THESIS

entitled

Studies on Fraction 1 Protein

of Beta vulgaris

Submitted for the degree of

DOCTOR OF PHILOSOPHY

by

Kenneth Edward Moon

1 970 TABLE OF CONTENTS

Acknowledgements i Abbreviations i

Summary ii Chapter 1 Introduction

1.1 Carbon Dioxide Fixation in the Calvin • 1 Cycle 1.2 Some Properties of RuDPcase and Similarity 2 to Fraction 1 Protein 1.3 Cellular Location of Fraction 1 Protein 3 1.4 The Mechanism of Action of RuDPcase 4

1.5 Structure of Fraction 1 Protein from 13 Higher Plants 1.6 Relationship of Structure to Enzymic 14 Activity and Role of Sulphydryl Groups

1.7 Studies on RuDPcase from Sources other 17 than the Higher Plants 1.8 Comparison of Some Kinetic and Physical 21 Constants of RuDPcase from Different Sources

1.9 Aim of This Research Project 23 Chapter 2 Methods and Materials 2.1 Plant Material 24 2.2 Preparation of Chloroplasts and Isolation 24 of Fraction 1 Protein 2.3 Concentrating Protein Solutions 26

2.4 Preparation of Reduced and Carboxy- 27 methylated Fraction 1 Protein

2.5 Preparation of Maleyl-carboxymethyl 28 Fraction 1 Protein

2.6 Removal of Maleyl Groups 29

2.7 Gel Electrophoresis 30

2.7 (i) Polyacrylamide Disk Electrophoresis 30

(ii) Polyacrylamide Electrophoresis in 31 the presence of SDS

(iii) Slab Acrylamide Gel Electrophoresis 32

(iv) Isoelectric Focusing 33

2.8 Ribulose-1,5-diphosphate Carboxylase 3^ Assay

2.9 Preparation of the S-Sulphenylsulphanate 34 Derivative of Fraction 1 Protein

2.10 Edman Degradations 35

2.11 Hydrazinolysis 35

2.12 Isolation of Blocked N-Terminal Peptides 36

2.13 Tryptic Digestion and Peptide Mapping 3®

2.14 Amino Acid Analysis

2.15 (i) Carboxypeptidase B Digestion 39

(ii) Carboxypeptidase A Digestion 39

2.16 Titration of Sulphydryl Groups 40

2.17 Protein Determinations 41

2.18 Chemicals 41 Chapter 3 Results

3.1 Preparation and Purity of Fraction 43 1 Protein

3.1 (i) Isoelectric Focuaing -of--Fraction 46 1 Protein

3.2 Amino Acid Analyses 50

3.3 Titration of Native Fraction 1 Protein 55

with 5 t5!-Dithiobis(2-nitrobenzoic .acid)

3.4 Studies on the Relationship of Enzymic 57 Activity to Sulphydryl Groups and Structure

3.5 Subunit Structure of Fraction 1 Protein 65

(i) Electrophoretic Studies on SCM- 65 Fraction 1 Protein

(ii) Molecular Weight Estimation of the 68 Subunits Resolved on Columns of Sephadex G-200

(iii) Amino Acid Composition of the 70 Subunit Components

(iv) Studies on Maleylated Fraction 1 72 Protein

(v) Further Characterisation of the 75 Subunits

3.6 Peptide Mapping of the Isolated Subunits 80

3.7 NH^-Terminal Amino Acid Analysis of 87 Subunit A 3.7 (i) Isolation of a Peptide with a Blocked 88 Amino Group

3.8 C-Terminal Residue 9k (i) Hydrazinolysis 9^

(ii) Carboxypeptidase B Digestion 95

(iii) Carboxypeptidase A Digestion 96

3.9 End Group Analysis of Subunit B 96

Chapter k Discussion 97 References 107 Publication 116 ACKNOWLEDGEMENTS

I wish to thank my supervisors, Professor E.O.P.

Thompson and Professor H. N. Barber, for their help and encouragement during the course of this study,

I am also grateful to my wife for typing this thesis and to Mr. P. M. Long for preparing the photo­ graphs appearing in the thesis.

ABBREVIATIONS

RuDPcase ribulose-1,5-diphosphate carboxylase

PCMB p-chloromercuribenzoic acid

SDS sodium dodecyl sulphate

DTNB 5,5!-dithiobis(2-nitrobenzoic acid)

SCM-cys teine S-carboxymethyl cysteine ii

SUMMARY

Fraction 1 protein isolated from spinach beet

chloroplasts was purified by ammonium sulphate fraction

ation and Sephadex G-200 gel filtration. The protein

was essentially homogeneous as judged by acrylamide

gel electrophoresis.

The RuDPcase activity was abolished by low or concentrations of urea and by exposure to a sulphydryl

blocking reagent, sodium tetrathionate. In this latter

case full RuDPcase activity was restored by incubating

the inactive protein with cysteine.

Structural studies have shown that if the isolated

protein was reduced and S-carboxymethylated in the

presence of 8M urea then two distinct subunits could

be resolved by gel filtration on Sephadex G-200 in an

8M urea buffer at pH 10.0.

A comparison of the amino acid composition of the

two subunits showed distinct differences between them.

The molecular weights of the two protein subunits were

estimated tobe 5^,000 and 16,000 by comparison of their

elution volume on gel filtration with elution volumes

of reduced carboxymethylated proteins of known molecular weight. iii

Further studies showed that when the protein was treated with maleic anhydride, after S-carboxymethlation, then the subunits of the protein could be separated on a

Sephadex G-100 column without the use of 8M urea in the column.

Electrophoretic examination of these two components in a Gradipore gel confirmed molecular weight dissimilarities. Precise values for their molecular weights were determined by electrophoresis in acrylamide gels containing SDS. Corrected values, i.e. allowances made for the introduction of the maleyl group, are

55»^00 and 12,000 for the large and small subunit respectively.

On an average dry weight basis the large subunit accounts for 68$ and the smaller subunit 32$. Using these values a tentative model of Fraction 1 protein consisting of six large subunits and twelve small subunits is proposed. It is suggested that the six larger subunits could lie on the six faces of a cube with the twelve smaller subunits occupying the twelve edges.

Peptide mapping of the tryptic digests of the isolated subunits gave different fingerprints, however in the case of the larger subunits the number of ninhydrin positive spots was only one half of the

theoretical number expected.

End group analysis, both amino and carboxyl, did not yield conclusive information. Evidence is presented

to suggest that the amino end of the larger subunit has its amino function masked. The techniques used in the analysis of the carboxyl end of the protein would not have detected a C-terminal basic residue which had as its pentultimate residue proline. 1

CHAPTER 1. INTRODUCTION

1.1 Carbon Dioxide Fixation in the Calvin Cycle

Ribulose 1,5-diphosphate carboxylase EC 4.1.1.39 occupies an important niche in the photosynthetic carbon reduction cycle of higher plants and other photosynthetic organisms. It is responsible for catalysing the carboxylation of ribulose 1,5-diphosphate to produce phosphoglyceric acid as shown in Reaction 1. (Calvin,

1962; Bassham, 1963; 1965).

H_COPO.H~ HoC0P0-H“ I 3 2. 3 c = 0 + CO ^2^ + 2H+ Reaction 1. I U2 2 HCOH H-C-OH I Mg2+ C00~ H-C-OH h2copo3h

The isolation from Chlorella of a cell-free extract that was capable of catalysing the above reaction was reported by Quayle et al. in 1954. Shortly after this, Weissbach et al. (1956) isolated the enzyme from spinach leaves and since its detection it has been the object of several investigations (Trown, 1965; Paulsen and Lane,

1966; Akoyunoglou and Calvin, 1963)* 2

1.2 Some Properties of RuDPcase and Similarity to

Fraction 1 Protein.

Wildman and Bonner (1947) investigating the soluble

cytoplasmic proteins of spinach leaves found that after

electrophoresis, two major groups of protein could be

distinguished. One of these behaved homogeneously

during electrophoretic and ultracentrifugation studies

and was called Fraction 1 protein. Subsequent studies

(Eggman et al. 1953) showed that Fraction 1 protein

exhibited a sedimentation constant of 18 to 19 Svedberg

units. The enzyme purified by Weissbach et al» (1956) was reported by them to behave as an almost homogeneous

protein on ultracentrifugation and electrophoresis and

its molecular weight was estimated to be about 300,000

and having a sedimentation coefficient of 17S. The

similarity of the physical properties of these two proteins was noted by Dorner et al. (1957) and later workers confirmed this observation and demonstrated that

RuDPcase activity was associated with the Fraction 1 protein (Lyttleton and T!so, 1958; Park and Pon, 1961;

Mendiola and Akazawa, 1964; Van Noort and Wildman,

1964; Trown, 1965; Thornber et al. 1967)* 3

1.3 Cellular Location of Fraction 1 Protein

It has been known for many years that chloroplasts isolated in an aqueous medium are leaky, i.e. a suspension of chloroplasts on standing will lose their outer membrane and virtually all the electron dense material (mainly protein) of the stroma (Kahn and von Wettstein, *196*1; and Jacobi and Perner, 1961). If however the chloroplasts are protected from swelling during isolation by a gentle method, a large proportion of Fraction 1 protein is found to be still within the chloroplasts (Lyttleton and Tfso, 1958; Park and

Pon, 1961).

Using non-aqueous medium it is possible to isolate chloroplasts which, when viewed under the electron microscope, are seen to have lost their outer membrane but retained their stroma contents. By means of this technique it has been shown that higher plant chloro­ plasts and chloroplasts isolated from the alga

Euglena gracilis contain most, and possibly all, of the

RuDPcase (Smillie and Fuller, 1959; Heber et al. 1963;

Smillie, 1963)* Furthermore, Ridley et al» (19^7) have produced results that indicate that the bulk of the

Fraction 1 protein occurs in the stroma, and is not 4.

associated with the lamellar structure.

1.4 The Mechanism of Action of RuDPcase

The stoichiometry of the reaction catalysed

(Reaction 1 ) as well a.s many of the properties of the

carboxylase have been known since the original study by Weissbach et al. (1956). Many of these findings

have been verified by subsequent workers, most recent

studies being those of Trown (1965) and Paulsen and

Lane (1966). Despite these studies, information on the

carboxylation mechanism or structural characteristics

of the enzyme is meagre. It is not evident from the

stoichiometry of the reaction which of the carbon -

''^rbon bonds are cleaved in ribulose diphosphate because of the formation of two identical products.

Cleavage could occur at the C2 - C3 or C3 - C4 bond.

That cleavage occurs at the C2 - C3 bond was shown by

Mullhofer and Rose (1965). Having established this

fact a possible mechanism can now be written in which it is envisaged (Calvin, 1956; Rabin and Trown, 1964)

that an enolization (Step 1 of Reaction 2) is followed by addition of CO^ and a shift of the double bond to

form a 6 carbon $-keto acid (Step 2 of Reaction 2) which then is cleaved at the C2 - C3 bond to produce 5.

2 molecules of phosphoglyceric acid.

The essential enolization step demands some attention. The proton lost from the 3 position of the substrate (Step 1 of Reaction 2) does not contribute to

the formation of product. Experiments by Fiedler et al.

(1967) using ribulose diphosphate labelled with tritium in the 3 position showed that 98$ of the radioactivity was found in the water, less than 0.1$ being in the phosphoglyceric acid. Enolization type reactions are often characterised by the fact that they undergo isotope exchange reactions with protons of the medium. Hurwitz et al. (1956) and Fiedler e t al. (1967) have shown that the rate of exchange of proton in the absence of CO^ is negligible. The mechanism as outlined in Reaction 2 does not envisage a CO^ requirement for proton exchange at C3, however it is possible that the CO^ causes some conformational alteration of the enzyme which allows the enolization step to proceed.

In opposition to this possible mechanism there is a report by Rabin and Trown (1964). They produced spectral evidence for an interaction between ribulose diphosphate and enzyme in the absence of Mg^+ or CO^.

They interpreted this result as due to an enolization reaction involving an essential sulphydryl group on the 6

enzyme. If this be the case it is necessary to postulate that in the absence of CO^ the enolization step occurs but that the proton is not free to exchange from its conjugate acid group (AH+) on the enzyme unless a subsequent reaction, which depends on CO^ being present in the reaction mixture, takes place. Reaction 2 shows a mechanism in which it is considered that only one enzymic site (a) undergoes complete exchange with the medium protons, H+m, in the conjugate acid form prior to proton transfer. Alternatively a second enzymic site BH+m which is independent of site A could function in the protolytic steps. Investigations on the mechanism however have been hampered by the failure to detect partial reactions and by the lack of conclusive information on specific amino acid residues at the active site(s) that are functionally involved in binding or catalysis. The number of reacting moieties in the mechanism has also contributed to the complexity of the problem.

The ribulose diphosphate binding site has had the most concerted study made of it but even in this case the nature of the reaction is uncertain. Trown and

Rabin (1964) suggested, from a study of the protective effect of ribulose diphosphate on the kinetics of 7.

REACTION Z

H COP *“■ BHm 11 y$ HOOP2 | c=o 1 cN 1 9-0" 2 H90H COH HCOH 1 \—ah+ HCOH v. H£OP h/op /„ //■✓/// n f**

Hm H,COP ■BHm n H,C0P n'C-C-O" *+* H ‘Mkr +-° c=o q-o H HCOH LAHm HCOH HaC0P H^OP 'TmTTTTm

HaC0P HQC-C-OH f4m 8. carboxylase inactivation by iodoacetamide, that two sulphydryl groups per mole of protein are involved in binding the diphosphate. This has been questioned by

Akoyunoglou1s group because protection from iodoaceta- mide inactivation can be afforded by carbamyl phosphate as well as different sugar phosphates, yet these compounds are not competitive inhibitors of the ensyme (Argyroudi-

Akoyunoglou and Akoyunoglou, 19^7? Akoyunoglou et al.

1967)* At the present time no decisive answer can be given as to whether a sulphydryl residue is essential in the binding of the diphosphate. The possibility that a Schiff base between a carboxylase amino group and ribulose diphosphate is involved in the carboxylation mechanism has been considered by Wishnick and Lane (1969) but attempts to demonstrate this type of reaction have not been success­ ful and although the participation of a Schiff base intermediate in the carboxylation mechanism cannot be excluded it seems unlikely. The answer to this intriguing problem will probably only be available when a small proportion of the protein containing the diphosphate bound in a stable form is separated and analysed.

Our knowledge about the reactions involving the 9.

other substrate, viz. C0o or HCO ^ , is also limited.

Akoyunoglou and Calvin (1963) were able to isolate a 2 + C0o enzyme complex, the formation of which was Mg dependent. Subsequent work (Akoyunoglou et al. 1967) showed that the CO^ was present in a yellow coloured peptide that could be isolated from a tryptic digest of the C0o enzyme complex. Promising though this result seems it was obtained using only a partially purified enzyme preparation. Since it is possible to prepare

Fraction 1 protein that is not coloured (Paulsen and

Lane, 1966; Sugiyama et al. 1968) the coloured pigment in the preparations of Akoyunoglou et al. (1967) must be considered a contaminant and the results obtained treated with reserve.

The effect of bicarbonate concentration on the reaction kinetics has recently been investigated

(Sugiyama et al. 1968a). A kinetic analysis of the

CO^- fixation reaction catalysed by RuDPcase prepared from spinach leaves showed that the reaction rate

(^COg- fixation) versus NaHCO^ concentrations curve deviated from the Michaelian type. A sigmoid curve was obtained which strongly suggests a homotropic interaction of NaHCO^ in the CO^- fixation reaction, indicating more than one substrate binding site inter- 10. acting co-operatively in the enzyme molecule. An analysis of the data using the Hill equation, gave an interaction coefficient (n) of approximately 2 . The authors suggested that RuDPcase is a regulatory enzyme, in which NaHCO^ functions as a homotropic effector subs tance.

Several studies have shown unequivocally the requirement of Mg^+ for the RuDPcase activity (Veissbach et al. 1956; Racker, 1957; Paulsen and Lane, 1966). Veissbach et al. (1956) were also able to observe some activation with Ni^+, which was as effective as Mg^+, Co^+, Mn^+ and Fe^+. Sugiyama et al. (1968a) however observed activation only by Mg 2 + . These latter workers have begun studies in an effort to elucidate the 2 + mechanism whereby Mg activates the carboxylase reaction. Pon et al. (1963) reported that preincubation of the

^ - carboxylase with bicarbonate in the presence of Mg induced greater activity and indicated that an enzyme - 2 + Mg -CO^ complex is formed just prior to the carboxylation of RuDP,

The effect of increasing Mg^+ concentration on the carboxylase reaction is threefold (Sugiyama et al. 1968 a,b). Firstly the specific activity of the enzyme was increased. An increase in the Mg^+ concentration from _ 2 0 to 10 M increased the specific activity of the enzyme nearly ten-fold at pH Secondly, a decrease in the apparent Km(NaHCO^) value was noted. Raising the 2 + -2 concentration of Mg from 0 to 2 x 10 M brought about a four-fold decrease in the apparent Km(NaHCO^) value.

These results suggest that perhaps Mg^ + facilitate the binding of CO^ (or HCO ) to the enzyme molecule.

Thirdly, a marked decrease in the pH optimum was observed. 2 + In the absence of Mg the activity was low and the optimum pH was about 8.5» Increasing the Mg^+ concentra- tion to 10 M caused the pH optimum to shift to 6.5*

The authors suggest that this phenomenum indicates a possible conformation change of the enzyme molecule induced by the Mg^+ binding.

The low activity of RuDPcase has always been a perturbing problem in our understanding of the mechanisms involved in the carbon reduction cycle. It has consistently been observed that the activities of several enzymes, amongst which is RuDPcase, are not high enough to satisfy the rate of overall CO^ fixation observed in intact cells (Peterkofsky and Racker, 1961;

Latzko and Gibbs, 1969)* It is possible that some effector substance(s) in the chloroplasts modifies the enzyme with the result that the low affinity for CO^ is 12.

enhanced.

The above results by Sugiyama e t al. (1968b) show

2 + that Mg alters the affinity of the enzyme for CO^ '

and increases the activity of the enzyme, and lends

support to the hypothesis that RuDPcase is a type of

regulatory enzyme. It is indeed difficult to accurately

assess the physiological status of Mg^+ within the

intact chloroplast and its overall relationship to

photosynthetic CO^- fixation.

It is only recently that the presence of tightly

bound copper has been detected in homogeneous prepara­

tions of RuDPcase (Wishnick et al. 1969). They found

that copper was bound in a stoichiometric amount, vis.

1 gm atom of Cu^+ per mole of protein (560,000 gm) and

suggested that the enzyme had a specific site for copp^x which fulfilled some functional role. Earlier work

(Wishnick and Lane, 1969) had demonstrated that

cyanide was an effective inhibitor of RuDPcase, which bound stoichiometrically to the enzyme (1 mole per mole

of enzyme) only in the presence of ribulose diphosphate.

This inhibition could be explained by the formation of a stable complex between cyanide and the tightly bound copper at a site on the enzyme which becomes accessible when ribulose diphosphate is bound by the enzyme. 13.

Clearly, further investigations will be necessary to

define the possible functional role of the bound copper.

1.5 Structure of Fraction 1 Protein From Higher Plants

There has been numerous studies on the molecular

weight of Fraction 1 protein but there are differences

between the published values. Eggman et al. (1953)

reported a molecular weight of 375,000 which is consider­

ably less than more recent reports; 590>000, 620,000

(Lyttleton, 1956); 515,000, 559,000 and 560,000 (Trown,

1965); 557,000 (Paulsen and Lane (1965)» 585,000,

561,000 (Ridley et al. 1967); 570,000 (Steer et al. 1968).

Electronmicroscopic studies of Fraction 1 protein

from various plant sources have consistently revealed

a regular subunit .arrangement (Haselkorn et al. 1965;

Ridley et al. 1967 5 Sugiyama and Akazawa, 1967; Steer,1968;

Matsumoto et al. 1969). The original work by Haselkorn

et al. (1965) showed that the oligomeric structure of

the protein consisted of a cube having sides of about

120$. These workers suggested that the arrangement

of the subunits is such that there are three subunits along each edge and one in the centre of each of four

faces of the cube - making a total of 24 subunits. Steer et al. (1968) produced electronmicrographs of Fraction 1 14 protein obtained from Avena sativa in which the edges appear sometimes to have two indentations visible along an edge. This suggests that at least part of the molecule consists of a row of three subunits. These authors have discussed various models to explain the observed features of Fraction 1 protein but the overall gross structure and the disposition of the subunits within the molecule can not be stated with any degree of certainty at this time. The proposed models all assume the presence of twenty four identical promoters and this may not be the case (Rutner and Lane, 1967)*

1.6 Relationship of Structure to Enzymic Activity and

the Role of Sulphydryl Groups

Several studies related to furthering our under­ standing of the relationship of structure and catalyt4c activity of Fraction 1 protein have been made (Sugiyama and Akazawa 19^7; Sugiyama et al. 1967a,bJ Sugiyama et al. 1968 c ; Akazawa et_al_. 1968). These studies have involved the use of chemical dissociating agents such as urea and sodium dodecyl sulphate, selective proteolytic digestion and sulphydryl blocking reagents.

Sugiyama et al, (i968c) found that carboxylase activity 15

of Fraction 1 protein was reduced considerably in the

presence of relatively low concentrations of urea (2M) and was instantaneously inactivated when the urea

concentration was made to 4M. Pre-incubation of the protein with either ribulose diphosphate or NaHCO^ plus Mg^ + afforded protection from the urea-induced inactivation when the protein was subsequently exposed to

3M urea. The protecting effect of the substrates became less as the concentration of urea was increased. Poly­ acrylamide gel electrophoresis suggested that subunits had formed in urea-treated samples but that in the cases where pre-incubation had taken place the pattern obtained was more like that for the native enzyme. Similar results were obtained when sodium dodecyl sulphate was used instead of urea. a These results clearly show that maintenance of the structural integrity of the protein is essential for cataltic activity and that the binding of either ribulose diphosphate or NaHCO^ and Mg^+ induces a stable conformation which is then partially resistant to attack by these dissociating agents.

The exact role of sulphydryl groups in the protein is not clearly defined. Numerous studies with sulphydryl blocking agents such as PCMB and iodoacetamide have indicated an important role for sulphydryl groups in 16.

the protein. (Mayauden e t al. 1957; Rabin and Trown, 196;-

Sugiyama and Akazawa 19^7 5 Sugiyama et al, 1968,d;

Akoyunoglou and Akoyunoglou 1968.

The results of Sugiyama and Akazawa (1967) have

shown that there is a correlation between the loss of

enzyme activity and the number of PCMB titratable

sulphydryl groups. Ribulose diphosphate protects the

enzyme from PCMB inactivation but the number of total

sulphydryl groups that could be titrated by PCMB was

the same regardless of the pre-incubation of the protein with ribulose diphosphate. Similar results using

iodoacetamide as the sulphydryl blocking agent have been

obtained by Akoyunoglou and Akoyunoglou (1968). Because

of the large number of sulphydryl groups pres3nt in the native enzyme, approximately 96 half-cystine residues per mole of protein, it is possible that only a small numbo-^

of sulphydryl groups are involved in binding the ribulose

diphosphate and these escape detection because of the

limits of the method for determining the number of sulphydryl groups that have reacted.

Proteolytic digestion of the protein, resulting in a decline of the RuDPcase activity, was found to be enhanced by either PCMB or iodoacetamide treatment.

Sugiyama et al. (1967b) has interpreted these results t^ 17.

indicate that some sulphydryl groups are involved in maintaining the tertiary structure of the protein and that reaction of these sulphydryl groups brings about a loosening or unfolding of the structure in the vicinity of the enzymic site.

The distribution of the important sulphydryl groups is not known and indeed it is a challenging problem to fully characterise the sulphydryl groups in Fraction

1 protein.

1.7 Studies on RuDPcase from Sources other than the

Higher Plants

The existence of RuDPcase in a number of photo­ synthetic micro-organisms has been known for a considerable time (Fuller and Gibbs, 1959; Fuller et al. 1961 ; Peterkofsky and Racker, 1961 5 Smillie et al. 1962). However, it is only recently that some of the physical properties of RuDPcase isolated from these organisms have been reported. Dorner et al. (1958) demonstrated that Fraction 1 protein was present in

Chlorella vulgaris. Sugiyama et al. (1969) isolated

RuDPcase from Chlorella ellipsoidea and found that it was similar to spinach leaf RuDPcase on the basis of its large molecular size (820= 18.4) and kinetic 18.

properties but there existed an immunological difference

between the two enzyme preparations. Subsequently a

small difference in the molecular weights of the two

enzymes was reported (Akazawa et al. 1969)* The RuDPcase enzyme from the green alga Chlamydomonas

reinhardi was studied by Kieras and Haselkorn (1968) who

reported that it exhibited a sedimentation coefficient

of 16-18S. These authors further stated that preliminary

electron microscopic examination revealed it to be a globular molecule and the molecular weight could be assumed to be about 500?000. The other green alga that has been studied is Euglena gracilis (Anderson et al. 1968). These workers observed that extracts of Euglena grown under phot©syn­ thetic conditions possessed a RuDPcase enzyme which had a sedimentation value of 19S. Extracts of Euglena that had been grown in the dark did not possess luch a c omponent.

Two studies on RuDPcase from the blue-green algae have been reported (Kieras and Haselkorn, 1968; Anderson et al. 1968). Kieras and Haselkorn (1968) reported that the enzyme from Plectonema boryanum had a sedimentation value of 16-18S while Anderson et al.

(1968) reported a sedimentation value of 19*5S for the 19.

RuDPcase from Anacystis nidulans.

Studies have also been made on RuDPcase isolated from both photosynthetic and chemosynthetic bacteria.

It is in these organisms that wide molecular diversities of RuDPcase occurs.

In marked contrast to the enzymes from plants and algae the RuDPcase isolated from the bacterium

Rhodospirillum rubrum was found to be small, having a molecular weight of 120,000 (Anderson et al. 1967)•

This was further verified end the study extended to include Rhodopseudomonas_spheroides Rhodopseudomonas polustris and Chromatium. The molecular weights for

RuDPcase prepared from the two Rhodopseudomonas species was 360,000 while that for Chromatium had a molecular weight nearer to that of plants (Anderson et al. 1968, Kieras and Haselkorn, 1968).

Akazawa et al. (1969) has presented data also on the properties of RuDPcase prepared from the two bacteria Rhodopseudomonas spheroides and Rhodospirillum rubrum which is at a slight variance to that by Anderson et al. (1968). Molecular weights for the RuDPcase of these two bacteria are stated by Akazawa et al. (1969) to be 260,000 and 68,000 respectively.

Characterisation of RuDPcase from chemosynthetic 20 bacteria has also been attempted. MacElroy et al. (1968) investigated RuDPcase from Thiobacillus thioparus and

Thiobacillus neapolitanus and found that while its sedimentation value was approximately 17S, in freshly prepared extracts an additional peak of carboxylase activity with a lower molecular weight was observed.

It is worth noting that these preparations of RuDPcase contained phosphoribulokinase activity, an enzyme which is usually separable in higher plant preparations.

Two other chemosynthetic bacteria, Hydrogenomonas facilis have been used to investigate the properties of RuDPcase. The molecular weights for these two enzymes are 515*000 and 551*000 respectively. Furthermore, it has been stated that the average subunit molecular weights were 40,700 and 38,000 respectively (Kuehn and McFadden, 1969). Thus, as shown by Anderson et al. (1968) we have an instance where a reaction is catalysed by three different molecular forms and it is of extreme interest to chemically characterise these proteins and perhaps elucidate the evolutionary development of the mechanisms of photosynthetic CO^ fixation. It is also known that erythrocytes contain RuDPcase but little is known about its molecular weight (Fortier et al. 19&7)* 1.8 Comparison of Some Kinetic and Physical Constants of RuDPcase from Different Sources

Source of Enzyme Sedimentation Molecular Km(M) Km(M) Co fact or(s) Inhibitor(s) Re ference Value Weight RuDP HC0 Km(M) Ki (M)

5 -4 -2 Spinach Leaf 1 7a ca 3x10 2.3x10 1.1x10 Weissbach et al. 1956.

18.6a 5.15x10^ Trown, 1963.

-4 -2 2+ - 3 2 1.0a 5.57x10^ 1.2x10 2.2x10 Mg 1.1x10 J Pi(K+)C4.2x10"3 3 Mn2+3.9x10"5 (NH4)2S04C8.1x10“ Paulsen & Lane, 1966. 3-PGAd8.3x10“3

3-PGAe9.3x10“3

Chinese Cabbage -4 -2 Leaf 17.oa 5.1x10^ 1.0x10 3x10 Kieras & Haselkorn, 1968

Chlorella ellips oidea 18.4a 4.7x10^ Akazawa et al. 1969.

Euglena gracilis I9b 6.6x105 Anderson et al. 1968.

Chlamydomonas reinhardi ca I8b ca 3x105 Kieras & Haselkorn, 1968

Plectonema boryanum ca I7b ca 3x10^ Kieras & Haselkorn, 1968

Anacystis nidulans 19.5b 6.3x10^ Anderson et al. 1968.

Chromatium I8b 6.6x1o5 Anderson et al. 1968.

Chromatium Strain D. ca I8b ca 3x10^ {k- 1.0)x10-Z+ (9.0±3.0)x10-2 Kieras & Haselkorn, 1968

Thiobacillus thioparus 1 7.5b 3x10^ McElroy et al. 1968.

Thiobacillus neapolitanus 1 7 - 5b 3x10^ McElroy et al. 1968. 22

Source of Enzyme Sedimentation Molecular Km(M) Km(M) Cofactor(s) Inhibitor(s) Reference Value Weight RuDP HCO“ Km(M) Ki (M)

Hydrogenomonas -4 eutropha 20 5.15x10 1.25x10 Kuehn & McFadden, 1969*

Hydrogenomonas -4 facilis 5.5xio5 2.35x10 4. 16x10"'3 Mg2+1.39x10“3 Pi(K+)( 10.Ox 10 3

(nh4)2so4c5.1x1o”3

Na2S04°5.1x10~3 Kuehn & McFadden, 1969*

3-PGAdi4.7x10“3 3-PGAei5.0x10~3

Rhodops eudomonas palus tris 12b 3.6x103 Anderson et al* 1968.

Rhodopseudomonas spheroides 14.5b 3.6x103 Anderson et al• 1968.

2.6x103 Akazawa et al. 1969*

Rhodospirillum rub rum 6.2b 1.2x103 0.8x10"^ 12.5x10-3 Anderson et al» 1968.

0.68x103 Akazawa et al. 1968.

b Relative sedimentation values determined by sucrose density centrifugation c Competitive inhibitor with respect to ribulose 1,5-diphosphate d Competitive inhibitor with respect to HCO^ e Noncompetitive inhibitor with respect to ribulose 1,5-diphosphate. 23.

1.9 Aim of This Research Project

It was felt that many of the observations made in

the numerous studies, that have been outlined in this

Chapter, could be better interpreted if a more detailed

description of the structure of the protein was

available.

At the commencement of the project there was a

paucity of information in the literature regarding dissociation studies on Fraction 1 protein. Trown

(1965) briefly mentioned that RuDPcase from spinach was broken down into subunits with a sedimentation coefficient of approximately 2S by treatment with sodium dodecyl sulphate.

The large molecular weight of Fraction 1 protein has hindered concise biochemical studies. However the techniques employed in protein chemistry have now reached a high degree of sophistication and standard­ isation and there is an ever growing tendency to tackle larger and larger proteins.

This study then was commenced in an effort to clarify the position regarding the nature of the subunit structure of the protein and to begin definite chemical investigations on the isolated subunits. 24

CHAPTER 2. METHODS AND MATERIALS

2.1 Plant Material

Young leaves from spinach beet (Beta vulgaris)grown out of doors either at local market garden or on the university site were used,

2_,_2_Preparation of Chloroplasts and Isolation of

Frac tion_1 Protein,

Chloroplasts were isolated from the leaves of spinach beet using essentially the method of Ridley,

Thornber and Bailey (19^7)»

Freshly collected leaves of spinach beet were deribbed and about 160 gm of the remaining tissue was homogenised in a Waring blender with about twice the volume of 0.5M sucrose - G.05M KHCO^ (pH 7«5) ioi'

30 seconds. The resultant slurry was filtered through two layers of cheese cloth and the liquid centrifuged at 50 g for 10 mir The supernatant was further subjected to centrifugation at 600 g for 20 min and the pellet consisting of chloroplasts kept. The isolated chloroplasts were ruptured in 0„01M Tris - 0.1M KC1 -

0.001M EDTA - 0.01M 2-mercaptoethanol (pH 8,3) and left for 30 min. The mixture was centrifuged at 38,000 g 25

Tor 20 min and the supernatant made 35$ saturated with

ammonium sulphate. ’’Saturated” being defined as 70 gm

of ammonium sulphate per 100 ml of original solution

(Trown 1965)* The pH of the solution was kept between

7.5 and 8.0 by additions of concentrated ammonia.

After 1 hr the precipitate was removed by centrifuging

at 38,000 g for 10 min. The supernatant was made 45$

saturated with ammonium sulphate and left for 1 hr.

The precipitate was collected by centrifuging at 12,000g

for 10 min and dissolved in the minimum amount of

homogenising buffer (3-5 ml). The solution was either

dialysed against homogenising buffer overnight and then

subjected to gel filtration through a Sephadex G-200

column or placed directly onto the column. For gel

filtration a column 44 by 2.5 cm was used. The

temperature was maintained at 0-4°C throughout the

preparation.

Elution was effected using the homogenising buffer,

and fractions monitored for protein at 280 mu and

ribulose-1,5-diphosphate carboxylase activity. The peak corresponding to carboxylase activity was collected

and aliquots taken for electrophoresis and protein estimation. Samples for electrophoresis were dialysed against electrophoresis buffer, whilst samples for protein estimation were dialysed against distilled water. 26

The Sephadex G-200 chromatography step was repeated if subsequent electrophoresis revealed other components

than Fraction 1 protein or the sample was contaminated with "nucleic acid material". The "nucleic acid material" eluted just before Fraction 1 protein but trailed into

tubes containing Fraction 1 protein.

In later experiments when large quantities of

Fraction 1 protein were required, a whole leaf homogenate was used instead of isolated chloroplasts. In this case a Sephadex column 44 x 5*0 cm was used and several cycles through the column were necessary to ensure homogeneity.

2.3 Concentrating Protein Solutions

When it was necessary to concentrate the protein solution (e.g. recycling through Sephadex columns) this was accomplished either by making the solution saturated with ammonium sulphate and centrifuging off the precipitated protein or by subjecting the solution to pressure dialysis (Berggard 1961) .

Freeze drying of Fraction 1 protein was performed in the presence of an amount of urea so as to facilitate subsequent solution of the material and to give an 8M urea solution with the added water 27

2.4 Preparation of Reduced and Carboxymethylated

Fraction 1 Protein.

The Fraction 1 protein freeze-dried in the presence

of urea was reduced and carboxymethylated by a method

similar to that of Crestfield, Moore, and Stein (l$)63)»

To samples of freeze-dried Fraction 1 protein (30-70 nig)

the correct amount of water was added to make the final

solution 8M with respect to urea. The flask was flushed

out with nitrogen and redistilled mercaptoethanol was

added to give a final concentration of 0.14M. The pH

was taken to 10.5 with 5M potassium hydroxide and left under nitrogen for 3 hours at room temperature. After

3 hours iodoacetic acid (2.68 gm per 1.0 ml mercapto-

ethanol) was dissolved in 3M Tris-HCl (pH 8.5) (8.3 ml

3M Tris-HCl per 2.68 gm iodoacetic acid), added to the

flask, and the pH adjusted to 8.5 with 5M potassium hydroxide if necessary. The alkylation was allowed to proceed 5-10 min at pH 8.5. The reagents were removed by dialysis for 3-4 hours against 8M urea.

The pH was raised to 10 by the addition of concentrated ammonia and the reaction mixture submitted to gel filtration through Sephadex G-200 containing 8M urea buffer at 4°C.

To ensure reasonable flow rates through the Sephadex 28

column, the Sephadex G-200 was first sieved and only that

retained by a 300-mesh sieve was used to pack a column

2.3 x 44 cm. The proteins run ahead of the buffer in

which they are applied and consequently are exposed to

at least 1 day old 8M urea solutions which would contain

a considerable amount of cyanate. To prevent carbamy-

lation of a and e amino groups on the protein by cyanate

during passage down the Sephadex column the following

buffer of high pH was used, 8M urea - 0.05M Tris-HCl -

0.001M EDTA - 0.1M KC1 - 0.2M ammonia (pH 10.0) (Thompson

and O'Donnell 1966). The fractions were monitored for

protein at 280 mu. The required fractions were pooled

and dialysed against water, 0.1M KC1 - 0.001M borate,

and finally water, before freeze-drying.

The S-carboxymethyl (SCM) derivates of bovine

plasma albumin, 3 -lactoglobulin, lysozyme, and ovalbumin were prepared in essentially the same manner as for

SCM-Fraction 1.

2.5 Preparation of Maleyl-carboxymethyl Fraction 1

Protein.

After carboxymethylation of the Fraction 1 protein as described above, the reaction mixture was dialysed against 0.4M boric acid - 0.001M EDTA adjusted to 29

pH 8.5 with NaOH. Maleylation of the protein was then carried out by

a method similar to that of Butler et al. (19^9)* To approximately 100 mg of carboxymethylated protein,

300 mg of resublimed maleic anhydride in dry acetone

was added dropvise over a 15-20 min period. The pH

of the solution was kept between 8.5 - 9»0 by the addition of 1M NaOH. After maleylation of the protein the

solution was freeze dried and then either dialysed against 0.01M-Tris-0.001M EDTA pH 8.3 to prepare the sample for Sephadex chromatography, or against dilute ammonia and freeze dried. A column of Sephadex G-100 (2.5 x 88 cm) was used to fractionate the dissociated products.

2.6 Removal of Maleyl Groups

When removal of the maleyl groups from the amino groups of the protein was required the protein was suspended in 0.2M acetic acid, adjusted to pH 3-5 with ammonia, and heated at 60° for 8 hr (Butler et al. 1969). The reaction was stopped by freeze-drying. 30

2.7 Gel Electrophoresis

2.7 (i) Polyacrylamide Disk Electrophoresis

During the course of this research both 7.5$ and

5.0% polyacrylamide gels were routinely employed. The

formulation of the gels were as published by Davis

(1964), the stock solution of acrylamide was diluted

appropriately with distilled water in the case of the

5.0$> gels. The buffer system employed was the

glycine-Tris discontinuous buffer of Davis (1964)

which gives a running pH of about 9»5» Gels which were homogeneous in respect to the buffer were also

employed. Two buffer systems were used

(a) 0.05M Tris-HCl (pH 8.3);

(b) 0.05M 2 methyl-2 amino propanol-HCl (pH 10.3)•

All buffers contained 0.001M sodium thioglycollate to prevent possible oxidation of protein SH groups. As an added precaution gels run in the homogeneous buffer systemswere equilibrated before use by application of a potential difference of 120V for 25 min.

Urea gels were prepared to give a final urea concentration of 8M. Electrophoresis in urea was performed using the glycine-Tris buffer.

Electrophoresis was carried out for approximately

60 min at 120V and 4-5 mA per tube. After electro- 31

phoresis the gels were stained for 15 min in a solution of 1% amido black in 7$ acetic acid and destained by washing for 24 hr with numerous changes of 7^0 acetic acid.

2.7 (ii) Polyacrylamide Electrophoresis in the

Presence of SDS Ten (10) percent acrylamide gels containing 0.1^o

SDS were used. The method of preparing the gels was essentially the same as Dunker and Rueckert ('{969) except "split" gels were not used. It was not possible to ensure that the wedge used in forming a "split" gel was absolutely leak proof. Protein samples were prepared for electrophoresis by first dissolving them at a concentration of about

2 mg per ml in 1 'jo 2-mercaptoethanol (v/v) 4M urea and about ^(/o SDS, and then left to stand at room temperature for approximately 1 hour. Electrophoresis was carried out at approximately 6-8 mA per tube for 4 hr.

After electrophoresis the gels were stained with

Coomassie blue in a methanol, acetic acid mixture and subsequently destained by extensive washing in a 7*5^

(v/v) acetic acid-5.0$ (v/v) methanol solution (Veber and Osborn, 1969). 32

2.7 (iii) Slab Acrylamide Gel Electrophoresis

Slab acrylamide gel electrophoresis (Margolis and

Kenwick, 1968) was performed in a Gradipore electro­

phoresis apparatus using commercially available gels having a continuous acrylamide concentration gradient.

The gels had a concave gel gradient of 4-26^. The buffer system used was 0.05M Tris-HCl - 0.01M 2- mercaptoethanol (pH 8.5) and gels were equilibrated with this buffer by a pre-run of 2 hours at 50V/20 mA.

After loading the proteins electrophoresis conditions were 75V/25 at the beginning of the run, no attempt was made to keep either voltage or current constant over the duration of the run. The current usually dropped to 15 niA after 10 hr. The buffer was cooled by circulating it through a heat exchange plastic bag immersed in ice.

At the completion of the run the gels were sliced horizontally into two pieces and the cut surface immediately stained with a saturated solution of Coomassie blue in 15^ TCA (Massaro and Market, 1968). The protein on the cut surface was in immediate contact with the dye and stained rapidly. The gels were destained by numerous washings with 7$ acetic acid. 33

2,7 (iv) Isoelectric Focusing

Isoelectric focusing was performed as described by

Wrigley (1968). In this technique polyacrylamide gel

was used instead of a sucrose density gradient to

stabilise the pH gradient. The ampholytes covered the

range pH 7-10* Fraction 1 protein was first electro-

phoresed, as previously described, in a 7*5$ acrylamide

gel. The location of Fraction 1 protein was found by

treating the gel with 20% TCA which precipitated the proteins in the gel. From another gel which had not been treated with 20% TCA the portion containing Fraction

1 protein was sliced out of the gel and applied to the top (cathode) of an isoelectric focusing column. The anodic and cathodic vessels were filled with 0,2%} sulphuric acid and 0, k%o ethanolamine respectively. The gels were run for 3-5 hours at 1-2 mA per tube.

The gels were removed and ampholytes washed out of the gel by extensive washing with 5% TCA for at least

24 hr. The fixed proteins were subsequently stained with amido black. The pH gradient was determined by slicing a gel that had not been washed with 5$ TCA, into pieces, immersing them into 3 ml of water overnight and determining the pH of the resultant solution. 34

2.8 Ribulose-1,5-diphosphate Carboxylase Assay

This determination involved the measurement of

acid-stable radioactivity produced in the reaction 14 / between ribulose-1,5-diphosphate and NaH CO^ (Rabin

and Trown 1964a), The components of the reaction mixture

and the procedure followed were the same as that used

by Ridley, Thornber, and Bailey (1967) except that 1 4 in some experiments only 1 uCi of NaH CO^ was used.

Aliquots of the acid-stable radioactivity were counted

in a Nuclear Chicago gas-flow counter.

2,9 Preparation of the S-Sulphenylsulphonate Derivative

of Fraction 1 Protein,

Fraction 1 protein (ca 10 mg) in 0.01M Tris-HCl-

0.001M EDTA-0.01M mercaptoethanol (pH 8,3) was treated

with 50 nig of sodium tetrathionate (Na^S^O^ . ^H^O)

prepared by the method of Gilman et al. (1946). The pH

of the solution was kept between 7-8. The reaction mixture was allowed to stand for 30 min at room

temperature and then applied to a column of Sephadex G-25

(30 x 1.5 cm) equilibrated with 0.05M Borate - 0.001M

EDTA (pH 8.5) at 4°C and developed. Tubes containing protein were pooled. 35

2.10 Edman Degradations

Degradations were carried out following the

procedure of Blomback et al. (1966) after coupling in

60% pyridine-water containing 6^ N-ethylmorpholine. A

portion of the ethyl acetate extract of the phenyl-

thiohydantoin was chromatographed on Eastman Chromogram

silica gel sheets with fluorescent indicator (jeppson

and Sjoquist 1967). Another aliquot was dried and

heated with 6N HC1 in vacuo at 150°C for 24 hr to

regenerate the amino acid which was then identified

on a Beckman amino acid analyser model 120C (Van Orden

and Carpenter, 1964).

2.11 Hydrazinolysis

The protein (ca 0.3 umole) and catalyst, 60 mg

Bio-rex 70 H+ form (Braun and Schroeder 1967) were weighed into a Pyrex test tube and then constricted.

The material was dried by heating for 24 hr in vacuum

in an Abderhalden pistol at 78°C. Anhydrous hydrazine

(0.5 ml, 97% 1 Matheson, Coleman and Bell Co.) was

added by a drawn out funnel through the constriction,

and the tubes sealed under vacuum.

The tubes were left for 48 hr in an 80°C oven.

After reaction, the tubes were opened and carefully 36

lyophilized to remove excess hydrazine. The residue was

dissolved in water, transferred to a 5- by 1-cm Bio-rex

70 column, and the amino acids eluted with approximately

30 ml of water. The eluate was lyophilized, dissolved

in 3 ml of pH 2.2 buffer, passed through a Millipore

filter, and a sample applied to the long column of the

amino acid analyser.

This method of separating amino acids from the

hydrazides (Blackburn and Lee, 1954; Silman, Cebra, and

Gival, 1962) will not detect C-terminal basic residues

which require special procedures (Braun and Schroeder,

1967).

2.12 Isolation of Blocked N-terminal Peptides

The protein was dissolved in 0.2M KHCO^ buffer

pH 8.2. Pronase was added at an enzyme/protein ratio

1.20 (w/W) and digestion carried out at 37°C for 3-4 hr.

The digest was then treated in one of two ways.

A column 2 x 30 cm of Dowex 50 X2,H+ form, 200-

400 mesh) was poured and washed well with water at 4°C.

The digest was run onto the column and washed through with water at 4°C (Ikenaka et al. 1966). Peptides with

the amino end blocked (provided it has no positively charged side chains) will not bind onto the column and 37

will be present in the water eluant.

The second way in which the digest was treated

was that of Wilkinson, Press and Porter (1966). The

digest was treated with fluorodinitrobenzene and

stirred vigorously for another 3 hr to block all free

amino groups. The reaction mixture was acidified to

pH 2 by the addition of 99$ (v/v) formic acid. Most of

the DNP-peptides and excess fluorodinitrobenzene were

removed by several extractions with ethyl acetate and

ether. The remaining solution was concentrated and

then passed through a small column (8.0 x 0.5 cm) of

talc (Sanger, 19^+9) in 1N acetic acid. The DNP-peptides,

but not other peptides, were retarded. The eluant was

concentrated and passed through a Dowex-50 column as be fore.

Aliquots of the water wash from the Dowex-50 columns were dried and hydrolysed in 6N HC1 for 24 hr at 110°C, As well, paper electrophoresis at pH 3*5 at 4000V for 45 min was performed. The peptides were detected by the chlorine/tolidine/KI technique

(Reindel and Hoppe, 195^0 after lightly spraying the paper with 0.05M sodium tetraborate solution (0!Donnell and Thompson, 1964). 38

2.13 Tryptic Digestion and Peptide Mapping

Digestions with TPCK - trypsin (Worthington) were carried out with 1 $ protein solutions in 1 $ ammonium carbonate - ammonia solution (pH 8.7) for 4 hr at 37°»

The ratio of trypsin to protein was 1:50 W/W.

The reaction was stopped by freeze-drying and residue dissolved in electrophoresis buffer (50 ul/2 mg) and centrifuged. The soluble peptide mixture was applied to Whatman 3MM paper and electrophoresis carried out in pyridine: acetic acid: water (25:1:225 V/V) pH 6.4, at 3000V for 45 min under Varsol in a Savant ionophoresis unit. The 1 inch strip containing the peptides was sewn onto a second sheet of paper and chromatographed in n-butanol-pyridine-acetic acid-water

(15:10:3:12 V/V - Hill et al. 196O; Thompson et al.

1969)* After drying, the papers were dipped in a 0,2% solution of ninhydrin in 95$ ethanol and heated at 55°C to develop the colour.

2.14 Amino Acid Analyses

These were carried out using a Beckman amino acid analyser, model 120C, and hydrolysates prepared under vacuum (Crestfield, Moore, and Stein 1963)* In the analysis of native Fraction 1 protein samples were 39

hydrolysed in redistilled 6N HC1 at 110°C for times of

12, 24, 48 hr. Tryptophan was determined from the tyrosine: tryptophan ratio (Beavan and Holiday 1952).

The cysteic acid content was determined from samples of Fraction 1 protein oxidised with performic acid following the method outlined by Thompson and OlDonnell

(1959) and 0lDonnell and Thompson (1962). Extrapolation of amino acids destroyed during hydrolysis, or which were slow to be liberated, followed the standard

procedure (Hill 1965). In most other analyses a standard time of 24 hr was used.

2.15 Carboxypeptidase Digestion

2.15 (i) Carboxypeptidase B Digestion The protein (0.1-0.2 umoles) was dissolved in 1 ml

of 0.2M N-ethylmorpholine acetate (pH 8.5) buffer and 0.1 ml (0.5 units) of carboxypeptidase B (Worthington DFP) in water was added. Incubation was for 4 hr at 37°C and the reaction stopped by freeze drying. The

residue was dissolved in pH 2.2 citrate buffer and a

suitable aliquot analysed on the amino acid analyser.

2.15 (ii) Carboxypeptidase A Digestion

Crystals of DFP-carboxypeptidase A (Worthington) 40

were brought into solution following the procedure outlined by Ambler (1967) adapted from Potts et al.

(1962). The protein to be digested was dissolved in

0.5M sodium phosphate buffer (pH 8.5) and sodium dodecyl sulphate. Sufficient water was added to reduce the sodium phosphate and sodium dodecyl sulphate concen­ trations to 0.2M and 0.1$ respectively. An aliquot of enzyme (0.1 mg) was added and the reaction mixture heated for 2 hr at 37°C. The reaction was stopped by the addition of Dowex-50 (X16,H+ form) to the mixture and the products of the reaction separated as outlined by Ambler (1967)*

2.16 Titration of Sulphydryl Gro~7

Titration of the SH groups of native Fraction 1

^rotein with 5>51-dithiobis(2-nitrobenzoic acid) was carried out according to the method of Ellman (1959)»

Fraction 1 protein was first chromatographed on a

Sephade:: G-25 column (30 x 1.5 cm) to remove 2-mercapto- ethanol. The protein (ca 0,5 mg) in 0.01M Tris-HCl-

0.001M EDTA (pH 8.0) was incubated with various amounts of 5-5T>dithiobis(2-nitrobenzoic acid) in a 3 ml cuvette at room temperature. The course of the reaction was followed at 412 mu on a Shimadzu spectrophotometer. 41

— 1 — 1 A molar extinction coefficient (E4i2mu^ of 13,600M cm

for the anion of thionitrobenzoic acid was used to

determine the number of moles of SH groups reacted

(Ellman 1959).

Sodium dodecyl sulphate (0.3^) was included in

the reaction mixture when total SH groups were being

investigated.

Protein concentration was accurately determined

by hydrolysing a portion of the solution with 6N HC1

and analysing on the amino acid analyser.

2.17 Protein Determinations

Wherever possible salt-free protein preparations

were weighed out. For early work involving solutions

of Fraction 1 protein the protein was estimated by

the method of Lowry et al. (193T) using bovine plasma

albumin as a standard.

2.18 Chemicals

Acrylamide monomer, N,N-methylenebisacrylamide,

and ammonium persulphate were purchased from Cyanimid

Australia Pty. Ltd., and N,N,N!, N1 -tetramethylethylene-

diamine from Eastman Organic Chemicals; all were used without further purification. D-Ribulose-1,5-diphosphate 42

tetrasodium salt was purchased from Sigma Chemical Co.

All urea solutions were deionized by passage through a mixed bed of ion-exchange resin before adding the buffer salts. The 18/32 Visking cellulose tubing was boiled in changes of distilled water before use (Hughes and

Klotz, 1956).

The proteins used for calibrating the Sephadex

G-200 column were commercial crystalline samples and consisted of bovine plasma albumin (Sigma Chemical Co.), lysozyme, ovalbumin, and p-lactoglobulin (Pentex Incorp.).

Carrier ampholytes (pH 7-10) used in isoelectric focusing was made by LKB-Stockholm, Sweden.

Gradipore gels and apparatus were purchased from

Towns on and Mercer Pty. Ltd. Sydney.

Proteolytic enzymes used in this study have included, TPCK-trypsin, DFP-Carboxypeptidase A, carboxypeptidase D (Worthington) and Pronase (Kaken

Chemical Co. Japan).

5,5'-dithiobis(2-nitrobenzoic acid) was bought from

Aldrick Chemical Co. Inc. Milwaukee, Wis. U.S.A. 43.

CHAPTER 3. RESULTS

3.1 Preparation and Purity of Fraction 1 Protein

The preparation of Fraction 1 protein from the chloroplasts of spinach beet depended upon isolating intact chloroplasts from the leaves. The amount of soluble protein that was subsequently extracted from the isolated chloroplasts was extremely variable.

From 160 gm of spinach beet leaf material a yield of ca 30 mg of the 35-45$ ammonium sulphate fraction would represent a maximum value, with many preparations giving considerably less. This variability could be due to excessive leakage of the soluble proteins from the chloroplasts during their isolation. Ridley, Thornber and Bailey (1967) reported as well that the protein content in aqueous washes of the chloroplasts varied from 90-210 mg depending on the season.

The supernatant from the 38,000 g centrifugation step was usually green in colour but all of this coloured material was precipitated when the solution was made 35$ saturated with ammonium sulphate.

The precipiobtained in the 35-45$ ammonium sulphate fraction was slightly yellow in colour and a considerable amount of it could be dialysed away. 44.

Electrophoretic examination of the 35-45$ ammonium

sulphate using a 7•5$ acrylamide gel showed it to be

heterogeneous. As well as containing Fraction 1 protein

numerous other components, Fraction II proteins (Singer

et al. 1952), were present.

Sephadex G-200 chromatography was used to separate

the Fraction 1 protein from those components possessing

lower molecular weights. A typical elution curve of the

35-45$ ammonium sulphate fraction from the ruptured

chloroplasts is shown in Figure 1. Despite its large molecular weight, Fraction 1 protein was retarded to

some extent on the Sephadex column. The curve shown in

Figure 1 reveals only one main peak of protein material, however this was not always the case. In many prepara­

tions variable amounts of material were eluted just prior to the elution of the main peak. This material had a maximum absorption at 260 mu and no doubt represented "nucleic acid material". The elution pattern after the main protein peak was also variable; this was probably due to the variation in the concentration of soluble proteins in the chloroplasts over the different seasons. Fraction II proteins are eluted after the main protein peak.

The ribulose-1,5-diphosphate carboxylase activity 45-

Fig. 1

Tube number

Fig.1 - Gel filtration on Sephadex G-200 of 22 mg of a 35-45$ ammonium sulphate fraction of the chloroplast extract. Absorbance at 280 mu. Ribulose-1,5- diphosphate carboxylase activity. See text for assay procedure. Fraction size 6 ml; flow rate 24 ml/hr; eluent 0.01M Tris-HCl-0.1M KC1-0.001M EDTA-0.01M mercapto- ethanol, pH 8.3» Column 2.5 hy 44 cm. 46.

was associated with the main protein peak, but had a slight tendency to be closer to the trailing boundary, a result similar to that found by Ridley, Thornber and

Bailey (1967) and Thornber, Ridley and Bailey (1966).

Aliquots of the main peak when examined by polyacrylamide gel electrophoresis showed only one band to be present (Figure 2). Occasionally a very minor band with a slower mobility was noted. This probably represented the dimer of Fraction 1 protein (Trown 1965) caused by S-S bond formation due to a small amount of oxidation occurring in the gel, its presence therefore was ignored.

3.1 (1) Isoelectric Focusing of Fraction 1 Protein

The results of an isoelectric focusing experiment are shown in Figure 3» Gel (a) is a normal electro­ phoresis of a partially purified Fraction 1 protein preparation. The band corresponding to Fraction 1 protein in another unstained gel was cut out and placed directly on to an isoelectric focusing column which was then run as described in the methods. Gel (b) shows the result; the gel was stained with amido black, however it was noticed that the main protein band had already precipitated and was rather broad and uneven. 47

(a) lb) (c)

Fig, 2 - Polyacrylamide gel electrophoresis patterns of Fraction 1 protein at various pH values, (a) 0.05M Tris-HCl, pH 8.3; (b) O.O^M 2-methyl-2- aminopropanol-HCl, pH 10.3; (c) glycine-Tris discontin uous system, which gives a running pH of about 9.5* Approximately 60 ug of protein loaded in each case. 48.

I Q.

SLICE N°.

Fig. 3 - Isoelectric focusing patterns of Fraction 1 protein. (a) Normal electrophoresis of Fraction 1 protein in 7*5i° acrylamide. (b) Isoelectric focusing pattern of Fraction 1 protein cut out of an unstained (a). (c) Human carboxyhaemoglobin sample. (d) Same as gel (b) except acrylamide concentration reduced to 5$« pH calibration curve is for a 7•5$ acrylamide gel. 49.

As well as the major band occurring at about slice No.10

there was another band occurring at about slice No.5»

It was felt that this latter band could represent the

dimer of Fraction 1 protein that was being retarded

in the 7*5$ acrylamide gel. To ensure that the

protein was being separated on charge only a 5$

acrylamide isoelectric column was run and the result

is shown in gel (d). Only one main band was observed

indicating that a molecular sieving effect was being exerted on the protein in the 7*5$ acrylamide gel.

The pH calibration curve is for a 4 hr run in a

7.5$ acrylamide gel, from this it was found that

Fraction 1 protein precipitated at a pH of 6.3.

This value is slightly high because a run for 2 hr showed that Fraction 1 protein had already precipitated and there is a gradual drift in pH value over the next two hours. From a 2 hr run, Fraction 1 protein was found to precipitate at pH 6.

Because of this tendency of Fraction 1 protein to precipitate at a pH value of about 6, the isoelectric focusing technique was not a very successful check on the homogeneity of Fraction 1 protein.

The ultraviolet spectrum of the preparation resulted in a typical protein curve, it had E max at 280 mu with 50.

a very slight shoulder at 290 mu. The 280 mu/260 mu

varied slightly but the protein usually contained less

than 0.5$ contamination with nucleic acid. Figure 4

shows a typical ultraviolet scan of Fraction 1 protein

having a 280/260 ratio of 1.7> i.e. containing 0.3$

nucleic acid (Warburg and Christian 1941 ).

When a whole leaf homogenate was used the fraction

ation was essentially the same except that numerous

passages through the larger Sephadex column were necessary to separate Fraction 1 protein from the

remainder of the components. The fractionation in

this case was followed by removing aliquots from the

tubes and electrophoresing directly in acrylamide gels .

3.2 Amino Acid Analyses

The amino acid analyses of Fraction 1 protein is presented in Table 1. The results are calculated assuming a minimum molecular weight of 24,400 (Ridley et al. 1967) and from the number of umoles of amino acid recovered from the column, assuming an average residue weight of 110. All the common amino acids are present in the hydrolysates. Serine and threonine content were determined by extrapolation of the 51 •

E-> M CO w£ D < U M Eh CU O

WAVE LENGTH iny)

Fig.4 - Ultraviolet Spectrum of Fraction 1 protein in 0.1M sodium borate, pH 8.6 52.

individual time values to zero hour. Threonine was destroyed at a rate of about 7.5$ of starting content per 24 hours and serine at a faster rate of about 18*$ per 24 hours. These values are considerably less than those presented by Ridley et al. (1967); from their data the destruction of threonine and serine were about 20.7$ and 48$> respectively.

Spectrophotometrie analysis of the enzyme by the method of Beavan and Holiday (1952), Figure 5m gave a tyrosine/tryptophan ratio of 3.04 (Cf. 2.99 Thornber et al. 1965) and knowing the tyrosine content from an acid hydrolysis the tryptophan content of the protein was calculated to be 3.^ residues per mole

(24,400 gm).

Half-cystine in the protein was determined as cysteic acid in performic acid-oxidised samples and as carboxymethylcysteine in reduced carboxymethylated protein. 53

TABLE 1. Amino Acid Composition of Fraction 1 Protein Values are residues per minimum molecular weight of

24,400 (Ridley et al, 1967) and were calculated from the umoles recovered from the column assuming an average residue weight of 110.

Time of Hydrolysis (hr) Mean,extra­ Amino Acid 12 24 48 polated or Ridley max, value Data * Tryptophan 2.2 1.6 1.3 3.4 3.0 Lysine 13.2 13.0 13.9 13.9 12.0 Histidine 6.3 5. 1 5.6 5.7 5.9 Arginine 12.6 10.9 12.0 11.8 10.2 Aspartic acid 20.5 20. 1 19.5 20.0 17.8 + 00 Thi*e onine 14.4 13.7 12.9 • 14.2 Serine 9.4 8.4 7.3 10.3 + 10.9 Glutamic acid 25.2 24.7 23.8 24.6 21.0 Proline 12.8 12.6 12.2 12.5 11.8 Glycine 21.4 20.6 20.5 21.1 19.7 Alanine 20.7 20. 1 19.8 20.2 18.2 Valine 15.8 16.6 16.9 16.9 18.2 Methionine 3.7 3.3 3.2 3.7 4.0 Methionine sulphone 3.7^ Isoleucine 7.6 8.5 9.3 9.3 10. 1 Leucine 20.2 20.0 19.6 19.9 19. 1

Tyrosine 10.3 10.4 10. 1 10.3 9.0 Phenylalanine 10.4 10.6 10.3 10.4 10. 1 SCM-cys teine 4.4 4.5 4.4 4.4 Cysteic acid 4. or 4.0

*Data taken from Ridley et al. (1967) for comparison. +0btained by extrapolation to zero time assuming first order breakdown kinetics. 24 hr values only. 54.

260 2 80 300 320 340 WAVE LENGTH dim)

Fig* 5 - Spectrophotometric determination of tryptophan by the method of Beavan and Holiday (1952). The cuvette contained approximately 1.4 mg of Fraction 1 protein in 0.1N NaOH. 55.

A carboxymethyleysteine content of 4.4 residues per

mole (24,400 gm) was determined. Reaction with iodo-

acetic acid was judged to be complete because no

half-cystine occurred in the acid hydrolysates of the

treated protein. A cysteic acid content of 4.0 residues

per mole (24,400 gm) was calculated by normalisation to

leucine content.

Methionine was determined as methionine sulphone in

performic acid-oxidised protein and gave a value of 3*7

residues per mole (24,400 gm), this agreed with the

value obtained from unoxidised samples.

3.3 Titration of Native Fraction 1 Protein with

5.5T-Pithiobis (2-nitrobenzoic acid)

The reactivity of the SH groups of native Fraction 1

protein was first examined at various molar ratios of

5,5 *-dithiobis(2-nitrobenzoic acid) to enzyme. The

results are shown in Figure 6. The profile of the reactivity curve was characterised by an initial fast reaction of 2-4 SH groups depending on the concentration

of' 5 y 5 1 -di thiobis ( 2-ni trobenzoic acid). This was

followed by a very much slower reaction in which only a total of about 18 SH groups reacted in 30 min when the concentration of 5»5 *-dithiobis(2-nitrobenzoic acid) was 56. protein

mole

/ 2 0-04 sh

moles

time (mm)

Fig.6 - Reaction of Fraction 1 protein in 0.01M Tris-0.001M EDTA (pH 8.3) with 5,5»-dithiobis(2- nitrobenzoic acid) at three different levels of concentration of 5>51-dithiobis(2-nitrobenzoic acic). Curves 1, 2 and 3 correspond to 100, 200, 400 molar excess of 5>5 *-dithiobis(2-nitrobenzoic acid) respectively. Absorbance values at 412 mu were con­ verted into moles of SH groups using a molar o extinction coefficient of 13.6 x 10^ (see Materials and Me thods). 57.

a 400 molar excess. If however the protein is denatured

in 0,3% sodium dodecyl sulphate, there is a large

increase in the number of reactive SH groups which

suggests that the native structure determines the lack

of reactivity of a large number of SH groups. If one

assumes a molecular weight of 550»000 for the protein,

then the total number of SH groups reacted with the

5,51-dithiobis(2-nitrobenzoic acid) was found to be 107*

3.4 Studies on the Relationship of Enzymic Activity to

Sulphydryl Groups and Structure

The reactivity of the sulphydryl groups of

Fraction 1 protein was further studied with the reagent

sodium tetrathionate which reacts with the sulphydryl

groups of the protein molecule to form a sulphenyl

thiosulphate derivative.

R-SH + S406= ______R-SS203“ + S203= + H+

When Fraction 1 protein was reacted with sodium tetra-

thionate a loss in the carboxylase activity resulted.

Figure 7 shows the rate of inactivation. In this experiment some residual activity {3%) remained after

60 min, but in most experiments activity was non existent after 60 min of incubation with sodium tetra- thionate. 58.

RATE OF INACTIVATION OF RuDP CASE BY TETRATHIONATE

TIME (MIN.)

Fig. 7 - Rate of inactivation of RuDRcase activity by sodium tetrathionate, Approximately 10 mg of protein was reacted with Na^S^O^. 211^0, 50 mg, keeping the pH of the solution between 7-8. Aliquots were removed at the times indicated and the RuDPcase activity determined (lower curve). The upper curve represents the activity of a control to which no sodium tetrathionate had been added. 59

In an- attempt to quantitate the number of sulphydryl groups that had reacted with the sodium tetrathionate, a sample of Fraction 1 protein was reacted for 15 min at room temperature which would reduce the carboxylase activity some 90%, Excess sodium tetrathionate was removed by gel filtration on a Sephadex G-25 column and the protein alkylated with iodoacetic acid in the presence of 8M urea in the normal manner. Sulphydryl groups that had not reacted with the sodium tetrathionate now reacted with the iodoacetic acid and occurred as carboxymethylcysteine in a subsequent acid hydrolysate of the protein. Approximately 1 residue of carboxymethyl cysteine per mole (24,400 gm) was present in an acid hydrolysate. By difference this would indicate that

3 cysteine residues per mole had reacted with rhe sodium tetrathionate in the 15 min incubation period.

Examination of the patterns obtained in a starch gel electrophoresis of both native and the sulphenyl thio­ sulphate derivative did not reveal any significant differences.

The reversibility of the reaction of Fraction 1 protein with sodium tetrathionate was next studied.

Incubating the inactive sulphenyl thiosulphate derivative of Fraction 1 protein with cysteine for 2 hr 60.

at 37°C restored fully the carboxylase activity of

Fraction 1 protein. Table II shows the results of such

an experiment. TABLE II Effect of Incubating Sulph nyl Thiosulphate Derivative With Cysteine

CO FIXED SYSTEM ^ * C.P.M. /mg protein

Native Enzyme + Cysteine 17,800

S-S^O^ Enzyme + Cysteine 18,900

* counts per minute. The values are the mean from duplicate assays, the counting of acid stable radio­ activity was also done in duplicate. The agreement between duplicate samples was 3 • 5$ 1 nil (ca 10 mg) of native Fraction 1 protein was reacted with 50 mg

(0.16 mmole) of Na^S^O^.2H^0 and left for 60 min. The pH was kept at pH 8 by the addition of 1N NaOH during the reaction. 0.5 ml of the solution was diluted with

4,5 nil of 0.1M Tris-HCl pH 8.0. Duplicate samples

0.8 ml each were removed and incubated with 0.2 ml of

0.6M cysteine (freshly neutralized) for 2 hr at 37°C.

At the end of 2 hr 50 ul aliquots were removed and assayed for RuDPcase activity. A control was likewise treated except sodium tetrathionate was omitted from the reaction mixture.

Despite the fact that Fraction 1 protein must be comprised of subunits, it has not been possible to dissociate the oligomeric protein molecule into enzymically active fragments although there could be the possibility that RuDPcase is only a small protein subunit that is associated with the larger moiety.

The effect of urea on the RuDPcase activity of

Fraction 1 protein was then studied and the results are shown in Figure 8. Fraction 1 protein was incubated at 0°C for 30 min with increasing amounts of deionised urea. Aliquots were removed and assayed for RuDPcase activity. The results show that urea even at quite low concentrations (1.5M) reduced considerably the RuDPca«e activity and a urea concentration of about 3M completely inactivated the enzyme.

Enzyme which had been partially inactivated by urea treatment was found to have regained full RuDPcase activity upon removal of the urea from the enzyme by

Sephadex G-25 chromatography.

However, no significant enzyme activity was detected in a solution that had been exposed to 4M urea and then passed through a Sephadex G-25 column. These results are summarised in Table III. 62

INHIBITION OP RuDP CASK ACTIVITY OP FRACTION I PROTEIN BY UREA

(UREA) M

Fig.8 - Inhibition of RuDPcase activity of Fraction 1 protein by urea. Aliquots of Fraction 1 protein in 0.01M Tris-0.001M EDTA (pH 8.3) were incubated at 0°C for 30 min with increasing amounts of urea. After 30 min aliquots were removed and assayed for RuDPcase activity. 63.

TABLE III

Restoration of RuDPcase Activity of Urea Treated Protein

C02 FIXED % SYSTEM C.P.M, /Mg Protein x 10 ^ ACTIVITY

Native Enzyme 24.6 100

Native Enzyme + 1.6M Urea 24.4 99

Native Enzyme + 4M Urea 0.34 1.3

Sulphenyl Thiosulphate

Derivative of Fraction 1

Protein + 4m Urea 2.8 11

* counts per min.

To duplicate 1,5 ml (20 mg) samples of enzyme solution sufficient 8M urea -0.01M Tris-HCl-0.001M EDTA (pH 8.0) to remove urea. 0.8 ml aliquots were removed containing

0.64 mg of protein and incubated with 0.2 ml of 0.6m cysteine for 2 hr at 37°C and then aliquots removed for

RuDPcase activity. Another tube containing the sulphenyl thiosulphate derivative of Fraction 1 protein was treated similarly. The control consisted of enzyme to which no urea was added. 64.

The sulphenyl thiosulphate derivative of Fraction 1 protein was also included in the test system for it was felt that the inability to isolate enzymically active fragments might be due to oxidation of sulphydryl groups at the active site during the dissociation of the protein by high concentrations of urea. Therefore protection of these sulphydryl groups by reacting with the reversible sulphydryl reagent sodium tetrathionate during the dissociation and then subsequent regeneration of the free sulphydryl groups before assaying for enzymic activity was attempted.

There was a slight increase in the sulphenyl thiosulphate system over the native enzyme system but it was concluded that strict maintenance of the tertiary structure of the protein was the overriding factor in determining the RuDPcase activity and that agents such as urea primarily inactivate the enzyme by disorganising this tertiary structure. 65.

3.5 Subunit Structure of Fraction 1 Protein

3.5 (i) Electrophoretic Studies on SCM- Fraction 1

Protein

When carboxymethylated Fraction 1 protein was examined in acrylamide gels containing 8M urea the protein migrated essentially as one band at high loadings of protein (Figure 9a). If the protein loading was decreased there was evidence that the band was split

(Figure 9h) and that a band which was not stained very intensely by the amido black was seen to move slightly faster than the major band towards the anode.

The separation achieved in the electrophoretic system was not entirely satisfactory so an attempt was made to separate the components in a system that separated on molecular weight differences only. Gel filtration of SCM-Fraction 1 protein on Sephadex G-200 in the presence of an 8M urea buffer resolved the protein into two major subunits of different molecular weight (Figure 10). Both these peaks ? when electro- phoretically analysed, revealed their correspondence to the two bands seen in the light-loaded 8M urea gels (Figure 9a, e, f), 66.

——2^. “ Gel electrophoresis pattern of S-carboxyraethyl Fraction 1 protein, (d) 8M urea gel of c.100 ug of SCM-Fraction 1. (e) and (f) 8M urea gels of c.50 ug of peaks A and B respectively as indicated in Fig. 10 from the gel filtration of SCM-Fraction 1 on Sephadex G.200.

FiiL:—9b - 8M urea-acrylamide gel pattern of a light- loaded (c.50 ug) SCM-Fraction 1, showing distinct splitting of the band into two major components. Origin is not shown; arrow indicates the direction of protein migration. 67.

TUBE NIMBER

Fig. 10 - Gel filtration on Sephadex G-20O olf 30 mg of SCM-fraction 1 in 8M urea-O.05M Tris-HC1— O. 0C1M EDTA-0.01M KC1-0.2M ammonia, pH 10.0, Fraction size 4 ml; column dimensions 2.5 by 44 cm. 68.

3.5 (ii) Molecular Weight Estimation of the Subunits

Resolved on Columns of Sephadex G-200

For globular proteins in aqueous solutions the

elution volumes from Sephadex columns of proteins which

are not involved in reversible association equilibria

are a function of their molecular weight (Andrews 1964;

Whitaker 1963)* They showed that the elution volume,

Ve, of the protein, is linearly related to the logarithm

of their molecular weight over a certain size range,

depending on the pore size of the Sephadex. The molecular weight of an unknown protein may therefore be estimated by means of a calibrated column. Thompson

and O’Donnell (1965) have shown this linear relation to hold also for reduced and carboxyme thylated proteins

in 8M urea in which the proteins are likely to behxve

randomly coiled molecules. The relationship between

the elution volume and the logarithm of molecular weight

for a series of reduced and carboxymethylated proteins

of known molecular weights is shown in Figure 11. The molecular weights taken for these proteins were plasma albumin 67,000 (Putman 1965)> ovalbumin 45,000 (Warner

1954), lysozyme 14,300 (Canfield 1963), and

-lactoglobulin 18,300 (Piez et al. 1961). Allowances were made for the introduction of the carboxymethyl groups. 69

>• LYSOZYMESCM FRACTION 1 (peak b) - _ i 'nrTrv' i r\r> t m '

S01 FRACTION 1

^ (PEAK a) 3OVINE PLASMA ALBUMIN

LOG (MOLECULAR WEIGHT)

Fig.11 - Graph of elution volume against logarithm of the molecular weight for a series of S-carboxymethyl proteins chromatographed on Sephadex G-200 in 8M urea buffer. Approximately 5 ml containing 30-60 mg of proteins was loaded in each case. 70.

The elution volumes of the two peaks (A and B) are consistent with proteins having molecular weights of approximately ,000 and 16,000 respectively.

3.5 ( iii ) Amiro_Acid Composition of the Subunit

Components

The tubes indicated by the bars in Figure 10 were pooled, and after dialysis and freeze drying were subjected to amino acid analyses and the results are given in Table IV.

Compositional differences between the two peaks are quite apparent. Peak B is rich in lysine and deficient in histidine when compared with peak A.

It is clear that the two peaks constitute different protein chains. TABLE IV.

Amino Acid Composition of the Subunit Fractions of Reduced and Carboxymethvlated Fraction 1 Protein.

Time of hydrolysis 24 hr at 110°C. Values are the number of residues per mole taking the molecular weights of

peaks A and B as 54,000 and 16,000 respectively. Average values from one hydrolysate from three different

preparations are given. Values were calculated from the number of umoles of amino acids recovered from the column

assuming an average residue weight of 110. Values in parentheses are calculated from the data of Rutner and Lane

(1967) for subunits obtained from Fraction 1 protein treated with sodium dodecyl sulphate assuming identical

phenylalanine contents.

PEAK B AMINO ACID PEAK A PEAK B AMINO ACID PEAK A

(47.2) 12.8 (9.3) Lysine 25.9 (26.6) 11.7 (9.8) Glycine 50.1 (48.6) 8.8 (7.0) Histidine 14.0 (15.1) 1 .7 (3.7) Alanine 47.9 11.4 - Arginine 28.3 (32.4) 4.7 (8.0) Valine 35-3 - 2.3 (3.7) S-Carboxymethylcys teine 7.9 - 2.8 - Methionine 7-5 (9-5) 5.7 (4.5) Aspartic acid 43.9 (49.1 11.7 (17.3) Is oleucine 19.7 (19-8) (47-2) 12.5 (13.2) Threonine* 31.6 (39.4) 6.2 (9.6) Leucine 43.0 8.0 (12.6) Serine* 19.9 (18.2) 7.4 (6.2) Tyrosine* 17.1 (20.7) 8.1 (8.1) Glutamic acid 51.8 (49.5) 17.6 (18.3) Phenylalanine 22.5 (22.5)

Proline 23.9 (25.4) 11.6 (12.6) 72.

3.5 (iv) Studies on Maleylated Fraction 1 Protein

The separation of the subunits of Fraction 1

protein on a preparative scale relies on gel filtration

on Sephadex columns in the presence of 8M urea. To be

able to use Sephadex columns in the absence of the

denaturing agent is a distinct advantage. The flow rate through Sephadex G-200 columns progressively becomes slower with the result that fractionation takes longer. It became desirous then to investigate an alternative method for preparing the subunits of

Fraction 1 protein.

Maleic anhydride has been shown to have a variety of different applications in protein chemistry

(Buter et al. 1969). The increased negative charge at neutral pH introduced onto the protein by the maleyl group means that protein-protein interactions are minimised and protein-water interactions are favoured.

Several proteins have been dissociated into soluble subunits by treatment with maleic anhydride (Butler

1967; Bruton and Hartley, 1968; Sia and Horecker, 1968).

Maleyl proteins and peptides can then be fractionated without use of disaggregating solvents. Proteins, dissociated by other means e.g., exposure to 8M urea and then subsequently maleylated, should be amenable 73. to separation on Sephadex columns containing a simple salt medium.

Fraction 1 protein was denatured in 8M urea and carboxymethylated as before. The 8M urea -Tris buffer was exchanged by dialysis for one which would not react with the maleic anhydride, 0.4M borate (pH 8.5) > and the protein was reacted with excess maleic anhydride.

After dialysis against water and freeze drying the protein was soluble in dilute Tris-HCl buffer, and could be fractionated on a Sephadex G-100 column (2.5 x 88 cm) into two main components. The result of a typical fractionation is shown in Figure 12.

Sephadex G-100 proved to be very successful in fractionating the components of maleylated carboxymethyl

Fraction 1 protein. Denaturation of proteins often results in increased physical dimensions with the result that the protein will now be excluded from a

Sephadex bead which normally would have allowed its admittance. Thus Sephadex G-100 will exclude uncoiled proteins of molecular weight greater than about 20,000

(Davison, 1968). Peak A was eluted in the void volume of the column, well separated from peak B. On a dry weight basis the yields of the two subunits were

65-70% of subunit A and 30-35*1° of subunit B. EDTA Column Fig. of

S-carboxymethyl-maleylated

12 E 280 mp . (pH

size -

8.3). Gel

2.5

filtration

x

88

74 cm. .

TUBE on Eluant

Sephadex

NC. Fraction

0.01M

G-

Tris-0.001M 1 1

00 protein.

of

50

rag

75.

The amino acid compositions of these two peaks were comparable with the subunits isolated from the

Sephadex column containing 8M urea.

3.5 (v) Further Characterisation of the Subunits

Electrophoretic examination of these two components in a Gradipore gel further revealed molecular weight p dissimilarities. Gradipore gel electrophoresis will separate proteins on the basis of molecular weight and the effect of charge differences will be minimised.

Figure 13 shows the results obtained in a seven hour electrophoresis. The proteins have not moved to their equilibrium positions in this period but clearly show good separation of the subunits from each other and from the native protein.

Subunit A migrated to a position between human carboxyhaemoglobin and Cohn's fraction V albumin while subunit B migrated to a position ahead of ovalbumin, but behind low molecular weight peptides; obtained when maleylated kangaroo $.-globin was digested with trypsin.

The denatured proteins, i.e. subunits A and B, are not as compact on the gels as was the native protein and standards. This is probably due to the fact that these denatured proteins are uncoiled and thread their 76.

Fig.13 - Seven hour electrophoresis at pH 8.5 in a Gradipore polyacrylamide slab of native Fraction 1 protein (1, 2) and maleylated subunits A and B (3> 5 and 4, 6) respectively. Standard proteins were albumin (7), ovalbumin (8), tryptic digest of maleylated kangaroo d-globin (9) and human carboxyhaemoglobin (10). Migration of the proteins is from left to right. The solid line through the gel indicates the acrylamide concentration at any particular length of gel. 77.

way through the gel pores.

Figure 14 shows the results obtained after 15 hr electrophoresis. Subunit B will migrate off the gel during this extended time; a further indication of its low molecular weight. Subunit A was found to have migrated until it reached an acrylamide gel concentration of about 17$ while the native protein did not penetrate the 10$ gel.

Molecular weight determinations of the polypeptide chains in oligomeric proteins using acrylamide electrophoresis in the presence of sodium dodecyl sulphate has been well characterised (Weber and Osborn,

1969; Bunker and Rueckert, 1969) and results obtained by this method correlate with results obtained using physicochemical .means.

The isolated subunits from carboxymethylated maleylated Fraction 1 protein were run on 10$ polyacryl­ amide gels containing sodium dodecyl sulphate to determine the molecular weight of the subunits. Figure 15 shows the separation of the isolated subunits A and B, as well as several standard marker proteins, on a single gel.

The identity of each band was established by running each protein on a different gel. A graph (Figure 16) 78.

5LAB LENJQTH (li.M.)

Fig. 14 - Fifteen hour electrophoresis at pH 8.5 in a p Gradipore polyacrylamide slab of native Fraction 1 protein (l, 2) and maleylated subunit A (3, 4). Standard proteins were albumin (5) oestrogen sulpho- transferase (E.C.2.8.2.4) (6), this enzyme has a molecular weight of about 70,000 (Adams and Low 1970, the enzyme was a gift from Dr. J. B. Adams), ovalbumin (7)» human carboxyhaemoglobin (8). 79

Fig. 15 - Electrophoresis in 10$ acrylamide gels containing SDS. The standard proteins were, (1) Albumin (3) Ovalbumin (4) Trypsin. Bands number 2 and 3 are subunit A and B respectively. In the densitometer trace, shown above the gel, subunit A occurs as a shoulder on the peak corresponding to albumin. 80.

of the relative mobility (trypsin = 1.0) of each protein

against the logarithm of their respective molecular

weights yielded a straight line which allowed the

molecular weights of the two subunits A and B to be

determined as 58,000 and 13>000 respectively.

The introduction of the maleyl group increases the

molecular weight of the protein by about 100 for each

free amino group that it reacts with. The lysine

content of the subunit A as determined from an amino

acid analysis was 26, thus the molecular weight of the

unmaleylated subunit A was estimated to be 55*^00, which compares favourably with the value determined using the calibrated Sephadex G-200 column.

The molecular weight of subunit B even without

correcting for the bound maleyl groups of subunit B was lower in the polyacrylamide system than in the

Sephadex G-200 system.

3.6 Peptide Mapping of the Isolated Subunits

The isolated subunits were treated with acetic acid

to regenerate the free lysine groups and then digested with trypsin and fingerprinted, the results of which

are presented in Figures 17* 18. These peptide maps or

fingerprints clearly show that these two subunits differ ) 000 , 13 (

B

by

the SUBUNIT

of

acrylamide composed mobility

) fo c

0 . ^O 1

weights

a

table

■•JRYPSIN

in (TRYPSIN relative

the

SDS

IN the

molecular from

MOBILITY of

proteins PEPSIN

The

. -•

RELATIVE 81 taken ) curve

ELECTROPHORESIS

000 SDS. , several

58 OVALBUMIN

( % were 1

(1969)*

A of

O.

)

0 . POLYACRYLAMIDE SUBUNIT 1 Calibration Osborn

= proteins

- and

6

1 containing

(trypsin standard Fig. gel Weber ALBUMIN' cnco r\ 82

Fig. 17 - Peptide map obtained by ionophoresis at pH 6.4 (i) and chromatography (c) in butanol-pyridine-acetic acid-water (15:10:3*12 V/V) of a tryptic digest of the isolated S-carboxymethylated maleylated subunit A. The maleyl groups were removed as outlined in the Methods, prior to tryptic digestion.

83.

Fig. 18 - Peptide map obtained by ionophoresis at pH 6.4

(i) and chromatography (c) in butanol-pyridine-acetic acid-water (15*10:3*12 V/V) of a tryptic digest of the isolated S-carboxymethylated maleylated subunit B. The maleyl groups were removed, as outlined in the Methods, prior to tryptic digestion.

84.

in structure from each other as would be predicted from their amino acid composition.

If the isolated subunit A, of molecular weight, approximately 5^>000, consisted of a single chain or two non identical chains, peptide mapping of the carboxy- methylated subunit A, completely hydrolysed by trypsin at all lysine and arginine bonds, should reveal approximately 55~6o ninhydrin-positive spots. In fact, the peptide map contained only about one half this number. After digestion with trypsin and freeze drying not all the digest was soluble in the pH 6.4 electrophoresis buffer. It was important to determine whether this insoluble fraction contained several inaccessible trypsin-sensitive bonds, thus rendering an estimate of the molecular weight of the subunit speciously low.

Relative total amino acid contents of 11.8$ and

88.2$ were obtained for the insoluble residue and soluble peptides, respectively, by acid hydrolysis and assay on the amino acid analyser.

The amino acid composition of the insoluble and soluble peptide fractions was determined and related to that of the undigested carboxymethylated subunit A.

The results are shown in Table V. The insoluble fraction 85.

TABLE V.

Amino Acid Composition of Carboxymethylated Subunit A and Derivative pH 6.4 Soluble Tryptic Peptides

(S6.4 Fraction) and pH 6.4 Insoluble Residue (R6.4 Fraction)

R6.4 S6.4 Sum of Carboxymethylated AMINO ACID Amount in Calculated No. Nearest Amount in Calculated No. Nearest Residues Subunit A umoles of Residues Integer. umoles of Residues Integer. S6 .4 + r6.4 No. of Residues

Lysine 0.0207 1.0 1 0.0450 25 25 26 l 26 Histidine 0.0456 2.2 2 0.2070 11.5 12 14 14 Arginine 0.0276 1.3 1 0.510 28.1 28 29 28 Aspartic acid 0.0852 4.1 4 0.7530 41.7 42 46 44 Threonine 0.0712 3.4 3 0.4760 26.5 27 30 32 Serine 0.558 2.7 3 0.2540 15.7 16 13 20 Glutamic acid 0.0776 3.8 4 0.8420 46.8 47 51 52 Proline 0.0713 3.4 3 0.3740 20.8 2 1 24 24 Glycine 0.1566 7.6 8 0.7520 41.8 42 50 50 Alanine 0.0924 4.5 5 0.8000 44.4 44 49 48 Valine 0.1017 4.9 5 0.5940 33.0 33 38 35 Methionine 0.0110 0.53 1 0.0860 4.8 5 6 8 Isoleucine 0.0576 2.8 3 0.3440 19.1 19 22 20 Leucine 0.0884 4.3 4 0.7250 40.3 40 44 43 Tyrosine 0.0400 1.9 2 0.3420 19.0 19 2 1 17 Phenylalanine 0.0556 2.7 3 0.3210 17.8 18 2 1 22 SCM-cys t eine trace 0.1150 6.4 6 6 8 Total No.of Residues 52 444 496 491 ______I 86.

contained a small amount of lysine and arginine. On the assumption that this fraction contained one lysine residue and the soluble peptides contained the remainir.g lysine residues present in the intact molecule, the sun of the residues in both fractions agreed well with that of the undigested subunit. The values of 52 residues for the insoluble fraction and 444 residues for the soluble peptides, 10.5$ and 89.5$ of the total respect­ ively, match the results of the acid hydrolysis quantit­ ative analysis closely.

Thus the potential maximum number of tryptic peptides contained in the insoluble residue is only three and consequently has no significant effect on a molecular weight estimate by peptide mapping. It would appear that the minimum molecular weight of the subunit A based on peptide maps, is 27,000, one half of its isolateable molecular weight, and that the identical chains are held together by some bond, probably covalent , other than a disulphide bond.

The peptide map of subunit B revealed 13-14 ninhydrin-positive spots in close agreement to that expected from the total lysine and arginine residues of the protein. 87.

3.7 NH^- terminal Amino Acid Analysis of Subunit A,

An Edman degradation of subunit A as performed in

an attempt to clarify the position regarding its molecular weight. An appropirate quantity of maleylated carboxy- methylated subunit A was treated with acetic acid to

regenerate free amino groups. After one step of the

Edman degradation considerable ultraviolet material was

present in the ethyl acetate extract. The absorption

curve gave a ratio of the minimum at 240 mu to maximum

at 267 mu of about O.87, which is considerably different

from that of pure amino acid derivatives (approximately

0.4, Sjoquist 1957). For this reason it was impossible

to directly quantitate the phenylthiohydantoin derivative

This deviation in the ultraviolet absorbance was due to contamination with side products of the reaction such as phenylthiourea formed from ammonia present in the sample.

Thin-layer chromatography revealed only one fluorescent spot in the region of phenylthiourea. No significant quantity of amino acid was recovered in an a£id hydrolysis of the phenylthiohydantoin derivative.

The aqueous phase, after ethyl acetate extraction of the phenylthiohydantoin, was made alkaline to pH 7*5 with phosphate buffer and then extracted with ethyl 88.

acetate to see if the phenylthiohydantoin of histidine was present. No ultraviolet absorbance was detected in the ethyl acetate indicating that histidine could not be the N-terminal amino acid.

This absence of a free N-terminal amino acid could not be due to failure to remove the maleyl group from the N-terminal amino acid for similar results have been obtained with the subunit A isolated on Sephadex G-200 columns in the presence of 8M urea.

3.7 (i) Isolation of a Peptide With a Blocked Amino

Group

The inability to detect an N-terminal amino acid with the Edman reagent could mean that the amino end of the protein is masked by some blocking group such as or formyl, acetyl or propionyl group the N-terminal residue is pyroglutamic acid.

Subunit A was digested for 3-4 hr with pronase and the digest mixture poured through a Dowex-50

(H+form) column at 4°C. Peptide material containing no free amino groups passed through the column without binding and was collected. An aliquot was taken and hydrolysed and the amino acid composition determined.

The peptides isolated on Dowex-50 from a pronase 89. digest of Subunit A contained several amino acids with

glutamic acid predominating (Column 1 Table Vi). The

presence of glutamic acid in the water wash from the

Dowex-50 column suggested that perhaps the absence of

an N-terminal amino acid in subunit A was due to a

terminal pyroglutamic acid residue. However there is

the possibility that the pyroglutamic acid is an artifact formed by the cyclisation of a glutaminyl residue under the acidic conditions on the Dowex-50 column. Thus a further pronase digest of subunit A was performed followed by the immediate addition of fluorodinitrobenzene to the reaction mixture to block all free amino groups (Press et al. 1966), It has been established that glutamine does not form any significant amounts of pyroglutamic acid under these conditions

(Press et al. 1966). After removal of DNP-peptides, the extract was passed through a Dowex-50 column as before and an aliquot hydrolysed and its amino acid composition determined. The result (column 2) were similar to that in which no fluorodinitrobenzene was added. The predominant amino acid was glutamic acid but the yield of aspartic acid was very low. The maleylation of the protein served as a precaution against there being a lysine residue close to the N-terminal. 90.

TABLE VI

Amino Acid Composition of Hydrolysed Acidic Peptides

Isolated by Chromatography on Dowex-50 from Subunit A

Digested with Pronase

Moles of Amino Acid/mole of Subunit A. Maleylated Amino Acid Carboxymethylated A Carboxymethylated A 1

Aspartic acid 0.45 trace 0.54 Threonine jad 0.87 0.69 0.64 Serine 0.56 0.60 0.64 Glutamic acid 1.20 1.20 1.76 0 xn O 0 0 • Proline 0.61 • O Glycine O 0.73 0.54 Alanine 0.44 0.44 0.39 Valine 0.30 0.43 - Is oleucine 0. 10 - - Leucine 0.21 - - Phenylalanine 0. 12 - -

Uncorrected for decomposition. * No treatment with fluorodinitrobenzene. $ Fluorodinitrobenzene added to digest after pronase digest. + For this experiment the time of digestion was 12 hr and carboxymethylated subunit A prepared on a Sephadex G-200 column in the presence of 8M urea was used. The values represent the mean of two experi­ ments. Fluorodinitrobenzene was added to the digestion mixture 91

A peptide containing a free lysine residue would bind onto the column of Dowex-50 and be undetected. However,

as column 3 shows it was possible to isolate acidic peptides from subunit A in which the lysine residues had not been modified.

It was clear from Table VI that the eluate from

the Dowex-50 column was a mixture of peptides and an attempt was made to purify the peptide mixture by

paper electrophoresis.

After electrophoresis, at pH 3»5» four bands which gave no colour with ninhydrin indicating no free amino groups but gave a positive reaction with the chlorine- tolidine-iodide reagent were detected on the paper. The mobilities of these bands in the relationship to pyroglutamic acid are presented in Table VII. 92.

TABLE VII

Mobilities at pH 3.5 of Acidic Peptides of Subunit A

With a Blocked Amino Group.

Peptide No. Rf. pyroglutamic acid = 1,0

1 0.4 1

2 O.56

3 0.70

4 0.82

The bands from an unstained paper were eluted with 6N

HC1 and hydrolysed for 24 hr at 110°C and the amino acid composition determined. These results are shown in

Table VIII. 93.

TABLE VIII

Amino Acid Composition of Peptides Containing No Free

Amino Group Separated by Paper Electrophoresis ,PH 3.5

Moles Amino Acid/Moie Subunit A Amino Acid Peptide No. 1 2 3 4 O Aspartic acid 0.015 O 0.023 0.012 Threonine 0.050 0.015 0.037 0.027 Serine 0.050 0.023 0.040 0.033 Glutamic acid 0.106 0.037 0.043 0.020 Proline 0.048 0.012 0.016 - Glycine 0.003 0.02 5 0.04 1 0.045 Alanine 0.002 0.02 1 0.029 0.014 Valine trace 0.008 0.023 trace Is oleucine trac e trace trace trace Leucine trace 0.009 0.008 trace Tyrosine trace trace trace trace Phenylalanine trace trace trace trace

The results show that there are compositional differences between the peptides. The yields based on moles of protein originally digested are extremely low, peptide number 1 being the major component but even in this case the yield is disappointingly low. Whether the low yield is a real value or due to poor elution from the paper is not known.

Consequently it was not practical at this time to further characterise the peptides, especially peptide No.1, as has been done with the N-terminal peptide of the heavy 94.

chain of rabbit immunoglobulin (Wilkinson, Press and

Porter, 1966 ) .

3.8 C-Terminal Residue

3.8 (i) Hydrazinolysis

The free amino acids determined in the hydrazinoly-

sates of subunit A are shown in Table IX.

TABLE IX

Yields of C-Terminal Amino Acids After Hydrazinolysis

Values are Mole/Kole of Subunit A ^rotein, Uncorrected

for Losses

Amino Acid Subunit A Lysozyme

Serine 0.22 trace

Glycine 0.13 trace

Alanine 0. 10 trace

Valine 0.12 -

Leucine - 0.79

Tyrosine 0. 1 5 -

As Braun and Schroeder (1967) have shown, there is considerable variation in yields depending on the time and temperature of the reaction. In these present experiments on subunit A no study of these variables was 95.

made, a temperature of 80°C with heating for 48 hr being

used.

Under these conditions lysozyme, as a control

protein, gave a 79$ yield of leucine which was less than

the 94-96$ reported by Braun and Schroder (1967) after

15“20 hr at 80°C. No one particular amino acid was

present in the hydrazinolysates of subunit A in high

enough yields to be the C-terminal amino acid.

Considerable amounts of serine, glycine and alanine were obtained along with some valine and tyrosine. The

first three amino acids occur as background in the hydrozinolysates of all proteins studied, in yields of about 5$ of one residue per mole of protein (Braun and

Schroeder 1967). As the method employed to separate the amino acids from the hydrazides only separated the acidic and neutrals any basic amino acid would not be detected. As no neutral or acidic amino acid was present in sufficient quantities to be considered a C-terminal amino acid it was thought that a basic amino acid was occupying the C-terminal position. Thus a carboxy- peptidase B digestion of subunit A was performed.

3.8 (ii) Carboxypeptidase B Digestion

The a-chain of kangaroo globin was used as a 96.

control for the carboxypeptidase B digestion and under the conditions of the experiment, three amino acids corresponding to the last three residues were liberated, viz. lysine (0.50)-tyrosine (0.51)-arginine (0.68).

The figures in brackets are moles of amino acid per mole of protein. However, no basic amino acid was liberated from the subunit A digestion with carboxypeptidase B.

3.8 (iii) Carboxypeptidase A Digestion

Carboxypeptidase A digestion for 2 hr at 35 C of subunit A in the presence of 0.1$> sodium dodecyl sulphate did not liberate any significant amount of amino acid.

3.9 End Group Analysis of Subunit B.

Both amino and carboxyl end group analyses of subunit B, using similar techniques to that employed for subunit A, were done, but no conclusive evidence to allow a definite amino acid to be ascribed to either end of the subunit was obtained. 97.

CHAPTER 4. DISCUSSION

The initial part of this study has confirmed many of the known facts about Fraction 1 protein. It was possible to prepare Fraction 1 protein from isolated chloroplasts or whole leaf extract by a combination of ammonium sulphate precipitation and gel filtration on

Sephadex G-200 in agreement with the earlier report by Trown (1965).

A molecular weight estimation of the native Fraction

1 protein was not attempted in this study but the amino acid analysis was in close agreement with that reported by Ridley et al. (1967). This would suggest that the protein isolated in this present study is comparable to that isolated by Ridley et al. (1967). The electro­ phoretic behaviour of the protein when examined in a number of systems gave no reason to suspect that the protein was heterogeneous.

The number of half-cystine residues in the protein, determined from the amino acid analysis was shown to be

4.0 per 24,400 gm. In the native state only a small number of SH groups are sufficiently exposed to allow them to react with the rather "bulky" thiol reagent,

5,5*-dithiobis(2-nitrobenzoic acid). Upon denaturation with sodium dodecyl sulphate more SH groups became 98.

exposed until a total of 107 per 550,000 gm had reacted with the thiol reagent, this would indicate that no SH groups are involved in disulphide bonds. Using another thiol reagent, PCMB, Sugiyama and Akazawa (1967)_ observed that there were 96 SH groups per molecule of

Fraction 1 protein and suggested that they were distributed evenly amongst twenty four hypothetical monomers.

The exact role of the cysteine residues in the protein molecule has been studied in some detail by the Japanese workers and they have postulated that only a few SH residues of the total are essential for catalytic activity (Sugiyama and Akazawa 1967)-

The differential reactivity of the cysteine residues of a protein toward a particular chemical reagent is attributable to a number of factors, most important of which are the state of ionisation and/or the accessibility of the SH group.

The properties which determine reactivity are conferred on the SH groups by their local environment, and, therefore depend upon the over-all three dimension­ al structure of the protein. It is a general notion that those functional groups in the active site are normally among the most reactive of their class. 99.

This is supported by the exceedingly high rate of reaction towards thiol reagents of the "active site" cysteine of muscle glyceraldehyde 3-phosphate dehydrogenase (Davidson et al. 1967; Wassarman and Major 1969) and papain (Smith, Light and Kimmel 1962).

However, this is not always the case. In some enzymes e.g. isocitrate dehydrogenase, there seems to be no large differences in the reactivity of five cysteine residues that are normally available to

5,51-dithiobis (2-nitrobenzoic acid), although one or two are actually involved in the catalytic function of the enzyme (Colman 1969)* Sugiyama et al. (l968d) were unable to demonstrate, using 14C-iodoacetamide, the presence of an extremely reactive cysteine residue and thus the sulphydryl groups of Fraction 1 protein could belong to the latter class. In this present study the importance of SH groups in maintaining catalytic activity was verified. Sodium tetrathionate was shown to react with the protein to form an inactive complex. This inactive form of the enzyme could be converted back to an active form by incubating with cysteine. There are no reports in the literature implicating other amino acid residues that 1 00

are essential for the catalytic activity of the enzyme.

In sulphydryl enzymes it is difficult to chemically modify such groups as a carboxyl or imidazole group because almost any reagent that will react with these groups will certainly react with a sulphydryl group first. The demonstration that sodium tetrathionate can be used as a reversible modifier of the sulphydryl groups of Fraction 1 protein opens up the possibility of studying other amino acid residues involved in the enzymic function such as has been done with Strepto­ coccal proteinase (Liu, 1967).

Perhaps the most significant contribution of this present study is the demonstration that Fraction 1 protein can be dissociated into two main components.

It has been the lack of knowledge concerning the subunit structure of Fraction 1 protein that has limited the interpretation of much of the chemical data which is available.

Gerhart and Schachraan (”1965) demonstrated that the addition of the mercurial, p-mercuribenzoate to the enzyme aspartate transcarbamylase caused the enzyme to dissociate into two different types of subunits. (Sug- iyama and Akazawa (1967) and Sugiyama et al.(1968) pursued this technique with Fraction 1 protein but were 101 . unsuccessful in obtaining clear evidence for protein dissociation by analytical ultracentrifugation. Using polyacrylamide gel electrophoresis they were able to observe a different banding pattern of Fraction 1 protein which had been in contact with the mercurial for 24 hr. Electron microscopic studies on Fraction

1 protein that had been exposed to high concentrations of p-chloromercuribenzoate revealed disordered aggregates (Sugiyama et al. 1968).

The first clear demonstration that Fraction 1 protein was composed of dissimilar subunits was the report by Rutner and Lane (1967)* Using sodium dodecyl sulphate they were able to dissociate the protein, after blocking the sulphydryl groups with ethylenimine, into two distinct subunits. These subunits were shown to have different amino acid composition and sedimentation velocities.

Sugiyama and Akazawa (1967) reported that they were unable to characterise any subunits identifiable by polyacrylamide gel electrophoresis from urea treated

Fraction 1 protein. This present study clearly shows that 8M urea dissociates the protein into two major components which can be separated from each other by gel filtration on a Sephadex G-200 column containing 8M 102

urea. These two subunits have different amino acid

compositions, in agreement with the work by Rutner and

Lane (1967)* A distinct advantage of using a Sephadex

column packed in 8M urea is that it prevents association

processes taking place and that protein behaves as

random coils. By comparison of their elution volumes

on gel filtration with those of other well-characterised

proteins a valid molecular weight can be determined.

Thus the molecular weights of the two subunits A and

B were estimated to be 5^+> 000 and 16,000 respectively.

An accurate molecular weight determination is

essential in any study of a protein. It is only

recently that another estimate of the molecular weight

of the subunit has been reported (Rutner, 1970) and the values are awaited with interest.

The use of maleic anhydride to introduce maleyl groups onto the amino groups of the protein allowed the subunits of the protein to be separated on a Sephadex

G-100 column without the use of 8M urea in the column.

On an average weight basis the large subunit accounts

for 68% and the smaller subunit 32%, The molecular weight of the large subunit determined by the two methods employed in this study show good correlation and an average value would be about 55,000. This would 103.

mean that in the native oligomeric structure there are

six of the large subunits. For the smaller subunit

the molecular weights were not in as good agreement as

for the larger subunits. However if one takes the

average of the two values i.e. 14,000 then the number

of smaller subunits is twelve. These calculations

assume a molecular weight of 500,000 for the native

protein. Caution must be exercised in accepting these

values,for calculations similar to this to determine

the number of each different subunit in aspartate

transcarbamylase suggested that there were four of

each type (Changeux and Gerhart 19^7). The amino acid

sequence of the regulatory subunit has now been

determined (Veber 1968) and consequently its true molecular weight known, It is now clear that there are

six of each kind of subunit in aspartate transcarbamylf'c.

(Weber 1968). The values assigned for the number of

each different kind of subunit in Fraction 1 protein will need to be verified by eventual amino acid sequence

determination on the individual chains.

The arrangement of the individual subunits to form

the regular cube like structure that is seen in electron microscopic studies of Fraction 1 protein is also an interesting problem. Accepting the tentative values 104.

reported in this study it is possible that the six large

subunits could occupy the six faces of a cube and the

twelve smaller subunits could then lie along the twelve

edges of the cube.

Chemical characterisation of the isolated subunits has been hampered by the inability to identify either

the amino or carboxyl terminal amino acids. The

isolation of peptides with no free amino groups, from

a pronase digest of subunit A would support the idea

that the amino group of subunit A is blocked and thus inaccessible to end group analysis. The carboxyl

terminal amino acid would have escaped detection if

the sequence is pro-X (where X is a basic amino acid).

The presence of a proline residue in the pentultimate position would render a carboxy-peptidase B digestion ineffectual. An acid or neutral amino acid at the carboxyl end should have been detected in the hydrazin-

olysis technique.

Recently there has been wide interest shown in comparative studies on RuDPcase (see 1,8 page 22) and attention has been focused on the evolutionary develop­ ment of the mechanisms of photosynthetic C0o fixation

(Anderson et al. 1968, Kieras and Hazelkorn 1968). The

finding by Anderson et al. (1968) of an RuDPcase with 105

a molecular weight of about 120,000, in Rhodospirilium

rubrum promoted Anderson and Fuller (1969) to suggest

that the RuDPcase with the larger molecular weight of

500,000 as found in other photosynthetic bacteria and

higher plants may have evolved from the small RuDPcase

in Rhodospirillum rubrum.

The subunit molecular weights of the RuDPcase from

Hydrogenomonas was reported to be about 40,000 (Kuehn

and McFadden 1969) an<3 that 12-14 of these subunits make up the native enzyme. These authors suggested that if

increasing molecular weight of an enzyme can be

considered a consequence of later evolutionary develop­ ment, then the Hydrogenomonas and higher plants may

represent a later form of autotrophy than do the purple, nonsulphur photosynthetic bacteria typified by R. rubrum.

The evolutionary interrelations may well be studied best at the protein amino acid sequence level. Thus a

concerted effort to characterise the subunits of the

carboxylase from many different sources should be attempted.

So far the subunit structure of Fraction 1 protein

from higher plants has only been investigated in spinach

(Rutner and Lane, 19^7) Kawashima 1969) tobacco

(Kawashima 1969) and silver beet. It has been shown that the amino acid composition of the large subunit of

Fraction 1 protein from these sources seems to be the same (although this does not mean the amino acid sequences are identical) while there seems to be a greater variation in amino acid composition in the smaller subunit.

This particular field of study is only just beginning to develop and should be an avenue of fruitful study in the future. In particular the isolation and characterisation of peptides from subunit A containing unique cysteine residues will clarify the position regarding the molecular weight of this subunit. The relatively low methionine content in both subunits will certainly make cyanogen bromide cleavage a favourable technique in determining the amino acid sequences. 10?.

REFERENCES Adams, J.B., & Low, J., Proc, Aust. Biochem. Soc. 71 (19?0)

Akazawa, T., Sato, K., & Sugiyama, T.t-Arch, Biochem^ Biophys. 132, 253 (1969).

Akazawa, T., Sugiyama, T., Nakayama, N, , & Oda, T, , Arch, Biochem. Biophys. 128, 646 (1968).

Akoyunoglou, G,, Argyroudi-Akoyunoglou, J.H., & * Methenitou, H., Biochim, Biophys. Acta. 132, 481 (1967)*

Akoyunoglou, G., & Calvin, M., Biochem, Z., 338, 20 (1963)

Ambler, R.P., Methods in Enzymology vol, XI, Enzyme Structure - Academic Press N.Y. ed. Hirs, C.H.W.,

p. 155 (1967).

Anderson, L.E., Price, G.B., & Fuller, R.C., Science 161, 482 (1968).

Anderson, L.E., Vorthen, L.E., & Fuller, R.C., Fed. Proc.

26, 725 (1967). \

Andrews, P., Biochem. J. 91. 222 (1964).

Argyroudi-Akoyunoglou, J.H., & Akoyunoglou, G., Nature 213. 287 (1967).

Bassham, J.A., Advan, Enzymol. 2 5. 39 (1963).

Bassham, J.A., Plant Biochemistry, p.875> eel. by Bonner, J, & Varner, J.E., New York Academic Press

(1965).

Beavan, G.H., & Holiday, E.R., Advan. Protein Chem. 2.r

319 (1952).

Berggard, I,, Arkiv Kemi. 18. 291 (1961).

Blackburn, S., & Lee, G.T., J, Text. Inst. 45, T487 (1954). 108

Blomback, B., Blomback, M., Edraan, P. , & Hesa^l, B., Biochim, Biophys. Acta JJJ5, 371 (1966).

Braun, V., & Schroeder, W.A., Arch. Biochem, Biophys. 118, 241 (1967).

Bruton, C.J., & Hartley, B.S., Biochem. J. 108, 281 (1968).

Butler, P.J.C., Ph.D Dissertation: University of Cambridge (1967)*

Butler, P.J.G., Harris, J.I., Hartley, B.S., & Leberraan R., Biochem. J. 112. 679 (1969).

Calvin, M., J. Chem. Soc. 1895 (1956).

Calvin, M., Science, 135* 879 (1962).

Canfield, R.E., J. Biol. Chem. 238. 2691 (1963).

Changeux, J.P., & Gerhart, J.C., "Regulation of Enzyme Activity and Allosteric Interactions". ed. E. Kvamme and A. Pihl. pei3* (Academic Press, New York) 1967*

Colman, R.F., Biochemis try 8, 888 (1969)*

Cr.estfield, A.M., Moore, S., & Stein, W.H., J. Biol. Chem. 238. 622 (1963).

Davidson, B.E., Sajgo, M., Noller, H.F., & Harris J.I., Nature 216. 1181 (1967).

Davis, B.J., Ann. N.Y. Acad. Sci. 121. 404 (1964).

Davison, P.F., Science 161. 906 (1968).

Dorner, R.W., Kahn, A., & Vildman, S.G., J, Biol. Chem. 229. 945 (1957).

Dorner, R.W., Kahn, A., & Wildman, S., Biochim. Biophys. Acta. 29. 240 (1958).

Dunker, A.K. & Rueckert, R.R,. J. Biol. Chem. 244. 5074 (1969). Eggman, L., Singer, S. J., & Wildman, S.G., J. Biol. Chera. 205, 969 (1953)-

Ellman, G. L., Arch. Biochem. Biophys. 82, 70 (1959).

Fiedler, F., Mullhofer, G., Trehst, A., & Rose, I. A., European J. Biochem. J_, 395 (1967).

Fortier, N.L., Gallard, L., & Lionetti, F.J., Arch. Biochem. Biophys. 119, 69 (19 6 7)•

Fuller, R. C., & Gibbs, M., Plant Physiol. 34, 324 (1959).

Fuller, R. C., Smillie, R. M., Sisler, E. C., & Kornberg, H.L., J. Biol. Chem. 236, 2140 (1961).

Gerhart, J. C., & Schachman, H.K., Biochemistry 1054 (1965).

Gilman, A., Philips, F.S., Koelle, E.S., Allen, R.P., & St. John, E. , Am. J. Physiol. 147, 1 15 (1946).

Hazelkorn, R., Fernandez-Moran, H,, Kieras, F.J., & Van Bruggen, E.F.J., Science 1 5Q, 1598 (1965)*

Heber, U., Pon, N.G., & Heber, M. Plant Physiol. 38,

355 (1963).

Hill, R.L., Advan._Protein Chen. 2_0, 37 (1965)*

Hill, R.L., Swe:. .son, R.T., & Schwartz, H.C., J. Biol. Chem. 235, 3182 (i960).

Hughes, T.R., & Klotz, I.M., Methods of Biochemical Analysis vol. 3 P« 265 (interscience Publishers, Inc.: New York). (1956).

Hurwitz, J., Jakoby, W.B., & Horecker, B.L., Biochim. Biophys. Acta 22, 194 (1956). 110.

Ikenaka, T., Baramerlin, H., Kaufmann, H., & Schmid, K., J. Biol. Chen. 24l_, 5560 ( 1966) .

Jacobi, G. & Perner, E. Flora 1J50,, 209 (1961).

Jeppson, J.O., & Sjoquist, J., Analyt. Biochem. 18,

264 (1967).

Kahn, A. & von Wettstein, D. J. Ultrastruci Res. J5,

557 (1961).

Kawashima, N., Plant & Cell Physiol. 10, 31 '1969)*

Kieras, F.J., & Haselkorn, R. , Plant Physiol. 43., 1264 (1968).

Kuehn, G.D., & McFadden, B.A., Biochemistry 8, 239^ (l969)»

Latzko, E., & Gibbs, M. , Plant Phy s i o 3.. 44, 295 ( 1 9 6 9 ) •

Liu, T J. Biol. Chom242, 402 9 (1967).

Lowry, O.H., Rosebrough, N.J», Farr, A.L., & Randall

R.J. , J . Biol. Chon. J£3, 265 (19 51).

Lyttleton, J.W., Bi_ochemJ. 64, 70 ( 1956).

Ly t tie ton, J . W. , & T'so, P.0.P, , Arch. Biocheni. Biophys .

23, 120 (1958).

MacElrcy, R.D. , Johnson, E . J . , & Johnson, M.K. Arch, Biochem Flo ■nays*„ 127, 3 JO ( 1 968 ) .

Margolis} J,, & Xenrick, K.G., Analy t_. Biochem. 25»

347 (1968).

Massaro, E.J., & Market, Cd. , J. Histochem. Cytochem. J_6_, 380 (1968).

Matsumoto, C., Sugiyama, T., Akazawa, T., & Miyachi, S,

Arc h, Bi. ochem. Biophys. 13 5, 282 (1969)* 111.

Mayaudon, J., Benson, A.A., & Calvin, M., Biochim. Biophys. Acta, 23, 342 (1957).

Mendiola, L. , and Akazawa, T. Biochemistry 3., 174 ( 1964) .

Mullhofer, G., & Rose, I.A., J, Biol. Chem. 240, 1341

(1965).

O'Donnell, I.J. & Thompson, E.O.P., Aust, J. Biol. Sci. 15, 740 (1962).

O'Donnell, I.J., & Thompson, E.O.P., Aust. J. Biol. Sc. 22, 271 (1964).

Park, R.B. & Pon, N.G., J. Mol. Biol. 2, 1 (1961)•

Paulsen, J.M., & Lane, M.D., Biochemistry 2, 2350 (1966).

Peterkofsky, A., & Racker, E., Plant Physiol. 36, 409 (1961).

Piez, K.A., Davie, E.W., Folk, J.E., & Gladner, J.A., J. Biol. Chem. 236. 2912 (1961).

Pon, N.G., Rabin, B.R., & Calvin, M. , Biochen. Z. 33.8, 7

(1963)-

Potts, J.T., Berger, A., Cooke, J., and Anfinsen, C.B., J. Biol. Chem. 237, 18-51 (1962)

Press, E.M., Piggot, P.J., & Porter, R.R., Biochem,J. 99, 356 (1966).

Putman, F.W. In "The Proteins" ed. H. Neurath vol. 3, p. 153 (Academic Press, Inc. New York) (1965)*

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

Rabin, B.R., & Trown, P.W., Nature 202, 1290 (1964). 112.

Rabin, B.R., & Trown, P.W., Proc, Natn, Acad. Sci. U.S.A. li, 497 (1964a)

Racker, E., Arch._Biochem. Biophys. 69, 300 (1957)•

Reindel, F. , & Hoppe, W. , Chern. Ber. 8_7, 1 103 (1954).

Ridley, S.M., Thornber, J.P., & Bailey, J.L., Biochim. Biophys. Acta. 140, 62 (1967)*

Rutner, A.C., Biochem. Biophys. Res. Commun. 391 923 (1970).

Rutner, A.C., & Lane, M.D., Biochem. Biophys. Res. Commun. 28, 531 (1967).

Sanger, F. , Biochem. J. 45 , 563 (1949)»

Sia, C.L., & Horecker, B.L., Biochem. Biophys. Res. Commun. 31, 731 (1968).

Silman, H.I., Cebra, J.J., & Givol, D., J. Biol, Chem. 237, 2196 (1962).

Singer, S.J., Eggman, L., Campbell, J.M. & Wiidman, S-^ J. Biol. Chem. 2.9.7, 233 (1952).

Sjoquist, J., Arkiv._Kemi V1_, 129 (1957)

Smillie, R.M. , Can. J. Bot. 41 , 123 (1963).

Smillie, R.M. , & Fuller, R.C., Plant Physiol. 3^• 551 (1959)

Smillie, R.M. , Rigopoulus, N., & Kelly, H., Biochim. Biophys. Acta. 56, 612 (1962).

Smith, E.L., Light, A., & Kimmel, J.R., Biochem. Soc. Symp, 21, 88 (1962).

Steer, M.W., Gunning, B.E.S., Graham, T.A., and Carr, D.J., Planta (Berl) 79, 254 (1968). Sugiyama, T., and Akazawa, T., J. Biochem, (Tokyo) 62■ , 474 (1967)-.

Sugiyama, T., Akazawa, T. , & Nakayama, N., Arch, Biochem. Biophys. 121, 522 (1967a).

Sugiyama, T,, Akazawa, T., Nakayama, N., & Tanaka, Y., Arch. Biochem. Biophys. 125, 107 (l968d)

Sugiyama, T,, Matsumoto, C., Akazawa, T., & Miyachi, S., Arch. Biochem. Biophys. 129. 597 (1969).

Sugiyama, T., Nakayama, N. & Akazawa, T., Arch, Biochem. Biophys, 126, 737 (1968a).

Sugiyama, T., Nakayama, N., & Akazawa, T., Biochem. Biophys. Res. Commun. 30, 118 (1968b).

Sugiyama, T., Nakayama, N., Ogawa, M., and Akazawa, T., Biochem. Biophys. Res. Commun. 28, 431 (1967b).

Sugiyama, T., Nakayama, N., Ogawa, M., Akazawa, T., and Oda, T. Arch. Biochem. Biophys. 125, 98 (1968).

Sugiyama, T., Nakayama, N., Tanaka, Y., & Akazawa, T., Arch, Biochem. Biophys. 126, 181 (1968c).

Thompson, E.O.P., Hosken, R., & Air, G.M., Aust, J. Bio1 Sc. 22, 449 (1969).

Thompson, E.O.P. & O'Donnell, I. J., Aust. J. Biol. Sc. J_2, 282 ( 1959).

Thompson, E.O.P., & O'Donnell, I.J., Aust. J. Biol. Sci. 18, 1207 (1965).

Thompson, E.O.P., & O'Donnell, I.J. Aust. J. Biol. Sci. 19, 1139 (1966). Thornber, J.P., Ridley, S.M., & Bailey, J. L., N.A.T.O. Advanced Study Institute, Biochemistry of Chloroplasts, Aberystwyth, vol. I, p. 275 (1966).

Thornber, J.P., Ridley, S.M., & Bailey, J.L., Biochem, J. 96, 29c (1965).

Trown, P.W., Biochemistry 4_, 908 (1965)* Trown, P.W., & Rabin, B.R., Proc. Natl. Acad. Sci. U.S., ^2, 88 (1964). Van Noort, G., & Wildman, S.G., Biochim. Biophys. Acta. 90, 309 (1964). Van Orden, H.O., and Carpenter, F.H., Biochem. Biophys. Res. Conmun. 14, 399 (19^4).

Warburg, 0., & Christian, W. , Biochem. Z. 310« 384 (1941). Warner, R.C., In "The Proteins" ed. H. Neurath & K. Bailey, vol. 2? pt. A, p. 435 (Academic Press, Inc. New York). Wassarman, P.M., & Major, J.P., Biochemistry 8_, 1076 (1969).

Weber, K., Nature 218, 1116 (1968).

Weber, K. , & Osborn, M. , _Biol. Chem. 244, 4406 (1969)* Weissbach, A., Horecker, B. L. & Hurwitz, J., J. Biol. Chem. 218, 795 (1956).

Whitaker, J. R., Anal. Chem. 35, 1950 (1963).

Wildman, S.G., & Bonner, J. Archs■ Biochem. 14, 3&1 (1947). Wilkinson, J.M., Press., E.M., & Porter, R.R., Biochem. j. 100, 303 (1966) 115.

WishJiick, M. , & Lane, M.D., J. Biol, Chem. 244, 55 (19^9)

Wish.nick, M. , Lane, M.D., Scrutton, M.C., & Mildvan, A.S. J. Biol. Chem. 244, 5761 (1969).

Wrigley, C., Science Tools 15, 17 (19^8). PUBLICATION

Moon, K.E., & Thompson, E.O.P., Subunits from reduced and S-carboxymethylated ribulose diphosphate carboxylase

(Fraction 1 protein) Aust, J. Biol, Sci. 22, 463 (1969)*