ingested preformed retlnol, which exists in foodstuffs largely in the

esterified form, is apparently hydrolyzed in the intestinal uwns and

re-esterifled in the intestinal wall. After absorption, esterified

retinal is transported via the blood stream to the liver where it is

stored largely as the ester (23). The vitamin A content of fasting
normal blood remains nearly constant and consists almost entirely of free

retinol, bound most probably to a high density al-globulin (24). On the

other hand, post absorptive blood contains an increased mount of

vitamin A in the ester form, which is probably bound to a low density

lipoprotein (25). Liver retinol ester apparently Is hydrolyzed by a
liver esterase (21) and released into the bloodstream as retinol when

needed by the tissues.
Retinal, although not a cooon dietary component, is absorbed

from the intestinal lumen, reduced d a esterifed in the intestinal
mucoss, and transported to the liver where it is stored as retinol

ester (26). Enzymes which carry out the reversible oxidation of retinol

to retinal are found In several tissues (26-29). Since the equilibrium
of this reaction strongly favors reduction of retinal to retinol (30),

retinal appears in low concentrations in mammal11a tissues indeed
outside the retina it has been found only in the liver (31).







3
Another factor reducing the amount of retinal in tissues is the
presence of enzymes in liver, kidney, and the mall intestine (31-36),
which catalyze the oxidation of retinal to retinelc acid. This
oxidative reaction is essentially irreversible (33). Retinol ester is
not recovered IM yjvi even after the administration of large doses of
retinotc acid (37-39).
Although enzymes which oxidize retinal and retinal to retinoic
acid exist in tissues, retinoic acid has never been found It any tissue
when animals were fed with normal leads of p-carotee or retinal ester.
On the basis of limited spectral evidence, Koizumi at I. .laimed that

-carotene was converted to retd to ti ad tin tissue homogenates (40).
Ofmitrovski also reported that retinolc acid was formed by the rat
i viv after the administration of 10 to 60 ag of retinal and L. vitro
by rat intestinal preparations (35). In this case as well, the charac-
terization of retinoic acid was limited to spectral analysis of crude
fractions obtained by the treatment of intestinal extracts by alumina
chromatography.

Even after massive dosing with retinoec acid, free retinotc acid
has not been readily detected in tissues (39, 41-43). Recently,
Jurbowitz, using an improved acidic extraction procedure, found retinoic
acid in human plasma up to 6 hrs. after the oral administration
of 100-200 ag of retinolc acid (44). After oral administration of 3 mg
of retlnoic acid to chicks, Krishnamurthy et al. (37) found retlnole
acid in the liver 6 hrs. later, but not 18 hrs. later. After daily
administration of $0-100 p&g of retlnotc acid to vittain A deficient rats
over a three week period, no retinal, retinol ester, or retinolc acid

could be detected in any tissue (37). The difficulty in finding retinoic






4

acid as a product of administered precursors (31, 37, 45) has led
to the speculation that free retinoic acid is not an obligatory
Intermdiate in the formation of a biologically active metabolite from
retinol, but rather my be converted to that metabolite by another
route (18, 22, 37).
A number of products of retinal mtabolis In jvtv have been

suggested as possible links to the biologically active form of vitamin A.
Wolf ("6, 47) found that 5% of the radioactivity of 14-CL-retinol
injected intrapertoneally into rats appeared n expired CO2 within
24 hrs. Appreciable amounts of radioactivity also appeared in the
urine, faces, carcass, lver, Intestine, blood, kidney, skin, and eyes.
To a large extent, the radioactivity was water soluble. Further
examination of radioactive components of the urine revealed two compounds
separable by paper chromatography. One was water and ether soluble
(WES), while the other was water soluble but not soluble in ether (WS).
On the basis of spot tests for functional groups, it was concluded that
WES contained double bonds, hydroxyl groupss, and an aldehyde group.
WS was crystallized as a dinitrophenylhydrazone derivative, seemed to
be a hydroxylated ester with a non-conjugated keto group, and had the
elementary composition, C 1lH 10.
Garbers found radioactivity in the urine as well as In the a,-

globulin fraction of serum after the administration of radioactive
retinol (24). In contrast, retinol was associated with the l-globulin
fraction of serum. Varandani (48) also reported water soluble
metabolites in the urine, but not in the liver, after the administration
of radioactive retinol. After the administration of radioactive
retinoic acid to vitamin A deficient chicks, Krishaurthy (37) found






5
radioactivity in a water soluble fraction after ether and acid-ethanol

extractions. This water soluble material was dialyzable but was not
biologically active.
After the intraperitoneal injection of ) mg doses of radioactive

retinoic acid, Yagishlta e dl. (41) found radioactive metabolites in
the liver and Intestine of rats. After Intestinal and liver extracts

were hydrolyzed and separated by thin layer chroemtography, ts radio-
active compounds were detected, both of which were reported to be free

of retinoic acid. The first was biologically active, was mre polar
than retinoic acid, contained hydroxyl groups as indicated by Infrared

analysis, and had an absorption nmmimum at 252 mp, suggesting the
presence of three conjugated double bonds. The second component was
slightly leu polar than retinolc acid on thin layer chromatography and
was biologically Inactive.
Radioactive retinolc acid adoinlstered to rats by Zile and

de Luce (43) was not recovered from tissues as free retinotc acid, but
as a group of mtabolites separable by silicic acid chromatography.
One metabolite was biologically active, non-acidic, less polar than
retinolc acid, and absorbed at 272 op. Apparently this radioactive
mtabolite, which was isolated from lver, is not identical with that
reported by Yagishita.
Ninety minutes after perfusion of an isolated rat liver with
radioactive retinol, Zachman found that 75% of the radioactivity was
recovered in a water soluble, hexane Insoluble fraction (49). Tie
percent of the radioactivity in the perfusate was secreted via the bile

duct.2 When he looked more closely at bile from bile duct cannulated

2R Zachmna, unpublished observation








rats, he found 20 to 40% of the injected radioactivity in the bile

12 hrs. after the injection of radioactive retinol or retmolic acid via

the portal vein (50). The radioactivity in the bile was separated into

feur fractions by step-wise anion exchange chromatography. The first
fraction, which was eluted with methanol and was therefore non-ionic,

contained about 10% of the recoverable radioactivity, but not retinal,

retinol, or retinal ester. The second fraction, which was eluted
with 1% acetic acid in mthanol, had 10-20% of the recovered radio

activity. Although free retinoic acid would be eluted in this fraction,
it was not looked for. The remaining fractions, accounting for 70-80%

of the recovered radioactivity, were not further examined, Since bile,
which has previously been ignored as a source of retinol metabolites,

contained such a large portion of the injected radioactivity, the

further characterization of these fractions should yield useful
information about the metabolism of retinol.

As indicated by the proceeding discussion, the metabelites of
retinal reported by various investigators apparently neither fit into a

unified scheme nor seem be related to rlad one another. Although sem of

these metabolites had biological activity, none were well characterized.

Unfortunately, the methods employed in these studies were rather crude.

Metabolites were separated and characterized on the basis of their
partition between various solvents or their extractability with organic

solvents under given conditions, and by reactivity in non-specific
tests such as the Carr-Price reaction. In the few cases where colaum

chroatography was used, all components were not clearly separated.
Thin layer chromatography, which is a very rapid and efficient method,
cannot be expected to separate retinol compounds in pure form from






7

complex lipid mixtures containing hundreds of compounds. The mass of
many of these components greatly exceeds that of the investigated
compound. Moreover, retinol derivatives separated by thin layer
chromtography are subject to destruction and isomerlzatlon (51).
Gas chromtography, which since its inception In 1952 (52) has

been a valuable tool In analysis, has not heretofore been used
successfully for the separation of retinol, its derivatives, and ts
metabolites. In the only reported application of gas chromatography to
retnoal separation, retinal, retinyl acetate, and retinyl pallmtate
were dehydrated rapidly to anhydro retinol (53).
As a result of the inadequate isolation and characterization of

retinol metabolites, little is known about the metabolic patihay of

retinol beyond the reversible oxidation of retinol to retinal. Although
retinoic acid or one of its metabolites must be the active form of
vitamin A in growth and epithelial tissue maintenance, free retinoic
acid has not been clearly demonstrated as a product of retinol or
retinal metabolism In yin. Hopefully, the determination of the
metabolic pathways involving retinal, retinal, retinoc acid, and
further oxidized metabolites will ultimately elucidate the functions of
vitamin A in metabolism.
Specifically, the present investigation concerns (a) the

recovery and unambiguous identification of free raetnoe acid in the
white rat after moderate doses of retlnol or retinoic acid were

administered, and (b) the characterization of metabolites fond in
liver, intestine, and bite after the administration of moderate doses of
retlnol, retinal, and retinolc acid.








The Investigation of these problem has depended to a large
degree upon the development of the following techniques for the separation
and isolation of retinol and (ts derivativess (a) gas liquid chrom-

tography, (b) anion exchange chromtography with a grdient elution

procedure, and (c) adaptation of existing techniques In thin layer
chromatography and silicic acid chromatography.














MATERIALS AND METHODS


Materials


Radioactive retinol derivatives

C -retinol and C -retinoic acid were generously supplied by

Hoffan-LaRoche, Inc., Basel, Slitzerland. The alcohol was further

purified by chromatography on alumina and the acid was further purified

by either Ion exchange chromatography or by silcicc acid chromtography.

CG-retinal was prepared by Mn02 oxidation (54) of C -retinol and was

isolated by alumina chromatography.

Non-radioactive retinol derivatives

Non-radioactive crystalline all-~gan ratinol, all-.tras retinyl

acetate, all-trans retinal, and all.tn rretinoic acid were purchased

from Distillation Products Industries, Rochester, Ne York. The

crystalline 13-S Isomer of retinoic acid was kindly supplied by

Hoffan-LaRoche, Inc., Basel, Switzerland. The physical properties

(melting point and ultraviolet absorption spectra) of these compounds

agreed very closely with reported values (51, 55). In addition, 94-100X

of the methyl ester of each compound was eluted as a single peak from
gas chromatographic columns. The compound collected from the gas

chromatographic eluate possessed the spectrum of the pure injected
material. Therefore, these substances were used without further

purification.








Solvents ffr colun chromtography
Reagent grade methanol and acetic acid were used without further

purification for ion exchange chromatography. For silicic acid chrome-

tography reagent grade absolute ethanol was used as purchased. Spectral
grade hexane was prepared from Skellysolve B (Skelly Oil Company,
Elddorado, Texas) by treatment with fuming sulfuric acid (30% S0 ).
Approximately 1500 I of Skellysolve I were shaken for one hour with
250 ml of fuming sulfuric acid. The organic phase was then neutralized
with 10% KOH, dried with Na2OR4, and distilled over KOH pellets.
Approximately 1000 ol were collected. The purified solvent had a boiling

range of 66-69 C and an absorbance of 0.05 or less at 230 up when
measured in a cuvette with a 1 ca light path against a water blank.

Methods


Metabolic studies on bile cannulated rat
Fasted male rats (Rolfamayer Farm, Madison, Wisconsin) weighing

from 150 to 300 g were anesthetized and their bile ducts were

cannulated.3 Then a solution of 1 ml of the appropriate substrate
suspended in Twe.n 80 was injected into the portal vein. After closing
the incision, the animal was placed in a restraining cage and the bile

was collected at desired intervals (50). At predetermined times, the

animals were killed, various organs were honogenlzed In CHCtl mathanol
(2sl). and the homogenates were filtered. The filtrates were
evaporated to dryness and taken up In methanol. Bile and urine samples
were diluted with 4 vols. of methanol. Diluted bile and urine and

3The author wishes to gratefully acknowledge the preparation of
bile duct cannulated rats by Richard 0. Zachman.






11

methanolic extracts of organs were separated by Ion exchange chrometo-

graphy or used for other analyses.

As. excite chrommtograph of tissues
Analytical and semi-preparative columns (1.9 a i.d. x 6.5 cm)

contained 200-400 mesh Slo-Red AG2-X8 anion exchange resin In the acetate
form (Bio-Rad Laboratories, Richmnd, California). Preparative

columns (5.0 ca .d. x 7.5-10 cm) contained either the above resin or
Ito-Red MAI-X4. Components of bile were separated well by both resins,
indicating that a8 cross-linking was not excessive. Samples were
eluted by gradient elution with increasing concentrations of acetic acid
-Vh
tn methanol according to the equations C 1 *- e (56) where C
is the concentration of the contributing solution found to the mixing
chamber at elution volume v, and vo is the voelum of the mixing chamber.
The elutlon system consisted of three chambers (Fig. l)s the wash
chamber the distributing chamber, and the mixing chamber. A solution

of 50% acetic acid in methanot (100% acetic acid for the preparative
colent) from the contributing chamber flowed into a closed 500 ml
mixing chamber which emptied Into the chrometographic column through one
side of a three-way Teflon valve. The other side of the three-way
valve was connected to the wash chamber, which contained methanol for
washing the column and for the Initial elution of sample. The effluent
from the column was collected nt 10 ml fractions by a Technicon dlop
counting fraction collector (Technicon Chromatography, Corp., Chauncey,

NMo York). All parts of the system which came in contact with acetic
acid were constructed of Teflon, glass, or stainless steel.
After equilibrating the resin column with methanol, I to 25 ml of
a bile solution or organ extract from one rat was added to the column.

























Wash -
chamber

Contributing


Fig. 1. Appwrtu for automatic gradient elution chromtogrphy.
The solution from the contributing chamber passes nto the mdxing
chamber, rmxes with its contents and passes onto the colm via the
three-way valve. The colum effluet is measured by the drop-couter
and is collcted I the fraction collector.


3-way valve


chamber


chamber


Column









- Drop
counter


- Mixing







13
When the sample had completely entered the column, the column surfaces

were washed twice with 2 to 5 al of methanol. Thereafter 10 om of
mthanol were added and the column was connected to the methanol

reservoir. After 50 of methanol, which includes the sample and wash
solutions, had flowed through the column into the fraction collector,
the gradient was begun by switching the three-way valve to the mining
chamber. Eighty fractions were collected from the analytical colums

and 150 from the preparative column. The column was regenerated by
washing first with a 10% solution of sodium acetate until the *uaet was
basic then it was equilibrated with methanol. Each tube was usually
mentored for 350 Up absorbing material with a Coleman Universal
Spectrophotometer and was further analyzed for radioactivity. Radioactive
fractions were frequently analyzed further by spectrophotometry or by
other techniques as described in later sections.

Slicic ad chrout m gera of. fraction III .ft.4

Silicic acid chromatography was performed with the sme apparatus
as that used for ion exchange chromatography. The sample was separated
on <300 mesh silicic acid which had been treated by the method of Hirsch
and Ahrens (57) and was suitable for Immediate use (Blo-Rd Laboratories,

Richmond, California). The mixing chamber contained 500 oll of heane,
and the contributing chamber contained 25% hexane in absolute ethanol.
Prior to staple separation, 20 g of silicic acid was poured Into the
column (1.9 em id.) which was filled with hmxane. After settling of
the silicic acid, the sample was applied as described for ion iechange
chromatography. A volume of 15 to 20 ml of heasne uas run through the

4The author wishes to thank E. Harris Meadows, Jr. for his
excellent assistance in the chromatographic analyses performed on
silicic acid colums and thin layer plates.







t4
column before beginning the gradient. Fractions of 5 al were collected
in the automatic fraction collector.

Synthess of retnol derivative
Anhydro retinol was prepared by the dehydration of retinol with
ethanolfc MCI (58, 59), The reaction was stopped by neutralization with
base, the mixture was extracted with benzene, md the anhydro retinal
extract was finally purified by chromatography on activated alumina.
Anhydro retinol was eluted with 0.5% acetone In hezane. The ultraviolet

spectrum of the product (maxim are 346, 365, and 388 Ip) agreed with
that reported by Shntz, Cwley, and Norris (58).
Methyl retlnyl ether was prepared frea retinol by a modified
Wlliamson's synthesis (60). Retnol was shaken with n-butyl lithium
in dry benzene for 5 minutes. Then dimthyl sulfate tn dry benzene was
added. The products were washed with acid and then base, dried over
Ne2SO4, and separated en activated alumna. Retinol ether, which was
eluted with 1% acetone In hexane, was well separated from retinal and
was used without further purification.

Methyl retinoate was prepared by treating retinoic acid with
diomthane th her. The disuomethane was released from 'tiazald"

(Aldrich Chemical Co, Inc., Milwaukee, Wisconsin) by treatment with
iNaH and was distilled into ether (61). Diazemethane prepared in this

manner was preserved for 2 or 3 days in the cold. Diazomothane was
generated and used to a hood because of Its poisonous and explosive
nature. After evaporation of ethyl ether n vamue, methyl retlnoate
was crystallized twice from mthanolth20 (Sl). Its malting point and
infrared and ultraviolet spectra agreed with those reported In the
literature (62, 63).









jg liSqud roItograp (GLC

A Model 600 Research Specialties Co. gas chromatograph was used
for all gas chromatography. Solutions containing pure anhydro retfnol,

methyl retinyl ether, retinal, methyl retinoate, retinol, or retinyl
acetate were used directly.

Some tissue detracts were analyzed without treatment, but most
were mthylated by addition of ethereal dizomethane until the evolution

of nitrogen gas ceased. The resulting mixture was evaporated to dryness
under Nt, and was dissolved In an appropriate amount of hexane,
isopropyl ether, or other volatile solvent. Hydroxylated compounds

were converted to the ore volatile trimethyl silyl ethers by treatment
with hexamethyldstlazane and trimethylchlorosllane in pyridine (64).

For solutions of pure compounds a capllary injection system

gave reproducible results with 0.5 to 2.0 lil of solutions, but not with

larger volumes. For tissue extracts an on-column injection system was

used. Samples were delivered from a needle tipped Halitton micro-
syringe through a rubber septum into the carrier gas stream. With this

procedure larger more accurately measured samples (5-50 pS) could be
analyzed with better reproducibility than with the capillary injection

system.

Volatilized components of the sample were separated on 2 to 4.5 ft.
colums containing a sllanized diatomeceous earth support coated with I

to 3% SE-30 (a non-polar silicone rubber gum made by General Electric
Corp.) unless otherwise stated. Colums were treated with p-carotene

to prevent the dehydration of retinal and retinyl acetate. From 50-300 pl
of p-carotene in benzene were added to the column by either Injection

system. After raising the column temperature to 250O C for 2-3 hours









and then cooling to normal operating temperature, unstable retinol

derivatives could be chromatographed with little destruction.
Components of the column effluent were detected in the gas phase

by an argon Ionizatlon detector or, alternatively, by a flae ionization
detector In parallel with an electron capture detector.
Individual components from the effluent stream were collected

when desired by one of four methods. (a) The conventional collecting
device was a glass U tube, shortened and tapered at the distal end,

which contained a 4 I layer of A120) over a glass wool plug. Components
were adsorbed onto the alumina and eluted with a minimum of 0.5 al of
absolute ethanol. When the collection device was properly made,
essentially all the liquid and solid components of the effluent stream
were adsorbed on the alumina. The alumna was frequently changed to

prevent contamination from previously collected samples. (b) A rapid,
inexpensive, and disposable collection system with which the collected

sale could be eluted with ess than 0.1 tl of most solvents was made
from a 1.5 m o.d. x 300 am glass capillary tube. The capillary was
connected to the 1/16 In. exit tubing from the detector by meMs of
a #15 AR Teflon tube, and could be rapidly inserted and removed.
(c) Unstable compounds such as methyl retinyl ether, rettnol, and

retinyl acetate were collected on a cotton pad placed nt the bottom of
a hypodermc needle which was Inverted over a short Teflon tube
attached directly to the exit end of the column. Although collection
was inefficient, unstable compounds which were destroyed in the
long (75 cm) emit system could be isolated Intact by this procedure.
Retinol, retinyl acetate, and, to som extent, methyl retlnyl ether were
dehydrated to anhydro retinol when collected In the conventional manner.







17
(d) An automatic gas fraction collector (Packard Instrument Co.) was

used to monitor the effluent stream for radioactivity. Sample

components were condensed in anthracene filled cartridges and the entire
cartridge was placed Into a vial for counting in a liquid scintillation

spectrometer. The efficiency of collection by this method was 75-85%,
approximately the sae as by adsorption on A1203 or by the capillary

collection method.

Thin layer chromtgrahy (TLC)
The apparatus was manufactured by Research Specialties Co.,
Richmend, California, and standard techniques wer used. Sales were
applied on TLC plates coated with Silica Gel 6 by means of 10 and 100 pl

Hamnlten syringes. After sapte application, the TLC plates were
developed with (a) the benzene, chloroform, methaol (4sll) system of

Wolf (41), (b) H20, or (c) benzene, chloroform, methanol, acetic

acid (55$t51). Separated coqmonents were detected by fluorescence

under ultraviolet light, by Iodine vapor, or by charring in the

presence of sulfuric acid. Detection by I2 vapors proved to be the mst
sensitive method. Charring nt the presence of H2504 did not reveal

cre spots then did I vapor. A permanent record of the TLC chreato-
gram was conveniently made by photography with a polaroid 110 B Land

Camera. Utilization of a blue filter (Corning No. 5-61 Signal Blue)

between the plate and the lens gve a photograph with high contrast

between the yellow 12 spots and the white background of the plate. to

may cases, spots which were invisible to the naked eye were revealed
by this type of photography. The adsorbent was scraped off tn suitable

increments, suspended in scintillation fluid containing 4% Cab-0-St (65),
and counted for radioactivity in a Trlcarb liquid scintillation spectro-

photomt r.


- -- M









Ulravieolet pectrophotemtry
Most ultraviolet spectra were obtained on the Perkin-Eflr
iHdel 1000 ultraviolet recording spectrophotomter. For cempounds

obtainable in sufficient quantity, standard ) ml cells with I ca light

paths ware employed. However, the smll quantities usually obtained

from the effluent of the gas chromstograph necessitated the use of
aicrocells containing 0.1 to 1 ml of solution. To adapt the recording

spectrophotometer for this purpose, a blackened shield was placed into
the light path of both the reference and sple beans. An orifice

eaatly I m in diameter was drilled into each shield and the shields
were adjusted to permit the light bem to pass through the center of

the mferoealls. The intensity of the hydrogen lp source was

sufficient for the required analyses without bea condensation. The
spectra of very dilute solutions obtained from ion exchange chromt-

graphy were measured in cells having 2 or 5 an light paths. 1We
single beam spectrophotomters, the eckma DU and the Zeiss PNQ II,

were also used.

tMds1a f. Ikremmt 9f radioactivity
Radioactivity was measured either as infinitely thin dry samples

plated on aluminum planchets In a windowless gas flow Ge ger-Huller
counter, or as a liquid suspension fn scintillation fluid by means of a

Tr-Cerb liquid scintillation spectrometer. One to 10 ml of the
radioactive solutions were diluted with either the diexane-naphthelene

scintillation fluid of Bray (66) or the toluene fluid recomeded by
Packard Instrumnt Co. (67). The former scintillation fluid was
superior for counting water soluble ampounds.


I







19
Fractions obtained from ion exchange chromatography were assayed
for radioactivity by liquid scintillation spectrometry as aliquots of

0.1 Il to 10 al. Since sequential fractions from an I o exchange

otlumn contained Increasing amounts of acetic acid, which appreciably
quenched the counting response, an internal standard (C '-toluene) was

necessarily added to each assay vial to determine the exact amount of
radioactivity. Alternatively, the mthanol-acatic acid eluent was
evaporated from each fraction before addition of the scintillation
fluid. Internal standards were always used in counting yellowish
solutions, which showed significant quenching.

Specific activities of retinoic acid derivatives were calculated
by dividing the radioactivity In a given solution by the concentration
of retinoate derivatives in that solution. The concentration of
retinoic acid was determined from the absorbance of the solution and the

extinction coefficient of all-ta.l retinoic acid determined in this
laboratory, namely an 1 % of 1470 at the wavelength of maxi
absorption, which varied from 337 to 350 %q depending on the solvent.

Hydrolysl ao t ise tract
Hethanolic solutions of fraction I and III of liver and bile or
of whole bile were treated with 0.1 to 0.2 a, of 50% NaOH and heated in
a water bath at 60 to 70 under a ntrogen stream for 45 to 180 minutes.
After neutralization of the hydrolysate with concentrated HCI and
evaporation to dryness, the residue was suspended In 2 to 3 al of
ethanol and was again evaporated to dryness. The anhydrous residue
was then extracted with methanol or ethanol and stored under nitrogen.













EXPERIMENTAL RESULTS


Analvyis of Six Retinal Derivatives by Gas Chroematgrahy

6g chro-mtography o individual derivatives gg hort colgg ua

By the use of short columns packed with an acid washed, base
washed, and silanized support coated with 1% SE-30, anhydro retinol,

ethyl retinyl ether, retinal, and methyl retfnoete chromatographd as

virtually one substance without destruction (Fig. 2). For successful

chromatography of methyl retinyl ether, the column was necessarily aged

by passing carrier gas through it under normal operating conditions for
one or two days prior to use. Retinol and retinyl acetate, on the other

hand, dehydrated to form anhydro retlnol even after prolonged aging of
the column, in agreement with Ninomiya's report (53). These compounds

could be chroumtographed only by pretreating the column with an anti-

oxidant (p-carotene) and increasing the carrier gas flow rate hbout

six fold (Fig. 3). The treatment minimized dehydration of retinol and

retinyl acetate for one to three days.
dentif nation of chromtographically sarated coMounds Iby utre-
idimctra

In order to ascertain whether or not retinol derivatives were
chemically changed by GLC treatment, the ultraviolet spectrum of each
compound collected from the gas chromatographic effluent was covered

with the spectrum of the corresponding original material. As shown in
Fig. 4, the specter of the pr o t collected and injected cosounds agreed

closely. The sall amount of anhydro retinol present In chromatographed

20






















METHYL RETINYL ETHER


5 tO
TIME (MINI


RETINAL


TIME (MIN)


5 10
TIME (MINI


METHYL RETINOATE


TIME (MIN)


Fig. 2. Gas chromatography of individual retinot derivatives on
Gas Chrom P. Approxamtely 1 pg of each dervative was chromatographed
on a column of 60-80 sh Gas Chrom P coated with 1% St-30. Colum
length 2 ft.I column tperatures ISO C; argon flew rate 150 m1/lmn.


ANHYDRO RETINOL





















RETINOL
2- RETI NYL ACETATE







0 2 4 6
TIME (MIN.)






Fig. Gas chromatography of retinol and retinyl acetate.
Approximately 5 pg of each compound were separated on a colum of
Gas Chrom P coated with 1% SE-30 and treated with -carotene as
described na the text. Column lengths 2 ft. column tmpeatures
150 Ci argon flow rate. 880 al/min.











23











S METHYL HfETlIN ETMH(

t10 ANYODRO RETINOL 72



o06

04" i





410 3O0 60 340 320 300 280 o 3036 o30 0 300 00
WAVE LENGTH (m( ) WAVE LENGTH (mI )


ATI NAL L
SETHYL RETIOATE
12-



Io- 1'
10


44-

02- 02-

400 4 10 I 0 3 000 20 450 400 37 ; 30 325 3;0 MO0
WAVE LE0TH 1m.) WAVE LENGTH (mR)

RETINOL
S16-

t4-

12-
RETINYL ACETATE
10- 10 t

08- 08-

06.- 06-

04- 04-

02 02-

032o 0o 410 380360 340 320 300 2;0
410 380 3 340 320 300 280
WAVE LENGTH (mp)











Fig. 4. Spectra of retinol derivatives before and after gas

chromtography.


Spectra of solutions applied to ga chromtography


........... Spectra of collected copounds









samples of retlnol, retfyl acetate, and methyl retinyl ether apparently
resulted from dehydration after separation on the column, When these

compounds passed through the conventional detection system nd were

collected from the conventional output port, anhydro retinol accounted

for 50 to 100% of each compound Injected.

Except for retinol and retinyl acetate, the relative absorbance
of the spectra shown in Fig. 4 indicate the amount recovered from the

gas chromatograph. The compounds tested and their percent recoveries
were anhydro retinol (78%), retinal (10%), methyl retineate (94%),

methyl retinyl ether (59%), retinol (<25%), and retinyl acetate (<25%).

The low recoveries for the latter three compounds were caused by an

inefficient collection system, and for the latter tweo by the high flow

rate and the high injection pressure as well. The above recovery data,
although only approxiamte, do show that the ore stable derivatives

were chromtographed and collected without appreciable alteration.

T relationship between the mount injected and detector Crsosoe

Although the recovery of injected material was complete only

under special conditions, smll mounts of anhydro retinol, methyl

retinyl ether, retinal, and methyl retinoate could be quantitated with

the capillary Injection system. A Mixture of these four compounds was

separated sufficiently (68) to permit qualitative and quantitative

analysis. Either peak height or peak area was proportional to the

amount of these compounds injected doa to 0.025 pg (Fig. 5). The

slight deviation from linearity is apparently characteristic of the

argon ionization detector, since the recovery of trapped material is
linear with the concentration of the Injected meterfal (Fig. 6). The

capillary injection system used for these experiments gave reproducible


I













































05 01 015 0.2
Amount Injected (0&g)


Amount Injected (pg)


Fig Calbration curves for four rotinol derivatives
separated by gas chroeography. One dcroliter of solutions
containing a mixture of the fur derivatives shm above were
separated at 150 on a 2 ft. column of Gas Chrom P coated with
1% S5-30.







26





0.4- -8





0.3- -6


I tt


S0.2- -4






I I I I I





0.2 0.4 0.6 0.8 1.0
Retinal Injected (p.g)





Fig. 6. Comprtson of the recorder response with the Mnt of
retinal recovered by gs chromtography. Retinal ws chromtoraphed
on a 2 ft. colin of Gas Chro P coated with 1% SE-30, collected on
altuin at the output ports eluted with 0.5 al of ethanol, end mmured
spectrophotmltrically at 380 up. The heights of corresponding peak
on the recorder wre directly meured.
0 0- Peak height

SI.. Absorbance of recovered retinal







27

results when the injection pressure was le or moderate but not when It
was high. Hence, the dose-response relationship for retlnol and retlnyl

acetate was not examined.
Electron capture detection of retinal ad ethyl raetlnate

Certain types of molecules which have a high electron affinity,

such as halogenated hydrocarbons, nitro compounds and conjugated

carbonyls can be specifically detected by the electron capture

method (69). Since retinal and methyl retinoate are conjugated
carbonyls, their chromatographic behavior was monitored by splitting the

column effluent through both flame ionization and electron capture
detectors. Both detectors respond well to small amounts of retinal and

methyl retinoate (Fig. 7). Since the electron capture detector responds

poorly to the solvent, early components which are usually masked by the
solvent response of the flame or argon ionization detectors can be

measured by this device.
Eff ct of various factors on the stability of retinal and ethyl
retlnoate duing o chromatography

Since retinal and methyl retinoate were stable under the initial

mild conditions employed for short and medium columns (68), attempts
were made to improve the resolution and efficiency of the column. A

number of polar substrates were tried, such as QF1 (Dow-Cornmng),a
fluoro*sillcone polymer; XE-60 (General Electric Corp.), a cyano-

silicone polymer; and ethylene glycol adipate (Applied Science Labs.),

a high molecular weight polyester. Polar coatings inverted the elutlon
order of retinal and methyl retinoate (compare Figs. I and 6) and

Increased the retention times relative to non-polar coatings.

























METHYL
1 RETINOATE
RETINAL


| EC





O 6 12 18 24 30 36
TIME (MIN.)







Fig. 7. Comparison of the responses of a flame onization
detector and a electron capture detector to a mxture of retinal and
methyl retinoate. Approximtely 0.2 Pl of each compoent were eprated
by gas chromtography at 150I on a 2 ft. column of sfilmized
Chromsorb W coated with 1% QF 1. The effluent streak was mattered nt
parallel with a flm ionization detector (F) and an electron capture
detector (EC).







29
Three other observations should be mentioned (a) Retinal and
methyl retinoate were rapidly destroyed on st lanzed supports which were
washed only with acid. Acid washed packing was only suitable for use

after several days of aging, whereas a support neutralized with base
after acid treatment was immediately useful. (b) The polyester type

liquid phase contained an appreciable amount of alcohol and base soluble
material which was probably unreacted adipic acid. Retinal and methyl

retlnoate were partially destroyed on this type of column, even after a
considerable period of aging. (c) Exposure of the staple to uncoated
adsorptive sites on the supporting medium seemed to increase the

destruction of methyl retinoete. Some destruction of methyl retlnoate
occurred at 200" C on a 1% SE-30 coated Gas Chrom P column, while none
was noticed at 230 C on a 3% SE-30 coated Gas Chrom P column with three

times the length. In addition, adsorptive sites present in 1% coated

supports Increased the telling of peaks, thereby reducing column
efficiency. Ultimately the following chroatographic system was devised. liq-

uid phase, 3% SE-30 on Gas Chrom P column, 4.5 ft. glass# temperature, 180
argon flow, 50-55 ml/min.; detector, argon ionization. The efficiency
of this column was 1000 theoretical plates, four fold greater than that

of earlier columns, and was suitable for the separation of methyl
retinoate from tissue extracts.

Control Studies 2 on-radioactive adRadfoctive Retinoo c Acid

Stability of retinol c acid t various conditions of analysis

Dilute solutions (5-10 ug) of retinolc acid were stored in the
dark In non-polar solvents (heane) at -20 for may months without any

appreciable change or loss in optical density. A dilute solution of








retinoc acid In ethanol also showed no detectable loss of optical
density or alteration of spectrum when stored in the cold under nitrogen

for eae week. On the other hand, when a similar solution was exposed
to air and light in a Pyrex container for S hours, the optical density
decreased about 4%/hr. A similar solution placed n a low actinic flask
under air for the same period of time showed no detectable loss ti

optical density.
Crystalline retinoic acid as well as other retinol derivatives

apparently can be kept for months or years when stored under nitrogen

in the dark at -20 C. After a commercially available vial of retinoic
acid was opened, the unused portion was pieced in a low actinic bottle,

stoppered with a serum cap, and then flushed with nitrogen through
entrance and exit needles placed in the cap. Even after several months

of storage, the crystals retained the sam general appearance and
mreting point.
Accordingly, retinol derivatives in crystalline form and in

solution as well as tissue extracts containing retlnol derivatives have
been handled in the dark and under N2 as much as practicable, and have

been stored in the cold. Since tissue extracts were occasionally
subjected to light for considerable periods and fractions from ion

exchange chromatography frequently remained in cold solutions of acetic

acid and methanol for may days, a significant amount of isomerlzatlon
or destruction of retinoic acid probably occurred.

Irradiation and gsoeriaton of retinoic acid
When rtinoic acid was applied to a TLC plate and left in air
under subdued light for 30 minutes, several distinct zones representing
materials more polar than retinoic acid appeared upon subsequent







31
development in benzenesCHCl Imethanol (4sll). These ones were not

present in retinotc acid chromtogras developed immediately after
spotting. Since application of tissue extracts to a single plate
frequently required 30-45 minutes, small amounts of retfnofc acid In
these extracts might be entirely transformed to more polar components
during application. Thus the failure of Yagishita 2t l. (41) to find
small aunts of retinoic acid in tissue extracts when analyzed by this
procedure is understandable.

Upon ultraviolet Irradiation of all-tragn retinoec acid in
ethanol (1 rg/a1) in a closed tube for 12 hours, extensive isomerizaton

and sonm destruction occurred. On TLC analysis a n zaen appeared
which migrated faster than the all-tms isomer. This fast migrating

zone fluoresced under UV light and probably is composed of cis

Itomrs (ti).

Three major components were separated when the methyl esters of
the isomerate were analyzed by gas chromatography (Fig. 8), The relative
retention times and relative amounts of each of these components, which
are designated components (or peaks) A, 8, and C, in the isemrate and
In preparations of all-trI n and 13-cis methyl retinoate are shew In
Table I. The ultraviolet absorption spectra of all the collected
components after gas chromatography of the methyl esters of the
i somrate, the all-tr* n. and the 13-cia compounds were typical of
methyl retfnoate (see Fig. 4) and differed only In the wavelength of
maximum aheerption (from 348 to 359 up). Component A, which was found

only in the isomerate, had an absorption maxium of 348 qp. Component B,
also found In smtler amounts as an impurity in preparations of the all-

VMM and 13-cL Isoaars, had an absorption maxiam ef 349 up.











































0 6 12 18 24 80 A3 42
TIME (MIN.)


Fig. 8. Gas chromatography of
of retinote acid. Preparation of the
Separation was performed on a 4.5 ft.
Chrom P coated with 3% SE$-0. Colum
rates 55 al/Min.


the methyl sters of an isomrate
Isomrs is described in the text.
col-m of 100-120 mes Gas
temperature 180 C; argon flew














Table I
Comparison of somers of methyl


retinoate


Retention Absorption Reported of total

time mxfau of maximum peak ware
Sample
collected

methyl ster


minutes


Isomerate
produced from
ultraviolet
Irradiation of
alI-trans
retfnoate

Peak A

Peak S

Peak C

A11 trans
retlnoate

Peak 8

Peek C


Peak B
Peak C


7.65

10.10

15.00


10.15

15.00


10.15
14.80


348

395

355.5


349

355.5




358.5


35M




359








Since several cfs isomers have this same absorption maximum (62),
absolute identification was not possible from spectra alone. Component C

of the isonarate probably consists mainly of all-trans retinoate, since
both its retention time and absorption maximum (355-356 up) agree with
those of standard all-LtrE retinoate. Since the major peak of the
1)3-cs isomer also coincided with component C, however, this component
of the isomerate may also contain one and perhaps more ci isemers.
Since the absorption maxima of the methyl esters of the all-JItcn
and 13*-g isomers are unchanged by GLC analysis, Iomerizatlon of these
Isomrs was apparently minimal during gas chromatography. This

agreeable finding supports the above Interpretation of multiple peaks
in retinoate preparations, and might reasonably be extended to include
the rapid separation of all six isomers of methyl retineate.
Efficacy of mthlation with duia ethane and recovery fre Is
chrobatography
One milligram of all-tren retinoac acid was esterified with
diaomethane in the manner described In the Methods section, and the
resultant solution was compared spectrophotometrically with the Initial

retinoate solution. No significant change in absorbance occurred,
although the absorption maximum shifted n the expected way for ester
formation, from 348 pi to 352 up. Upon gas chromatography of the
esterlfied retlnoic acid solution, components b and C accounted for 2%

and 764 of the Injected sample, respectively. Under similar conditions,
pure crystal line methyl retinoate gave isomers B and C In yields of 2%
and 79% of the Injected dose, respectively. Thus diazomethane completely
esterifies retinoic acid without destruction or Isomerizatton. The 80
yields observed In these experiments were typical of a system consisting

of top-of-column Injection and capillary collection.








Analysis ifa pifled radioactive retlnoic aid
Since the 6,7.C14-retinoc acid (reported specific activity.
43,800 d'e used in these experiments hed been stored in a benzene
solution for about years prior to usage, it was purified before
administration to test animals. When purified by anion exchange
chromatography, most of the radioactivity appeared In fraction XI, where
non-radioactive retinolc acid is also eluted. Fraction XI was extracted
with ether, the ether was evaporated, and the residue was dissolved In
ethanol to yield a stock solution of purified undiluted radioactive
retinoic acid. Upon rechromatography of this purified undiluted radio-
active retinolc acid on an anion exchange column, 99.8 of the recovered
radioactivity was again found in fraction II. Upon gas chromatography
of the methyl ester of the purified undiluted labeled retinoic acid,
three maJor radioactive components appeared (Fig. 9). These components
corresponded exactly with those found in the isomerate of unlabeled
retinoate (Fig. 8). The last two components B and C, had the typical
absorption spectrum of methyl retinoate with absorption maxima of 348
and 354 ML, respectively. Component A was present in Insufficient
quantity for spectrophotometric analysis. Of the total radioactivity
collected in the eluate from the gas chromatograph, 6% was found in
component A, 10% tn B, and 33% in C. The overall recovery was much
lower than with diluted samples of labeled retinoate, probably as a
result of the destruction or adsorption during GLC analysis of part of
the small quantity (about 0.2 pg) of retinoic acid Injected.
For metabolti studies, purified undiluted radioactive retinoic
acid was diluted with pure non-radioactive all- retlnolc acid to
give a solution of labeled substrate retinoic acid with a specific

























I A
SB
L C



I-
I I I I I I II I
2000- 0 6 12 18 24 30 36 42 48
TIME (MIN.)

I-
1000









Fig. 9. Gas chromtography of radioactive retinoic acid
purified by ion exchange chromatography. TInty microliters containing
approximately 0.2 pg of mthylated radioactive retinolc acid were
injected Into the gas chromtograph and collected on anthracene crystals.
Radioactivity was measured as described in the text. The upper trace
Is the response from the argon detector. The tower trace shows the
corresponding radioactivity.








activity of 237 d-. Upon gas chromatography of this solution,
pg
componet B contained 15% and component C, 67% of the total injected

dose. When labeled substrate retinoic acid was recrystallized with 5 ag
of non-radioactive all-mtras retinoic acid, the specific activity after
four crystallizations was 53% that of the original solution. This
procedure measures only the radioactivity in the all-trans. somer.

Thus, component C of the labeled substrate retinofc acid apparently
contained 53% of the total radioactivity as all-t rretfnoate and ll4

as the 13-cfl or as other cis asomers.

Undiluted radioactive retinoic acid (43,800 ) was isomerizad
to a lesser degree when purified by silicic acid chromatography than by
Ion exchange chromatography. Radioactive retlnofc acid was eluted from
silicic acid with 5% acetone in hmaane, methylated, and separated by gas
chromatography (Fig. 10). In this case components A, B, and C contained

2%, 7%, and 45% of the recovered radioactivity, respectively. When
undiluted radioactive retinoic acid purified by silicic acid chromato-

graphy was diluted with non-radfoactive all-trans retinolc acid to a
specific activity of 522 d- and analyzed by gas chromtography,
components B and C accounted for 7% and 60% of the recovered radio-
activity, respectively. In both instances component C contained relatively

more radioactivity and components A and B less than when ion exchange
columns were used.

The presence of Isomers of retinoic acid in the labeled substrate
used for metabolic studies was both a help and a hindrance. The
Isolation of three isomeric components of retineate from biological
materials was useful as an additional criterion for identifying retnoate
and its conjugates. On the other hand, the presence of isemrs with



















Lj












\c
A B




0 6 12 18 24 30 36 42
TIME (MIN.)

S12,000-
-s
8,000-

4,000-







Fig. 10. Gas chromtography of radioactive retinolc acid
purified by silticc acid chromtography. Radioactive retinolc acid
was purified on a stidcie acid coluho mthylated, separated by gas
chromtography and collected on nthracene crystals. The upper trace
is the argon detector response. The lowr trace shows the corresponding
radioactivity.









widely different specific activities in the Injected substrate

complicated the Interpretation of radioactivity data In biological
experiments. Since a major objective of this study was the elucidation

of the metabolic pathway for retinoec acid, whatever its Isomeric form,
the existence of isomers in the substrate did not appreciably affect

the experimental approach or the conclusions reached,

Distribution Radoactivity tn. Various Tigsuesa go the Mdministration
fLabeed Substrates

Moderate doses of retinol, retinal, and retinoic acid were
administered to bile duct cnnulated rats in the manner described It

the Methods section. Gradient elutlon ion exchange chromatagraphy of

extracts of liver and Intestine, and of methanolic solutions of bile,
resolved the radioactivity Into three distinct fractions fraction I

contained non-Ionic compounds, fraction It contained any free retlnolc
acid and other material with similar Ionic properties, and fraction XII

contained more acidic compounds. The relative mount of radioactivity
found in various tissues after the administration of each substrate is

shown in Table II.

The bile became Increasingly radioactive for the first 6 hours
after the administration of 1-5 mg of retinol, retinal, or retinoic

acid. The cumulative amount excreted tn the bile in 6 hours was 16%
with retinol, 28% with retinal, and 38% with retinoic acid (Fig. 11).

The distribution of bile radioactivity on ion exchange chromatography

was similar with all substrates (Fig. 12). Of the recovered radio-
activity, approximately 10-15% was found in fraction X, 15-20% was

found in fraction II, and 60-70% was found in fraction III.
















0


I
4*

'S


ii



1

11d
t= X1
i at


b

I
'S
i



0
I


CU
0LP












I




b
U


.10
3






r.


I I I #


(4 *d 1% (4 *
.' 0 0L 0dr


iosm a- we
-% No
(4-c o r (4


4(405%


0% W %




-%* (4 *
-A 4 *
~G J `~ ~`o


g. U O U I Ub U O XI U




F f O OA O O & C& r &O
( N M M 4 1 m

NOmeanwe
4 #1 a a
rc(N~r~cvr e '40


Ir


eu




















60 Retinoic Acid


50 50


S40 -
/Retinal g

30 -
0 Retinol
o 20
20 -





4 8 12 16 20 24
Time after Injection







Fig. 11. The cumulative secretion rate of radioactivity into
the bile after the adantstration of I to ) ag of Cl4-retfnol, C14.
retinal, or Cl1-retinoic acid. Plotted values represent averages of
two, four, or twelve rats, respectively.






























000- e' o
(20%.) -
.20

0 -


5 10 15 20 25 30 35 40 45 50 55 60
Froction Number



~m-















S C i 15 2 2 a S S s S
FrCtion Nru


RETINOIC ACID




]AMB 11J






o io 5o 25s S5 o S b S b
Fracion Number






Fig. 12. Sawration of bl y Ion exchange chroatography after
administration of Cl-retinol, Cl%-retinal, or Ca-retnlec acid. B1le
was collected for a 6 hr. period lMdiately following administration
of eck substrate, and was treated a described In the txt prior to
chromtography.








The liver contained approximately 30% of the Injected dose
within 6 hours after the intravenous Injection of retinal or retinal,

with 97% of the recovered radioactivity nt fraction I (Fig. 13).
Virtually all of the radioactivity in fraction I was found in the

retinal ester fraction on alumina chromatography.5 On the other hand,

4.5 hours after the injection of retinoic acid, only 7% of the injected
dose was recovered in the liver, most of which appeared in fraction II
where free retlnoic acid is eluted.
The small intestine contained only 1% of the injected dose

6 hours after administration of retinol or retinal (Table I1). Ion
exchange chromatography showed that most of this radioactivity was in
fraction I (Fig. 14). On the other hand, 4.i hours after retinoic acid
administration, about 4% of the injected radioactivity appeared in the
smell intestine, where it was distributed evenly between fractions It
and III (Fig. 14).

Charetartzation of Fre Retinolic cid In i oln BL g'je. Zn ltin
and tl~w *ftr IntraveouL J96 ntraction f Botinotc Acid

Four and one-half hours after the administration of 4.5 qg of
retinolc acid to a rat with a cannulated bile duct, tissues were

examined for free retc acid. Portions of free r ad. rtn o action It from bile,
liver, and small intestine were treated with diazomethane and analyzed
by gas chromatography. The fraction eluted from gas chromtography
with the retention time of all-trans methyl retinoate was collected, was
dissolved In ethanol, and was examined spectrophotometrically.
Radioactivity was measured n corresponding fractions and components.

SR. 0. Zachuma, personal communication.
























RETINOL


6i 26 36
Fraction Numer


RETINAL


Fr10 2 Nu b
Fraction NurmMr


A rb S


RETINOIC ACID


II
I1n9v


0 10 NR


Fraction Number


40 53 63


Fig, 13. Separation of a liver vract by eon exchange
c photography after administration of C' -retinol, C -retinal, or
Cl -retinoc acid. The lver extract which was prepared as described
In the twt, was examined 24. hrs after the administration of retinot,

6 hra. after the administration of retinal and 4.5 hrs. after the
administration of ratinoic acid.


5000 -


a


0o 5b


5 o00.000 -
te


a
be


(411


"I I


--


I
























RETINOX


-In


n-


FraIion Nunr
Fraction Nutter


RETINAL


I III
.r .- -,


rtion Nl 3
Fraction Numorr


A (o 60


RETINOIC ACID


III,


0 10 23 30
Fraction Number


Fig. 14. Separation of a intestinal extract by fon exchange
c t ntgraphy after adldnItration of C14-retinel, Ca*retinal or
C' -rttnoc acid. The intestinal extract, which w prepared a
described i the text, was eamind 24 br. after thte adinistratfon
of retnol, 6 hrs. after the administration of retinal, and 4.5 rs.
after the dalmnistratfon of retfnolc acid.


45 h 6I







46
The specific activity of retinoate found in fraction II was increased

slightly by gas chromatographic purification to a value approximately
equal that of the Injected substrate (Table III). The absorption

maximum of retinoic acid at 347 qp was clearly evident in the ultra-
violet spectra of fraction XI of the small Intestine and liver (Fig. 15),

but was obscured In fraction II of bile by some material absorbing
primarily at 400 ma (Fig. 15). This material was also found In normal

bile of untreated animals (Fig. 15).
Thus, free retinoic acid persisted In tissues of the rat for at
least 4.5 hours after Its Intravenous Injection. During this period the

endogenous production of retinoec acid was small in comparison with the
administered dose.

Characterization of etabolites of Retinoate In Bite fter the iAdmIat
tration of CR1-RetInoic Acid

Experimental procedure

Bile, which contained more radioactivity than any other source
when investigated 6 hours after the administration of labeled ratlnoate,

was carefully analyzed qualitatively and quantitatively by Ion exchange,
gas, and thin layer chromatography to ascertain the nature of all the
radioactive metabolites of retinoic acid. The bile from six rats

dosed with 2 to 3 ag each of labeled retinoic acid was pooled and
examined according to the scheme shown in Fig. 16. The bile from six
additional rats dosed similarly was also separated by ton exchange
chromatography, and the resultant fraction III was further purified by

silicic acid chromatography. This purified fraction I] was treated
with various reagents to determine the nature of Its functional groups.


















Table III
Specific activities of retinoic acid from various tissues

4.5 hrs. after the Injection of 6,7-C 1-retinolc acid


Radioactivity Specific activity Specific activity of

recovered In of fraction I fr fr a action I after gas
Tissue
fraction II anion exchange chromatography

chromatography

% of initial dose dpm/pg dpm/Pg

Original
retinoate
substrata 139 -

Liver 5.2 110 125


intestine 2.1 106 128

Bile 3.2 133 134


I












































I I I I II I I
450 400 375 350 325 300 275
Wavelength (mp)


Fig. 15. Ultraviolet
and intestine 4.5 hrs. after


spectra of fraction II of bile, liver.
the admfnfstration of retinoic acid.


B Bile

L Liver


SI Sall intestine


BC Bite from a rat to which no retinoate was administered


1.6 -


1.4


1.2


1.0


S 0.8 -


S0.6 -


0.4


0.2


0.0 -









6,7.C 1-Retnoic acid

6 Rats

1. Collect bile for 10 hrs. and combine

2. Ion exchange


III

so


26%
N.


TLC 1. M
Tc, 4
2. Gd
3. C


1. OH' hydrolysis
2. Ion exchange


IX X7X' 1XX

I. Methylate
2. Gas chromatography
3. Collect and analyze


ithylate
is Chromatogra
>llect and anal
I
5%


/
c TLC TLC
piy 1. OH" hydrolysis
lyze 2. Ion exchange
II III

73% 22%

1. Hethylate
2. Gas chromtography
3. Collect and analyze


Fig. 16. Flow sheet for the treatment of bile after injection
of 67Ci-retinolc acid. Percetages refer to the relative amounts
of radioactivity found In each fraction after ton exchange
chroimtography.








The bilt accumulated for up to 10 hours after the administration of

radioactive retanoic acid to each of these twelve rats, was separated

by ion exchange. Bite collected for longer periods of tim was assayed

for total radioactivity only, since it was relatively non-radioactive.
Several organs from these twelve rats were also monitored for total

radioactivity for period up to 96 hours after administration of ratlnoic
acid.

The r o disapp ance of labeled retinoic acid from various ti, as
I extreti In the -bt

The total radioactivity found in the tissues of twelve bile duct
cannulated rats after the administration of 2-3 ag of retinolc acid is

shown as a function of time in Fig. 17. Radioactivity appeared rapidly

in the liver, kidney, mall intestine, and the bil. Thereafter, the
radioactivity In the tissues decreased at similar rates. Within 96

hours virtually 100% of the injected dose of C was excreted in the
bile. Apparently, the animal s entire store of Injected retlnoic acid

is eventually transported to the river and ec to th vr t crtd n the bile, largely

as a component of fraction XlI.

Fraction I of bile
Fraction I of bile, which contained the majority of the non-

radioactive components of bile, was difficult to analyze. The

characteristic absorption maximum of retinotc acid appeared in the
ultraviolet spectrum of fraction I from treated animals but net in
that from control animals (Fig. 18). Therefore fraction I of bile
appeared to be an ester of retinolc acid. To test this hypothesis,

fraction I was hydrolyzed with base, neutralized, and separated by ion
exchange chromatography. Seventy-four percent of the recovered radio-
activity was found in fraction It (Fig. 16). The ultraviolet absorption





















- I I


10 20


30 40


Liver


* Intestine
f Kidney


10 20 30 40
Time after Injection
Hours






Fig. 17. Effect of time on the recovery of injected radioactive
retinoic acid in various tissues. To to three aili|gres were
administered to each of twelve rats and the cited tissues were analyzed
for total radioactivity at the specified times.
























1.6 -


1.4 -




1.0 /



8 0.8


S0.6 0-


0.4 -


0.2 -


0.0 -
450 400 375 350 325 300 275 250
Wavelength (mp)












Fig. 18. Spectra of fraction I of bite.

............. Fraction I of bile collected for 10 hrs. after the
administration of retlnoic acid.

Fraction I of bile obtained from a rat not
Injected with retinoic acid.







53

spectrum of this fraction was characteristic of retinoic acid, although

it was partially obscured by material absorbing primarily at lower
wavelengths.
Upon gas chromatography of the methylated basic hydrolysate of

fraction I, several radioactive components appeared (Fig. 19). The
retention times and absorption maenl of the two main radleactive peaks

corresponded exactly to components 8 end C of standard methyl
retinoate (Fig. 20). A third radioactive peak, apparently present In

substrate retinoate (Fig. 9) but not in an somerate of non-radioactive
retinolc acid (Fig. 8), was situated between the last two ratinoate

Isomers. This compound did not absorb appreciably at 350 wi. All

three components absorbed extensively at lower wavelengths, however,
and were probably contaminated with components of fraction I which

overloaded the column and slowly bled through it. Although a radio-
active peak corresponding to component A appeared in the radlogram, so

much other lipid material was eluted in the sam fraction that spectral
analysis was not worthwhile. By summation of the radioactivity In
characteristic components of retinoate, retinoate isomers account for

at least 2$% of the C recovered upon gas chromatography of methylated
fraction I (Table IV), confirming the hypothesis that fraction I

contains an ester of retinoic acid.
The average specific activities of the retinoate in fraction I of

bile before and after hydrolysis and of the retinoate isomers found
after gas chromatographic analysis of fraction I hydrolysate were
approximately four times higher than that of the substrate retinol

acid (Table IV). Since the specific activity of the cls isomers of the
labeled substrate retinolc acid was much higher than that of the





















L44

(6)






L B


.C




80- 0 6 12 18 24 30 36 42 48 54
TIME (MIN.)
S60-

3 40-
20








Fig. 19. Gas chromtography of fraction II obtained after the
hydrolysis of fraction I of bile collected free rats injected with C'4
retlnoic acid. Whole bito was chroatographed on an ion exchange coluir
the resultant fraction I was hydrolysed with NaOH, neitralzed, and
rechroimtographed on an ion exchange colum. This final fraction II
was mthylated, and 10 Il were applied to gas chromtography. The
upper trace shows the response of the argon detector, while the lower
trace shows the corresponding radioactivity trapped on anthracene
crystals.










































I I I I I I
450 400 375 350 325 300 275 2
Wavelength (mn)


Fig. 20.
collected after
fraction 1.


Ultraviolet spectra of Isomrs of retlnoate
gas chroatography of the mthylated hydrolysate of


************ Component B


Cowonent C


1.2-


1.0 -


0.8-


0.6 -


0.4-


0.2-


0.0-















Table V
Specific activities of 6,7-C1-retinolc acid found in fractions of bile

After the administration of 6,7-C -retlnoic acid, bile collected
from six rats for 10 hrs. was pooled, chromatographed on an anion
exchange column, hydrolyzed by base, mthylated, and analyzed by gas
chromatography. Since anion exchange fractions contained ultraviolet
absorbing fipurities, reported specific actvites are miniem values.
The percentage of the total recovered radioactivity which was found in
each 6LC fraction is given in parenthesis. Average deviations were
calculated from duplicate assays


Specific activities

Average Average
Sample Methylated Cokmonants of 6LC
initially after basic
A B C
hydrolysis


dpWm/g


Injected
substrate


237t2


555 (15%)


190217 (67%)


Fractions from
anion exchange
column


>970


-(3%) 20801470 (6%) 12302250 (17%)


II >-725
III >725


---*(Q) 207t27 (21%)


-2- 861*1 (18%) 611tl0 (26%)


***


>762








all-trras somer, the above results favor the direct biological
esterification of ci isomers of retinoate without equilibration with
the all-trans form.
Fraction of bikL
The characteristic absorption maximum of retinoc acid was

virtually obscured in fraction It of this bile sample (Fig. 21), unlike
the previously cited experiment (Fig. 15) in which the amunt of

injected substrat was greater. Upon gas chromatographic analysts,
fraction IX contained four radioactive peaks, the last three of which
corresponded with peaks on the gas chrometogrna (Fig. 22) and had the
same relative retention times as components A, B and C of methyl
retinoate. Since the ultraviolet spectrum of collected componet C had
the absorption maximnu (352 qn) of methyl retlnoate (Fig. 21), the
presence of retinoic acid in fraction lI is confirmed.
Assuming that the last three radioactive peaks are somers of

retlnoate, approximately 28% of the recovered radioactivity of
fraction IX consisted of retinoic acid (Table IV). Since fraction XX
was found to be grossly contaminated with fraction Ut in this case by
TLC analysis, the reported content of retinoic acid is a amnmal- value.

The specific activity of component C derived from fraction It of
bile (207 d s) t approximately the saen as that of the starting
pg
material (190 l). Moreover, the relative distribution of retlnoate
isomers found in fraction XX of bile s nearly the same as that present
in the Injected substrate. These findings suggest that retinoic acid
was absorbed by the liver from the blood and directly excreted into the
bile without appreciable isomerization or dilution by endogenous
retinoie acid.


















1.2



1.0


0.8

u
| 0.6
0
0.4



0.2


0.0

450 400 375 350 325 300 2?5
Wavelength (mp)









Fig. 21. The ultraviolet spectra of fraction II of bile before
nd after gas chromtographic analysis.

.____ Spectrum of fraction I o.f be

.*.**., *Spectrm of the major radio.tiveo component
(peak C) of mthylated fraction It after gas chromtograplh

























A B


I I I I I
0 6 12 18 24 30 36 42 48
TIME (MIN.)



r HtA f 'T~n


Flg. 22. Ga chromtography of fraction IX from bile after the
administration of Clb-retinoic acid. Whole bile collected from almtls
inJected with C14-retinoic acid was separated by ton exchange
chrew tography. The resultant fraction U was methylated and 1O pl
were applied t the gas chromtograph. The top trace shos the
response of the argao detector while the lower trace shows corresponding
radioactivity collected on anthracene crystals.









Fracgiong III b Il
The ultraviolet spectrum of fraction 11I of bile clearly showed
the absorption maxfimu (35) mg) of retlnolc acid (Fig. 23), Indicating

that fraction I mst contain n acidic conjugate of retinolc acid.
No radioactive components were detected by gas chromatography of

fraction XX, even after methylatton and preparation of the salyl ether.
Apparently, the volatility of the radioactive components tn fraction III
is low, perhaps due to their high molecular weight or polarity. After
anon exchange chromatography of the basic hydrolysate of fraction rII,
however, 73% of the recovered radioactivity appeared In fraction II

(Fig. 16). This fraction had the characteristic ultraviolet absorption
maxisum of retinoec acid.

Upon methylatlon and gas chromatography of this resultant
fraction UI, the to major peaks of mass and radioactivity which
appeared (Fig. 24) corresponded exactly In retention times to
components B and C of methyl retnoate. These two major components,
which contained 18% and 26% of the recovered radioactivity, possessed

the absorption maxima of retinoate Isomers at 345 nq and 353 i!,
respectively (Fig. 25).

A smaller radioactive component, which contained 13% of the
recovered radioactivity, was also present nt resultant fraction 1X.
A similar component was also present In labeled substrate retlnolc
acid, but not in the retinoate tsomereat (Fig. 7).
The calculated specific activity of retinoate in fraction XII
before hydrolysis, in fraction IX after on exchange chromatography of
hydrolyzed fraction IXX, and in methyl rtfneate after methylatfon and
gas chromatography of the resultant fraction IX was about few times








61









1.2


LO -


0.8


S0.6-
0

0.4 -


0.2 -

.............. ................. **. ,ae **
0.0

450 400 375 350 325 300 75
Wavelength (mp)










Fig. 23. Ultrviolet pectrum of fraction III of bile after
admrinltration of Ci -retinoic acid.

ret_ cId. Fraction III of bile after the admitnatration of
rotismoic acid.

......... Fraction IX of bile from a rat not Injected with
retinelc acid.






























I -4 I I I 1 1 1 1 1 1


I I I I
0 6 12 18
TIME (MIN.)


fIlhmTfl4bhr~1


I I 1 1 I
24 30 36 42 48






U LM


Fig. 24. Gas chromtography of mthyleted fraction II obtained
by ion exchange chomatoraphy %f a basic hydrolysate of fraction III of
bile from a rat treated with C1-retineic acid. Whole bile collected
from animals injected with C1 -retinofc acid was separated by on
change chromatography. The reultant fraction III was hydrolyzed with
MaH nd eparated on an Ion exchange column. The resultant fraction IX
wa mthylated, and 10 Il were applied to the gas chromtegraph. The
upper trace shows the response of the rgo detector while the lower
trace shows the corresponding radioactivity collected on anthracene
crystals.


75-

50-

25-


LalffolnA








63








1.2


1.0-


08-


E 0.6-


0.4


0.2-


0.0-

450 400 375 350 325 300 275
Wavelength (rUp)










Fig. 25. Ultraviolet spectrum of retinoate Isom&ers collected
from gas chromatogrphy of methylated fraction III hydrolysate.

............ Component 8


Compoent C






64

that of the Injected labeled substrata retinoic acid (Table IV). As in
the case of fraction I, Isoers of substrate retinoate are apparently
incorporated into the metabolites of fraction III without prior
equilibration with the all-trans isomer.
A inimu of 7 of the recovered radioactivity from fraction IIX

consisted of retinoic acid. Since no other obvious radioactive peaks
appeared upon gas chromatography of the mathylated hydrolysate of

fraction IIt, the radioactivity found in fraction III of bile after the
administration of 2-3 mg of C4-rotinoic acid seemingly resides almost
exclusively in an acidic conjugate, or conjugates, of retinoic acid.
Investigation of b by thin lan chromatography (TLC)
Bile from six rats dosed with 2-3 ug of C -retinoic acid was

separated by ion exchange chromatography. The resulting fractions were
further separated by three different TLC system. The distribution of
radioactivity In each sample was obtained by counting sequential samples
of silica gel scraped in 0.5 cm increments along the migration path.
The Rf values of the mjer radioactive maximum of the components on
the thin layer chromatogrms are listed in Table V.
When fractions III and whole bile were developed with water, the

predominant radioactive maxima were associated with fluorescent spots
with an Rf of 0.75. In the same system, the Rf of radioactive maxim from
fractions I and 1 and of retinoic acid was zero.
When developed with benzeneechloroformmsthanol (4Iltl) (Fig. 26), a

relatively non-polar solvent system, the R of the major radioactive
maximum of fraction I was about 0.75, that of fraction IX was 0.20, and
that of whole bile and fraction III was only 0.03. The radioactive
maxmm of fraction III, which contains relatively few components, also











Table V

Thin layer chromatography of rat bile 10 hrs. after administration of
C 1-retnoic acid

After separation of various bile fractions by TLC, uniform
incremnts of the adsorbant were scraped into counting vials and assayed
for radioactivity by liquid scintlllation spectrometry. The relative
amunts of radioactivity In the radioactive maxim of each ample are
listed with the corresponding R'*s.

Developing system


H20 Benzeneschloroforma BenIeneschloroferm
Simple ethanol (4lslt) methanoliecetic

acid (555isl1)
I I II I I lll~ ll I I II I III I ll I I I I II [ I.


Rf % Rf


Whole bile




Fraction I




Fraction 11


Fraction III


0.00

0.55-0.75


0.00




0.02


0.15-0.40

0.70-0.85
0.00


0.02-0.15

0.84-0.92


60 0.08

0.72

0.84

20 0.20


20 0.02-0.10

35


Retlnoic acid


0.44

0.87

0.97
0.25

0.93


0.05

0.80

0.46


0.34


0.91


















I .".: 1~.:


6


Fig. 26. Thin layer chronatographic comparison of whole bile and
fractions of bile obtained by Ion exchange chromtography. Saples were
developed In benzeneschloroformiethanol (I 1st), detected by 12 vapors,
and photographed through a blue filter. Samples were
1 and 6 5 jg of retinoic acid


Fraction I
Fraction 1I
Fraction III


5 Whole bile


a


t



I


4
4


.., .....:..:. *[ :, ...
i: '.. i: .. ..*. ..:.
'+ .
: N: :.m .
V.% m.


io~-g






67

fluoresced under ultraviolet light like retinoic acid and stained with
iodine vapor. The Rf of the radioactivity In fraction XI was invariably

lower (0.20) than that of retinoic acid (0.34), but corresponded to an
artifact of all-tL retinoic acid which usually occurred after

exposure of the plate during application of sample.
Modification of the previous developing system by adding acetic

acid and reducing the amount of benzene increased the migration rate of
all components (Table V). The major radioactive components of fractions!
and I migrate near the solvent front, whereas the Rf of radioactive

maxima in fraction III and whole bile is about 0.S. The radioactivity
of fraction III was associated with a fluorescent area (Rf 0.44) which

apparently consisted of two components.
The following information was obtained about the composition of

fractions I, IX, and III as a result of thin layer chromatographic

analysis. (a) Adequate separation of whole bile and fractions I and It
is difficult due to the presence of a large amount of non-radioactive

material. On the other hand, fraction XII, which has fewer components,
was separated well on TLC. (b) Fraction I is composed of at least two

radioactive compounds, a water insoluble, non-polar component, and a
so-what more polar component. (c) The radioactive maximum of fraction XI
always corresponded to an artifact in standard all-trans retinoic acid

when analyzed by the benzeneechloroformtmethanol (41.t1) system.
Apparently the mall amount of retinoic acid on the plate is oxidized

during exposure to air and light and thus migrated differently from the
standard retinoate. (d) The major radioactive component of fraction III

was water soluble, was very polar, and fluoresced like retinoic acid
when irradiated by ultraviolet light. (e) The radioactivity of whole






68

bile largely consisted of fraction III, which is in agreement with the
results of Ion exchange chromatography.
Purification and characterization of fraction 11 o bf ie. h chromto-
r.anht siliciec acid
When fraction III, obtained by Ion exchange chromatography of

pooled bile, was chromatographed on a column of silicic acid, three major
radioactive components were found (Fig. 27). The first component
absorbed ultraviolet light non-specifically, but when analyzed by TLC

(Fig. 28), contained a radioactive substance with the Rf of free
retinolc acid as well as several other substances. The second and

third radioactive components from the silicic acid column possessed the
absorption spectrum of retinoic acid with maxima at 345 and 355 qp,
respectively, and contained the major portion of the radioactivity in
fraction III. Upon thin layer chromatographic separation and iodine
staining, no major impurities were evident (Fig. 28). Thus separation
of fraction III by silitic acid chromatography removed much of the
impurities including some free retinoic acid.
In an attempt to determine the nature of the functional groups
of the retinoic acid conjugate of fraction RIl, several derivatives of
purified fraction III were analyzed by thin layer chromatography.
Fraction III, purified on a silici acid column, was first treated with
diazomothene, which would make methyl esters of any acidic groups, and
then with hexamethyldistltaane in pyrldine, which would make ethers of
any free hydroxyl groups. Upon comparison of these derivatives with
untreated fraction III by thin layer chromtography (Fig. 29), the Rf
values and relative rates of migration were ester-ether (0.78)>ester
(0.29)>untreated fraction 11 (0.03). Retinoic acid in the conjugate
was unaffected by this treatment, since the radioactivity maximum and








69










8000 -


7000 Radioactivity

.......... Absorbance
6000 -


'9 5000 -
0


| 4000 -

3000 -


2000-


1000 1.0
S0.5


0 10 20 39 40 50 60 70 80
Fraction Number









Fig. 27. Silicic acid chromtography of fraction III of bile
after thi administration of C4-reatnoe ac id. The gradient elution
solvent system Is described In the methods section.

__Radioactivity

*..*.. ..*..... OOptical density at 350 uI













i,


*
:
*;;<*".


1 2


3 4 5


B





I


Fig. 28. Thin layer chromatography of c nponents Isolated by
silttl acid chromtography of fraction In from bie. The dweeloping
system in A was baieneichloreforrmsthanol (4Ilst), and in B wa
besmeneschloreformmethanolacetic acid (55s;5t). Samples wars

1, 5 Retinolc acid

2 Fraction III unpurtfied

3 Eluate from fraction 27


4 Eluate from fractions 35-50













1 P

2

30'

4

5
5 OW




100

o0
00-


1 o- 4^ --- r N


200

100
O -
I I I
0 5 10
Distance from origin (cm)


Fig. 29. Behavior of derivatives of purified radioactive
fraction IX which were separated on thin layer plates of Silica Get 6
by development with benzeneschloooroformmmthano (4itll). A photograph
of the lodinated plate is given In the upper figure, and the distribution
of radioactivity In various compounds is given belew. Samples weare

1,5 Non-radioactive retinoic acid

2 Fraction III, methyl ester

3 Fraction III, lslyl ether of the methyl ester


4 Fraction III, untreated







72

fluorescence exactly coincided in each of the three spots. Evidently
the non-radioactive moety of the fraction III conjugate possesses

carboxyl group or other acidic function and one or mere hydroxyl
groups.
Since glucuronides of several lipid alcohols and acids are
formed in biological systems, the corresponding derivatives of
glucuronic acid were analyzed in the Identical manner. Glucuronic

acid, its methyl ester, and the ether of its methyl ester migrated on
thin layer plates with Rf values of 0.00, 0.07, and 0.80, respectively.

Whereas the methyl glucuronate tetrasilyl ether and the methyl ester-
ether of fraction XII migrated similarly, methylated fraction III was

less polar than methyl glucuronate, and the untreated fraction III was
slightly less polar than free glucuronic acid. On structural grounds
a conjugate of retinoic acid and glucuronic acid would behave similarly

to glucuronic acidl therefore the behavior of fraction XI Is consistent
with a hypothetical conjugate of retlnoic acid and glucurenic acid.

Characterization of Retinalt atabolitos


Fraction a oft
Fraction I obtained from Ion exchange chromatography of rat bile
after administration of ) ig of C l-retinal failed to show the spectrum

of retinal or retinol ester. Apparently neither the substrate nor the
storage formed of vitamin A s excreted in the bile. Further studies

were not conducted because of the small mount of radioactivity found
in this fraction.








Fractain of a~iL
Fraction II did not possess the characteristic absorption
spectrum of retimic acid. After the addition of 5 ag of non-radio-
active s5% all-t agm retinolc acid to fraction IX, retinolc acid wa
crystallized four times to constant specific activity. The fourth
crystals had a specific activity 8.5% that of the original
material (Table VI). A non-radloactive impurity which c-crystallized
with retinote acid, absorbed slightly at 345 i and thereby wered
somewhat the specific activity of the final crystals. The specific
activity of the supernatant solution from the fourth crystallization,
which was free from this impurity, was 11.5% that of the original
material (Table VI). Upon gas chromatographic analysis of the final
crystals, 15% of the recovered radioactivity appeared In component
and 85% in C. The specific activity of the collected methyl ester of
component C agreed with that of the final supernatant solution (Table VI).
Since the crystals failed to account for mst of the radio-
activity in fraction I, supernatant solutions from the first three
crystallizations were combined, methylated, and analyzed by gas
chromatography. Qualitatively the sa peaks occurred in the chroaeto.
grai of the combined supernatant solutions as in those of the final
crystals (Fig. 30). Of the recovered radioactivity, 13% appeared in
retinoate component I and 23% in component C. The average weighted
specific activity of collected methyl retioate Isomers was 17.5 dJ ,
lpg
or 53% of the specific activity of the original solution (Table VI).
In summary 53% of the radioactivity In fraction I of bile
after C -retinal administration apparently consisted of rettnesl acid,
of which 8.5-11.5% was the all-tran Isomer and 35-40% were other Isomers.















Table VI

Specific activity of retinoic acid found In fraction II

of bite 8 hrs. after injection of 67-C l-retial

Material analyzed Specific activity Total retinolc
acid present

dpelpg Ig

Original fraction It diluted with
aon-radioactive retinoic acid 33.0 5.15
1st Crystallization 4.4 2.90
2nd Crystallization 3.1 2.18
3rd Crystallization 3.2 1.74
4th Crystalllzation 2.8 1.06

4th Crystals
Componnt C by GLC 3.8

4th Supernatant solution 3.8 0.42
Combined supernatant solutions 1-3
Compoent B by GLC 36.4
Componnt C by GLC 12.9

























tB
LL















S0-









FIg 30., Gas chromatography of fraction II obtained bCl lon
exchange chromatography of bitle after the admdnlsratton of Ci -rotinal,
Whole bCle collected from animals Injected with Cl4-retinal was
rated by Ion enacMge chromatgraphy. Fract on I so obtained was
diluted with non-radioactive retnoic acid, was mthylated, and 10 pil
I--















wereapplied the gas chromgtograph. o he f o trace s o s tin response
of the argon detector, while the lower trace shows corresponding
radioactivity collected on anthraseme crystals.






76

Whether this distribution in the isomers has biological significance
or only reflects isomrization during analysis has not been ascertained.
Fraction II of bilo
To ascertain whether fraction III of bile contained conjugated
retinoic acid after administration of C 1-retinal, fraction III was
hydrolyzed with base followed by ion exchange chromtography. Seventy
percent of the radioactivity in the hydrolysate was extracted with

methanol and of this methanolic extract, 53% of the radioactivity
recovered from ion exchange chromatography was eluted in fraction II.
The retinoate spectrum, although obscured by ultraviolet absorbing

compounds in unhydrolyzed fraction IIi, was clearly evident in this
resultant fraction II. After crystallization of this resultant

fraction II with 5 ng of non-radioactive 95% all-granS retinofc acid,
the specific activity of the crystals after crystallization was reduced

to less than 5% of that found in the original diluted fraction Ii
(Table VII). Essentially the sam low specific activity was found in
the methylated crystals purified by GLC.
Upon gas chromatography of the combined supernatant solutions
from the first four crystallizations, the two major radioactive peaks,
which contained 10% and 13%. respectively, of the recovered radio-
activity, corresponded exactly to components B and C of added non-radio-

active retinoic acid (Fig. 31). On the basis of data given in Table VI,
the average specific activity of these two peaks was calculated to be
2.2 Since the specific activity of the original solution was 10 I,

22% of the radioactivity of fraction III was retinoic aid, in close
agreement with the recovery data which indicated that 2! of fraction III
was retinoic acid. Apparently fraction III as well as fraction 1I
















Table VII

Specific activity of retinoic acid found in fraction III
of bile 8 lhr. after Injection of 6,7-C 1-rettal

Material analyzed Specific activity Total retinetl
acid present


dp/pg


Fraction III after hydrolysis,
rechromatography on ton exchange,
and dilution of the resultant
fraction l with non-radioactive
retinoic acid

lit Crystallization
2nd Crystalliztaton

3rd Crystallization

4th Crystallization

5th Crystalllzatlon

Sth Crystals
Copumont C by OLC

5th Supernatant solution
Cobined sApernatMt solutions 1-4

Component 8 by SLC

Compoaent C by GLC


10.1
4.24

1.69

0.78

0.60

0.40


5.02

2.04
2.25

1.59
1.00

0.67


0.34

0.87


0.17


11.5
1.4




























A




80- 0 6 12 18 24 30 36 42 48
TIME (MIN.)
60-

S40

20-





Fig. 31t. Ga chromtography of fraction It obtained by toI
exchange chromtogrphy of the basic hydrolysate of fraction IIt after
administration of Cl-retial. Whole bile collected from anmils
injected with C -retinal was separated by ion exchange chrostography,
Fraction III so obtained was hydro yzed with NaOHp neutralized, and
rechromtographed on an Ion exchange column, The resultant fraction II
was diluted with non-radloactive retinolc acid, mthylated, and 10 I&l
wore applied to the gas chromtograph. The upper trace shows the
response of the argon detector, while the tmer trace shows the
corresponding radioactivity collected on anthracene crystals.









contain Isaemrs of retinoic acid In addition to the all-.n fora.
In fraction III the radioactivity consisted of about 4-5% of all-

trs retinoate and about 18% of other isomers.













DISCUSSION

Methods for detecting mall amounts of retinol derivatives

In biological tissues generally depend on the unique spectral

properties of the conjugated alkne linkage. The absorption maximum
ts read either directly or after reaction with antimony trichloride

or similar reagents (70). Because these assays are relatively non-
specific and are affected by interfering chromophores (71), retinol

derivatives are often purified before analysis. Several procedures
have been employed (a) differential solubility between various

organic solvents, (b) adsorption chromatography on columns (72) or

thin layer plates (51) of alumina or silicic acid, and (c) partition
on liquid-liquid partition columns (73).
Since these methods separate compounds with relatively poor
resolution on the basis of their polarity, solubility, and adsorption
tenacity, retinol fractions are invariably contaminated with large
quantities of other lipids. Thin layer chromatography has the
additional disadvantage of exposing retinol derivatives to air and

light during application of the sample and development of the plate (51).
Gas liquid chromatography, although of great utility in
analyzing many lipids, has not previously been applied successfully to
vitamin A. Ninomya, in the only published report on the use of this
method, noted eKtensive dehydration of retinol and its esters to anhydro
retinol upon analysis by gas liquid chromatography (53). Because of

the high sensitivity and resolution of this technique, however, further








study of its possible applicability to ratinol analysis was deemed
worthwhile. In the present Investigation, factors which influence the

behavior of retinol derivatives on GLC columns were carefully examined,
ad suitable procedures were developed for the isolation of several

retinol derivatives with little destruction.
Many problem were posed tn applying this technique to vitamin A.

Although retinol derivatives can be volatilized, their instability imposes

narrow limitations on the conditions which can be used. The
conjugated alkene linkage, characteristic of retinol derivatives, Is

susceptible to free radical or acid catalyzed polymerization,
isomrization, oxidation, and addition reactions (74). Moreover,

compounds having a hydroxyl group alpha to an unsaturated bond easily
undergo acid catalyzed dehydration (59, 75). High temperatures
accelerate the rae of these reactions. Hence, the extensive formation
of anhydro retinal observed by Ninemiya (53) is expected from a
consideration of the chemical properties of this class of substances.
In order to minimize the destruction of retinol derivatives

during gas chromatographic analysis, rapid elutlon of compounds at low

temperatures was necessary. By employing low levels of liquid

loading (1% SE-30), high flow rates (150 nl/mnute), and short
olumns (2 ft.), short retention times for retinol derivatives were

obtained. Destruction of ratinol derivatives was further reduced by use
of a solid support (Gas Chrom P) which was base washed to remove acidic
sites and silanized to decrease the number of adsorptive sites. Under
properly selected conditions anhydro retinol, methyl retinyl ether,
retinal, and methyl retinoate were successfully separated by gas

chromatography. In this system, however, retinyl acetate and retinol







82

were largely dehydrated. By pretreating the colum with an antioxidant

such as p-carotene or hydroquinone and by increasing the flow rate six
fold, these compounds were also separated with little destruction.

P-Carotene proved to be the most useful antioxidant since it remained
on the column for several days, whereas hydroquinome exerted Its
protective effect for only an hour or to. Thus by eliminatIng

conditions which promoted aeid and free radical catalyzed reactions,
even the more unstable rettnol derivatives ware separated by gas

chromatography.
Unfortunately the efficiency of these columns was low. By

Increasing the concentration of the liquid phase, SE-30, to 3%,

destructive, adsorptive sites on the column were further reduced. This
column could be lengthened to 4.5 ft. and operated at a temperature of

18d0 and at a flow rate of 50 ml/minute without destruction of methyl
ratnoeate, the most stable of the retinol derivatives tested. The

column efficiency was therby increased from 250 to 1000 theoretical
plates, sufficient for the separation of methyl retinoste from most
other components, if not all, in tissue extracts.
Although the acidic character of same metabolites of retinsta

acid has been recognized (41), anion exchange resins have not previously

been employed for their purification. The preponderance of tissue
lipids ar non-ionic, however, and this technique allows the separation

of mall amounts of anionic compounds from the great mass of tissue
Itpids. Although acetic acid used in the eluant tended to catalyze the
somerization of retinolc acid, this effect was minimized by evaporating

the acidic solvent in the dark at a low tperature jn vauo.









Thus, the procedure of chromatographing tissue extracts on anion

exchange columns followed by chemical treatment of individual fractions
and ultimately by gas chromatography of methylated derivatives was of

critical importance in tracing the metabolin and fate of vitamin A
derivatives in the rat.
Like retinol, retinoic acid supports growth (14-16), although

it cannot be reduced to retinal in viv.~ Presumably retinal is oxidized
through retinal to retinoic acid, which then exerts a biological effect.

Surprisingly, free retinoic acid has not been detected in tissues in

vavI as a product of administered radioactive retinol, even when non-
radioactive retinolc acid was also given as a trapping agent (22, 41).

In those studies the lipid extracts were analyzed by thin layer chrome-

tography. In view of our results, this negative finding Is probably

due to the use of a method which caused destruction of retinoic acid,
and in addition was not adequately sensitive to detect the smal amount

of retinoic acid present in the liver and simll intestine. In the
present study, for example, the spectrum of retinoate was even obscured

in crude fraction II after separation of the great mass of lipids from
the acidic fraction of bile by ion exchange chrometography, and became

clearly evident only upon purification by gas chromatography.
Crystallization procedures must also be used with care. Wright

was unable to trap radioactivity in crystalline presumably all-trans

retinoic acid after the administration of C14-retinyl acetate to rats,
or after its incubation with pig adrenal homogenates (45). In the
present study crystallization of fraction It of bile derived from retinal
treated animals with non-radioactive all-trans retnoic acid indicated









that about 10% of this fraction was free all- a retinoic acid. But by
gas chroemtographic analysis of the remaining radioactivity n fraction It
of bile, at least 40% ore of fraction I was found to be g isomers of
retinoate. Thus great care mast be used in Interpreting crystallization
data, since the radioactivity my largely reside In ltomirs other than

the one specifically measured. In the present study retlnoec acid has
been unambiguously established as an intermediate in retinol mtaboliou
in vivo. Whether it Is an obligatory intermediate, or is only

Interconvertible with an unknown obligatory intermediate, has not yet
been establt shed.
In early studies retinoec acid could not be detected in tissues

even after the administration of large doses of retinoec acid (39, 41-43).
However, Jurkowitz, who devised a new acidic extraction method, did find
free retnoeic acid In human plasm after the administration of 100-120 ag
of the acid. He Justifiably criticized the extraction procedure used

by other investigators (44). Using Jurkowltz's procedure,
Krishnamurthy (37) also found free retinoic acid in chicks after the
administration of large doses of the free acid. In the rat, however,
free retinoic acid still could not be found. Using thin layer chrem-

tography, Yagishita at aL. were unable to detect free retinoate in
extracts of rat live r or intestine as little as 5 minutes after the
administration of I to 3 ug of retinoic acid (41). Similar results
were reported by Zile and DeLuca (43). In their procedure a liver extract
obtained from rats dosed with 1.5 mg of retinoec acid was passed
through a series of organic extractions and finally purified by silicic
acid and thin layer chreatography.








I was also unable to identify retinoic acid unambiguously In

tissue extracts by thin layer chromatography. By use of a combination

of ton exchange and gas chromatography, however, its presence t liver,
bile, and Intestine was clearly evident up to 4.5 hours after the
administration of 4 mg of retinoic acid to bile duct cannulated rats.

Since the amount of free acid in tissues is small, the inability of
others to detect it my be attributed to its Instability to light,
heat and acid, poor extractability, and the presence of large quantities

of lipid contaminants which obscured Its spectrum and made Its isolation
difficult.

Although small amounts of retinol acid persist in tissues, the
major portion of an Injected dose uast be rapidly metabolized. A

number of products of retinoate, several of which are water soluble,
have been reported (24, 37, 41, 43, 46-50). Most of these metabolltes

have been poorly characterized, d ndndeed my be chemical artifacts of

Isolation Instead of biological products. However, two groups of
investigators have studied products of retinoic acid metabolism In

sufficient detail to make worthwhile a comparison of their compounds
with the compounds isolated tn this investigation.

Zile and DeLuca (43) found a biologically active, radioactive
product of retinoic acid in liver after fractionation of a lipid extract

by silicti acid chromatography. The biologically active product was
slightly less polar than retinoic acid and my well be stiilar to my
fraction I, which had similar chromatographic properties on a thin layer

plate of Silica Gel 0. Smaller amounts of more polar but biologically
inactive metabolites which were noticed, were not further characterized.
Yagashita at ja. (41) also reported the presence of a biologically








active product of retinolc acid In the liver and the Intestine in
addition to a non-polar, non-biologically active product. From its

behavior on thin layer chromatography, their biologically active
product seem similar to my fraction III.
The ultraviolet maxima of the compounds Isolated by these two

groups were well below 300 ap, however, in sharp contrast to the
maxima of 345 to 356 pt found in all the major radioactive components

Isolated in this study. On the basis of spectra, these groups
suggested that the metabolism of retlnoate involves a shortening of the
conjugated double bond system and perhaps cleavage of the side

chain (41, 43). Present results do not support this view. Indeed the
majority of the radioactivity In bile after the administration of
retinoic acid was present in free or conjugated rtlnolc acid, and not
in same oxidized mntabolite. Since all of the radioactivity was not
found in retinoate no our study, the possibility exists that a minor
portion of retinoic acid may be reduced, hydroxylated, or cleavd. In
all likelihood, however, the spectra observed by Yagishita (41) and
Zile (43) were caused by ultraviolet absorbing contaminants which
obscured the spectrum of retinolc acid. These contaminants were also
found in all of the fractions Isolated in the present study; some even
co-crystallized with retinoic acid, a ndndeed were only eliminated
after methylation and purification by gas liquid chromatography.
The distribution of radioactivity in Ion exchange fractions of
tissues Is similar with retinal and retlnol, but markedly different
with retlnoic acid. These patterns are explicable in terms of the
known irreversibility of retinal oxidation to retinoate. On the other
hand, the pattern of radioactivity in anion exchange fractions of bile






87

is approximately the sam when any of the three substrates is injected.
In addition, the relative amount and pattern of radioactivity excreted
In the bile was the same whether large amounts (3 mg) or mal amounts
(10-25 pg) of retinoate were injected into the portal vein or into
intestinal loops in ljg (50). Apparently the formation n liver of a
water soluble mntabollte, which is the major radioactive component of
bile and was identified as a conjugate of retinoic acid, is a pro-
dominamt pathway for the metabolism of various form of vitamin A, once
they have been converted to retinolc, or to a copound of the same
oxidation-reduction state. Indeed the oxidation of retinal to ratinoic
acid seem to be the rate limiting step in this sequence.
The high excretory rate of retlnot derivatives in the bite,
which amounted to 20 to 60% of the injected dose in 24 hours, should
not necessarily be interpreted as elimination of an undesirable,
inactivated substance. In the Intact amimul Zachmn (50) has shown
that the radioactivity excreted in the bile is largely reabsorbed by
the gut and re-excreted into the bile. Thus, these metabolites are
involved in an enterohepatic circulation similar to that of the
eonjugted bile salts. Unlike retinol ester, retimoic acid may be
"stored" in the form of a circulating water soluble entity.
Two bits of evidence suggest that some biologically active form

of retinoic acid, although not necessarily the free acid itself, remains
in tissues n appreciable amounts for rather extensive periods.
Retinoic acid is toxic to rats when administered in daily doses of 2 .g
per 100 g of body weight (76). Symptoms of toxicity only appear with
relatively high doses of vitamin A derivatives. In the case of retlnol,
massive amounts of retinol ester are deposited in the liver and other
organs. With retinoic acid, however, little free acid is found in








tissues within hours after injection. Since retinoc acid s mOre
toxic than retinol (76), some form of the vitamin must persist in

tissues to produce these symptoms. In regard to growth response,
Nalathi t al. (15) have shown that a single 500 lg dose of retineic
acid promotes the growth of vitamin A deficient rats for a period of 4
weeks, while a similar mount of retinl sustains growth for 5 weeks.
Since retinetc acid is almost quantitatively converted to a water
soluble conjugate within 48 hours it is unlikely that the free acid
produces thee effects. It is attractive to propose that the retinolc
acid conjugate or sme product of It, which is produced rapidly and tn
large amunts n the liver, is reabsorbed by the Intestine, and is
readily soluble in plama, is the biologically active entity responsible

for both growth and toxic responses.
The retlnolc acid conjugate of fraction III possesses a acidic
function and one or more hydroxyl groups, properties which are
consistent with only one conjugate commonly found tn mnmmllan tissues,

nmely the glucuronlds (77). lucuronides are formed pririly in the
liver and are excreted n the bile and urine (78, 79). Since p-
glucuroandase has been found In almost every tissue and body fluid (80)
a conjugate of gluurenic acid and retinoc acid would most likely
be biologically active, or would become biologically active after
enzymatic hydrolysis. The structure of the conjugate has not been
unabiguously determined, however, nor has retinoyl glucurontde boen
synthestzed chemically. Hence, the possibility cannot be overlooked
that the polar molety of the retineate conjugate of fraction III may
yet be unknown.






89

Although the all-tras isomer of retinol derivatives are more

active than the el form In biological tests, the only well defined

function of retinol, the visual cycle, requires the l-cl- Isomer of

rettnal (20). The low activity of the ci. isomers, including the 11-cl_

form, when orally administered in growth tests, has been attributed to
an obligatory transformation to the all-raL forms presumably In the
liver and Intestine, prior to its transport to the eye and other

tissues (20.22). In the present Investigation, however, pil Isomers
were shown to be incorporated directly into conjugates of retlnoic acid

found in the bile without previous equilibration with the allt-tan
form. Hence, the condensation reactions to form conjugates is

apparently not specific for the all-trens forn. Whether thee.ecI
retinoyl conjugates re active in biological systems Wr #, after

isomerization to the all-trans form, or not at all, Is presently

unknown.














SUJMARY


1. Anhydro retinol, methyl retlnyl father, retinal, and methyl retinoate
were separated by gas chroatography on short columns of sllanized Gas

Chrom P coated with 1% SE-30. The operating temperature was 150 C and

the flow rate of carrier gas was 150 al/minute. Retinol and retinyl
acetate, which were dehydrated when chromatographed In this annerp, wre

successfully separated after the colu was treated with p-carotene md
the flow rate was Increased 6 fold. Methyl retlnoate was Isolated from

tissue extracts on a longer, i fold more efficient column which contained
3% SE-30 and was operated at 180 with a flow rate of 50 m1/minute.

2. Ionic metabolites of retinol, retinal, ad retinoic acid were

separated from non-Ionic metabolites and from massive amounts of non-

Ionic lipids by means of a gradient elution, anion exchange chromate-

graphic system. Compounds In tissue extracts, when eluted with

increasing concentrations of acetic acid in methanol, were separated
reproducibly into three major fractions on columns of Bio-Rad AG2X8

In yields of 80-100%.

3. After the administration of 3 mg of C 1-retinal to bile duct
cannulated rats, free radioactive retinole acid was recovered from the

bile. Thus free retinoic acid has been established as an Intermediate
in retinol metabolic, although not necessarily an obligatory

intermediate.






91

4. The presence of mall amounts of radioactive, free retlnoie acid

was unambiguously demonstrated in liver, the small Intestine, and bile

up to 6 hours after the administration of 2-4 mg of C -retinoic eaid

to bile duct canulated rats.

5. Non-acidic conjugates (fraction I) were found in the tissues after
the administration of retinot, retinal, or retinoic acid. After the

administration of retinol or retinal, retinol ester appeared in the

liver and the Intestine but not in the bile. After retinoate

administration, an ester of retinoic acid was found in the liver and

in the bile.

6. Acidic polar conjugates (fraction III) were found in the bile after

the administration of radioactive retinol or retinal, and in the liver

and intestine as well as the bile after the administration of radio.
active retinolc acid. An acidic conjugate of retinoic acid accounted

for 23% of fraction IXt of bile after administration of retinal and >$0%
of fraction UII of bile after retinoate administration. This

conjugate was water oltuble, polar, possessed a carboxyl group or
other acidic function, and contained one or mwre hydroxyl groups.

7. After the administration of C -retinoic acid to bile duct

camulated rats, the major recovered components were free rettnoie acid
and Its conjugates, rather than further oxidized or partially degraded
metabolites.













REFERENCES

1. International Union of PAre and Applied Chreistry, DefniStiv
Rule* fM t manclature of the Vil minsp JA Chem-

2. McCollum, E. V., and Davis, M., .~L Siot. Ch.ap, j 245 (1914).
3. McCollum, E. V., and Davis 0 J. eL CIah.. t l 181 (1915).
4. Osborne, T. B., and Hd el, L. B., j.Bolg Cheim. 20, 379 (1915).
5. McCollum, E. V., Simonds, N., Decker, J. E., and Shipley, P. G.,
J. Bola. Chw. lb 292 (1922).
6. Moore, T., SVtB A Elsevir Publishing Company, Amtwrdm,
1957, p. 5.
7. Wolbach, S. B., ind Howm, P. R., J Ex2ptl. )d 43 753 (1925).
8. Ialbah, S. B., and Hwe, P. R., J. Extl. H~ 5.. 511 (1933).
9. Mbrlae, S. 8. 0., Nutr.~ 45 (1934).
10, Coetie, W. H. K., Bgcham. J9, f 628 (1949).
Il. Dowling J. 1., ad Wald, 6., Proc. Nat. Aced. YS 6648
(1958).
12. Ames, s. R.*, Ann R. Slochem.. 3 371 (1958).
13. More, T., Vitg n A, Elsevier Publishing Company, Amstwdml,
197P, p. 375.
14. Van Dorp, 0. A, nd Arens, J. F., Nature 18. 60 (1946).
15. M.lath l P., Ro K. S., Sestry, P. S., and Gnguly, J., BIDoch. J
8jL 305 (1963).
16. de in, T. J., van Lees n, P. H. and Robergh, J. R., Nature. 201
77 (196).
17. Murray, T. K., Pro. So. Exptl. Blot. MI2L 111j 609 (1962).




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EZ4ZQOQHX_F70S13 INGEST_TIME 2012-03-02T22:40:51Z PACKAGE AA00009500_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES