AROMATIC AMINO ACID STUDIES

APPROVED;

Graduate Committee:

'.u —I "I Major Pr^fessot

,4——

Committee Member

(AJ. Committee Mem^r

Committee Member

Director of the Department of Chemistry

Dean1 of the Graduate School 7

Sullivan, P. Timothy, Aromatic Amino Acid Studies.

Doctor of Philosophy (Chemistry), December, 1970, 1^0 pp.,

18 tables, 21 figures, 89 references.

Pyridine ring analogs of the aromatic amino acids-

phenylalanine and tyrosine were synthesized and studied in

microbiological and mammalian systems.- Among the synthetic

analogs prepared were the isomeric pyridylalanine N-oxides,

(3-(2-, 3-, or ^-pyridyl 1-oxide)-DL-alanine. Bacterial

growth inhibition studies in Escherichia coli 9723 revealed

that ^-pyridylalanine N-oxide (I) is a fairly active phenyl-

alanine antagonist to this organism. However, the toxicity

of ^-pyridylalanine N-oxide was shown to be due to enzymatic

reduction of I to the more toxic ^-pyridylalanine.

In a second study, 5-hydroxy-2-pyridylalanine, a pre-

viously reported tyrosine antagonist, and its N-oxide der-

ivative were prepared. Iodination of these compounds

resulted in monoiodo derivatives whose structures were

determined by nmr spectroscopy. The new iodo amino acids,

B-(5-hydroxy-6-iodo-2-pyridvl)-DL-alanine and {3-(5-hydroxy-

6-iodo-2-pyridyl 1-oxide)-DL-alanine, were found to be very

weak inhibitors to the growth of E. coli 9723.

Also synthesized were the isomeric a-fluoro~, and a-

hydroxypyridylalanines. The fluoro amino acids were found

to be active phenylalanine antagonists in the microorganisms:

E. coli 9723, L. arabinosus 17-5? and L. dextranicum 8086.

These analogs, plus others synthesized, the a-bromo-, and

a-chloropyridylalanines, were also studied in the rat liver

tryptophan hydroxylase system. An anomalous behavior of the

hydroxylase system was observed, and thus a more detailed

investigation of this enzymatic process was undertaken.

Typical substrate-velocity response plots for the

hydroxylation of tryptophan are sigmoidal rather than

hyperbolic. Various factors which might be responsible for

the non-Michaelis-Menton (hyperbolic) kinetics were examined.

Evidence strongly indicates that the sigmoidal kinetics are

the result only of the hydroxylation reaction and not a loss

of tryptophan or product (5-hydroxyindoles) through inter-

fering processes. It was also found that various agents

such as guanidine hydrochloride, sodium dodecyl sulfate,

alcohols (especially 1-propanol), and high reaction temper-

atures give rise to hyperbolic kinetics. Thus it is believed

that rat liver tryptophan hydroxylase is subject to allo-

steric regulation, especially in light of the effects of the

known desensitization agents-previously mentioned which have

been shown to modify and/or destroy enzyme cooperativity

through tertiary and/or quaternary structure alterations.

AROMATIC AMINO ACID STUDIES

DISSERTATION

Presented to the Graduate Council of the

North Texas State University in Partial

Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

By

P. Timothy Sullivan, B. S, 'I

Denton, Texas

December, 1970

PREFACE

The author wishes to express his utmost appreciation

for the very able and inspirational guidance of this re-

search by his supervising professor, Dr. Scott J. Norton.

In addition the author expresses his gratitude to Mrs.

Marian Kester for her aid in some of the microbiological

assays and Mrs. Mary Moore for her aid in the reproduction

of the many figures in Part IV of this dissertation.

Further, the author wishes to express his gratitude to

Mrs. Cynthia Sullivan for her help with many of the micro-

biological assays and the compilation of this dissertation

into its final form.

The financial support of this investigation by the

North Texas State University" Faculty Research Fund, and

the Robert A. Welch Foundation of Texas is gratefully

acknowledged.

Tim Sullivan

December, 1970

iii

PART

TABLE OP CONTENTS '

Page

I . SYNTHESIS AND STUDY OF PYRIDYLALANINE

N-OXIDES 1

INTRODUCTION * 2

EXPERIMENTAL 6

Experimental Techniques 6

Microbiological Assays 6

Assay Media . . . . . . . . . . . . . . . 7 Preparation of Bacterial Inoculum

Suspensions . . . . . 7 Assay Procedures 8

Organic Syntheses 9

*f-Chloromethylpyridine 1 -Oxide Hydrochloride . 9

Ethyl 2-Acetamido-2-(1+-pyridylmethyl 1-oxide )malonate 9

p_(l+_Pyridyl 1 -oxide) -DL-alanine 12

RESULTS AND DISCUSSION 13

SUMMARY 29

REFERENCES 30

II. PYRIDINE RING ANALOGS OF IODOTYROSINE . . . . 32

INTRODUCTION 33

EXPERIMENTAL 3I+

Experimental Techniques 31*

Microbiological Assays 3*+

iv

TABLE OF CONTENTS (Continued)

Page PART

Organic Syntheses 3*+

Organic Intermediates 3*+ ^-Chloro-2-chloromethyl-5-methoxy-pyridine 1-Oxide 35

Ethyl 2-Acetamido-2-(l+-chloro-5-niethoxy-2-pyridylmethyl 1-oxide)malonate . . . . 35

j3-( 5-Hydroxy-2-pyridyl 1 -oxide)-DL-alanine 36

P-(5-Hydroxy-6-iodo-2-pyridyl 1-oxide)-PL-alanine 37

Etfiyl 2-Acetamido-2-(1+-chloro-5-methoxy-2-pyridylmethyl)-malonate . 38

p-(1f-Chloro-5-hydroxy-2-pyridyl)-DL-alanine 38

p-T5-Hydroxy-2-pyridyl)-D^-alanine . . . . 39 p~( 5-Hydroxy-6-iodo-2-pyndyl)-DL-alanine 39

2-Todo-6-methyl-3-pyridinol h-0

RESULTS AND DISCUSSION . . . ! Vl

SUMMARY *+8

REFERENCES -9

III. a-FLUORO- AND a-HYDROXYPYRIDYL ALANINES . . . . 50

INTRODUCTION 51

EXPERIMENTAL 55

Experimental Techniques 55

Microbiological Assays . . . . . 55

Organic Syntheses 55 4

a-Fluoropicolines . . . . . 56 2-Fluoro-3-bromomethylpyridine Hydrobromide . . . . . 56

V

PART

TABLE OP CONTENTS (Continued)

Page

Ethyl 2-Acetamido-2-(2-fluoro-3-pyridylmethyl)malonate 58

Ethyl 2-Acetamido-2-(2-fluoro-3-pyridylmethyl)cyanoacetate 58

P-(2-Hydroxy-3-pyridyl)-DL-alanine 58

B-??-Fluoro~ V- cvridyl)-DL-alanine . 63

RESULTS AND DISCUSSION 6b

SUMMARY . . 73

REFERENCES . . yk

APPENDIX . 76

IV. A REEXAMINATION OF THE PHENYLALANINE-TRYPTOPHAN HYDROXYLASE SYSTEM OF RAT LIVER 77

INTRODUCTION 78 .

EXPERIMENTAL METHODS 90

Instruments and Equipment 90

Abbreviations Used 90

Chemicals 90

Preparation of Rat Liver Homogenates . . . . 91

Enzymatic Reaction Procedures 92

Colorimetric Assays . 93

RESULTS 95

INTERPRETATION AND DISCUSSION 128

REFERENCES 137,

vi

LIST OF TABLES

Page Table

PART I

I. Pyridylalanine N-Oxides and Intermediates . . 10

II. Analytical Data for Pyridylalanine N-Oxides and Intermediates . . . 11

III. Reversal of ^-Pyridylalanine N-Oxide Toxicity in E. coli 9723 by Phenyl-alanine and Tyrosine 1*+

IV. Absorbance Maxima for Substances Tested Employing a Pyridine Ring-Specific Color Reaction 21

PART II

I. NMR Spectra

PART III

I. Intermediate Bromomethylpyridine

Hydrobromides 57

II. Intermediate Ethylacetamidomalonates and

Ethylacetamidocyanoacetates 59

III. Fluoro and Hydroxy Substituted Pyridylalanines 60

IV. Analytical Data for Synthetic Intermediates . 61

V. Analytical Data for the Fluoro and Hydroxy Substituted Pyridylalanines 62

VI. Summary of Microbial Growth Inhibitions i by Fluoro- and Hydroxypyridylalanines « . . . 66

VI1

Table

LIST OF TABLES (Continued)

Page

VII. Reversal of Fluoropyridylalanine Toxicities in Lactobacillus arabinosus 17-5 by DL-Phenylalanine 68

VIII. Reversal of |3-(6-Fluoro-3-pyridyl)-DL-alanine Toxicity in leaeonostoe dextranicunf~S086 by DL-Phenylalanine 70

IX. Reversal of B-(6-Fluoro-2-pyridyl)-DL-alanine Toxicity in Escherichia coli 9723 by DL-Phenylalanine 71

PART IV

Tryptophan Hydroxylase Activity of Various Mammalian Tissues 87

Studies on the Presence of Tryptophan-Utilizing-Enzymes in the Rat Liver Preparation 123

Paper Chromatographic Data Comparing Rf Values of Tryptophan, Serotonin, and Tryptamine with Compounds Produced During a One Hour Incubation of Tryptophan with Rat Liver Preparations 12^

IV. Studies on the Catabolism of 5-Hydroxy-tryptophan by Enzymes Present in the Crude Rat Liver Preparation 126

I.

II.

III.

viii

Figure

LIST OF FIGURES

Page

PART I

1. Reaction Sequence Leading to the Isomeric Pyridylalanine N-Oxides . . . b

2. Ascending Paper Chromatogram Tracing Indicating the Reduction of "Various Substituted Pyridine N-Oxides to the Corresponding Substituted Pyridines by Growing Cells of E. coli 9723 18

3. The Effect of e-Hydroxymethylpyridine N-Oxide on the Inhibition of E. coli 9723 by ^-Pyridylalanine N-Oxide 2b

b. Effects of Increasing Concentrations of 3-Hydroxymethylpyridine N-Oxide on the Toxicity of ^-Pyridylalanine to E. coli 9723 27

PART II

Sequence of Reactions Leading to the Iodopyridylalanines ^3

PART III

Sequence of Reactions Leading to the a-Fluoro- and a-Hydroxypyridyl-alanines 53

IX

Figure

LIST OF FIGURES (Continued)

Page

PART IV

1. Tryptophan Hydroxylase—Typical Tryptophan Response Plot for One Hour Reaction at 30 . . 80

2. The Phenylalanine Hydroxylation System

of Rat Liver 83

3. The Biosynthetic Pathway of Serotonin . . . . 8*+

*+. Effect of Phenylalanine on the Hydroxylation of Tryptophan 97

5. Effect of Enzyme Pretreatment with Guanidine Hydrochloride on the Hydroxylation of Tryptophan 100

6. Effect of Enzyme Pretreatment with Sodium Dodecyl Sulfate on the Hydroxylation of Tryptophan 102

7. Effects of Concentration of Reduced Pteridine on the Hydroxylation of Tryptophan by

- Nontreated and Guanidine Hydrochloride Pretreated Enzyme 10^

8. Phenylalanine Hydroxylase—Effect of Pre-treatment of Enzyme with Guanidine Hydrochloride on Hydroxylation of Phenylalanine 107

9. Effect of Incubation Temperature on Hydroxylation of Tryptophan 109

10. Effect of Enzyme Pretreatment by Heat on the Hydroxylation of Tryptophan 111

11. Effect of 1-Propanol on the Hydroxylation , of Tryptophan 11 if

x

Figure

LIST OF FIGURES (Continued)

Page

12. Effect of Various Primary Alcohols on the Hydroxylation of Tryptophan 117

13. Effect of Certain Alcohols on the Hydroxylation of Tryptophan 119

1*+. Effect of 1-Propanol on the Hydroxylation of Phenylalanine 121

15 Tryptophan Hydroxylation Employing Partially Purified Enzyme Preparations . . . . . . . . . 133

xi

PART I

SYNTHESIS AND STUDY OF PYRIDYLALANINE

N-OXIDES

INTRODUCTION*

Although the synthesis and biological study of struc-

tural analogs of the aromatic amino acids, phenylalanine and

tyrosine, have been conducted in many laboratories for over

two decades, the study of new s-tructura-l analogs remains

interesting and often enligii tuung. The three pyridy 1 alanines

(the alanine side chain substituted in the 2, 3, and b positions

of the pyridine ring) have been synthesized, and the 2- and h-2-i+

pyridylalanines are well-documented antagonists of phenyl-

alanine. Certain of these pyridylalanines have been employed

in enzyme specificity studies and also serve as false feedback 5-7

inhibitors. Similarly, a tyrosine analog containing the

pyridine ring, 5-hydroxy-2-pyridylalanine, has been a useful

tool in the study of biological processes.Further, *f,5-

dihydroxy-2-pyridylalanine, a structural analog of Sj^-dihy-

droxyphenylalanine (DOPA), has been found to serve as a

substrate for DOPA decarboxylase, while it inhibits the oxi-

dation of DOPA by the enzyme, tyrosinase.^ Certain substi-

tuted pyridine and quinoline N-oxides have been studied for

fungistatic and bacteriostatic properties.^ The N-oxides

that were found to be active in the study were classified as

"wide-spectrum" compounds^ . This work has recently been published: P. Timothy Sullivan, Marian Kester, and S. J. Norton, J. Med. Chem., 11, 1172, (1968). *" '

The synthesis (Figure 1) and the determination of the

biological activity of the three pyridylalanine N-oxides,

p-(2-,3-, or ^-pyridyl 1-oxide)-PL-alanine, was interesting,

therefore, in that these compounds could serve reasonably as

either phenylalanine or tyrosine antagonists. The ^-pyridyl-

alanine N-oxide, for example, might be expected to act as a

tyrosine antagonist because of the presence of an oxygen

atom para to the alanine subsbituent on the pyridine ring.

The other two pyridylalanine N-oxides (alanine substitution

in either the 2 or 3 position) would not be expected to be

as effective tyrosine antagonists as the ^-substituted analog,

because of the ortho and meta orientations of the oxygen

atom5 however, no selection of any of the three analogs as

a most probable antagonist of phenylalanine can be made

easily.

In the present study the three pyridylalanine N-oxides

were synthesized,* and microbial growth inhibition studies

were conducted. ^-Pyridylalanine N-oxide was found to be

antagonistic to both phenylalanine and tyrosine in Esche-

richia coli. However, a strictly competitive reversal of

the toxicity of ^-pyridylalanine N-oxide by either phenyl-

alanine or tyrosine (or combinations of both) could not be

demonstrated. The 2- and 3-pyridylalanine N-oxides are

f

^-Pyridylalanine N-oxide has been synthesized by a differ-ent procedure than that reported herein." .(See reference 12) The compound has not previously been studied biologically.

Figure 1

Reaction Sequence Leading to the Isomeric

Pyridylalanine N-Oxides

pCH20H S0C12 ^ [| -jj-CH2Cl

I X - h o 1

I a,b,c II a,b,c

0(COOCpH^)o

-CH 2C1 -j- (C00C2H^) 2 F^^CHg-C-NHCOCH^ H-C-NHCOCH^ NAOCOH* J

6 - H C I 3 J II a,b,c III a,b,c

(COOC2H?)2

R^-CH 2-C-NHC0CH3 HCI ^ [F^-CH2-CH-C00H

i i NHa

H I a,b,c IV a,b,c

a — ^-substitution

b — 3-substitution

c — 2-substitution

only weakly inhibitory to this organism, and the toxicities

of the latter two compounds are reversed by low concentra-

tions of either phenylalanine or tyrosine. Evidence is also

given for the enzymatic reduction of the alanine-substituted

pyridine N-oxides by E. coli to produce the corresponding

pyridylalanines. Further evidence indicates that the rate

of conversion of the pyridylalanine N-oxides to the corre-

sponding pyridylalanines is an important factor, in the in-

hibition of E. coli.

EXPERIMENTAL

Experimental Techniques

A Thomas-Hoover capillary melting point apparatus was

employed for all melting point determinations, and the

melting points reported are uncorrected. Paper .chromato-

graphic studies were conducted by the ascending technique

using Whatman No. 1 chromatographic paper. Uv spectra were

determined with a Beckman DBG recording spectrophotometer.

All compounds synthesized in this study were analyzed for car-

bon, hydrogen, and nitrogen. The data are presented in Table

II. The hydroxymethylpyridines and their N-oxides were ob-

tained from Aldrich Chemical Co., Inc.

Microbiological Assays

Stock cultures of.Escherichia coli 9723 were maintained

on agar slants of a glucose-inorganic salts medium.^

Monthly transfers of the organism were made to new agar slants,

the slants allowed to incubate at 37° for about 2^ hours to

insure full growth of the organism, and the slants stored at o

5 • Cultures of Lactobacillus arabinosus 17-5 were main-

tained and stored in a like manner after growth on agar <

stabs of a glucose-yeast extract medium for ^8 hours at 30°.

Assay Media.— For the assays employing E. coli 9723 as

test organism a previously reported inorganic salts-glucose

medium13 Was employed. When L. arabinosus 17-5 served as

the test organism, a previously described amino acid medium1^

was used, except that phenylalanine and tyrosine were omitted

from the basal medium. Phenylalanine and tyrosine were sup-

plemented as required for limited growth of the organism,

and the tryptophan and aspartic acid concentrations were in-

creased threefold.

Preparation of Bacterial Inoculum Suspensions.— In pre-

paration for a microbial assay, transfers were made from

stock cultures of E. coli into 10 ml of a sterilized glucose-

inorganic salts solution, previously described. The inoculum

o

tube was then incubated at 37 until a heavy growth of cells

had developed (usually 12 to 15 hours were required). One

ml of this culture was then transferred aseptically to a

tube containing 10 ml of the same medium, and the incubation

was continued for 6 to 8 hours. This culture was centri-

fuged in a clinical centrifuge at about 2,000 x g for 6 to

8 minutes, and the bacterial cell pellet was washed by sus-

pending it in 10 ml of sterile 0.85$ sodium chloride. The

cells were again centrifuged, the supernatant solution de-

canted, and the cell pellet resuspended in 10 ml of the

sterile sodium chloride solution. Finally, 0.1 ml of this

suspension was added to another tube containing 10 ml of the

sterile sodium chloride solution. The tube was vortexed

8

to insure complete mixing of the contents and one drop of

this suspension was used to inoculate each assay tube.

For the preparation of the L, dextranicum inoculum, the

same procedure as described above was employed. However, the

bacterial cells were grown in a previously reported enriched

amino acid medium,^ and incubated at 30° for about 18 hours.

The pellet was washed exactly as described in the E. coli pre-

paration. The final inoculum suspension was made by diluting

3 to 5 drops of the bacterial suspension into 10 ml of

enriched medium. Again, one drop of this suspension was

used to inoculate each assay tube.

Assay Procedures.— Exactly 2.5 ml of the double strength

medium were added to lipless pyrex tubes. Sterile water was

then added in such volumes that upon the later addition of

solutions of compounds to be tested, the total volume would

be 5.0 ml. The tubes, covered with aluminum caps, were then

autoclaved at a steam pressure of 15 pounds per square inch

for 7 minutes. The amino acid analogs were dissolved in

sterile water and added aseptically to the assay tubes. The

tubes were then inoculated as described earlier, shaken, and

incubated at the appropriate temperature and for the ap-

propriate length of time for the particular microorganism

(see above section). The amount of growth was determined

photometrically at 600 my. after the incubation period. The

optical density readings wera then .converted to $ inhibitions

by comparing with control tubes which contained no inhibitor.

Organic Syntheses

The chemical procedures for the organic synthesis for

all three of the pyridylalanine N-oxides are the same. The

experimental procedures below describe the synthesis of

^-pyridylalanine N-oxide. Analyses and physical constants

for all of the synthetic compounds are given in Table I.

^-Chloromethvlpvridine 1-Oxide Hydrochloride (Ila).—

^-Pyridylcarbinol 1-oxide (15.0 g, 0.12 mole), was added to

thionyl chloride (~80 ml) with stirring over a period of

about 10 minutes. After the initial exothermic reaction

had subsided, the reaction mixture was heated for an ad-

ditional 10 minutes. The clear solution was cooled in an

ice bath and overlayed with an equal volume of ligroin.

After vigorous scratching and stirring with a glass rod, the

lower layer solidified to a light brown material. The solid

was filtered off and washed several times with ligroin and

ether to remove residual thionyl chloride. After dissolving

the solid in ethanol and decolorizing with Norit A, the

product was crystallized from a 1:1 solution of ethanol-

ethylacetate to yield 8.6 g of product.

Ethvl 2-Acetamido-2-(I+-pvridvlmethvl 1-oxide)malonate

(Ilia)— To a cool solution of 12.1 g (0.056 mole) of ethyl

acetamidomalonate in 200 ml of dry ethanol containing 2.57 g

(0.112 g-atom) of sodium was added slowly with stirring 10 g

(0.062 mole) of ^-chloromethylpyridine 1-oxide hydrochloride.

After the addition was completed, the reaction mixture was

w hi

to 0 43

Cd • H nd 0)

a Jh 0

- P d

CO 0

nd

T l O

1 » 0

• H

§ H 03

H > >

• H Jh

n ,

o cd H

I O

fa

H <H 0 -H

• H

H 2 P *

£ & U hO tO O 0) ~ p p j

c d H 3 05

2 >

• d ^ O K

in 0)

& p *

o

o i s ;

H H H O O

W W w VO v O v O • • • O 0 O

O O 0 CM CM CM |25 «—I

& rH & H % % % O O O CM CM CM V0

m v O

W VO

W l r \ W LfN &

v O VO VO T— T— T—

O O O 0 O O

a

0 0 n o r—

I v O n o

H

W

o o CN O .

o o ON

I T— O

cd ,jQ M H M M

I ON o

n o n o n o O O O

CM CM CM }25 S 3 S2J

O O O V"* «r— r »

w w w 0 0 0 0 c o

0 0 0

o o n o

I r s Jt

o o i t n \TN \Ts iTs

n o ON

CM O N

VO \jrs

I l f \ t f \

v O

I J -

o 0

Td

n o

( M *

CM Jr CM

o M M

cd rQ O * 0 H M W cd H M M > H M M W

rO , 0 O J -

rQ T—

O O O

cd cd cd v O 0 0 J -•r— * — CM

• • •

O O O

0 0

T3

ITN J " CM

1

rO >

0 0

T5

O CM CM

1 n o CO

J - r -CM CM

O >

10

I J -

u 0 • p

cd £

» •H O cd

o •H - P 0 0 cd 1

•

0 • to

n— cd 1 0

ON T~" K*"*

U «r% 0

0 > r ^

1 0

W d O •H fH r& 0 S nd

cd a

0 •H u d 0 0 > i to 3 0

to X) t» 0 H

43 cd cd

d 0 o d o o *

d 0 fcuO o

• p

•H d

£ j d nd • Cu cd 2 0

- P & CO 0 4J nd

rO 0 <r» 0 d CM

to to 0 • cd cd 50 C30 > > 0 n o

}H CM 43 4J> £3 d P i 0 0 r d a > > a

rH i H • > CM O O d v— to to 0 «s

,Q •

0 0 •P cd

EH E~f 0 cd rO 0 Td

11

TABLE II

Analytical Data for Pyridylalanine

N-Oxides and Intermediates

tfo Calculated No

C H N c H N

Ila >+0.0 3.9 7.8 ^0.0 3.8 7.6 lib ^0.0 3.9 7.8 ^0.0 3.9 7 A

lie *+0.0 3.9 7.8 *+0.2 3.8 7 A

I l i a 55.5 6.2 8.6 55.5 6.5 8.7 I l l b 55.5 6.2 8.6 55.6 5.9 Q.k

I I Ic 55.5 6.2 8.6 55.5 6.3 8.8 IVa 52.7 5.5 15.^ 52.6 5.6 15.5 IVb M+.o 5.1 12.8 ^3.9 5.6 12.5 IVc 52.7 5.5 15.^ 52.7 5.7 15.6

12

heated under reflux for about ^ hours until the pH Qf an

aliquot dissolved in water had decreased to approximately

pH 5. The precipitated sodium chloride was filtered off and

the filtrate was taken to dryness in vacuo. The tan residue

was dissolved in ethanol, decolorized with Norit A, and cry-

stallized from ethanol-ether. Recrystallization yielded

5.5 g of product.

B-(*f-Pvridvl 1-oxide)-PL-alanine (IVa).— A solution of

*+.6 g (0.01 mole) of Ilia in 50 ml of 6 N hydrochloric acid

was heated under reflux for 10 hours. The solution was con-

centrated in vacuo to approximately 10 ml, 50 ml of water

was added, and the solution was taken to dryness, in vacuo.

The resulting solid was dissolved in a small volume of water

and neutralized with 10$ sodium hydroxide. The crude product

was precipitated from the cold solution by the addition of

1:1 ethanol-acetone. The amino acid was suspended in boiling

ethanol, and water was added dropwise until dissolution was

complete. The hot solution was decolorized with Darco G-60

and the pure amino acid was crystallized from water-ethanol-

acetone, yielding 1.0 g of white crystals, ninhydrin positive

(red-brown color).

13

RESULTS AND DISCUSSION

The synthesis of the pyridylalanine N-oxides, Figure 1,

was accomplished by employing the appropriate hydroxymethyl-

substituted pyridine N-oxides as starting material. Conver-

sion of the hydroxymethyl grouping to the chloromethyl

grouping was accomplished by use of thionyl chloride without

removal of the N-oxide function. The corresponding chloro-

methylpyridine N-oxides thus produced were then condensed

with sodio ethyl acetamidomalonate. Acid hydrolysis of the

resulting condensation products gave the desired pyridyl-

alanine N-oxides. Certain physical data of the intermediates

and final products are summarized in Table I and Table II.

All three of the synthesized pyridylalanine N-oxides

were tested for biological activity in E. coli 9723. Both

3-pyridylalanine N-oxide (IVb) and 2-pyridylalanine N-oxide

(IVc) were found to be only slightly inhibitory to this

organism at concentrations up to 1000 M.g/ml in the growth

medium; the 2-substituted pyridine N-oxide was somewhat more

inhibitory than the 3-substituted compound. The toxicity

due to either the 2- or 3-pyridylalanine N-oxide could be

reversed by supplements of either phenylalanine or tyrosine.

In contrast to 2- or 3-pyridylalanine N-oxide, -pyridyl-

alanine N-oxide is a fairly good inhibitor in E. coli 9723

(see Table III); complete inhibition of growth usually resulted

1*f

M

w

en

0 d d •H cd d rH 3 b rH d cd 0 rH *c| hi •H £>JI JH rQ !>>

PL, no 1 CM

J" £N On

<H O •H

H H O cd O CO

•

0 Wl >

Wl 0 «

43 a 0 a 0 H R § CO

<D d •H to 0 {* EH1

1 Q

fcuC H

0 a •H d cd H cd rH w d 0

P* I

h-3 A

o sO

O CM

vQ

O vO

eg

0 a o S3

I H K*> T5 •H t* r*> Pi I

I aX

0 d •H

H cd I

1-3 Q I 0 •H X O

d O •H -P #H ,Q

o ^ CM

VO

H

fcuO zt

O 00 Irs j -r- no j -

O O IN £N ON ON J" EN- ON ON ON

O O CM ON vO ON VO 00 00 ON ON

tr\ O en IfN lr\

O O ON 00 00 ON r- <M UN IfN

O "N \0 CN CN lA CM Jh IfN V0 ON

O ^N ON CM CM en £N IN 00 ON on

O ON O VO EN O vO ON ON ON O

O no O O no O O

O O no

O O O

d O •P O 0

CO

•P d 0 a •H fn 0

W •

O d EN •H no

T5 -P 0 cd rQ

•H w JH a 2 CO o 0 rd

V0 cd r— •H

*0 no 0 0 a -p a cd rd

-P rO rd -P 3 T* a o

d FH M O cd rQ

1?

at concentrations of 30 ng/ml and higher in the growth

medium. Although either phenylalanine or tyrosine reverse

the inhibition of the analog, as shown in Table II, the

reversal is not strictly competitive. It can be seen from

this experiment that increasing concentrations of either

phenylalanine or tyrosine tend to overcome the effect of the

inhibitor in a noncompetitive fashion. The results of in-

hibition experiments in E. coli were quite varied, and a

consistent pattern of reversal of toxicity by either phenyl-

alanine or tyrosine (or combinations of both) could not be

obtained.

Biological activities of the three analogs were also

determined in L. arabinosus 17-5 employing an amino acid

medium^ in which phenylalanine and tyrosine had been omitted

from the basal solution. The assay medium prepared therefrom

was supplemented with 10 jig/ml each of DL-phenylalanine and

PL-tyrosine to promote growth of the organism. All three of

the pyridylalanine N-oxides were only very slightly inhibitory

at concentrations up to 1000 H-g/ml of the assay medium. 3-

Pyridylalanine N-oxide was slightly more inhibitory in this

organism than either 2- or ^-pyridylalanine N-oxide.

Because of the inconsistency of both phenylalanine and

tyrosine to reverse competitively if-pyridyl alanine N-oxide

toxicity in E. coli 9723? the question arose as to whether

any enzymatic chemical modification of the pyridylalanine

N-oxide was occurring. Such a process affecting the chemical

16

nature of the inhibitor might explain the inconsistencies in

the '•>oassays, since the rate of inhibitor modification might

be affected by a variety of factors (£.£., slight variations

in incubation temperature, size and age of the inoculum, etc.).

Paper chromatographic analyses were conducted on the E.

coli assay media (after bacterial growth) which had been sup-

plemented with phenylalanine to prevent inhibition. In all

cases it was found that the pyridylalanine N-oxides tested

were almost completely converted to a ninhydrin-positive sub-

stance which exhibited a higher Rf value in acidic, basic,

and neutral chromatographic solvents. Except with very high

concentrations of either 2- or 3-pyridylalanine N-oxide, no

phenylalanine supplementation was necessary because of the

low toxicity of these compounds; however, conversion of these

compounds to ones of unknown structure resulted in all cases.

It was reasonable to assume that enzymatic reduction of

the N-oxide function of the pyridine ring might occur. If

such were the case, the enzyme(s) responsible might be expected

to act on any pyridine N-oxide regardless of the substitution

on the pyridine ring and to function as a pyridine N-oxide

reductase. As shown in Figure 2 (a paper chromatogram

tracing), certain substituted pyridine N-oxides were tested

in growing cultures of E. coli 9723 to determine whether they

could be reduced to the corresponding substituted pyridines.

It can be seen that 2-hydroxymethylpyridine N-oxide is ap-

parently converted to the 2-hydroxymethylpyridine. Also

Figure 2

Ascending Paper Chromatogram Tracing Indicating the Reduction

of Various Substituted Pyridine N-Oxides tp the

Corresponding Substituted Pyridines

by Growing Cells of E. coli 9723

2-pyridylalanine N-oxide 2-pyridylalanine N-oxide added to growth medium 3-pyridylalanine N-oxide 3-pyridylalanine N-oxide added to growth medium *+-pyridylalanine N-oxide *4~pyridylalanine N-oxide added to growth medium ^-pyridylalanine 2-hydroxymethylpyridine N-oxide 2-hydroxymethylpyridine added to growth medium 2-hydroxymethylpyridine

Those compounds added to the growth medium were incubated (1 mg/ml) with growing cells of E. coli 9723. The inor-ganic salts-glucose medium, to which the pyridylalanine N-oxides were added, was supplemented with 100 jig/ml phen-ylalanine to prevent growth inhibition. After heavy growth (approx. 16 hours), cells were removed by centri-fugation, and aliquots of the supernatants were spotted on paper. All spots were detected after chromatography in ethanol-ammonium hydroxide (19:1) by ninhydrin and/or uv light.

©

18

*

X J C

> : o »

> k ©

0 > : T 5

> i o

> C - Q

> k O

19

tested, but not shown in the tracing, were the 3- and

hydroxymethylpyridine N-oxides. Both compounds were also

apparently converted to the corresponding hydroxymethyl-

pyridines by growing cultures of the organism as determined

by paper chromatography.

In all solvent systems tested the Rf values for the

three pyridylalanine N-oxides were found to be similar; the

same is true for the enzymatic products obtained from these

N-oxides. The R^ values for a given pyridylalanine N-oxide

and its enzymatic product, however, are significantly dif-

ferent. The general correspondence between Rf values for the

pyridylalanine N-oxides and the like correspondence of Rf val-

ues for the enzymatic products might be taken to indicate that

a similar enzymatic conversion is occurring in each case.

To confirm that the pyridylalanine N-oxides were being

converted by the bacterial enzyme activity to the correspond-

ing pyridylalanines, several studies were initiated. ^-Pyri-

dylalanine was synthesized according to the procedure of Bix-

ler and Niemann12 and compared (see Figure 2) with the enzy-

matic product of ^-pyridylalanine N-oxide. It became evident

after comparing Rf values in several chromatographic solvents

that enzymatic reduction of ^-pyridylalanine N-oxide was oc-

curring, resulting in the formation of pyridylalanine. Fur-

ther substantiation of this conversion is that the compound en-

zymatic ally produced from ^-pyridylalanine N-oxide is converted

photochemically by uv light into a yellow product (as yet

20

unidentified) which has an absorption spectrum identical

with that of uv-irradiated ^-pyridylalanine, X m a x M+0 mji.

Uv irradiation of ^-pyridylalanine N-oxide has no obser-

able effect on its absorption spectrum.

Information concerning the nature of the enzymatic

products from the pyridylalanine N-oxides was also obtained

by chemical reduction studies.- The 2--and 3-pyridylalanine

N-oxides were reduced (iron-acetio acid) under conditions

in which only the N-oxide function is affected.^ The

resulting 2- and 3-pyridylalanines were compared by paper

chromatography with the corresponding enzymatic products and

were found to have identical values.

A color reaction,specific for the pyridine ring, was

employed to compare the visible absorption spectra of the

chemically reduced 2- and 3-pyridylalanine N-oxides (and the

synthetic !+-pyridyl alanine) with the absorption spectra ob-

tained from the enzymatic products of the 2-, 3-, and *+-

pyridylalanine N-oxides. Table IV summarizes the findings

of this study. These data leave little doubt that the re-

duction of pyridylalanine N-oxides (and other pyridine N-

oxides) to the corresponding pyridylalanines (or other sub-

stituted pyridines) does occur in E. coli 9723.

That reduction of the N-oxide is enzyme catalyzed is

shown by the fact that when sufficiently high concentrations

of the analogs are employed to effect complete inhibition of

bacterial growth, there is very little or no chromatographic

evidence of pyridine N-oxide reduction.

21

TABLE IV .

ABSORBANCE MAXIMA FOR SUBSTANCES TESTED EMPLOYING

A PYRIDINE RING-SPECIFIC COLOR REACTION9-

Substance Tested*-* Absorbance Maximum, mji.

Enzymatic Product of 2-Pyridylalanine N-Oxide H-85

Chemical Reduction Product of 2-Pyridyl-alanine N-Oxide *+85

Enzymatic Product of 3-Pyridylalanine N-Oxide >+85

Chemical Reduction Product of 3-Pyridyl-alanine N-Oxide *+85

Enzymatic Product of >+-Pyridyl alanine N-Oxide 60*+

^f-Pyridylalanine 606

Enzymatic Product of >+-Hydroxymethyl-pyridine N-Oxide 61*+

^-Hydroxymethylpyridine 61k-

aThe color reaction for the pyridine ring is a reported procedure employing hydrochloric acid, chloramine, cyanide, and barbituric acid as reagents. The wavelengths of max-mum absorbance were determined by scanning on a recording spectrophotometer. Pyridine N-oxides do not give the color reaction.

ID All the enzymatic products tested were from the growth media of E. coli 9723 which had been supplemented with a, substituted pyridine N-oxide. Chemical reductions of t 1b 2- and 3-pyridylalanines were conducted by use of iron and acetic acid. The ^-pyridylalanine was synthesized by the procedure of Bixler and Niemann.

22

Another question of interest was whether the toxicity

of the ^-pyridylalanine N-oxide in E. coli is due to the un-

altered compound or to its reduction product. This question

might be resolved in the studies with E. coli if the con-

version of ^-pyridylalanine N-oxide to H-pyridylalanine

could be prevented. In other experiments it was found that

the various hydroxymethylpyridine N-oxides (employed as

starting compounds in the synthesis of the pyridylalanine

N-oxides) were nontoxic to E. coli even at very high concen-

tra'.. ns. As has been shown earlier (see Figure 2) the

hydroxymethylpyridine N-oxides are reduced in growing cul-

tures of E. coli to the corresponding hydroxymethylpyridines.

It was decided to test whether the hydroxymethylpyridine

N-oxides at high concentrations could exert a sparing effect

on the reduction of ^-pyridylalanine N-oxide. In other words,

by greatly increasing the total concentration of pyridine

N-oxides available for enzymatic reduction, the rate of

reduction of the ^-pyridylalanine N-oxide might be diminished

significantly. Further, if it is ^-pyridylalanine and not

its N-oxide that is the more potent inhibitor, a diminished

rate of reduction of ^-pyridylalanine N-oxide should be re-

flected in a decreased growth inhibition in the microorganism.

This hypothesis was tested employing 3-hydroxymethyl-

pyridine N-oxide to retard the reduction of ^-pyridylalanine

N-oxide; the results are summarized in Figure 3 . Increasing

concentrations of 3-hydroxymethylpyridine N-oxide have vir-

tually no effect on the growth of E. coli 9723 in the absence

Figure 3

The Effect of 3-Hydroxymethylpyridine N-Oxide

on the Inhibition of E. coli 9723 '

by -Pyridylalanine N-Oxide

Duplicate fubes at each concentration level of 3-hydroxy-methylpyridine N-oxide were run, and optical density readings are averages of the individual readings at each level. See the experimental section and references 13 and 17 for the basal inorganic salts-glucose medium and the experimental details.

*f-Pyridylalanine N-oxide concentrations:

@ none •

O—O , 10 M-g/ml.

^ ^ j 30 ng/ml.

7 5 100 ng/ml.

2b

PLUS 4-PYRIDYLALANINE N-OXIDE

0.0 0 1 2 3 4

3-HYDROXYMETHYL PYRIDINE N-OXIDE, mg/ml

25

of"inhibitor; however, the toxicity of the inhibitor is

greatly decreased as the concentration of the hydroxymethyl-

pyridine N-oxide increases. In another experiment, Figure *+,

the effects of increasing concentrations of 3-h.ydroxymethyl-

pyridine N-oxide on the toxicity of *+-pyridylalanine to E.

coli were tested. It was found that supplements of the non-

toxic N-oxide were completely without effect on the toxicity

of ^-pyridylalanine. Inhibition studies have shown that

both phenylalanine and tyrosine prevent the toxic effect of

^-pyridylalanine in E. coli 9723*.

A logical interpretation of these findings is that the

toxicity of *+-pyr idyl alanine N-oxide to E. coli 9723 is

largely due to the conversion of this compound to the toxic

^-pyridylalanine. It is to be remembered that none of the

three pyridylalanine N-oxides that were studied are toxic to

Ii* arabinosus 17-5 at concentrations below 1,000 ng/ml.

Paper chromatographic studies of the growth medium employed

for L. arabinosus have indicated that this organism does

not catalyze any detectable reduction of substituted pyridine

N-oxides during growth to the corresponding substituted

pyridines. The failure of this reduction process to occur

may explain the poor inhibition properties of the pyridyl-

alanine N-oxides in this organism. It was found that the

reduction product, pyridyl alanine, is fairly toxic to L;

arabinosus (over 90$ inhibition at 200 M-g/ml).

Figure b

Effects of Increasing Concentrations of

3-Hydroxymethylpyridine N-Oxide on the

Toxicity of H-Pyridylalanine

to E. coli 9723

Duplicate_tubes at each concentration level of 3-hydroxy-methylpyridine N-oxide were run, and optical density readings are averages_of the individual readings at each level. See the experimental section and references 13 and 17 for the basal inorganic salts-glucose medium and the experimental details.

27

,0

0.8f

Q d 0.6

x t -

o (T O

0.4

0.2

0.0

PLUS 4-PYRIDYLALANINE

A

0 1 2 3 3-HYDR0XYMETHYL

PYRIDINE N-OXIDE, mg/ml

4

28

It is tempting to speculate on the possible use of the

pyridylalanine N-oxides for the selective inhibition of those

microorganisms which have a pyridine N-oxide reductase. It

may well be that H-pyridylalanine ft-oxide could be employed

as a reagent to determine the presence of such an enzyme

activity in a given microorganism (based on the ability of

the compound to inhibit growth of the organism). However, a

simple colorimetric p r o c e d u r e 1 6 is more suitable for determin-

ing pyridine N-oxide reductase activity in whole cell sus-

pensions of a given microorganism. ^-Hydroxymethylpyridine

N-oxide has been employed as substrate; the enzymatically

produced ^-hydroxymethylpyridine may then be determined

spectrophotometryally.

29

SUMMARY

The three pyridylalanine N-oxides, p~(2-, 3-, or ^-pyridyl

1-oxide)-DL-alanine, have been synthesized, and their bio-

logical activities in Escherichia coli 9723 and Lactobacillus

arabinosus 17-5 have been determined. ^-Pyridylalanine N-

oxldu inhabits growth of H. coli at coneentratiQns of 30 jxg/ml.

The 2- and 3-pyridylalanine N-oxides are less effective inhi-

bitors in this organism; the toxicities of all three N-oxides

are reversed by supplements of phenylalanine or tyrosine.

Evidence is given for the enzymatic reduction of the pyridyla-

lanine N-oxides to the corresponding pyridylalanines in E. coli

and for the probability that the toxicity of ^-pyridylalanine .

N-oxide in this organism is due to the formation of the more

toxic ^-pyridylalanine. The pyridylalanine N-oxides are inhi-

bitory to L. arabinosus only at very high concentrations; the

organism does not reduce these compounds to the pyridylalanines,

30

REFERENCES

1. Shive, ¥. and Skinner, C. G., "Metabolic Inhibitors. A

Comprehensive Treatise," Vol. I, Academic Press Inc., *

New York, N. Y., 1963, pp. 2-73-

2. Lansford, E. M., Jr. and Shive, ¥., Arch. Biochem.

Biophvs.. 38. 3^7 (1952).

3. Elliot, D. F., Fuller, A. T., and Harrington, C. R.,

J. Chem. Soc.. 85 (19W". J "" . .

*+. Niemann, C., Lewis, R. N., and Hays, J. T., J. Am. Chem.

Soc.. 6if, 1678 (19^2).

5. Conway, T. ¥., Lansford, E. M., Jr., and Shive, ¥.,

J. Biol. Chem.. 237. 2850 (1962).

6. Conway, T. ¥., Lansford, E. M., Jr., and Shive, ¥., Arch.

Biochem. Biouhys.. 107. 120 (196*+).

7. Moyed, H. S., J. Biol. Chem.. 236. 2261 (1961).

8. Norton, S. J., Skinner, C. G., and Shive, ¥., J. Org.

Chem., 26, 1^95 (1961).

9. Ravel, J. M. and Shive, ¥., Biochem. Bio-phvs. Res.

Commun.. 20, 352 (1965).

10. Norton, S. J. and Sanders, E., J. Med. Chem., 1_0, 961

(1967).

11. Leonard, F., Barkley, F. A., Brown, E. V., Anderson, F. E.,

and Green, D. M., Antibiot. Chemotherapy. i>, 261 (1956).

12. Bixler, R. L. and Niemann, C., J. Org. Chem., 23., 575 (1958).

31

13. Anderson, E. H., Proc. Natl. Acad. Sci. U. S., 32, 120

(19^6).

1^. Ravel, J. M., Woods, L., Pelsing, B., and Shive, ¥.,

J. Biol. Chem.. 206. 391 (1951+).

15. den Hertog, H. J. and Combe, ¥. P., Rec. Trav. Chim..

2Q, 581 (1951).

16. Asmus, E. and Garschagen, H., Z. Anal. Chem.. nq. 81

(1953); Chem. Abstr.. V7. 10 -101 (1953).

17. Dunn, F. ¥., Ravel, J. M., and Shive, ¥., J. Biol. Chem..

2H, 809 (1956).

PART II

PYRIDINE RING ANALOGS OF IODOTYROSINE

33

INTRODUCTION*

Many structural analogs of iodotyrosine and diiodo-

tyrosine as well as of the iodinated thyronines, have ap-

peared in the literature.^ Since all of these analogs con-

tain the benzene ring, it seemed worthy to attempt the

synthesis of iodotyrosine analogs containing the pyridine

ring. Effective analogs of phenylalanine, tyrosine, and

DOPA have been prepared by substituting the pyridine moiety

for the benzene moiety of these compounds Such suc-

cessful substitutions in the preparation of iodotyrosine

analogs would support the feasibility of the preparation of

analogs of the thyronines in which one or both of the aro-

matic rings of these hormones were substituted by the pyri-

dine ring. Also, due to the recent synthesis of various

N-oxides of certain natural and synthetic nitrogen hetero-Ef A

cycles, some of which have had striking biological activity,

the synthesis of iodotyrosine analogs containing the pyri-

dine N-oxide-moiety was of interest. The synthesis and

structural proofs of the compounds p-(5-hydroxy-6-iodo~2-

pyridyl 1 -oxide)-PL-alanine (VII), and {3-(5-hydroxy-6-iodo-

2-pyridyl)-DL-alanine (XI), analogs of iodotyrosine, are

described in the present study.

This work has recently been published: S. J. Norton and P. Timothy Sullivan, J. Heterocyclic Chem.. 2, &99 (1970).

3^

EXPERIMENTAL ' '

Experimental Techniques

A Thomas-Hoover capillary melting point apparatus was

employed for all melting point determinations, and the melt-

ing points reported are uncorrected. Uv spectra -were deter-

mined with a Beckman DBG recording spectrophotometer. The

nmr spectra were obtained on a Varian T-60 spectrometer.

All compounds previously unreported in.the literature were

analyzed for carbon, hydrogen, and nitrogen. Kojic acid

was obtained from Aldrich Chemical Co., Inc.

Microbiological Assays

The same general procedures and techniques for con-

ducting the microbiological assays given in Part I were em-

ployed in this study. A previously reported inorganic salts-

glucose medium was used in the assays in which E. coli 9723

served as the test organism.7

Organic Syntheses

Organic Intermediates.— 2-HydroxyTiiethyl-5-methoxy-1+H-o !

pyran-M—one, mp 157-158 , was prepared by treatment of kojic

acid with dimethyl sulfate in potassium .hydroxide solution.®

3?

The latter derivative was then reacted with concentrated o

ammonium hydroxide in a stainless steel bomb at 90 for two

hours to form 2-hydroxymethyl-5~methoxy-1+-pyridinol, mp 172-

175°.^ Reaction of this compound with phosphorous oxy-

chloride under reflux conditions yielded ^-chloro-2-chloro-

methyl-5-methoxypyridine (II), mp 72-73°»^

i+-Chloro-2-chloromethyl-I5-methoxyr)yridine 1 -Oxide

(III).-- To a solution of 50 g (0.260 mole) of II in 280 ml

of glacial acetic acid was added 50 ml of a 30$ aqueous hy-

drogen peroxide solution. The reaction mixture was heated o

with stirring in an oil bath at 70 . After three hours an

additional 30 ml of hydrogen peroxide was added, and the o

reaction mixture was kept at 70 with stirring overnight.

The solution was taken to dryness in vacuo., the residue ex-

tracted with acetone, decolorized with Norit A, and cooled

to yield a white crystalline product. Recrystallization

from acetone-ether yielded 20.0 g, mp 163-16^°. Workup

of the mother liquor gave another 12.3 g resulting in an

overall yield of 32.3 fX60%). ~

Anal Calcd for C7H7C12N02: C, UoA; H, 3.If; N, 6.7.

Found: C, ^0.5; H, 3.*+$ N, 6.>+.

Ethyl 2-Acetamido-2-(i+-chloro-5-methoxv-2-i)vridvlmethvl

1-oxide)malonate (IV).-- About 250 ml of dry ethanol was col-

lected in a 3-neck flask and was purged with dry nitrogen.-'

Sodium, 2.17 g (0.09^ g-atom), was dissolved in the ethanol

and then 20.3 g (0.09^ mole) of 1+-chloro-2-chloromethyl-5-

36

methoxypyridine 1-oxide was added and the reaction mixture

was heated under reflux for about four hours (until the pH

of an aliquot dissolved in distilled water had decreased to

approximately pll 5). The solution was concentrated to about

75 ml and then poured over ice. The resulting crystalline

material was removed by filtration, washed with cold water,

and dried to yield 19*5 8- Concentration and cooling of the

filtrate yielded an additional 9*5 g "to give an"overall yield

of 79%. Recrystallization from ethanol-water afforded white o

needles, mp 156-158 ; uv Xmax (water) 269 mM-.

Anal Calcd for C16H21C1N207: C, H, 5^; N, 7.2.

Found: C, ^9.2; H, 5-5; N, 7-3.

g-(5-Hydroxy-2-pyridyl 1-oxide)-PL-alanine (VI).— Com-

pound IV, *+.0 g (0.01 mole) was dissolved in 100 ml of 25%

hydrochloric acid in a 125 ml flask, placed in a stainless o

steel bomb, and heated in an oven at 160 for four hours.

The light yellow solution was then taken to dryness in vacuo

and the residue dissolved in water.* About 200 mg of pal-

ladium black was added to the aqueous solution and the mix-

ture was treated with hydrogen under three atmospheres of

pressure for four hours. The catalyst was filtered off and

the solution evaporated to dryness in vacuo. The solid was

dissolved in water and again taken to dryness. The residue

>jc * f

Compound V was not usually isolated; however, in one instance the compound was isolated forjise in nmr spectral studies. The product melted at 222-225 dec., and the carbon, hydro-gen, and nitrogen analysis was in agreement with the molec-ular formula CgH^Cl^Oi^-l-B^O.

37

was redissolved in water and neutralized with Amberlite IR-1+5»

The resin was removed by filtration and the filtrate concen-

trated to a smaller volume. Upon addition of acetone, fol-

lowed by cooling, a light pink solid separated and was fil-

tered off, washed with acetone, and dried to yield 1.1 g

(5k%); mp 255-258° dec. A solution of the compound in water

gave a positive ferric chloride test and showed only one

spot on a paper chromatogram developed with ninhydrin reagent.

Anal Calcd for CqH-jo^Ol,.: C, 1+8.5; H, 5.1; N, 1^.2.

Found: C, ^8.5; H, 5-0; N, 1*+.2.

B-(5-Hydroxy-6-iodo-2-uvridyl 1-oxide)-DL-alanine (VII).-

Compound VI, 0.396 g (0.003 mole), was dissolved in ^0 ml of

concentrated ammonium hydroxide, and b ml of a 1 molar iodine

solution was added at such a rate that the solution was de-

colorized between each successive addition of the iodine

solution. The complete addition required about three hours,

and the reaction mixture was left stirring for several hours.

The solution was then taken to dryness in vacuo, and the

residue was dissolved in a small volume of water. Acetone

was added to the turbidity point, and the solution was stored

in the refrigerator. A light tan solid subsequently sepa-

rated and was filtered off, washed with acetone and ether,

and dried to yield 0.3 g {b6%)\ mp 2^5-250° dec. The com-

pound showed only a single ninhydrin positive spot on a ,

paper chromatogram.

Anal Calcd for CgH^I^O^.: c , 29.6; H, 2.8; N, 8.7.

Pound: C, 29-5; H, 2.9; N, 8.7.

38

Ethvl 2-Acetamido-2-(1+-chloro-5-methoxy-2--Dyridylmethyl)-

malonate (VIII).— Compound IV, 10.15 g (0.026 mole), was

dissolved in 100 ml of chloroform, and 10 ml of phosphorous

trichloride was added. The reaction mixture was heated under

reflux for one hour, and at the end of this period the dark

solution was evaporated to near dryness in vacuo. Ice was

added to the residue to decompose any residual phosphorous

trichloride. Finally, water was added to the cooled solution,

and it was heated for a few minutes. Filtration removed

some insoluble material and the filtrate was neutralized in

the cold by the addition of 30$ sodium hydroxide solution.

The resulting precipitate was recovered by filtration, washed

with cold water, and dried over phosphorous pentoxide to

yield 8.5 g (87$); mp 152-153° (Lit.3, mp 150-151°); uv

Xmax (water) 281 m|i (Lit.3, \max (water) 282 mn).

B-(*f-Chloro-5-hydroxv-2--pyridyl)-PL-alanine (IX). — Com-

pound VIII, 1+.5 g (0.012 mole), was dissolved in 100 ml of

25$ hydrochloric acid in a 125 ml flask, placed in a stain-

o

less steel bomb, and heated in an oven at 160 for four hours.

The yellow solution was taken to dryness in vacuo and the re-,

sidue dissolved in water. This solution was neutralized with

Amberlite IR-J+5 and then concentrated in vacuo. Upon cool-

ing for several hours a light pink solid formed. The solid

was filtered, washed with acetone and ether, and dried ov^r

phosphorous pentoxide to yield 0.52 g (20$); mp 260-262° dec.

The product gave a positive ferric chloride" test and a pos-

itive reaction with ninhydrin.

39

Anal Calcd for CgH^Cl^O^ 2H20: C, 38.1; H, 5.2; N, 11.1.

Found. C, 38.25 H, 5.1; N, 11.2.

B-(5-Hvdroxy-2-pvridvl)~DL-alanine (X).— Hydrogenol-

ysis of compound IX was accomplished as previously described.3

After removal of the catalyst by filtration, the solution was

neutralized with Amberlite IR-*+5. The resin was filtered

off and washed twice with hot water. The washings were com-

bined and concentrated in vacuo and left in the refrigerator

for several hours. A solid came out of solution and was re-

covered by filtration, washed with acetone, and dried over

phosphorous pentoxide to yield 1.0 g (60$); mp 2l+0-2)+1° dec;

uv Xmax (water, pH 3) 285 mn, Xmax (water, pH 12) 306 mji

(Lit.3, \max (water, pH 3) 288-289 mji, Xmax (water, pH 12)

303 mil.

Anal Calcd for 031^^203: C, 52.7; H, 5.5; N, 15.if.

Found: C, 51.1*; H, 5.^5 N, 15.1.

B-( 5-Hvdroxv-6-iodo-2-t)vridvl)-PL-alanine (XI).— Com-

pound X, 0.091 g (0.0005 mole), was dissolved in 10 ml of

concentrated ammonium hydroxide solution. One ml (0.001 mole)

of an aqueous one molar iodine solution was then added with

stirring over a period of about one hour. After the ad-

dition was completed, the reaction mixture was evaporated

to dryness in vacuo and then dissolved in water. Acetone

5jC | On the basis of paper chromatography, uv spectra, and the microbiological inhibition index (see reference 3), and also on the basis of nmr data (see Table I), it was concluded that the compound was sufficiently pure for further syn-thetic work.

to

was added to the turbidity point and the dark solution was

stored in the refrigerator. A dark material came out of

solution and was removed by filtration. Acetone was added

to 'iv nitrate, and after standing in the cold, more of the

dark, ninhydrin-negative material separated out of solution.

This solid was removed by filtration, and acetone was added

to the now light colored filtrate. After cooling for several

days a light tan solid w~a <v(- dried (ninhydrin-positive).

The solid weighed 0.050 g (32$); mp 19^-195° dec.

Anal Calcd for C8H9IF2O3: C, 31.2; H, 2.9; N, 9.1.

Found: C, 30.9; H, 3.1; N, 9-0.

2-Iodo-6-methvl-^-r)vridinol.— 6-Methyl-3-pyridinol,

5.*+5 g (0.050 mole) was reacted with iodine in a manner anal-

ogous to the iodination of 3-pyridinol reported by Schickh.11

The isolated 2-iodo-6-methyl-3-pyridinol weighed 6.8 g (58$);

mp 179-183° dec.

Anal Calcd for C5H5INO: C, 30.7; H, 2.6; N, 6.0.

Found: C, 30.9; H, 2.6; N, 5.8.

1+1

RESULTS AND DISCUSSION

The iodo amino acids, VII and XI, were synthesized

through a sequence of reactions utilizing kojic acid (5-

hydroxy-2-hydroxymethyl-*fH-pyran-*f-one) as the starting

material, and the complete sequence of reactions is shown

in Figure 1. For the preparation of VII the N-oxide func-

tion was introduced early in the reaction sequence; per-

acetic acid oxidation of lf-chloro-2-chloromethyl-5-meth-

oxypyridine (II) afforded III in good yield. Condensation

of III with sodio ethyl acetamidomalonate gave the key in-

termediate IV. Acid hydrolysis of IV followed by hydrogen-

olysis yielded (3-( 5-hydroxy-2-pyridyl 1-oxide)-DL-alanine (VI)

Removal of the N-oxide function for the preparation of

XI was smoothly accomplished by treating a portion of IV

with phosphorous trichloride in chloroform. The resulting

compound, VIII, was also subjected to acid hydrolysis fol-

lowed by hydrogenolysis to give p-(5-hydroxy-2-pyridyl)-

DL-alanine (X), which has previously been prepared as the

dihydrochloride salt.3

Compounds VI and X were iodinated in a manner analo-

gous to that reported for the iodination of tyrosine.1^ It

was found however, that two mole equivalents of iodine to i

one mole equivalent of substrate was necessary for the

complete mono-iodination (based on carbon, hydrogen, and

Figure

Sequence of Reactions Leading

HO CI

^ stepsa CH3 ! ° 0

%>ch 2CI

HQOQ

it OAc

II

^-^CHa-CH-COOH

i hh2

**2. Pd

CI H O ^

^•>CHo-CH-C00H 2 I

HC1

i NH.

VI

I2, NH^OH

HO

I'v»#v3H2-CH-COOH

NH2 i VII

V

H0]P1 I-ki^C I^v^CHp-CH-COOH . I0

NH2 NHi+OH

XI

>j< '| Condensation of III with Sodio

^3

to the lodopyridylalanines

CI CH

n-^ch2CI

III

DEAM

CH3a CI

4 IV

(C00C2Ht>) 2

h2-c-nhcoch:

pci^ ch3°]^>| (cooc2H5)2 CHC13 ^>ch2-c-nhcoch3

VIII

HC1

T 1 W - c

kCH2-CH-C00H

NH2

X

^-52. Pd

ch2-ch-cooh

nh2

IX

Diethylacetamidomalonate in NaOC2H^.

nitrogen analyses) of VI and X to yield VII and XI respec-

tively. Schickh reported that iodination of 3-pyridinol

yielded 2-iodo-3-pyridinol and 2,6-diiodo-3-pyridinol at

elevated temperatures,^ Thus,- it was anticipated that

iodination of X would yield a monoiodo derivative with the

iodine atom in the C^-position of the pyridine ring. It is

well documented that pyridine N-oxide nitrates readily at

the C^-position of the ring; little 2-nitropyridine N-oxide

is isolated J ^ Thus, due to an enhanced activation of the

(^-position for electrophilic substitutions in pyridine

N-oxides, it was not certain whether iodination of VI would

yield the 6-iodo or *+-iodo derivative. The position of the

iodine atom in compounds VII and XI was determined by nmr

spectroscopy.

Because of the low solubility of the substituted pyrl-

dylalanines (Figure 1) in water, especially the iodo amino

acids, the hydrochloride salts of these compounds were pre-

pared. The salts were found to be extremely hygroscopic and

water soluble. Table I shows the aromatic region in the nmr

spectrum of the pertinent compounds employed to establish

the position of iodine in compounds VII and XI.

Compound VII has a very symmetrical splitting pattern

in the aromatic region. There is a doublet centered at

6 7-20 (J = 9 cps) and another doublet centered at 6 7.55 >

(J = 9 cps). Due to the small difference in chemical shifts

of the two doublets, the pattern resembles a quartet. The

TABLE I

NMR Spectra

k5

Coiiijjouiid a)b

V

VII

IX

X

XI

2-iodo-6-methyl 3-pyridinolc

H a

8.27

8 A3

8.3^

5-value (rmm)

Hy

7.82

7.55d 7.20d

8.18

8.00 8.00

7.5^ 7.5^

7.00 7.00

aThe spectra of the amino acid hydrochlorides were determined m D2O solution containing 1% sodium-2, 2-dimethyl-2-sila-pentane-5-sulfonate as reference standard.

bThe alanine moietyproton chemical shifts (and integrations) for each of the amino acids above are as follows: triplet. 6 4.6 (1 proton); doublet, 6 3.7 (2 protons).

c The spectrum was obtained in DMSO-dg (TMS standard)#

dHa was assigned to doublet at 6 -7.55 (J =~9. cps). Hv was assigned to doublet at 6 5.20 (J = 9 cp-s). : -

1+6;

two doublets could only arise from mutual splitting of pro-

tons on adjacent carbon atoms, namely the C^- and expos-

itions of the pyridine ring. By comparison with V, in

which two singlets appear in the aromatic region, b 8.27 and

6 7.82, it is logical to assume that the iodine was substi-

tuted at the C^-position to yield p-(5-hydroxy-6-iodo-2-pyri-

dyl 1-oxide)-DL-alanine (VII).

for compound XI, only a single peak appears in the aro-

matic region, at 5 7 >5*+' Integration reveals that two

protons are present. A spectrum of p-(5-hydroxy-2-pyridyl)-

DL-alanine (X) indicates two peaks in the aromatic region:

a singlet at 6 8.3 +, integration for one proton, and a pos-

sibly unresolved multiplyt at 6 8.00, integration for two

protons, ft- (^-Ghloro- 17r-hvdroxv-2-T)vrldvl)-rDL-alanine (IX)

has singlets at b 8.*+3 and 6 8.18. Thus, it would appear

that due to the absence of the low field proton (Ha) in the

spectrum of X, the iodine atom occupies the Cg-position of

the pyridine ring. As a further proof for this structure an

nmr spectrum of 2-iodo-6-methyr-3-pyridinol„:in deuterated

DMSO was obtained. This compound has a single resonance in

the aromatic region, a singlet at 6 7*00 (integration for two

protons). A comparison with 6-methyl-3-pyridinol revealed

that the a-proton is absent in the corresponding iodo com-

pound which had been prepared. Single peaks were present 'in

the aromatic region of 6-methyl-3~pyridinol at 6 8.07 (inte-

gration for one proton) and 6 7.07 (integration for two

1*7

protons). Thus, compound XI is assumed to be (3-(5-hydroxy-

6-iodo-2-pyridyl)-DL-alanine.

Several of the new amino acid analogs were studied for

inhibition of bacterial growth. Compounds VI, VII, and XI

completely inhibit the growth of Escherichia coli 9723 at

concentrations of 20, 60, and 600 Jig per ml respectively.

Inhibitions by compounds VII and XI are reversed in an ap-

parently competitive <;u by tyrosine. Paper chromato-

graphic studies have indicated that no bacterial deiodina-

tion of the compounds occurs during the incubation period.

It appears therefore, that the bulky iodine atom in these

compounds does not interfere with their being recognized by

tyrosine-utilizing enzymes of the test organism. The tox-

icity of compound VI is also reversed by tyrosine, but the

reversal is apparently not competitive. It was shown in

Part I that cells of E. coli catalyze a reduction of several

pyridine N-oxides; reduction of the N-oxide function of com-

pound VI by E. coli was confirmed in the present study. It

is probable that the enzymatic reduction product of the lat-

ter compound is actually the inhibitory agent, since supple-

ments of certain noninhibitory pyridine N-oxides to the

growth medium of E. coli prevent growth inhibition by com-

pound VI.

kQ

SUMMARY

B-(5•-Hydroxy-2-pyridvl)-PL-alanine and (3-(5-hydroxy-2-

pyridyl 1-oxide-PL-alanine were prepared through a sequence

of reactions utilizing kojic acid as the starting material.

Iodlmtion of the substituted pyridylalanines yielded mono-

iodo derivatives, analogs of iuu.o tyrosine, which were, proven

by nmr spectroscopy to be P-(?-hydroxy-6-iodo-2-pyridyl)-PL-

alanine and p-( 5-hyd<roxy-6-iodo-2-pyridyl 1-oxide)-PL-alanine,

respectively. Preliminary microbiological investigations are

reported.

^9

REFERENCES

1. Barker, S. B., "Metabolic Inhibitors. A Comprehensive

Treatise," Vol. I., Academic Press Inc., New York,

N. Y. 1963, pp. 535-566.

2. Lansford, E. M., Jr. and Shive, W., Arch Biochem.

Biophvs.. 38) 3^7 f'!'\ru?) •

3. Norton, S. J., Skinner, C. G., and Shive, W., J. Org.

Chem.. 26, 1^95 (1961).

1+. Norton, S. J. and Sanders, E., *J. Med. Chem.. 10. 961

(1967).

5. Sugiura, K. and Brown, G. B., Cancer Res.. 27. 925

(1967).

6. Fujiwara, A. N., Acton, E. M., and Goodman, L., J. Het-

erocyclic Chem.. 6, 389 (1969).

7. Anderson, E. H., Proc. Natl. Acad. Sci. U. S., 32 , 120

(19^6).

8. Campbell, K. N., Ackerman, F. J., and Campbell, B. K.,

J. Org. Chem.. 15. 221 (1950) report a melting point of

157-158°.

9. Armit, J. ¥. and Nolan, T. J., J. Chem. Soc.. 3023

(1931) report a melting point of 173-175°.

10. Pitt-Rivers, R., Chem. Ind. (London), 21 (1956). ,

11. Schickh, 0. v., Binz A., and Schulz, A., Chem. Ber.. 69.

2593 (1936).

12. Ochiai," E., J. Org. Chem.. 18. 53^ (1953).

PART III

a-FLUORO- AND a-HYDROXYPYRIDYLALANINES

51

INTRODUCTION*

The replacement of a hydrogen atom with a fluorine atom

is often advantageous in the constructing of compounds dis-

playing antimetabolic properties,1"11" Certain of the fluoro-

phenylalanines and fluorotyrosines have exhibited quite

striking biological activity '*"• microbial and/or mammalian

systems.5-7 Certain other phenylalanine and tyrosine antag-

onists have been prepared by the replacement of the benzene

ring of phenylalanine or tyrosine with the pyridine ring or

the hydroxy-substituted pyridine ring.®'9 Because of pre-

vious successes with both fluorine atom and pyridine ring re-

placements, the synthesis and biological study of certain fluo-

ropyridylalanines (Figure 1) was undertaken. The compounds

synthesized were the a-fluoropyridylalanines with the alanine

substituent positioned on the ring in such a manner that pseu-

d o ortho. meta. and para isomers resulted. In addition, the

four isomeric a-hydroxypyridylalanines+ were synthesized.

It was anticipated that certain of the latter compounds,

*This work is currently being published: P. Timothy Sullivan, Cynthia B. Sullivan, and S. J, Norton, J. Med. Chem. (in press)

t It is well established fact that the tautomeric equilibrium favors the pyridone over the pyridinol structure in neutral solution (Lit.'/). However, for simplicity and for com-, parative purposes with the a-fluoro-x-pyridylalanines, the name, a-hydroxy-x-pyridylalanine, will be employed through-out the discussion rather than the more .correct name, p-(1,2-dihydro-2-oxo-x-pyridyl)-DL-alanine.

Figure

Sequence of Reactions Leading to the

Hjr*N.

CH-

dr HBF^

NaNO,

(COOC2H5)2

CH2-C-NHCOCH3

DEAM1

(5,7,8,10)

HC1

HO

CH2-CH-COOH

NH2

(12-15)

^Condensation of (1-»t) with Sodio Ethylacetamidomalonate/

a-Fluoro- and a-Hydroxypyridylalanines

53

CH-

N-bromo sue c inimide

anhyd HBr

HBr EACA,

C00C2H5

P

CH9-C-WHCOCH. I I 3

II CH

( 1 -^) (6,9,11)

Ba(0H)2

F

CHo-CH-COOH ^ I

NH0

(16-18)

Condensation of (1-^-) with Sodio Ethylacetamidocyano-acetate.

5^

especially p-(6-hydroxy-3-pyrldyl)-^-alanine (the para

isomer), would "be effective tyrosine antagonists; P-(5-

hydroxy-2-pyridyl)-DL-alanine was previously found to be a

potent tyrosine antagonist in certain microorganisms.9 All

of the amino acid analogs were studied for growth inhibition

properties in Escherichia coli 9723 > Leuconostoc dextran-

icum 8086, and Lactobacillus arabinosus 17-5.

5?

EXPERIMENTAL

Experimental Techniques

A Thomas-Hoover capillary melting point apparatus was

employed for all melting point determinations, and the melt-

ing points reported are uncorrected. Uv spectra were deter-

mined with a Beckman DBG recording spectrophotometer. All

compounds synthesized in this study were analyzed for carbon,

hydrogen, and nitrogen. The data'are presented in Tables IV

and V. The aminopicolines were obtained from Aldrich Chemical

Co., Inc. and J. T. Baker Laboratory Chemicals.

Microbiological Assays

The same general procedures and techniques for conduct-

ing the microbiological assays are given in Part I. For the

assay of Leuconostoc dextranicum 8086 the same incubation

temperatures and times as described for L. arabinosus 17-5

were used. The same amino acid medium was also employed,

except that 0.05 Hg/ml panththeine was added.

Organic Syntheses

The following reaction procedures are given for specific

compounds; compounds indicated by reference to the particular

table were prepared in like manner.

56

g-Fluoropicolines.— The appropriate aminopicoline was

diazotized as previously reported^*^^ utilizing fluoboric

acid and sodium nitrite. The boiling points agreed in all

cases with those reported above.

2-Fluoro-^-bromomethylpyridine Hvdrobromide (Table I,

— To 11.1 g (0.10 mole) of 2-fluoro-3-methylpyridine

in 300 ml of magnesium sulfate-dried carbon tetrachloride

was added 17.8 g (0.10 mole) of N-bromosuccinimide and 1 g

benzoyl peroxide as catalyst. After heating under reflux

for several hours, the succinimide was removed by filtration,

and the filtrate was concentrated in vacuo to about 50 ml.

The carbon tetrachloride solution was then washed with an

equal volume of each of the following: b% sodium hydroxide,

water, and 2% hydrobromic acid. To the carbon tetrachloride

solution ether was added to make a total volume of 150 ml

and the solution was dried over anhydrous magnesium sulfate.

The dried solution was then saturated with anhydrous hydro-o

bromic acid 0 . The precipitated salt was then rapidly

filtered by suction, washed several times with anhydrous

ether, and stored in a dessicator over phosphorus pentoxide.

The product was extremely hygroscopic and a powerful lachry-

mator. Attempts to recrystallize the product resulted in ap-

preciable decomposition; however, it was sufficiently pure

(physical constants and analyses are given in Tables I and IV)

for further synthetic work.

H

m hi

9 EH

m d)

•H a o

RQ o Ph

£

0)

*a • H U >» P*

H

£ •P <D a o a o u

m

<D p ed

• H •rJ <D

CD p

fl

CM

cd cd

H 3

l §

•%

V3 H 0)

•H H

a 0

T5

a

§

fl o .p

• H CO O ft

PS

o 125

57

fX| pq CM CM U Jh

CM fn

W PQ p0 . H I P jp 1

w w w \D \D \D

a o o

(M l r \ l f \ £S-

IN- CM IN- IfN

O CM

l£s O CM O CM r -

i r i ^ m j -O O N T-

I 00 o

CM no J -

CM no J -

> M

0) H RQ ctf

EH

a? o w

S 2 5

<r» W

•N o fH o «H

*0 0 N K**

H cd

0)

0)

cd

g o

<m

cd

o CD

i—I O a CD

.p

rO

*0 0 •p cd o

• H TJ

•H

CO H3

O ft a o o

H H

58

Ethyl 2-Acetamido-2-(2-fluoro-^-pvridvlmethYl)maTnnntft

(Table II, 5, 7, 8, 10).— To a solution of 6.51 g (0.03 mole)

of ethyl acetamidomalonate in 180 ml of magnesium-dried

ethanol containing 1.38 g (0.06 g-atom) of sodium was added

8.13 g (0.03 mole) of 2-fluoro~3-bromomethylpyridine hydro-

bromide. The reaction mixture was heated under reflux until

the pH of an aliquot dissolved in distilled water had de-

creased to approximately pH 5-6. The reaction mixture was

taken to dryness in vacuo, and the product was extracted into

ether. The product was then crystallized from ether-pet ether

and reerystallized from water. The condensation leading to

compounds 7 and 10 was carried out in the same volume (as above)

of 1:1 benzene-ethanol. For compound 7 a molar excess of ethyl

acetamidomalonate and sodium was used and the halide was added

portionwise over a period of 1 hour. Physical constants and

analyses are given in Tables II and IV.

Ethyl 2-Acetamido-2-(2-fluoro-3-pvridvlmftthvl)cvann-

acetate (Table II, 6, 9, 11).-- The same reaction procedure

was followed as for the "corresponding malonate intermediate,

5, except that ethyl acetamidocyanoacetate was employed as

condensing reagent. A crystalline product could not be ob-

tained for compound 6 so the oil was used directly in the

barium hydroxide hydrolysis. A 1:1 benzene-ethanol solvent

was used in the preparation of 11. Physical constants and

analyses are given in Table II and Table IV.

P,-(2-Hydroxy-Vpyrldyl)-PL-alanine (Table III, 12-15).--

Compound 5, 3.5 g (0.011 mole) was hydrolyzed in the presence

H W

w

EH

t o 0 4 3

cd • p

0 o cd O

o o T3 • H

• P 0 o cd

H

I f 4 3

W

' O

CO 0

4 3

cd C o

H cd

a o

• H

4 3 0 O cd

I—1 i>>

•CJ - p

P£j

0 4 3

cd

• H XS 0

e 0

4 3 &

M

n o

, W

l r % o W O

CM o o s n

2 ^ O I }2J O - O - O

I C\J

W o

1

O J «

CM n o

^ US I T N O

W o C\J O

o W o g O I 0 - 0 < W |

C\J

W o

I

I I

«

CM

cd cd

o

T 3 ncf H 0 * H

• H

ft

r O

S + J - P C I CO 0 !>> >

H O O 0 t o

«

O

£

O • H 4 3

CO

O

P k

«

O !2S

t r \ O

O J

S J fo

O N T—

w I r s T—

O

l r \ I r s

O O CM CM

£ 3 S25

O N O N

n o " ^ N O O

n o CM }25 fe

O N

n o O

n o

a fe

J -

W « W « w I r s I r s r o \ j r \ 0 0

o ~ o " o ~ o ~ o "

J - O J -

c o c m J r n i w

I s I s w

w

0 c o v O O CM CM CM n o T~* t — T—

T~ r— v— 1 0 r » 1 1

O N n o O - J - C D r— O J CM n o O r ~

r t -V"* t — T—

r o O N

I O J O N

c m a a j -

t - CM 1— t — CM t — CM

( i j w pc ; & pc; p c j CE;

^ MD t > - C O O N o

59

H

0 H r O cd

EH

cd

0 4 3

o ti

• p

o o

<4H

0 0

CO cd

*

\ £ >

0

> txO

O 4 3

>> H • P O 0

0 N !>>

H O

T 5

>>

2 r Q

• €%

T 5 0 CsJ

• s Ph H 0 cd

- P cd S > 11 0

> JH 0

<r» si - p

0 fH r*H • P <+~f

0 r&

I I 0

• H w < H

• H r% U

0 2 f k

O - P 4 3 0 O O cd

c* l i r H

• H <4 O

r O O

M H

W

EH

to 0 a §

73 H

t 3

U >>>

Ph

• a 0 43

2 - p •H - P tO

rQ 3

CO

o u V3 i ?

no

o JH o

(H

s o 0 J . ^ w w O - S 2 5

I CM

W a

I

II «

CM

n o

cd cd

«—I

1 in O

fx,

H3 X) H <h CD *H

*H S>H 2 ft

& - P - p JZ| CO CD K*& K*

h o o 0 to «

9 a

o CD

x) o

o •H 4-3

« *H CO O ft

X

o JS|

O 0 O O CM

CM CM CM w W W W »h|C\2 CM

« « % % O n o n o n o 0 0 CM

O O O 0 CM CM CM CM fe

J25 S25 #25 ON O O O O w T— Y— v— 0 0

w W {xj w 0 CO 0 0 CO 0 0

0

0 O 0 0

\o ON \ 0 0 0 Jr I r s CN vO

CM O

CM

O

ON 0 0 CM ON CM CM n o CM CM CM CM CM

•V WS <r* 0 CN. CM n o 0 ON O O r o CM n o n o

CM vO CM

CM O

CM

w co o

CM CO j -

ON O J - J -

<3j <Ej <j| <4 <jJ I I £ > I I I

> 5b ^ & 5B

^ J -vD \ D CM CM

ON ^ t \D 0 0 O n o ^ — CM 0 0 T— ITN O I tn CM CM

1 CM

1 CM

1 CM a CM CM CM

O - n o 1

I r s 1

CN-1

o o 1

CM O CM 0 0 lT\ ON

V"* IfN CM

CM CM CM CM ON V"* CM CM

0 0 J -

W W M W li) j i , 111 0 0 0 0

CM r o l r \ \ o f>- CO

60

0 H rQ cd

EH

0 0 to

w

c* o

fH 0

CH

T3 0 &i »>>

§ 0 Ph

S cd

H

1 O

<H

fH cd H 0 0 0

H

1

0 a +3 »>>

rQ • *

rO 1—1 H 0 M 1—1

~P cd 0 0 O H H

•H rQ rQ cd cd

£ EH EH •H

rQ O to T3 0 0 El . p - P p O Q O d ft •P 43 a 0 0 0 0 0 0

rH 0 0 rH 0 0

t o CO cd rQ 0

61

TABLE IV

Analytical Data for Synthetic Intermediates

% Calculated No

C H N C H N

1 26.6 2.2 5.2 27.0 2.1 5.0

2 26.6 2.2 5-2 27.0 1.9 *+.8

3 26.6 2.2 5.2 26.3 2.0 5.0

b 26.6 2.2 5.2 not analyzed

5 55.5 5.9 8.6 55.5 5.8 8.l+

6 55.9 5.1 15.1 oil, not analyzed

7 55.2 5.9 8.6 55A 6.1 8.6

8 55.2 5.9 8.6 5 .9 6.1 8A

9 55.9 5.1 15.1 55.6 5.2 1 M

10 55.2 5.9 8.6 55.1 6.1 8.6

11 55.9 5.1 15.1 55.6 b.9 15.1

62

TABLE V

Analytical Data for the Fluoro and Hydroxy

Substituted Pyridylalanines

io Calculated No

C H N C H N

12 1*8.0 6.0 1^.0 kS.b 6.3 1*f.2

13 M3.0 6.0 1*f.O MJ.1* 6.2 13.9

1^ *+8 * 0 6.0 1^.0 **7.6 6.0 13.8

15 50.3 5.8 1^.7 50.8 5.5 15.0

16 52.2 15.2 52.k 5.2 1^.9

17 52.2 ^.9 15.2 52.1 >+.9 1 5.^

18 52.2 15.2 52.3 5.2 15.5

63

of 50 ml of refluxing 6 N hydrochloric acid for 8 hours. The

solution was then concentrated to dryness in vacuo and the

residue dissolved in 100 ml of water. The warm aqueous sol-

ution was neutralized with Amberlite IR-1+5) and the resulting

filtrate was decolorized with Darco G-60 and concentrated to

a smaller volume. Acetone was added to the turbidity point

and the amino acid crystallized out in the cold. Physical

constants and analyses are reported in Table III and Table V.

B-(2-Fluoro-^-pyridyl)-PL-alanine (Table III, 16-18).—

The cyanoacetate intermediate, 6, was added to a warm slurry

of 15# barium hydroxide in water, and the reaction mixture

heated at 70° with stirring for 3 days. Periodically, during

the course of the reaction, aliquots were removed from the

reaction mixture and the uv spectra determined to insure that

hydrolysis of the fluoro substituent was not occurring. At

the completion of the reaction, insolubles were removed by

filtration and chunks of dry ice added to the filtrate until

the pH had fallen to approximately pH 7. After removing the

precipitated barium carbonate, the pH was carefully lowered

in the cold to pH if.5 by the addition of 10# sulfuric acid.

The barium sulfate was filtered off and the filtrate concen-

trated to dryness in vacuo keeping the amino acid solution

o

at 35 or less. The product was dissolved in a minimal

amount of water and acetone added to near the cloud point.

After standing in the refrigerator for several hours, the'

amino acid crystallized out of solution. Physical constants

and analyses are reported in Table III and Table V.

6k

RESULTS AND DISCUSSION

The fluoro- and hydroxypyridylalanines were synthesized

through the usual malonic ester condensation synthesis. Al-

lylic bromination with n-bromosuccinimide of the appropriate

fluoropicoline gave the desired bromomethyl derivative. This

derivative was then converted to the hydrobromide salt since

the free base was too unstable to purify. For the synthesis

of the fluoropyridylalanines, the, bromomethyl intermediates

(1-*+) were then condensed with sodio ethyl acetamidocyano-

acetate. Bradlow et al.^ reported that fluorine in the ex-

position of the pyridine ring is stable under basic conditions

but is readily hydrolyzed to the pyridone under acidic con-

ditions. Thus, It was initially anticipated that treatment

of the cyanoacetate intermediate (Table II) with refluxing

15% barium hydroxide would afford the desired fluoropyridyl-

alanine. However, under these conditions hydrolysis of the

fluoro substituent occurred as evidenced by uv spectra. It

was finally found that hydrolysis could be effected under

less vigorous conditions, leaving the fluoro substituent

intact. Thus, the desired amino acids were obtained by heat-

ing in 15% barium hydroxide at 70° for three days.

Repeated attempts to prepare 3-(2-fluoro-Lf-pyridvl)-DL-

alanine were unsuccessful. The corresponding cyanoacetate

intermediate could not be crystallized, as it formed an oil.

65

Hydrolysis of the oil did apparently give the amino acid,

since the solution was ninhydrin positive. However, workup

of the reaction mixture in the usual manner failed to yield

a pure product as judged by carbon, hydrogen, and nitrogen

analysis.

For the preparation of the hydroxypyridylalanines the

bromomethyl intermediates (1- )- were condensed-with sodio

ethyl acetamidomalonate, followed by hydrolysis in 6 N hy-

drochloric acid. Confirmation of fluoro substituent hydrol-

ysis was made by uv spectroscopy..

A summary of the biological activities in several micro-

organisms of the fluoro- and hydroxypyridylalanines is pre-

sented in Table VI. The fluoro analogs are active to some

extent in all of the microorganisms tested. On the other

hand, the hydroxy analogs are completely inactive in E. coli

9723 and L. dextranicum 8086, but do exhibit some inhibitory

properties in L. arabinosus 17-5.

It is surprising that p-(6-hydroxy-3-pyridyl)-^-alanine

(1 f) exhibits little if any inhibitory activity in the organ-

isms studied, since 3-( 5-hydroxy-2--pyrldvl)-DL-alanine is a

potent tyrosine antagonist in E. coli 9723 and L. dextran-

icum 8086.9 A plausible explanation for this finding may

be that the a-hydroxyl group of compound 11)- exists predom-

inantly as the pyridone tautomer at the physiological pH of

the growth medium. The p-hydroxyl group of p~(5-hydroxy-2-

pyridyl)-DL-alanine, however, exists as the enol tautomer

and thus more closely resembles tyrosine structurally.

66

w P3

Eh

cd tf\ i CN. p T«. O

•H W «P P •H CO

O •H P rP •H P rO •H 05 p rP cd -P £ •

O P3i P

P3i U

0 rO P 0 w H p w P o 0 a cd

•H p o vo P •H o CO •H P O rO cd p 00 •H H O •3 d <H S c H p H {>* * o

<T3 H •H rP •H •5 P -P P •5 CO £ !>> p o n p p }>> « o jxj * 0 O a H P o cd •H •

•H , { > > -P Ptf pQ W Cd Ptf O P P •r? •P o Q P •H 0 a

• O H CH i o P O cd o P o no o OJ

r** p H O-P H cd ON cd a ON § •H •H S P H p •H O CD a O

H|

O

n3 O P. a o o

O T5 O O o O vO VO o O o VO • •

VO VO vO o o A

CM

O O O vO A

O OJ O o o vO A

O O o o C\J vO

A

vO

CM n i lA VO O- CO

•

/*"N H

fctf) :±

O * O

vO

T3 0 P tn 0 p H 0 > 0 H

S3 g o 3 •H a p •H O M 0 cd CO a H 0 cd p •p P 0 p a cd •H P P •

0 o 1—1 •H \ M P \

W •H to rO n

P •H •3 o c o

vO 0

rO ,P P •H *P cd P £ O O P W P o 0 fctf) •H T3 43 cd 0 •H cd H rO •H rQ •H nd cd rP 0 > P a P

0 rP w "£ & O £ O 00 O 1 P o o tS & J-cd ,P o

H to n o CM -p cd p o •H *P •H rQ

O ON -P P o rQ <

67

Each of the fluoropyridylalanin.es is quite toxic to the

growth of L. arabinosus 17-5« The data in Table VI indicate

that in certain experiments compounds 16, p-(2-fluoro-3-

•pyridyl)-PL-alanine, and 17, p-(6-fluoro-3-pyridyl)-DL-ala-

nine, completely inhibited the growth of this organism at

concentrations as low as 0.6 jig/ml. In other more detailed

experiments (Table VII) attempts were made to reverse the in-

hibitions of 16 and 17 "by phenylalanine. Virtually complete

growth inhibition occurs at 2 (ig/ml for both 16 and 17.

Phenylalanine reverses the inhibition of both analogs in a

competitive fashion over a 10-fold range of increasing

phenylalanine concentrations. The inhibition index (ratio

of inhibitor to substrate necessary for complete inhibition

of growth) was found to be between 100 and 300 for both ana-

logs. p-Fluorophenylalanine has been reported to inhibit

growth of L. arabinosus at 16 |ig/ml and to be reversed com-

petitively by phenylalanine with an inhibition index of ap-

proximately 10. 3

It is apparent that the fluoropyridylalanines, 16 and 17,

are more inhibitory to L. arabinosus at lower concentrations

than is p-fluorophenylalanine, while the inhibition indices

of compounds 16 and 17 are higher than that exhibited by p-

fluorophenylalanine inhibition.13 This paradox may pos-

sibly be due to the actual utilization of the latter antag-

onist in protein biosynthesis, which is known to occur in

this microorganism. This could necessitate a larger

68 TABLE VII

Reversal of Fluoropyridylalanine Toxicities in Lactobacillus

arabinosus 17-5 by DL-Phenylalaninea?^

M-g/ml None 0.20 0.60 2.00

P-(6-Fluoro-3-pyri dyl)-PL-alanine (17)

% Inhibition

0 0 0 0 0

0.2 0

0.6 83 ^3

2 96 88 68

6 100 96 75 35

20 100 88 51

60 100 99 58

200 100 89

600 100

P-(2-Fluoro-3-pyridyl)-PL-alanine (16)

0 0 0 0 0

0.2 16

0.6 9lt 31 \

2 100 79 58

6 100 87 7b ^9

20 96 77 59

60 100 90 58

200 100 77

600 100

aIncubated 36 hr at 30°.

Growth media was supplemented with O.O1!- ng/ml phenylalanine.

69

relative concentration of phenylalanine to reverse growth

inhibition caused by both the blocking of phenylalanine

utilisation as well as the formation of nonfunctional pro-

tein. '^ Further studies to determine whether compounds 16

and 17 are utilized by L. arabinosus could be informative in

this regard.

The variations in the toxicities of compounds 17 and

18, p~(6-fluoro-2-pyridyl)-DL-alanine in E. coli and L. dex-

tranicum were not anticipated (Table VI). Compound 17 com-

petitively antagonizes phenylalanine in L. dextranicum, with

an inhibition index of 300, over a 30-fold range of increas-

ing phenylalanine concentration (TableVIII). There seems to

be no apparent explanation for the complete lack of toxicity

of 17 in E. coli. p-Fluorophenylalanine, isosteric to 17,

1

was reported by Bergmann to completely inhibit the growth

°£ !!• coli (ATCC 9637) at an inhibitor concentration of

60 jig/ml. On the other hand, compound 18 (fluorine meta to

the alanine side-chain) is fairly toxic to E. coli as shown

in Table IX. An inhibition index of 3>000 was demonstrated

for 18, and it was also found that tyrosine and tryptophan

reversed the inhibition to some extent.

A preliminary investigation of the activities of the

synthetic phenylalanine analogs in the phenylalanine and

tryptophan hydroxylase system of rat liver was conducted.,

The results of this study are given in the appendix to Part

III. Observations of anomolous behavior of the phenylalanine-

tryptophan hydroxylase system of rat liver led to further

70

TABLE VIII

Reversal of p-(6-Fluoro-3-pyridyl)-^-alanine Toxicity in

' mconostoc dextranicum 8086 by DL-Phenylalanineajb

p_(6-Fluoro-

3-pyridyl)-

DL-alanine (17)

|ig/ml

Sudd lenient. DL-Phenylalanine, us/ml

p_(6-Fluoro-

3-pyridyl)-

DL-alanine (17)

|ig/ml

None 0.20 %

0.60

Inhibition

2.00 6.00

0 0 0 0 0 0

2 9 9

6 37 22 16 0

20 100 81 3^ 0 11

60 100 100 80 23

200 100 36 16

600 85 65

2000 97

aIncubated 30 hr at 37 .

^Growth media described in Experimental Section.

71

TABLE IX

Reversal of p-(6-Fluoro-2-pyridyl)-DL-alanlne Toxicity in

Escherichia coli 9723 by DL-Phenylalaninea>b

p-(6-Fluoro-

2-pyridyl)-

alanine (18)

»ig/ml

Supplement. DL-Phenvlalanine« ug/ml

None 0.02 0.06 0.20 0.60

% Inhibition

0 0 0 0 0 0

0.6 0 0

2 2? 17 18 12

6 98 ^9 39 13

20 100 71 k? 35 *+1

60 95 71 if7 58

200 93 73 65

600 88 75

2000 9^

aIncubated 15 hr at 37°.

^Growth media described in Experimental Section. 1

72

enzyinatic studies of this system. The results of the latter

studies are reported in Part IV of this dissertation.

73

SUMMARY

The four isomeric a-hydroxypyridylalanines and three of

the four isomeric a-fluoropyridylalanines have been synthe-

sized. All of the synthetic amino acids have been studied

for growth inhibition properties in Escherichia coli 9723,

Leuconostoc dextranicum 8086, and Lactobacillus arabinosus

17-5. The a-hydroxypyridylalanines possess very little in-

hibitory activity, whereas certain of the a-fluoropyridyl-

alanines were found to be competitive antagonists of phenyl-

alanine .

7^

REFERENCES

1. Heidelberger, C., Chaudhuri, N. K., Dannenberg, P.,

Mooren, D., Griesbach, L., Duschinsky, R., Schnitzer,

R. J., Pleven, E., and Scheiner, J., Nature. 179. 663

(1957).

2. Miller, J. A., Miller, E. C., and Finger, G. C., Cancer

Res.. JQ, 93 (1951).

3. Bergmann, E. D., Koninkl. Ned. Akad. Wetenschap. Proc.,

57c. 108 (195*+); Chem. Abstr., ;J+8,'8325b (195*0.

b. Coulson, W. F., Wardlee, E., and Jepson, J. B., Biochim.

Biophvs. Acta. 167. 99 (1968).

5. Mitchell, H. K. and Niemann, C., J. Am. Chem. Soc.. 69.

1232 (19^7).

6. Armstrong, M. D. and Lewis, J. D., J. Biol. Chem.. 190.

*+61 (1951).

7. Saari, W. S., Williams, J., Britcher, S. F., Wolf, D. E.,

and Kuehl, F. A., Jr., J. Med. Chem., .10, 1008 (1967).

8. Niemann, C., Lewis, R. N., and Hays, J. T., J. Am. Chem.

Soc.. 6k, 1678 (19^2).

9. Norton, S. J., Skinner, C. G., and Shive, W., J. Org. Chem..

26, 1^95 (1961).

10. Miner, J. T., Hawkins, G. E., Vanderwerf, C. A., Roe,,A.,

J. M ' Chem. Soc.. 21, 1125 (19^9).

11. Roe, A., Cheek, P. H., Hawkins, G. F., J. Am. Chem. Soc..

75

21, ^152 (19^9).

12. Bradlow, H. L. and Vanderwerf, C. A., J. Org. Chem.. 1*f.

509 (19^9).

13. Atkinson, D. E., Melvin, S., and Fox, S. ¥., Arch.

Biochem. Bionhys.. . 205 (1951 )•

1*+. Johnson, J. E. and Fox, S. ¥., Biochim, Biophvs. Actsi,

28, 318 (1958).

15. Fowden, L., Neale, S., and Tristram, H., Nature, 199.

35 (1963). 16. Bergmann, E. D., Sicher, S., and Volcani, B. E., Biochem.

J., ik, 1 (1953).

17. Degraw, J. I., Cory, M., Skinner, W. A., Theisen, M. C.,

and Mitoma, C., J.'Med. Chem., 10. 6^ 0967) •

'76

APPENDIX

Summary of Inhibition of Phenylalanine Hydroxylase

by Synthetic Substituted Pyridylalanines

Analogaid Amount of Inhibition^'0

2-fluo ro-3-pyri dy1alanine +

6-fluoro-3-pyri dylalanine +

6-fluoro-2-pyridylalanine -

2-hydroxy-3-pyridylalanine +

2-hydroxy-1f-pyridylalanine. +

6-hydroxy-3-pyridylalanine +

6-hydroxy-2-pyridylalanine +

2-bromo-3-pyridylalanine -

2-bromo-1+-pyridylal anine ++

6-bromo-3-pyridylalanine +

6-bromo-3-pyridylalanine -

2-chloro-3-pyridylalanine +

2-chloro-if~pyridylal anine H—f

6-chloro-3-pyridylalanine +

6-chloro-2-pyridylalanine

3, The bromo- and chloropyridylalanines "were synthesized in a maimer similar to the fluoro- and hydroxypyridylalanines.

no inhibition, + 0-25% inhibition, ++ 26-50% inhibition.

cThe concentration of phenylalanine and analog were equal in all cases.

dLf~Fluorophenylalanine has been shown to give a 50% inhibition of phenylalanine hydroxylase at a substrate to inhibitor 2? t X O O f 1 0 .1 • .

PART IV

A Reexamination of the Phenylalanine-

Tryptophan Hydroxylase Systfem

of Rat Liver

78

INTRODUCTION

Tn the course of a preliminary study (Part III, Appen-

dix) of the effects of the halogen-substituted and hydroxyl-

substituted pyridylalanines on the hydroxylation of

phenylalanine and tryptophan by crude rat liver preparations,

unanticipated kinetic behavior of this hydroxylase system

was observed. As shown in Figure 1, the substrate response

plot was not typical of Michaelis-Menton kinetics (hyperbolic

plot); but rather a sigmoidal plot was obtained. Such sig-

moidal plots are often observed for those enzymes which

exhibit cooperativity among multiple binding (or catalytic)

sites. Because this observation had not been reported in

the literature, and because of the implications associated

with sigmoidal kinetics, an investigation of the phenomenon

was undertaken.

In 1952 the demonstration of the in vitro conversion of •1

phenylalanine to tyrosine was made by Udenfriend and Cooper.

These investigators found that liver from the following

sources had phenylalanine hydroxylase activity: rat, guinea

pig, rabbit, dog, and human. Lung, kidney, brain, and muscle

tissues from the rat lacked this enzymatic activity. Using

a "13,000 rpm supernatant" from rat liver, tyrosine formation

could be detected after incubation of the supernatant with

L-phenylalanine, DPN or TPN, and oxygen. The dependence of

o o • CM.

O O • o _

80

U J i— ZD

o o

2

•—i ID

ac

B— (T) UJ

o —1 O

O O

o 2 1 O

CO

1 <r UJ o

> *—*

o o SP"

- j

fMMJ s

^.00 8.00 16.00 24.00 32.00 tlO.OO

T R Y P T O P H A N (MILLIMQLAR)

81

the hydroxylation upon pyridine nucleotide and oxygen impli-

cated the necessity of two enzymes for the enzymatic process. p

In 1955, Mitoma was able to purify partially phenylala-

nine hydroxylase from rat liver. He obtained two protein

fractions from ammonium sulfate precipitation of a 15)000 x g

rat liver supernatant. Zero to saturation with ammonium

sulfate yielded Fraction I (a labile fraction) and kj to 60%

saturation yielded Fraction II (a relatively stable fraction),

Neither of the fractions alone incubated with phenylalanine

was catalytically active in the hydroxylation reaction, but

Fractions I and II in the presence of phenylalanine effected

the conversion to tyrosine. A requirement for ferrous ion

was also demonstrated for the system through use of the iron

chelator a,a'-dipyridyl.

Subsequent to the work of Mitoma, Kaufman found that

sheep liver homogenates could replace Fraction II of rat

liver.3 Further, he obtained an approximately 20-fold puri-

fication of each fraction. In the course of further purifi-

cation of the rat liver enzyme it was observed that another

cofactor, in addition to reduced pyridine nucleotide, was

involved in the enzymatic conversion of phenylalanine to ) 1

tyrosine. With the report that tetrahydrofolic acid could

replace the natural cofactor,-' other substituted tetrahydro-

pteridines were studied for cofactor activity. f

Kaufman found that 2-amino-6,7-dimethyl-1+-hydroxy-5,6,

7,8-tetrahydropteridine possessed cofactor activity.6

Further, he obtained evidence that neither reduced pyridine

82

nucleotide nor the sheep liver enzyme is involved in the

2-amino-6,7-dimethyl-lf-hydroxy-!?, 6,7 5 8-tetrahydropteridine

hydroxylation reaction; the initial rate of tyrosine forma-

tion with stoichiometric amounts of tetrahydropteridine is

independent of both the sheep liver enzyme and reduced pyri-

dine nucleotide. Thus he postulated that the role of the

tetrahydropteridine is that of an electron donor. The com-

plete phenylalanine hydroxylase system is shown in Figure 2.

The biosynthesis of the powerful vasoconstrictor and

neurohumor, serotonin (5-hydroxytryptamine), involves as a

preliminary step th® hydroxylation of L-tryptophan to ^-hy-

droxy tryptophan followed by decarboxylation of the 5-hydroxy-

tryptophan'' (see Figure 3). The decarboxylase enzyme has O

been found in many tissues and studied extensively. In 1961

two investigations on the hydroxylation of tryptophan in cell-

free systems were reported. Cooper and Melcer,^ using a par-

ticulate fraction of intestinal mucosa from rat or guinea

pig, observed that the reaction required ascorbic acid or

its analogs, cuprous ion, and functioned anaerobically. The

soluble fraction of the intestinal mucosal cell inhibited < the reaction. In the second report, Freedland, Wadzinski,

1 0 ~ ' ~ "

and Waisman using a 20,000 x g supernatant from rat liver

o CM

W

83

C\J

<I> u t*0

•H

fn d) >

•H

cti «

O

a 0 -p C/3 s

CO

a o •H -P cti

.H

s?

s *a

o a

•H

s

H &

CD A CU

<D . 3 EH

W - { 3 fe-w

w A — f - W <7"> oo

W W o o

m

Figure 3

The Biosynthetic Pathway of Serotonin

0g> I H ~ "

•CH2-CH-COOH

NH,

Tryp tophan

Tryptophan ^-Hydroxylase

CH2-CH-COOH

iT M 2 I H

5-Hydroxytryptophan

Aromatic Amino Acid

Decarboxylase

HO.-r •CH2-CH2-NH2

•r i H

Serotonin (5-Hydroxytryptamine)

85

demonstrated the formation of 5-hydroxyindoles employing a

relatively high concentration of tryptophan, pyridine nucleo-

t i d e , n n d o x y g ' - i t .

11

The following year Udenfriend presented evidence to

suggest that the liver enzyme which catalyzes the hydroxyla-

tion of tryptophan is in fact phenylalanine hydroxylase.

hydroxylation of both amino acids exhibits the r rne re-

quirements: Fraction I, Fraction II, pteridine'cofactor,

and reduced pyridine nucleotide. Consistent with the fore-

going hypothesis was the finding that -phenylalanine, but 1 2

not D-phenylalanine inhibits tryptophan hydroxylation.

As further evidence for the nonspecificity of phenylalanine

hydroxylase, Kaufman^ has shown that -fluoro- and ]3-chloro-

phenylalanine are also hydroxylated by the enzyme.

Since many tissues contain appreciable amounts of sero-

tonin, investigators have examined these tissues for

tryptophan hydroxylase activity. Malignant mast cells of

the mouse which contain very high levels of serotonin have

been found by Day and Green,11+ and Levine1^ to possess tryp-

tophan hydroxylase activity. The tryptophan hydroxylating

enzyme from the mast cell tumor was partially purified and

its properties determined by Sato et al.16 The enzyme activ-

ity showed complete dependence on the presence of oxygen, a

reduced pteridine, and ferrous ion. 2-Mercaptoethanol (0,05 M)

was also required for optimal, in vitro enzyme activity.

Particular attention has been devoted to hydroxylation

of tryptophan in the brain because of the postulated role of

86

serotonin as a neurohumoral agent. The observation by

Bertaccini1? in i960 on the serotonin content in brain of

totally gastroentereotomisfd rats suggested its independent

synthesis in this organ. Subsequently, several groups of

investigators have detected tryptophan hydroxylase activity

in the mitochondrial fraction of brainstem of several

animals."18-21 Using a radioassay technique, Lovenberg et

aj..have examined a number of tissues for tryptophan

hydroxylation. The results of this study are summarized in

Table I. Using the same radioassay method, it was shown that

when mitochondria-free brainstem fractions were dialyzed

overnight against 0.1 M 2-mercaptoethanol in pH 7.*+ buffer,

the enzyme activity in the supernatent fraction concentrated

by ammonium sulfate precipitation (60$), the mitochondrial

fraction contained only 30$ of the tryptophan hydroxylase

activity whereas the soluble fraction contained about 70%.

Many inconsistencies exist in the literature concerning

the requirements for the brainstem tryptophan hydroxylase.

Some w o r k e r s I ® h a v e reported the absolute necessity for

TPNH with slight stimulation upon the addition of reduced

pteridine cofactors. On the other hand, other investigators^O

observed no stimulation of activity either by TPNH or tetra-

hydropteridines. Lovenberg .et ai.,23 found that the hydrox-

ylase enzyme from a variety of tissues exhibited a nearly 1

complete dependence upon reduced pteridine cofactors with

essentially no effect due to the addition of TPNH.

cd H

W

EH

d N dp fH > O «H 0) -P ft O

m cd

0 2 to to

•H EH

O ^ •H -P

•H > O -H 0 -P ft O

CO 05

CD 2 0} to

•H EH

g S l g l s l i • • • • • • • * O O O O O O O O

V V V

cd to o 0

1

to

I cd

r d ' •P fn O 0) ft >

rCj H

to P H 0 cd H d CD .H

S - P P 0 Cd tO S r i 0)

ft-P

a s - s

a p cd

kO-P O Cd

jxjfr:

-p 05

cd d

•H -P to CD -P d .

•H

bO d •H CD -P ft CD *h

i—I Cd Cd ft0 CD CO^ d •H -P -P 2 cd cd

O r t p c ;

OCX) ItnITN OS on novo I n cm T-

l A 4 - l f N C \ I O O O O V O O O O O O O O

0 o

fH o

cd H <Dcd d CD »H •H d o ft*H d

ft«H to <*h O 2 0 - P N o 0 cd cd

S p q p S O

-p to OS a< 0

a 0 •p to d •H -P cd

to 3 0 H O

H 0

to -p d cd

•H P H TJ cd 2 fH Cd

cd fn -P rQHtQ o

rO p rQ ho bo cd cd o o

« « Q Q

^of

cd rd Ph

>

< •%

cd a to

T3 fH 0 o »l \ r ¥

to •

<1}

9 «r»

fH 0

•H

o4

0 hs

•

W

fciiO fH 0

rQ d 0 •

—s o o o

MlvO ON

• -r—

a j -O Cvi <H •s

d < 0

rM-p cdct,

EHW cd

I ft 0

0 tO & to to

87

O S _ •H -P *H P

4~>-H ^ 8 8 -P (D p,

r—I CQ <H O -P

ctfno CD CM +3 «H •

O HI Ci 05 <D 0 fctO bO-Pl o cd qjI a ph • Q <D tuogn « > P-i 2

<d.O) O m <D „ rnxl <U

-P ctJ

XI rj d

fH 0 >

O ft -Pi-q

to d cd 0 cn t ^ to o

0 ?H 0 ft U 0 pj

nd 0

d •H 0

-P O fH ft

to 4-3 d 0 a

•H 0 a ^ > O fH 0O!^fc iO ftrQ ft*H M Cd 0 !>>H

m oJ»H 0 0 to g to 0 . , rd Cd p > 0

,d <H 0 P O rd

fH 0 ft

to 0

,d o cd • 0 J -

EH mh d o to §. 2 a

d N w " ^ w g 0 ft*H O 4-> Pn ^ cd 2 W £ O *H *H tO fH 0

-P fH 0 0 a s -p R 0 tKltr\rt3 -p d o «H 0 • 0 >

O - P - H 0 Oj ~P XI

rd

to cd

d 0 >

to to H

s s s

88

Nagatsu est were the first to report the use of

mercaptoethanol to stabilize and measure tyrosine hydroxylase

activity. Since this report, most investigators working with

the tryptophan hydroxylase enzyme have used mercaptoethanol

in their investigations, although its exact role is unknown.

It may be that it functions both in stabilizing the enzyme

and cofactor. This would explain the lack of stimulation by

TPNH observed by Lovenberg et al.,23 since these investiga-

tors carried out the reactions in the presence of high con-

centrations of 2-mercaptoethanol.

Finally in the consideration of the importance of the

hydroxylation of tryptophan by the rat liver enzyme, the

hydroxylase system of brainstem or pineal gland was examined

for phenylalanine hydroxylase activity. Lovenberg, jet al.,22

using conditions optimal for tryptophan hydroxylation and a

new assay procedure of Guroff and Abramowitz,2? demonstrated

a fairly active phenylalanine hydroxylation by pineal gland.

Phenylalanine is a typical competitive inhibitor of the

pineal hydroxylase enzyme. -

An enlightening investigation of the rat liver phenyl-

alanine- tryptophan hydroxylase systems was recently reported

by Coulson, Wardle, and Jepson.26 These workers were able

to demonstrate a stimulation in the production of serotonin

using phenylalanine at a concentration reported to give a i

50% inhibition of tryptophan hydroxylase, while employing

five times the usual concentration of reduced pteridine.

89

Their contention is that the inhibition of tryptophan hydrox-

ylase by phenylalanine reported previously11 is not a com-

petition between the two substrates for a common binding sit#

on the enzyme; but rather that it is a competition by each

substrate for limiting quantities of the pteridine cofactor.

Further, these investigators have proposed a model for the

rat liver phenylalanine hydroxylase system -with separate

hydroxylating sites for benzenoid and indolic substrates.

It is also proposed that the hydroxylation at each site is

allosterically modified by binding at the other site, al-

though the evidence for the proposal is not entirely unequiv-

ocal.

In view of the findings of Coulson et al., and the other',,./-

investigators cited, and especially in light of the pre- .

viously unreported sigmoidal substrate response plot ob-

tained for tryptophan hydroxylase of rat liver (Figure 1),

a reexamination of the phenylalanine-tryptophan hydroxylase

system of rat liver was proposed. Studies were undertaken

to determine what factors or reaction conditions give rise

to the observed sigmoidal kinetics of tryptophan hydroxy-

lation by rat liver extracts. Evidence is given supporting

the conclusion that rat liver tryptophan hydroxylase is sub-

ject to homotropic and homotropic-heterotropic regulatory

controls. Finally the overall significance of these findings

is discussed in light of the presently known roles of sero-

tonin in the central nervous system.

90

EXPERIMENTAL METHODS

Instruments and Equipment.— All of the colorimetric as-

says, as well as uv spectra, were determined on a Beckman DBG

recording spectrophotometer. Paper chromatographic studies

were conducted by the ascending method on Whatman No. 1 chro-

matographic paper. The chromatograms were developed with

ninhydrin spray reagent. Homogenization of liver tissue was

conducted in a Virtis nl+5" homogenizer. An International

HR-1 refrigerated centrifuge was used for the 25,000 x g

centrifugations. For the 100,000 x g centrifugations a

Beckman Model L ultracentrifuge was used.

Abbreviations Used.— The following abbreviations have

been used throughout this study: G»HC1, guanidine hydro-

chloride; SDS, sodium dodecyl sulfate; DMPHi,., 6,7-dimethyl-

5,6,7,8-tetrahydropteridine; AHDMPHi , 2-amino-6,7-dimethyl-

l+-hydroxy-5,6,7,8-tetrahydropteridine;' DPN, diphosphopyridine

nucleotide; and finally, TCA for trichloroacetic acid

(routinely used as a 20$ w/v solution).

Chemicals.— The chemicals for this investigation were

purchased from the following sources: L-tryptophan, L-phenyl-

alanine, serotonin creatinine sulfate complex, and 6,7-

dimethyl-5,6,7,8-tetrahydropteridine*HCl, from Calbiochem;

DPN, from Nutritional Biochemical Corporation; 5-hydroxy-

tryptophan, iproniazid phosphate, tryptamine hydrochloride,

91

and DPNH from Sigma Chemical Company; guanidine hydrochloride,

from Reasearch Plus; sodium dodecyl sulfate from Sargent

Chemical Co., 1 -nitroso-2-naphthol, and .-dimethylamino-

benzaldehyde from Matheson, Coleman, and Bell. The 2-amino-

6,7-dimethyl-lf-hydroxy-5)6,7>8-tetrahydropteridine hydro-

chloride was prepared by catalytic hydrogenation^ of 2-

amino-S^-dimethyl-^-hydroxypteridine, obtained from Aldrich

Chemical Company.

Preparation of Rat Liver Homogenates.— Male, random bred

Sprague-Dawley rats weighing from 250 to 300 g were used in

this investigation. All operations were conducted either in o

a 5 cold room or in ice baths. The standard buffer rou-

tinely used was 0.075 M potassium phosphate, pH 7.!+, con-

taining glycerol (10$ by volume).

The animal was sacrificed by a blow to the head, the

liver removed immediately and placed in a cold saline (0.85$

sodium chloride) solution. The liver was then cut into

small slices, weighed in saline, and added to 3 volumes of

standard buffer. Homogenization of the liver slices was

then conducted in a Virtis ,,lf5" homogenizer at medium speed

for about 30 seconds. Finally the homogenate was centri-

fuged at 0° at 25,000 x g for one hour. At the end of this

period the supernatant was drawn off with a disposable pipet

taking care not to remove the lipid floating on the surface.

The crude enzyme preparations were always stored in small

flasks at -*+0° . Usually the homogenate was divided into

92

several smaller portions for storage so that a given pre-

paration would not be exposed to excessive freezing and

thawing. The honogenatea were generally dlalys#a fos? on©

to two hours against about 30 volumes of standard buffer.

In those cases where a microsome-free supernatant was

desired, the homogenate was centrifuged at about 100,000 x

g for 60 minutes.

In those reactions in which partially purified enzymes

were used, Fraction I (the hydroxylase enzyme) from rat liver

2

was prepared by ammonium sulfate fractionation. The ac-

tivity is found in the 33-^5% fraction. Sheep liver ho-

mogenate (prepared analogously to the crude rat liver enzyme)

was substituted for rat liver Fraction II (the pteridine-

reducing enzyme). The sheep liver homogenate has been

shown to be virtually free of Fraction I.^

Enzymatic Reaction Procedures.— For the hydroxylation

of tryptophan the reaction mixtures routinely contained:

DMPHij., 0.7 mM (or AHDMPH^, 1.0 mM); DPN, 1.3 mM; iproniazid

phosphate, 10 mM; L-tryptophan, 2 to 2k mM; and a rate-

limiting amount of enzyme (0.6 to 0.75 ml) preparation in

a total volume of 1.25 ml. The pH of the reaction mixtures

was 7.1+. For the hydroxylation of phenylalanine the com-

ponents of the reaction mixtures were identical to the above

except the concentration of L-phenylalanine was varied frpm

0.1 to 0.8 mM, The reaction was initiated by the addition

of the enzyme preparation to the other reaction components.

93

The reaction mixtures were incubated in small scintillation

counting vials placed in a 13 x 16 cm cylindrical container

fitted with a gas-tight lid. The container was purged with

oxygen for two minutes and shaken on a metabolic shaker at

low speed in a controlled temperature incubator room for one o , o

hour. The incubation temperatures ranged from 30 to 4-3 .

The enzymatic reaction was terminated by the addition of

0.5 ml of 20% TCA.

Golorimetric Assays.— The colorimetric assays used for

the determination of enzymatic products, tyrosine and hy-

droxyindoles, are modifications of published procedures 29

The deproteinized solution was transferred to a small test

tube and centrifuged at about 2,000 rpm for 5 to 8 minutes.

The supernatant was transferred to another small test tube

and 1.0 ml removed by pipet and placed in a large pyrex

tube. For the phenylalanine hydroxylations, 1.0 ml of a

1-nitroso-2-naphthol solution followed by 1.0 ml of the

nitric acid-sodium nitrite reagent were added to the large

pyrex tube. The contents of the tubes were shaken and main-O

tained in a water bath at 55 for 30 minutes. The same pro-

cedure was followed for the tryptophan hydroxylations except

that a sulfuric acid-sodium nitrite reagent was employed

in lieu of the nitric acid-sodium nitrite reagent, and the

colorimetric reaction was allowed to develop for only a few

minutes at room temperature. In both cases at the completion

of the colorimetric reaction, 10 ml of ethylene dichloride

9If

was added to each tube and the tube vortexed 30 seconds to

extract any unreacted nitrosonaphthol. The aqueous layer

was removed by a disposable plpet and transferred to a

small test tube and centrifuged at 2,000 rpm several minutes

to clarify the solution,. The absorbances were then deter-

mined (tyrosine at k^O mjx, hydroxyindoles at 5^0 mji).

For the assays in which total indole contents were de-

termined, the reaction of £-dimethylaminobenzaldehyde with

the indole ring was employed.30 To 0.9 ml of the deprotein-

ized supernatant was added 0.1 ml 10 N sodium hydroxide.

This solution was then added to 9»0 ml of 21 A N sulfuric

acid solution containing 30 mg of jD-dimethylaminobenzaldehyde.

The solution was stoppered, shaken, and allowed to stand at

room temperature for one hour in the dark. The color was

developed by the addition of 0.1 ml of a 0.0*+$ sodium nitrite

solution. After 20 to 30 minutes the absorbances were de-

termined at 600 m|i. All absorbancy readings were converted

to mumoles of product (tyrosine, hydroxyindole, indole) by

use of standard curves.

95

RESULTS

Effects of Phenylalanine and Other Epinephrine Pre-

cursors.— In view of the stimulation of tryptophan hydrox-

26

ylase activity by phenylalanine as noted by Coulson ejb al.,

it seemed worthy to determine if phenylalanine had any

effects on the kinetics of tryptophan hydroxylation under

conditions normally giving rise to sigmoidal response plots.

Figure k shows the results of such an experiment. A phenyl-

alanine level was chosen which had given approximately 50$

inhibition of tryptophan hydroxylase employing 0.7 mM DMPH^

in the reaction mixture$ but in this experiment the DMPH^

concentration was increased to 3.5 mM. Thus under these

reaction conditions, O A mM phenylalanine and 3*5 mM DMPH^,

a nearly linear plot results. It is apparent that the con-

centration of DMPH^ is significant in determining whether

phenylalanine is inhibitory or stimulatory to rat liver tryp-

tophan hydroxylase.

It was thought that other precursors of epinephrine, or

even epinephrine itself, might act as positive effectors,

the action of which would be reflected in the tryptophan re-

sponse plots. Various concentrations of either DOPA, nor-

epinephrine, or epinephrine were added to the reaction ,

mixture employing relatively high levels of DMPH^. No stim-

ulation was observed for any of these compounds, but inhibi-

tion did occur when the concentration of DMPH^ was fairly low.

Figure b

Effect of Phenylalanine on the

Hydroxylation of Tryptophan _

Reaction velocity is expressed in terms of mMmoles of 5-hydroxyindole formed/minute. The usual procedures and incubation mixtures were employed as described in Exper-imental Methods. The reaction mixtures were incubated at 37 for 1 hour.

© — © , hydroxylation of tryptophan.

H—• , hydroxylation of tryptophan in the presence of phenylalanine (OA mM), and DMPE ' (3.5 mM).

97

IxJ

tn UJ

LJ Q o 31

^ o cc LU °

4.00 8.00 12.00

TRYPT OPHRN (MILL I MOLAR)

16.00 20.00

98

Effects of Guanidine Hydrochloride and Sodium Dodecyl

Sulfate.— Several reagents are known to modify enzyme coop-

erativity by altering the tertiary and/or quaternary structure!

guanidine hydrochloride and sodium dodecyl sulfate have been

shown to effect such enzyme modifications.-^ >32 Figure 5

shows the results of pretreatment of the enzyme preparation

with G-HC1 at a final concentration of 0.25M, Analogously,

pretreatment of the crude preparation with SDS (O.yfa concen-

tration, w/v) gives a hyperbolic substrate response plot

(Figure 6)

A question arose as to whether the concentration of

DMPH^ would have an effect upon the G»HCl-treated enzyme

preparation. An answer for this question is given in Figure

7, where it is apparent that the DMPH^-concentration is very

important in determining the kinetics of the reaction. Pre-

treatment of the crude preparation with G«HC1 results in a

hyperbolic substrate response plot when the concentration of

DMPH^ in the assay is 0.7 mM. However, when the DMPH^ con-

centration is 3.5 mM the G*HCl-treated preparation exhibits

an even more sigmoidal plot that the non-treated preparation

(low DMPH^). The interpretation of these findings is not

presently clear. It may be that the function of the reduced

pteridine is not only that of a cosubstrate, but also that

the pteridine aids in maintaining the protein in a conforma-

tion such that enzyme-substrate cooperativity is operative.

The G*HC1 effect may be to promote a disaggregation of the

Figure 5

Effect of Enzyme Pretreatment with Guanidine Hydrochloride

on the Hydroxylation of Tryptophan

Reaction velocity is expressed in terms of mumoles of 5-hydroxyindole formed/minute. The usual procedures and incubation mixtures were employed as described in Exper-imental Methods. The reaction mixtures were incubated at 36 for 1.hour. Thirty minutes prior to the addition of the enzymes to the reaction mixtures, the enzyme preparation was divi-nwni eclual aliquots. To one aliquot sufficient G-HC1 was added such that the final concentration of 1? was 0.25 M. Each of the two aliquots was then stirred for 30 minutes at 5 .

5 hydroxylation of tryptophan by untreated enzyme.

' tion of tryptophan by enzyme pretreated with G«HC1 (O.JM),

S—g

100

o o • C\J_

o o

IxJ o o

Z »—§ oo

>— X • \ 1— 01 1 -1 LlJ o CJ

1 O

o

O x: o

(D

•mmmmJI QC LLJ O

> •—i

M O O sr"

I—! w

5 .00 10.00 IS .00 20 .00 25 .00

T R Y P T O P H A N (MILLIMQLflR)

Figure 6

Effect of Enzyme Pretreatment -with

Sodium Dodecyl Sulfate on the

Hydroxylation of Tryptophan

Reaction velocity is expressed in terms of m^moles of 5-hydroxyindole formed/minute. The usual procedures and incubation mixtures were employed as described in Exper-imental Methods. The reaction mixtures were incubated at 30° for 1 hour. Thirty minutes prior to the enzyme addition to the reac-tion mixtures, the enzyme preparation was divided into two equal aliquots. To one aliquot sufficient SDS was added such that the final concentration of SDS was 0., Each of the two aliquots was then stirred for 30 min-utes at 5°.

^—0 , hydroxylation of tryptophan by untreated en-zyme.

B—• , hydroxylation of tryptophan by enzyme pre-treated with SDS (0.3$).

102

LU 1 —

3 o 3* Z • -

•—< cu >-1— s:

N in

I—1 CJ

LLJ -J O

o 00

ED MMMMJ

s o 0£

»"#

LLJ o 1MB*

> 1 1 s o

<M 1—I

-J _J •—«

£

2.00 1.00 6.00 8.00 10.00

TRTPTOPHFIN (MILLIM0LRR)

Figure 7

Effects of Concentration of Reduced Pteridine on the

Hydroxy!ation of Tryptophan by Nontreated and

Guanidine Hydrochloride Pretreated Enzyme

Reaction velocity is expressed in terms of mjJmoles of 5-hydroxyindole formed/minute. The usual procedures and incubation mixtures were employed as described in Exper-imental Methods. The reaction mixtures were incubated at 360 for 1 hour. The enzyme pretreatment by G*HC1 was conducted as described in the legend for Figure 5.

Q 9 , hydroxylation of tryptophan by untreated enzyme employing 0.7 mM DMPH^ in the reaction mixture.

• h , hydroxylation of tryptophan by enzyme pretreated with G«HC1 (0.25 M) employing 0.7 mM DMPH^ in the reaction mixture.

ib ± , hydroxylation of tryptophan by untreated enzyme employing 3.5 mM DMPH^ in the reaction mixture.

6 O , hydroxylation of tryptophan by enzyme pretreated with G-HC1 (0.25 M) employing 3.5 mM DMPHi^ in the reaction mixture.

10

L L J °

^.00 5.00 10.00 15.00 20.00 25.00

T R Y P T O P H A N ( M I L L I M O L R R )

105

native enzyme into subunits with concomitant loss of homo-

tropic cooperativity. It was subsequently shown that G*HC1-

treatment does not give a permanent effect. G*HC1-treated

enzyme was dialyzed to remove the G«HC1, assayed, and the

resulting tryptophan response plot was again sigmoidal.

In conjunction with the tryptophan hydroxylase experi-

ments, reactions were run in which phenylalanine served as

the substrate. A typical response plot for the'phenylalanine

reaction is seen in Figure 8. It is interesting that the

plot lacks the sigmoidicity of the tryptophan system, and

that G*HC1 exerts almost no effect on the kinetics of the

system. It might also be pointed out that phenylalanine

hydroxylation by the crude enzyme preparations does not re-

quire the addition of reduced pteridine to the reaction

mixture.

Temperature Effects.— A striking temperature phenomenon

is observed in the tryptophan hydroxylating system in the

temperature range of 30° to ^3°. Figure 9 shows the rather

dramatic variation in reaction kinetics with varying temper-

ature. An attempt was made to change irreversibly the

system by pretreatment of the enzyme at a relatively high

temperature for several minutes, then performing the usual

reaction at 30°. A considerable stimulation of hydroxylase

activity results at low tryptophan concentrations upon prp-

treatment of the enzyme preparation at *+7° for short incu-

bation periods, and a definite hyperbolic curve results

(Figure 10). Accompanying this irreversible change in

Figure 8

Phenylalanine Hydroxylase—Effect of Pretreatment of

Enzyme with. Guanidine Hydrochloride

on Hydroxylation of Phenylalanine

Reaction velocity is expressed in terms of mMmoles of tyrosine formed/minute. The usual procedures and in-cubation mixtures were employed as described in Experi-mental Methods. The reaction mixtures were incubated at 36° for 1 hour. The enzyme pretreatment by G»HC1 was conducted as de-scribed in the legend for Figure 5.

9 — o , hydroxylation of phenylalanine by untreated enzyme. '

a—• , hydroxylation of phenylalanine by enzyme pre-treated with G*HC1 (0.5 M).

107

o o

tjJ t— 3 Z

CO Ixi — J •

CJ O C\J O § -J cc UJ <-> M — M

> Z

.00 0.20 0.U0 0.60 0.80 1.00

PHENYLALANINE (MILLIM0LRR)

Figure 9

i

Effect of Incubation Temperature on

Hydroxylation of Tryptophan

Reaction velocity is expressed in terms of mMmoles of 5-hydroxyindole formed/minute. The usual procedures and incubation mixtures were employed as described in Experimental Methods. The reaction mixtures were in-cubated 1 hour at each of the reaction temperatures.

0 — • , hydroxylation of tryptophan at 30°.

±—* ? hydroxylation of tryptophan at 37°.

0—e , hydroxylation of tryptophan at ^3°.

o o *

CM_

O o

109

LU P

15.00 20.00 25 on

TRTPTQPHfiN (MILLlMOLflR)

Figure 10

Effect of Enzyme Pretreatment by Heat

on the Hydroxylation of Tryptophan

Reaction velocity is expressed in terms of mMmoles of 5-hydroxyindoles formed/minute. The usual procedures and incubation mixtures were employed as described in Exper-imental Methods. The reaction mixtures were incubated at 30° for 1 hour. The enzyme preparation was divided into three equal ali-quots. One of the aliquots was incubated 5 minutes at 47°, a second aliquot was incubated 8 minutes at b7 , and the third aliquot was maintained at 0 .

, hydroxylation of tryptophan by -untreated enzyme,

5 hydroxylation of tryptophan by enzyme pretreated at k7° for 8 minutes.

• • , hydroxylation of tryptophan by enzyme pretreated at ^7° for 5 minutes.

111

o o

.00 1.20 2.40 3.60

TRYPTOPHAN ( M I L L I M O L f l R )

U.80 6.00

112

reaction kinetics is a significant hydroxylase destruction,

since the maximum velocities are considerably lowered by

the high enzyme pretreatment temperature.

Alcohol Effects.— The effects of~ethanol and 1-propanol

on the hydroxylation kinetics of rat liver tryptophan hydrox-

ylase were studied. Of the two alcohols, propanol exerted

the more profound effect upon the system. Typical Michai-

lis-Menton kinetics are observed when propanol is included

in the reaction mixture at a final concentration of 0.5 M

(Figure 11).

Initially it was thought that possibly the alcohols were

serving to maintain a high concentration of the reduced pyri-

dine nucleotide, DPNH, which would in conjunction with the

pteridine reductase enzyme, maintain a high level of reduced

pteridine. However upon adding DPNH to the reaction mixture,

at a concentration five times that of the DPN concentration

routinely added, no change in the sigmoidal response curve

was observed. In another study DPNH was incubated with a

rat liver enzyme preparation and disappearance of absor-

bance at 3 *f0: m|-i followed. The rate of conversion of the DPNH

to DPN, however, was not sufficiently rapid to account for

the lack of stimulation of added DPNH in the hydroxylation

reactions.

In view of the autocatalytic oxidation of the tetra- ,

hydropteridines to the dihydropteridines reported by Kaufman,6

it seemed possible that the effect of either alcohol might

be manifested in preventing or retarding this oxidation.

Figure 11

Effect of 1-Propanol on the Hydroxylation

of Tryptophan

Reaction velocity is expressed in terms of m^moles of 5-hydroxyindole formed/minute. The usual procedures and incubation mixtures were employed as described in Experimental Methods. The reaction mixtures were in-cubated at 30° for 1 hour.

9—9 , hydroxylation of tryptophan.

g—a , hydroxylation of tryptophan in the presence of 1-propanol (0.5 M).

111+

o o

LlJ °

.00 U.00 8.00 12.00

TRYPTOPHAN CMILLIMOLRRJ

16.00 30.00

115

However, uv spectral analysis indicated that ethanol or

propanol neither prevented nor retarded the non-enzymatic

oxidation of the tetrahydropteridine.

In light of the better stimulatory effect of propanol

over ethanol, a variety of alcohols were examined. Figure 12

shows the results obtained when each alcohol was added to

the reaction mixture at a concentration of b% of the total

volume. Since a structure-activity trend was apparent em-

ploying the primary alcohols (Figure 12), other alcohols

related in structure were studied. It is interesting that

2-butanol (Figure 13) is almost as effective as 1-propanol

in perturbing the sigmoidal kinetics of the hydroxylation

system. However, without a more extensive investigation of

this phenomenon, it is difficult to find a rationale for

this observation. Finally, phenylalanine hydroxylation was

studied in the presence of 1-propanol; there was no observa-

ble effect (Figure 1*+).

Studies on the Nature and Possible Cause of the Observed

Sigmoidal Kinetics.— An examination of the factors and/or

conditions, which might give rise to the sigmoidal plot for

tryptophan hydroxylation (Figure 1) was made. The sigmoid

curve may be the result of any of the following: limitations

of the colorimetric assay for product formation, competition

of other tryptophan-utilizing-enzymes for substrate, destruc-

tion or loss of some of the product formed, nonlinear multi-

substrate mechanisms, or cooperative (allosteric) effects.

Figure 12

Effect of Various Primary Alcohols on the

Hydroxylation of Tryptophan

Reaction velocity is expressed in terms of mumoles of 5-hydroxyindole formed/minute. The usual procedures and incubation mixtures were employed as described in Exper-imental Methods. The reaction mixtures were incubated at 36 for 1 hour.

X—* , hydroxylation of tryptophan.

, hydroxylation of tryptophan in the presence of methanol (k% v/v).

, hydroxylation of tryptophan in the presence of ethanol (b% v/v).

B—a > hydroxylation of tryptophan in the presence of 1-propanol (b% v/v).

hydroxylation of tryptophan in the presence of 1-butanol v/v).

117

o o

W VSrnJ • ,

O S 4 0

L±J °

^ . 0 0 1.20 2.40 3.60 4.80 6.00

T R Y P T O P H A N (MILLIM0LAR)

Figure 13

Effect of Certain Alcohols on the

Hydroxylation of Tryptophan

Reaction velocity is expressed in terms of mjimoles of 5-hydroxyindole formed/minute. The usual procedures and incubation mixtures were employed as described in Exper-imental Methods. The reaction mixtures were incubated at 36° for 1 hour. ___ _ -

, hydroxylation of tryptophan in the presence of t-butyl alcohol (*+% v/v).

, hydroxylation of tryptophan in the presence of isobutyl alcohol iS% v/v).

, hydroxylation of tryptophan in the presence of isopropyl alcohol v/v). .

g g , hydroxylation of tryptophan in the presence of sec-butyl alcohol (b% v/v).

119

o CO

LLJ i— O ZD CM Z

mrnrn

»—i CO > - n

\ h- CO 1—1 CJ

LlJ —1 o 1—1 CJ O • *

o £ o CM 1 01

LU O «—i > £ O CD

»—<

_J -J * *

s: w

^.00 1.20 2.U0 3.60 4.80 6.00

TRYPTOPHAN (MILLIMOLflR)

Figure 1*+

Effect of 1-Propanol on the Hydroxylation

of Phenylalanine

Reaction velocity is expressed in terms of miomoles of tyrosine formed/minute. The usual procedures and incu-bation mixtures were employed as described in Experi-mental Methods. The reaction mixtures were incubated at 30° for 1 hour.

o © , hydroxylation of phenylalanine.

0—• , hydroxylation of phenylalanine in the presence of 1-propanol (O.f M).

121

o o

txJ I— o

o 3 •

2 CD. H—t

>-h-

s \ CD

i—i UJ i

o o

CJ mmmS

o OJ. O

1

z: o QZ

* - •

LU O •—i

> 2Z o o

i—i 9 mm

-J CD —1 i—i 5

^ . 0 0 0 .20 0 .40 0 .60

P H E N Y L R L R N I N E (MILLIM0LRR)

0.80 1.00

122

These possible explanations were considered in devising

experiments to ascertain causes of the sigmoidal response

curve. The velocity of the hydroxylation reaction was mea-

sured by the amount of 5-h.ydroxyindole formed. Since a

standard curve employing 5-hydroxytryptophan as substrate

for the nitrosonaphthol reaction was linear, the possibility

of the colorimetric assay being insensitive at low 5-hydroxy-

indole concentration was eliminated.

Competition of tryptophan hydroxylase with other trypto-

phan-utilizing-emzymes was examined next. A significant

decarboxylation of tryptophan could give rise to the sig-

moidal response plot under appropriate conditions. If the

decarboxylase enzyme were to become saturated at relatively

low concentrations of tryptophan, appreciable hydroxylation'

could only occur at higher tryptophan concentrations. Since

it has been shown that'tryptamine does" not serve as a pre-

cursor to serotonin,33 the argument appears reasonably valid.

However, it is shown in Table II that the total indole con-

centration, after extracting indolealkylamines (tryptamine

and/or serotonin) from the one hour reaction mixture, remains

nearly constant. Generally about a 2% conversion of trypto-

phan to serotonin is achieved by the rat liver preparation.

That propanol had no effect in the postulated decarboxylation

is also seen in Table II. f

Paper chromatographic studies (Table III) also indicate

that no appreciable decarboxylation of tryptophan occurs.

1 2 3

H I—I

m § EH

I hJO a

#r-j

N •H r—I •H -P . P

I

9 p . O -p. £ U

&H

<4H O

CD O

a) CO 0 u

cu

0 , d 43

a o

to 0

•H <V £

-P CO

d o

»H 43 05 U cd P i 0 u

pL4

0 >

•H

-P cd

PS

0 A •P

d •H

CO 0

fcsl

a 0 o

H «H O -P

^ cd d

M -P a 0 O a o

o

0 a

•H ***> tO Sh

d . d O s—/

#r^l - p o 05 O

rO r o 2 O P Pi cd

H

rH O d

cd cd P<

CO O - p

PH 0 d o fk a o PL,

o Q

d o

•H -P nd cd •H

rQ JS3 P5 cd O •H fl d

M o u

M

O !2i

CO J " I r s J - r o • • • • • •

r n no n o n o n o n o

O O

O O

na T3

n3

CM r o 4 * VO

g l O v CM

CM •

Ch - O

!2J ^ 0 « |

, # S H S o a > O H

• cd J " -P

o •N-p

cd r d * P 4 S 1 S O *H -P l r \ jpt «rs • S>»rH O fh a -P «

i l r \ cd

• O n3 CO > O H iH O

<H

CO cd

0 to

0 0

iS3 O d d 0 CD

p< •«> o 2 PH

a ft CD Fh CD O fH £ • O

o Uir- d CD O PH *~*H 2 fflp -P -P cd M cd PH

•H ^ P S Pkti

m cd d Q O O ,£} £3

r l P i O P O cdT3

^ • r l r l pi N cd o o5 d Pj -H -H

•H fl<H O

© SH © , 3 PiX3 E-l *H EH cd

<r3 CD CD T3 > T5 CD f>» O <D W f l a ,Q Cd CD CD -H & T5 PH PH ftH

O Cd C! w w ^ •H 0) CD <CS

P O W

pi Pi O CD CD rQ ^ O

PH cd O* Pi p , cd »H

rH 8 ® o o 8 . A C RD H p C3P i>>

p . 3 *- ps pj p

< j ,Q -H CD o i a EH ' txO-H

S 13 •H I H RCJ

a P Pi •H «H

ITS Is Cd • a

O t > CD 0) Ph

«M P O O f j r l

cd O P i a ph-'H a O -P P

•H X cd P CD PH •H P -T3 P T3 Pj cd cd

4 3 0 cd

rCj a •P 5h

0 ^ P l H

rQ £ O co ""d

•o a 0 0 «H

43 .JCj

ft

•H &

0

a w

0 CO a cd C o o O I H

{H 0 - p

0 a H PH O O

H r O

© P

cd P H O pj cd o

• H " C P <D a g o © ^ j PH Cd P ,C| <H

P PH ~CD O

P! ^ ! !>» W to O E H P ® cd ' H T3 ^ P T 3

CdCJN QJ CJ M M P! H O PJ * * H W

• h «h >> a Pi P *H rH PH O O PH CO CD -H cd P pi P > © P3 O © © PH © -H T3 PH

O > P I © © W

• 3 £2 '[S EH RP PR IS cd

^3

i vU

ncJ •H

cd •rH d

• o 0 f-i £ Pk

IS} d 0 0 &

43 <H o •H

<3 d o o EH

•H - P CH •H. O ro •tJ d cd o

•H 0 4*3

•H 43

no cd

0 •P 0 <4H r d cd 43

0 >> •p RQ

d 0

a 43 a cd •

Y— d 0 •H ?H

!>» a ^ »—1 ?H43 0 0 W

4 > 43 . H cd a a CO

•H Cd 0 M ^ r d O 43

d PH O O P I •rH 43 cd 43

O T5 T5 cd 0 0 0 T 3

T 3 SH^O NCJ cd cd 0

0 CO 43 SH cd 0 £

0 '<$ 4 3 JS;

o EH O T5

12b

TABLE III

Paper Chromatographic Data Comparing Rf Values of Tryptophan

Serotonin, and Tryptamine -with Compounds Produced

During a One Hour Incubation of-Tryptophan

•with Hat Liver Preparations

Substance or Reaction

Mixture Chromatographeda Rf

b>°

reaction mixture minus propanol' 0.68

reaction mixture plus propanol 0.67

tryptophan 0.67

serotonin 0.66

tryptamine 0.77

aThe reaction mixture from the 1 hour incubation described in Table II (Nos. 1 and 2), after precipitation of protein with TCA. was extracted into 1-butanol as previously de-scribed. 29 When tryptophan, serotonin, or tryptamine Rf values were determined, a 1.0 mg/1.25 ml solution was pre-pared in standard buffer and 0.5 ml TCA added. This so-lution was then extracted with butanol as above. 5-Hydroxy-tryptophan is not extractable into butanol under these con-ditions and thus was not determined.

^All R^ data were obtained employing Whatman No 1 Chromato-graphic paper. Ten overlays of the butanol extraction phase were required.

cThe solvent employed was butanol, acetic acid, water; !+: 1 :1 .

125

According to data, the butanol extractable material,

after the one hour incubation reaction, corresponds to

either substrate (tryptophan) or product (serotonin), but

not to tryptamine.

A second major tryptophan-utilizing-enzyme is tryptophan

pyrrolase which oxidizes tryptophan to N-formylkynurenine,

2-amino-lf-oxo-!+-(2-formylamidophenyl)butanoic acid. However,

the action of this enzyme can be discounted as a depleter of

the substrate, tryptophan, by the data presented in Table II.

Studies were also conducted to determine if the hydroxyl-

ation product was being removed or destroyed through further

enzyme processes. The data in Table IV, however, indicate

that the 5-hydroxyindole moiety is not catabolized by enzymes

present in the crude rat liver preparation.

Further, the possibility of protein and/or lipoprotein

tenaciously binding a portion of the 5-hydroxyindoles formed

was considered. Routinely, the reaction was terminated by

precipitating the protein with TCA, and the supernatant as-

sayed for 5-hydroxyindoles. A product adsorption onto the

protein would then result in an erroneous measure of the ex-

tent of hydroxylation. Allowing the TCA-treated reaction

mixtures to stand for 18 hours before removing the protein

in no way altered the hydroxylation kinetics normally ob-

served (Figure 1). Also, no 5-hydroxyindoles could be exT

tracted from the precipitated protein. Lastly, reactions

employing a 100,000 x g supernatant (microsome-free) as

1 2 6

w

m m & fcsl a

w

rO

Xi P4 O -P Pi j>> u •p

f? o u

13

£ ?

i A <H o a co

•H rH O

rQ cd -p cd o 0

•P

C o co 0

•H nd $ -P CD

a o •H •P cd ?H cd P< 0

u 0 >

•H

•P cd

ps

0 nD 0

fn O

0 . 3 •p

PJ *H

-P a 0 CO 0 u

Ph

I a

cd a 0 tf

05 0 «—I o nd

. 3

6 0 fH «&

1 ITS

a o

•H -P cd fQ

a a

u &

lH 0

-p Ch cd

&j0

w 0

H O

o S 3 u \ p, . o o Shv^

Ph

T5 0 cd

r& CO nd 0 cd U

fl •p cd M

.a •H Pi a •"""N O CO

•p A 0 Pi 0 rH s •H 0 5H 43 a +3 cd n h rO "W K O 0

fl nd 1—1

W 1

0 -p lr\

XTMrsDvOO O i O O O O O l

\ 0 J" tN » t~ r— C\j CNJ

o o o o o o o o o o

I + ! + I + I +

vD\O ON ON OJ CM OD CO O O O O O O r - r r r r O r O • • • • • • • • • • o o o o o o o o o o

c* • 0

0 rH »£j •p s - p cd

CMM rO CO * O r - T J rd 0 Pk*H 43

O cd nd •H

£S3 cd

•H a o u Pi H

•H cd -P

0

rH o >

<H O

0 O C 0 CO 0 U Pi*H 43

fl H a

0 •p

0 1H 0 £

CO 0 CO CO 0 a O U Pi

O cd *h Jh -P cd cd P < 3 0 U N Pi £

0 !h 0 r> > o •H O

? H r O a

-p cd

0 ;d •p

•H

Td 0 43 Cd

rQ

4> cd

CO cd 0 £ rd

rCj P-i rH o a -p

>>o-U •

-P O

g s nd

J?! t

ITS

C3 O •H

pl 43 in cd OrQ<l|

0

3 -P

<+H o

a o H 4 3 Ph

O 0 -H 0 • 4> Id

EH r - <Jj cd cd

1 o

P<

o

4 3

Pi cd a o CO o SH 4 3

•H a

0 XJ 4 3

rO

*0 0 a •H a Jh 0

-P • 0 CO T5 nd

O CO ,d cd -p £ 0

O rH •H cd 4 3 4 3

cd S U ' -p C 0

O EH S

•H rH

^ l r \ O • rd o

0 a H ?H

O 0 &

o M o p q 0

rH *H O T3 nd fi ^ fH fO h ' H K Jh 0 O U CO ^ Q rM

1 (D 1r\ fn

3 0> 'O

CD EH O

127

enzyme source were assayed; the kinetics were found to de-

viate only slightly from those typically obtained employing

the 25,000 x g supernatant.

It is well known that nonlinear kinetic behavior can

arise from two-substrate reactions when a ternary complex is

formed in a random order (pathway) process. The substrate-

velocity plot may in some cases be sigmoidal. Since the

phenylalanine-tryptophan hydroxylase system is complicated by

the requirement of three substrates (tryptophan, reduced

pteridine, and oxygen) for reaction, a complete kinetic an-

alysis is obviously very difficult. In view of the pro-

nounced effects of G*HC1, SDS, heat treatment, and alcohols

on the hydroxylation kinetics of tryptophan hydroxylase, and

in view of the fact that these agents give rise to linear

double reciprocal plots, it is reasonable to assume that

their effect on the system is not simply in establishing an

obligatory pathway in the reaction mechanism. Rather, it

is believed that enzyme-substrate cooperativity is modified

or lost such that typieal Michaelis-Menton- kinetics are

observed.

128

INTERPRETATION AND DISCUSSION

The "biological activity of many proteins is regulated

by naturally occurring metabolites which may or may not be

associated with the particular biosynthetic pathway in which

the protein functions. The binding of the regulatory metab-

olite to the protein results in a conformational alteration

or allosteric transition of the tertiary and/or quaternary

structure of the protein. This mechanism was suggested by

Monod, Changeux, and JacobS^ to play an essential role in

the regulation of metabolic activity.

In the enzyme, aspartate*transcarbamylase, which cat-

alyzes the initial step in pyrimidine biosynthesis, such a

biological control mechanism is operative.3? Feedback in-

hibition is exerted on aspartate transcarbamylase by cytidine

triphosphate, an end product of the biosynthetic pathway.

The negative allosteric effector thus inhibits the formation

of pyrimidines. However if insufficient pyrimidine is being

synthesized, adenosine triphosphate, a purine derivative, acts

to accelerate the velocity of the asparatate transcarbamylase

reaction. Thus, by the action of adenosine triphosphate as

a positive allosteric effector, a purine-pyrimidine balance

is maintained in the physiological system. ?

Many studies of the phenylalanine-tryptophan hydroxylase

system of rat liver have indicated that the hydroxylation is

129

mediated by a single nonspecific enzyme protein,^,11,13

even though the two hydroxylase activities have different

aotlvfttsofu and However, the wo^k by CouliOft

et al-.26 implicates a more complex system than a single

protein with a nonspecific catalytic site.

It is felt that the results of this investigation may

shed new light on the phenylalanine-tryptophan system of

rat liver. The most striking observation is that phenyl-

alanine hydroxylation and tryptophan hydroxylation do not

behave similarly. In repeated experiments the phenylalanine

hydroxylase response plots were only, at best, very slightly

sigmoidal. In contrast, the response plot for tryptophan

hydroxylation was reproducibly highly sigmoidal. Secondly,

agents such as guanidine hydrochloride and sodium dodecyl

sulfate, known to affect enzyme cooperativity through ter-

tiary or quaternary conformational modifications,31?32

exerted profound alterations on the tryptophan hydroxylation

kinetics. No such effects were observed for phenylalanine

hydroxylation. The response elicited by alcohols, in par-

ticular 1-propanol, on the tryptophan system was absent in

the phenylalanine system. The high concentrations

routinely employed (0.5 M for propanol) do not seem con-

sistent with an enzyme active site phenomenon, although this

possibility can not be completely discounted with the avail-

able data. It is felt that the alcohol effect is possibly

that of disruption of hydrogen and/or ionic bonding

130

associated with tertiary or quaternary structure, perhaps

through dielectric constant effects. Since it was shown

that the sigmoidal kinetics are not the result of loss of

either substrate or product through interferring processes,

it is believed that the hydroxylation of tryptophan is

mediated by homotropic and probably homotropic-heterotropic

allosteric regulation.

That homotropic regulation is apparently involved is

demonstrated by the sigmoidal tryptophan response plot

(Figure 1). That is, the binding of tryptophan to the en-

zyme facilitates further tryptophan binding resulting in a

sharp increase in velocity before the maximum velocity is

attained. Homotropic regulation is often accompanied by

homotropic-hetereotropic allosteric regulation.37 As was

previously mentioned, epinephrine and its precursors were

tested as possible allosteric activators, but with the ex-

ception of phenylalanine (Figure *0 , failed to stimulate the

hydroxylation. It is difficult to explain the positive

effect of phenylalanine on the tryptophan hydroxylase system.

It would be anticipated that certain end products of phenyl-

alanine, such as dopamine, norepinephrine or epinephrine,

which are known neurohumors, should be positive effectors of

the first step involved in serotonin biosynthesis.

A logical negative effector for tryptophan hydroxylation

might be serotonin itself. However, employing the present

colorimetric assay, 5-hydroxyindoles can not be added to the

1,31

reaction mixtures and obtain meaningful results. A radio-

chemical assay would be desirable in conducting such a study.

The most intriguing question arising during the course

of this investigation was that of the failure of other in-

vestigators to have observed the phenomenon of sigmoidal

substrate kinetics. There are some possible explanations,

however. First of all, most investigators working with the

phenylalanine hydroxylase system incubated the hydroxylation . O o

mixtures at 25 to 30 . But in the first reports of tryp-

tophan hydroxylation,''®?1"' and in all of the subsequent re-

ports, the reactions were performed at approximately 37°.

It is clearly evident in Figure 9 that only a slightly sig-

moidal curve results when the reaction.JLs run at 37° in

contrast to the obvious sigmoidicity of substrate response

for the reaction at 30°. The higher incubation temperatures

obviously cause a decrease in substrate cooperativity which

under certain conditions is irreversible (Figure 10). These

studies at higher incubation temperatures may explain the

failure of earlier investigators to observe the anomalous

behavior in the kinetics of tryptophan hydroxylation by rat

liver extracts.

The failure to observe cooperativity in the tryptophan

hydroxylase activity may be due in part to the use of par-

tially purified enzymes by other workers. In Figure 15 it

is seen that the reaction kinetics are only slightly sigmoidal

when Fraction I is employed as the hydroxylating enzyme and

Figure 1 5

Tryptophan Hydroxylation Employing Partially

Purified Enzyme Preparations

Reaction velocity is expressed in terms of mumoles of 5-hydroxyindole formed/minute. The usual procedures and incubation mixtures were employed as described in Exper-imental Methods. The reaction mixtures were incubated at 30° for 1 hour. The procedure for the preparation of Fraction I is described in Experimental Methods.

Q gp , hydroxylation of tryptophan by crude rat liver preparation.

, hydroxylation of tryptophan employing Fraction I and crude sheep liver preparation.

133

O CD

L J °

^ . 0 0 5.00 10.00 15.00

T R T P T O P H R N CMILL IMOLf lR)

20.00 25.00

13^

crude sheep liver homogenate as the pteridine-reducing

enzyme. In the same experiment the crude preparation dis-

plays the usual sigmoidal Kinetics, It should also b©

pointed out that these studies were conducted at 30° 5 it is

probable that at 37° the kinetics of the reaction employing

partially purified enzymes would be nonsigmoidal (i.e,

hyperbolic).

The logical question to be answered is why the tryp-

tophan hydroxylating system employing partially purified

enzymes obeys typical Michaelis-Menton kinetics. It has

been shown that several reagents or reaction conditions

have perturbed the apparent cooperativity of the tryptophan

hydroxylase system. It is believed that during routine

purification prcedures the cooperativity is destroyed or

altered by the usual methods of protein purification. It

was found, as has been previously reported,^"3 that the

rat liver hydroxylase enzyme activities are greatly decreased

upon long-term dialysis. Dialysis of enzyme preparations

in 0.075 M potassium phosphate buffer (pH 7.if) for 2k hours

results in a complete loss of activity. However, upon

carrying out the dialysis in the same manner employing

buffer containing 10$ glycerol (an agent known to stabilize

quaternary structure), only approximately half of the ac-

tivity is lost. It was recently found that buffer containing

both glycerol and 2-mercaptoethanol can be employed as a

dialysis medium with, no loss of hydroxylase activity after

135

dialyzing 2b hours. Also, no loss in cooperativity is ob-

served. Thus, perhaps the complete purification of the

hydroxylase system of rat liver can be conducted in the

presence of glycerol and 2-mercaptoethanol with no cor-

responding loss of enzyme cooperativity.

Since the brilliant studies of Page, Rapport, and

ErspamerB^-^O which led to the isolation and character-

ization of serotonin, many new fields of investigation at

biochemical, pharmacologic, and psychic levels have been

developed. As usual the elucidation of- the-biochemistry

has advanced the most rapidly although many of the details

are yet lacking. Surprisingly little is known about the

functions of serotonin. However, the demonstration that

hallucinagenic indole drugs, such as lysergic acid diethyl-

amide, inhibit the actions of serotonin on smooth muscle

suggested that the central action of such drugs results from

antagonism of serotonin centrally.^1Subsequently,

Pletscher et al.^ showed that reserpine, an indole tran-

quilizing drug, released bound serotonin from the brain.

With these findings the concept of a possible role of ser-

otonin in the central nervous system developed.

If serotonin is indeed a factor in maintaining a central

nervous system "balance" with the other biogenic amines,

particularly the catecholamines, a regulation of its bio-

synthesis as well as catabolism would not be too surprising.

The results of this investigation imply that the biosynthesis

136

of serotonin in rat liver is subject to allosteric regu-

lation. Other investigators have suggested that liver

tryptophan hydroxylase is a negligible contributor to over-

all serotonin production in view of the low specific ac-

tivity of the liver enzyme (Table I). However, upon con-

sidering the possible allosteric nature of the liver enzyme,

and the actual size and mass of this organ in contrast to

other organs of the body, it may well be that liver tryp-

tophan hydroxylase is a major contributor to the total

serotonin levels present in the organism.

137

REFERENCES

1. Udenfriend, S., and Cooper, J. R., J. BioL Chem.. 12l±?

503 (1952).

2. Mitoma, C., Arch. Biochem. Biophys.. 60. V76 (1956).

3. Kaufman, S., J. Biol. Chem., 226. 511 (1957).

k. Kaufman, S., J. Biol. Chem.. 2^0. 931 (1958).

5. Kaufman, S., Biochim. Bio-phvs. Acta. 27. 28 (1958).

6. Kaufman, S., J. Biol. Chem.. 2^. 2677 (1959).

7. Udenfriend, S., Titus, E., Weissbach, H., and Peterson,

R. E., J. Biol. Chem.. 219. 33? (1956).

8. Lovenberg, W., Weissbach, H., and Udenfriend, S., J.

Biol. Chem.. 2^7. 89 (1962).

9. Cooper, J. R., and Melcer, I., J. Pharmocal. Exptl.

Therat).. H2. 265 (1961).

10. Freedland, R. A., Wadzinski, I. M., and Waisman, H. A.,

Biochem. Bio-phvs. Res. Commun. . 9^ (1961).

11. Renson, J., Weissbach, H., and Udenfriend, S., J. Biol.

Chem.. 2^7. 2261 (1962).

12. Renson, J., Goodwin, F., Weissbach, H., and Udenfriend, S.,

Biochem. Biophvs. Res. Commun.. 6, 20 (1961).

13. Kaufman, S., Biochim. Biophys. Acta, 51. 619 (1961).

1*+. Day, M. and Greem, J. P., J. Physiol.. 16>+. 210 (1962).

15. Levine, R. J., Lovenberg, W., and Sjoerdsma, A., Biochem..

Pharmacol.. 1 k. 887 (1965).

138

16. Sato, T. L., Jequier, E., Lovenberg, W., and Sjoerdsma,

A., European J. Pharmacol., 1, 18 (1967)*

17. Bertaccini, G., J. Phvsiol. (London), U£i, 239 (1960).

18. Gal, E. M., Armstrong, J. C., and Ginsburg, B., J.

Neurochem., 13.? 6 +3 (1966).

19. Nakamura, S., Ichiyama, A., and Hayaishi, 0., Fed. Proc.,

2it, 60^ (1965). " " ' •

20. Green, H., and Sawyer, J. L., Anal. Biochem., JLl, 53

(1966).

21. Grahme-Smith, D. G., Biochem. Biophys. Hes. Commun., 16,

586 (196V).

22. Lovenberg, ¥., Jequier, E., and Sjoerdsma, A., Advan.

Pharmacol., 6 (Pt. A), 21_ (1968).

23. Lovenberg, W., Jequier, E., and Sjoerdsma, A., Science,

155. 217 (1967).

2k. Nagatsu, T., Levite, M., Udenfriend, S., J. Biol. Chem.,

232, 2910 (196*0.

25. Guroff, G. and Abramowitz, W., Anal. Biochem., 12, 5*+8

(1967).

26. Coulson, ¥. F., Wardle, E., and Je-pson, J. B., Biochim.

Biophys. Acta, 167, 99 (1968).

27. Pohland, A., Flynn, E. H., Jones, R. G., and Shive, W.,

J. Am. Chem. Soc., 21, 32k7 (1951).

28. Udenfriend, S. and Cooper, J., J. Biol. Chem., 196.

227 (1952).

139

29. Udenfriend, S., Weissbach, H., and Clark, C. T., J.

Biol. Chem.. 215, 337 (1955).

30. Greenstein, J. P., and Winitz, M., "Chemistry of the

Amino Acids," Vol. 3, John Wiley and Sons, Inc., New

York, N. Y., 1961 , p. 2323.

31. Tsao, T. C., Biochim. Bionhys. Acta. 11 . 368 (195*+) •

32.' Hersh, R. T., Biochim. Biophys. Acta. 58. 353 (1962).

33. Udenfriend, S., Creveling, C. R., Posner, H., Redfield,

B. G., Daly, J., and Widkop, B., Arch. Biochem. Biophys..

M , 501 (1959).

3^. Monod, J., Changeux, J. -P., and Jacob, F., J. Mol. Biol..

6, 306 (1963) .

35- Changeux, J. -P. and Gerhart, J. C., "Allosteric Inter-

actions in Aspartate Transcarbamylase," Proceedings of

the Fourth Meeting of the Federation of European Biochem-

ical Societies (July 3-7? 1967), pp. 13-38.

36. Freedland, R. A., Biochim. Biophys. Acta. 23.5 71 (1962).

37. Monod, J., Wyman, J., and Changeuz, J. -P., J. Mol. I

Biol.. 12, 88 (1965).

38. Rapport, M. M., Green, A. A., and Page, I. H., J. Biol.

Chem.. IZit, 735 (19W.

39. Rapport, M. M., J. Biol. Chem., 180. 961 (19*+9).

^0. Erspamer, V. and Asero, B., Nature. 169. 800 (1952).

*+1 . Gaddum, J. H., Ciba Foundation Symposium on Hypertension,

Humoral, and Neurogenic Factors, 75 (195^).

*+2. Wooley, D. ¥., and Shaw, E., Proc. Natl. Acad. Sci.. UO,

228 (195^).

1*fO

1*3. Pletseo• x», A., Shore, P. A., and Brodie, B. B., J.

Pharmacol. Exptl. Therap.. 116. 81* (1956).