^*- II

1B3Q^^^^^£

] Marine Biological Laboratory Library [

B Woods Hole, Mass. E

n ra

n

I

JJ Presented by li

] The Publisher C

II [0

D John Wiley and Sons, Inc

a August 1965 11

II II

a

iE3^^^^^^3^^^^^^^SI

Metabolic Pathways
in Microorganisms

E. R. SQUIBB LECTURES ON

Presented at the Institute of Microbiology
Rutgers, the State University of New Jersey

F, M. Strong, Topics in Microbial Chemistry, 1956

F. H. Stodola, Chemical Transformations
by Microorganisms, 1957

V. H. Cheldelin, Metabolic Pathways

in Microorganisms, 1960

Roger J. Williams

CHEMISTRY OF MICROBIAL PRODUCTS ""

Metabolic
Pathways
in Microorganisms

By VERNON H. CHELDELIN

Director, Science Research Institute
Oregon State University

NEW YORK • LONDON, JOHN WILEY & SONS, INC.

Copyright © 1961 by John Wiley & Sons, Inc.

All rights reserved.

This book or any part thereof must not be reproduced in
any form without the written permission of the publisher.

Library of Congress Catalog Card Number: 61-16617
Printed in the United States of America

In recognition of the importance of cooperation between
chemist and microbiologist the E. R. Squibb Lectures on
Chemistry of Microbial Products were established with the
support of The Squibb Institute for Medical Research in
1955. The lectures are presented annually in the fall at
the Institute of Microbiology, Rutgers, the State University
of New Jersey, New Brunswick, New Jersey.

PREFACE

I am happy to have the opportunity to present this work on
pathways of carbohydrate metabolism, which has been carried
out, for the most part, in our laboratories (Science Research
Institute, Oregon State University) during the last eight
years. This w^ork emphasizes heavily the peculiarities of me-
tabolism that characterize the acetic acid bacteria, as well as
emphasizing observations on microbial systems, which lend
themselves especially to radiorespirometric studies. Finally,
it presents a general discussion on various aspects of metab-
olism, which seems appropriate in this series of lectures.
Because of the personal nature of the experiences described,
this will not represent an attempt to review exhaustively
the literature in the field, although it is hoped that this
work may be of some aid in that direction.

Being selected as a lecturer to continue the series ably
begun by the other speakers in the Squibb series produces
in one mixed feelings— diffidence, yet at the same time the
need for confidence to provide some useful thoughts about

PREFACE

the subject chosen for discussion. The first reaction is
spontaneous, and is so well recognized by everyone as to
require no further elaboration. In marshaling my confi-
dences to present this material, I am reinforced by some
of the thoughts laid before me in my graduate study by
Professor Roger J. Williams, to whom these lectures are ded-
icated, plus the sound, carefully fabricated experimental
work by many of my colleagues over a period of fifteen
years. Among these are Drs. Tsoo E. King, Chih H. Wang,
and R. W. Newburgh, to mention only three, and also
many former graduate students who have contributed to
the over-all knowledge about the problem, especially Drs.
Jens G. Hauge, Paul A. Kitos, and Joseph T. Cummins.
To these persons, and others of our group, I am deeply
grateful.

V. H. Cheldelin
August, 1961

CONTENTS

CHAPTER 1

The Acetic Acid Bacteria 1

CHAPTER 2

Evaluation of Metabolic Pathways 30

CHAPTER 3

Carbohydrate Metabolic Pathways:

General Considerations 64

Index 89

CHAPTER

THE ACETIC ACID BACTERIA

These bacteria, like several other species of microorgan-
isms, have been utilized by man since antiquity; wild cul-
tures invade cider, for example, to promote the acetic acid fer-
mentation, and thus have been responsible for commercial
vinegar production.

Acetobacter suhoxydans, one of the better known mem-
bers of this group, was discovered in 1924 by Kluyver and
de Leeuw (1). Its discovery did much to broaden the
industrial uses for the acetic acid bacteria, for in addition
to the vinegar fermentation it carried out several others,
mostly one- or two-step oxidations of polyols (2):

Mannitol — > fructose (1)

Sorbitol — » sorbose (2)

Glycerol — > dihydroxyacetone (3)

Erytheritol -^ erythrulose (4)

Glucose -^ gluconic acid — ^ ketogluconate (5)

2,3-Butanediol — ^ acetoin (6)

1

2 METABOLIC PATHWAYS IN MICROORGANISMS

as well as several others. The organism was at first regarded
as a most versatile one, being adaptable to so many sub-
strates; but, it may be more nearly correct to consider it as
a metabolic "cripple," not particularly better able to attack
substrates than many other organisms, and unable to oxi-
dize these beyond the first one or two steps. Perhaps the
chief reason for this is the fact, which will be discussed
later, that the organism has no Krebs cycle, and thus lacks
the "prairie fire" of terminal oxidation that most other
organisms, whether men or mice, enjoy.

Our interest in this organism began, however, not with
the oxidations that it carried out, but with finding a
reason for its unusual pantothenic acid requirement. An
earlier lecturer in this series, Dr. Frank M. Strong (3), has
described much of the literature dealing with coenzyme A,
so I will dwell on it only for a moment. Suffice it to say
here that this organism is ten to twenty times as sensitive
to bound forms of pantothenic acid [coenzyme A, pante-
theine {L. bulgaricus factor, LBF), pantothenyl cysteine] as
it is to the free vitamin (4). This fact was discovered in
our laboratory, where it gave rise to the description of a
pantothenic acid conjugate which we abbreviated PAC
(5). This conjugate was not fully characterized, but it is
now regarded as a fragment of the coenzyme A molecule.
The enhanced activity of conjugates of pantothenic acid
toward A. suhoxydans has been of aid to various investiga-
tors in their studies of derivatives of pantoic acid that lead
to coenzyme A (3, 6, 7).

The observed superiority of coenzyme A over the free
vitamin as a growth promoting agent, may be rationalized
by the fact that cells grown deficient in pantothenate (and
hence coenzyme A) have a lower lipid content than normal

THE ACETIC ACID BACTERIA

cells. In view of the known function of this coenzyme
in fat formation (8) this seems reasonable. Other activities,
such as glucose oxidation, appear not to be influenced
by coenzyme A deficiency, although the oxidation of glyc-
erol is markedly reduced (see Fig. 1.1) (9) in a manner
that is not yet understood.

100
Time (minutes)

200

Fig. 1.1. Glycerol oxidation by pantothenate-deficient A. suboxydans
cells. Deficient cells were grown in 0.0006 y pantothenic acid equiv-
alent per milliliter of medium. This concentration permitted growth
of only 0.33 g. per liter, in contrast to sufficient cells which were grown
in excess CoA and yielded a crop of over 1 g. per liter.

METABOLIC PATHWAYS IN MICROORGANISMS

CARBOHYDRATE AND POLYOL OXIDATIONS

Testing of various carbon compounds soon revealed that
several were inert, whereas others were susceptible to at-
tack to only a limited degree. Table 1.1 indicates the
extent of such oxidations. Where extensive oxidation oc-

TABLE 1.1
Oxidation of Various Substrates by A. suboxydans

O2 Consumption: ^tatoms/)umole

Substrate

Glycerol

Dihydroxyacetone

Ethanol

Acetaldehyde

Sorbitol

Erythritol

Pyruvate

Lactate

Acetate

Ketoglutarate

Malate

Succinate

Fumarate

Citrate

The systems contained 0.05M phosphate, 0.01 A/ MgCl2, 10-^A/
DPN, and 10 mg. dry weight of washed cells. Volume = 2.8 ml.,
pH = 6.0, temperature = 29° C. The substrate was tipped into the
main compartment containing dinitrophenol (lO^^Af) after 5 minutes'
preincubation. All values corrected for endogenous blanks, which
were about 0.1 fiatom oxygen//xmole substrate.

Dinitrophenol

Addition

1 X lO-^M

3.7

1.0

3.0

0.3

1.9

2.0

1.0

4.0

1.2

1.0

0.9

2.0

0.4

THE ACETIC ACID BACTERIA

curred, as in glycerol or sorbitol, this could be reduced to
one atom of oxygen per molecule of substrate by including
dinitrophenol in the medium. The further oxidation of
the one-step oxidation products (dihydroxyacetone or sor-
bose) was virtually completely repressed in dinitrophenol
solutions, which presumably prevented coenzyme-linked
phosphorylations.

The cells were therefore broken, and cell-free prepara-
tions were made. This was accomplished either by dis-
integrating the cells in a 10-kc. Raytheon sonic oscillator
or by grinding with alumina. The broken cell suspension
was mixed with phosphate buffer and centrifuged at 20,000g
for an hour; then the residue was re-extracted and the com-
bined extracts pooled.

Particulate Enzymes

These extracts revealed that the oxidizing enzymes of
A. suboxydans varied considerably, more than had been
suspected. Two, and sometimes three, systems existed side
by side in the organism for the breakdown of individual
polyhydroxy compounds. There were, for example, a num-
ber of particle-bound, phosphate-independent dehydro-
genases (10) which oxidized mannitol, sorbitol, erythritol,
glycerol, and glucose to the extent of one atom of oxygen
per molecule of substrate. Two atoms of oxygen were
used per molecule of ethanol or propanol. The dehydro-
genases appeared different from each other, since purifica-
tion of the particulate suspensions effected a ten-fold in-
crease in the concentration of glucose dehydrogenase (oxi-
dase), four-fold for erythritol, three-fold for glycerol, but
did not affect the concentration of ethanol dehydrogenase.

8-D-Gluconolactone has been indicated as the product of

METABOLIC PATHWAYS IN MICROORGANISMS

H— C=0

I
H— C— OH

I
HO— C— H

I
H — C — OH

I
H— C— OH

I
H— C— OH

I
H

-2H

Flavin?
Cytochrome?

c-

I

c-

I

HO— C-

//

H-

■OH
-H

-OH

H

H— C-

I
C

I
H— C— OH

I
H

HoO

^.

-OH

H— C— OH

I
HO— C— H

H-

-OH

H— C— OH

I
H— C— OH

I
H

6 - D - Gluconolactone

Gluconic acid

Fig. 1.2. Oxidation of glucose by particulate glucose oxidase in
A. siiboxydans.

D-glucose oxidation by a particulate enzyme (see Fig. 1.2).
The optimum pH was 5.5 (11). A separate enzyme also
exists in this fraction, which hydrolyzes the lactone to d-
gluconic acid. Deoxycholate extracts of the particulate
fraction retained activity for oxidation of glucose, but they
no longer contained the hydrolyzing enzyme.

"Soluble" Enzymes

In addition, the cell-free extracts contained several soluble
enzymes that cooperate to effect the terminal oxidation of
glucose, other carbohydrates, and polyols. These were, for
the most part, phosphate-dependent and DPN- or TPN-
linked. Upon fractionation, the extracts were found to
contain the entire pentose cycle complex of enzymes (Fig.
1.3).

THE ACETIC ACID BACTERIA

ATP

Glycerol-a-P04

DPN

DHA

GLA-P ^

^ R-5-P

DPN
TEN

6-PGA ^

DPN

TPN

-> FDP

Mg

F-6-P

G-6-P

Fig. 1.3. The pentose cycle in A. suboxydans.

Glucose. The first example of a soluble dehydrogenase
is an exception to the generalization just stated: phosphate
does not participate in glucose oxidation to gluconate,
though pyridine nucleotide (TPN) does. The soluble de-
hydrogenase catalyzing this oxidation is a separate enzyme
from the particulate enzyme already discussed, since it has

8 METABOLIC PATHWAYS IN MICROORGANISMS

its pH optimum at 8.6. We have purified this enzyme
about 100-fold and have found it to be strictly TPN-specific.
The reaction is reversible, as tested with 8-D-gluconolactone
and TPNH, but the y-lactone is less active. Among the
sugars and their phosphates tested, only D-glucose and 2-
deoxy-D-glucose were oxidized, whereas glucose-6-phosphate,
glucose- 1 -phosphate, gluconolactone, and gluconic acid
were not attacked. This enzyme therefore appears different
from those described by de Ley and Stouthamer (12) for
this organism in the oxidation of gluconate.

Glucose-6-Phosphate and 6-P ho spho gluconate. These
two soluble enzymes usually are closely associated in A.
suhoxydans, as they are in most organisms. They have
been highly purified, to the point that they catalyze the dis-
appearance of about 100 and 50 /xmoles of substrate per
minute, respectively, per milligram of enzyme. With a pH
optimum at 8.0, the turnover numbers for G-6-P dehydro-
genase are about equal with DPN and TPN, whereas DPN
is much superior with 6-phosphogluconic dehydrogenase.
The purifications are effected by combinations of protamine
precipitation, ammonium sulfate fractionation, calcium
phosphate gel adsorption, and column electrophoresis (13).
A comparison of properties of G-6-P dehydrogenase from
various sources is given in Table 1.2, whereas Table 1.3
contains a similar listing of properties of 6-PGA dehydro-
genases.

For clarification, the enzymes that oxidize glucose or glu-
cose-6-phosphate, as reported from different laboratories,
are listed in Table 1.4 along with some of their distinguish-
ing properties. It is evident from Table 1.4 that A.
suboxydans is well endowed with separate enzymes all of
which attack these substrates. The reasons for this diversity
are not yet clear. It is possible that there is little or no

THE ACETIC ACID BACTERIA

^

un

fa

o

G

S-i

C^

t;

.bp

g

z

1

P4

V

G

«H

4;

^~^

G

9

1

3

T T T

XXX

'g

be

C/2

G

CN

CO

CN

h

so

CO

II

%

^

>-

X

CM

1

9i

Q

a

i

c;

sO

^

3

cG

«

t

.§K

in

06

q

'u

H,

a

00

r-;

06

^

r-'

CO

^

&N

i

i

en

tu:

eg

) u

*

-0
00

r^

CM

en

CQ

0)

K

C/2

1

"^

V
faD

^

3

^

H

^

^

j:3

•M

G
.0

,. — s

>^

Q

.^

^

G

*i5

"0

"c^

CN

T— 1

^

.2

1

<2

0"

-a

C8

a

S

08

^

u

Cli

^'

CJ

5J

>.

bXD

«M

a

bic

tf.

.Q

N

G

1

■ G

^-0

■55

.*

u

C/2

&H

Oh PL, 0^

G

'u

h

Hhh

G
n

P £^ «3 ^ G .

g ^ - 1 s ^ ^ .a

ao

^

iS

10 METABOLIC PATHWAYS IN MICROORGANISMS

^i> o r- 00 u-5

o

03 CI. pH

gj ^^O OO OO

• &9 ^^_^vo or-

CD

b/) bO

W

2 ^ X XX XX

^^ in 00 CO CO en

T3

w B^^^oor-o u^

go ^ ^

o

j:3

.2 a

5 '-'II

i g I I ■= •

O ^ s ^ 3 c

THE ACETIC ACID BACTERIA 11

^ m (N CN] (N

^ S ffi "1 ^ q I I I q

O w ex LO 00 00 I I I 00

6£ fl * V

a 5 i; 1) ■'-'

.^O ^ a V V V V ci

0-5 2 o O "bfi "He "biD "be o

k>CL to ^ so CN CN in Ph

-jet- «j >. ii B

3§is|2Q§^ H^.

G

O

C^i 03OO03OOO

OJ ^ tL> ^ "^2 -^

a

fa i

- - o o u ii

SS .a 33. a 333

.-5 w c^ o o «^ o o o

b£

©

3

so ^ ^ ^

A cd c^ c^

So C G G

000000

biD

J2

C/3 ^3:3G3G^

O O O O O O o

12 METABOLIC PATHWAYS IN MICROORGANISMS

free passage of glucose or its phosphate from one oxidizing
system to another, yet this imphes that no "pool" of glu-
cose derivatives exists. While this has neither been proved
nor disproved, it seems not to be a tempting conclusion.
In the organism B. suhtilis, kinetic experiments with iso-
topically labeled gluconate suggest strongly that a "pool"
of glucose exists, and that gluconate when reconverted to
hexose via the pentose cycle is labeled in accordance with
expected patterns, as will be seen in the next chapter. Spa-
tial separation might seem to provide a plausible explana-
tion for the plethora of separate enzymes that exists for
oxidation of individual, or closely related, substrates.

Dihydroxyacetone Phosphate, Dihydroxyacetone phos-
phate was converted, after its formation from glycerol, to
fructose 1, 6-diphosphate through the action of the iso-
merase-aldolase system, which was found to be very active
in this organism (19). With dihydroxyacetone as the start-
ing material, it was possible, in the presence of ATP and
Mg- + , to show the formation of fructose in amounts (as
measured by the resorcinol test) approaching theory based
on the ATP present, assuming activity of ADP. The re-
actions could also be followed through measurement of the
inorganic phosphate released. Under oxidative conditions
(DPN and triphenyltetrazolium) a further, slower release of
P^ was observed, as would be expected if the pentose cycle
were operating, and hexose accumulation dropped to about
one-tenth of the amount formed non-oxidatively. In the
presence of Mg- + , a material was formed that produced two
paper chromatogram spots characteristic of an authentic
mixture of glucose-6-P and fructose-6-P. The scheme is
shown in Fig. 1.3, together with subsequent reactions of the
pentose cycle.

When the oxidation of glucose-6-phosphate was followed

THE ACETIC ACID BACTERIA 13

manometrically (by the cell-free extract), the CO2 produc-
tion lagged initially, compared to the Oo consumption.
Chromatogiaphy of the oxidation products revealed zones
corresponding to 6-phosphogluconate and ribose-5-phos-
phate. Both DPN and TPN appeared active in the dehy-
drogenases for glucose-6-phosphate and 6-phosphogluconate
(see Table 1.4).

The transketolase-transaldolase reactions that characterize
the pentose cycle were demonstrated by chromatography
and the appropriate color reactions. Sedoheptulose was
determined by the cystine-sulfuric acid reaction, at 415 lUfi
and 505 uifx, and ribose was measured with orcinol. Sedohep-
tulose and the phosphates of fructose, glucose, and dihy-
droxyacetone were also measured chromatographically.
The recovery of total sugar and the measurement of each
sugar derivative with time is shown in Table 1.5, where
ribose-5-phosphate is added non-oxidatively to the cell-free
extract and its rate of disappearance is followed together
with appearance of other sugars.

Other ancillary reactions leading into the pentose cycle,
such as kinases for ribose, erythritol, and glucose, have been
identified. The glucokinase has been partly purified (19).

Each reaction of the pentose cycle, plus related ones listed
in Fig. 1.4, has thus been documented in soluble extracts
of the organism. The quantitative importance of the pen-
tose cycle as a terminal respiratory mechanism in A. suh-
oxydans has been demonstrated (23) through the use of
specifically C^Mabeled glucose and gluconate as substrates
for aerated resting cells: for every 100 molecules admin-
istered 28 were oxidized to 2-ketogluconate, presumably
by the particulate dehydrogenases. Of the remaining 72,
63 (equals 88%) entered the pentose cycle. As a later
chapter will reveal, we have calculated that essentially all

14 METABOLIC PATHWAYS IN MICROORGANISMS

TABLE 1.5

Non-oxidative Breakdown of Ribose-5-Phosphate
by A. suhoxydans Extract

Time (Minutes)

2

5

20

80

Sugar

yuMoles Sugar

Pentose

10.0

8.9

7.3

4.4

0.8

Hexose

0.0

0.0

1.1

2.6

4.7

Sedoheptulose

0.0

0.4

0.7

0.9

0.6

Triose-phosphate

0.0

0.5

0.6

0.6

0.8

Total *

10.0

9.8

10.0

9.1

7.8

* Values for total sugars are given as pentose equivalents (jumoles
carbon/5).

The tubes were incubated in vacuo with 10 ^tmoles ribose-5-phos-
phate, 100 /xmoles tris buffer at pH 8.0, 20 jumoles sodium fluoride,
0.15 ml. cell-free extract, and water to 2 ml. The reaction was
stopped by adding 1.6 ml. of 10% trichloroacetic acid. Pentose was
assayed in the reaction mixtures according to Mejbaum (20), hexose
and sedoheptulose with the sulfuric acid-cysteine reaction (21), and
triose-phosphate as alkali-hydrolyzable phosphate. Magnesium was
omitted. [After Hauge et al. (22).]

of the CO2 produced from glucose arises via the pentose
cycle; in fact, A. suhoxydans is unique in this respect among
organisms studied to date. The active existence of the
pentose cycle in this organism makes more reasonable the
finding that the Krebs cycle appears absent. This latter
difference is the first clear one between A. suhoxydans and
related Pseudomonas species, several of which rely heavily
on the Entner-Doudoroff pathway for glucose breakdown
(24).

Other results, summarized in Fig. 1.4, which deserve spe-
cial mention are the following.

THE ACETIC ACID BACTERIA

15

ATP

Mg/

Glycerol Sorbitol Glucose

\ DPN / \ TPN

Glycerol-CK-P DHA D -Fructose L-Sorbose
DPN^

ATP

Glucose-6-P

Pentose
cycle

TPN

Fructose-6-P 6-P-Gluconate

/TPN
DPN

DPT

Mg

CH3CHO

DPT
DPN

CH3COOH

Fructose- 1,6-di-P Ribose-5-P

ATP
Ribose

O OH

II I

,A ., w f • ^ C— C— C— COOH ^Valine

(Acetolactate-forming enzyme I) 1

CH3

/ (Acetoin-forming enzyme I)

CH3COCOOH

CH,CHO

(Acetoin-forming enzyme II)

O OH

II I

^ c— c— c— c

CoASH

-^ CHaC^SCoA

No citrate

No acetyl
sulfanilamide

Fig. 1.4. Carbohydrate dissimilation in soluble extracts of A. sub-
oxydans.

16 METABOLIC PATHWAYS IN MICROORGANISMS

Glycerol, Glycerol enters the pentose cycle via dihy-
droxyacetone (Figs. 1.3 and 1.4). This conversion is reached
by two alternate pathways (25), both of which are in the
soluble fraction (soluble at 30,000g). One, active at pH
6.0, is independent of ATP and DPN, and yields dihydrox-
yacetone directly. The other, with a pH optimum around
8.5, requires the participation of ATP and Mg^+, in a
kinase reaction, to yield glycerol-a-phosphate. The latter
is then oxidized by a DPN-dependent dehydrogenase to
form DHA-PO4.

Sorbitol, Sorbitol (Fig. 1.4) is oxidized on either end of
the molecule by soluble extracts of A. suhoxydans, depend-
ing upon which pyridine nucleotide is present (26, 27). In
the presence of TPN, L-sorbose is formed; with DPN, d-
fructose is produced (see Table 1.6). The fructose can then
be phosphorylated or further oxidized via the pentose cycle.
These two pathways occur in addition to the previously
demonstrated sorbitol dehydrogenase in the particulate frac-
tion of the cells, which form L-sorbose.

The DPN-linked enzyme has been purified about sixteen-
fold so that it is free from the TPN (sorbose-producing)
enzyme, and is also free of mannitol dehydrogenase activity
(27). An effective purification step consists of heating the
enzyme in the presence of a polyhydroxy compound and
pyridine nucleotide; the sorbitol dehydrogenase is pro-
tected by these compounds, while many other proteins are
denatured and may be removed by centrifuging.

Although earlier work (30, 31) has suggested the presence
of some relatively non-specific polyol dehydrogenases in
mammalian tissues, the present experience with A. suhoxy-
dans indicates a higher degree of specificity at least as far as
sorbitol is concerned. The purified DPN enzyme, active
for sorbitol, was completely inactive toward mannitol.

THE ACETIC ACID BACTERIA

17

TABLE 1.6

Identification of Products of Sorbitol Oxidation
by A. suboxydans

Pyridine Nucleotide
Added

D-Fruc-

tose L-Sorbose
Standard Standard

0.93
-102

0.69
-48

1.00
-92

0.73

-42

DPN TPN

Position constant in

phenol-HsO (4:1) *
la]'S (28), degrees
Crystal form of

osazone (29) Rosettes Amorphous Rosettes Amorphous

Mehing point of

osazone (28), °C. 206 161.5-163 206 163

* The position constant is the distance traveled by the compound
divided by the distance traveled by fructose.

The reaction mixtures contained the following: 180 /umoles of
sorbitol, 10 jumolesof TPN or DPN, 3 ml. of CFE, 100 )umoles of TTZ,
500 )umoles of MgCl2, 1 mmole of tris buffer, pH 8.5. The total volume
was 10 ml.; temperature, 30°; time, 4 hours. 1 ml. of 50% TCA
was added, and the mixture v/as centrifuged and extracted with ether.
The [ajr? was calculated on the basis of sorbose and fructose deter-
mined both by the cysteine-H2S04 and resorcinol methods. [After
Cummins et al. (26).]

ribitol, dulcitol, perseitol, glycerol, ethanol, acetaldehyde,
or 2-butene-l, 4-dioL Mannitol was oxidized as extensively
as fructose by whole cells (8.7 atoms of oxygen per molecule
substrate) but this may reflect only the easy conversion of
mannitol to fructose by this organism. Mannitol oxidation
has been found in our laboratory to be completely TPN-
specific and completely separable from the TPN-sorbitol
enzyme.

The influence of pyridine nucleotide in guiding the oxi-
dation toward one end or the other of the sorbitol molecule

18 METABOLIC PATHWAYS IN MICROORGANISMS

is unusual, although a related directing of metabolism
toward glycolysis (DPN) or the pentose cycle (TPN) has been
noted in animal systems in our laboratory (32), as well as
by Wenner and Weinhouse (33).

Pyruvate, Pyruvate conversion to acetate in A. suboxy-
dans follows an interesting course (34, 35). Whereas in
animals and most other aerobic organisms, the highlight
of pyruvate oxidation is terminal consumption through the
Krebs cycle, in this organism there is ample opportunity
to recognize breakdown products from pyruvate that have
more than a transient existence. The details of formation
of some of the minor products also differ. Although in
animals and most bacteria, pyruvate is converted in small

TABLE 1.7

Comparison of Substrate Specificity of A, suhoxydans
and Yeast Pyruvic Carboxylase

Yeast *

Green

Kobaya

Acid Added

A. suboxydans

et al. (36)

(37)

Pyruvic

1.00

1.00

1.00

a-Ketobutyric

0.75

0.80

a-Ketoglutaric

0.00

0.01

0.04

Q;-Ketoisovaleric

0.00

0.88

0.26

a-Ketoisocaproic

0.00

0.05

Oxalacetic

0.7

0.32

0.54

Phenylpyruvic

0.00

0.00

* Recalculated from the data of Green et al. (36) and Kobayasi

A direct quantitative comparison is not possible because of differ-
ences in test conditions. The numbers represent relative rates of
decarboxylation compared to the rate with pyruvic acid. [After King
and Cheldelin (34).]

THE ACETIC ACID BACTERIA 19

measure to acetoin, in A. suhoxydans pyruvate is first de-
carboxylated to acetaldehyde and the latter compound is
then oxidized to acetate. A. suboxydans is thus one of a
very few with a yeast-type decarboxylation (also shared by
higher plants); it is probably associated with the failure
of the organism to form acetyl phosphate or acetyl CoA
in this organism (see below).

Pyruvic carboxylase has been prepared from this organ-
ism, in a high state of purity (turnover = 300 moles COo
produced/minute/1 00,000 g. of enzyme). Thiamin pyro-
phosphate and a divalent ion serve as cofactors. Such prep-
arations show a higher degree of specificity than does yeast
carboxylase, since the bacterial enzyme responds only to
a-ketobutyrate, oxalacetate, and pyruvate, as shown in
Table 1.7.

Acetaldehyde, The oxidation of acetaldehyde may pro-
ceed, as do several other oxidations in A. suhoxydans, by
two routes: one with TPN as coenzyme (more active), the
other with DPN, although it has not been possible to de-
termine with certainty whether two apoenzymes are pres-
ent (35). The specific activity of the purest preparations
[about 140 spectrophotometric "units" (Table 1.2)] is con-
siderably higher than any reported for yeast or liver.

Acetoin, Acetoin formation (38), as depicted in Fig. 1.4,
has revealed that this metabolite appears to arise differently
from pyruvate, acetolactate, or from acetaldehyde. We
believe that two acetoin-forming enzymes may be present:
one that employs acetolactate as the preferred substrate,
and the other which employs acetaldehyde. A final decision
on this point will probably have to await further fractiona-
tion of the partially purified enzymes.

Acetate, The utilization, or rather non-utilization of
acetate by A. suhoxydans has received much study. In ad-

29 METABOLIC PATHWAYS IN MICROORGANISMS

dition to the stoichiometry already quoted (23) which
demonstrates the preponderant use of the pentose cycle for
terminal oxidation of carbohydrate, there are several other
lines of evidence which indicate that acetate cannot be oxi-
dized by this organism. These are:

1. Added acetate is not oxidized in simple manometric
experiments by intact cells or by cell-free extracts, either
alone or in the presence of glucose or glycerol as a potential
"sparker."

2. When CHgCi^OOH is added to respiring cells, only
0.013% of the added C^* appears in the respiratory CO2,
even under conditions where 25% of the added C^* is
incorporated into the lipid fraction of the cells (see Table
1.8) (39).

3. A. suboxydans extracts can form acetyl CoA in good
yield by the ATP-acetate-CoA reaction, yet the acetyl CoA
formed cannot be converted to citrate or acetyl sulfanila-
mide. The organism evidently lacks a suitable acceptor
system; only when the acceptor fraction from pigeon liver
[Chou and Lipmann (40)] is added can acetyl sulfanilamide
be produced, and only when pigeon liver condensing en-
zyme is added can citrate be formed.

4. Pyruvate does not form acetyl CoA during oxidation
unless ATP is added (39). This suggests that free acetate
is produced before acetyl CoA is formed.

The Citric Acid Cycle, The statements in the four pre-
ceding paragraphs imply that the Krebs citric acid cycle may
not function in A, suboxydans. This is of course surprising,
for other species of Acetobacter such as A. pasteurianiim
(41) and A. aceti (42) have been shown to possess a full
complement of Krebs cycle enzymes, as indeed virtually all

THE ACETIC ACID BACTERIA

21

H

0)

fl .2

1^

u

^

o
u

_o

^tS

hO

Xo

XX

^3 2S
<^h o o

S2

-°^ XX

-^ so O

CN

T3 "

^ s

S-,

^ a
8>

C^ O

G so

■«> 2

a o

«gs

>u s
2 2^

or, •• <U

o fc J

"sf Oh V3

7, X O

^« 5:^

3h . be
O a; W5

s i

CO

V

^ O

" 9

<N

be o

>.^

. o

ho ,

B S
10 ^

4J

3
G

o .> .5

so C i^
. U V

o ^ ex
e o «->

O^

.2 o

« 'oQ "t; hn"

t; O hU

Xi

-G y O
-« " LO

T^ O <N

un G

is

G -G

" O

G O ^

^ O

^ O 03

^^
.S bo

a, .

+-' 1;
03 ^

22 METABOLIC PATHWAYS IN MICROORGANISMS

organisms do. Moreover, the idea that such a universally
important complex of metabolic machinery should be com-
pletely absent from any aerobic organism that can oxidize
glucose to CO2 and H2O, is not likely to gain full accept-
ance—even by those who work with the organism— without
careful checking. We have scrutinized this problem care-
fully, but all data point to the absence of a traditional citric
acid cycle in A. suboxydans [to the extent that these reac-
tions have been studied, they have in general been verified
by Rao (42)]. The pertinent data are:

1. Failure to form citrate from acetyl CoA, as outlined in
part 4 of the previous section.

2. Failure of either whole cells or cell-free extracts of A.
suboxydans to oxidize, succinate, fumarate, malate, or a-
ketoglutarate. Oxalacetate is oxidized to pyruvate and ace-
tate. The only other citric acid cycle member to be oxidized
at all is citrate, and this only slightly. The possible signifi-
cance of this oxidation is being examined further.

3. The quantitative data on C^^02 arising from cells
oxidizing glucose- or gluconate- 1-, 2-, 3,4-, 6-, or U-C^-^
indicate that the pentose cycle is the terminal oxidation
route in this organism. There is no indication that glycoly-
sis or the Krebs cycle are operative (23).

Glycolysis, A. suboxydans displays high aldolase and tri-
ose phosphate isomerase activity, and might be supposed
to carry out glycolysis in the usual manner. However, the
failure to produce significant amounts of acetate from
glucose (lactate and pyruvate are quantitatively converted
to acetate) has cast doubt upon the presence of a typical
Embden-Meyerhof dissimilation scheme. The presence of
individual enzymes characteristic of glycolysis does not of
itself assure glycolytic action, since every enzyme in the

THE ACETIC ACID BACTERIA 23

scheme except phosphofructokinase may also be used by
either the pentose cycle or the Entner-Doudoroff pathway.
The isotope data of Kitos et al. (23) argue strongly
against the use of glycolysis in whole A. suboxydans cells.
Likewise, when extracts are employed to oxidize glucose or
glycerol, the pentose cycle prevails (no acetate is formed
unless lactate or pyruvate is added). However, when the
cell-free extract is treated with Dowex-50 to remove Mg,
DPT, and pyridine nucleotides, it becomes possible to show
fructose diphosphate disappearance in a DPN-dependent
reaction, and the accumulation of 3-hour stable phosphate,
suggestive of 3-PGA formation. Both aldolase and triose
phosphate dehydrogenase are active under these conditions
(43). It thus appears that at least some of the reactions
of glycolysis can function in the traditional manner when
external conditions permit (or force?) their operation. The
importance of Mg in this test is clearly evident; in its pres-
ence the pentose cycle is favored to the virtual exclusion of
glycolysis.

THE BIOSYNTHESIS OF AMINO ACIDS

The foregoing sections on acetate non-utilization and lack
of a citric acid cycle logically raise the question of the ori-
gin of amino acids that normally arise from the cycle, espe-
cially glutamate and aspartate. These reactions are under
study, although at present the crucial experiments remain
to be carried out. Several bits of information are known
—for example, that this organism has great synthetic powers,
requiring [in contrast to the findings from an earlier study
(44)] only serine plus either glutamate, histidine, or pro-

24 METABOLIC PATHWAYS IN MICROORGANISMS

line for healthy growth, with glycerol as substrate. [When
glucose is the carbon source, serine may be omitted (45).]
The earlier report (44) listed valine as an essential growth
substance, but we have found it to be formed from acetolac-
tate, probably similarly to the formation of this amino acid
in yeast (46). When isotopic glucose is administered, the
label appears first (among amino acids) in aspartate and
glutamate; the same is true of administered C^^Oo. Also
(although this may not be related) despite the fact that
acetate is not oxidized by A. suhoxydans, CHgCi^OOH can
nonetheless give rise to labeled succinate.

At one point we had surmised that the isocitrase reaction
(Fig. 1.5) (47) might be operative in A. suboxydans, and
that it might aid in isocitrate formation by acting in reverse
to form isocitrate from glyoxylate -f succinate, or even to
form oxalsuccinate from glyoxylate -|- malate. However,
although one successful experiment (48) was performed that
tended to support this idea, subsequent efforts have failed.
What has been found instead is that glyoxylate has only

OH

H— c— cr^
c-cf

OH

H— C— C^
I ^OH
H

H

1

OH

1

1
C-

1

-c

II

4

1?

c-

1

-c-

-c-

II

-c

1

1
OH

OH

Fig. 1.5. The isocitrase reaction.

THE ACETIC ACID BACTERIA 25

a transient existence, due to the presence of a very active
glyoxylate reductase (glycolic dehydrogenase) in the organ-
ism (49). This enzyme, studied in detail in plants by Zelitch
(50), has an equilibrium constant of 1 X 10~^'^ in the
direction glycolate -^ glyoxylate; hence only traces of gly-
oxylate could be present at any moment.

PHOSPHORYLATION

The non-phosphorylative oxidations of this organism
presumably proceed without capture of the released energy,
at least as nucleotide polyphosphates. These oxidations
are not inhibited by dinitrophenol (51). In contrast, the
oxidations beyond the first step, i.e., those in the pentose
cycle, are phosphate dependent and, as mentioned earlier,
are dinitrophenol sensitive.

This observation prompted a more complete investigation
of oxidative phosphorylation in A. suhoxydans. Using
glucose, glycerol, and fructose as substrates and P^-04^^
as a tracer, it was found (52) that oxidative phosphorylation
in this organism differs in certain important respects from
the corresponding process in animal tissues. Thus,

1. Unlike animal systems, but like many other microbial
systems, the P/O ratios are low, averaging about 0.5.

2. The nucleotides that are active in phosphorylation
are present in low concentrations (about 1.5 /^moles ATP
per gram of respiring cells). It has not been possible to
increase these values through the addition of acceptors.

3. Inorganic pyrophosphate is present in relatively large
amounts (about one-eighth that of ATP) and turns over

26 METABOLIC PATHWAYS IN MICROORGANISMS

rapidly. However, its specific activity does not reach that
of orthophosphate, but levels off at about half that value.

4. After prolonged incubation, the specific activities of
the labile groups in ATP and ADP approach that of inor-
ganic p))?"ophosphate, not orthophosphate.

5. About five to ten times as much high polymer meta-
phosphate is found as is ATP. This material is slowly
labeled by added P^-04^~, although there is some evidence
that metaphosphate may be formed in resting cells oxidiz-
ing fructose and glycerol.

These data point strongly to inorganic pyrophosphate as
an intermediate in oxidative phosphorylation, in A. siib-
oxydans as well as in yeast. The origin of pyrophosphate
remains to be studied; it may arise through interaction of
polymetaphosphate with P^-0/~.

SUMMARY

The organism A. suhoxydans thus presents a variety of
behaviors that makes it seem curious indeed. Its special
behavior toward glucose has made it possible to recognize a
number of reactions, which, although present in other
organisms as well, are easier to view here because of the
absence of a functioning Krebs cycle. The Krebs cycle,
if present, might serve to remove many of the metabolites
that are now recognizable. The documentation of the
pentose cycle, both with respect to its extent and its abun-
dance, will be treated in the next chapter, where attention
is paid to the complex problem of evaluating pathway par-
ticipation in whole cells through the use of isotopes.

THE ACETIC ACID BACTERIA 27

REFERENCES

1. Kluyver, A. J., and F. J. de Leeuw, Tijdschr. Vergelijk. Geneesk.,
10, 170 (1924); Kluyver, A. J., The Chemical Activity of Microor-
ganisms, University of London Press, London, 1931.

2. Fulmer, E. I., and L. A. Underkofler, loiua State Coll. J. Sci., 21,
251 (1947).

3. Strong, F. M., Topics in Microbial Chemistry, E. R. Squibb Lec-
tures on Chemistry of Microbial Products, Wiley, New York, 1958.

4. King, T. E., and V. H. Cheldelin, Proc. Soc. Exptl. Biol. Med., 84,
591 (1953).

5. King, T. E., L. M. Locher, and V. H. Cheldelin, Arch. Biochem.,
17, 483 (1948); King, T. E., I. G. Eels, and V. H. Cheldelin, /. Am.
Chem. Soc, 71, 131 (1949).

6. Baddiley, J., and E. M. Thain, /. Chem. Soc, 1610 (1953).

7. Brown, G. M., and E. E. Snell, /. Bacteriol, 6Y, 465 (1954).

8. Green, D. E., Biol. Revs. Cambridge Phil. Soc, 29, 330 (1954).

9. Cheldelin, V. H., J. G. Hauge, and T. E. King, Proc. Soc Exptl.
Biol. Med., 82, 144 (1953).

10. Widmer, C, T. E. King, and V. H. Cheldelin, /. Bacteriol., 71, 737
(1956).

11. King, T. E., and V. H. Cheldelin, Biochem. J., 68, 31P (1958) .

12. de Ley, J., and A. M. Stouthamer, Biochim. et Biophys. Acta, 34,
171 (1959).

13. King, T. E., and V. H. Cheldelin, Federation Proc, 16, 204 (1957).

14. Glaser, L., and D. H. Brown, J. Biol. Chem., 216, 67 (1955).

15. Scott, D. B. M., and S. S. Cohen, Biochem. ]., 55, 23 (1953).

16. JuHan, G., R. G. Wolfe, and F. J. Reithel, /• Biol. Chem., 236, 754
(1961).

17. Horecker, B. L., and P. Z. Smyrniotis, /. Biol. Chem., 193, 371
(1951).

18. DeMoss, R., in Colowick, S. P., and N. O. Kaplan, Methods in
Enzymology, XoX. I, Academic Press, New York, 1955, p. 328.

19. Hauge, J. G., T. E. King, and V. H. Cheldelin, /. Biol. Chem., 214,
11 (1955).

20. Mejbaum, W., Z. physiol. Chem., 258, 117 (1939).

28 METABOLIC PATHWAYS IN MICROORGANISMS

21. Axeliod, B., R. S. Bandurski, C. M. Greiner, and R. Jong, /. Biol.
Chem., 202, 619 (1953).

22. Hauge, J. G., T. E. King, and V. H. Cheldelin, Nature, 174, 1104
(1954).

23. Kitos, P. A., C. H. Wang, B. A. Mohler, T. E. King, and V. H.
Cheldelin, /• Biol. Chem., 233, 1295 (1958).

24. Wood, W. A., Bacteriol. Revs., 19, 222 (1955).

25. Hauge, J. G., T. E. King, and V. H. Cheldelin, /. Biol. Chem., 214,
1 (1955).

26. Cummins, J. T., T. E. King, and \'. H. Cheldelin, /. Biol. Chem.,
224, 323 (1957).

27. Cummins, J. T., V. H. Cheldelin, and T. E. King, /. Biol. Chem.,
226, 301 (1957).

28. Heilbron, I., and H. M. Bunbury, Dictionary of Organic Com-
pounds, Oxford University Press, New York, 1953.

29. Hassid, W. Z., and R. M. McCready, Ind. Eng. Chem., Anal. Ed., 14,
683 (1942).

30. Blakley, R. L., Biochem. J., 49, 257 (1951).

31. Hollmann, S., and O. Touster, /. Biol. Chem., 225, 87 (1957).

32. Jolley, R. L., V. H. Cheldelin, and R. W. Newburgh, /. Biol. Chem.,
233, 1289 (1958).

33. Wenner, C. E., and S. Weinhouse, /. Biol. Chem., 219, 691 (1956).

34. King, T. E., and V. H. CheldeHn, /. Biol. Chem., 208, 821 (1954).

35. King, T. E., and V. H. Cheldelin, /. Biol. Chem., 220, 177 (1956).

36. Green, D. E., D. Herbert, and V. Subrahmanyan, /. Biol. Chem.,
138, 327 (1941).

37. Kobayasi, S., /. Biochem. (Japan), 33, 301 (1941).

38. King, T. E., M. K. Devlin, and V. H. Cheldelin, Federation Proc,
14, 236 (1955).

39. Kitos, P. A., T. E. King, and V. H. Cheldelin, /. Bacteriol., 74, 565
(1957).

40. Chou, T. C, and F. Lipmann, /. Biol. Chem., 214, 11 (1952).

41. King, T. E., E. H. Kawasaki, and V. H. Cheldelin, /. Bacteriol., 72,
418 (1956).

42. Rao, M. R. R., Doctoral Thesis, Univ. of Illinois, 1955; Ann. Rev.
Microbiol., 11, 317 (1957).

43. King, T. E., and V. H. Cheldelin, /. Biol. Chem., 198, 127 (1952).

44. Stokes, J. L., and A. Larsen, /. Bacteriol., 49, 495 (1945).

THE ACETIC ACID BACTERIA 29

45. Shamberger, R. L., Master's thesis, Oregon State College (1959);
Shamberger, R. L., and S. S. Kerwar, unpublished results.

46. Strassman, M., J. B. Shatton, and S. W'einhouse, /. Biol. Chem.,
235, 700 (1960).

47. Smith, R. A., and I. C. Gunsalus, /. Am. Chem. Soc, 76, 5002 (1954);
Wong, D. T. O., and S. J. Ajl, Nature, 176, 970 (1955).

48. Flikke, M., unpublished observations.

49. Baich, A. S., unpublished observations.

50. Zelitch, I., and S. Ochoa, /. Biol. Chem., 201, 707 (1953).

51. King, T. E., and V. H. CheldeHn, /. Biol. Chem., 198, 135 (1952).

52. Klungsoyr, L., T. E. King, and V. H. Cheldelin, /. Biol. Chem., 227,
135 (1957).

CHAPTER

EVALUATION OF METABOLIC
PATHWAYS

Our attention shifts in this chapter to a subject arising
out of the previous discussion, namely, an answer to the
question how to tell how much each given avenue of
metabolism is employed in the normal cell. In other words,
have traffic counting techniques been established, are they
reliable, and what do the counts show? The answers to
these questions are varied and almost too numerous; this
tends to reflect incomplete satisfaction with many of the
methods so far developed.

Metabolic inhibitors, or "traffic blocks," have been used
from time to time to study metabolic pathways. They can-
not be used satisfactorily to study the pentose cycle because
suitable specific inhibitors of this cycle have not been de-
veloped. Beyond this, it is questionable whether their use
would be warranted even if appropriate inhibitors were
developed, since it cannot be readily determined whether
the unphysiological conditions that would result from the
accumulation of metabolites might introduce spurious
readings of the "traffic count."

30

EVALUATION OF METABOLIC PATHWAYS 31

SPECIFIC ACTIVITY OF Ci^Q.

In principle, a distinction among the pathways should
be easy with isotopic tracers. As shown in Fig. 2.1, if a
molecule of glucose tagged in carbon 1 undergoes glycolysis,
the carbon chain is split in half; carbons 3 and 4 of glucose
become the carboxyl group of pyruvate, carbons 2 and 5
become the carbonyl group, and carbons 1 and 6 the methyl

®
CO2

c® 1 C® 1 c®

I i

C 2 C 2 C

I I I

C 3 ^ Pentose C 3 Glycolysis C

I cycle I

C 4 C 4 C

C 5 C 5 C

C 6 C 6 C

C CI c

— I I

C C 2 C

i ' I

C C 3 C

I Pentose | Glycolysis

9 cycle 9 ^

->■

C C 5 C

C® C® 6 C®

Fig. 2.1. Sketch of the routes of breakdown of glucose by glycolysis
and by phosphogluconate cleavage.

32

METABOLIC PATHWAYS IN MICROORGANISMS

1 1

O Q W fe

+

©

(M CO PQ ■>* lO CD

'n

\

\

Tt lO «£)
+

M CO CQ O Q W fe

CQ O Q H fL^

(M CO -* lO CO

pQ O P W P^

(M CO T}< lO CO

w ^

Q CO

r? CO

?3 II

c^] II

I 1

o o w fe

(M CO PQ ■<* lo to 23

CO

■>* lo CO

+

M CO PQ CJ Q W PiH

pq O Q H CiH

+

(M CO ■<** lO CO

pq O Q W Pm

(N CO -<r UO CD

EVALUATION OF METABOLIC PATHWAYS

33

•-• --H ;= <L) rt^

eg

CO '^

r lO

CO

3 a
">> o

"8

a;

O

o o c

; o o

3

•-^

<M CO T

r ic

CO

o

*

a

o o o o o 8

3
DQ (J Q W fc O

Q

h

fa

OJ

G

-^

M

O

•

•

Q

V3

rs

^

'f

U

2

"bf)

qj

g

.

<u

O

(U

<u

s

bx;

c

>

?5

c

CL

r--

5

^"^

co

^

C

O

c1

.2

O

• 2

fcC

OJ

far

-^

^

O

lU

f3
>

so

CO

r.

8
(J

JJ

7!

TJ

a;

,o

rt

u

^

3

"rt

OJ

i-(

£

qj

O

'~'

rt

• ^

dJ

S

o

03

-Td

c

"S

QJ

-—

;= a-

o - :^ .2

° CO

o^i:

C 3
^ O

^ ^ -rt

(L)

cu o

rC CO

H .Si

•- fa
^ fa

CO

s

2

o

c

o

n

S

CO

rt

OJ

<i)

1— H

■^

QJ OJ
OJ

£ o

■5 H

34 METABOLIC PATHWAYS IN MICROORGANISMS

C 1 COOH

I I

C 2 C^O Pyruvate

I I

C ^ 3 CH3

+ < +

C "^4 CHO

I I

C 5 CHOH Glyceraldehyde-3 -phosphate

C 6 CH2OPO3H2

2-Keto-3-deoxy-
6 - phosphogluconate

Fig. 2.4. The fate of the carbon skeleton of ghicose in the Entner-
Dotidoroff pathway.

group. The specific activity of the respiratory C^^Oo should
be identical whether glucose- l-C^^ or glucose-6-Ci^ is ad-
ministered to the test system.

If, on the other hand, phosphogluconate decarboxylation
prevails, carbon 1 of glucose will be quickly converted
to respiratory COo. If this carbon is labeled, the CO2 will
be substantially enriched with C^^, whereas if carbon 6
is labeled instead, the CO2 will have little or no activity
at first, and will become more active only as the pentose
cycle operates to relocate active carbon atoms into the oxi-
dizable position of the glucose molecule. Figure 2.2 indi-
cates the fate of individual carbon atoms as they traverse
the cycle; this scheme has been reported by Beevers (1)
for plant systems, and work in Dr. Wang's laboratory with
B. siibtilis oxidations indicates that the scheme outlined
may operate in this fashion in this organism as well.^

1 Beevers' scheme is sHghtly different from the one presented by Kitos
et al. (4); see Fig. 2.3. The latter, which showed one molecule of hexose

EVALUATION OF METABOLIC PATHWAYS 35

A third possibility involves the operation of the Entner-
Doudoroff pathway (2) (Fig. 2.4). Here the yield of COo
will be identical from carbons 1 and 4 of glucose, from
carbons 2 and 5, and from 3 and 6. Administration of glu-
cose labeled specifically in these positions should reveal
the presence of this route.

LABELING OF FERMENTATION PRODUCTS

Finally, use may be made of labeling of products isolated
from fermentation or other degradation of glucose. Thus,
Blumenthal, Lewis, and Weinhouse (3) have used the intra-
molecular distribution of isotope in acetate derived from
labeled substrate as the criterion for identifying the meta-
bolic routes that are being followed. Carbon 1 of glucose,
for example, would be converted to carbon 2 of acetate by
glycolysis, whereas it would not get into acetate at all via
the pentose cycle; and so on. A summary of locations
of specific glucose carbon atoms in various metabolic end
products by different pathways is given in Fig. 2.5.

Actually, the evaluation of experimental findings is not
quite so simple as might be expected from a cursory in-
spection of Fig. 2.1. Early workers, in comparing the
specific activities of Ci^02 arising from C-1 and C-6 of
glucose, often found the value to be several-fold higher
from C-1. The conclusion w^as reached that the major
portion of glucose metabolism was via phosphogluconate
decarboxylation. This, as is now recognized, was in error,

catalyzing the decomposition of all the others, seems attractive for this
reason; however, the data of Beevers (1) as well as those to be pre-
sented below indicate that his suggestion (Fig. 2.2) should probably
rate as the preferred mechanism.

36

METABOLIC PATHWAYS IN MICROORGANISMS

EMP

1 CHO

I

2 CHOH

I

3 CHOH

I

4 CHOH

I

5 CHOH

I

6 CH2OH

ED

EMP

1 CH3

2 CO

I

3 COOH

4 COOH

I

5 CO

I

6 CH3

1 COOH

i CO

3 CH3

4 COOH

5 CO

6 CHa

■CO.

+ [0]

■C05

•[O]

■C02

+[0]

-C02

+ [0]

C02

Extensive

pentose

cycle

CHO

nwc\u

CHOH
CHOH
CH2-O-P

Limited

pentose

cycle

22 C

_^33 C

32 C

44 C

55 C

66 C

2 C

C2-C3

cleavage

4 COOH

5 CHOH

6 CH3

EVALUATION OF METABOLIC PATHWAYS 37

1 CH3

TCA ^

1 CO2
2 CO2

2 COOH

cycle

3 CO2

4 CO2

5 COOH

TCA ^

5 CO2

6 CO2

6 CH3

cycle

1 C02

2 COOH TCA

■^ 2 CO2

3 CH3 cycle

4 CO2

5 COOH TCA ^ g

6 CH3 cycle

3 CO2

6 CO2

-^ 2 CO2

3 CO2

6 COoetc.

4 C 4 CO2

"5 C '^^f > 5 CO2

" 6 C ^'^'^ 6 CO2

22 CH3 _co 22 CH3 TCA ^ 22 CO2

33 CO 7;^ 33 COOH cycle 33 CO2

EMP _ ^^^TT +[^J

-^ 32 COOH

32 CO2

44 COOH ^^ 44 CO2
55 CO

+ [0]
66 CH3 55 COOH TCA ^ 55 CO2

66 CH3 cycle 66 CO2

Fig. 2.5. Relative rates of CO^, production from glucose via different
pathways; identity of glucose carbons in other intermediates.

EMP = Glycolysis (Embden-Meyerhof-Parnas)
ED = Entner-Doudoroff

HMP = Hexose Monophosphate Pathway (phosphogluconate cleavage)
(including pentose cycle)

38 METABOLIC PATHWAYS IN MICROORGANISMS

since the ratio of Ci^02 production, Ci/Cq, does not carry
any quantitative significance unless the actual amounts of
substrate glucose are known. Moreover, the evolution of
COo from carbons 3 and 4 (reflecting glycolysis) was not
taken into account.

Major objections to the use of specific activities of C^^Oo
to measure pathway participation are two: (a) whenever
more than one pathway is operative, as is usually true,
regeneration of hexoses by the pentose cycle results in
unpredictable dilution of the respiratory COo, from glucose
carbon atoms that undergo glycolysis or other breakdown;
(b) there is a dilution of all C^-^ atoms (5) by endogenous
metabolites, which are oxidized and enter the total CO2
pool.

The latter objection may also be raised against methods
that employ measurement of the specific activity of such
intermediates as pyruvate, lactate, or alanine. Weinhouse
and co-workers (3) have met this by the ingenious device
of using glucose-U-C^* in concurrent experiments as a ref-
erence standard, so that the dilution may be recognized
and compensated. However, this method also assumes that
there exists no drainage of pentose cycle intermediates for
synthetic purposes, and it assumes further that pentose P
does not re-form hexose P. The error caused by this last
assumption will be small if the contribution that is made
by the pentose cycle to total metabolism is small. Dawes
and Holms (6-8) have assumed, correctly, I believe, that
any regenerated hexose will be catabolized by botJi gly-
colysis and the pentose pathway; however, they have also
assumed no drainage of pentose cycle intermediates for syn-
thesis.

In a comprehensive, thoughtful analysis of this subject,
Katz and Wood (9) have recognized the dilution problem

EVALUATION OF METABOLIC PATHWAYS 39

caused by oxidation of regenerated endogenous metabolites,
and have attempted to account for the recycling of carbon
atoms that occurs in the pentose cycle; extensive calculations
are offered to show how this recycling would influence the
specific activities of respiratory CO2 from carbons 1, 2,
and 3 of glucose. However, although these workers assume
that glucose-6-P and fructose-6-P are in complete equilib-
rium, their treatment of data does not seem to allow ade-
quately for the possibility of diversion of hexose-P from
the pentose cycle into glycolysis. Their calculations appear,
in effect, to assume that carbon in the pentose cycle tends
to be recycled until it is oxidized to CO2. Moreover, no
drainage of the cycle intermediates is envisaged for other
cellular functions, and it is assumed that pentose cannot
be formed by reversal of the cycle, that is, by reversal of
the transaldolase-transketolase sequence. These points will
be dealt with in greater detail in the discussion on B. sub-
tilis oxidations, where the bulk of glucose catabolism is
routed via glycolysis, and also under the topic of the role
of the reductive pentose cycle in organisms.

RADIORESPIROMETRY

The third method has been called by us the radiorespiro-
metric method. You might expect that we would favor it,
since it was developed in our laboratories; but I believe
it to possess certain attractive features, and since we are
especially familiar with the method, I would like to explain
it to you.

As developed by Dr. Wang and his colleagues, the radio-
respirometric method does not stress specific activities; it
measures instead the yields of 0^*02 from various metabo-

40

METABOLIC PATHWAYS IN MICROORGANISMS

lites. The range of readily available compounds includes
glucose 1-, 2-, 3-, 3,4-, 6-, and U-C^^; similarly labeled glu-
conate; pyruvate 1-, 2-, and 3-Ci^; and acetate 1- or 2-Ci*.
These substrates represent all major sources of carbon in the
currently recognized metabolic pathways.

The apparatus used is represented in the sketch in Fig.
2.6. As will be seen, it consists of a modified Warburg

Fig. 2.6. The interval racliorespirometcr. (Courtesy of Krishell Lab-
oratories, Portland, Oregon.)

EVALUATION OF METABOLIC PATHWAYS 41

apparatus, in which the reaction flask and CO2 receiver are
constructed as shown in Fig. 2.7. It may be noted that the
manometer o£ the usual apparatus has been exchanged
for a system for trapping respiratory CO^; by proper manip-
ulation of the stopcock above the CO2 trap (G in Fig. 2.7),
shifts can be made to alternate receivers. (For operating in-
structions, see reference 11.) Shifts to new receivers permit
analysis of the KOH solution holding the recovered CO2,
and thus the CO2 obtained in a desired interval of time may
be recorded; the apparatus becomes an interval radio-
respirometer. Plots of the interval yields of Ci^02 from the
labeled substrate obtained automatically display the kinet-
ics of the process. It will also be noted that several simul-
taneous experiments may be carried out; by using differ-
ently labeled substrates in each flask, a battery of experi-
ments may be conducted that will give a simultaneous pic-
ture of the kinetics of oxidation of each of several carbon
atoms of a given substrate. The uncontrollable complex-
ities of metabolism that cause dilution of individual carbon
atoms and spuriously influence specific activities may be ig-
nored since the instrument permits a measure of the total
yield of CO2 from a given labeled position of a substrate
molecule.

It is reasoned that oxidation of glucose exclusively at
carbon 1 is indicative of phosphogluconate cleavage. Any
entrance of carbon 1 into CO2 via the Embden-Meyerhof
route followed by the TCA cycle should also be reflected in
equal C^ oxidation; so the difference between the Ci and
C(> yield in respiratory CO2 should give a measure of the
extent of phosphogluconate cleavage. The per cent of the
total metabolism of glucose that this process supports, on
the other hand, will be chiefly influenced by the appearance
of carbons 3 and 4 in CO2 (indicative of glycolysis). The

AIR PUMP OR
AIR SUPPLY MANIFOLD

TRAP

FLOWMETER

CONNECTING ARM

TRAP SOLUTION
RECEIVER

P"

HOLDER

Fig. 2.7. Details of reaction flask and CO^ traps in interval radio-
respirometer. a, incubation flask and side arm; b, head for connection

42

EVALUATION OF METABOLIC PATHWAYS 43

calculation of the fraction of the total metabolism that
travels by the pentose pathway (with and without recycling)
is measured as

Gp —

Gi — Gq

Gt

where Gp = fraction of glucose participating in phospho-
gluconate decarboxylation
Gi\ Gq^ = total activity in respiratory CO2 from cells uti-
lizing equal amounts of C-1 or C-6 labeled
glucose
Gt = total activity of the administered substrate (al-
ways taken as 100%, or unity)

The fraction of glucose catabolized by glycolysis, G^ (Emb-
den-Meyerhof-Parnas pathway) is

Ge = 1 — Gp

This assumes that the Entner-Doudoroff pathway does not
simultaneously operate. In our experience W'ith various
organisms, this assumption seems generally safe. When the
Entner-Doudoroff pathway is present, glycolysis is seen gen-
erally to have subsided, and the recognition of Entner-Dou-
doroff activity (separately from the pentose cycle) is made
by noting the difference between the yield of CO2 from car-
bons 1 and 4 of glucose, since exclusive Entner-Doudoroff

of incubation flask; c and p, aeration tubing (glass) with lower end
protruding into incubation flask; d, rubber tubing; e, air inlet; /, flow-
meter; g, three-way stopcock; h, CO2 trap; i, needle valve, to regulate
gas flow; /, tension clamp for release of trap solution; k, clamp for
holding trap solution receiver; n, trap solution receiver. (Courtesy of
Krishell Laboratories, Portland, Oregon.)

44 METABOLIC PATHWAYS IN MICROORGANISMS

activity would show equal yields of CO2 from carbons 1 and
4 of glucose.

Gp —

Gi — G4

Gt

Ged (Entner-Doudoroff) is taken as the remainder

The types of CO2 production curves obtained vary from
organism to organism; four of these are reproduced here
to show the characteristic differences that exist.

1. In bakers' yeast, Fig. 2.SA shows that the greatest evo-
lution of CO2 is from carbons 3 and 4, characteristic of
glycolysis. (The general availability of this species of glu-
cose dictates its routine use over glucose-S-C^*; it is as-
sumed here that the yield in CO2 is equal from both car-
bon atoms.) C^^02 production from C-1, although higher
at first than from C-6, is much lower than from C-3,4- The
curve for interval recovery of CO2 from C-3,4 reaches a
maximum at about 2 hours, then drops to a minimum at
4 hours; the latter time coincides with complete removal
of the administered glucose from the medium. This time
is defined as I relative time unit (1 RTU) and is used to
make comparisons more relative among different organisms
(Streptomyces griseus, for example, exhibited a much slower
metabolism under the experimental conditions used than
did bakers' yeast; 1 RTU was about 10 hours). After 1
RTU the accumulated radioactivity in the cells and medium
undergoes depletion; yields of Ci^02 from carbons 2 and 6
increase in that order, as would be expected from oxidation
of acetate via the TCA cycle.

2. The second type is exemplified by Zymomonas mo-
bilis; the C^^02 yields agree with the contention that the

EVALUATION OF METABOLIC PATHWAYS 45

Entner-Doudoroff pathway predominates, since the amounts
are essentially equal from C-1 and C-4 of glucose. This
is revealed in Fig. 2.8^^ since there is reported to be no
Krebs TCA cycle in this organism. There is therefore no
second phase after 1 RTU in which C14O2 is formed from
carbons 2 or 6, as in yeast.

.8. Phosphogluconate cleavage, followed by the complete
pentose cycle, is seen in A. suhoxydans fFig. 2.8C). In this
curve, both interval and total yields are greatest from glu-
cose C-1, >C-2, >C-3,4, >C-6. This is precisely in accord-
ance with expectations. The greater yield of C-1 over
C-2 in the respiratory CO2 suggests that pentose cycle ac-
tivity is not as great as phosphogluconate cleavage; some
C-2 is evidently utilized in pentose that undergoes assimila-
tion. The Entner-Doudoroff route is ruled out because
of the high yield of CO2 from glucose C-2 (higher than
from C-4). If A. suhoxydans utilized the Entner-Doudoroff
pathway, and then diverted the triose formed (from carbons
4, 5, and 6) into hexose phosphate via aldolase condensation
and degraded the latter via the pentose cycle, the C^^Oo
would be fairly rich from carbon 6. C-2, however, would
be retained as acetate, since pyruvate is converted nearly
quantitatively to acetate in this organism (12).

With these alternate pathways eliminated, the contribu-
tion of the pentose cycle to total glucose dissimilation in A.
suhoxydans can be estimated quantitatively with the use of
the cumulative recovery data of Table 2.1. Thus, in an oxy-
gen atmosphere, at 1 RTU, conversion of glucose to pentose
phosphate = C-1 recovery = 72%.- Pentose phosphate un-
dergoing cycling = C-2 recovery = 63% of glucose adminis-
tered (hence, (63 X 100)/72 = 88% of pentose formed).

2 2-Ketogluconate accounted for all (28%) of the glucose that was
not converted to pentose in this experiment.

46

METABOLIC PATHWAYS IN MICROORGANISMS

CO

^Oj^tO |0 AJ8A008J IBAJSJU!

EVALUATION OF METABOLIC PATHWAYS 47

Fig. 2.8. Time-course plots of radiochemical recoveries in CO2 from
metabolizing specifically labeled glucose.

(A) Bakers' yeast: 1 RTU = 4 hours.

(B) Zymomonas mohilis (formerly called Pseudomonas lindneri):
1 RTU = 4 hours.

(C) A. suboxydans: 1 RTU = 4 hours.

(D) B. subtilis: 1 RTU = 4.5 hours.

In all systems, the following legends are employed: .,

glucose-l-Ci4; -, glucose-2-Ci4; , glucose-3,4-Ci4;

, glucose-4-Ci4; and — • — , glucose-6-Ci4. Inserts: cumu-
lative recoveries of Ci40o.

Since oxidation of C-6, -5, or -4 required recycling, we can
assume that the maximal yield of C-4 in COo will be ap-
proximately 88% of C-6, = (33/100) X 0.88 = 29%. Also,
since C-3, 4 "average" recovery = 46%, recovery of C-3 -f
C-4 = 92%. Therefore, minimal recovery of C-3 = (92 —
29)/100 = 63%.

This calculated value for recovery of C-3 in the respira-
tory COo agrees well with the observed value for C-2 re-
covery from glucose (63%; Table 2.1) and thus confirms
the pentose cycle as the only significant pathway for com-
plete oxidation of glucose in the organism concerned under
the conditions which prevailed.

This calculation, although different from that of Katz
and Wood (9) nevertheless takes into account much of the
complexity of recycling of F-6-P. In an organism that relies
on the pentose cycle as extensively for oxidation of glucose,
one would expect that in resting cells the combustion of
glucose carbon atoms 2 through 6 should approach equal
magnitude with carbon 1, although they may do so more
slowly. The fact that they do not, even at the end of the

48

METABOLIC PATHWAYS IN MICROORGANISMS

^OfciO P AJ8A039J IBAJ81U! %

EVALUATION OF METABOLIC PATHWAYS

49

00

(2

^O^lD P AJ9A009J IBAJa^U! %

50

METABOLIC PATHWAYS IN MICROORGANISMS

"* E

) o o c

AJ3A0D8J 'inLunD %

\\\

20„3 p AjsAOoaj leAjgjui

EVALUATION OF METABOLIC PATHWAYS

51

g
be

^'o

CN T— 1 O '— '

g CO

>i

Cs O O O

^ «

13

Si

T-l T-H ,— 1

(N CD ^r^ C\

00

^1 .

ago
gt^ ^

2

<

O O O CN

Cn

Mi

1

i?

a c s

s

«5

u

S

c
a;
bjo

r- \0 r-H LO

(N CO in NO

mediu
[ask cc
the te

Ik.

^

"5

^^^

Q)

o

<u

O J3 C

!-i

O Tj- Tj- ^

\o

milliliter
ours. Eac
is 10 ml. a

O

<

<N en rt o

CO

i-H

en

G

■M

G

O CM CO CO

eight, per
e was 5 hi
olume wa

(N

,^

^ ^ V

s

<

(N rl- sO LO

r~-

CQ

•^

«rH .

*

G

CN CO O CO

mg. of cell
;ntrations.
ffer, pH 6.0

4^

8

*

r^ NO "^i- CO

§

u

O IT) T— CO

IT)

C ° V

<

OO O U^ cO

LO

o

«4H

til

S3

>>!

5l

§§88

o
o

o^ o_ o o
r^ t-~ r- o

o

B V o

C8

a -^ ^

^M

'"^ -^ V

.§

!s|

OD

■2^ h =3.

^

'2

G

"2

c ^ o

?^18

3 ^ OS
OU4U

1

repre
atmos
and 5

■r^ CM CO^nO

o

>s

CO

y ii

u o o o

o

■^u

3 3 D G

* >. y

X -i

ooooo

bD

52 METABOLIC PATHWAYS IN MICROORGANISMS

time-course experiment, strongly suggests drainage of carbon
skeletons from the cycle.

4. The fourth example is given by B. subtilis, which ap-
pears to utilize a combination of pathways, much as yeast
does. It is reproduced here because (a) the glycolysis pat-
tern is less pronounced than in yeast, and {b) the data
permit a testing of one of the main implications of the paper
of Katz and Wood, namely that carbon atoms traversing the
pentose cycle seemingly do so until oxidized, without escape
of triose units.

One can readily recognize two distinct phases of glucose
oxidation in B. suhtilis (Fig. 2.8/)), as in yeast, with the
division occurring approximately at 1 RTU. At first,
glycolysis is clearly indicated by the high recovery of glu-
cose-C-3 and -4 in CO2. Meanwhile, the presence of an
alternate pathway is recognized, since the yields of Ci^02
from these carbon atoms are not equal, nor are they equal
between C-1 and C-6. The excess yield of C-1 over C-6
points to phosphogluconate cleavage as the alternate route
involved. The fact that C-2 oxidation was less than C-1
again suggests that some pentose phosphate was assimilated
—a conclusion to be expected since some cell proliferation
took place in this experiment.

After exhaustion of the administered glucose in Fig. 2.8£)
there is a resumption of oxidation of carbons 2 and 6, and
to some extent carbon 1. The order of release of these car-
bons into COo (C-2 ^ C-6 > C-1 > C-3,4) reflects the opera-
tion of the TCA cycle.

The usefulness of the radiorespirometric experiments is
revealed from the experiments in Fig, 2.9 on the utilization
of gluconate by B. suhtilis. Gluconate presumably cannot
be converted directly back to glucose; its oxidation would
seem to be obligatory by the pentose cycle, at least as far

EVALUATION OF METABOLIC PATHWAYS

53

AJ8A009J •|nujno %

.2 U
be oo

n ^

O ^

U

c3 ..

-^ I

-r. U

o ^

in "

o o

u

i2 ^
^ 'be

'H,

be

^Of^rD 10 AJ8A039J |BAJ8}U! ^

WD -2

54

METABOLIC PATHWAYS IN MICROORGANISMS

as the formation of triose phosphate or fructose-6-phosphate.
Now the disposition of these species is crucial: will the F-6-P
re-enter the pentose cycle as seems obligatory from the paper
by Katz and Wood, or will the triose be subjected in large
part to degradation via the Krebs cycle, as indicated for
this organism in the foregoing paragraphs?

The radiorespirometric patterns in Fig. 2.9 and Table
2.2 provide the following information: (a) gluconate is
readily broken down, presumably after initial phosphoryla-
tion; (b) the route employed is not the Entner-Doudoroff
pathway, either alone or in combination with the pentose
cycle. In the Entner-Doudoroff pathway, there should be
metabolic equivalence between C-1 and C-4, C-2 and C-5,
or C-3 and C-6. This was not found. Instead, gluconate
appears to be metabolized via a sequence (Fig. 2.10):

Glucose,

G\ycoly

Gluconate

6-P

Mg

F-6>P

i/f/ Glycolysis

^ ^ Triose P

i=F-l,6-P

Glycolysis

TCA Cycle
CO2, etc.

Fig. 2.10. Simultaneous operation of glycol)sis and the pentose cycle,
and their effect on metabolism of administered glucose and gluconate.
Both circular loops represent pentose cycle operation: the outer loop
for the degradation of gluconate and a part of glucose; the inner, for
the portion of both glucose and gluconate metabolism that escapes
glycolysis after the first formation of triose phosphate, i.e., the portion
that recycles.

EVALUATION OF METABOLIC PATHWAYS 55

pentose cycle

Gluconate-6-P > fructose-6-P

major

7^ > F-1, 6-diP -^ glycolysis + TCA cycle

Fructose-6-P \

\^ minor

> G-6-P -^ pentose cycle

This conclusion is dra\vn from the consideration that the
fructose-6-P derived from the labeled gluconate should
bear the following labeling patterns (1) (see also Fig. 2.2):

1 C 1 CO2

+
2C 2C 2222

3 3 3 3
3C-^6 3C ->2233+C44
6 4444C55

4G 4C 5555 C66

6 6 6 6

5C 5C

6 C 6 C 4 F-6-P + 2 triose-P

Further catabolism of F-6-P may be predicted as follows.
Let us consider the fate of gluconate-C-2:

In Table 2.2, data are provided on the total recoveries
of C^^Oo from individual carbon atoms of glucose and
gluconate. For gluconate, the maximum fraction decar-
boxylated is 88%. This represents the maximum amount
of pentose phosphate that could have traversed the cycle,
and the probable actual amount that did so.

According to the scheme just presented, the re-formed
fructose-P (from gluconate) will have the original C-2 from
gluconate in carbons A and C (1 and 3). Sixty-seven per
cent of this will be in carbon A, 33% in carbon C.

56

METABOLIC PATHWAYS IN MICROORGANISMS

TABLE 2.2

Yields of Individual Atoms from Glucose and Gluconate:
B. subtilis Oxidations

(Synthetic Medium)

C^^02 Recovery

Substrate

CO2

Cells

Medium

Total

Gluconate-1

88%

1%

11%

100%

Gluconate-2

54%

22%

18%

94%

Gluconate-3

54%

31%

17%

102%

Gluconate-3,4

63%

19%

15%

97%

Gluconate-4

72%

7%

13%

92% (calc.)

Gluconate-6

44%

36%

18%

98%

Glucose-1

60%

22%

13%

95%

Glucose-2

60%

25%

16%

101%

Glucose-3

67%

19%

16%

102%

Glucose-3,4

77%

15%

12%

104%

Glucose-4

87%

11%

8%

106% (calc.)

Glucose-6

41%

23%

41%

105%

If we now assume that re-formed fructose behaves like
administered glucose, Table 2.3 tells us that carbons A and
C of glucose will be recovered in COo to the extent of 60
and 67%, respectively. Multiplying

88% X 67% X 60% = 35%
and multiplying
88% X 33% X 67% = 20%

55%

calculated recovery of C^'^O^
from C-2 of gluconate

This agrees well with the observed recovery of 54% from
C-2 of gluconate (Table 2.3).

Formation of C^^Oo from

EVALUATION OF METABOLIC PATHWAYS 57

O ^

•o

>>-Cr

s

%6

o t! M

u^Oir!Ot-'+OvOOC>

f0(MP0(>Jt-'O — rr3rr^, Q

|5

y rt •

dodddcdddd

^

^

P^fe

s

.2

rt 2

cS

^ ?n

<M 3

Ot-Or-r-t-t^-H^O

A

11

8^^

d d d d d d d d d d

|5

d o

^ .2

H ^

^ . -^ g 5 ^ « -S .S .S .S u .S .S 6 .S .S
I tLi^O.c dddd-ndd-^dd

(I!)
VI
>>.

£■ e rt V

3 rc•^" OCOOOCOOOOOCOOOOOOOO

■^ -"oOoJ O0O0XO0OCXMC3C0OO0

o 3 U d d d d d d d d d d

g.2

j2 rs

y 6

t. o iy —

aj y S _:

o o o o o o

(LI o o u u u u

C p 3 3 3 3 3

►^ o o GO 55

o-g o c

« 3 C .S -S

o 9, %%ou

i-pi 1-H (C tn >< ii

W ►Jh tc en y j_|

UUUOUO <<<c:i
< CQ U Q W ta

58 METABOLIC PATHWAYS IN MICROORGANISMS

other positions in gluconate also agrees fairly well with
predictions. Nearly identical results were obtained in a
repeat experiment, not shown here, performed six weeks
later.

It would appear from the above paragraph that fructose
exists in a general pool in this organism, and that it is
treated the same whether it is administered as substrate
or re-formed from the pentose cycle; it is apparently un-
necessary to make so extensive allowance for recycling as
Katz and Wood have done. Instead of a perfect equilibrium
existing among fructose- 1, 6-diP, glycolysis, and the pentose
cycle, there appears instead to be a steady state that, in
B. suhtilis at least, is shifted toward glycolysis and the TCA
cycle. These data support the assumption made by Dawes
and Holms (13) in Sarcina lutea, that re-formed fructose will
be degraded both by glycolysis and the pentose cycle. It
would appear likely that this condition may obtain in
animal systems as well; work is being planned in our
laboratories to test this assumption further.

QUANTITATIVE EXPRESSION OF PATHWAY
PARTICIPATION

The reader may now inquire whether so simple a cal-
culation as we have previously employed (11), can really
represent the complex events that take place in organisms,

Gp = -^ ^ (footnote 3) (1)

in spite of the seeming reinforcement given by the B. suh-
tilis experiment quoted. The fact is that the B. suhtilis
3 In the original report (11), G, and Gg were written as G^^' and G^'.

EVALUATION OF METABOLIC PATHWAYS 59

experiment merely equates fructose regenerated in the pen-
tose cycle with the glucose administered as substrate. The
equation used previously is indeed too simple; data obtained
through its use are somewhat in error, as follows:

When an organism degrades glucose by both the gly-
colysis-Krebs cycle and the pentose cycle routes, some triose
phosphate will be formed by either pathway. Glucose
carbons 1 and 6 will appear in triose in equal amounts by
glycolysis, but the triose from the pentose cycle will con-
tain only carbons 4, 5, and 6 (carbon 1 will have been con-
verted to COo). The total triose pool will therefore be
enriched with carbon 6, and carbon 6 yields in CO2 will
be too high.

This means that corrections should be provided which
enlarge the differences between Gi and G^, specifically by
reducing Gq-, in other words, the reported contribution of
phosphogluconate cleavage, as calculated by equation 1,
is spuriously low. Such refinements have been made (12, 14),
and it is believed that the newer equations will give a more
nearly correct evaluation of simultaneously participating
pathways. Briefly, where glycolysis and phosphogluconate
cleavage occur together, Wang (12, 14) has developed the
following expression:

G, = ^- - ^^- - ^-^-^ (2)

Gt — Grj"

where G^ = total radioactivity of each administered substrate
(always taken as 100%, or unity)
Gt' = fraction of the labeled substrate engaged in ana-
bolism
Gi, Ge = per cent total yields of C^^02 from systems me-
tabolizing equal amounts of glucose labeled in
these positions

60 METABOLIC PATHWAYS IN MICROORGANISMS

Ag = per cent total C^^02 yields from systems metabo-
lizing equal amounts of gluconate labeled in these
positions

Gp = fraction of the administered glucose catabolized
via phosphogluconate cleavage

In equation 2 a correction factor has been introduced
into the term Gq (equation 1). The coiTection arises from
the understanding that the term Gq in reality represents
a total conversion of C-6 of glucose to CO2 via both glycoly-
sis and the pentose cycle pathways, whereas the derivation
of equation 1 calls for the consideration of CO2 produc-
tion from C-6 of glucose via the glycolysis exclusively. The
COo production from C-6 of glucose via the pentose cycle
reactions can be represented as Aq{Gp) since G^ represents
the fraction of glucose routed into the pentose cycle and Aq
represents the CO2 yield from C-6 of glucose by way of
phosphogluconate. Consequently, the term (Gq-AqGp) rep-
resents the net production of CO2 from C-6 of glucose via
the glycolysis-Krebs cycle pathway which is theoretically
identical to the production of CO2 from C-1 of glucose via
the same metabolic route.

When equation 2 is used in place of equation 1 to cal-
culate the pathway participation in B. suhtilis, the previous
value of 26% for phosphogluconate decarboxylation rises to
41%. Similar refinements (including a determination of
gluconate conversion to COo) are in progress with several
other organisms. Table 2.4 lists several organisms in which
the extent of participating pathways is reasonably well
known, although several may be revised when adequate
values for gluconate oxidation are available.

Knowledge of catabolism of glucose via catabolic path-
ways in animals is in a less satisfactory state, although in

EVALUATION OF METABOLIC PATHWAYS

61

TABLE 2.4

Estimation of Catabolic Pathways of Glucose

Pathway Estimation

Refer-

Microorganism

Glycolysij

5 HMP

ED

Method

ence

Saccharomyces

88

12

Radiorespirometry

11

cerevisiae

83-100=

*= 0-17 =

¥

Specific activity of
intermediates

3

Candida utilis

66-81

19-34

Specific activity of
intermediates

3

Streptomyces griseus

97

3

Radiorespirometry

11

Penicillium chrysogenum

77

23

Radiorespirometry

11

Penicillium digitatum

77-83

17-23

Radiorespirometry

11

Fusarium lini

83

17

Specific activity of

16

Escherichia coli

72

28

Radiorespirometry

11

Sarcina lutea

70

30

Radiochemical
yield of inter-
mediates

6

Bacillus subtilis

74

26

Radiorespirometry ;
equation 1

59

41

Radiorespirometry ;
equation 2 f

Pseudomonas reptilivora

28

72

Radiorespirometry

15

Pseudomonas aeruginosa

29

71

Radiorespirometry

15

Acetobacter suboxydans

100

Radiorespirometry

4

Zymomonas mobilis

100

Radiorespirometry

15

Pseudomonas

100

Radiorespirometry

15

saccharophila

* Experimental conditions vari

ied from

growing to depleted resting

cells.

t Equation 2 is used

I with the

assumpt

ion that Gr' = 0.

general it may be stated that it appears that glucose is
catabolized mainly by way of glycolysis and the Krebs cycle;
phosphogluconate cleavage may participate to a minor,
variable extent (17). Qualitative experiments have been
reported by Bloom (18) on a variety of rat tissues; by others
on rats (19-23), blood cells (24), insects (25), cow's mammary

62 METABOLIC PATHWAYS IN MICROORGANISMS

gland (26), retina (27), and carp (28). Interesting side re-
actions have been uncovered; the occurrence of the pentose
cycle as well as the extent of recycling of pentose phosphate
in cows were reported by Wood (29). By studying the iso-
topic distribution patterns in amino acids, glycerol, and
glucose isolated from milk derived from cows fed with spe-
cifically labeled substrate, it was possible to detect trans-
aldolase exchange, and to enable the author to examine the
possible effect of side reactions upon the estimate of glucose
metabolic pathways.

Other reports in the literature permit speculation on the
probable role of the pentose cycle in over-all metabolism.
These speculations will be reserved for the next lecture,
where a more extensive discussion of this cycle will be made.

REFERENCES

1. Beevers, H., Plant Physiol., 31, 334 (1956).

2. Entner, N., and M. Doudoroff, /. Biol. Che??!., 196, 853 (1952).

3. Blumenthal, H. J., K. F. Lewis, and S. Weinhouse, /. Am. Chem.
Soc, 76, 6043 (1954).

4. Kitos, P. A., C. H. Wang, B. A. Mohler, T. E. King, and V. H.
Cheldelin, /. Biol. Chem., 233, 1295 (1958).

5. Wood, H. G., Physiol. Rev., 35, 84 (1955).

6. Dawes, E. A., and W. H. Holms, Biochim. et Biophys. Acta, 29, 82
(1958).

7. Dawes, E. A., and W. H. Holms, Biochim. et Biophys. Acta, 34, 551
(1959).

8. Dawes, E. A., and W. H. Holms, personal communication.

9. Katz, J., and H. G. Wood, /. Biol. Chem., 235, 2165 (1960).

10. Wang, C. H., and R. D. Barbour, unpublished.

11. Wang, C. H., I. Stern, C. M. Gilmour, S. Klungsoyr, D. J. Reed,
J. }. Bialy, B. E. Christensen, and V. H. Cheldelin, /. Bacteriol., 76,
207 (1958).

EVALUATION OF METABOLIC PATHWAYS 63

12. Cheldelin, V. H., C. H. Wang, and T. E. King, in Florkin, M., and
H. S. Mason, Comparative Biochemistry, Vol. Ill, in press.

13. Dawes, E. A., and W. H. Holms, Biochim. et Biophys. Acta, 34, 551
(1959).

14. Wang, C. H., personal communication.

15. Stern, I., C. H. Wang, and C. M. Gilmoiu', /. Bacterial., 79, 601
(1960).

16. Heath, E. C, DeL. Nasser, and H. Koffler, Arch. Biochem. Biophys.,
64, 80 (1956).

17. Dickens, F., Proc. jrd Intern. Congr. Biochem., p. 171 (1955).

18. Bloom, B., Proc. Soc. Exptl. Biol. Med., 88, 317 (1955).

19. Katz, J., S. Abraham, and R. Hill, and I. L. Chaikoff, /. Biol. Chem.,
214, 853 (1955).

20. Bloom, B., and D. Stetten, Jr., /. Biol. Chem., 212, 555 (1955).

21. Bloom, B., F. Eisenberg, Jr., and D. Stetten, Jr., /. Biol. Chem., 222,
301 (1956).

22. Ashmore, J., G. F. Cahill, Jr., and A. B. Hastings, Recent Progr. in
Hormone Research, 16, 547 (1960) .

23. Abraham, S., P. F. Hirsch, and I. L. Chaikoff, /. Biol. Chem., 211,
31 (1954).

24. Dickens, F., Ann. N. Y. Acad. Sci., 75, 71 (1958).

25. Silva, G. M., W. P. Doyle, and C. H. Wang, Nature, 182, 102 (1958).

26. Wood, H. G., R. Gillespie, R. G. Hansen, W. A. Wood, and H. J.
Hardenbrook, Biochem. J., 73, 694 (1959).

27. Kerly, M., and M. A. Rahman, Biochem. J., 74, 16P (1960).

28. Brown, W. D., /. Cell. Comp. Physiol., 55, 81 (1960).

29. \Vood, H. G., Second Intern. Conf. on Peaceful Uses of Atomic En-
ergy, A/Conf. 15/P/852 (1958).

CHAPTER

CARBOHYDRATE
METABOLIC PATHWAYS:
GENERAL CONSIDERATIONS

In the two previous chapters we have discussed the
pecuHar metabolic events that occur in Acetobacter suh-
oxydans, and how some of the major carbohydrate degrada-
tion routes are measured in respiring organisms. In this
chapter I should like to strike a somewhat more general
note, to discuss some of the processes w^ith which most
students of carbohydrate metabolism are already familiar,
but in the hope that our fresh discussion may divert a few
rays of light in a new manner upon the complexities of
total metabolism in a whole tissue or organism.

Let us begin this discussion by again referring to the
Krebs tricarboxylic acid cycle. This mechanism deserves
special recognition because its discovery, by Krebs and
Johnson almost a quarter century ago (1), ranks as one
of the major landmarks in all biochemistry. Its primary
importance to students of carbohydrate metabolism can best

64

CARBOHYDRATE METABOLIC PATHWAYS 65

be emphasized by the simple diagram of the Krebs cycle
shown in Fig. 3.1. This sequence shows how compound
A (pyruvate) can be converted to B (acetyl CoA) which, in
turn, combines with a molecule of C (oxalacetate) to pro-
duce D (citric acid) and continue stepwise to the re-formation
of compound C.

Viewed in this light the Krebs cycle is revealed for what
it really is— a catalytic cycle. As has been pointed out
previously (2), the removal of a few molecules of any of
the intermediates pictured here for synthesis stops the
cycle, unless there is some provision for the independent
formation of di- or tricarboxylic acids. Thus we find in
most organisms a functioning malic enzyme, or oxalacetic
decarboxylase, or malate synthetase system; if one or more
of these is operating there is an additional source of Krebs
cycle intermediates, so that removal of one or more of the
intermediates for synthesis of amino acids, for example,
may be possible. Without such independent mechanisms,
the Krebs cycle as wt usually picture it could not operate,
and energy metabolism therefrom would soon also grind to
a halt.

Another reason for making initial mention of the citric
acid cycle is that it helps to emphasize the importance
that this mechanism has held, not only as a pathway of
metabolism, but also perhaps even more importantly as a

H

>■

B

+ V -

— ^

D

^ E

_^

F

_^

G

cj

\

^

K

^—

- J

^

—

I

^

—

H

Fig. 3.1 The Krebs tricarboxylic acid cycle— a schematic diagram.

66

METABOLIC PATHWAYS IN MICROORGANISMS

conceptual contrivance that has made possible many ex-
tensions of knowledge about metabolism in general. Be-
cause the individual steps in the pathway have been verified,
the viewpoint is sometimes diminished that the cycle is
not only a proven piece of metabolic machinery, but also
may be a powerful psychological tool in helping to chart
new ground.

THE MITOCHONDRION AND THE CITRIC ACID CYCLE

The ubiquity of the Krebs cycle was given deepened ap-
preciation during the early 1940's, with the discovery by

Mitochondria

Nucleolus

Microsomes

Cell
membrane

Fig. 3.2. Diagram of a generalized cell.

r

r

t

Fig. 3.3. Mitochondria and microsomes in the liver. Ultrathin sec-
tion of mitochondria from a rat liver cell. Mitochondria showing
internal cristae. In the surrounding cytoplasm canaliculi of the endo-
plasmic reticulum with dense granular material in the outer surface
are seen. Magnification = 25,500x. [After Palade (5).]

67

68

METABOLIC PATHWAYS IN MICROORGANISMS

Claude (3, 4) and others, of the close association between
the group of enzymes necessary for the process, and their
presence in mitochondria.

Consider for a moment the inclusions, or subcellular
particles, that may be observed in a typical cell. A glance
at the diagram of a generalized cell in Fig. 3.2 will reveal
the nucleus, the large moon-shaped object in the upper
center; the mitochondria, or "large granules," as these are
called, which are smaller, yet easily visible under an ordi-
nary high-power microscope; and the microsomes, or endo-
plasmic reticula, which resemble small strings of spaghetti,
and which can be viewed in some detail in an electron micro-
graph. Other inclusions may be seen at times, such as food
particles; but the foregoing are the principal ones. Pal-
ade's work, one of the plates of which is reproduced in
Fig. 3.3 (5), has contributed much to our knowledge in
this area.

Examples of the association between Krebs cycle enzymes
and mitochondrial activity were to be found easily. With

TABLE 3.1

Size, Sedimentation Time, Gross Composition, and
Succinoxidase in Liver Cellular Fractions

Fraction (per cent)

Diameter,

Sedimentation

of Total

Time,

Accel-

Succin-

Fraction

M

minutes

eration

N

RNA

oxidase

Nuclei

50-100

10

600^

15

10

10

Mitochondria

1-3

20

24,000^

30

25

90

Microsomes

0.06-0.15

120

41,000^

15

>35

Soluble

—

—

—

40

25

After Lehninger (6).

CARBOHYDRATE METABOLIC PATHWAYS

69

advances in refrigerated centrifugation, it was discovered
that the cell fractions discussed could be separated readily:
the nuclei and cell debris at centrifugal speeds of lOOOg
or less; mitochondria at 5000-20,000g; and microsomes at
25,000-50,000g-. When this was done, Lehninger found
(Table 3.1) (6) that whereas the nuclei, microsomes, and

TABLE 3.2
Complete Oxidation of Members of the Citric Acid Cycle

Oxygen Consumption

Amount

Theory for Com-

Added,

Observed,

plete Oxidation,

Substrate

//moles

yuatoms

/xatoms

a-Ketoglutarate

5

38.1

40.0

Malate

5

30.7

30.0

Citrate

5

47.1

45.0

Isocitrate

5

45.3

45.0

c/i-'Aconitate

5

48.4

45.0

Succinate

5

34.3

35.0

Fumarate

5

34.0

34.0

Pyruvate

5

23.9

25.0

Oxalacetate

2.5

12.2

12.5

Substrate

Carbon Dioxide

Oxidized,

jumoles

Mmoles

/xmoles

a-Ketoglutarate

4.37

21.3

21.9

m-Aconitate

5.00

27.9

30.0

Isocitrate

3.79

21.7

22.9

Succinate

3.77

15.1

17.2

Fumarate

4.73

16.4

18.9

Each flask contained 1.5 ml. of kidney mitochondria, 3 yumoles
ATP, 4 yumoles MgS04, 50 /xmoles phosphate buff'er, pH 7.2, and
substrate as indicated.

After Green et al. (7).

70 METABOLIC PATHWAYS IN MICROORGANISMS

soluble enzymes had virtually no power to oxidize Krebs
cycle intermediates (succinoxidase activity), the mitochon-
dria by contrast showed a great ability to do so. In Table
3.2, Green et al. (7) have assembled all of the Krebs cycle
intermediates and have shown that kidney mitochondria
can oxidize these substrates completely, or, as shown in
the lower half of the table, can produce COo from these
oxidations in theoretical yields. If one assumes the pres-
ence of enzymes that will produce pyruvate from glucose in
an organism, it is clear that the oxidation of carbohydrates,
the first great class of foodstuffs, can be explained by the
activity of mitochondria; extramitochondrial Krebs cycle
oxidations are rare in most organisms.

The mitochondria may be shown to be potent fatty acid
oxidizers also. As Table 3.3 shows, caproic acid is oxi-
dized by rabbit liver mitochondria, in amounts equivalent

TABLE 3.3

Oxidation of Caproic Acid to Acetoacetic Acid, Carbon
Dioxide, and Water in Rabbit Liver Mitochondria

Caproate

Theoret-

Oxidized

ical

Calculated

Oxygen

Oxygen

from

Require-

Con-

AcAcOH

Oxygen

Caproate,

ment,

sumed,

Produced,

Consumed,

O2/CO2

/xmoles

diatoms

/xatoms

/xmoles

)umoles

(theory = 4)

15

120

105

13.2

13.1

3.97

15

120

101

13.1

12.7

3.85

15

120

96

13.1

12.0

3.68

30

240

92

12.1

11.5

3.80

30

240

142

17.9

17.8

3.93

30

240

163

21.9

20.4

3.74

After Cheldehn et al. (8).

CARBOHYDRATE METABOLIC PATHWAYS

71

to the acetoacetate produced (acetoacetate is generally
recognized as a by-product of fatty acid oxidation in the
liver). Moreover, it may be seen that all of the oxidized
fatty acid follows this route. The mechanism of fatty acid
oxidation is depicted in Fig. 3.4, and it will be observed
that there is continuing production of a two-carbon moiety
(acetyl CoA) which, as discussed above, is burned completely
to CO2, H2O, and energy via the Krebs cycle. All of the
enzymes connected with this ^-oxidation of fatty acids are
in the mitochondria, and although a few examples may be
found throughout the literature of extramitochondrial oxi-
dation of fatty acids, e.g., in peanut microsomes (9), the great
bulk of fatty acid oxidation appears to be of mitochondrial
origin. The oxidation of the second major class of food-
stuffs, viz., fats, thus appears to fall into line with carbo-

Fig. 3.4. Mechanism of fatty acid oxidation: the fatty acid spiral.
Example: octanoic acid.

72

METABOLIC PATHWAYS IN MICROORGANISMS

hydrate oxidation, so far as intracellular location of the en-
zymes is concerned.

The oxidation of the third major class of foodstuffs,
namely the amino acids, is siuiimarized briefly in Fig. 3.5.
Here it will be seen that many of the common amino acids
are in equilibrium with specific compounds in the Krebs
cycle (pyruvate, a-ketoglutarate, or oxalacetate) and exten-
sive research has shown that for the most part the specific
amino acid dehydrogenases, as well as the transaminases
necessary to produce these key intermediates, are located in
the mitochondria.

Alanine

Valine

Serine

C4

Threonine
Aspartic acid

Glutamic acid

Proline

Hydroxyproline

Arginine

Histidine

Lysine

Serine

^

Fig. 3.5. Protein degradation and tlic TCA c\cle.

CARBOHYDRATE METABOLIC PATHWAYS 73

CO2

Fig. 3.6 The Krebs cycle— an energy-producing pathway.

These experiments demonstrate what was said before
about the value of a concept: the idea gradually developed
that the mitochondria, and the mitochondria alone, could
furnish the crucial enzymes necessary for the oxidative
breakdown of the major foodstuffs. The scheme in Fig. 3.6
formalizes this; the Krebs cycle may also be considered as
an energy cycle, whose enzymes reside in the mitochondria
—as Claude has said, these are the power plants of the cell.
We will also see how concepts may become top-heavy; i.e.,
as students of comparative biochemistry added their find-
ings during the 1940's to the pool of knowledge, namely
that virtually every organism examined possessed an active,
functioning Krebs cycle, the notion began to prevail that
this pathway was the "prairie fire" of all terminal oxidation.

That the importance of the Krebs cycle may have been
oversold to biochemists (or oversubscribed by them) is
revealed in their reaction to the discovery of the Zwischen-

74 METABOLIC PATHWAYS IN MICROORGANISMS

terment enzyme, or the "direct" oxidation of glucose. This
pathway was discovered independently by Warburg, Lip-
mann, and by Dickens; it referred to the conversion of glu-
cose-6-P04 to pentose phosphate, using (usually) TPN as co-
enzyme. This work was done in 1935-1936 (10-12); yet
some fifteen years passed before most of the workers in
the field appreciated its significance, although Engelhardt
and Barkhash (13), as well as Dickens (14), felt that the
new oxidation of glucose might be charted by quite a dif-
ferent pathway from the glycolysis-Krebs cycle sequence. In
retrospect, it seems that the finding of a mechanism that
accounted for the formation of pentose satisfied the work-
ers of the day; certainly few people appreciated the possi-
bility that the "direct" oxidation of glucose might take on
major importance, since the Krebs cycle seemed adequately
to account for the terminal oxidation of sugars.

The findings of Claude's successors merely solidified this
view: the discovery that pentose phosphate could give rise
to sedoheptulose phosphate and tetrose phosphate was
brushed aside, and instead the new findings were empha-
sized for their simultaneous formation of triose phosphate.
The latter compound seemed important because it gave
rise easily to pyruvate, which in turn semed to provide a
ready entree to the Krebs cycle; it appeared, in short, as
if the excess pentose were simply being shunted into the
Krebs cycle via triose phosphate, and the name "hexose
monophosphate shunt" became for a time a popular term
to alternately describe the "direct" oxidation of glucose.

The work of Horecker and Racker and their respective
colleagues (15-17) changed this concept, for they discovered
that the enzymes involved were not of the nature of a shunt
mechanism at all, but instead constituted an intact system
that might be considered completely separately. Although
triose phosphate is formed in the transketolase reaction,

CARBOHYDRATE METABOLIC PATHWAYS 75

the consumption of this metabohte is assured ii transaldo-
lase is also functional. Although very recent years have
seen the determination that triose phosphate may feed into
either the Krebs cycle or the pentose cycle or both (Fig.
2.10), it is clear from Figs. 2.2 and 2.3 that these cycles need
not be interdependent; operation of the pentose pathway
can under certain circumstances lead to quantitative pro-
duction of CO2, as discussed in Chapter 2.

In attempting to assess the over-all importance of the
pentose cycle, a logical first question might be raised re-
garding the ubiquity of this complex of enzymes in living
tissue. Several laboratories, including our own, have sought
to demonstrate its activity in a variety of organisms, with
generally complete success. In addition to the isotopic
evidence cited in the previous chapter, enzymes of the pen-
tose cycle have been found in higher animals (18-32), insects
(33-37), marine organisms (38, 39), worms (40, 41), higher
plants (42-48), yeast and other fungi (49-58), as well as
bacteria (59-63). The enzymes of this complex seem, in
short, to be at least as widespread as those of the Krebs
cycle.

The demonstration of a metabolic pathway in an organ-
ism is fairly complicated. To carry out a complete search
for the Krebs cycle in a new organism, for example, would
probably involve at least five sets of operations: (a) attempts
to obtain complete oxidation of each intermediate; (b)
search for the individual enzymes, with (c) enough purifi-
cation of each so that the suspected intermediates may
accumulate and their formation be proved; (d) use of
certain inhibitors, e.g., malonate, to permit the recognition
of intermediates; and (e) the use of one or more isotopically
labeled substrates to determine whether the intramolecular
distribution of products is in agreement with expectations.
Likewise, a search for the pentose cycle would follow this

76

METABOLIC PATHWAYS IN MICROORGANISMS

general outline, except for the use of specific inhibitors;
there are no good ones that have been discovered that have
an exclusive effect on the pentose cycle.

We have carried out this detailed type of investigation in
a few situations; but I think you will agree that if one wishes
to survey a large number of new organisms, progress by this
means will be painfully slow. As opposed to this rigorous
method of investigation, we usually use a more "vigorous"
one, in which two general operations are employed: (a)
demonstration of the presence of the dehydrogenases for
G-6-P and 6-PGA; and (b) measurement of the non-oxida-
tive disappearance of ribose, followed by a search for the
transient formation of sedoheptulose and triose phosphates,
and the somewhat delayed accumulation of hexose phos-
phate. This type of behavior is noted in Fig. 3.7, which is

I I

I

+ — + Ribose-5-phosphate
* • Hexose phosphate

Sedoheptulose-7- phosphate

Triose phosphate

Fig. 3.7. Non-oxidative disappearance of added pentose from extracts
of pea aphid. Evidence for the pentose cycle in soluble extracts of the
pea aphid.

CARBOHYDRATE METABOLIC PATHWAYS

77

400

300-

S 200

QjO

100

jGlucose-6-P04
/'lRibose-5-P04

20

40 60 80

Time (minutes)

100

120

Fig. 3.8. Oxidations in rabbit liver mitochondria.

a diagram of these transformations in the pea aphid. The
accumulation of sugars proceeds as outhned in accordance
with expectations based on operation of the cycle.

In this connection, it seems appropriate to point out
that merely to find the individual enzymes of a pathway in
question is not enough to establish the operation of the
pathway. The enzymes formerly regarded as "belonging"
to glycolysis, for example, are, with the exception of phos-

78

METABOLIC PATHWAYS IN MICROORGANISMS

phofructokinase, all shared by either the pentose cycle
or the Entner-Doudoroff pathway.

A second question to which we have sought an answer
is that of intracellular location of the respective enzymes.
Whereas we have seen how the Krebs cycle complex is
located in the mitochondria, Fig. 3.8 shows that this is
obviously not true of the enzymes dissimilating G-6-P or
R-5-P. Citrate added to this preparation is oxidized over-
70% of theory, thus assuring the integrity of the prepara-
tion. In Fig. 3.9, the disappearance of pentose and accumu-

1

1

(M)

1 1

R-5-P.

o

- \

\.

^^

JS}^--

Hexose

- /

/^

• ^

(S)

^

~~~^

^Is)

^.^_R-5-P

^SH-7-P

1 1

10 20

30 40 50

Time (minutes)

60

Fig. 3.9. Pentose cycle reactions in rabbit kidney tractions (non-
oxidative).

M = mitochondria (residue, washed three times ^vith isotonic KCl)
S — kidnev fraction, sohible at 105,000fi^

CARBOHYDRATE METABOLIC PATHWAYS

79

R-5-P

S 105,000
S 25,000

10 20 30 40

Time (minutes)

50

Fig. 3.10. Disappearance of added pentose in rabbit kidney (non-
oxidative).

Sio5 000 = soluble after centrifuging 1 hour at 105,000g
S05 000 = soluble after centrifuging 30 minutes at 25,000g
Sqqq = soluble after centrifuging 20 minutes at 600g

lation non-oxidatively of SH-7-P and hexoses are seen to
be associated with the fraction sohible (in a centrifugal field
of 105,000g); in other words, the fraction of the cytoplasm
from which both the mitochondria and microsomes have
been removed. The 8105,000 fraction appears to be as active
as the 8(300 or 805,000 fractions, which contain either micro-
somes (805000) or also the mitochondria (Sqoo) (Fig. 3.10).

80 METABOLIC PATHWAYS IN MICROORGANISMS

This has been verified quantitatively (24). When the gravi-
tational field is increased to 144,000g, all the pentose cycle
enzymes are precipitated together (not shown); for this
reason we are often tempted to refer to the pentose cycle as
a complex; and although it may not be safe to draw firm
conclusions on such indirect evidence, nonetheless we are
taking as a working hypothesis the view that the entire com-
plex may be an intact submicroscopic unit within the cell.
These two major dissimilation pathways thus appear to
exist side by side in different parts of the cell. An inter-
esting point would be the determination of the control
agents that may be operating in vivo to direct the sub-
strate into one or the other pathway. This is as yet un-

TABLE 3.4

Effect of Pyridine Nucleotides on Specific Activity of C^^Oo
Arising from Labeled Glucose in Pig Heart Homogenates

CiVCe' *

No Pyridine
Size Nucleotides DPN TPN

Adult 1.00 1.02 2.29

Fetal (5.5 in.) 1.09 1.47 14.24

Flask components: 460 /zmoles of KCI; 20 //moles of phosphate
buffer, pH 7.4; 200 /xmoles of nicotinamide; 10 //moles of MgClo;
0.14 /imole of cytochrome c; 10 //moles of ATP (K salt); 10 /imoles
of glucose, 3 //moles of pyridine nucleotide; 2 ml. of homogenate in
tris buffer; 0.3 //c. of labeled glucose. The volume was 4.15 ml.;
the temperature, 37°; and the time, 5 hours.

* d' specific activity of C'^O? from glucose- l-C^"*
Ce' specific activity of C^'*02 from glucose-6-C^^
After Jolley, Cheldelin, and Newburgh (29).

CARBOHYDRATE METABOLIC PATHWAYS 81

certain, but in vitro experiments have pointed to the in-
fluence of CO factors such as Mg- + , DPT, and pyridine nu-
cleotides. We have already discussed some of these in the
first lecture, where in A. siiboxydans normally all the
carbohydrate that is broken down to CO^ and H^O tra-
verses the pentose cycle; but if Dowex adsorption is used
to remove Mg, DPT, and the pyridine nucleotides, then
re-addition of DPN allows some glycolytic reactions to oc-
cur. In the developing foetus, Jolley et al. (29) have found
that increasing the concentration of TPN can greatly in-
crease the ratio of pentose cycle "traffic" to the glycolysis—
Krebs cycle route. Table 3.4 shows that the oxidation of
glucose carbon 1 compared to C-6 in pig foetus hearts is
increased more than ten-fold by adding extra TPN to
the medium. Although this effect is greatest with homog-
enates, where the influence of TPN might be expected to
be felt to a relatively larger extent, the effect has also been
demonstrated in w^hole perfused hearts; Table 3.5 describes
this, where added TPN directs a doubling of the normal
ratio of oxidation of C-1 compared to C-6. The effect
is specific for TPN, and for the oxidation on the first car-
bon, as may be seen from the table in experiments with
DPN. This effect of TPN on oxidation routes has also
been noted by Wenner and Weinhouse (64), w^orking with
rat liver systems.

This last point gives rise to a third question: that of the
share of total cellular oxidations that may normally be
carried by the pentose cycle. This w^as discussed in Chap-
ter 2; it is apparently not a large figure in most organ-
isms, yet it is a significant amount. However, the reductive
pentose cycle may also be important, as is suggested by at
least three different experiments that are recorded in the
literature.

82 METABOLIC PATHWAYS IN MICROORGANISMS

TABLE 3.5

Effect of Pyridine Nucleotide on €^^02 Production from
Labeled Glucose by Perfused Adult Rat Hearts

Pyridine

Substrate

Nucleotide

Ratio

Glucose-1-Ci^

TPN

2.29

Glucose-l-Gi4

TPN

2.54

Glucose-6-Ci4

TPN

1.39

Glucose-6-Ci4

TPN

1.00

Glucose-l-G^^

DPN

1.32

Glucose-6-Ci4

DPN

1.39

Components per liter of medium: 140 mmoles of NaGl; 5.4 mmoles
of KGl, 3 mmoles of CaCl2; 80 //moles of nicotinamide; 900 /imoles
of phosphate buffer, pH 7.2; 1.15 mmoles of barbital buffer, pH 7.2.
The temperature was 37°. CO2 samples were withdrawn at 30-
minute intervals over a 5-hour period. The pyridine nucleotide con-
centration was 174 )umoles per liter.

[specific activity of G^^02 from heart 1
^ . 1 with pyridine nucleotide J

'specific activity of C^^02 from heart
without pyridine nucleotide

After Jolley, Gheldelin, and Newburgh (29).

1. The first observations pointing to a synthetic role for
the pentose cycle relate to the obvious need for ribose in
nucleotide synthesis, and for shikimic acid. Though the
biosynthesis of ribose would appear to be most simply
effected by the action of G-6-P and 6-PGA dehydrogenases,
indications are that this may not be the normal route.
Bernstein (65), in a study of the intramolecular distribu-
tion of C^^ in nucleotides of chicks fed labeled acetate,
glycine, and formate, found it impossible to relate his
findings to any oxidative scheme then known, but later

CARBOHYDRATE METABOLIC PATHWAYS 83

Horecker and Mehler (66) showed that Bernstein's results
were in agreement with expectations from the transaldolase-
transketolase sequence operating in reverse.

2. In higher plants, a close relative of the cycle, ribulose
1,5-diphosphate, acts as a primary acceptor of COo (67, 68);
the subsequent reactions leading to polysaccharide synthesis
appear to employ the reverse, or reductive pentose cycle
(69):

3 Pentose-P + 3 ATP -^ 3 Ru-di-P + 3 ADP

3 Ru-di-P + 3C02 + 3H2O -> 6 P-glycerate

6 P-glycerate + 6 ATP -> 6 di-P glycerate + 6 ADP

6 di-P glycerate + 6 TPNH + 6 H+

^ 6 GLA-P + 6 TPN+ + 6 PO/"
2 GLA-P ;=± 2 DHA-P
2 GLA-P + 2 DHA-P ^ 2 F-1, 6-di P
2 Fl,6-di P + 2H2O -^ 2 F-6-P + 2 Pi

F-6-P + GLA-P ^ Xu-5-P + E-4-P

F-6-P + E-4-P ^ SH-7-P + GLA-P

SH-7-P + GLA-P ^ 2 pentose-P

Sum: 3 CO2 + 9 ATP + 5 H2O + 6 TPNH + 6 H+

-^ GLA-P + 9 ADP + 6 TPN+ + 8 PO/~

3. A third experiment which points to a synthetic role
for the pentose cycle is the finding by Jolley et al. (29)
that this complex of enzymes is much more active in the
mammalian foetus than in the adult, and is most active
in the very young embryo (about one-fourth of the total
glucose oxidation appears to proceed by phosphogluco-
nate cleavage). There is a gradation toward adult character-
istics which is reached about three-fourths of the way
through the gestation period. From this point onward the
metabolic "traffic" in the heart of the foetus becomes vir-

84 METABOLIC PATHWAYS IN MICROORGANISMS

tually exclusively one following glycolysis and the Krebs
cycle, which is maintained throughout post-paitum life.
It is not understood why phosphogluconate cleavage dis-
appears altogether in the late foetal stages; the need for
rapid synthesis is much reduced at this point as compared
to the situation in the earlier foetus, and it seems even pos-
sible that the synthesis observed in a young animal is so
small (relative to oxidation) as to escape measurement by
the methods used.

Finally, as suggested in the review by Horecker and Hiatt
(70) a possible role of the pentose cycle (here operating
clockwise, or in the forward direction) may be to supply
TPNH to the system. In many species, the pentose cycle
is TPN-specific; glycolysis and Krebs cycle oxidations are,
on the other hand, largely DPN-requiring (a possible ex-
ception in some systems is isocitric dehydrogenase) so that
processes such as fatty acid synthesis may derive significant
amounts of needed TPNH from phosphogluconate cleavage.

CONCLUSION

Our ideas about carbohydrate metabolism have under-
gone considerable shifting during the past quarter century,
as glycolysis, the Krebs cycle, and the pentose cycle have
successively emerged as major pathways. Each of these
appears to contribute to the breakdown of carbohydrates;
in selected organisms, major use appears to be made
of one or another of these pathways. In many organisms,
however, a strong case appears to be developing for the
employment of the pentose cycle largely as a synthetic ap-
paratus. Time and the diligence of researchers in this
area will determine whether overlapping may occur with

CARBOHYDRATE METABOLIC PATHWAYS 85

other pathways yet undiscovered. Meanwhile, the three
major routes presently recognized seem to fill the con-
ceptual roles required for oxidation or synthesis, as well
as the interchange of metabolites that is convenient to the
over-all economy of the cell.

REFERENCES

1. Krebs, H. A., and W. A. Johnson, Enzymologia, 4, 148 (1937).

2. Wang, C. H., R. C. Thomas, V. H. Cheldelin, and B. E. Christensen,
/. Biol. Chetn., 197, 663 (1952).

3. Claude, A., Cold Spring Harbor Symp. duant. Biol, 9, 263 (1941).

4. Claude, A., /. Exptl. Med., 84, 51 (1946).

5. Palade, G. E., /. Histochem. Cytochem., 1, 188 (1953).

6. Lehninger, A. L., in Edsall, J. T., Enzymes and Enzyme Systems,
Harvard University Press, Cambridge, 1951.

7. Green, D. E., W. F. Loomis, and V. H. Auerbach, /. Biol. Chem.,
172, 389 (1948).

8. Cheldelin, V. H., I. S. Mirviss, and D. E. Green, Abstr. llSth
meeting, Am. Chem. Soc, No. 11, 5C (1950); Green, D. E., in
Edsall, J. T., Enzymes and Enzyme Systems, Harvard University
Press, Cambridge, 1951.

9. Humphreys, T. E., E. H. Newcomb, A. H. Bokman, and P. K.
Stumpf, J. Biol. Chem., 210, 941 (1954).

10. Warburg., C, W. Christian, and A. Griese, Biochem. Z., 282, 157
(1935).

11. Lipmann, F., Nature, 138, 588 (1936).

12. Dickens, F., Nature, 138, 1057 (1936).

13. Engelhardt, V. A., and A. P. Barkhash, Biokhimiya, 3, 500 (1938).

14. Dickens, F., and G. E. Clock, Biochem. J., 50, 81 (1951).

15. Horecker, B. L., and P. Z. Smyrniotis, /. Am. Chem. Soc, 74, 2123
(1952).

16. Racker, E. G. de La Haba, and I. G. Leder, /. Am. Chem. Soc, 75,
1010 (1953).

17. Horecker, B. L., Brewers' Digest, 28, 214 (1953).

86 METABOLIC PATHWAYS IN MICROORGANISMS

18. Abraham, S., P. Cady, and I. L. Chaikoff, /. Biol. Chem., 224, 955
(1957).

19. Black, A. L., M. Kleiber, E. M. Butterwoith, G. B. Brubaker, and
J. J. Kaneko, /. Biol. Chem., 227, 537 (1957).

20. Dische, Z., Naturwiss., 26, 252 (1938).

21. Racker, E., Advances in Enzymol., 15, 141 (1954).

22. Dickens, F., and D. H. Williamson, Biochem, J., 64, 567 (1956).

23. Srere, P. A., J. Cooper, V. Klybas, and E. Racker, Arch. Biochem.
Biophys., 59, 535 (1955).

24. Newburgh, R. W., and V. H. Cheldelin, /. Biol. Chem., 218, 89
(1956).

25. Murphy, J. R., and J. A. Muntz, /. Biol. Chem., 224, 987 (1957).

26. King, H. K., Sci. Progr., 45, 663 (1957) .

27. Villavicencio, M., and E. S. G. Barron, Arch. Biochem. Biophys., 67,
121 (1957).

28. Utter, M. F., A7in. Rev. Biochem., 27, 245 (1958).

29. Jolley, R. L., V. H. Cheldelin, and R. W. Newburgh, /. Biol. Chem.,
233, 1289 (1958); Biochim. et Biophys. Acta, 33, 64 (1959).

30. Dickens, F., Ann. N. Y. Acad. Sci., 75, 71 (1958).

31. Berardinis, de, E., Ophthalmologica, 138, 419 (1959).

32. Horecker, B. L., J. Hurwitz, and P. Z. Smyrniotis, /. Am. Chem.
Soc, 78, 692 (1956).

33. Chefurka, W., Biochim. et Biophys. Acta, 17, 294 (1955); Eyizy-
mologia, 18, 209 (1957).

34. Newburgh, R. W., and V. H. Cheldelin, /. Biol. Chem., 214, 37
(1955).

35. McGinnis, A. J., V. H. Cheldelin, and R. W. Newburgh, Arch.
Biochem. Biophys., 63, 427 (1956).

36. Hoskins, D. D., V. H. Cheldelin, and R. W. Newburgh, /. Gen.
Physiol., 39, 705 (1956).

37. Ito, T., and Y. Horie, Arch. Biochem. Biophys., 80, 174 (1959).

38. Krahl, M. E., A. K. Keltch, C. P. Walters, and G. H. A. Klowes,
/. Gen. Physiol., 38, 431 (1955).

39. Backstrom, S., Exptl. Cell Res., 18, 347 (1959).

40. Ley, de, J., and R. Vercruysse, Biochim. et Biophys. Acta, 16, 615
(1955).

41. Entner, N., Arch. Biochem. Biophys., 71, 52 (1957).

42. Conn, E. E., and B. Vennesland, /. Biol. Chem., 192, 17 (1951).

CARBOHYDRATE METABOLIC PATHWAYS 87

43. Anderson, D. G., H. A. Stafford, E. E. Conn, and B. Vennesland,
Plant Physiol., 27, 675 (1952).

44. Axelrod, B., R. S. Bandurski, C. M. Greiner, and R. Jang, /. Biol.
Chem., 202, 619 (1953).

45. Gibbs, M., Plant Physiol, 29, 34 (1954).

46. Vishniac, W., B. L. Horecker, and S. Ochoa, Advances in EnzymoL,
19, 1 (1957).

47. Clayton, R. A., Arch. Biochem. Biophys., 79, HI (1959).

48. Tabachnick, M., P. A. Srere, J. Cooper, and E. Racker, Arch. Bio-
chem. Biophys., 74, 315 (1958).

49. Warburg, O., and W. Christian, Biochem. Z., 238, 131 (1931); 242,
206 (1931); 287, 440 (1936).

50. Horecker, B. L., and P. Z. Smyrniotis, /. Biol. Chem., 193, 371
(1951).

51. Horecker, B. L., P. Z. Smyrniotis, and J. C. Seegmiller, /. Biol.
Chem., 193, 383 (1951).

52. de La Haba, G., L G. Leder, and E. Racker, /. Biol. Chem., 214,
409 (1955).

53. Glaser, L., and D. H. Brown, /. Biol. Chem., 216, 67 (1955).

54. Cooper, J., P. A. Srere, M. Tabachnick, and E. Racker, Arch.
Biochem. Biophys., 74, 306 (1958).

55. Hurwitz, J., and B. L. Horecker, /. Biol. Chem., 223, 993 (1956).

56. Couri, D., and E. Racker, Arch. Biochem. Biophys., 83, 195 (1959).

57. Newburgh, R. W., C. A. Claridge, and V. H. Cheldelin, /. Biol.
Chem., 214, 27 (1955).

58. Newburgh, R. W., and V. H. Cheldelin, Can. J. Microbiol, 5, 415
(1959).

59. MacDonald, J. K., V. H. Cheldelin, and T. E. King, /. Bacteriol,
80, 61 (1960).

60. Scott, D. B. AL, and S. S. Cohen, Biochem. J., 55, 23 (1953).

61. Mortenson, L. E., and P. W. Wilson, Arch. Biochem. Biophys., 53,
425 (1954).

62. Hauge, J. G., T. E. King, and V. H. Cheldelin, /. Biol. Chem., 214,
11 (1955).

63. Katznelson, H., Can. J. Microbiol, 4, 25 (1958).

64. Wenner, C. E., and S. Weinhouse, /. Biol. Chem., 219, 691 (1956).

65. Bernstein, I. A., /. Biol. Chem., 205, 317 (1953).

66. Horecker, B. L., and A. H. Mehler, Ann. Rev. Biochem., 24, 207
(1955).

88 METABOLIC PATHWAYS IN MICROORGANISMS

67. Quayle, J. R., R. C. Fuller, A. A. Benson, and M. Calvin, /. Am.
Chem. Soc, 76, 3610 (1954).

68. Weissbach, A., P. Z. Smyrniotis, and B. L. Horecker, /. Am. Chem.
Soc, 76, 3611 (1954).

69. Wilson, A. T., and M. Calvin, /. Am. Chem. Soc, 77, 5948 (1955).

70. Horecker, B. L., and H. H. Hiatt, New England J. Med., 258, 177
(1958).

INDEX

Italic page numbers indicate tables and figures.

Acetate, non-utilization by A. sub-

oxydans, 18, 19
Acetic acid bacteria, 1
Acetobacter suboxydans, acti\ity of

pantothenic acid conjugates, 2
calculation of carbohydrate oxi-
dative pathways, 47, 49
coenzyme-A-deficient cells of, 3
enzymes oxidizing glucose and

derivatives, 11
glycolysis in, 22

oxidation of by enzymes, acetal-
dehyde, 18

acetate, 18, 21

dihydroxyacetone, 12

glucose, 7

glucose-6-phosphate, 8

glycerol, 16

6-phosphogluconate, 8

pyruvate, 18

ribose phosphate, 14

Acetobacter suboxydans, oxida-
tion of by enzymes, sorbitol,
16, 17
oxidations of dinitrophenol, 4
oxidations in, general, 1
particulate enzymes, 5
pentose cycle in, 7, 14, 15
soluble enzymes in, 6
Acetoin, formation in A. suboxy-
dans, 19
Amino acids, formation in A. sub-
oxydans, 23

Bacillus subtilis, calculation of
carbohydrate breakdown path-
ways, 50, 51, 53

citric acid cycle in, 52

glycolysis in, 52

Calculation of pathway participa-
tion in glucose oxidation, 45,
46, 48, 49

89

90

INDEX

Carbohydrate oxidative pathways,

evaluation of, 30, 43, 44, 58, 59

Carbon atoms, of glucose, fate

when oxidized, 36, 37
Cell, diagram of generalized, 66

fractions, oxidative behavior, 68
Citric acid cycle, in acetic acid bac-
teria, 20
in A. suboxydans, absence of, 20
energy-generating mechanism of

cell, 73
in mitochondria, 68, 69
oxidation of amino acids by, 72
oxidation of fatty acids by, 70,

71
oxidation of intermediates of, 69

Dihydroxyacetone phosphate, oxi-
dation by A. suboxydans, 12

Dinitrophenol, effect on A. sub-
oxydans oxidations, 4

Entner-Doudoroff pathway of glu-
cose oxidation, 33
fate of individual glucose C
atoms in, 37

Equations describing oxidation of
labeled glucose, 43, 44, 58, 59,
61

Fatty acid oxidation, in mitochon-
dria, 70, 71

Flow sheet of oxidative reactions
in A. suboxydans, 15

Fructose-6-phosphate, oxidized by
pentose cycle, 32, 33

Gluconate, oxidation by B. sub-
tilis, 53, 56, 57

Glucose, isotopic, breakdown by
different pathways, 31
oxidation, by A. suboxydans, 7,

12, 45, 49 J

by B. subtilis, 50, 56 \

by bakers' yeast, 46
by Zymomonas ynobilis, 48
Gl ucose-6-phosphate, dehydrogen-
ase, properties, 9
oxidation in A. suboxydans, 8
Glycerol, oxidation by A. suboxy-
dans, 3

Isocitrase reaction, 24

Krebs cycle, see Citric acid cycle

Labeled glucose, breakdown by

different pathways, 31
Labeling, isotopic, fermentation

products from glucose, 35

Microsomes, in cells, diagram, 66,
67

oxidative behavior, 68
Mitochondria, in cells, diagram,
66, 67

oxidation of citric acid cycle in-
termediates, 68, 69

oxidation of glucose-6-phos-
phate, 77

oxidation of ribose-5-phosphate,
77

Oxidations, by acetic acid bac-
teria, 1

Pantothenic acid conjugates, activ-
ity for A. suboxydans, 2

INDEX

91

Pentose cycle, absence in mito-
chondria, 71 , 79
in A. suboxydans, 7 , 14, 15
in cells, 74, 76
in embryo tissue, 81
fate of individual glucose C

atoms in oxidation, 32, 33
non-oxidative portion of, 76
in photosynthesis, reductive, 81,

83
possible function in cells, 81, 83
reductive, 81, 55

simultaneous operation of gly-
colysis with, 54
Phosphogluconate cleavage, fate of

labeled glucose in, 31, 37
6- Phosphogluconate, dehydrogen-
ase, properties, 10
oxidation in A. suboxydans, 8
Phosphorylation, in A. suboxy-
dans, 25
Protein breakdown, in mitochon-
dria, 72
via citric acid cycle, 72
Pyruvate, oxidation by A. suboxy-
dans, 18
Pyruvic carboxylase, in A. suboxy-
dans, 18

Quantitative expression of path-
way participation in carbohy-
drate oxidation, 43, 44, 58, 59

Radiorespirometer, diagram of ap-
paratus for, 40, 42

Radiorespirometry, 39

ec|uations describing, 43
Recovery of CO^ (labeled) during

glucose oxidation, 56, 57
Ribose-5-phosphate, oxidation by

A. suboxydans, 14

Sedoheptulose, relation to other
sugars, 32, 33

Soluble fraction of cells, oxidation
in, 78, 79
pentose cycle in, 78, 79

Sorbitol, conversion to fructose,
16, 17
formation of sorbose from, 16,
17

Sinnmary of carbohydrate oxida-
tive pathways in microorgan-
isms, 61

Synthetic role of pentose cycle, 82,
83, 84

TPN, effect on pentose cycle, 80,

81
Triose phosphate, oxidized by

pentose cycle, 32, 33

Yeast, calculation of carbohydrate
breakdown pathways, 46

Zymomonas mobilis, calculation of
carbohydrate breakdown path-
ways, 48

I

[S^iSI

— ■ v^jia

1 0^WzW.':^^^^S^^^B-

m^mmmif^

-p?:r/f;^-=^il:iteS;^::^?;jis2

ffiiiliiiiniBliS^fiB