Temperature and Enzyme Activity in Poikilotherms: Liver Soluble

TEMPERATURE AND ENZYME ACTIVITY IN POIKILOTHERMS: LIVER SOLUBLE

NADP+-LINKED ISOCITRATE DEHYDROGENASE FROM TROUT

by

THOMAS WILLIAM MOON

B.Sc. Oregon State University, 1966 M.A* Oregon State University, 1968

A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

IN THE DEPARTMENT of ZOOLOGY

We accept this thesis as conforming to the required standard:

THE UNIVERSITY OF BRITISH COLUMBIA

September 1971 In presenting this thesis in partial fulfilment of the requirements for r

an advanced degree at the University of British Columbia, I agree that

the Library shall make it freely available for reference and study.

I further agree that permission for extensive copying of this thesis

for scholarly purposes may be granted by the Head of my Department or

by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.

Department of

The University of British Columbia Vancouver 8, Canada

Date £&jk.QJd. ABSTRACT

The effect of temperature on the oxidative-decarboxylation of isocitrate by the soluble NADP+-linked isocitrate dehydrogenase (NADP-

IDH, EC 1.1.1.42) from rainbow trout (Salmo gairdnerii) liver has been investigated. A particular interest was given those properties of the enzyme which might help to explain the temperature-independent function of the Krebs cycle and the large increase in fatty acid synthesis known to occur during low temperature acclimation.

Within the thermal range experienced by rainbow trout, control of catalysis by this enzyme is temperature-independent. Acclimation to an altered thermal regime is accompanied by an increase in the relative proportion of the slowest migrating isozyme of liver NADP-IDH on starch- gel electrophoresis. These cold- and warm-isozyme variants display different and adaptive Km-temperature relationships, and allow for temperature-independent modulation of enzyme activity through the entire thermal range this species is likely to encounter in nature.

Other trout species, including the brook (Salvelinus frontinalis), lake (Salmo namaycush) and their hybrid, the splake trout, were invest• igated for similar responses. The elaboration of enzyme variants in brook and splake trout are complexly regulated by temperature changes, but the lake trout genome contains a single gene coding for liver NADP-

IDH which is not affected by temperature.

Catalysis by the trout liver enzyme is modulated not only by temp• erature, but also ADP and^X-KGA. Both of these metabolites alter the

Km of DL-isocitrate; at physiological ADP concentrations, the Km is reduced as it is with^K-KGA below 0.05 mM, but at higher 0(-KGA concent- rations it is markedly increased. These two controls suggest this

enzyme may be important for the Krebs cycle oxidation of isocitrate.

The availability of a purified NADP-IDH from pig heart allowed

a study of the kinetic properties of homologous enzymes from both a

poikilotherm and a homeotherm. Even though the molecular weights, Ea values, substrate, cofactor and inhibitor specificities are similar,

subtle changes in enzyme structure and/or conformation as identified by

electrophoresis, may result in the observed differences in temperature

characteristics. These apparent adaptive enzyme responses are of

importance to the rainbow trout which lives in a fluctuating thermal

regime, but not the pig which does not experience these changes.

In vivo, the response of enzymes to temperature fluctuations may

be quite different to those seen in vitro. The locus(i) coding for

rainbow trout liver NADP-IDH was found to contain a large amount of

heterogeneity; in fact, seven distinct phenotypes were found to co•

exist in one hatchery population. The kinetics of three of these

phenotypes were investigated and it was found that by increasing the

number of slow moving isozymes, an increase in Km(DL-isocit) at high

assay temperatures occurs. This suggests that irrespective of changes

in the cellular milieu, isozymal content can determine the Km-temperature

response.

The data from this study suggest that changes in enzyme-substrate

affinity with temperature as a result of either the temperature directed

production of enzyme variants and/or their genetic expression, are

important in controlling the catalytic activity of NADP-IDH from the

eurythermal rainbow trout. Also, unlike the mammalian enzyme, the trout

liver enzyme may be important in both fatty acid synthesis and the Krebs

cycle oxidation of isocitrate. iii

TABLE OF CONTENTS

Page

Abstract i

List of Tables vi

List of Figures viii

Acknowledgements xi

Chapter I: Introduction 1

Temperature and the branch point of lipogenesis and the

Krebs cycle 3

The NADP+-linked isocitrate dehydrogenase 7

Chapter II: Materials and Methods 13

Fish and the acclimation process 13

Preparation of NADP-IDH 14

Assay of NADP-IDH activity 15

Estimates of protein content 16

Starch-gel electrophoresis 16

Isoelectric focusing of NADP-IDH 17

DEAE-cellulose chromatography 18

Sucrose gradient ultracentrifugation 18

Gel-filtration of NADP-IDH 19

Chapter III: Effects of thermal acclimation on multiple forms

of trout liver NADP-IDH 20

Introduction 20

Results 22

Winter lake, splake and brook trout S-form IDH 22

Spring lake, splake and brook trout S-form IDH 24

Summer splake and brook trout S-form IDH 26 IV

Page

Discussion 31

Chapter IV: Temperature and enzyme activity in poikilotherms 34

Introduction 34

Results and discussion 36

NAD- vs NADP-IDH 36

Isozymes of liver NADP-IDH 36

Effects of temperature on Vm 42

Effects of temperature on Km 42

Effects of temperature on the Michaelis constants

of Mg"1^ and Mn++ 49

Effects of temperature on the pH optimum 51

Effects of other metabolites 51

Effects of temperature acclimation on enzyme activities 58

Chapter V: Comparison of the pig heart and rainbow trout

liver NADP-IDH 59

Introduction 59

Results 61

Characterization of NADP-IDH 61

Comparison of kinetic constants 67

Effects of inhibitors 67

Discussion 76

Chapter VI: Effects of temperature on individual NADP-IDH

phenotypes of rainbow trout liver 79

Introduction 79

Results 81

Changes in the cellular milieu 81 Page

Kinetics of trout liver NADP-IDH isozymes 84

Discussion 89

Chapter VII: Summating remarks 93

Chapter VIII: Literature cited 102 VI

LIST OF TABLES

Table Page

III, 1. Frequency distribution of the three S-form IDH

isozyme patterns from summer brook trout liver 30

IV, 1. Tissue-specific NADP-IDH activities of high

speed supernatant from rainbow trout 37

IV, 2. Q-J^Q values for the warm- and cold-enzyme variants

at various assay temperatures 48

IV, 3. Km(cations) for the cold- and warm-enzyme variants

at various assay temperatures 50

IV, 4. The effects of various metabolites on the activity

of trout liver NADP-IDH 53

IV, 5. The effects of ADP on the Km(DL-isocit) of the v

cold-NADP-IDH variant of trout liver 56

V, 1. QIQ values for the pig heart NADP-IDH at various

DL-isocitrate concentrations 69

V, 2. The effects of certain citric acid cycle inter•

mediates on the activity of pig heart NADP-IDH 70

V, 3. The effects of temperature on the Km(DL-isocit)

in the presence of 0.75 mM (C<-KGA) 73

V, 4. The effects of temperature on the Ki(°<-KGA) for

pig heart and trout liver NADP-IDH 75

VI, 1. The relative distribution of isozymic forms of

trout liver NADP-IDH from a hatchery population 86

VI, 2. Q-J^Q values between 10 and 20°C for the A2 and

A2,B2>C2 trout liver NADP-IDH at various

DL-isocitrate concentrations 91 Table

VII, 1. Comparative properties of rainbow trout liver

and pig heart NADP-IDH viii

LIST OF FIGURES

Figure Page

III, 1. Composite electrophoretogram of winter lake,

splake and brook trout liver NADP-IDH 23

III, 2. Resolution of spring acclimated brook trout

liver S-form IDH 25

III, 3. Diagrammatical representation of spring brook

trout liver S-form IDH 27

III, 4. Diagrammatical representation of S-form IDH

isozymes from summer brook trout liver 28

III, 5, a. Composite electrophoretogram of liver

S-form IDH from the family Salmonidae 32

b, Diagrammatical representation of possible

liver S-form IDH subunit types 32

IV, 1. Starch-gel electrophoretogram of the cold- and

warm-enzyme variants with densitometer traces 38

IV, 2. Concurrent electrofocus separations of trout liver

cold- and warm-NADP-IDH enzyme variants 40

IV, 3. DEAE-cellulose chromatography of NADP-IDH from

trout liver 41

IV, 4. Arrhenius plots of trout liver NADP-IDH from

cold- and warm-acclimated animals 43

IV, 5. DL-isocitrate saturation curves and Lineweaver-

Burk plots at different assay temperatures for the

cold- and warm-acclimated trout liver NADP-IDH 44

IV, 6. NADP+ saturation curves and Lineweaver-Burk

plots at different assay temperatures 45 ix

F igure Page

IV, 7. Km(DL-isocitrate) for the cold- and warm-NADP-IDH

variants from trout liver at different temperatures 47

IV, 8. Relative NADP-IDH activities plotted at two

temperatures and various pH values 52

IV, 9. The effect of CX -KGA on the Km(DL-isocit) for

the cold-NADP-IDH variant 57

V, 1. Starch-gel electrophoresis of NADP-IDH from pig

heart and trout liver 62

V, 2. Sucrose density ultracentri fu*at ion and Sephadex

G-100 gel filtration of trout liver and pig heart

NADP-IDH 63

V, 3. Sucrose density ultracentrifugation of trout liver

NADP-IDH in the presence of substrates, cofactors

and/pr inhibitors 65

V, 4. Arrhenius plots of trout liver and pig heart NADP-IDH 66

V, 5. The effects of temperature on DL-isocitrate saturation

curves for pig heart NADP-IDH 68

V, 6, The effects of°(-KGA inhibition on DL-isocitrate

saturation curves for pig heart NADP-IDH 72

VI, 1. The effects of changing substrate concentrations

on the Km of the alternate substrate for trout

liver NADP-IDH 82

VI, 2. The effects of enzyme protein concentration upon

the Km(DL-isocit) of trout liver NADP-IDH 83

VI, 3. Electrophoretogram of six trout liver NADP-IDH

phenotypes 85 Figure Page

VI, 4. The effects of tissue isozymal content on the

Km(DL-isocit) vs temperature relationship for

trout liver NADP-IDH 87 ACKNOWLEDGEMENTS

I would like to thank the members of my committee, Drs. G. I.

Drummond, J. E. Phillips and D. J. Randall for their comments concerning

this text. Special thanks goes to my supervisor, Dr. P. W. Hochachka,

whose comments and encouragement have allowed this thesis work to

come this far, and who has made my graduate training such a rewarding

and exciting experience. Also, I would like to thank the other members of Dr. Hochachka's group, including Drs. G. N.Somero,

J, Baldwin and H. Behrisch for their criticism and help with this study.

Tariq Mustafa, as a fellow commrade in arms, has helped to make my work

at UBC interesting. I must thank my parents, wife and child for their moral support throughout my graduate career. And finally, the

Fisheries Research Board of Canada which has supported both myself

and this study throughout. CHAPTER I:

Introduction " the cold-blooded animal successfully adopting ingenious mechanisms, first biochemical, then physiological, in order to adapt its heart to the variations of its environment; the warm-blooded animal discarding what its cold-blooded predecessor has laboriously beaten out, invoking the nervous system to reverse the normal biochemical relationship and gaining a new freedom by adapting, not itself to the internal environment, but the internal environment to itself."

Barcroft (1934, p. 62)

The initial response of cold-blooded or poikilothermic animals to a change in environmental temperature is a rapid alteration in its physiological processes. This instantaneous or immediate response is - f ollowed by a series of slower biochemical and physiological changes in the direction of the original state resulting in a new equilibrium condition. The term used to define this slower process is acclimatiz- a tion, if the process occurs in the organism's habitat, or acclimation, if iti is induced in the laboratory.

Thermal acclimatization occurs seasonally in aquatic poikilotherms residing in temperate freshwater streams. It is possible, however, that the thermal stress will be of longer duration resulting in evolutionary adaptation, a process requiring many generations to complete. In all cases, thermal compensation, be it immediate, acclimation or evolutionary adaptation, counteracts the possible deleterious effects of a fluctuating or constantly reduced thermal regime encountered in Nature by poikilo• thermic animals.

Early investigations into the physiological basis of thermal compen•

sation and particularly acclimation, dealt mainly with definitions of the

"scope for activity" (Fry, 1947), thermal lethal limits (Brett, 1956), 2

and other whole organism functions (Prosser & Brown, 1961; Brett, 1970;

Fry & Hochachka, 1970). In recent years an interest has arisen in the cellular mechanisms of these changes.

Subsequently, the process of thermal acclimation has been found to be accompanied by a fundamental reorganization of cellular metabolism

(Hochachka, 1967). These changes are exceedingly complex and have been reviewed recently by Hochachka & Somero (1971) and Somero & Hochachka

(1971).

The studies reported in this thesis will deal with a single aspect of this problem: the control of enzyme activity. What mechanisms are

available to allow enzymes to function efficiently in the face of environmental temperature fluctuations? Has the process of evolution through natural selection resulted in poikilothermic enzymes that are

fundamentally different from the homologous homeothermic enzymes? That

is, can adaptation to temperature be observed at the enzyme level?

Isocitrate dehydrogenase (threo-Ds isocitrate: NADP oxidoreductase

(decarboxylating), EC 1.1.1.42) from the liver of rainbow trout (Salmo

gardnerii) was selected because of its potential function in both the

tricarboxylic acid cycle (Krebs cycle) and lipogenesis. 3

Temperature and the branch point of lipogenesis and the Krebs cycle.

Citrate occupies a metabolic branch point of considerable importance, leading not only to its further oxidation and the eventual production of

energy in the form of ATP, but also to fatty acids (Atkinson, 1966; 1968).

According to Atkinson, competition for citrate by ATP-citrate lyase and

NAD+-linked isocitrate dehydrogenase will be regulated by the adenylate

energy control system. When ATP concentrations are high, citrate

traverses the mitochondrial barrier freely and is converted by ATP-citrate lyase to acetyl-CoA which serves as the precursor of fatty acids. As

ADP or AMP levels increase, ATP-citrate lyase is strongly inhibited, and the NAD+-linked isocitrate dehydrogenase is activated. This mechanism assures that citrate and ATP will be converted to storage compounds only when the level of ATP is high. Competition for citrate by these

enzymes may be an important mechanism to explain the observed effects of temperature on lipogenesis and Krebs cycle activity known to occur during thermal acclimation (Hochachka, 1968; Dean, 1969; Newell & Pye,

1971).

The effects of thermal acclimation on the participation of these d ifferent metabolic pathways is dependent upon the tissue examined

(Hochachka, 1967). Ekberg (1958) reported warm-adapted gill tissue of goldfish to be more resistant to 10 ^M cyanide, but more senstive to

iodoacetate (5.4 x 10-^ M) when compared to the same tissue in cold-

adapted goldfish. It was implied that in the warm-adapted gill tissue,

either the cytochrome system was altered or another oxidative pathway

less sensitive to cyanide was utilized. Similarly, carbon flow through

the pentose pathway was diverted in these warm-adapted fish. These

oberservations could not be made for liver or brain preparations under 4 identical conditions.

The rate of oxygen consumption in goldfish liver homogenates was f ound by Kanungo & Prosser (1959) to be higher in cold-adapted than in warm-adapted fish, although the efficiency of phosphorylation ( as measured by P/0 ratios) was decreased in the cold. A number of possible mechanisms leading to decreased P/0 ratios in the cold were discussed, including uncoupling of phosphorylation, increasing mitochondrial ATPase activity, increasing the "Calorigenic shunt" for NADPH oxidation via a cytochrome c system, or an increased activity of the pyridine nucleotide transhydrogenase. None of these alternatives, however, were examined.

They concluded by suggesting that metabolic adaptation to temperature in goldfish occurs by the quantitative change in activities of serveral enzyme systems.

Oxygen consumption studies, at best, are not adequate to indicate specific shifts in metabolic pathways during thermal compensation.

Isotopic incorporation into specific metabolic intermediates have lead to

some understanding of the changes occurring at this time.

Hochachka & Hayes (1962) demonstrated directly a shift favoring higher levels of pentose cycle participation accompanying acclimation to low temperatures from the predominant Embden-Meyerhof and Krebs cycle carbon flow normally observed in the eastern brook trout, Salvelinus frontinalis. These data suggested an altered acetate metabolism with low temperature adaptation.

Acetate-l-^C metabolism has been examined by Dean (1969) in both muscle and liver tissue from warm- and cold-adapted rainbow trout, Salmo ga'rdnerii. Cold-adapted fish muscle showed higher turnover values for both labelled acetate and palmitate compared wJ :h warm-adapted tissue, 5 again suggesting an alter Krebs cycle activity. Liver tissue from cold- adapted fish also demonstrated a higher oxidative activity, although at an intermediate temperature (11.5°C), there was no difference between the cold- and warm-adapted tissues. At this temperature (11.5°C), there was no difference between the amount of acetate incorporated into ^4C02, but a marked difference in incorporation into total lipids. This again suggests a shift favoring lipogenesis resulting from the more efficient competition by this pathway for common metabolites.

These data generally conclude that carbon flow through the Krebs cycle is more or less independent of temperature. Hochachka (1968) suggests that citrate and ATP reduce the effectiveness of the Krebs cycle competing for acetyl-CoA, resulting in increased conversion to lipids.

This mechanism is so efficient that in the electric organ of Electrophorus electricus, mM concentrations of citrate allow for acetate oxidation to preceed independent of temperature over a 20°C range. Hochachka did not establish the precise mode of action of ATP or citrate, but it is thought to be primarily on acetyl-CoA-carboxylase (Waite & Wakil, 1962), the citrate cleavage enzyme (ATP-citrate lyase) or citrate synthase (Hathaway

& Atkinson, 1965). Citrate synthase inhibition by ATP has been investigated by Hochachka & Lewis (1970) and is found to be complexly regulated by pH and temperature, as was suggested earlier (Hochachka, 1968).

The increase in lipogenesis observed during incorporation studies results in an altered lipid environment at low temperatures. In all cases reported, the proportion of unsaturated to saturated fatty acids increase in the cold. This altered lipid content has been found for goldfish brain lipids (Johnston & Roots, 1964), muscle and organ lipids of rainbow trout (Knipprath & Mead, 1965), total lipids of mosquito-fish 6 and guppies (Knipprath & Mead, 1966), goldfish intestinal lipids (Kemp

& Smith, 1970), and goldfish gill mitochondrial lipids (Anderson, 1970;

Caldwell & Vernberg, 1970). It is also known that very specific lipid

changes can occur (Knipprath & Mead, 1968; Roots, 1968; Anderson, 1970).

Roots (1968) found that specific phospholipids are altered, probably as

related to nerve function.

Membrane lipid changes are extremely important due to the number of

enzymes associated with them. Smith & Kemp (1969) have found that membrane ATPase behaves differently in membranes containing different

fatty acids. This type of data may be explained by alteration in the

conformation of the ATPase protein resulting from phase changes in the membrane lipid component (Kumamoto, Raison & Lyons, 1971), as recently

seen for the mitochondrial succinate dehydrogenase and cytochrome oxidase

systems (Raison, Lyons & Thomson, 1971).

Transition temperatures in Arrhenius plots for the transport of

lactose or lactose analogs have been found for various unsaturated fatty

acid auxotrophs of E_. coli. The type of fatty acid supplied in the growth medium will determine the transition temperature (Schairer & Overath,

1969; Fox, Law, Tsukagoshi & Wilson, 1970; Overath,; Schairer & Stoffel,

1970; Wilson, Rose & Fox, 1970). Esfahani, Barnes & Wakil (1970) have

reported an increase in trans-fatty acid incorporation into an E_. coli

fatty acid auxotrophy at 37°C compared to 27°C; at 27°C there was an increase in cis-fatty acid incorporation over 37°C,

These changes in total lipids and specific membrane lipids are

extremely important to the field of thermal adaptation since they demon• strate unequivocally (1) that the temperature range of a given kind of

enzyme function can be determined by the associated phospholipids, and 7

(2) that the physical state of the membrane depends upon temperature

(Esfahani, Limbrick, Knutton & Wakil, 1971). It had been assumed that

these membrane properties were critical to thermal adaptation (Hochachka,

1967; Hochachka & Somero, 1971), but only with the work of Fox, Wilson,

Esfahani,.-Wakil and others have these suggestions been given...an experimental basis.

In order for extramitochondrial lipogenesisto continue, reducing

equivalents in the form of NADPH are necessary (Wakil, Titchiener &

Gibson, 1959). One enzyme which has this activity is the NADP+-linked isocitrate dehydrogenase (Lowenstein, 1961a,b; Savard, Marsh & Howell,

1963; Hanson & Ballard, 1967; O'Hea & Leveille, 1968; Walker & Bailey,

1969; Bauman, Brown & Davis, 1970; Flint & Denton, 1970).

The NADP"1"-linked isocitrate dehydrogenase.

The presence o f NADP+-1inked isocitrate dehydrogenase (NADP-IDH)

activity has been noted from most tissues of vertebrates thus far

examined (Plaut, 1963) and many microorganisms (Hampton & Hanson, 1969).

In microorganisms and plants, the partitioning of isocitrate between the

Krebs and glyoxylate cycles occurs through the concerted inhibition of

NADP-IDH by glyoxylate and oxaloacetate (Ozaki & Shiio, 1968). The vert•

ebrate enzyme is likewise subjected to concerted inhibition by these

metabolites (Ozaki & Shiio, 1968; Shiio & Ozaki, 1968; Hampton & Hanson,

1969; Marr & Weber, 1969a,b; Charles, 1970) even though the glyoxylate '

cycle is absent (Mahler & Chordes, 1966). This apparent general property

of all NADP-IDHs may result from the effects of glyoxylate and oxaloacetate

on the aggregation state of the enzyme subunits (Kemper & Kaplan, 1971).

This enzyme is under adenylate control in microorganisms. Ones 8 reason for the acceptance of the NAD+-linked IDH enzyme being important

in isocitrate oxidation within the mitochondria is that the enzyme is modulated by ADP or AMP (Plaut & Aogaichi, 1968). Parker & Weitzman

(1970) identified two kinetically, molecularly and electrophoretically

distinct NADP-IDH enzymes in Acinetobacter lwoff; isozyme II (the

higher molecular weight species), in the presence of 1 mM AMP or ADP,

is activated five-fold and two-fold respectively, while 1 mM ATP had

essentially no effect. These results resemble closely those obtained by Atkinson, Hathaway & Smith (1965) for the NAD-IDH from yeast.

Marr & Weber have identified ATP as a noncompetitive inhibitor of Crithidia fasciculata (a protozoan) NADP-IDH (1969a,c) and ADP and -

ATP as inhibitors of the same enzyme in Salmonella typhimurium (1968) .

Although not conclusive, entropy and free energy changes are consistent

with probable physical changes in the enzyme during binding of these

inhibitors. Higashi, Maruyama, Otani & Sakamoto (1965) attribute adenylate

inhibition of this enzyme with metal ion chelation.

The only established function for metazoan NADP-IDH is in the pro• duction of reducing equivalents for lipogenesis as previously mentioned.

There is considerable disagreement among authors with regard to the

relative contribution of the NAD- and NADP-linked IDH enzymes towards

oxidation of isocitrate in mitochondria (Nicholls & Garland, 1969).

However, with the recognition of the activation of NAD-IDH by ADP

(Chen & Plaut, 1963) or phosphate (Goebell & Klingenberg, 1963), the cold-

lability of NAD-IDH (Plaut & Aogaichi, 1967) and the activation by malate

of isocitrate permeation into rat liver mitochondria (Chappell &

Robinson, 1968), the significance of the NAD-IDH enzyme in mammalian mitochondrial oxidative metabolism is widely accepted, while the question 9 of potential roles for the NADP-IDH enzyme has been left largely unconsid• ered. In fact, Colman, Szeto & Cohen (1970) suggest the importance of this enzyme is in its functional similarities with NAD-IDH which will allow the comparative study of an allosteric y_s non-allosteric enzyme.

In lower vertebrates, including fish, the function of this enzyme is unknown, but the lack of an active NAD-IDH in these organisms (Crabtree

& Newsholme, 1970) may put this enzyme in an unique position.

The enzyme has been purified from a number of microorganisms, but the major source has been the pig heart cytoplasm (Plaut, 1963). Colman c(1968) and Colman, Szeto & Cohen (1970) have investigated the properties of this enzyme under a number of conditions and suggest that it is a single polypeptide chain with a molecular weight of 58,000. Reports of a similar molecular weight are available (Moyle & Dixon, 1956), although

Magar & Robbins (1969) suggest the enzyme is a dimer. The subunit structure of NADP-IDH from pig heart has recently been found to result from enzyme aggregation and that the extent of aggregation controls the direction of enzyme catalysis (Kemper & Kaplan, 1971). The active species catalyzing the reduction of NADP+ is a 30,000 molecular weight monomer; whereas, the active species catalyzing the oxidation of NADPH is a

120,000 molecular weight tetramer. These conflicting data concerning the subunit composition of this pig heart enzyme have yet to be resolved.

The pig liver enzyme has been studied by Illingworth & Tipton (1970) and its properties vary somewhat from the heart enzyme: a molecular weight 1.3 times that reported for the pig heart enzyme, and a dimer subunit structure have been accepted. The dimeric form of pig liver and mouse muscle soluble NADP-IDH has been confirmed by genetic analyses

(Henderson, 1965; Mintz & Baker, 1967). In bacteria, a similar molecular 10 weight as that found by Illingworth & Tipton has been reported (Chung

& Franzen, 1969; Howard & Becker, 1970; Barrera & Jurtshuk, 1971).

Early work by Lowenstein & Smith (1962) and Baron & Bell (1962)

indicated that NADP-IDH existed as immunologically distinct enzymes, or

isozymes. These forms are found in the soluble and mitochondrial portion

of higher vertebrate cells. Using breeding experiments between normal and mutant allele-carrying mice strains, Henderson (1965) found that only

the soluble enzyme from both heart and liver appeared as heterozygous

forms. Such evidence indicated that the soluble and mitochondrial IDH

enzymes are not genetically linked, and that the soluble form is a dimer.

Similar experiments by Mintz & Baker (1967) on muscle NADP-IDH gave

identical conclusions. Higashi, Maruyama, Otani & Sakamoto (1965)

studied beef heart soluble and mitochondrial isozymes of NADP-IDH and found that each had similar catalytic properties, but the soluble enzyme was more sensitive to inhibitors.

Multiple forms of NADP-IDH have also been investigated in micro• organisms. A strain of E_. coli grown on acetate as its sole carbon

source produces an enzyme variant completely different than the two pro• duced by the same organism with glucose as a carbon source (Reeves,

Brehmeyer & Ajl, 1968a,b). This suggests that NADP-IDH may be inducible

in these bacteria. Self & Weitzman (1970) found two isozymes of

Acinetobacter lwoff by zonal centrifugation which differed kinetically, molecularly and electrophoretically. Also, Tait (1970) has shown mitochondrial and soluble isozymes in Paramecium aurelia which are not genetically linked.

Some electrophoretic investigations have been carried out on the

enzyme from salmonid fishes. Wolf, Engel & Faust (1970) observed poly- 11 morphic forms of the soluble liver NADP-IDH of the rainbow trout Salmo

irideus, results suggesting only a single locus for the enzyme existed.

However, the heart muscle mitochondrial NADP-IDH could be shown to be derived from two different gene loci, always giving three electrophoretically distinct isozymes in all individuals. The liver of this fish contained

no mitochondrial NADP-IDH, although the heart showed both forms of the ,

enzyme. The authors concluded that the locus coding for the heart mitochondrial NADP-IDH has undergone complete diploidization to two

separate loci; however, the liver soluble NADP-IDH is still subjected to

;tetrasomic inheritance. This diploidization mechanism of phylogenetically

tetraploid organisms has been discussed by Ohno (1970).

Other tetraploid fishes have also been studied. Quiroz-Gutierrez

& Ohno (1970) found that the tetraploid goldfish and carp are both endowed

with two gene loci for the soluble NADP-IDH in liver and heart. In both

studies, the enzyme was found to be a dimer.

Lin, Schipmann, Kittrell & Ohno (1969) have found that at least

5 0% of a wild population of Lake Erie goldfish are heterozygous at the

sorbitol dehydrogenase locus. It has been suggested by Quiroz-Gutierrez

6 Ohno (1970) that an identical situation exists with regard to the

soluble IDH locus of this^ame fish population. A polluted environment,

it was felt, may strongly select for heterozygosity at a number of gene

loci, including that coding for soluble IDH. However, no other parameter,

such as for example temperature, was implicated in the maintenance of

heterozygosity at this locus.

The advantage of using the rainbow trout Salmo gairdnerii as an

experimental organism relates to its apparent tetraploid evolution, a topic

discussed in Ch, III of this thesis. The tetrasomic locus consists of 12 initially four identical alleles in which the chance of mutation is greater than in a diploid organism since each allele is subject to less selective pressure (Ohno, 1970; Wolf, Engel & Faust, 1970). This increase in genetic material is the basis for natural selection.

Haldane (1955) and Mayr (.1963) have emphasized that geographical and temporal changes in the environment could place a premium on metabolic flexibility. As a result, the large amount of potential gene products

(i.e., proteins) available in a tetraploid organism such as the rainbow trout, may be of an adaptive significance.

With these considerations in mind, a study was initiated to examine the properties and particularly the thermal behaviour of NADP-

IDH from the liver of the rainbow trout, Salmo gairdnerii, a highly eurythermal species of the family Salmonidae. CHAPTER II:

Materials and Methods 13

MATERIALS AND METHODS

Fish and the acclimation process.

Rainbow trout (Salmo gairdneri) were purchased from the Sun Valley

Trout Farm, Mission, B.C. for all experiments. Prior, to acclimation,

the trout were stored in a large outside holding tank supplied with a constant flow of subterranean water and fed twice weekly with New Age

Fish pellets (Moore-Clark Co., Salt Lake City, Utah). These fish were used as a source of fresh tissue when required. Also, trips were made to the trout farm in order to excise a large number of trout livers to be used as a source of frozen tissue. No differences could be detected between the isozyme patterns in the fresh or the frozen liver tissue, so

the frozen tissue was used preferentially.

Acclimation experiments were carried out on rainbow trout during early fall of 1968 and 1969. It has been observed that warm-acclimation

is extremely difficult in late fall or winter (Baldwin & Hochachka, 1970).

Two non-circulating 150 gal stainless-steel tanks were used to acclimate approximately 12 fish at a time under a constant 14 hr light- 10 hr dark photo regime. One tank was maintained at 18°C (+ 2°C) and the other at

2°C (+ 0.5°C). Approximately one-third of the water in each tank was exchanged daily, and feed was given four-times weekly. The period of acclimation was at least four weeks in length. These temperatures are consistent with fluctuations seen in many Western Canadian waters.

All individuals of brook- (Salvelinus frontinalis), lake- (Salmo n amaycush), and splake trout used in these experiments were kindly donated by Dr. F.E.J. Fry, University of Toronoto, Canada, to whom I am greatly indebted. 14

Acclimation of the brook-, lake- and splake trout was carried out for at least one month at the Maple Laboratories, Toronoto, Ontario, by a number of Dr. Fry's co-workers. Large 100 gal concrete tanks supplied by a constant temperature water source were used in all acclimation studies. These tanks were maintained on a 12 hr light- 12 hr dark photo regime no matter what time of year the fish were acclimated. The fish were all approximately five inches long and two years of age, and while acclimating we're fed fish pellets, daily. At least fifty fish could be acclimated in each tank.

Preparation of NADP-IDH.

All rainbow trout tissues were homogenized in four volumes of ice cold buffer consisting of 100 mM tris, 2 mM EDTA, 2 mM glutathione

(reduced), 0.25 M sucrose, titrated to pH 7.0 with 1 M HC1, with an

Omni-Mixer (Sorvall) at medium speeds for short time intervals. This

preparation was centrifiiged at 40,000g for 30 min (Sorvall, RC-2B) and

further at 105,000g for 1 hr (Spinco, Preparative Ultracentrifuge,

Model L). All further preparative steps were carried out in an ice bath.

The enzyme was purified by (NH^^SO^ fractionation between 35 and 60%

saturation. The final extract was resuspended in a minimal quantity of

the above buffer, minus sucrose, and dialyzed against the same buffer

f or two hours to eliminate salts (this procedure tends to reduce the

enzyme activity, so it is not recommended to increase the dialysis period).

This enzyme preparation is stable for at least one month when stored at

-30°C.

Preparation of livers from brook, lake and splake trout depended

upon the time of year collected. All winter animals were sacrificed, livers 15 removed and quickly frozen as a group. Spring animals were initially frozen in liquid nitrogen and at a later time, the livers were individually removed. The liver enzyme from individual summer animals was prepared and subjected to electrophoresis before and after freezing of the tissues.

No differences in electrophoretic mobility could be detected between freshly prepared and frozen samples.

In all cases, the liver tissues were homogenized in four volumes of

50 mM tris containing 0.25 M sucrose and titrated to pH 7.2 with 1 M HC1, according to the method of de Duve, et al. (1955). Since no unique mitochondrial NADP-IDH pattern could be detected, the supernatant remaining after a 30 min centrifugation at 27,000g was used as a source of enzyme for electrophoresis.

Pig heart soluble NADP-IDH is available commercially from Sigma

Chemical Co. (St. Louis, Missouri, No. 1-2002) in a 50% glycerol solution.

This preparation was used as a source of the enzyme for column chromato• graphy and electrophoresis following proper dilution with 50mM tris-HCl buffer, pH 7.3.

Assay of NADP-IDH activity.

NADP-IDH catalyses the reversible oxidative decarboxylation of threo-

Dg isocitrate according to this reaction sequence which proceeds through an unstable intermediate (Ochoa, 1948) :

E + threo-D isocitrate + NADP+ E*(oxalosuccinate ) + NADPH + H+ s \ Me4"*" or Mn4"4" I

C( -ketoglutarate + C09 + E 16

The specific absorbancy of NAI(E)H at 340 nm (E34Q) can be used as a means of following this reaction. All assays were carried out on either

a Unicam SP 800 or SP 1800 spectrophotometer (Pye Unicam, Ltd., Cambridge,

E ngland).

The basic reaction mixture contained 50 mM tris-HCl buffer, pH

8.0 (temperature adjusted according to Sigma Technical Bulletin No. 106B,

1967), 1 mM MgCl2» 0.15 mM NADP+, various DL-isocitrate concentrations,

and enzyme, added last in a total volume of 2.0 ml. All chemicals were

purchased from Sigma Chemical Co., St. Louis, Missouri. Ln all cases,

isocitrate concentrations are given as DL-isocitrate, even though only

5 0% of this is the enzymatically active threo-Ds form (Plaut, 1963).

Cuvette temperatures were accurately controlled by the use of a circulating water bath (Lauda Brinkman, K-2/R) coupled to the cuvette holder.

Estimates of protein content.

The protein content of all enzyme solutions was determined by the method of Lowry, Rosebrough, Farr & Randal (1951). All samples were

diluted with water to approximately 50 yugm/ml protein, and compared with

a standard curve determined from 0 to 100 }jgm/ml bovine serum albumen

p rotein.

Starch-gel electrophoresis.

Horizontal starch-gel electrophoresis was carried out according to

Smithies (1955) using a citrate-phosphate buffer system at pH 7.0 (tank

buffer was 9 mM citric acid and 90 mM dibasic sodium phosphate; gel

buffer is a 20-fold dilution of this tank buffer). Alternate buffer

systems were employed, but resolution was found to be the best with the citrate-phosphate system. Supernatant samples were applied to three pieces of 5 mm square Whatman No. 1 filter paper, and two thicknesses of this same paper were used as a bridge between the electrode tanks and the gel. Adequate separation of isozymes was obtained using a 13% starch-gel (Hydrolysed, Connaught Medical Research Laboratories, Toronto,

Ontairo) and electrophoresing at 200 V ( approximately 20mA) for 17 hr at

4°C.

The gels were stained for NADP-IDH activity using a modified histochemical stain developed by Hunter & Markert (1957). The staining solution contained 62 mgm DL-isocitrate, 15 mgm NADP+, 15 mgm nitro blue tetrazolium, approximately 5 ^ugm phenazine methosulfate, and 2.5 ml

of 20 mM MgCl2 solution in 50 ml of 0.1 M tris-HCl buffer, pH 7.0. All chemicals were obtained from Sigma Chemical Co. Staining was carried out on sliced gels in the dark at room temperature for 2 hr.

Numerous NADP-IDH isozyme patterns (zymograms) were scanned using a recording microdensitometer (Joyce, Loebl, Burlington, Mass.). This method can be used to quantify the distribution of isozymic bands. The area under each peak is determined and expressed as a per cent of the entire area.

Isoelectric focusing of NADP-IDH.

Electrofocusing of the rainbow trout liver enzyme was carried out according to the method of Haglund (1967) to estimate pi values for each liver isozyme. The best resolution was attained using a pH 3-10 gradient

(LKB Ampholine 8141), run at 300 V for 32 hr. Longer electrofocusing as well as higher voltages gave lower enzyme recovery even though the column was maintained at 5°C throughout the experiment. 18

Each 2 ml fraction was collected using a LKB ultrorac fraction collector and assayed for NADP-IDH activity by the standard method. The

isoelectric point of each isozyme peak was estimated from the pH profile.

DEAE-cellulose chromatography.

The enzyme was prepared as above, except that the homogenization buffer was 1 mM tris + 1 mM EDTA + 1 mM ^-mercaptoethanol + 0.25 M sucrose,

titrated to pH 7.5 with 1 M HC1. The 35-60% (NH4)2S04 precipitated fraction was resuspended in a minimal quantity of the above buffer

(minus sucrose) and dialyzed for 12 hr against eight liters of the same buffer. The dialyzed sample was applied to a column(2.0 x 25 cm) of

DEAE-cellulose (Whatman, Microgranular) previously equilibrated to 4°C with the same buffer. At least two column volumes of the initial buffer were necessary to wash out all non-adhering protein prior to elution of

the NADP--IDH isozymes, The enzyme was eluted with a linear gradient in which the tris concentration was increased to 0.3 M. Each 2 ml fraction was collected and assayed for NADP-IDH activity as above. The gradient was determined by conductivity measurements of eluted samples and compared to those of standard NaCl solutions.

Sucrose gradient centrifugation.

Sucrose gradient centrifugation was undertaken in order to compare the molecular weights of the NADP-IDH from rainbow trout and the purified pig heart enzyme, as well as to determine if the isozymes of the trout enzyme had identical weights. Samples were run using a Spinco model L preparative ultracentrifuge equipped with a SW 39 rotor. The 5 to 20% gradients were prepared with a dual chamber gradient maker in 100 mM tris 19 buffer containing 1 mM ^-mercaptoethanol and 2 mM EDTA, titrated with

1 M HC1 to pH 7.0. A 0.2 ml enzyme sample was layered onto the top of each gradient and run for ten hours at 35,000 RPM and 0°C. Drops were collected by gravity from a hole punched in the bottom of each tube and assayed for NADP-IDH activity using the standard assay. Gradient densities were calculated from refractometer readings.

Gel-filtration of NADP-IDH.

Sephadex G-100 (Pharmacia, Uppsala, Sweden) gel-filtration was carried out as another method to determine comparable size of the rainbow

trout liver isozymes. A 1 x 32 cm column was employed with a VQ of approximately 6.5 ml as determined by dye dilution using Dextran Blue 2000

(Pharmacia). The preswollen gel was packed under gravity, and equilibrated at k°C with 50 mM tris buffer with 2 mM ^-mercaptoethanol titrated to pH 7-2 with 1 M HC1. A 1 ml sample of enzyme was applied to the top of the column, and the activity was eluted with the same buffer. One ml fractions were collected with a LKB ultrorac fraction collector and assayed for NADP-IDH activity using the standard assay. CHAPTER III:

Effects of Thermal Acclimation on Multiple Forms of the Liver Soluble

NADP+-Linked Isocitrate Dehydrogenase in the Family Salmonidae 20

INTRODUCTION

The existence of a mitochondrial (M-IDH) and a supernatant (S-IDH)

NADP+-specific isocitrate dehydrogenae (IDH) (EC 1.1.1.42) has been demonstrated in mice strains (Henderson, 1965; Mintz & Baker, 1967), in Paramecium aurelia (Tait, 1970) and in smelt, carp and goldfish

(Quiroz-Gutierrez & Ohno, 1970). Multiple forms of this enzyme are also known in a number of bacteria (Reeves, Brehmeyer & Ajl, 1968;

Self & Weitzman, 1970). In all cases thus far examined, these multiple forms differ in kinetic, molecular and electrophoretic properties, although their functional significance has not been clarified.

In those organisms where the S-form IDH has been extensively studied, the enzyme appears as a dimer, with a molecular weight of approximately 75,000 (Illingworth & Tipton, 1970; Howard & Becker, 1970).

Genetic analysis carried out using mice strains (Mintz & Baker, 1967;

Henderson, 1965) and smelt, a diploid fish (Quiroz-Gutierrez & Ohno,

1970j, suggest that the enzyme is specified by a single gene locus.

Multiplicity of protein structure has been found to occur frequently in Salmonid fishes. This group of fishes are believed to be tetraploid, mainly from the finding by Ohno, Wolf & Atkin (1968) that these animals maintain about twice as much DNA/cell as found in most vertebrates. Enzymes such as malate dehydrogenase (Bailey, Cocks &

Wilson, 1969), lactate dehydrogenase (Massaro & Markert, 1968), enolase

(Tsuyuki & Wold, 1964), aldolase (Lebherz & Rutter, 1969), and creatine kinase (Eppenberger, Scholl & Ursprung, 1971) are known to occur as multiple enzyme systems, supporting this tetraploid hypothesis. Work 21

mentioned by Quiroz-Gutierrez & Ohno (1970), but not reported, and by

Wolf, Engel & Faust (1970) on Salmo irideus indicate that the gene loci

coding for the heart M-IDH has also been duplicated, although the liver

S-IDH apparently has not.

Carp and goldfish, also believed to be tetraploid fish, were

shown to be endowed with two separate gene loci for S-form IDH (Quiroz-

Gutierrez & Ohno, 1970). A part of the evidence in favour of this

hypothesis is that the predicted frequency of phenotypic expression of

the enzyme variant was in fact observed in a wild population of goldfish.

In this study, as in others of its type, the implicit assumption is

that extrinsic factors have no influence on the expression of these genes. Previous studies of rainbow trout S-IDH (Moon & Hochachka,

1971) and other isozyme systems in fish (Hochachka, 1965; Baldwin &

Hochachka, 1970; Hochachka & Somero, 1971) indicate that the expression of different enzyme variants can depend upon such environmental parameters as season and thermal acclimation. To gain further insight into this

problem, four members of the Salmonidae family are examined for multiple

S-form IDH isozymes. Based upon the assumption that the liver S-form

IDH is a dimer, the IDH of lake trout (Salmo namaycush) is assembled

from a single subunit type whose expression is temperature independent;

the IDHs of brook trout (Salvelinus frontinalis) are formed by dimer

assembly from subunit types whose distribution is temperature dependent;

assembly of the isozymes in splake trout requires at least two subunit

types, the expression of which is complexly affected by temperature. 22

RESULTS

Winter Lake, Splake and Brook Trout S-form IDH.

Pooled samples of winter brook trout yield a pattern of five anodally

moving isozymes (Pig. Ill, ID). This pattern is consistent with the idea

of heterozygosity at one of the two gene loci as suggested by Quiroz-

Gutierrez & Ohno (1970) for goldfish. Since individuals were not assayed,

i t is impossible to determine whether or not polymorphic .specimens

are seen in this group. There are no detectable effects of- acclimation

seen in these winter brook trout.

Winter splake trout demonstrate an exceedingly complex acclimation

pattern. Fig. Ill, 1 shows that there has been a shift in the isozyme

pattern towards those forms with lower mobility at the extreme acclimation

temperatures (4° and 17°C, panels B and E). The three bands seen in

the 9°C trout correspond to the three fastest moving anodal bands in

the brook trout, and the 4°C bands show homologies with the three

slower migrating brook trout bands (see Fig. Ill, 1). The 17°C pattern

overlaps only its most anodal band with the brook trout pattern. This is

unexpected since the splake is a brook trout-lake trout hybrid cross,

and neither of the parental species expresses this slowest moving band

(or subunit). Whether this 17°C isozyme pattern is due to genetic or i physical characteristics of the enzyme is unknown, although numerous

electrophoresis runs over long time periods show the same separation.

The interesting feature of the lake trout S-form IDH from liver,

is that genome duplication, if it did occur, did not result in the

production of different electrophoretic forms as in the other salmonids

studied here (Fig. Ill, 1A). Studies by Wolf, Engel & Faust (1970) suggest 2 3

Fig. Ill, 1. Composite starch-gel electrophoretogram of winter acclimated lake, splake and brook trout liver S-form IDH. Electrophoresis conditions: 17 hr at 15 mA and 200 V, in phosphate-citrate buffer, pH 7.0. Gel temperature constant at 5-6°C. Anode at bottom of all electrophoretograms, Origin marked by 0. A. 9°C-acclimated lake trout; B. 4°C-acclimated splake trout; C. 9°C-acclimated splake trout; D. 9°C-acclimated brook trout; and E. 17°C-acclimated splake trout. o

A B C D E 24

that the liver S-form IDH locus of Salmo irideus similarly has not undergone duplication. Lake and rainbow trout liver LDHs exhibit identical

single band patterns (Hochachka, 1966). Again, as in brook trout, there is apparently no acclimation effect, with the isozyme pattern being the same at 4° and 17°C.

In summary, gene duplication resulting in variant enzyme forms has apparently occurred in brook trout liver S-form IDH, but not in . lake trout. Since splake trout is a hybrid, this characteristic pattern seen in the brook trout is carried over with the addition of a non• homologous subunit type in the case of the 17°C splake trout. Only in

the case of splake trout liver S-form IDH is there a shift in pattern with temperature acclimation, although the genetic basis for this is not understood. It may be, however, that the 17°C pattern seen in splake trout represents a mutated locus derived from the brook trout, which is not expressed except under proper conditions of season and

temperature.

Spring Lake, Splake and Brook Trout S-form IDH.

Lake trout acclimated to 4° and 17°C in the spring exhibit identical

electrophoretic patterns, and, in fact, these patterns are unchanged

from the winter group. Again, only a single enzyme form is present.

Splake trout acclimated under identical conditions also exhibit

similar spring patterns, but these are unlike those seen in winter fish.

The spring pattern is identical to that seen at 9°C during the winter.

The spring brook trout acclimated to 4° and 17°C demonstrated at

least four different electrophoretic patterns (Fig. Ill, 2) based on

individual samples. Statistical interpretation of these data was 25

Fig. Ill, 2. Resolution of spring acclimated brook trout liver S-form

IDH on starch-gel electrophoresis. Electrophoretic conditions as in

Fig. Ill, 1. Examples of four electrophoretic patterns are: Br-1 in column C, D and G; Br-2a in A and F; Br-3 in E and H; and Br-4 in B.

26 difficult because of the small sample size (only eight fish in each acclimation group). The dominant pattern remained that seen in winter brook trout £five isozymes) and is called Br-1; this group consisted of 6 of the 16 specimens used. The second most abundant pattern (Br-2,

Fig. Ill, 2) made up five of the 16 specimens. Br-3 consisted of 3 of 16, and was a variant of pattern Br-1 without the expression of the most anodal homodimer. Br-4 (occurring only 2 of 16 times) consists of a six-isozyme pattern, also similar to Br-1.

Fig. Ill, 3 is a diagrammatic representation of these four distinct banding patterns, using a nomenclature similar to that of Quiroz-

Gutierrez & Ohno (1970). It can be seen that the patterns are consistent with the existence of at least four homodimers (AA, aa, BB and bb) plus the numerous hybrid forms of each homodimer.

Summer Splake and Brook Trout S-form IDH.

Since no unusual isozyme patterns were seen in lake trout during the winter and spring testing, no work was carried out on these trout

during the summer. Also, lake trout are extremely hard to warm acclimate

(Peter Ihssen, personal communication) apparently due to their long ancestral history at the bottom of lakes where temperatures are very low. Also, no differences in isozymic patterns were detected in summer splake as compared to those run in the spring, so they will not be dealt with further.

Again, individual brook trout showed variant S-form IDH patterns, although the number of variants was reduced from spring. Unlike the

spring experiments, a large sample size was collected allowing for a statistical treatment of the data. Fig. Ill, 4 represents a diagrammatical 27

Fig. Ill, 3. Diagrammatical representation of spring acclimated brook trout liver S-form isozyme patterns seen in Fig. Ill, 2. < o jQ < < o < l_ m

< < o ro i

GO a CD CD o CM I CD

Fig. Ill, 4. A diagrammatical representation of S-form IDH isozymes from summer acclimated brook trout liver observed after starch-gel electro• phoresis. Only relative positions of the bands are shown in the three distinct patterns. representation of the three patterns, with the relative staining intensities indicated. In Table 1, the frequency distribution of these three variant patterns is given, correlating both temperature acclimation in one case, and sex in the other.

According to X^ values, the frequency distribution between the

4° and 17°C acclimated groups is significantly different, with the 17°C group having more of the Br-2 pattern and the 4°C group more of the

Br-1 pattern. There is little difference between the sexes as far as the frequency of bands is concerned. 30

Table III, 1. Frequency distribution of the three S-form IDH isozyme patterns from summer acclimated brook'trout liver, with Chi square c alculations.

Br-1 Br-2 Br-3 Total X2-Value

Acclimation

temperature

4°C 13 5 5 23 28.9

17°C 14 24 5 43

Sex

Male 16 11 3 30 6.5 Female 11 16 7 34

= 10,6 0.005(2) 31

DISCUSSION

The results of these electrophoretic experiments tend to support the observations of Ohno e_t al_. (1968) as to the tetraploid nature of salmonid fishes. As can be seen in the composite electro• phoretogram (Fig. Ill, 5a) and the accompanying diagrammatical represen• tation of the typical four trout patterns (Fig. Ill, 5b), all results can be explained if two separate gene loci for S-form IDH are maintained in the salmonid genome (Quiroz-Gutierrez & Ohno, 1970) as in heart

M-form IDH (Wolf, Engel & Faust, 1970). The splake and rainbow trout liver S-form IDH show a two-subunit pattern, whereas four subunits in the brook trout are not unusual (spring experiment, Fig. Ill, 2).

The lake trout is unique in producing only a single subunit type, a finding similar to that of Wolf, Engel &. Faust (1970) for liver S-form

IDH in Salmo irideus. The genetic model of Quiroz-Gutierrez & Ohno (1970) is compatible with the typical brook pattern seen in Fig. Ill, 5a,b,

i.e., two gene loci, which are heterozygous in one or the other allele.

However, such a model cannot explain the results seen in the spring

experiment, nor for that matter, in the 17°C-acclimated summer brook

t rout.

Tentatively, we suggest that the brook trout liver S-form IDH

consists of four basic subunits (although not all fish will express

them all at any one time), whose expression is induced by temperature

change over long time periods (i.e., acclimation). This would explain

the differences seen in the frequency distribution data given in Table

III, 1 between the 4° and 17°C-acclimated brook trout. Such a shift

in isozyme pattern has been seen in rainbow trout liver S-form IDH Fig. Ill, 5a. Composite electrophoretogram of liver S-form IDH from the

family Salmonidae on starch-gel. Electrophoretic conditions as in Fig.

Ill, 1. Lake trout, A and H. Typical splake trout 2°C pattern, B.

Equal amounts of winter acclima ted 9° and 17°C splake trout, C. 2oc- "

acclimated rainbow trout, D. 17°C-acclimated rainbow trout, E. Dominant

brook trout patterns, F and G.

b. Diagrammatical representation of possible liver S-form IDH subunit

types from the family Salmonidae observed from various starch-gel

electrophoretograms. A BCD E F G H

AA

A'A

Ba • m

Brook Splake Lake Rainbow 33

(Moon & Hochachka, 1971) and brain acetylcholinesterase (Baldwin &

Hochachka, 1970), although these systems have fewer isozymic forms.

Splake trout, a brook trout-lake trout hybrid, is complex only in the winter acclimated animals (Fig. Ill, 1). In all other seasons, no qualitative acclimation phenomenon is apparent, and the isozyme system can be defined as a simple two homodimer system. The winter pattern is unusual, in that it requires the expression of another subunit not present in the liver IDH from either parental stock. Breeding exper• iments are necessary to determine what possibly controls the expression of this slowest migrating subunit in the splake trout liver. As demonstrated in the winter experiments, these subunits are present, but their expression appears to be tightly regulated.

Ecologically the lake trout is unique among the salmonid fishes s tudied in being adapted to the benthic environment which in deep lakes often displays near constancy of temperature, salinity, pressure, light, etc. And, indeed, of the species examined in this study, it is known to be so stenothermal that it is difficult to warm acclimate.

The occurrence of a single S-form IDH in this species is consistent with the idea that organisms which have evolved in stable environments show little enzyme heterogeneity. In contrast, in species such as the rainbow and brook trout, both of which are distinctly eurythermal, selection may favour enzyme heterogeneity in order to maintain greater physiological flexibility. CHAPTER IV:

Temperature and Enzyme Activity in Poikilotherms: Isocitrate Dehydrogenases

in the Rainbow Trout Liver 34

INTRODUCTION

The relative roles of the NAD-linked isocitrate dehydrogenase

(NAD-IDH, EC 1.1.1.41) and the NADP-1inked IDH (NADP-IDH, EC 1.1.1.42)

in the metabolism of most organisms are unknown. Largely on the basis of

tissue and intracellular distribution, Goebell & Klingenberg (1963;

1964) proposed that only the NAD-IDH was operative in the Krebs cycle,

and this proposal seems to be supported by recent studies (see, for

example, Nicholls & Garland, 1969). However, in a comprehensive survey

of vertebrate muscle types, Crabtree & Newsholme (1970) point out that

NADP-IDH activities exceed NAD-IDH activities by some 10-100 times in mammalian tissues and by up to 300 times in fishes. During initial

phases of this study, I confirmed the observations of Crabtree & Newsholme

(1970) . I therefore initiated detailed studies of the kinetic properties

of the NADP-IDHs in fish tissues in order to gain some further insight

into the possible functional significance of these enzymes in the metabolism of vertebrate tissues.

In mammals, NADP-IDH occurs in multimolecular forms with electro-

phoretically distinct supernatant and mitochondrial isozymes (Henderson,

1965) . Recently, the supernatant NADP-IDH has also been reported to occur

in isozymic forms in bacteria (Reeves, Brehmeyer & Ajl, 1968; Self &

Weitzman, 1970), in the carp and in the trout, both of which are thought

to be tetraploid (Quiroz-Gutierrez & Ohno, 1970; Wolf, Engel & Faust,

1970) .

Previous work from this laboratory (see Hochachka & Somero, 1971)

indicate that isozymic changes may be correlated with changes in acclimat- 35 ization temperature and that these changes may be adaptive for the survival of the fish. For this reason, and because NADPH produced by

IDH may be utilized during the increased lipogenesis that occurs in fishes at low temperatures (Knipprath & Mead, 1968; Dean, 1969), I was particularly interested in the relationship of this enzyme activity to the acclimatization state of the organism. The results are consistent with the thermal induction of NADP-IDH variants which are well suited for function at their respective acclimatization temperatures. A preliminary report of this work was presented recently (Moon, 1970). RESULTS AND DISCUSSION

NAD- versus NADP-IDH.

The NAD-IDH does not appear to be present in rainbow trout tissues.

Enzyme activity could not be detected using the methods of isolation and assay described by Chen & Plaut (1963), nor after attempts at

stabilization of the enzyme with substrate, cofactor, cations (Mg"*"*" or

Mn^) or a known positive modulator (AMP or ADP) . Glutathione and

-mercaptoethanol were also without effect. Various tissues (liver, gill, brain, heart and skeletal muscle) known to differ in oxidative potential were examined. In no case was there any significant NAD-IDH activity measurable, although the activity of the NADP-IDH enzyme was always relatively high (estimates of NADP-IDH activities are given in

Table IV, 1). These results are in close agreement with those of

Crabtree & Newsholme (1970), and suggest either (1) that the NAD-IDH does not occur in fish tissues, or (2) that it is highly unstable and

its activity in homogenized tissue preparations is no measure of its activity in. vivo.

Isozymes of liver NADP-IDH.

In preparations of rainbow trout liver, three bands of NADP-IDH activity are typically stained in both warm-(17°C) and cold-(2°C) acclimated trout, but the distribution of these bands is somewhat different.

Estimates made from densitometer traces (see Fig. IV, 1) indicate a

5-fold increase in the relative amount of the low mobility isozyme in the cold-acclimated trout, whereas the warm pattern shows approximately 37

Table IV, 1, Tissue-specific NADP-IDH activities of high speed

(35,000g) supernatant from rainbow trout. Isolation and assay as in

Materials and Methods. Temperature of assay: 25°C.

Tissue limol NADPH/min/g wet wt.

brain 2.86

liver 4.76

gill 1.58

heart 90.50

skeletal muscle 0.30 38

Fig. IV, 1. Starch-gel electrophoretograms of the cold- and warm- enzyme variants with the corresponding densitometer traces. Electro• phoresis run at 4°C for 17 hr at 200 V and 20 mA using a citrate/ phosphate buffer system, pH 7.0.

39

9 0% of all activity in the fastest moving band. If a system of two sub- units, aggregating into three types of dimers is assumed, the data are consistent with two gene loci coding for S-form IDH, a finding in agree• ment with that proposed for goldfish (Quiroz-Gutierrez & Ohno, 1970) and trout heart M-form IDH (Wolf, Engel & Faust, 1970).

The rainbow trout liver NADP-IDH is also interesting in that the homodimers appear to be stained more intensely than does the heterodimer

(see Fig. IV, 1). This suggests either (1) that each band represents more than a single isozyme; or (2) that assembly of polypeptide subunits i s non-random.

Unlike mammalian NADP-IDH (Henderson, 1965), rainbow trout liver does not display an unique mitochondrial enzyme form. Similar results have been reported by Wolf, Engel & Faust (1970) for liver from Salmo irideus, but not heart where distinct mitochondrial forms are present.

When the enzyme is isolated from carefully prepared trout liver mito• chondria, the Km for substrate is similar to values obtained for super• natant IDH (Moon, unpublished observations). Similar results are available for beef heart NADP-IDH (Higashi, Maruyama, Otani & Sakamoto,

1965). This suggests contamination of the mitochondrial pellet with the S-form IDH.

The results of electrofocusing are consistent with the existence of multiple forms of liver IDH (Fig. IV, 2). However, the recovery of

enzyme activity is quite low, so that only shoulders are visible at the pI of the minor enzyme components. The results of DEAE-cellulose chromatography (Fig. IV, 3) are unequivocal in demonstrating three isozymic forms in liver of cold-adapted fish. As in starch-gel electrophoresis, 4 0T

Fig. IV, 2. Concurrent electrofocus separations on a pH 3-10

Ampholyte gradient of the cold (upper panel) and warm (lower panel) enzyme variants. Run at 5°C for 32 hr at 300 V. Fractions were assayed at pH 8.0, 50mM tris-HCl buffer, 1 mM DL-isocitrate, 1 mM

MgCl2, and 0.15 mM NADP+.

41

Fig. IV, 3. DEAE-cellulose chromatography of the cold-enzyme variant from rainbow trout liver. NADP-IDH isozymes were eluted from a 2.0 x

25 cm column with increasing salt concentrations (from 1.0 to 300 mM tris). Initial buffer consisted of 1.0 mM tris +1.0 mM EDTA + 1.0 mM

&-mercaptoethanol, titrated to pH 7.1 (at 5°C) with 1 M HC1. i4.0

3.0

E A 2.05 o o

1.0

ih 120 130 140 150 160 170 180 190 200 210 Fraction Number 42 the heterodimer (middle peak in Fig. IV, 3) occurs in reduced activity.

The major peak in Fig. IV, 3 corresponds to the fast-moving band in Fig.

IV, 1. The existence of these isozymes of liver NADP-IDH led to a

kinetic study of the cold- and warm-enzyme forms.

Effects of temperature on Vm.

One method of increasing the catalytic efficiency of enzymes

which work at low environmental temperatures is to decrease the energy

of activation necessary for the formation of the enzyme-substrate

complex. Such an idea has been suggested by Vroman & Brown (1963),

although in enzymes examined in this laboratory no general relationship

between environmental temperature and Ea has been confirmed (Hochachka

& Somero, 1971). Arrhenius values (Ea) of 18kcal/mol for both the warm-

and cold-enzyme variants were determined from a plot of log Vopt vs

1/T (Fig. IV, 4).

Effects of temperature on Km.

Fig. IV, 5 shows substrate saturation curves of DL-isocitrate at

different assay temperatures and their Lineweaver-Burk treatments. In the

case of the warm-enzyme curves, the activity at 17°C (the temperature at

which the fish were acclimatized) is as high as the activity at 25°C

when examined at low substrate concentrations (i.e., less than 10 yuM DL-

isocitrate). A similar finding is suggested from Fig. IV, 6 which plots

NADP+ saturation curves at different temperatures. Again, activity is as

high at 15° and 17°C as it is at 25°C. This has been found in a number

of enzymes when substrate saturation curves are plotted at different temp•

eratures, including phosphofructokinase from goldfish (Freed, 1969), A3

Fig. IV, 4. Arrhenius plots of trout liver NADP-IDH from cold- (closed circles) and warm- (open circles) acclimated animals. Assay carried out at pH 8.0, 50 mM tris-HCl buffer, in the presence of saturating substrate concentrations. Ea determined from the slope of the line.

44

Fig. IV, 5. DL-isocitrate saturation curves and Lineweaver-Burk plots at different assay temperatures for the cold- (lower panel) and warm-

(upper panel) acclimated rainbow trout NADP-IDH. Assay carried out at pH 8.0, 50 mM tris-HCl buffer (temperature adjusted), 0.15 mM NADP+, and 1.0 mM MgCl~. Equal enzyme quantities added in all cases.

45

Fig. IV, 6. NADP saturation curves and Lineweaver-Burk plots at different assay temperatures for the cold- (lower panel) and warm-

(upper panel) acclimated rainbow trout NADP-IDH. Assay carried out at pH 8.0, 50 mM tris-HCl buffer (temperature adjusted), 1.0 mM DL- isocitrate and 1.0 mM MgC^. Equal enzyme quantities added in all cases.

46

fructose diphosphatase from lungfish liver (Behrisch & Hochachka, 1969) and trout brain acetylcholinesterase (Baldwin & Hochachka, 1970).

When the Michaelis constant, Km, derived from Lineweaver-Burk plots (Fig. IV, 5 and 6), is plotted as a function of assay temperature, complex curves are found (Fig. IV, 7). At the upper biological temper• ature extreme, Km varies directly with temperature, approaching a minimal

Km (maximal apparent affinity) within the temperature range at which acclimatization occurs. This is particularly marked for the Km of EXD-

isocitrate (Fig. IV, 7), but the relationship is not so extreme in the case of the cofactor, NADP+ (see Fig. IV, 6). Similar complex Km vs

temperature curves have been seen for other enzymes from several different

tissues of aquatic organisms (Hochachka & Somero, 1971; Somero &

Hochachka, 1971).

The significance of this upper shift in Km at high biological

temperatures has been previously discussed (Baldwin & Hochachka, 1970;

Hochachka & Somero, 1971). By increasing the apparent Km at these high

temperatures, the thermal effects on velocity are reduced, thereby

leaving the reaction relatively temperature-independent. The upswing

seen at the lower biological temperature range for the warm-enzyme (Fig.

IV, 7) is interesting and may be coupled with the high Ea value seen in

this region for this enzyme variant (Fig. IV, 4). Table IV, 2, showing

QIQ values for both enzymes, demonstrates that for the warm-variant between 5 and 10°C, Q-^Q increases with decreasing substrate (DL-isocitrate)

concentrations because Km increases and thermal energy decreases (Fig.

IV, 7). These two effects, high Km and high Q^Q (at physiological

concentrations of substrate) probably set a lower thermal limit for 47

Fig. IV, 7. Km of DL-isocitrate for the cold- (open squares) and warm-

(open circles) NADP-IDH enzyme variants from rainbow trout liver as a function of assay temperature. Km was estimated from Lineweaver-Burk plots in Fig. IV, 5.

Table IV, 2. Q^o values for the warm- and cold-enzyme variants at various DL-isocitrate concentrations.

QJQ; Temperature range

5-10°C 15-20°C

DL-isocitrate C-enzyme W-enzyme C-enzyme W-enzyme

1.0 mM 4.0 5.14 2.92 3.55

0.1 mM 4.0 4.0 2.65 3.47

0.075mM 3.44 4.0 2.34 3.62

0.05 mM 3.41 7.1 2.0 2.56

4>- 00

\ controlled catalytic function by this warm-enzyme variant.

In Fig. IV, 7 it is evident that the minimal Km values for the two enzyme preparations are similar. In consequence, the rate of catalysis by the cold-enzyme at 2°C (minimal Km for this form) will be much lower than the rate catalyzed by the warm-variant at 17°C (minimal Km for the warm-enzyme). -If acclimatization is to lead to significant rate compensation at any given substrate concentration (Hazel & Prosser,

1970), the Km of the cold-enzyme variant would have to be much lower than for the warm-form. Since this is not the case, we have considered the possible roles of cations, other metabolites, pH and enzyme concent• ration in affecting NADP-IDH activity during the thermal acclimatization p rocess.

Effect of temperature on the Michaelis constants of Mg"""*1 " and Mn

As in the case with NADP-IDH from other organisms (Plaut, 1963), the isozymes from trout liver display an absolute requirement for a

I | 4-(- divalent cation, which can be fulfilled by either Mg or Mn . It is

I | evident from Table IV, 3 that the Ka values for Mg are essentially

temperature independent. Calculated Ka values are in the range of

3.3-3.8 x 10"5M for the cold-variant and 2.3-2.6 x 10"5M for the warm- variant (Table IV, 3). Although the Ka of Mn has not been calculated for the warm-variant, that for the cold-variant is five times smaller than the Ka of Mg"*""*" (Table IV, 3). Since distribution and concentration of cations is thought to be adjusted during thermal acclimatization in fishes (Hickman, McNabb, Nelson, van Breeman & Comfort, 1964; Houston,

Madden & DeWilde, 1970), it is possible that appropriate adjustments in 50

Table IV, 3. Ka (cations) for the cold- and warm-enzyme variants at various assay temperatures.

Ka (cations)

Mg^ x 105M Mn44" x 105M

Temperature C-enzyme W-enzyme C-enzyme

5°C 3.3 2.5 6.7

10°C 3.5 2.6 6.4

15°C 3.3 2.5

20°C 3.8 2.6 6.4

25°C 3.3 2.3 " 6.5

30°C 6.4 51

Mg or Mn concentrations in liver during thermal acclimatization could lead to some compensation of NADP-IDH activity.

Effect of temperature on the pH optimum.

Fig. IV, 8 is a plot of the relative IDH activity of both enzyme preparations as a function of pH and temperature, and demonstrates typical pH profiles for NADP-IDH (Higashi, et al., 1965). In the case of trout liver IDH, both forms display broad pH optima in the pH range

7.0 to 9.0. An interesting feature of these curves is that as the temperature is decreased from 25° to 15°C, the pH optimum for the cold- enzyme is extended towards lower pH values. Work by Rahn (1965) and

Reeves & Wilson (1969) has shown that the intracellular and blood pH of fishes is increased as temperature decreases. For the cold-form of trout liver IDH, the pH characteristics may exaggrate the effects seen by Rahn and coworkers, thus increasing the catalytic rates at this low temperature. This effect, unlike the Km-temperature relationship, would thermally stabilize the reaction irrespective of substrate concentrations.

Effects of other metabolites.

Since a part of the argument favouring NAD-IDH function in the

Krebs cycle is based on its responsiveness to regulatory metabolites such as the adenylates (Nicholls & Garland, 1969), a systematic search for compounds which may serve a regulatory role in the case of trout liver

NADP-IDH was initiated. Table IV,4 gives a list of the various metabolites which were tested against trout liver NADP-IDH at two different concent• rations of substrate. The apparent inhibition of the enzyme by the Fig. IV, 8. Relative NADP-IDH activities plotted at two temperatures and various pH values for the cold- (open squares) and warm-(closed circles) enzyme variants from rainbow trout liver. Assay carried out with 50 mM tris-HCl buffer (temperature adjusted), 1.0 mM DL-isocitrate,

+ 0.15 mM NADP , 1.0 mM MgCl2 and equal enzyme quantities. Relative IDH Activity

-o Table IV, 4. The effects of various metabolites on the activity of

NADP-IDH from rainbow trout liver. Activity expressed as per cent c

control in absence of metabolite.

DL-isocitrate concentration Metabolite 1 mM 0.1 mM

glyoxylate—5 mM 100.0 96.0 1 mM 101.0 96.0 .5 mM 104.5 106.0

oxaloacetate—5 mM 100.0 92.5 1 mM 102.5 92.5 .5 mM 104.5 92.5

glyoxylate + oxaloacetate—1 mM(each) 42.0 0.0

citric acid—5 mM 100.0 100.0 1 mM 100.0 100.0 .5 mM 101.0 100.0

glutamic acid—5 mM 102.5 88.5 1 mM 100.0 96.0 .5 mM 102.5 106.0 o\ -ketoglutarate—5mM 55.5 0.0 1 mM 80.0 34.6 .5 mM 86.5 50.0

pyruvate—5 mM 91.0 92.5 1 mM 91.0 96.0 .5 mM 95.5 92.5

fructose diphosphate—5mM 95.5 77.0 1 mM 91.0 80.5 .5 mM 82.0 80.5

phosphoenol pyruvate—5 mM 66.7 57.5 1 mM 82.0 88.5 .5 mM 78.0 88.5

NAD+—5mM 95.5 115.0 1 mM 91.0 100.0 .5 mM 95.5 104.0 54

Table IV, 4. Continued

NADH— 0.25 mM 84.5 81.0 0.1 mM 91.0 100.0 0.05 mM 91.0 92.5

ATP—1 mM 94.5 62.5 .5 mM 96.5 94.0

ADP--5 mM 100.0 65.5 1 mM 101.0 94.0 .5 mM 96.0 100.0

AMP—1 mM 100.0 115.0 .5 mM 94.5 106.0 55 l | adenylates, particularly ATP, is due entirely to chelation of Mg ; at high Mg concentrations, no poitive or negative adenylate effects are seen, unlike the case in some bacteria (Hampton & Hanson, 1969; Parker

& Weitzman, 1970) and protozoa (Marr & Weber, 1969a,b).

However, at high Mg concentrations, physiological concentrations of ADP tend to decrease the Km of DL-isocitrate. Table IV, 5 suggests that at low ADP (0.4 mM and less), the Km of the cold-enzyme variant at 10°C is decreased up to three-fold, whereas higher concentrations are not as effective. Such an effect is known for the NAD-IDH from animal tissues (Chen & Plaut, 1963), but has not been reported for the NADP-IDH enzyme.

Also, Fig. IV, 9 demonstrates that for the cold-enzyme variant,

&• -ketoglutarate (°(-KGA) is a competitive inhibitor of NADP-IDH, increasing the Km of DL-isocitrate by a factor of 2.5 at 1 mM and 10 at

5 mM °( -KGA. These concentrations of -KGA appear to be in the physiol• ogical range (Williamson, Scholz & Browning, 1969). This inhibition of

NADP-IDH by ^ -KGA is of particular interest since in vivo, control of trout liver IDH could be coordinated with adenylate control of citrate synthase (Hochachka & Lewis, 1970). In vivo under most circumstances, high ^ -KGA concentrations would be synonymous with high ATP concentrations.

Hence, inhibition of the citrate synthase could be achieved by high ATP concentrations in concert with NADP-IDH inhibition by high C^-KGA concentrations.

Trout liver NADP-IDH is also inhibited by the concerted action of glyoxylate and oxaloacetate (Table IV, 4). This inhibition has also been observed in bacteria (Hampton & Hanson, 1969; Ozaki & Shiio, 1968) and by

Marr & Weber (1969c) in protozoa. The significance of this action is not known in higher animals. 56

Table IV, 5. The effects of ADP on the Km of DL-isocitrate in the cold-enzyme variant of rainbow trout liver NADP-IDH. Assayed at 10°C

w ith 0.15 mM NADP+, 1 mM MgCl2 and 50 mM tris-HCl buffer, pH 8.0.

ADP concentration Km(DL-isocitrate) x 10^M

0 4.8

0.1 mM 2.8

0.4 mM 1.57

0.8 mM 2.58 57

Fig. IV, 9. The effect of 0(-KGA on the Km of DL-isocitrate for the cold-NADP-IDH enzyme variant from rainbow trout liver assayed at 10°C.

Assay carried out at pH 8.0 (50 mM tris-HCl buffer, temperature adjusted)

+ with 0.15 mM NADP and 1.0 mM MgCl2- -20 -15 -10 -5 0 5 10 15 20

[DL-lsocit]xl05M 58

Effects of temperature acclimatization on enzyme activities.

One general kind of mechanism which has been proposed previously for achieving rate compensation during thermal acclimatization involves

changes in the steady state activities of enzymes. Thus it is known

that total 6-phosphogluconate dehydrogenase and aldolase activities

(Jankowsky, 1968), total phosphofructokinase activities (Freed, 1969) and total lactate dehydrogenase activities (Hochachka, 1965) appear to be increased during cold-acclimatization. Although we have not examined

this problem in the case of trout liver NADP-IDH in detail, it is

evident from Fig. IV, 5 that total NADP-IDH activities in liver prep•

arations from both cold- and warm-acclimatized trout seen to be about .

the same. If there is any acclimatization effect it is to slightly r educe the activity of the enzyme during cold-acclimatization. And

indeed as we have argued above, the two basic forms of this IDH appear

to be best suited for function at their respective thermal ranges.

A_ priori it is clear that the production of more of the warm-enzyme variant during cold-acclimatization (or vice versa) would not make

sense biologically. As in the case of trout muscle pyruvate kinase (Somero,

1969) and brain acetylcholinesterase (Baldwin & Hochachka, 1970), a more

functional solution to the problem of low temperature appears to be to

increase the relative steady state concentrations of an enzyme variant x

encountered by the organism in Nature. CHAPTER V:

Comparison of the Soluble NADP+-Linked Isocitrate Dehydrogenase of

Rainbow Trout Liver with the Purified Pig Heart Enzyme 59

INTRODUCTION

The changes in enzyme-substrate affinity which have been found to occur in the soluble rainbow trout liver NADP-IDH are in an adaptive direction (Ch. IV); i.e., by maintaining a constant Km over the thermal range the fish is likely to encounter in nature, control of catalysis by this enzyme is essentially temperature-independent, and by increasing

Km at higher temperatures, the effects of temperature increases are minimized'. Is this property a general feature of all soluble NADP-

IDH enzymes, or alternatively is it a particular feature of the rainbow trout enzyme?

It has been assumed (Hochachka & Somero, 1971) that properties unique to poikilothermic enzymes are responsible for this kinetic behaviour, since as with other phenotypic characters, enzymes are thought to be shaped by the evolutionary process (Atkinson, 1968). Few comparative studies are available with regards to the physical and chemical characteristics ofriomologous enzymes from poikilothermic and homeothermic tissues. When they are, temperature is usually considered of secondary importance (Assaf & Graves, 1969; Kaloustian & Kaplan, 1969).

Most mammalian tissues have been found to contain a soluble NADP-

IDH, but a highly purified source of this enzyme has only been prepared from pig heart (Plaut, 1963). In recent years, Colman and coworkers have reported the isolation and characterization of this purified enzyme

(Colman, 1968; Colman, Szeto & Cohen, 1970). It was of interest to compare the properties of this enzyme available on a commercial basis with the rainbow trout enzyme in order to begin to answer the question , of chemical similarity vs true adaptation. 60

The results suggest that even though the two enzymes are similar in molecular weights and response to inhibitors, they differ in electro• phoretic mobilities and response of affinity constants to temperature changes. It therefore appears that the changes seen in the control of the NADP-IDH enzyme from rainbow trout liver are an outcome of the selective process. 61

RESULTS

Characterization of NADP-IDH.

The electrophoretic mobilities of pig heart and rainbow trout

liver NADP-IDH differ significantly (Fig. V, 1). Trout liver NADP-IDH migrates anodally as three distinct isozymes, whereas the pig heart enzyme moves cathodally as a single staining band of activity. This difference in mobility can be attributed to the basic nature of the pig heart enzyme resulting from its amide content (Colman, 1968). Other workers have reported little or no electrophoretic mobility between pH

5,6 and 8.5 and an isoelectric point of 7.8 (Plaut, 1963). In contrast,

the pi of the trout liver enzymes are below 7.0 (Fig. IV, 2) and, there• fore, less basic in structure, The IDHs from two bacteria, B_. stearo-

t hemophilus (Howard & Becker, 1970) and A, vinelandii (Chung & Franzen,

1969), are similarily more acidic and migrate anodally during electrophoresis as compared to the pig heart enzyme.

In contrast to electrophoretic mobility, the differences in molecular weight are small. Based on a number of physical criteria, Colman and co• workers (1968; 1970) have determined the molecular weight of the pig heart enzyme as 58,000, and suggested it consists of a single polypeptide chain. However, the enzyme has has been found to aggregate (Kemper &

Kaplan, 1971), so its actual subunit structure is still in doubt.

Although Fig. V, 2 does not allow an accurate assesment, it is apparent

from both Sephadex G-100 gel filtration and ultracentrifugation, that the

trout liver enzyme is slightly larger than the 58,000 value reported for

the pig heart enzyme,

Illingworth & Tipton (1970) have reported the molecular weight of 62

Fig, V, 1. Starch-gel electrophoresis of the soluble NADP-IDH from

2°C-acclimated rainbow trout liver and pig heart. Run for 20 hr at

2 00V (approximately 20 mA, 5°C) using a citrate/phosphate buffer system, pH 7.0, Rainbow Trout Liver

Pig Heart origin 63

Fig, V, 2, Sucrose density ultracentrifugat ion (right panel) and

Sephadex G-100 gel filtration (left panel) of the soluble NADP-IDH enzyme from 2°C-acclimated rainbow trout liver (R.T. liver) and pig heart (P.- heart). Experimental protocol can be found in Materials and Methods. Ultracentrifugation

©t*- P Heart

1.345 1.350 1.355 Elution Volume (ml) Refractive Index 64 pig liver NADP-IDH as 76,000 and that the enzyme is a dimer (subunit molecular weight of 36,000). The molecular weight reported for bacterial

NADP-IDH is similar to this value or exceeds it (Chung & Franzen, 1969;

Barrera & Jurtshuk, 1970; Howard & Becker, 1970), suggesting that the trout enzyme may be in this same range. Electrophoretic data indicate that the trout enzyme is a dimer (see Ch. IV), which is consistent with its higher molecular weight.

Recently, Kemper & Kaplan (1971) have suggested that the pig heart

NADP-IDH aggregates, and the degree of aggregation is determined by substrate and cofactor binding. In an attempt to varify if the rainbow trout enzyme behaves in an identical manner, ultracentrifugation experiments in the presence of various substrate, cofactors, and/or inhibitors were initiated. Fig. V, 3 shows that if aggregation does occur, it results in a small mobility change under the present experimental conditions.

Electrophoretic experiments using a similar set of conditions again did not suggest changes in subunit structure.

Pig heart and trout liver NADP-IDH demonstrate different temper• ature dependencies. The Ea or activation energy for the pig heart enzyme is approximately 5 kcal/mole greater than that seen for the trout enzyme (Fig. V, 4). Catalytic efficiency can not be extrapolated from this data since additional thermodynamic parameters are involved

(Wr6blewski & Gregory, 1961). It is interesting to note that both enzymes obey the Arrhenius law (i.e., a linear response of log reaction rate with respect to temperature) well beyond their respective thermal range, again suggesting that enzyme activity at saturating substrate concentrations can not be correlated with habitat preferrence (Hochachka

& Somero, 1971). 6 5

Fig. V, 3. Sucrose density ultracentrifugation of the liver soluble

NADP-IDH from 2°C-acclimated rainbow trout, in the presence of substrates, cofactors, and/or inhibitors. The 5-20% sucrose gradients were run at 40,000 RPM for 8 hr (0°C) using a SW 50.1 rotor (Spinco

Model L preparative ultracentrifuge). Code for symbols: O no added components; © 1 x 10"5 M NADP+, 2 x 10"5 M DL-isocitrate, 1 x 10"5 M

5 5 5 MgCl2; • 2 x 10" M DL-isocitrate, 1 x 10~ M MgCl2; H 1 x 10" M

NADP+; A 2 x 10"5 M DLJisocitrate, 5 x 10~3 M OC -KGA. Both IDH

activity and fraction number are plotted as per cent of total.

66

Fig. V, 4. Arrhenius plots of the soluble NADP-IDH from 2°C-acclimated rainbow trout liver and pig heart. Assayed in the presence of 50 mM tris-HCl buffer (pH 8.0 for trout and pH 7.5 for pig heart NADP-IDH),

1.0 mM DL-isocitrate, 0.15 mM NADP+, and 1.0 mM MgCl2- Ea values calculated from the slope of the line. 30

1 1—' ' '—i—i— • 50 40 30 20 10 0 Temp. (°C)

310 330 350 370 I09/T CK"1) 67

Comparison of kinetic constants,

At the higher temperature range, trout liver and pig heart NADP-

IDH demonstrate similar Km(isocit) vs_ temperature responses; i.e., a

temperature decrease results in increased enzyme-isocitrate affinity.

However, unlike the trout enzyme, the pig heart enzyme does not level

off, but continues to increase enzyme-isocitrate affinities at all temp•

eratures assayed (Pig. V, 5). Q1Q values calculated from Fig, V,5

tend to decrease as DL-isocitrate concentrations decrease (Table V, 1) over all temperature ranges. This thermal stabilization of reaction rates is a necessary outcome of a decreasing Km with decreasing temperature response. At the high temperature interyal (40^-45°C,

Table V, 1), Q^g values are greatly reduced.

The value of the Km of DL-isocitrate calculated from Lineweaver-

Burk plots of Fig. V, 5, are approximately lO^fold higher than those previously reported for this enzyme (Moyle, 1956; Plaut, 1963).

Commercial sources of this enzyme are stabilized with glycerol which may account for this altered affinity constant.

Effects of inhibitors.

The most potent inhibitor of trout liver NADP-IDH as well as other vertebrate and bacterial IDHs, is the concerted inhibition by oxaloacetate and glyoxylate. Table V, 2 shows that even though separately oxaloacetate and glyoxylate do not show a significant inhibition, together at low

substrate concentrations the enzyme is completely inhibited. This inhibition

is in the concentration range reported for the trout liver enzyme (Table.

IV, 4) and other IDHs (Ozaki & Shiio, 1968; Shiio & Ozaki, 1968; Marr &

Weber, 1969c), Although the physiological importance of this inhibition

is not known in vertebrates which lack a glyoxylate cycle, it may be 68

Fig. V, 5. The effects of temperature on DL-isocitrate saturation curves for the purified NADP-IDH from pig heart. Assayed with 50 mM tris-HCl buffer (pH 7.5, temperature adjusted), 0.15 mM NADP+, and

1.0 mM MgCl2> The Km(DL-isocit) was calculated from Lineweaver-Burk plots. Specific Activity (ymols NADPHAnin/mg protein) ro & cn GO o F3

Km(S)xl08M o — ro CM 4^ CJI i i

8

O Table V, 1. Q^Q values for pig heart NADP-IDH at various DL-isocitrate concentrations. Values determined from substrate saturation curves in Fig. V, 5.

Q10: Temperature Range (°C)

[DL-isocitrate] mM 20-30° 30-35° 35-40° 40-45°

1.0 3.76 3.34 2.90 1.74

0.1 3.76 2.75 3.10 1,93

0.05 3.84 2.27 2.85 1.90

0.025 3.50 2.35 2.60 1.87

ON 70

Table V, 2. The effects of certain citric acid cycle intermediates

on the activity of pig heart NADP-IDH. Values are given as per cent

of the activity in the absence of the particular metabolite. The assay

+ contained 50 mM tris-HCl buffer, pH 7.5, 0.15 mM NADP , 1.0 mM MgCl2,

DL-isocitrate as specified, and was initiated by the addition of 0.08 mg

of the purified enzyme. The cuvette temperature was 25°C,

Metabolite 1.0 mM DL-isocitrate 0.04 mM DL-isocitrate

C{ -KGA--5.0 mM 71.0 10.3

1.0 mM 87.0 27.6

0.5 mM 100.0 44.8

oxaloacetate—1.0 mM 100.0 86.2

glyoxylate-—1.0 mM 100.0 100.0

(oxaloacetate +

glyoxylate)—1,0 mM 51.6 0.0 71 -t related to the sate of IDH aggregation (Kemper & Kaplan, 1971).

As was reported previously (Fig. IV, 9), IDH from trout liver is markedly inhibited by CA.-ketoglutarate (°(-KGA) . The pig heart enzyme

shows a similar behaviour in the presence of this metabolite (Table V,

2) even at low inhibitor concentrations. -KGA is a competitive

inhibitor with respect to DL-isocitrate as seen from Fig. V, 6, but

unlike trout liver IDH, the Km of DL-isocitrate increases as a linear

function of the-KGA concentration (insert, Fig, V, 6). In the case

of trout IDH (Fig. IV, 9), at low CX -KGA concentrations (0.5 mM and less),

the Km of DL-isocitrate remains constant or decreases slightly, and only

at high ^X ^KGA concentrations does the Km increase drastically. The

differences in Km response (12-fold for trout liver and 19-fold for pig

heart NADP-IDH.at 5 mM -KGA) would suggest that C>{-KGA is a more

effective inhibitor of the pig heart enzyme.

When the Km of DL-isocitrate is determined in the presence of 0,75 mM

Q>i -KGA, Km for both the trout liver and pig heart enzyme is temperature

independent (Table V, 3), This is a direct effect of the competitive

binding of C\-KGA to the DL-isocitrate site, since both Rm(isocit) and

Ki(0(-KGA) are complexly effected by temperature (Table V, 4).

The change in the Km of DL-isocitrate when compared in the presence

and absence of CX.-KGA, however, is different. Pig heart IDH shows a four-

fco six-fold increase in Km (compare values from Table V, 3 with those of

Fig. V, 5), whereas the trout liver shows a maximum two-fold increase

at low temperatures and nearly no change above 15°C (compare Table V, 3

with Fig. IV, 7). This large difference again suggests the NADP-IDH

from pig heart is more tightly controlled by (X -KGA than is the trout

enzyme. This inhibition of IDH by CX-KGA has not been reported for any 7 2

Fig. V, 6. The effects of

+ 5 0 mM tris-HCl buffer, pH 7.5, 0.15 mM NADP and 1.0 mM MgCl2.

Cuvette temperature was 25°C. Specific Activity (jimols NADPH/rnin/mg protein) 73

Table V, 3. The effects of temperature on the Km of DL-isocitrate in the presence of 0.75 mM^-KGA.- The Km values were calculated from Lineweaver-

+ Burk plots using 0.15 mM NADP , 1.0 mM MgCl2 and 50 mM tris-HCl buffer,

temperature adjusted (pH 8.0 for.2°C-trout liver and pH 7.5 for pig heart NADP-IDH).

Km(DL-isocitrate) x 103M

Temp. °C Pig heart IDH Trout liver IDH

1.0 6.5

5.0 6.5

. 10.0 ' 6.5

15,0 6.5

20.0 12.5 6.5

25.0 12.5 6.5

30,0 12.5

35.0 12.5

40,0 12.5

45.0 12.5 other NADP-IDH.

It has been suggested (Hochachka & Somero, 1971) that in order for inhibitors to be equally effective at all environmental temperatures, the

Ki of the inhibitor should be relatively temperature independent. Table

V, 4 is a list of Ki^-KGA) values -for both pig heart and trout liver

NADP-IDH. These values are in the range reported by Williamson, Scholz

& Browning (1969) forC^-KGA in mammalian tissues. The trend is towards an increased Ki at the extreme habitat temperature for both enzymes, with n-values (representing the number of sites at which the metabolite may bind (Atkinson, 1966) approaching 1.0 in all cases except at low temperatures for the trout liver enzyme. At temperatures below 5°C, o( -KGA is more effective as an inhibitor of trout liver NADP-IDH since both the n-value increases and Ki decreases. This would suggest in the absence of other control factors, the ^{-KGA and temperature decreases may tightly regulate catalysis by the trout enzyme under these conditions.

Since the Ki represents a binding constant (Dixon & Webb, 1964), the change in Ki with temperature should and does follow the same pattern as does the Km of DL-isocitrate. This in turn indicates that the Km of

DL-isocitrate supplies a good approximation of enzyme-substrate affinity. 75

Table V, 4. The effects of temperature on the Ki(c^-KGA) for NADP-IDH from pig heart and 2°C-acclimated trout liver. Ki values calculated from Hill plots in presence of 0.06 mM DL-isocitrate, 0.15 mM NADP+,

1.0 mM MgCl2 and 50 mM tris-HCl, temperature adjusted (pH 8.0 for 2oc- trout liver and pH 7.5 for pig heart NADP-IDH). N-values calculated from the slope of the Hill plot line and given in brackets following

Ki values.

Ki(<*.-KGA) x 104M

Temp. °C Pig heart IDH Trout liver IDH

3.0 6.8(2.1)

5,0 4.5(3.2)

10.0 6.0(1.3)

15,0 — 8.1(.77)

20.0 7.0(0.9)

25.0 6,4(1.0)

30,0 7.3(1.0)

35.0 7.3(1.0)

40.0 9.5(1.0)

45,0 9.2(1.0) 76

DISCUSSION

It is apparent that the control of catalysis by NADP-IDH from pig heart and trout liver differs. Values of activation energies and molecular weights, although not identical, do not differ significantly.

Also, the magnitude of inhibition byO^-KGA and oxaloacetate + glyoxylate are similar, suggesting these specific properties allow for IDH to function as an oxidoreductase of isocitrate and are to be expected wherever the enzyme is found, Similar findings have been demonstrated for certain pressure responses at saturating substrate concentrations for fructose-diphosphatase and pyruvate kinase (Hochachka, Schneider &

Moon, 1971; Moon, Mustafa & Hochachka, 1971), and probably are related to the steric placement of substrates and cofactors.

However, the response of the Km of DL-isocitrate to temperature differs as does the magnitude of the response in the presence of the inhibitor ^-KGA. Little work has been reported regarding the effects of temperature on the affinity constants of other vertebrate enzymes.

The Km of fructose-diphosphate for rabbit muscle aldolase exhibits a complex U-shaped curve when determined at temperatures between 5 and

5 0°C (Lehrer & Barker, 1970), a result which mimics the response of the binding constant. These results suggested to Lehrer & Barker that changes in the ionization and solvation of the substrate or inhibitor were responsible for the temperature effects, An equally valid interpretation might assume the same direct effects on the enzyme conformation.

Cowey (1967) found a 15-fold increase in the Km of D-glyceraldehyde-

3-phosphate for D-glyceraldehyde-3-phosphate dehydrogenase in rabbit muscle between 5 and 35°C. When these data were compared with the homologous lobster and cod fish enzymes, increases of less than 10-fold

were noted between the same temperatures. The Km of phosphoenol pyruvate

for rat muscle pyruvate kinase varies by less than a factor of two

between 10 and 40°C, unlike the response seen in poikilothermic pyruvate

kinases (Somero & Hochachka, 1968).

No generalized pattern can be derived from these mammalian studies.

However, for NADP-IDH, it is apparent that modulation of enzyme activities

are different, depending upon the organism examined. In the pig heart

enzyme, little or no selective pressure would be applied to the temper•

ature characteristics of the Km; instead, the thermal properties observed

are merely the expression of the chemistry of the reaction, For trout

liver NADP-IDH, the maintence of a constant Km for efficient modulation

at low substrate concentrations at normal biological temperatures, and

the reduction of Qio by increasing the Km at the high temperatures is

under strict selective pressures. These differences call for different

evolutionary strategies which may be an outcome of the increased potential

for gene expression in the tetraploid fish (Ch. VI) as well as their

thermal history.

Another aspect of catalytic control by NADP-IDH from pig heart and

trout liver is the difference in the Km of DL-isocitrate in the presence

of °(-KGA. Even though the Km iat.emperature independent in the presence

of the inhibitor (Table V, 3), the Ki of 0(-KGA is temperature dependent,

which is particularly marked at the low temperatures (Table V, 4). This

may suggest that ^ -KGA is an important controlling metabolite at this

environmental temperature (5°C), although other factors such as pH or

ion concentration may reverse the effectiveness of this inhibitor as has

been seen for AMP modulation of salmon fructose-diphosphatase (Behrisch, 78

1969).

The binding site for both the substrate, isocitrate and the inhibitor,

CX -KGA, are probably similar for both NADP-IDHs examined. Estimates of

activity at saturating substrate concentrations and inhibition studies

support this conclusion. However, the tertiary structures undoubtedly

differ. Opposite electrophoretic mobilities at pH 7.0, high n-values

for enzyme-C\-KGA affinities for the trout liver enzyme, and altered

temperature-Km responses, all substantiate this proposal. The trout

enzyme has evolved a protein structure which is more flexible possibly

as a result of alteration in the exposed amino acids, but probably not

by a change in the active site. This idea of alteration of the exterior

amino acids and conservation of active site groups has been found to be

a general rule when protein sequences are investigated (Hill, et al.,

1969). These alterations in primary sequence may be responsible for

functional adaptation of the enzyme to environmental parameters such as

temperature or hydrostatic pressure. CHAPTER VI;

The Effects of Temperature on the Kinetic Properties of Various Isozymes of the Liver Soluble NADP+-Linked Isocitrate Dehydrogenase from

Rainbow Trout 79

INTRODUCTION

Poikilothermic enzymes appear to be evolutionarily tailored through natural selection for thermally independent function. Kinetic studies on a number of such enzymes, particularly from the temperate rainbow trout, have demonstrated complex U-shaped curves when the apparent Km is plotted against temperature. This work has recently been reviewed by Hochachka & Somero (1971) and Somero & Hochachka (1971). At the upper biological thermal range, temperature decreases act analogously to positive modulators by increasing enzyme-substrate (ES) affinity.

Near the lower range, the opposite is true; temperature decreases result in ES-affinity decreases. Between these two extremes, which normally coincides with the animal's habitat temperature, ES-affinity remains constant, thus assuring efficient control of catalysis at substrate concentrations found intracellularly,

There are three general mechanisms available to the poikilotherm for achieving thermally-compensated enzyme catalysis: (1) a single enzyme suited for function over the entire biological range; (2) expression of isozymes with different thermal "preferences"; or (3) a single enzyme with an adjustment in the cellular milieu. A priori it appeared that each of these mechanisms are equally likely to occur, although in previous work (Hochachka & Somero, 1971; Ch. IV, this thesis) the latter have been emphasized. However, an adequate test of these alternatives for any given enzyme system is not yet available.

In a preliminary examination, a hatchery grown population of rainbow trout, Salmo gairdnerii, was found to contain a large amount of hetero• geneity at the gene locus(i) coding for the liver soluble NADP-IDH. It 80 was of interest to determine whether or not all individuals, irrespective of their NADP-IDH isozymal content, showed identical IDH kinetic pro• perties. In this way, it may be possible to suggest whether the presence of isozymes, which are particularly numerous in these tetraploid fish, may result in the unusual kinetic behaviour exhibited by this rainbow trout enzyme system as previously reported (Ch. IV).

The results suggest that although changes in the cellular milieu may alter the Km-temperature response, alteration in tissue isozyme content are of more probable significance. By increasing the number of the slowest migrating isozymic forms, the Km-temperature response tends to increase at high assay temperatures. Also, individuals which maintain a single staining band of the liver soluble NADP-IDH show temperature-independent ES-affinity, a response which would appear to be selectively disadvantageous in eurythermal species such as the r ainbow trout. 8.1

RESULTS

Changes in the cellular milieu.

One possible mechanism leading to an alteration in the apparent Km-

temperature response is a change in the cellular milieu. Fig. VI, 1,

suggests that by changing either the NADP+ or DL-isocitrate concentration

independent of one another, the Km for the alternate substrate at 15°C will be reduced to minimal values reported for the enzyme system from

pooled 2°C-acclimated rainbow trout livers (see Fig, IV, 7). In either

case, the response is again a complex function of temperature; at 15°C,

the Km vs substrate curve is U-shaped, whereas at 3°C the relationship-

is linear at all measured substrate concentrations, As has been

previously shown (Ch, IV), the fluctuations in the Km of NADP+ are not

as great as seen for the Km of DL-isocitrate, This result suggest that

a minimal value of Km for substrate can be obtained irrespective of

temperature, and this value may approximate the true intracellular

isocitrate and NADP* levels. Changes in metabolite concentrations are

known to occur during altered tissue metabolism (Williamson, Herczeg,

Coles & Cheung, 1967) and in some cases are known to result in modified

affinity constants (Gumaa, McLean & Greenbaum, 1971),

Theoretically, changing the enzyme concentration may likewise modify

the Km-temperature response. However, Fig. VI, 2 shows that although

Vmax is linear when plotted against enzyme-protein concentration, the Km

is invariant. This mechanism is probably of little importance since

changes in NADP-IDH activities during acclimation are minimal (Ch. IV). 82

Fig. VI, 1. The effects of changing substrate concentrations on the

Km of the alternate substrate for the liver soluble NADP-IDH from 2°C~- acclimated rainbow trout. In the solid symbols ( @ 15°C and H 3°C assay temperatures), the Km(DL-isocit) is determined at a number of

NADP+ concentrations; the open symbols ( O 15°C and G 3°C assay temperatures), the Km(NADP+) is determined at a number of DL-isocit• rate concentrations.

83

Fig. VI, 2, The effects of enzyme protein concentration upon the DL- isocitrate saturation curves of the liver soluble NADP-IDH from 2°C- acclimated rainbow trout. Assayed at 10°C with 50 mM tris-HCl, pH 8.0,

+ 0.15 mM NADP and 1.0 mM MgCl2. The insert is a plot of Vopt (•) and Km(DL-isocit) CO) at three enzyme protein concentrations. Velocity is given in absorbancy units. II A h oocP*o"° o jj—O 20 X f I I ' I ' I H I •1 -2 -3 -4 -5 " 10 M(DL-isocitrate) X 103 84

Kinetics of trout liver NADP-IDH isozymes.

Heterogeneity at the liver NADP-IDH locus(i) has been observed and

Fig. VI, 3 is a composite electrophoretogram of five out of the six isozymic patterns expressed in the trout hatchery population. The three-

(A2,B-2,C2) and one-band (A2) phenotypes occur in the greatest abundance during both late winter and late spring (Table VI, 1). It is assumed that trout NADP-IDH is similar to bacterial and vertebrates IDHs in being a dimer (e.g., Henderson, 1965; Howard & Becker, 1970; Quiroz-

Gutierrez & Ohno, 1970), so there are at least four possible subunit types. The absence of the expression of the B and C subunits as individuals would suggest, (1) the A subunit is the primitive or ancestral subunit, (2) epigenetic control of gene expression of these loci as has been identified for the LDH system (Whitt, 1970; Rosenberg,

1971), and/or (3) that only under the extremes of summer acclimatization will these subunits be expressed.

The three isozymic patterns used in the kinetic analysis are seen

in Fig. VI, 4 together with a composite plot of the Km of DL-isocitrate v s temperature relationship. It is apparent that the A2-NADP-IDH from

spring rainbow trout liver, whether as a crude high speed supernatant or a partially purified preparation (i.e., (NH^^SO^ percipitation

followed by dialysis), shows temperature independent kinetic behaviour.

Unlike the A2, as the number of B and C subunits increase, an upswing in

Km at the high temperatures is noted. Therefore, at 25°C, the A2,B2,C2

isozymes show a three-fold increase in Km over the value at 10°C. With

only the A2>AB,B2 isozyme system, the upswing is less pronounced and

begins at 15°C instead of 10°C. This observation makes evident that

for the NADP-IDH from trout liver, at least, the particular Km-temperature 85

Fig. VI, 3. Electrophoretogram of five out of the six observed liver soluble NADP-IDH phenotypes in the Sun Valley rainbow trout hatchery population. Run for 20 hr at 200 V (approximately 20 mA, 5°C) using

a citrate/phosphate buffer system, pH 7.0. 1. A2,AB,B2,BC,C2;

2. A2; 3. A2,B2,C2; 4. A'2,A'A,A2,AB,B2,BC,C2; 5. A2',A'A,A2.

86 Table VI, 1. The relative distribution (as per cent of total) of isozymic forms of the liver soluble NADP-IDH from a hatchery population of rainbow

trout (S. gairdnerii). Distributions determined from the presence or absence of stain on starch-gel electrophoresis. Winter fish were collected and assayed on March 24, 1971 and spring fish May 31, 1971,

Total number of fish in each group was fifty individuals.

Pattern Winter Spring

27.0 20.0

A2 ,B2,C2 41.7 42.5

A2,AB,B2 8,3 12.5

A2,AB,B2,BC,C2 6.3 15.0

1 A2 ,A'A,A2 10.4 2.5

A2',A'A,A2,AB,B2,BC,C2 6.3 7.5 87

Fig. VI, 4. The effects of the tissue isozymal content on the Km

(DL-isocit) vs temperature relationship for the liver soluble NADP-IDH of spring rainbow trout. Assayed in 50 mM tris-HCl buffer (pH 8.0,

temperature adjusted), 0.15 mMNADP+ and 1.0 mM MgCl2» The curve for

A2-NADP-IDH is given for both the crude (O) and partially purified

(®) preparations. Electrophoretic insert: (1) A2,B2,C25 (2) A2,AB,B2J

(3) partially purified A2; and (4) crude A2.

88 curve obtained is determined by the isozymal composition of the tissues, not by an adjustment in the cellular milieu. 89

DISCUSSION

The evolution of the family Salmonidae by the process of tetra- ploidization is well documented (see Ohno, Wolf & Atkin, 1968; Ohno,

1970) and multiple gene loci coding for a large number of isozymes of particular enzymes attest to this theory (e.g., Massaro & Markert, 1968;

Bailey, Cocks & Wilson, 1969; Engle, Hof & Wolf, 1970; Wolf, Engel &

Faust, 1970). Duplication of the NADP-IDH locus in salmonids is complex, since different tissues do not necessarily show the same rate of drifting apart of the copied loci resulting in the establishment of a new gene locus in the population (Wolf, Engel & Faust, 1970).

S almo gairdnerii liver soluble NADP-IDH does not exhibit diploidization s ince a single activity band for this enzyme is seen in some members of a hatchery population (Fig. VI, 1). It is interesting to note, however, that nearly three-quarters of the population maintain some polymorphism at this gene locus, suggesting a selective advantage may be conferred upon these individuals.

Polymorphic enzyme systems have been investigated in a number of natural populations, and the results are as numerous as the enzymes themselves. Heterozygote advantage is maintained by temperature at the serum esterase locus in the fresh water fish Catastomus clarkii (Koehn,

1 969), but the kinetic differences between variant heart LDHs are minimal in geographically separated populations of the frog Rana pipiens, all of which maintain specific temperature preferences.(Levy & Salthe,

1971) . It is difficult in this case to define the advantage of maintain• ing polymorphic liver soluble NADP-IDH. However, it is possible that the individuals containing only a single activity band, and therefore, 90 presumably homozygous at the IDH locus, can not tolerate extreme temper• ature changes, and certainly temperature compensation by Km modulation is not possible.

Table VI, 2 shows that the Q1Q between 10 and 20°C for the A2 and

A2>B2,C2 isozymal systems differ. Since the Km is invariant, the temper• ature coeficient is increased above that found for the A2,B2>C2 system near Km concentrations of substrates. As substrate concentrations approach in vivo conditions, QIQ values for A2,B2,C2-NADP-IDH would further decrease, resulting in an even greater difference between the two enzyme variants. Thus, those individuals which are heterozygous at this locus(i) may be conferred a selective advantage over those that are homozygous. A similar argument has been made at the low temperature range, where both Km and thermal energies are changing, resulting in extremely high Q-^Q values (Hochachka & Somero, 1971). It has been reported (Ch. IV) that in cold-acclimation, the amount of the slowest migrating liver soluble NADP-IDH isozymes increase, again indicating the importance of the temperature-isozyme relationship.

Lake trout populations have been observed to maintain a single staining activity band of liver soluble NADP-IDH (Ch. Ill) and LDH

(Hochachka, 1966). This fish is known to be stenothermal (Peter Ihssen, personal communication), living at the bottom of deep lakes where environmental fluctuations are minimal. The genotype showing this identical single activity band for liver NADP-IDH in rainbow trout may have resided at sometime in their previous history in a similar constant environment, and processes of natural selection may not have had the necessary time to eliminate this phenotypic pattern from the hatchery population where it appears to be at a disadvantage. Table VI, 2. Q1Q values between 10 and 20°C for the A2 and A2,B2,C, liver soluble NADP-IDH isozyme systems from spring rainbow trout at various DL-isocitrate concentrations. Assays carried out with 1 mM

+ MgCl2, 0.15 mM NADP and 50 mM tris-HCl buffer, titrated to pH 8.0 at the respective temperatures.

Ql0 (10-20°C)

[DL-isocitrate] mM A2 A2,B2,C2

0.1 3.6 3.8

0.05 3.4 3.3

0.025 4.2 2.8 Alterations in substrate or cofactor concentrations may also be responsible for the complex kinetic properties of this enzyme. From the results reported here, it is apparent that at 15°C, levels of NADP+ can determine the Km of DL-isocitrate; the minimal Km value isfdentical to the lowest value at 3°C. By altering the NADP+ concentration at 3°C, however, no change in Km occurs, suggesting that a specific enzyme may have a characteristic substrate affinity constant which allows for maximal enzyme effectiveness. Temperature and the presence of non- physiological substrate concentrations may alter the binding of substrate to enzyme by putting the enzyme into unfavorable conformation, thereby reducing its efficiency. These metastable enzyme conformations may be of importance to the enzyme in vivo (Nickerson & Day, 1969; Ibsen,

Schiller & Haas, 1971; Ikai & Tanford, 1971).

The importance of these experiments is that the complex kinetic behaviour of certain poikilothermic enzymes need not be specifically related to changes in the primary structure of the enzyme. Instead, alterations in the genetic make-up and the substrate/cofactor concent• rations must be considered as possible mechanisms. These parameters are known to change in natural populations and may be used as a sensitive measure of kinetic changes in enzyme function. CHAPTER VII:

Summating Remarks 93

These studies on rainbow trout liver NADP-IDH have suggested a number of conclusions concerning both the control of isocitrate oxidation and possible mechanisms of enzyme adaptation to fluctuations in environmental temperatures. To conclude this study, a number of these will be discussed as well as questions initiating further experimentation.

Homologous enzymes can show similar physical and catalytic properties as a result of the strong selective pressure for the main- tanence of the structural intregrity of the active site. According to Lipscomb (1971), the active site usually is in a localized depressed area of the enzyme and not involved in molecular interactions. Active site analysis has so predisposed biochemists for the last decade that it has only recently been recognized that structural alterations of the entire molecule are equally important. For example, the filling of the active site cavity by the substrate of carboxypeptidase A results in localized movement of specific groups which is amplified many fold throughout the entire molecule (Lipscomb, 1971). These distortions of enzyme by substrate complexing, results in a decrease in the energy of activation, a theory used by Pauling in 1948 to explain the catalytic efficiency of enzymes. These events appear to be so basic, and the catalytic site so conservative, that active site analysis, per se, may not give us any new information concerning the unique nature of poikilo- thermic enzymes.

In fact, Table VII, 1 indicates this very conclusion. Each parameter listed on this table for rainbow trout liver NADP-IDH has a near mirror image in the pig heart enzyme, except electrophoretic mobility. Molecular weights are similar, although not identical, as are the Km of DL-isocit- Table VII, 1. Comparative properties of rainbow trout liver and pig heart NADP-IDH,

Trout Liver Pig Heart NADP-IDH NADP-IDH

Approximate m.w. > 60,000 60,000

Electrophoretic mobility at pH 7,0 anodally cathodally

Ea value 19 Kcal/mole 25.Kcal/mole

Km(DL-isocit) at 2.5 x 10"5M(10°C) 3,2 x 10~5M(37 ambient temperatures absolute absolute Cation requirement present present <^ -KGA inhibiton present ADP activation

OXA + glyoxylate complete complete inhibition (1 mM) 95 rate and Ea values. The substrate, cofactor and inhibitor specificities are identical. Chung & Franzen (1970) have studied NADP+-binding and H+ release from bacterial (A. vinelandii) and mammalian (pig heart) NADP-

IDH and found that they are identical in both cases. These properties can be termed general enzyme characteristics; that is, the invariant properties which differentiate IDH from other enzymes. Undoubtedly, there are strong selective pressures for their maintanence, and if through mutations they are altered, natural selection will take its toll.

These types of mutations according to Ohno (1970) are termed forbidden mutations since they are always selected against.

Tolerable mutations, whether neutral or favored, can accumulate in other parts of the amino acid sequence by missense or samesense mutations. Fitch (1966) found that 40% of all missense mutations result in changing the net charge of the polypeptide chain specified by that gene locus. It is not surprising, therefore, to find a difference in electrophoretic mobilities of homologous enzymes from widely separated species.

These subtle changes in primary sequence may be responsible for the unique properties of poikilothermic enzymes. The hypothesis that the primary sequence of a protein determines its conformation in a given environment has been proven in many cases (Epstein, Goldberger &

Anfinsen, 1963). However, in vivo folding of a polypeptide chain could lead to a stable active conformation at an energy minimum, which need not necessarily be the thermodynamically most stable enzyme con• formation (Nickerson & Day, 1969). An example of such a "metastable" but active enzyme species, has been found for chicken mitochondrial malate dehydrogenase (Kitto, Wassarman & Kaplan, 1966). This increased conformational flexibility, persumably under strong positive selective 96 pressures (Alexandrov, 1969), may be responsible for the altered functional properties of rainbow trout liver NADP-IDH compared to the pig heart enzyme.

As has been argued by Hochachka & Somero (1971), the adaptive significance of the Km-temperature response resides in two points;

(1) at the thermal range encountered by the organism, a constant Km for temperature-independent control of catalysis is favored, and (2). at the extreme temperatures, a reduction in Q-^Q at low substrate concentrations minimizing the increase in temperature is favored.

Trout liver NADP-IDH, as seen in Ch. IV, exhibits this same response as seen for a large number of other enzymes from this species (Somero •

& Hochachka, 1971). It is apparent that selective pressures maintain this adaptive response only when necessary; in pig heart NADP-IDH, Km decreases directly with temperature, so that control of catalysis is not temperature-independent at any assay temperature. Other controlling parameters, such as the degree of product inhibition, also fall into this area of enzyme properties which are under the control of natural selection, and will vary depending upon the organism's environment and physiology.

Before this theoretical treatment can have experimental basis, homologous enzymes from poikilothermic and homeothermic tissues must be examined with the methods presently available to biochemists (Lipscomb, 1971).

One postulated mechanism for the origin of metazoans, vertebrates and finally mammals from unicellular organisms is evolution through tetraploidization (Ohno, 1970). The duplication of the entire genome results in the required redundancy needed for the creation of new functional genes. Many fish of the family Salmonidae, including the rainbow and other trout examined in this study (see Ch. Ill), are 97 autotetraploid, which explains the previously mentioned large number of multiple enzyme forms as well as a DNA content per cell twice that of most vertebrates. If evolution from this tetraploid stage is to occur,

the disomic state must eventually be re-established by functional diversification of the four original homologues, so that one original linkage group is split into two separate linkage groups. This process of diploidization is occurring at different rates not only in the family

Salmonidae, but also at specific gene loci within a single species

(Ohno, 1970).

Rainbow trout liver soluble NADP-IDH appears to be subject to

tetrasomic inheritance; that is, diploidization of this locus is not complete as has been found by Wolf, Engel & Faust (1970) in J3. irideus.

The evidence for this conclusion is the existence of various isozymal phenotypes found in individual hatchery trout, as reported in Ch. VI.

Brook trout are in a similar process of diploidization, but both splake

and lake trout (if tetraploidization really occurred in this species) have disomic inheritance, with the occurrence of a single locus in

lake and two loci in splake trout liver NADP-IDH. The splake trout phenotype is unique in its response to temperature; here, three activity bands are always seen, only the relative mobilities are altered. Such

an alteration may be similar to the conformational isozymes seen for

chicken mitochondrial MDH (Kitto, Wassarman & Kaplan, 1966), but

further study of this isozyme system is necessary.

Acclimation to a new thermal regime has been found to alter the

expression of trout liver NADP-IDH in a number of species (Ch. Ill and

IV). The increase in the relative amount of the slowest migrating

activity band in the rainbow trout enzyme alters the Km-temperature 98 response of 2°C- vs 17°C-acclimated fish. During certain seasons, the

17°C-acclimated brook trout enzyme shows a similar increase in the number of NADP-IDH isozymes, The splake trout is different in that temperature either alters the enzyme already present, as suggested above, or that the expression of an entirely new set of isozymes occur.

In wild populations of Lake Erie goldfish, NADP-IDH heterogeneity has been correlated with the high levels of pollution, so that increased flexibility will be selectively advantageous (Quiroz-Gutierrez & Ohno,

1970). Thermal fluctuations can also put a premium on enzyme hetero• geneity, as seen from these results in direct support of the theoreties of Haldane (1955) and Mayr (1963). Lake trout, however, being extremely s tenothermal have not been forced to maintain large amounts of enzyme heterogeneity.

Taken together, the kinetic results and the extent of heterogeneity at the NADP-IDH locus(i) suggest either that this enzyme is evolving at a rapid rate, or that selective pressures to maintain one specific phenotype are not great. Studies on hatchery populations of trout can not differentiate between these alternatives. The most important point here, however, is that the in vivo alterations in the genotypic expression of these isozymes, irrespective of cellular milieu changes, can determine the kinetic response of NADP-IDH to temperature. This

suggests the importance of the Km-temperature response to the in vivo operation of this enzyme.

As mentioned in the introduction, the role of NADP-IDH in the

cellular metabolism of vertebrates (except for ruminants) is unknown.

NAD-IDH apparently functions in the Krebs cycle oxidation of isocitrate 99

(Nicholls & Garland, 1969), and glucose-6-phosphate dehydrogenase of the pentose shunt and "malic" enzyme provide sufficient reducing equivalents for fatty acid synthesis (Flatt & Ball, 1964; Wise & Ball, 1964; Katz

& Rognstad, 1966).

The rainbow trout liver enzyme may play a more important role in metabolism than seen in other vertebrates for three reasons: (1) the apparent absence of detectable NAD-IDH activity; (2) the observed control properties of the enzyme; and (3) the requirement for increased fatty acid synthesis during thermal acclimation. Gumbman & Tappel

(1962) investigated the Krebs cycle of fish tissues, but did not report the presence of a NAD-IDH, only the NADP-dependent enzyme. Further studies are in order to completely eliminate the possibility of this enzyme activity in fish mitochondria, but present evidence is all negative (Ch. IV; Crabtree & Newsholme, 1970).

Modulation of the Km of DL-isocitrate by ADP and -KGA (Ch. IV) both suggest the functioning of this enzyme when adenylate charge is low; i.e., during periods when ATP production is required (Atkinson,

1968). Recent studies on the homologous enzyme from Alaskan king crab

(Moon, unpublished data) tissues suggest NADPH is a potent inhibitor of

NADP-IDH, as is 0( -KGA, again implying that high energy levels inhibit this enzyme. A similar observation of NADPH inhibition of A. vinelandii

NADP-IDH has been reported by Franzen,. Wichen & Chung (1971). A NADP-

IDH enzyme with these characteristics would function well within the

Krebs cycle (Atkinson, 1968); in fact, similar reasoning has been used in assigning this function to the NAD-IDH in mammalian mitochondria

(Nicholls & Garland, 1969).

Ratios of NADPH/NADP within the cytoplasm are extremely high, 100

approaching 1000 (Krebs & Veech, 1967), Even at ratios of 1:1, king crab IDH (Moon, unpublished data), "malic" enzyme (Behrisch, unpublished data) and glucose-6-phosphate dehydrogenase (Hochachka, unpublished data) are 30-50% inhibited. Lipogenesis, per se, must control the activity of the NADPH producing enzymes by de-inhibition through alter• ing the NADPH/NADP ratios. These cytoplasmically located enzymes would be under extremely rigid control and could function only when lipogenetic activity is high.

In the absence of further evidence, it is difficult to unequivocally assign a specific in vivo function to NADP-IDH in rainbow trout liver.

The kinetic data presented here would suggest it functions mainly in the Krebs cycle oxidation of isocitrate, while it may have an additional potential role in lipogenesis.

The question remains as to what mechanisms allow for the altered expression and kinetic properties of trout liver NADP-IDH following acclimation to a new thermal regime. For instance, does temperature have a direct cellular effect or does it function coincidently with photoperiod? Recent studies on the altered expression of lactate dehydrogenase isozymes with changing photoperiod (Massaro & Markert,

1971) has an interesting similarity to some of these data. This initiation response is followed by a systemic response, undoubtedly mediated through the action of hormones and/or the nervous system

(Somero & Hochachka, 1971), A series of biochemical responses, culminating in enzyme properties similar to those reported here, produce a new steady state different from that found before the acclimation process began. The gap between initiation and culmination of the process is 101 wide, although studies by Lehmann (1970a,b) on the transitory changes in enzyme activities are a start in the right direction. However, a total approach to this transition period thus far has not been initiated. CHAPTER VIII:

Literature Cited 102

Alexandrov, V. Ya. 1969. Conformational flexability of proteins, their resistance to proteinases and temperature conditions of life. Curr. Mod. Biol. 3_:9-19.

Anderson, T. R. 1970. Temperature adaptation and the phospholipids of membranes in goldfish (Carassius auratus). Comp. Biochem. Physiol. 33:663-687.

Assaf, S. A. & Graves, D. J. 1969. Structural and catalytic properties of lobster muscle glycogen phosphorylase. J. Biol. Chem. 244:5544- 5555.

Atkinson, D. E. 1966. Regulation of enzyme activity. Ann. Rev. Biochem. 35:85-124.

Atkinson, D. E. 1968. Citrate and the citrate cycle in the regulation of energy metabolism. In: "Metabolic Roles of Citrate" (T. W. Goodwin, ed.), Biochemical Society Symposia 2J_:23-40. Academic Press, New York.

Atkinson, D. E., Hathaway, J. A, & Smith, E. C. 1965. Kinetics of regulatory enzymes. Kinetic order of the yeast NAD-isocitrate dehydrogenase reaction and a model for the reaction. J. Biol, Chem. 240:2682-2690.

Bailey, G, S., Cocks, C, T, & Wilson, A. C, 1969. Gene duplication in fishes: malate dehydrogenase of salmon and trout. Biochem. Biophys. Res. Comm. 34;605-612.

Baldwin, J. & Hochachka, P, W. 1970. Functional significance of iso• enzymes in thermal acclimatization. Acetylcholinesterase from trout brain. Biochem. J. 116:883-887.

Barcroft, J. 1934. Features in the Architecture of Physiological Function, 368 pp. Cambridge University Press, Cambridge.

Baron, D. N. & Bell, J. L. 1962. Isoenzymes of isocitrate dehydrogenase. Proc. Assoc. Clin. Biochem. 2^:8-10.

Barrera, C. R. & Jurtshuk, P. 1970. Characterization of the highly active isocitrate (NADP+) dehydrogenase of Azotobacter vinelandii. Biochim. Biophys. Acta 220:416-429.

Bauman, D. E., Brown, R. E, & Davis, C. L. 1970. Pathways of fatty 'acid synthesis and reducing equivalent generation in mammary gland of rat, sow and cow. Arch. Biochem. Biophys. 140:237-244.

Behrisch, H. W. 1969. Temperature and the regulation of enzyme activity in poikilotherms. Fructose-diphosphatase from migrating salmon. Biochem. J. 117:687-696.

Behrisch, H. W. & Hochachka, P. W. 1969. Temperature and regulation of enzyme activity in poikilotherms. Properties of lungfish fructose- diphosphatase. . Biochem. J. 112:601-607. 103

Brett, J. R. 1956. Some principles in the thermal requirements of fishes. Quart. Rev. Biol. 31:75-87.

Brett, J. R. 1970. Temperature. Fishes. In: "Marine Ecology" (0. Kinne, ed.), pp. 515-560. Wiley-Interscience, London.

Caldwell, R. S. & Vernberg, F. J. 1970. The influence of acclimation temperature on the lipid composition of fish gill mitochondria. Comp. Biochem. Physiol. 34:179-191,

Chappell, J. B. & Robinson, B. H. 1968, Penetration of the mitochondrial membrane by tricarboxylic acid anions. In: "Metabolic Roles of Citrate" (T. W. Goodwin, ed.), Biochemical Society Symposium 27_; 123-134 . Academic Press, New York.

Charles, A. M. 1970. Properties of an NADP-specific isocitrate dehydrogenase from Thiobacillus novellus. Can. J. Biochem. 48: 95-103.

Chen, R. F. & Plaut, G. W. E. 1963. Studies on the H+-transfer mediated by the NAD-isocitrate dehydrogenase of heart. Biochem. 2^:752-756.

Chung, A. E, & Franzen, J, S. 1969. Oxidized TPN-specific isocitrate dehydrogenase from Azotobacter yinelandii. Isolation and character• ization. Biochem. 8^:3175-3184.

Chung, A. E, & Franzen, J, S. 1970. Oxidized TPN-specific isocitrate dehydrogenase from Azotobacter yinelandii. The pathway of hydrogen in the enzymatic reaction. Arch. Biochem. Biophys. 141:416-422.

- Colman, R. F. 1968. Effect of modification of a methionyl residue on the kinetic and molecular properties of isocitrate dehydrogenase. J, Biol. Chem. 234:2454-2464.

Colman, R. F., Szeto, R. C. & Cohen, P. 1970. Pig heart TPN-specific isocitrate dehydrogenase. A single polypeptide chain. Biochem. £:A945-4949.

Cowey, C. B. 1967. Comparative studies on the activity of D-glycer- aldehyde-3-phosphate dehydrogenase from cold- and warm-blooded animals with reference to temperature. Comp. Biochem. Physiol. 23:969-976.

Crabtree, B, & Newsholme, E. A. 1970. The activities of NAD- and NADP- linked isocitrate dehydrogenase in insect and vertebrate muscle. Biochem. J. 116:22P-23P.

Dean, J. M. 1969. The metabolism of tissues of thermally acclimated trout (Salmo gairdnerii). Comp, Biochem. Physiol. 29:185-196. 104

DeDuve, C, Pressman, B. C, Gianetto, R. , Wattlaux, R. & Appelmans, F. 1955. Tissue fractionation studies. VI. Intracellular distribution patterns of enzymes in rat liver tissue. Biochem. J. 60:604-617.

Dixon, M. & Webb, E. C. 1964. Enzymes, 2nd edition, 950 pp. Longmans, Green and Co. Ltd., London.

Ekberg, D. R. 1958. Respiration in tissues of goldfish adapted to high and low temperatures. Biol. Bull., Woods Hole 114:308-316.

Engel, W., Hof, J. Op't & Wolf, U. 1970. Genduplikation durch poly- ploide Evolution: die Isoenzyme der Sorbitdehydrogenase bei herings- und lachsartigen Fischen (Isospondyli). Humangenetik _9:157-163.

Eppenberger, H. M., Scholl, A. & Ursprung, H. 1971. Tissue-specific isoenzyme patterns of creatine kinase in trout. FEBS Letters 14:317-319.

Epstein, C. J., Goldberger, R. F, & Anfinsen, C. B. 1963. The genetic control of tertiary protein structure studies with model systems. Cold Spring Harbor Symp. Quant. Biol. 28:439-450.

Esfahani, M., Barnes, E. M., Jr. & Wakil, S. J. 1970. Control of fatty acid composition in phospholipids of E_. coli: Response to fatty acid supplements in a fatty acid auxotroph. Proc. Natl. Acad. Sci., U. S, A. 64:1057-1064.

Esfahani, M., Linbrick, A. R., Knutton, S. & Wakil, S. J. 1971. Molecular organization of lipids in E, coli membranes. Fedn. Am. Soc. exp. Biol., Proc, 30:390.

. Fitch, W. M, 1966. An improved method of testing for evolutionary homology. J. Mol. Biol. 16:9-16.

Flatt, J, P. & Ball, E, G. 1966, Studies on the metabolism of adipose tissue. XIX. An evaluation of the major pathways of glucose catabolism as influenced by acetate in the presence of insulin. J. Biol. Chem. 241:2862-2869.

Flint, A. P. F. & Denton, R. M. 1970. The role of NADP-dependent malate dehydrogenase and isocitrate dehydrogenase in the supply of reduced NADP for steroidogenesis in the superovulated rat ovary. Biochem. J, 117:73-83.

Fox, C. F. , Law, J. H., Tsukagoshi, N. & Wilson, G. 1970. A density label for membranes. Proc. Natl. Acad. Sci., U. S. A. 67:598-605.

Franzen, J, S., Wicken, J. S. & Chung, A. E. 1971. TPN-isocitrate dehydrogenase from A. vinelandii: Order of substrate addition and product release. Fedn. Am. Soc. exp. Biol,, Proc, 30:1131. 105

Freed, J. M. 1969. Comparative study of temperature effects on phosphofructokinase of Paralithodes camtshatica and cold- and warm- acclimated Carassius auratus. Ph.D. Thesis, University of Illinois.

Fry, F. E. J. 1947. The effects of environment on animal activity. Publ. Ontario Fisheries Res. Lab. 68:5-62.

Fry, F. E. J. & Hochachka, P.W. 1970. Fish. In: "Comparative Physiology of Thermal Regulation" (E. Whitton, ed.), pp. 139- 179. Academic Press, New York.

Goebell, H. & Klingenberg, M. 1963. High activity of NAD-isocitrate dehydrogenase in mitochondria from various organs. Biochem. Biophys. Res. Comm. 13:209-212.

Goebell, H. & Klingenberg, M, 1964. NAD-isocitrate dehydrogenase of mitocondria. Biochem. Z. 340:441-464.

Gumaa, K. A., McLean, P. & Greenbaum, A. D. 1971. The changes of metabolite content and the control of phosphofructokinase in rat liver in different dietary and hormonal conditions. FEBS Letters 13:5-9.

Gumbman, M. & Tappel, A. L. 1962. The tricarboxylic acid cycle in fish. Arch. Biochem. Biophys. 98:262-270.

Haglund, H. 1967. Isoelectric focusing in natural pH gradients: A technique of growing importance for fractionation and characteriz• ation of proteins, Sci, Tools 14:17-25.

Haldane, J. B. S. 1955. On the biochemistry of heterosis, and the stabilization of polymorphism. Proc. Roy. Soc, B 144:217-220,

Hampton, M. L. & Hanson, R. S. 1969. Regulation of isocitrate dehydro• genase from Thiobacillus thiooxidaus and Pseudomonas fluorescens. Biochem. Biophys. Res. Comm. 36:296-305.

Hanson, R. W. & Ballard, F. J. 1967. The relative significance of acetate and glucose as precursors for lipid synthesis in liver and adipose tissues from ruminants. Biochem. J. 1_05:529-536.

Hathaway, J. A. & Atkinson, D. E. 1965. Kinetics of regulatory enzymes: Effect of adenosine triphosphate on yeast citrate synthase. Biochem. Biophys. Res. Comm. 20:661-665.

Hazel, J. & Prosser, C. L. 1970. Interpretation of inverse acclimation to temperature. Z. vergl. Physiol. 67:217-278.

Henderson, N. S. 1965. Isozymes of isocitrate dehydrogenase. Subunit structure and intracellular location. J. Exp. Zool. 158:263-274. 106

Hickman, C. P., Jr., McNabb, R, A., Nelson, J, S., van Breeman, E, D. & Comfort, D. 1964. Effect of cold-acclimation on electrolyte distribution in rainbow trout (Salmo gairdnerii). Can. J. Zool. 42:577-597.

Higashi, T., Maruyama, E., Otani, T. & Sakamoto, Y. 1965. Studies on isocitrate dehydrogenase. I. Some properties of isocitrate dehyd• rogenase from beef heart muscle. J, Biochem. 57:793-798.

Hill, R. L., Brew, K. , Vanaman, T. C, Trayer, I. P. & Mattock, P. 1968. The structure, function and evolution of -lactalbumin. Brookhaven Symp. Biol. 21:139-154.

Hochachka, P. W. 1965. Isoenzymes in metabolic adaptation of poikilo- therms: Subunit relationships in lactate dehydrogenases of goldfish. Arch. Biochem. Biophys. 111:96-103.

Hochachka, P. W. 1966. Lactate dehydrogenases in poikilotherms: Definition of a complex isozyme system. Comp. Biochem. Physiol. 18:261-269,

Hochachka, P, W. 1967, Organization of metabolism during temperature compensation. In: "Molecular Mechanisms of Temperature Adaptation" (C. L. Prosser, ed.), Publ, No, 84, pp, 177-203. Am, Assoc. Advance. Sci., Washington, D.C.

Hochachka, P.W. 1968. Action of temperature on branch points in glucose and acetate metabolism. Comp. Biochem. Physiol. 25:107-118.

Hochachka, P.W. & Hayes, F. R. 1962. The effect of temperature accl• imation on pathways of glucose metabolism in the trout. Can. J. Zool. 20:261-270.

Hochachka, P. W. & Lewis, J. K. 1970. Enzyme variants in thermal acclimation: Trout liver citrate synthases. J. Biol. Chem. 245: 6567-6*573.

Hochachka, P. W., Schneider, D. E. & Moon, T. W. 1971. The adaptation of enzymes to pressure. I. A comparison of trout liver fructose- diphosphatase with the homologous enzyme from an off-shore benthic fish. Am, Zool. (In press)

Hochachka, P. W. & Somero, G. N. 1971. Biochemical adaptation to the environment. In: "Fish Physiology" (W. S. Hoar & D. J. Randall, ed.), Vol. VI. Academic Press, New York. (In press)

Houston, A, M., Madden, J. A, & DeWilde, M. A. 1970. Environmental temperature and the body fluid system of the freshwater teleost. IV. Water-electrolyte regulation in thermally acclimated carp, Cyprinus carpio. Comp. Biochem. Physiol. 34:805-818.

Howard, R. L. & Becker, R. R. 1970. Isolation and some properties of the TPN-isocitrate dehydrogenase from Bacillus stearothermophilus. J. Biol. Chem, 245:3186-3194. 107

Hunter, R. L. & Markert, C, L, 1957. Histochemical demonstration of enzymes separated by zone electrophoresis in starch gels. Science 125:1294-1295.

Ibsen, K. H. , Schiller, K. W. & Haas, T. A. 1971. Interconvertible kinetic and physical forms of human erythrocyte pyruvate kinase. J. Biol. Chem. 246:1233-1240.

Ikai, A. & Tanford, C. 1971. Kinetic evidence for incorrectly folded intermediate states in the refolding of denatured proteins. Nature, London 230:100-102.

Illingworth, J. A. & Tipton, K. F. 1970. Purification and properties of the NADP-dependent isocitrate dehydrogenase from pig liver cytoplasm. Biochem. J. 118:253-258.

Jankowsky, H. D. 1968. Versuche zur Adaptation der Fische im normalen Temperaturbereich. HelgolHnder wiss. Meeresunters 1_8_: 317-362.

Johnston, P. V. & Roots, B. I. 1964. Brain lipids, fatty acids and temperature acclimation. Comp. Biochem. Physiol. 11:303-310.

Kaloustian, H. D. & Kaplan, N. 0. 1969. Lactate dehydrogenase of lobster (Homarus americanus) tail muscle. II, Kinetics and regulatory properties. J. Biol. Chem, 244^:2902-2910.

Kanungo, M. S. & Prosser, C. L, 1959. Physiological and biochemical adaptation of goldfish to cold and warm temperatures, II. Oxygen consumption of liver homogenates, oxygen consumption and oxidative phosphorylation of liver mitochondria. J. cell, Comp. Physiol. 54:265-274.

Katz, J. & Rognstad, R. 1966. The metabolism of tritiated glucose by rat adipose tissue. J. Biol. Chem. 241:3600-3610.

Kemp, P. & Smith, M. W, 1970. Effect of temperature acclimatization on the fatty acid composition of goldfish intestinal lipids. Biochem. J. 117:9-15.

Kemper, D. L. & Kaplan, N. 0. 1971. The control of TPN-isocitrate dehydrogenase and glucose-6-phosphate dehydrogenase through changes in aggregation state. Fedn. Am. Soc. exp. Biol., Proc. 30:1059.

Kitto, G. B., Wassarman, P, M. & Kaplan, N. 0. 1966. Enzymatically active conformers of mitochondrial malate dehydrogenase. Proc. Natl. Acad, Sci., U. S. A. 56:578-585.

Knipprath, W. G. & Mead, J. F. 1966. Influence of temperature on the fatty acid pattern of mosquito fish (Gambusia affinis) and guppies (Lebistes reticulatus). Lipids 1_:113-117.

Knipprath, W, G. & Mead, J. F. 1968. The effect of environmental temp• erature on the fatty acid composition and on the in_ vivo incorporation of 1-^C-acetate in goldfish (Carassius auratus L.). Lipids 3_:121- 128. 108

Koehn, R. K. 1969. Esterase Heterogeneity: Dynamics of a polymorphism. Science 163:943-944.

Krebs, H. A. & Veech, R. L. 1968. Equilibrium relations between pyridine nucleotides and adenine nucleotides and their roles in the regulation of metabolic processes. Adv. Enz. Reg. 7^397-413.

Kumamoto, J., Raison, J. K. & Lyons, J. M. 1971. Temperature "breaks" in Arrhenius plots: A thermodynamic consequence of a phase change. J. Theor. Biol. 31:47-51.

Lebherz, H. G. & Rutter, W. J, 1967. Glyceraldehyde-3-phosphate dehydrogenase variants in phyletically diverse organisms. Science 157:1198-1200.

Lehmann, J. 1970a. Uber die AbhHngigkeit der EnzymaktivitMten im Rumpfmuskel der Goldorfe (Idus idus L.) von der Vorbehandlungs- temperatur. Int. Revue ges. Hydrobiol. 55_: 413-429.

Lehmann, J. 1970b. VerHnderungen der EnzymaktivitMten nach einem Wechsel der Adaptationstermperatur, untersucht im Seitenrumpfmuskel des Goldfisches (Carassius auratus L.). Int. Revue ges. Hydrobiol. 52:763-781.

Lehrer, G. M. & Barker, R. 1970. Conformational changes in rabbit muscle aldolase. Biochem. 9^1533-1540.

Levy, P. L. & Salthe, S. N. 1971. Kinetic studies on variant heart- type lactate dehydrogenases in the frog, Rana pipiens. Comp. Biochem. Physiol. 39B:343-355.

Lin, C, C, Schipmann, G. , Kittrell, W. A. & Ohno, S. 1969. The predominance of heterozygotes found in wild goldfish of Lake Erie at the gene locus for sorbitol dehydrogenase. Biochem. Genet. 3_: 603-610.

Lipscomb, W. N. 1971. Relationship among diffraction, chemical, spectro• scopic and theoretical studies of enzymes. Fedn. Am. Soc. exp. Biol., Proc. 30:1042Abs.

Lowenstein, J. M. 1961a. The pathway of hydrogen in biosynthesis. I. Experiments with glucose-l-H and lactate-l-H"^. J. Biol. Chem. 236:1213-1216.

Lowenstein, J. M. 1961b. The pathway of hydrogen in biosynthesis. II. Extramitochondrial isocitrate dehydrogenase. J. Biol. Chem. 236: 1217-1219.

Lowenstein, J. M. & Smith, S. R. 1962. Intra- and extramitochondrial isocitrate dehydrogenase. Biochim. Biophys. Acta 56^:385-387.

Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-273. 109

Magar, M. E. & Robbins, J. C. 1969. The subunits of porcine heart TPN-linked isocitrate dehydrogenase. Biochim. Biophys. Acta 191: 173-176.

Mahler, H. R. & Cordes, E. H. 1966. Biological Chemistry. Harper and Row, New York.

Marr, J. J. & Weber, M. M. 1968. Nucleotide control of a NADP- specific isocitrate dehydrogenase. Bact. Proc. 1968, 117.

Marr, J. J. & Weber, M. M. 1969a. Allosteric inhibition of a TPN- specific isocitrate dehydrogenase from Crithidia fasciculata by nucleoside triphosphates. J. Biol. Chem. 244:2503-2509.

Marr, J. J. & Weber, M...M. 1969b. Feedback inhibition of an allosteric TPN-specific isocitrate dehydrogenase. J. Biol. Chem. 244:5709- 5712.

Marr, J. J. & Weber, M. M. 1969c. Concerted inhibition of a NADP- specific isocitrate dehydrogenase and the implication, for metabolic regulation. Biochem. Biophys. Res. Comm. 35.: 12-19.

Massaro, E. J. & Markert, C. L. 1968. Isozyme patterns of salmonid fishes: Evidence for multiple cistrons for lactate dehydrogenase polypeptides. J. Exp. Zool. 168^:223-238.

Mayr, E. 1963. Animal Species and Evolution. Harvard University Press, Cambridge, Massachusettes.

Mintz, B. & Baker, W. W. 1967. Normal mammalian muscle differentiation and gene control of isocitrate dehydrogenase synthesis. Proc. Natl. Acad. Sci., U. S. A. 58:592-598.

Moon, T. W. 1970. Isozymes in temperature acclimation: Trout NADP- linked isocitrate dehydrogenases. Fedn. Am. Soc. exp. Biol., Proc. 29:869.'

Moon, T. W. & Hochachka, P. W. 1971. Temperature and enzyme activity in poikilotherms: Isocitrate dehydrogenases in rainbow trout liver. Biochem. J. (In press)

Moon, T. W., Mustafa, T. & Hochachka, P. W. 1971. The adaptation of enzymes to pressure. II. A comparison of muscle pyruvate kinases from surface and mid-water fishes with the homologous enzyme from an off-shore benthic species. Am. Zool. (In press)

Moyle, J. & Dixon, M. 1956. Purification of the isocitric enzyme (TPN- linked IDH-oxalosuccinic carboxylase). Biochem. J. 63:548-552.

Newell, R. C. & Pye, V. I. 1971. Qualitative aspects of the relation• ship between metabolism and temperature in the winkle Littorina littorea (L). Comp. Biochem. Physiol. 38:635-650. 110

Nicholls, D. G. & Garland, P. B. 1969. The control of isocitrate oxid• ation by rat liver mitochondria. Biochem. J. 114:215-225.

Nickerson, K. W. & Day, R. A. 1969. Possible biological roles for metastable proteins. Curr. Mod. Biol. 2^:303-306.

Ochoa, S. 1948. Biosynthesis of tricarboxylic acids by CO2 fixation. III. Enzymatic mechanisms. J. Biol. Chem. 174:133-157.

O'Hea, E. K. & Leveille, G. A. 1968, Lipogenesis in isolated adipose tissue of the domestic chick (Gallus domesticus). Comp. Biochem. Physiol. 26:111-120.

Ohno, S, 1970. Evolution by Gene Duplication. Springer-Verlag, New York.

Ohno, S,, Wolf, U. & Atkin, N. B. 1968. Evolution from fish to mammals by gene duplication. Hereditas _59:169-187.

Overath, P., Schairer, H. U. & Stoffel, W. 1970. Correlation of in vivo and in vitro phase transitions of membrane lipids in E_. coli. Proc. Natl. Acad. Sci., U. S. A. 67 -.606-612.

Ozaki, H. & Shiio, I. 1968. Regulation of the tricarboxylic acid cycle and glyoxylate cycles in Brevibacterium flavum. J. Biochem, 64: 355-363.

Parker, M. G. & Weitzman, P. D. J. 1970. Regulation of NADP-linked isocitrate dehydrogenase activity in Acinetobacter. FEBS Letters 7^:324-326.

Pauling, L. 1948. Nature of forces between large molecules of biological interest. Nature, London l6l:707-709.

Plaut, G. W. E. 1963. Isocitrate dehydrogenas. In: "The Enzymes" (P. D. Boyer, H. Lardy & K. MyrbMck, ed.), pp. 105-126, Vol. 5. Academic Press, New York.

Plaut, G. W. E. & Aogaichi, T. 1967. The separation of NAD- and NADP- isocitrate dehydrogenase activities of mammalian liver. Biochem. Biophys. Res. Comm. 28:628-634.

Plaut, G. W. E. & Aogaichi, T. 1968. Purification and properties of NAD-isocitrate dehydrogenase of mammalian liver. J. Biol. Chem. 243:5572-5583.

Prosser, CL. & Brown, F. A., Jr. 1962. Comparative Animal Physiology, 2nd. edition. W. B. Saunders Co., Philadelphia.

Quiroz-Gutierrez, A. & Ohno, S. 1970. The evidence of gene duplication for the S-form NADP-linked isocitrate dehydrogenase in carp and goldfish. Biochem. Genet. 4^:93-99. Ill

Rahn, H. 1965. Gas transport from the external environment to the cell. In: "Development of the Lung" (A. V. S. DeReuch & R. Porter, ed.), Ciba Foundation Symp., pp. 3-20. J. and A. Churchill Ltd., London.

Raison, J. K., Lyons, J. M. & Thomson, W. W. 1971. The influence of membranes on the temperature-induced changes in the kinetics of some respiratory enzymes of mitochondria. Arch. Biochem. Biophys. 142:83-90.

Reeves, H. C, Brehmeyer, B. A. & Ajl, S. J. 1968. Multiple forms of bacterial NADP-specific isocitrate dehydrogenase. Science 162:359-360.

Reeves, H. C, Brehmeyer, B. A. & Ail, S. J. 1968. Multiple forms of isocitrate dehydrogenase (NADP+) in E. coli. Bact. Proc. 1968, 117.

Reeves, R. B. & Wilson, T. L. 1969. Intracellular pH in bullfrog striated and cardiac muscle as a function of body temperature. Fedn. Am. Soc. exp. Biol., Proc. 28:2927A.

Roots, B. I. 1968. Phospholipids of goldfish (Carassius auratus L.) brain: The influence of environmental temperature. Comp. Biochem. Physiol. 25:457-466.

Rosenberg, M. 1971. Epigenetic control of lactate dehydrogenase subunit assembly. Nature, New Biology 230:12-14.

Savard, K. , Marsh, J. M, & Howell, D. S. 1963. Progesterone biosyn• thesis in luteal tissue: Role of NADP+ and NADP-linked dehydrogen• ases. Endocrin. 73 :554-563.

Schairer, H. U. & Overath, P. 1969. Lipids containing trans- unsaturated fatty acids change the temperature characteristics of thiomethylgalactoside accumulation in E_. coli. J. Mol. Biol. 44:209-214.

Self, C. H. & Weitzman, P. D. J. 1970. Separation of isoenzymes by zonal centrifugation. Nature, London 225 :644-645.

Shiio, I. & Ozaki, H. 1968. Concerted inhibition of isocitrate dehydrogenase by glyoxylate plus oxaloacetate. J. Biochem. 64: 45-53.

Smith, M. W. & Kemp, P. 1969. Phospholipase C-induced changes in intestinal adenosine triphosphatase prepared from goldfish acclimated to different temperatures. Biochem. J. 114:659-661.

Smithies, 0. 1955. Zone electrophoresis in starch gels: Group variations in the serum proteins of normal human adults. Biochem. JT 61: 629-641. 112

Somero, G. N. & Hochachka, P°. W. 1968. The effect of temperature on catalytic and regulatory functions of pyruvate kinases of the rainbow trout and the Antarctic fish (Trematomus bernacchii). Biochem. J. 110:395-400.

Somero, G. N. & Hochachka, P. W. 1971. Biochemical adaptation to the environment. Am. Zool. 11:159-167.

Tait, A. 1970. Genetics of NADP-isocitrate dehydrogenase in Paramecium aurelia. Nature, London 225:181-182.

Tsuyuki, H. & Wold, F. 1964. Enolase: Multiple molecular forms in fish muscle. Science 146:535-537.

Vroman, H. E. & Brown, J. R. C. 1963. The effect of temperature on the activity of succinic dehydrogenase from livers of rats and frogs. J. cell. Comp. Physiol. 61:129-131.

Waite, M. & Wakil, S. J. 1962. Studies on the mechanism of fatty acid synthesis. XII. Acetyl Coenzyme A carboxylase. J. Biol. Chem. 237:2750-2757.

Wakil, S.J., Titchener, E. B. & Gibson, D. M. 1959. Studies on the mechanism of fatty acid synthesis. VI. Spectrophotometric assay and stoichiometry of fatty acid synthesis. Biochim. Biophys. Acta 34:227-233.

Walker, P. R. & Bailey, E. 1969. Role of "malic enzyme" in lipogenesis. Biochim. Biophys. Acta 187^:591-593.

Whitt, G. S. 1970. Directed assembly of polypeptides of the isozyme lactate dehydrogenase. Arch. Biochem. Biophys. 138:352-354.

Williamson, J. R., Herczeg, B. E., Coles, H. J. & Cheung, Y. 1967. Glycolytic control mechanisms. V. Kinetics of high energy-phosphate intermediate changes during electrical discharge and recovery in the main organ of Electrophorus electricus. J. Biol. Chem. 242: 5119-5124.

Williamson, J. R., Scholz, R. & Browning, E. T. 1969. Control mech• anisms of gluconeogenesis and ketogenesis. II. Interaction between fatty acid oxidation and the citric acid cycle in perfused rat liver. J. Biol. Chem. 244:4617-4627.

Wilson, G., Rose, S. P. & Fox, F. 1970. The effect of membrane lipid unsaturation on glycoside transport. Biochem. Biophys. Res. Comm. 38:617-623.

Wise, E. M., Jr. & Ball, E. G. 1964. Malic enzyme and lipogenesis. Proc. Natl. Acad. Sci., U. S. A. 52:1255-1263. 113

Wolf, U., Engel, W. & Faust, J. 1970, Zum mechanismus der diploid- isierung in der Wirbeltierevolution: Koexistenz von tetrasomen und disomen genloci der isocitrate dehydrogenase bei der Regen- bogenforelle (Salmo irideus). Humangenetik _9:150-156,

Wr6blewski, F. & Gregory, K. F. 1961. Lactate dehydrogenase isozymes and their distribution in normal tissues and plasma and in disease states. N, Y. Acad. Sci., Ann. 94:912-932.