THE SYNTHESIS AND REACTIVITIES 0F OXIDIZED
NICOTINAMIDE ADENINE DINUCLEOTIDE ANALOGS
WITH LIVER ALCOHOL DEHYDROGENASE

Thesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
MICHAEL KENNETH MAY
1976

 

Iv A' ' L [B R A R Y
IIIIIIII III III I 5 Michigan State

I .

 

' This is to certify that the ‘ H3 j - '.

thesis entitled

The Synthesis and Reactivities of Oxidized
Nicotinamide Adenine Dinucleotide Analogs
With Liver Alcohol Dehydrogenase

presented by
Michael Kenneth May

has been accepted towards fulfillment
of the requirements for

 

 

 

Ph.D. degree in Chemistry
(, / , “L
v ‘1

Major professor

Date 1/19/77

 

0-7639

 

 

 

. “J

ififiir‘
V:

i. f j ,w .‘i
”I .-’ P :93)

e: u a

 

C9/04059

ABSTRACT

THE SYNTHESIS AND REACTIVITIES OF OXIDIZED NICOTINAMIDE
ADENINE DINUCLEOTIDE ANALOGS WIflH LIVER ALCOHOL DEHYDROGENASE

By

Michael Kenneth May

The stereospecific1ty of the nicotinamide adenine dinucleotide
(NAD+-NADH) coenzyme oxidation-reduction of carbonyl compounds with the
dehydrogenase enzymes has been the topic of considerable study since
these reactions were discovered in 1935. werk done initially by Prelog
and extended by Karabatsos and other investigators has led to the pro-
posal of a model which might explain the substrate-coenzyme orientation

during the oxidation-reduction process (I).

H,’ HA
" CONH2.
T O L = larger group
Z s = smaller group

2 = ribose-phosphate-phosphate-ribose-adenine

I.
The model, if proven correct, should give information which might allow

predictions as to the stereochemistry of the products from the reaction.

Michael Kenneth May

One of the questions that must be answered is the spacial rela-

tionship of the substrate to the enzyme-coenzyme during the oxidation-

reduction process (11, 111).

R2 0
CONH2 C0NH2
R1 I R] KN I
"II o I R2
2 z
n HI

This question might be answered if the substrate were covalently bonded
to the coenzyme in such a manner as to limit possible alignments by
denying the substrate its freedom to rotate within its plane of the car-

bonyl. To accomplish this purpose, oxidized NAD+ analogs were syn-
thesized (1V3.

CONH
/ \
\ I (CH2)I1
'1'+ HOCIZH/ n = 3-8
1 CH3

IV

Several of the analogs prepared react in a manner similar to the
parent system but at a relatively slower rate when reacted with horse
liver alcohol dehydrogenase (liver ADH, E.C. 1.1.1.1.). The most reac-

tive ones are those with n = S and 7.

THE SYNTHESIS AND REACTIVITIES OF OXIDIZED
NICOTINAMIDE.ADENINE DINUCLEOTIDE ANALOGS
WITH LIVER.ALCOHOL DEHYDROGENASE

BY

Michael Kenneth May

A THESIS

Submitted to

Nfichigan State University

in partial fulfillment of the requirements

fer the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1976

To Shellye

 

ANIMAL CRACKERS by Rog BoIIci

ELEPHNOTS DON'T
CLIMB TREES! .. ,.
EUERQ ANIMAL.

BOOK PUBLtSI-IED
Nu. BEAR 'ME OUT!

 

 

ACKNOWLEDGMENTS

The author wishes to express his sincere appreciation to Professor
Gerasimos John Karabatsos for his guidance and encouragement throughout
this investigation as well as to Professor Harold Hart for serving as
second reader of this thesis.

Financial support provided by the Department of Chemistry, Michigan

State University, is gratefully acknowledged.

iii

TABLE OF CONTENTS

Page

ACKNOWLEDGMBNHS ............................................. iii
LIST OF TABLES .............................................. vi

LIST OF FIGURES ............................................. vii
INTRODUCTION ................................................ 1
EXPERIMENTAL ................................................ 12
Part I. Synthesis and purification of NAD+ analogs ....... 12
1. Synthesis of B,w-Hydroxyamines ...................... 12
A, Alkylations of Ethyl Acetoacetate ............... 12

B. Decarbethoxylation of alkylated B-ketoesters.... 13

C. Reduction of Ketonitriles ....................... 15
D. Reduction of B-hydroxynitriles .................. 16
II. Preparation of 8-Hydroxy-1-octylamine ............... 18
A. Alkylation of Ethyl Malonate .................... 18

B. Decarbethoxylation of Ethyl(S-Cyanopentyl)-
malonate ........................................ 19
C. Preparation of 8-Hydroxy-l-octylamine ........... 19

1. Preparation of Raney Nickel CW-4) Catalyst.. 19

2. Reduction of Ethyl 7-Cyanoheptanoate ........ 20
III. Synthesis of N-(hydroxyalkyl)nicotinamides .......... 20
A, Preparation of Nicotinyl Chloride ............... 20

B. Preparation of the Trimethylsilyl Ethers of the
AminoalcOhols ................................... 21

C. Preparation of N-(trimethylsiloxyalkyl)
nicotinamides ................................... 21

D. Preparation of Ne(hydroxyalkyl)nicotinamides.... 22
IV. Preparation of NAD+ analogs ......................... 23
V1 Isolation and purification of NAD+ analOgs .......... 24

iv

TABLE OF CONTENTS (Continued)

Page
Part II. Initial reactivities of coenzyme analogs ......... 27
A” thhods ........................................ 27
B. Materials and sample preparation ............... 28
RESULTS AND DISCUSSION ...................................... 30
BIBLIOGRAPHY ................................................ 40

LIST OF TABLES

TABLE PAGE
1. Effects of hydrophilic groups on product stereo-
specificity ............................................. 9
2. Alkylations of Ethyl Acetoacetate ....................... l3
3. Decarbethoxylation of Alkylated B-ketoesters ............ 15
4. Reduction of Ketonitriles ............................... 16
5. Reduction of B-hydroxynitriles .......................... l7
6. Melting points of N-(hydroxyalkyl)nicotinamides ......... 23
7. Fluorescence study of analog reactivity ................. 38

vi

LIST OF FIGURES

FIGURE PAGE
1. Structure of Oxidized Nicotinamide Adenine
Dinucleotide ............................................ 1
2. Reversible Oxidation and reduction of NAD+-NADH ......... 2
3. Enzymatic oxidation-reduction of Acetaldehyde-l-d ....... 3

4. Stereochemical course of YAD-I, NAIH, Acetaldehyde
enzymatic reaction ...................................... 3

5. Absolute configuration of the pro-R and pro-S hydrogens

of NAIH .................................................
6. Prelog Model for substrate-NADH orientation with Class B
enzymes ................................................. 5
7 . Kosower's model compounds used to compare hypsochromic
shift to NAIH ........................................... 6
8. Kosower' s postulate for carbonyl alignment in binary
complexes ......................................... . ..... 6
9. Arrangement suggested by Karabatsos ..................... 7

10. Stereospecificity of NAD+-YAI}I oxidation of Z-octanol. . . 7
ll. Enzymatic oxidation-reduction of lactaldehyde ........... 8
12. Enzymatic oxidation-reduction of pyruvic acid ........... 9

13. Possible carbonyl orientations of non-bonded substrates. 10

14. NAD+-Analogs with covalently bonded srbstrated .......... 10
15. Possible orientations of the carbonyl in covalently bon-
ded NAD -Analogs ........................................ 11
16. Elution curves for analog purification. ................. 32
17. Typical cyanide addition complex of NAD+ analogs ........ 33

18. Spectrum of N-(6-hydroxyhepty1) NAD+ and NAD+ with
ethanol ................................................. 35,36

19. Spectrum of N-(6-hydroxyhepty1) NAD+ without ethanol. . . . 37

vii

INTRODUCTION

The naturally occurring pyridine nucleotides nicotinamide adenine
dinucleotide (NAD+) and its phosphoric acid derivative (NADP+), with
their reduced forms (NADH and NADPH, respectively), are the coenzymes in
over a hundred enzyme catalyzed reactions. Whereas the structure of the
pyridine nucleotides was postulated as early as 1931 and the chemical
principles connected with the biological oxidation reactions were dis-
covered by warberg et a1.(1) in 1935, it was not until 1957 that the
structure was confirmed through synthesis by Todd.(2) The structure of
the oxidized fbrm of nicotinamide adenine dinucleotide is shown in Figure 1.

hfli2

C) C) /
g II II < I ,J
+N H H Hzo-IID—o-r—ocwz N N
W . . ea
f1 I1 f1 IND
(3')(2')
Figure l. - Structure of Oxidized Nicotinamide

Adenine Dinucleotide

The presence of a phosphoric acid group at the 2'-position of the
ribose in the adenylic acid moiety is the only difference between NADP+
and NAD+. With respect to reactions and properties associated with the
nicotinamide moiety the two coenzymes are identical.

The nicotinamide adenine dinucleotides are important participants
in many biological oxidation-reduction reactions. Dehydrogenase enzymes
(DH) catalyze the reversible electron transfer — as a proton or hydride

from various substrates to the C-4 position of the nicotinamide ring of

these cofactors to produce the reduced 1,4-dihydronicotinamide structures

of NADH or NADPH (Figure 2).

CONH
2
@ ‘DH
\N

l+
Z

 

(z = ribose-phosphate-phosphate-ribose-adenine)

Figure 2. - Reversible oxidation and reduction of NAD+-NADH

These enzyme catalyzed hydrogen transfers are stereospecific in that the
enzymes differentiate between the pair of hydrogens at C-4 of the 1,4-
dihydronicotinamide ring. Biological stereospecificity toward enantio-
topic and diastereotopic hydrogens was first observed and investigated
in the enzymatic reaction between yeast alcohol dehydrogenase

(YADH), NAD+, and ethanol by Leowus, westheimer, and vennesland.(3)
These investigators fbund that the 1-deuterioethanol fermed by the

YADH catalyzed reduction of l-deuterioacetaldehyde by NADH was
distinguishable from the l-deuterioethanol product fermed by the
enzymatic reduction of unlabeled acetaldehyde by NADD (the reduced form
of nicotinamide adenine dinucleotide with deuterium at C-4 of the nico-
tinamide moiety). The two products could be clearly differentiated by
running the reverse enzymatic reaction, the oxidation of ethanol by NAD+
in the presence of YADH, the 1-deuterioethanol formed enzymatically from
1-deuterioaceta1dehyde yielded non-deuterated NADH and l-deuterioacet-
aldehyde; the other l-deuterioethanol sample yielded NADD. The reaction

sequence is summarized in Figure 3.

o

a. misc-D + mm + u“ ——+ (11391011 + NAD+
9 . D .

b. CHgC-H + NADD + H ——- 01391011 + NAD

o D

c. (EH-SCHOH + NAD” ——. CHEEi-D + NADH + H+
D o

d. (11391011 + NAD+ —» digit-H + NADD + H+
D

Figure 3. - Enzymatic oxidation-reduction of Acetaldehyde-l-d

Although the results of westheimer and co—workers established that the
YADH reaction displayed stereospecificity toward the pro-chiral acet-
aldehyde and ethanol, they did not establish the stereochemical course of
the reaction.

The absolute configuration of 1-deuterioethanol obtained from the
reaction of YADH, NAH—I and acetaldehyde-l-d was established by Levy
and workers(4) who feund that the l-deuterioethanol fermed was levorota-
tory, and by the imaginative and definitive stereospecific synthesis by
Lemieux and Howard(5) of (+) R-l-deuterioethanol. These results require
that an electron from NADH be transferred onto the rgfface of acetaldehyde
and in the reverse reaction the pro-R hydrogen of ethanol to be transferred

to NAD+ (Figure 4). Similar experiments with other enzymatic preparations

 

 

Hs H, HS
I I CON"2 / ONH2
H CH
N + \m area-smog ‘ J + ”Xe”?!
I (3 “ I I1:i (3f!
2 1

Figure 4. - Stereochemical course of YADH, NADH,

Acetaldehyde enzymatic reaction

4

have established that biological stereospecificity toward the diastereo-
topic hydrogens at C-4 of the 1,4-dihydronicotinamide ring of NADH (and
NADPH) is not a unique property of‘YADH but a general enzymatic phenom-
enon.(6’7’8) These biologically stereospecific enzymatic reactions
were classified as "C1ass:A", having the same stereospecificity as

YADH; or "Class-B", having opposite stereospecificity of YADH(9) (Figure
5). The absolute configuration of the pro-R.and pro-S hydrogen (Figure
5) of the dihydronicotinamide ring of NADH was established in 1964

PI

'1
a

 

(I) ' (II)

Figure 5. - Absolute configuration of the pro-RF
and pro-8* hydrogens of NADH
by Cornfbrth and co-workers.(11)
, Prelog, in studying the requirements of the enzyme active site and
substrate orientation, determined the configuration of the product al-
cohols formed from a variety of reductions of cyclohexanones and decalones
with NADH and enzymes (Class B) isolated from Curvularia falcata.(6’12)

On the basis of the compounds used in these studies Prelog postulated a

 

*There is considerable disagreement in the literature as to the labelling
of the paired.methy1ene hydrogen at C-4 of the 1,4-dihydronicotinamide
ring. A.more descri tive - and correct - labelling would be the method
developed by Hanson('0) shown in Figure 5-1. For simplification in des-
cribing those hydrogens acted upon by Class-A enzymes (HA) or Class-B
enzymes (H3), the alternate notation in Figure 5-11 will be used throughout.

model which was to define the spacial requirements of the substrate to

the binary coenzyme-enzyme complex (Figure 6).

HA’ H B

HZNOC L
s I.“

| 0
Z

Figure 6. - Prelog Model for substrate-NADH orientation

with Class B enzymes (L=larger group, s=smaller group)

Prelog justified his model by noting that steric interactions between
coenzyme and substrate would be smaller when the smaller group, "s", was

over the carbamido group and the carbonyl pointed down toward the pyridin-
ium nitrogen of the dihydronicotinamide moiety, as proposed by Kosower.(13’14)
He predicted that Artype enzymes would lead to products of opposite con-
figuration, as the position of "s” and "L" would have to change.

Kosower proposed the model with the carbonyl p01nted down toward
the pyridinium nitrogen in order to explain the observed shift in absorp-
tion from 340 nm fer the unbound NADH to 325 nm for the bound NADH-
horse liver alcohol dehydrogenase (LADH) complex (this observation had

(15) and Theorell(16)). He sugges-

also been noted preViously by Kaplan
ted that a positive charge, in close proximity (ca. 3A) to the nitrogen
of the dihydropyridine ring, could account fer the hypsochromic shift,

as observed with model compounds A and B below (Figure 7).(14) The cal-

culated increase in the energy of the n-———’ ‘n* transition in the

presence of an alkylammonium ion, hydrogen bonded to two groups as shown

O O
)‘max = 2210 A Amax = 2380 A
Figure 7. - Kosower's model compounds used to compare

hypsochromic shift to NADH

in Figure 8, corresponds closely to the observed energy change.(14)

 

ICCDBH42
""H\ s‘H
/,h4
0----” ‘enzyme
HO I”phoqelmto
Figure 8. - Kosower's postulate for carbonyl

alignment in binary complexes

Support for much of this work is based only on identification of
products obtained from the previously mentioned reactions, and by analogy
to the absorption spectra of reduced and oxidized dihydropyridines and
pyridinium salts. It is evident therefore that the model having the
carbonyl group pointed downward toward the nitrogen of the ring is not
based upon the best evidence. From his model studies, Prelog further sug-

gested that the position of the substrate groups would be controlled by

their hydrophilic-hydrophobic interactions with the enzyme. This
suggestion had been made earlier by several investigators.(3’4’ls)
Reflecting upon the results obtained by Levyc4) on the enzymatic
reduction of acetaldehyde-l-d with NADH and YADH, Karabatsos pointed out
the contradictory results obtained by applying the Prelog model in this
case. He suggested(l7) that the carbamido group of the coenzyme played
no role in determining the stereochemistry of the product alcohols
since application of the Prelog model should predict alcohols of the
R-configuration. Experimental evidence clearly showed that these
products had the S-configuration. The spacial arrangement of the

substrate-coenzyme system was suggested to be the one shown in Figure 9,

an arrangement opposite to the one Prelog had suggested.

HB,’ HA
’ CONH2
s L
N
I o
2

Figure 9. - Arrangement suggested by Karabatsos

As support fer the model shown in Figure 9, the following evidence may be
cited: Kaplan and van Eys(l8) found that in the oxidation of (R,S)-2-octanol
with YADH and NAD+ only the (S)-2-octanol reacted to give the corresponding

ketone (Figure 10).

 

 

'15; fhk
+ I yeast ADH C H
as on \N ‘T Q H3c u 6(1)]3
.+ (s) T 0
Z Z

Figure 10. - Stereospecificity of NAD+-YADH oxidation of Z-octanol

Meshenflg’ZO’Zl) by using fermenting yeast (a Class A enzyme) to reduce
trimethylacetaldehyde-l-d, benzaldehyde-o-d, and butyraldehyde-l-d,
obtained products of the S-configuration. Donniger and Ryback, (22) using
LADH (a Class A enzyme) to reduce geraniol-l-t, obtained alcohol of the
S-configuration; and recently Gunther(23) and co-workers with 1-propanol-
l-d and YADH feund results which also supported the model by Karabatsos.
The evidence presented thus far in support of a model for the sub-
strate NADH spacial requirements has generally ignored the importance of
hydrophilic-hydrophobic interactions since the substrates used contained
relatively nonpolar alkyl substituents bonded to the carbonyl carbon of
the molecules. Karabatsos et al.(24’25) determined the absolute con-
figuration at C-1 of 1,2-propanediol-1-d from the reduction of D- and L-
lactaldehyde with NADD and LADH. These investigators found the absolute
configuration at C-1 of the 1,2-propanediol—l-d produced from the reac-
tion to be the R-configuration. This finding showed that hydrophobic
interactions were less important than steric interactions as far as

product stereospecificity was concerned (Figure 11).

 

l1: [) (?f1 PI
’ CONH
2 IiverADH H CHCH3 / CON”2
H \ ‘fHCHa ‘:___~. §< + l
N 0 0H 0‘ 0H N
I p“
Z 2

Figure 11. - Enzymatic oxidation-reduction

of lactaldehyde

However, work by van Eys and Kaplan(18) with pyruvic acid and YADH yielded

D-(-)-1actic acid which had the R-configuration (Figure 12). These

 

 

 

PL_ I4
’ 0N”? yeastADHA "02C CHa / CONH2
11CD2CZ fl ()FI -
N N
l |+
Z 2
Figure 12. - Enzymatic oxidation-reduction of pyruvic acid

as

contradictory findings by Kaplan and others 31) led workers in the

Karabatsos laboratories(32’33)

to test the relative importance of hydro-
phobic interactions as well as steric interactions on substrates with

other groups of comparable size but differing hydrophilicities (Table 1).

Table 1. - Effects of hydrophilic groups on product

stereospecificity (Ref. 33)

 

 

 

 

Ha’ HA
‘ CONH
R1 R2 | I cmNHiDH / | 2
Dr + :RIRQCHOH + \
(3 hi I INI-
I HA
1 Z
R1 R2 Enzyme(s) Class %R-c0nfig. %S-config.
CH3- HOCHZ- Gly-DH A mlOO O
CH?)— CHSCHZ- LADH,Gly DH A 33—36 64-67
CH3- ClCHZ- LADH A 48 52

The findings summarized in Table 1 above might imply that the orientation
predicted from either model may be completely reversed if the substituents

on the carbonyl bearing substrate are sufficiently hydrophilic. Indeed,

10

a model may be needed for every enzyme tested.

There is, however, one obvious fact which has not been taken into
account in the studies of the various model systems yet proposed. To
date there has been no model presented which may account fer the pos-
sibility of having the carbonyl of the substrate aligned in a manner
other than that suggested by Kosower.(13’l4) Any model now proposed
must account for the possibility of having the carbonyl group either "up“
of "down", or similar orientations, on face A or face B of the nicotin-
amide ring of the coenzyme. Available data from product stereospecificities

agree with either orientation of the carbonyl group (Figure 13).

 

H, HA
5; ‘ L .gflEfli. L CIDBU42
'*$‘ (3?. F4 5
‘2
Figure 13. - Possible carbonyl orientations of non-bonded substrates

This thesis describes an attempt to answer the question of the
substrate alignment by covalently bonding the substrate to the coenzyme

and thus removing its freedom to rotate within its plane (Figure 14).

/ C0NH\ / CONH\
\ l (0+2)n \ I (Cf-l2)"
Z z I
CH3
(1 I (II)
n::4-7 rl=:3-£3

Figure 14. - NAD+-Analogs with covalently bonded substrates

11

By restricting the "face" which the bonded substrate must use, with ap-
propriate choice of enzyme, only two orientations of the carbonyl should

be allowed (Figure 15). Isolation of products with concomitant analysis

 

ft, '1
I '——* I I
N R N R
.+ “V I 7r“
1 OH Z 0
,z’ CK35fl1-\\
\ I (CH2)n
N HO
|+ H“‘\
Z R
Figure 15. - Possible orientations of the carbonyl

in covalently bonded NAD+-Analogs

of the configuration of the unreactive alcohols should give a clearer
indication of substrate and coenzyme arrangement. The NAD+ analogs
shown in Figure 14 have been synthesized and their relative reaction
rates determined.(34) The results of these studies should provide fur-

ther insight into the elucidation of a model to fit this system.

EXPERIMENTAL

PART 1. Synthesis and purification of NAD+ analogs

I. Synthesis of B,w-Hydroxyamines

 

A. .Alkylations of Ethyl Acetoacetate

 

To a three-necked round-bottomed flask fitted with an addition fun-
nel, mechanical stirrer, and Friedricks Condenser was added 400 m1 of
absolute ethanol which had been refluxed six hours over magnesium.meth-
oxide and distilled prior to use.(35) The reaction vessel was maintained
under a slow stream of dry nitrogen while 11.5 g (0.500 mol) of cleaned
sodium metal was added and allowed to dissolve with gentle stirring.
After the solution had cooled to room temperature, 122.3g (0.940 mol,

119 ml) of freshly distilled ethyl acetoacetate (Aldrich) was added and
the solution was stirred fer 5-10 minutes at reflux on a steam bath. To
the heated solution was added 25 mol percent of potassium iodide (Mal-
linkrodt, reagent grade) which had been ground and oven dried at 150°
for two hours.

The stirring was increased as 0.500 mol of the appropriate pure
halonitrile was added to the heated solution of the sodio-salt of the
ester over a period of 10-15 minutes. The nitrogen flow was discontinued
and reflux was maintained until the mixture was neutral to moist litmus.
The reflux period varied from 1-4 hours depending upon.which halonitrile
was used. In all cases a white precipitate of the insoluble sodium iodide
began to ferm within 30 minutes.

Upon completion of the reaction the majority of the ethanol was dis-
tilled from the mixture and the residue was allowed to cool to room tem-

perature. After cooling, the residue - and precipitated salts - were

12

13

transferred to a separatory fUnnel. A 1:1 mixture of 320 ml benzene-
water was added to the separatory funnel whereupon two layers fbrmed.
The organic layer was separated and the aqueous layer extracted with
three lOO-ml portions of ether. The ethereal extractions and organic
layer were combined and dried over anhydrous magnesium sulfate.

The solvents were removed by rotary evaporator and the yellow, oily
residue was distilled under high vacuum. Three fractions were obtained

with the last fraction containing the desired products. (Table 2)

Table 2. - Alkylations of Ethyl Acetoacetate
X-(CHZ)n-CN + CH COGI CO CH CH -> (H COCHCO CH CH

 

 

3 2 2 2 3 3 . Z Z 3
(CH2)n-CN

n X Product b.p.,°C/mm %Yield
1 Cl ethyl (1-cyanomethyl)acetoacetate 9S.5°-100.5°/0.18 38.5
2 -- ethyl (Z-cyanoethyl)acetoacetate(36a) 132.0°-143.o°/o.zo 42.7
3 Br ethyl (3-cyanopropy1)acetoacetate 80.0°-82.5° /0.02 60.4

(C1) ethyl (3-cyanopropyl)acetoacetate ------- * (36.2)
4 Br ethyl (4-cyanobutyl)acetoacetate ------- * -—--
5 Br ethyl (S-cyanopentyl)acetoacetate 136.S°-l39.0°/0.02 69.0
6 Br ethyl (6-cyanohexyl)acetoacetate ------- * ----

 

* Product could not be distilled, crude residue used without further
purification

B. Decarbethoxylation of alkylated B-ketoesters
Procedure I:(38)
To a solution of 200 g of sodium carbonate in 1800 ml of water was

added 1.09 mols of the ester. The two layer mixture was refluxed fer

14

four hours during which time it became homogeneous. The product was
salted out with potassium carbonate and the aqueous layer extracted with
three lOO-ml portions of ether. The organic layer and ether extracts
were combined and dried over anhydrous potassium carbonate. The ether
was removed by rotary evaporator and the product distilled under vacuum.
In all cases, only one fraction was obtained with a small amount of
residue remaining (Table 3).

Procedure II:(39)

A solution containing 50 m1 of dimethyISulfoxide (Fischer certified),
0.20 mol of water, 0.07 mol of sodium chloride, and 0.065 mol of the ester
was added to a 3-necked round-bottomed flask fitted with a magnetic stir
bar, thermometer, and condenser attached to a trap filled with a saturated
solution of barium hydroxide. This mixture was heated slowly at 150°-
170° on an oil bath with stirring and kept at that temperature until the
evolution of carbon dioxide ceased. Heating was continued for 1-6 hours
depending upon the ester used. The solution remained colorless or
slightly yellow throughout the reaction period.

On completion of the reaction, the solution was cooled to room
temperature and transferred to a separatory funnel. Addition of a volume
of cold.water equal to the volume of dimethylsulfoxide used resulted in
the formation of two layers. The organic layer was separated and the
aqueous layer was extracted with three 50-m1 portions of ether. The
extracts were combined and washed with several small portions of saturated
sodium chloride solution and dried over anhydrous potassium carbonate.

The ether was removed under rotary evaporation and the product distilled.
Only one fraction was recovered (Table 3).
It is of interest that for the substrates where n = 1-3 (Table 2),

Procedure II caused decomposition of the B—ketoester; and for substrates

15

where n = 4—6 (Table 2), Procedure I failed to accomplish the desired

decarbethoxylation reaction.

Table 3. - Decarbethoxylation of Alkylated B-ketoesters

CHSCOCHCO CH CH ——a» CH3C0(CH2)n+l-CN

 

 

. Z 2 3
(CH2)n-CN

n Procedure b.p. °C/mm % Yield
1* I ..... --..-

II ----- ----
2 I 66.0°-68.5°/1.0 78.0(36)
3 I 65.5°-67.0°/0.10 85.3
4 II 88.0°-89.5°/0.10 86.0
5 II 91.2°-94.5°/0.20 75.9
6 II 90.5°-94.0°/0.0S 62.0

 

* Product hydrolyzed when either method was used.

C. Reduction of ketonitriles

 

Into a 3-necked round-bottomed flask fitted with an addition fun-
nel, thermometer, and magnetic stir bar was added 200 ml of absolute
methanol (Matheson, Coleman, and Bell) and 0.57 mol of the ketonitrile.
The solution was cooled to 0° in an ice-salt water bath with rapid stir-
ring. To the cooled mixture was added in small amounts a solution con-
taining 0.19 mol of sodium borohydride (Metal Hydrides, Inc.) in cold
water maintained at 0°. Only the minimum amount of water required to
dissolve the sodium borohydride was used. Hydride was added at such
a rate as to maintain the reaction mixture at a temperature less than

10°. Addition of the hydride solution was achieved over a period of

16

30-50 minutes in all cases.

Upon completion of the addition, the stirred solution was allowed
to gradually warm to room temperature. A solution of cold 10% sulfuric
acid was added until the solution reached a pH of 5-6. This solution
was then stirred for approximately 3—4 hours - or until the flocculent
precipitate became granular. The solid was filtered from the mixture
with a vacuum aspirator.

After most of the solvent was removed by rotary evaporator, the
residue was taken up in a small amount of ether and dried over anhydrous
magnesium sulfate. The ether was removed and the products were
distilled under vacuum (Table 4).

Attempts to reduce both the carbonyl and nitrile groups in one
reaction, using a variety of reducing agents, failed in all cases.

Only cyclised or polymerized products were obtained.

Table 4. - Reduction of Ketonitriles

CHSCO(CHZ)n-CN ———> GISCH(OH)(CHZ)n-CN

 

n b.p. °C/mm % Yield

 

3 55.0°-56.6°/0.10 95.0(36)
4 81.5°-83.0°/0.20 91.0
5 lO6.0°—106.5°/0.20 84.7
6 ll8.0°-ll9.5°/0.20 86.7

7 116.S°-118.0°/0.10 85.8

 

D. Reduction of B-hydroxynitriles with Raney Nickel (W—Z) catalyst

 

A solution of 0.1 mol of the nitrile in approximately 400 ml of ION

17

methanolic ammonia(40)

and 2-3 g of freshly prepared W—Z Raney nickel
catalyst (PCR Chemicals)(41) was placed under 50 psi of hydrogen pressure
in a Parr apparatus at room temperature. Reductions were generally com-
pleted within 18 hours with a pressure drop to within 2 psi of the cal-
culated value. Yields were quantitative for the liquid aminoalcohols
(Table 5).

After evaporation of the solvent, aminoalcohols were recovered
initially as pale green oils. These oils were dissolved in acetonitrile
(Fisher) to which a small amount of decolorizing charcoal had been added,

boiled for a short period, and precipitated as solids. Recrystallization

of the solids led to constant melting points (Table 5).

Table 5. - Reduction of B-hydroxynitriles

 

 

on on
ogrnwoggfoi——>(m§o+ampnrnfmg

n mtp.,°C[b.p.,°C] % Yield

2 [56°-59°/0.2-0.4](a) 62.08

3 [62.0°-62.5°/0.10] N100

4 [68.0°-69.0°/0.10] 92.0

37°-39°(b)

5 43.3°-44.1° 75.4

6 47.9°-48.1° 52.2

7 ss.3°-54.0° 68.6

 

a. Prepared by J. Miedema, Ph.D. Thesis, Michigan State
University, 1974, p. 26.

b. Distillable liquid which solidifies on standing at
room temperature

18

11. Preparation of 8-Hydroxy-l-octylamine

 

.A. .Alkylation of Ethyl Malonate

 

Into a 3-necked round-bottomed flask fitted with an addition fUn-
nel, mechanical stirrer, and Friedricks condenser was added 400 m1 of
freshly distilled dry ethanolCSS) and 1.05 moles of cleaned sodium metal.
The system was protected from moisture with dry nitrogen. When the metal
had completely dissolved, 27.66 g of pre-dried potassium iodide and 1.05
moles of freshly distilled ethyl malonate (Matheson, Coleman, and Bell)
were added and the solution was stirred for 15 minutes.

.A solution of 1.00 mol of 6-bromocapronitrile (Fairfield) in 120 ml
dry ethanol was added slowly to the rapidly stirring solution. On com-
pletion of the addition the reaction mixture was refluxed on a steam
bath until neutral to moist litmus. .A precipitate of the insoluble
sodium iodide began to be fermed immediately upon heating.

.Most of the ethanol was distilled from.the mixture and the residue
was allowed to cool to room temperature. A volume of water equal to the
volume of ethanol removed was added to the cooled residue without removal
of the precipitated salts. Two layers formed on addition of water and
the organic layer was separated from the mixture. This layer was washed
with three 50-m1 portions of a saturated sodium chloride solution and the
washings combined with the aqueous layer previously separated. The com-
bined aqueous solution was extracted with three 50-ml portions of ether
which were combined with the organic layer and dried over anhydrous
potassium carbonate.

The solvent was removed by rotary evaporation and the product was
obtained by vacuum distillation; b.p. 131.2°-132.0°/0.16 mm, in 71.7%

yield.

19

B. Decarbethoxylation of Ethyl (S-Cyanopentyl) malonate(39)

 

In a 3-necked round-bottomed flask outfitted with magnetic stir-bar,
condenser, and thermometer were mixed 40 ml of dimethylsulfoxide (Fischer),
0.15 mol of water, 0.05 mol of sodium chloride, and 0.04 mol of the diester.
The mixture was stirred vigorously as the temperature was raised to 150°-
170° in an oil bath. Heating was continued until the evolution of carbon
dioxide, monitored by means of a trap containing a saturated solution of
barium hydroxide, ceased.

The colorless solution was cooled in an ice-water bath to 0° and
transferred to a separatory funnel. A volume of cold water equal to the
dimethylsulfoxide used was added, resulting in the fermation of two layers.
The organic layer was separated and the aqueous layer was extracted with
three SO-ml portions of ether and combined with the organic layer. The
combined extracts were dried over anhydrous potassium carbonate.

The ether was removed by rotary evaporation and the residue was
distilled under vacuum. The one recovered fraction contained the product,

b.p. 93.0-94.0°/0.2 mm, in 73.9% yield.

C. Preparation of 8-Hydroxy-l-octy1amine

 

1. Preparation of Raney Nickel (W-4) Catalyst(42)

 

To a well-stirred solution containing 128 g of sodium hydroxide
in 500 ml of water was added 100 g of Raney Nickel catalyst powder (PCR
Chemicals) at such a rate as to maintain the temperature at 50° :_2°. A
few drops of ethanol were added to quench subsequent feaming when necessary.
After the addition had been completed, the mixture was digested fer 50
minutes at 50° with moderate stirring.

The catalyst was washed by decantation several times with water and

transferred to a 500-ml graduated cylinder. Approximately 15 liters of

20

water was passed through the suspended catalyst while the solution was
stirred in an arrangement described by Adkins. (42) The solid was kept
below the surface of the liquid at all times. When the washings became
neutral to litmus, the catalyst was transferred to a 250-ml centrifuge
bottle and stirred (not shaken) with three lSO-ml portions of 95% ethanol.
The catalyst was centrifuged after each washing at 10,000 rpm for 10
minutes.

The washing and centrifugation was repeated with three lSO-ml portions
of absolute ethanol and. the activated powder stored at 5° in a tightly
closed bottle filled to the top with absolute ethanol. This highly active
form of Raney catalyst was used inmediately upon preparation since it
tended to lose its activity upon storage.

2. Reduction of Ethyl 7-Cyanoheptanoate

 

A solution containing 0.10 mol of the ester in 400 ml of lON meth-
anoic anmonia with 5-6 g of the Raney Nickel (W- 4) catalyst was placed in
a Parr apparatus under 50 psi of hydrogen pressure. The reduction of
both the ester function and the nitrile was completed in 36 hours as
determined by the recorded hydrogen uptake .

The solvent was removed and the dark green oil boiled for a few
minutes in acetonitrile to which a small amount of decolorizing charcoal
had been added. The white crystals were obtained after several recrystal-

lizations from acetonitrile in 60.0% yield; m.p. 48.0°-48.3°.

III. Synthesis of N- (hydroxyalkyl)nicotinamides

 

A. Preparation of Nicotinyl Chloride

 

To a suspension of 400 ml of carbon tetrachloride and 64.8 g
(0.400 mol) of potassium nicotinate was added slowly, with stirring,

36.0 ml (0.500 mol) of thionyl chloride. The system was protected from

atmospheric moisture with dry nitrogen. After the addition was completed,

21

the solution was allowed to reflux fer 12-24 hours during which time
the evolution of sulfur dioxide ceased. The precipitated salts were
filtered from the mixture in an inert atmosphere and the solvent was
removed under vacuum. The residue was again filtered and distilled

to yield a colorless liquid, b.p. 30.0°-31.5°/0.08 mm. The yield was

not determined since the product hydrolyzes readily.

B. Preparation of the Trimethylsilyl Ethers of the Aminoalcohols

 

Equimolar amounts (ca. 0.10 mol) of the aminoalcohol and l,1,1,3,3,3,-
hexamethyldisilizane (Aldrich) with approximately 0.1 g of dried ammonium
chloride were placed in a three-necked round-bottomed flask equipped with
mechanical stirrer, condenser, and rubber septum. All glass-to-glass con-
nections were fitted with teflon sleeves (Chem. Tec). The heterogeneous
mixture was allowed to heat slowly to 125° with stirring, during which
time a vigorous evolution of ammonia was observed. When the evolution
of ammonia had ceased and the solution had become homogeneous, the mix-
ture was allowed to cool to room temperature. In order to avoid losses
due to hydrolysis, the product was not isolated (previous attempts at

(36b)

preparation and purification showed the product to be fermed in

nearly quantitative yield but to hydrolyze during purification).

C. Preparation of N-(trimethylsiloxyalkyl) nicotinamides

 

To the reaction vessel described above was added 70 ml of pyridine
(distilled from barium oxide) and 30 ml of triethylamine (distilled from
sodium metal). The flask was heated to 105° and an equivalent amount of
nicotinyl chloride (0.10 mol) was added slowly to the solution by injec-
tion through the septum using a disposable syringe with a neoprene

plunger. When the addition of nicotinyl chloride had been completed,

22

the septum was replaced with a glass stopper and the mixture was refluxed
fer one half-hour. The solution was cooled to 0° in an iceiwater bath
and the precipitated triethylammonium chloride filtered from the mixture.

The product was not isolated.

D. Preparation of N-(hydroxylalkyl) nicotinamides

 

The residue from the previous filtration was treated with 10 m1 of
water and the solution was refluxed fer one hour. The pyridine-water
azeotrope was distilled from the solution and the remaining solvent was
removed under vacuum with heating to yield a dark brown oil. The oily
residue was dissolved in chlorofbrm to a concentration of approximately
0.5 g/ml based upon the theoretical yield.

.Aliquots containing approximately 10 g of the crude product were
applied to a 2.5 x 80 cm column packed with alumina (Matheson, Coleman,
and Bell, Alcoa type F-20, 80-200 mesh) prepared with 100:2 chloroform-
methanol. Collection of the product by gravity elution using 100:2
chlorofOrmrmethanol as solvent was monitored by absorption at 254 nm
using a 1.0 mm flow cell with a Beckmann DB spectrophotometer.

Removal of the solvent by rotary evaporator left a pale yellow oil.
The oil was taken up in acetone and recrystallized twice in the same sol-
vent. The shorter chain N-(hydroxalkyl) nicotinamides crystallized spon-
taneously upon treatment with acetone. The longer chain substrates had
to be chilled to below 0° or allowed to stand for 10-14 days at room
temperature in a minimum of solvent. The yields, after successive re-
crystallizations, were typically 55-60% except fer N-(8-hydroxynony1)

nicotinamide. This compound could not be recovered as a white crystalline

23

solid in any solvent system tried. The brown oily residue solidified on
standing at room temperature after one month, but 1H-nmr of the tan solid
showed it to be reasonably free of impurities. The products and their

melting points are listed in Table 6.

Table 6. - Melting points of N-(hydroxyalkyl)nicotinamides

K;;\:[]//CONH-(CH2)m-CHZOH
K.

m Product m.p., °C

4 N-(S-hydroxypentyl)nicotinamide* 91.0°-92.0°
5 N-(6-hydroxyhexyl)nicotinamide* 79.2°-80.0°
6 N-(7-hydroxyheptyl)nicotinamide 60.1°-6l.0°
7 N-(8—hydroxyocty1)nicotinamide 90.1°-91.0°
*

Amino alcohols purchased from Aldrich

7T CONH- (CHZ)n-C'IH-OH
( I a.

\\ 3

N

n Product m.p., °C

3 N-(4-hydroxypenty1)nicotinamide 64.8°-66.0°(43)
4 N-(S-hydroxyhexyl)nicotinamide 48.9°-50.5°(36b)
5 N-(6-hydroxyhepty1)nicotinamide 41.1°-42.0°

6 N-(7-hydroxyocty1)nicotinamide 51.5°-52.1°

7 N-(8-hydroxynonyl)nicotinamide -----

8 N-(9-hydroxydecy1)nicotinamide 69.0°-70.3°

IV. Preparation of NAD+ analogs(42)

 

To a 25 ml Reacti-flask (Pierce) with a screw cap having a teflon-

rubber laminated insert and magnetic stir-bar was added 1.144 mmols of

24

the nicotinamide dissolved in 6.0 ml of 0.DM sodium phosphate buffer, pH
8.0. .A second solution was prepared containing 0.286 mmols of oxidized
B-nicotinamide adenine dinucleotide (Sigma, grade III) in 6.5 ml of the
same buffer. On mixing the solution had a pH of 7.6. To this solution
was added 0.5000 g (0.015 unit/mg) pig brain NADase (Sigma, E.C. 3.2.2.5.,
NAD glycohydrolase) and the heterogeneous mixture was incubated at 37.00°
:_0.10° fer 3.0 hours.

.At the end of the incubation period the solution was made up to 5%
in trichloroacetic acid (ca. 0.5 g) and chilled to 5°. The denatured
protein was removed from the mixture by refrigerated centrifugation at
10,000 rpm fer 10 minutes. The supernatant liquid was transferred to a
250-ml polyethylene centrifuge bottle with screw cap and 5 volumes (62.5
ml) of cold acetone was added to precipitate the solid products. After
standing at 5° for a few minutes the suspension was centrifuged at 5,000
rpm for 15 minutes and the supernatant liquid discarded. The residual
acetone was evaporated with a stream of nitrogen and the solid was dis-
solved in 10.0 ml of cold, de-ionized distilled water. The solutions

were stable for several months when stored at 5°.

V. Isolation and purification of'NAD+ analogs

 

The coenzyme analogs were separated on a 0.9 x 55 cm Polyethylene-
imine cellulose (PEI cellulose, Sigma, 1.20 meq/g) column prepared in
the following manner. .All operations were carried out at 5° unless
otherwise noted.

One and one-half column volume of dry resin was allowed to stand in
220 m1 of 0.003M ammonium bicarbonate solution for one-half hour. The

solvent was decanted from the settled suspension and the process repeated.

25

After the second decantation, the resin was suspended in 100 ml of the
bicarbonate solution and the entire mixture was poured into the column

as a slurry. The resin was allowed to settle for approximately two hours
without solvent flow. .A small layer of sand was placed above the settled
resin and the column was fitted with a pressure head connected to a
Polystaltic pump (Bfichler instruments). The solvent pressure was slowly
increased to allow even settling of the column until a flow rate of 1.8
ml/min was obtained. Column equilibration was continued until 1 liter

of 0.003M ammonium bicarbonate had passed through the resin.

.A 1.00 ml aliquot of the crude reaction mixture of a coenzyme analog
was loaded onto the column and those fractions containing analog and un-
reacted NAD+ were collected separately. Previous identification of the
fractions had been determined.(4z) The eluate was monitored at 260 nm
in a 0.1 cm flow cell using a Beckmann DB-G spectrophotometer. Fractions
from three column loadings were combined fer each analog isolated. The
column was washed for five hours with the eluting solvent between successive
separations. Of the substrates prepared, those analogs with seven- and
eight- carbon alkyl chains and a primary hydroxyl group, and the ten-
carbon chain with a secondary hydroxyl group could not be resolved.
variation of concentration of the eluting buffer as well as several
gradient elution methods failed to resolve the analog and NAD+ fractions.

.A 50-ml aliquot of each of the eluted fractions was further purified
by molecular filtration. The solution was placed in a 47 mm stirred cell
containing a PSAC membrane (Millipore, nominal molecular weight limit
1000). The solution was concentrated to approximately 10 m1 under 50 psi

nitrogen pressure with stirring and then diluted to 60 ml with 0.10M

26

sodium phosphate buffer, pH 7.5, which was 0.01M with respect to
semicarbazide. Dilution and concentration steps were repeated 5 times
to insure a sample purity of greater than 95% in terms of buffer exchange

and contaminating impurities. The final volumes of all samples purified

in this manner were 10 m1.

27

Part II. Initial reactivities of coenzyme analogs
Al Methods

Initial reaction velocities were principally determined by moni-
toring the increase in fluorescence due to the formation of the reduced
1,4-dihydropyridinium moeity reported to be at 462 mm. The excitation
wavelength used was 350 nm while recording fluorescence at 450 nm, A
computer centered spectrophotometer-spectrofluorimeter with PDP-8/I
(43)

unit was used to record any changes in fluorescence in samples whose

average concentration was approximately 8.4 x lO-DM (based on e
6

260 NAD+=
cm2 mol-l). Absolute concentrations could not be determined

18 x 10
since the reactions were run on a scale which precluded isolation of
adequate quantities for accurate determinations of extinction coeffic-
ients for each of the analogs. Extreme care was taken to exclude par-
ticulate matter which would have distorted the trace at the high sensi-
tivities employed.

Those analogs tested were run by using the fellowing methods: (a)
Freshly prepared enzyme solution was mixed with the analog and the scan
begun immediately upon mixing. Approximately 2-3 seconds lapsed between
mixing and starting the scan, but, since the reactions were relatively
slow, this proved to be no handicap. Samples were scanned for 5 minutes
with data points taken every 3.00 seconds. (b) A second sample of the
analog was incubated at 31.0° for two hours prior to the run. Fluores-
cence readings were taken fer 10.0 seconds to check fer the presence of
reduced 1,4-dihydropyridinium in the analog preparation. Baseline
checks were run on buffer solution alone, buffer and analog alone, buffer

with enzyme added, and buffer, purified NAD+, and enzyme to correct for

impurities that might affect the results.

28

.As a secondary check on the reactivity of the analogs an ultra-
violet spectrum was run on the samples by using Unicam SP800
spectrophotometer with a 1.0 cm cell. The spectra were recorded directly
as absorbance versus a constant buffer reference. Recordings were made
of sample absorbance alone, sample with added enzyme, and sample with
enzyme and ethanol as a check to determine whether the analog would
fUnction as a natural substrate in the liver ADH system. Readings were
recorded of analog alone in buffer solution,and analog with enzyme added
at time intervals t = 0, l, 3, 5, 7, and 9 minutes of mixing. Solutions
were transferred to small vials and 1ncubated at 31.0° with readings
made at 1.0 hour and 24.0 hour intervals. Baselines were run against
buffer solutions with care taken to maintain that baseline fer incubated
samples.

Separate samples containing analog and ethanol were run at t = 0, l,
3, and 5 minutes after addition of enzyme. No incubation of these samples
was attempted.

A diagnostic check was made by observing the fermation of the
fluorescent cyanide addition complex reported to show an absorption in-

crease at 325 nm similar to the enzymatic reduction of NAD+ to NADH.(44)

B. Materials and sample preparation

 

Samples used for both absorption and fluorimetry were prepared in a
similar manner. .A 1.50 ml aliquot of the purified coenzyme analog was
diluted prior to use with 0.10M sodium phosphate buffer which was 0.01M
in semicarbazide-HCI to a total volume of 3.00 ml. The buffer had been
previously titrated to pH 7.50 :_0.05 using a 0.10M sodium hydroxide.

Solutions were allowed to equilibrate at ambient temperature fer 15 minutes

29

before study without added enzyme or incubated with enzyme added for
the appropriate time. Ethanol, when present, was added as soul of a
95% solution.

Horse liver alcohol dehydrogenase (Sigma, liver ADH, E.C. 1.1.1.1)
was obtained as the 1yophilized powder containing 20 units activity and
reconstituted using 1.00 ml buffer stock solution. The enzyme solutions
were used without further purification since a fluorimetric scan of a
solution containing purified B-NAD+ (Sigma) and the enzyme showed no
fluorescence increase due to the presence of oxidizable impurities. One
unit, SOul, of enzyme was added to all samples checked fer reactivity.

Boiled, deionized-distilled water was used in preparation of all
solutions used throughout the study. Samples used in the cyanide test
were prepared by dissolving 0.60-0.80 ml in enough 1.0M potassium
cyanide solution (pH 12.0) to bring the total volume to 3.00 m1.(44)

Spectra were run against a potassium cyanide blank.

Results and Discussion

The work presented here represents a continuing effort to under-
stand the mechanism.of enzymatic oxidation and reduction in one of the
earliest discovered and most studied systems in biological chemistry.

The initial approach to the problem, first suggested by Karabatsos, was
to restrict the many degrees of freedom allowed the substrate. This move
toward eliminating some of the variables was first attempted with amino
alcohols possessing a primary amino and primary hydroxy function.(32’42)
Many of these primary amino alcohols were commercially available and it
was believed that through isotopic labelling experiments, stereochemical
assignments could be obtained. Once the initial reactivity of these
coenzyme analogs had been established, we realized that amdno alcohols
bearing a secondary hydroxyl group would simplify the problem of stereo-
chemistry by eliminating, fbr a time, the need fer labelling experiments.
The absolute configurations of a large number of asymmetric alcohols have
been in the literature fer years. ‘Methods fer the resolution of racemic
compounds also appear throughout much of the current literature; and, if
a general method fer the preparation of these so-called secondary amino
alcohols could be obtained, it remained then a simple matter of relating
the stereochemistry to findings already established.

Since no commercial preparation of secondary amino alcohols was

available, work was started by Maya’é)

to outline a good general method
for the preparation of these compounds in reasonable yields. Initial
attempts to synthesize secondary amino alcohols resulted in low yields
fer several reasons: (a) the lack of selectivity in group transfermation

(b) the tendency towards cyclization due to the nucleophilic character
30

31

of these groups, and (c) the extreme water solubility and hygroscopicity
of the compounds which complicated isolation procedures.

The synthetic route in equations 1-4 below, appears to have solved
the problems mentioned. All compounds used in this study have been

prepared by these methods and obtained in yields greater than 50%.

1. CHsCOCH CO C H + X- (CHZ)n-CN —-> CH COCHCO C H

2 2 2 5 3 ' 2 2 5
(CH2)n-CN
2. CHSCOCHCOZCZHS -——+» CHSCOCHZ-(CH2)n-CN
(CH2)n-CN
OH
I
3. CHSCOCHz-(CHZ)n-CN ——a» CHSCH-CHZ-(CH2)n-CN

OH OH

I
CHCHZ-(CH2)n-CHZNH

4. CHSCH~CH2-(CH2)n-CN' m—a- CH3 2
variations in the alkyl chain length, n, and the halide, X, have rela-
tively little effect in the preparation of these compounds.

The oxidized nicotinamide adenine dinucleotide analogs were prepared
by a method outlined by Kaplan(45) and isolated by column chromatography
on a PEI cellulose column. Eluates were monitored by untraviolet
spectrophotometer and the fractions identified as (1) unresolved
nicotinamide and N-(alkylhydroxy) nicotinamide, (2) NAD+ analog, and
(3) unreacted NAD+ (Figure 16). The retention times and peak elution
profiles fer all analogs isolated were identical. Additionally, thin
layer chromatography of crude reaction mixtures on prefermed PEI cel-
lulose 300 (UV254) plates developed with 0.15M ammonium chloride were
shown to have Rf values fer each of the fractions which were essentially

identical. Yields were typically 20-30% by this method.

32

(L6?

(L5

9
(d

ABSORIANCE
c:
N

 

 

 

0.1 .
1 2 3 p
0 25 45 65 85 105 125
ml cums

Figure 16. - Elution curves for analog purification

Preliminary characterization of the analogs was made by analogy to the
parent system. All compounds obtained showed a Amax at 260 nm and
formed fluorescent addition complexes with cyanide leading to an increase
in absorbance at 325 nm with a corresponding increase in extinction at
260 nm (Figure 17). This ability to form a complex with cyanide is a
general property of N-substituted nicotinamide analogs as wellas NAD+
and has been used as the basis for the analytical determination of NAD+
(44)

in unknown samples .

An earlier study in this laboratory(42) showed that all the analogs

33

 

 

 

 

 

If)
-I N
0')
E
-* i.
I
'l—
0
2:
I.“
.J
"'I LU
>>
<(
3
O
- W
N
(V l—
O
-‘ O
N
oi —-= 6 O
BDNVSUOSHV

Figure 17. - Typical cyanide addition complex of NAD+ analogs

1. NAD: analog alone
2. NAD analog with cyanide

34

from C5 - through C7 (i.e.: possessing an alkyl chain with 5-7 total
carbon atoms) underwent complete reduction in pyrophosphate buffer with
liver.ADH and ethanol on incubation. That study was extended to include
the highly purified substrates through C9 in phosphate buffer and those
original findings have been confirmed. This indeed shows that the
analogs can fUnction alone as active cofactors in the natural system;
although at a much slower rate in most cases. The longer chained N-
alkylhydroxy derivatives appear to react at nearly identical rates as
NAD+ itself when ethanol is added (Figure 18), and without the need for
incubation.

Information regarding the intra (or inter) molecular reaction of

the substrates without external ethanol added is much less conclusive.

Because of the small scale on which the reactions have been run (ca. 3 x
-7

 

10 to 8 x 10-5M) and an unfavorable equilibrium constant,
4.
K = [NADH][CH3CHO][H ].b 8 x 10-1%8
eq ’
[NAD+] [CHSCHZOH]

fer the ethanol system,(46) the accurate determination of any reaction
by ultraviolet methods has proven inconclusive. The general phenomena
of increasing extinction at 340 nm could not be accurately measured since
the sample concentrations were well below - or at the extreme lower
limits of - the sensitivity of ultraviolet spectroscopy. In the case of
N-(6-hydroxyalkyl)- and N-(8-hydroxyalkyl)nicotinamide adenine dinucleo-
tides a decrease at 260 nm was observed on incubation of the substrates
with enzyme for 24 hours (Figure 19). No usch decrease was seen for any
of the other tested substrates.

Due to the greater sensitivity of fluorescence analysis (10'9 to

lO-IJND it was decided that infermation could be obtained as to initial

3S

 

 

 

 

 

 

 

0

«an

m

t",
N

E
_. =1
5
0
Z
I.“
.4
g‘
<
3

o

-In

N

n I l
c. 0. o
N

BDNVGHOSBV

Figure 18. - Spectrum of N- (6-hydroxyheptyl)NAD+ with added ethanol

1. Increase in absorbance after 1 min.
2. Increase in absorbance after 3 mins.
3. Increase in absorbance after 5 mins.

36

 

 

 

 

 

 

 

 

(3
U1
V
O
u)
('0
E
1.
3E
.—
0
; I!
I/ 3
[LI
r >'
<(
3
A .
.\ _. a)
(‘4
\%§\
\\\‘j,»
l.
o. 0. 0.
N F- O
ADNVBUOSBV

Figure 18. (cont.) - Spectrum of B-NAD+ with added ethanol

37

 

 

 

 

I 8

N
E
1
3E
.—
§ 6
i!
I-IJ
.1
“J
i
3

N I— n

O

_. In

N

/.
/
K
1 l l
o, in 0. ‘9. 0.
N l- '- O O
ADNVBUOSSV

Figure 19. - Spectrum of N-(6-hydroxyhepty1)NAD+ without ethanol

1. Analog alone
2. Analog with HLAD after 1, 3, 5, 7, and 9 mins.
3. Analog with HLAD after incubation for 24 hrs.

38

velocities under extremely dilute conditions. Spectra were run for five
minutes of continuous scan on a computer centered instrument which gave
simultaneous, corrected fluorescence and absorbance data. Scanning time
was restricted to five minutes due to instrumental limitations. This
process attenuates the excitation beam such that the detector observation
geometry changes as the concentration of the absorbing species changes
which eliminates wavelength dependent errors.(43) As a check on the
reaction beyond the five minute limitation, samples were incubated two
hours prior to investigation with enzyme in buffer that was 0.01M in
semicarbazide. It was hoped that by trapping the carbonyl as it was
fermed, the equilibrium could be shifted enough to determine whether a
reaction had occurred. The results of these experiments are shown in
Table 7. No attempts have been made to quantitate the data with repet-
itive experimentation. Of the compounds tested only 6-HHNAD+ and 8-

HNNAD+ showed any significant reactivity, thus confirming preliminary

Table 7. - Fluorescence study of analog reactivity

 

Compound* Fluorescence increase

 

t = Smins t = 2hrs

 

4-HPNAD“ - -
5-HHNAD+ - -
6-HHNAD+ + +
74101141)+ - -
8-HNNAD+ + +

 

* Abbreviations used: 4-HPNAD+= N-(4-hydroxypenty1)NAD+
5-HHNAD+ = N-(S-hydroxyhexyl)NAD+; 6-HHNAD+ = N-(hydroxyheptyl)NAD+
7-H0NAD” = N-(7-hydroxyocty1)NAD+; 8-HNNAD+ = N- (8—hydroxynony1)NAD+

39

work done by ultraviolet spectroscopy. Studies made with NAD+ and yeast
ADH on alcohols of varying chain lengths lend some support to these
findings. Kaplan observed a similar alternating trend for the free
alcohols when the chain length was varied.(18)

Space filling models of the various compounds would seem to indicate
that 4-HPNAD1+ would have an alkyl chain too short to place the hydroxyl
group near the presumed reaction site. The fact that this compound does
not show a reaction under the conditions outlined is encouraging in terms
of tentatively eliminating a possible intermolecular reaction. This
same argument may be used for 7-HONAD+ since the alkyl chain length would
undoubted enable the hydroxy bearing carbon to fit into the enzyme active
site. Admittedly these are naive assumptions which require more exten-
sive investigation.

Several additional anomolies still exist and much more information
will be required. Essentially no infermation is known about the pH
dependence, buffer effects, or inhibition due to enzyme or cofactor con-
centrations in the systems studies. It is fortunate that work done here
supports that done previously in this laboratory under significnatly dif-
ferent conditions.(32’33’42)

No conclusions should be drawn, however, from work done to this point
other than that the substrates prepared do mimic the natural system without
the necessity of added substrate. Considering also that the N-alkylhydroxy
analogs of NAD+ were reacted as d,1 mixtures, the influence of one or
the other of these forms could drastically alter the reactions shown here.
werk is currently being done to optimize the preparation and isolation of
all the analogs as well as the resolution of the racemic aminoalcohols.

It is hoped that future work will soon be able to answer the many unsolved

jproblems presented here.

BIBLIOGRAPHY

 

10.
11.

12.
13.
14.
15.

16.
17.

18.
19.

20.

21.

1,5,, 5018 (1953).

Bibliography

0. warburg, W. Christian, and A. Griese, Biochem Z, 282, 157 (1935).
A. Todd, J. Roy. Inst. Chem., 82, 309 (1958).

 

F. A. Loewus, F. H. Westheimer, and B. vannesland, J. Amer. Chem. Soc.,

 

 

H. R. Levy, F. A. Loewus, and B. Vannesland, J. Amer. Chem. Soc., 79,
2947 (1957).

R. V. Lemieux and J. Howard, Can. J. Chem., 51, 308 (1963).

 

V. Prelog, Ind. Chim. Belge., 27, 1309 (1962).

 

 

J. W. Cornforth, Bioch. Bioph. Res. Comm., 2, 371 (1962).
G. J. Karabatsos and H. Nunez, unpub. results (1971).

For a listing of Class A and Class B enzymes see R. Bentley, "Molec-
ular Assymetry in Biology," V61 II, Academic Press, New York, 1970,
Table I, pp. 6-9; Table II, p. 11 and N. 0. Kaplan in P. D. Boyer,

H. Lardy, K. Myrback, "The Enzymes," V01 3, Academic Press, New York,
1960, Table II, p. 115.

K. R. Hanson, J. Amer. Chem. Soc., 88, 2731 (1966).

 

J. W. Cornforth, R. H. Cornforth, C. Donniger, G. Popjak, G. Ryback,
and G. J. Schroepter, Jr., Proc. pr. Soc. Ser. B.,,lg3, 436 (1966).

 

V. Prelog, Pure Appl. Chem., 9 (l), 119 (1964).
E. M. Kosower and D. C. Remy, Tetrahedron,,§, 281 (1959).

 

E. M. Kosower, Bioch. Biophys.Acta,,36, 474 (1962).

 

J. van Eys, F. E. Stolzenbach, L. Sherwood, and N. 0. Kaplan, Biochem
Biophys. Acta,,€z, 63 (1958).

 

H. Theorell and R. Bonnichsen, Acta. Chem. Scand.,,§, 1105 (1951).

 

G. J. Karabatsos, J. S. Fleming, N. Hsi, R. H. Abeles, J. Amer. Chem. Soc.,
88, 849 (1966) .

 

J. van Eys and N.O. Kaplan, J. Amer. Chem. Soc., (’9, 2782 (1957).

 

 

V. E..Althouse, K. veda, and H. S. Mbsher, J. Amer. Chem. Soc., 82,
5938 (1960).

R. MacLeod, H. Prosser, L. Fikentscher, J. La-yi, and H. S. Mosher,
Biochem.,,3, 838 (1964).

V. B. Althouse, D. M. Fiegel, and W. A. Sanderson, and H. S. Mosher,
J. Amer. Chem. Soc., 88, 3595 (1966).

 

4O

22.
23.

24.

25.

26.
27.
28.
29.
30.
31.
32.

33.

34.

35.

36.

37.
38.
39.

40.
41.
42.

41
C. Donniger and G Ryback, Biochem. J., 9,1,, 115 (1964).

 

A. Gunther, F. Biller, M. Kellmer, and H. Simon, Angew Chem. Inter.
31., 1,2, 146 (1973).

 

P. A. Frey, G. J. Karabatsos, and R. H. Abeles, Biochem Biophys Res. Comm.,
118, 551 (1965).

 

B. Zagalak, P. A. Frey, G. J. Karabatsos and R. H. Abeles, J. Biol. Chem.,
214/1, 3028 (1966).

 

. Neuberg, Adv. Carbohyd. Chem., ,4, 7S (1949).

 

. Neuberg, and F. F. Nord, Bil}, 512/13,, 2248 (1919).
. A. Levene and A. Walti, J. Biol. Chem., 94, 361 (1933).

 

Grzycki, Biochem. 2., 2,85,: 195 (1933).

 

Fumise, Biochem. 2., 2,36, 231 (1931).

 

. V. Lemieux and J. Giguerre, Can. J. Chem., 2,9, 678 (1951).

 

. Nunez, unpub. results 1972.

C. Stamoudis, Ph.D. Thesis, Michigan State University, 1973, Table 14,
57.

"d< {1370951) "U 00

See B. M. Anderson and N. 0. Kaplan, J. Biol. Chem., , 1226 (1958) and
refs. therein for prep. , rel. rates of redn. , and Amax or other NAD+
Analogs; also H. Sund, "Pyridine Nucleotide Dependent Dehydrogenases,"
Springer-Verlag, Heidelberg, 1970, Table I, pp. 606-607.

 

A. J. Gordon and R. A. Ford, "The Chemists Companion," J. Wiley and
Sons, New York, 1972, p. 433.

a) Prepared via Michael Condensation with acrylonitrile, see M. K. May,
M.S. Thesis, Michigan State University, 1973, p. 12.

b) Ibid., Idem., pp. 17-18.
N. F. Albertson, J. Amer. Chem. Soc., L2,, 2594 (1950).

 

A. P. Krapcho and A. J. Levy, Tet. Lett., 957 (1973).

. F. Fieser and M. Fieser, "Reagents for Organic Synthesis," Vol I,
. Wiley and Sons, New York, 1967, p. 725.

. A. Pavlic and H. Adkins, J. Amer. Chem. Soc., 6,8, 1471 (1946).

 

L
J
R. Monzingo, "Organic Synthesis Coll. Vol II," 181 (1955).
A
J

. Miedema, Ph.D. Thesis, Michigan State University, 1974.

43.

44.

45.

46.

42

J. F. Holland, R. E. Teets, and A. Timnick, Anal. Chem., 4,5 (l), 145
(1973).

 

S. P. Colowick, N. 0. Kaplan, and M. M. Ciotti, J. Biol. Chem.,,lgz,
447 (1951).

 

L. J. Zatman, N. 0. Kaplan, S. P. Colowick, and M. M. Ciotti, J. Biol.
Chem., 289, 467 (1954).

H. Theorell, A. P. Nygaard, and R. Bonnischen,.Acta. Chem. Scand., 9,
1148 (1955).

 

y ., it . 5.6!...

‘ .4ll!’ ~1.'."nf' .l‘r E‘i-

"IIIIIIIIIII'IIIIIIIII