Molecular and Biochemical Parasitology, 14 (1985) 219-230 219 Elsevier

MBP 00513

THE PRESENCE OF ct-GLYCEROPHOSPHATE DEHYDROGENASE (NAD+-LINKED) AND ADENYLATE KINASE AS CORE AND INTEGRAL MEMBRANE ENZYMES RESPECTIVELY IN THE GLYCOSOMES OF TRYPANOSOMA RHODESIENSE

JOHN McLAUGHLIN

Department of Microbiology and Immunology, School of Medicine, University of Miami, Box 016960, Miami, Fla. 33101, U.S.A.

(Received 3 July 1984; accepted 5 October 1984)

A subcellular fraction enriched 12 times in glycosomes (NAD*qinked a-glycerophosphate dehydroge- nase) and devoid of detectable contamination from other subcellular components, was prepared from bloodstream Trypanosoma rhodesiense. Using a method employing exposure to toluene as a means of studying normally latent glycosomal enzymes, and phospholipase A2 as a membrane probe, the association of adenylate kinase and ct-glycerophosphate dehydrogenase with the glycosome was studied. The normally latent glycerophosphate dehydrogenase (NAD ÷ linked), it is proposed, is an intraglycosomal enzyme having no membrane association, but bound to the core by weak ionic linkages. As such it is possible to release the enzyme from permeable (toluene treated) glycosomes using CI-, with a resulting 4-fold increase in the K m for dihydroxyacetone phosphate. The presence of CI- also stimulates an increase in specific activity, but this is observed before any release of enzyme. In contrast adenylate kinase, a non-latent glycosomal enzyme, is clearly membrane associated, the use of phospholipase A2 revealing an absolute dependence on phospholipid for activity. Restoration of activity appears to specifically require phosphati- dyl choline and to be co-operative in nature (n H -- 1.56). It is proposed that adenylate kinase is an integral glycosomal membrane enzyme, probably affecting the control of intra-glycosomal ADP/ATP levels.

Key words: Trypanosoma rhodesiense; Glycosome; Adenylate kinase; Phospholipid dependence; Glycerophosphate dehydrogenase; Core enzyme

INTRODUCTION

Glycosomes are microbody-like organelles (e.g. peroxisomes, hydrogenosomes) apparently unique to the Trypanosomatidae, originally of singular importance as being the principal site for most of the enzymes involved in glycolysis [1,2]. Further

Abbreviations: EGTA, ethylene glycol bis([5-amino ethyl)-N,N,N',N',-tetraacetic acid; CDTA, trans-cyclo- hexane-l,2-diamine N,N,N',N'-tetraacetic acid; DCPIP, dichlorophenolindophenol; GPD, a-glycerophos- phate dehydrogenase (NAD ÷ linked); PNS, post-nuclear supernatant fraction.

0166-6851/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

220

investigation has established the presence of several non-glycolytic enzymes [3-8], though in some instances the association is temporal.

Only limited information is available describing the functional organization of microbodies, including glycosomes, especially the disposition of associated enzymes. Relevant investigations of peroxisomes suggest the membrane to possess two enzymes concerned with lipid metabolism, enoyl-CoA hydratase [9] and dihydroxyacetone phosphate acyl transferase [ 10,1 1] the latter enzyme now known to be localized in the glycosomes of procyclic Trypanosoma brucei [8]. The peroxisomal membrane poly- peptide composition is quite distinct from either mitochondrial or more significantly, endoplasmic reticulum membrane [12], a fact of much pertinence to the process of peroxisomal biogenesis [13]. As well as a limiting membrane an insoluble core has been isolated containing urate oxidase and a purported ' framework' protein [14].

In a previous study of bloodstream Trypanosoma rhodesiense [3] evidence was presented that not only is adenylate kinase a predominantly glycosomal enzyme, but its localization within the organelle was on the outer surface. In contrast several glycolytic enzymes, including et-glycerophosphate dehydrogenase (GPD) appeared to be localized within the interior of the glycosomes, bound by non-hydrophobic interac- tions.

The present study confirms the surface localization of adenylate kinase, but in addition establishes an intimate association with the glycosomal membrane, manifest- ed by an absolute dependence on phosphatidyl choline for enzyme activity. There is no such evidence for GPD being membrane associated, and by using permeable (toluene-treated) glycosomes it has been possible to study certain in situ properties of GPD.

METHODS

The Wellcome CT strain of T. rhodesiense is maintained in Sprague Dawley rats at The Rane Laboratory (U.M.). For the preparation of glycosomes, infected rat blood (approx. 150 ml) was processed to yield purified trypanosomes which were then disrupted with glass beads, and a post-nuclear supernatant fraction (PNS) prepared as described previously [15].

The PNS fraction was centrifuged at 6 500 rpm (5 100 X g) in a Sorvall SS-34 rotor fitted with 145 adapters for 10 min. The resulting supernatant fraction was removed and 8.5 ml layered over each of two discontinuous gradients: 7.5 ml, 54% (1.252); 10 ml, 42% (1.187); and 6.5 ml, 35% (1.112). Sucrose concentrations are expressed as w/w with the specific gravities (g cm -3) shown in parentheses. After centrifugation at 18000 rpm (33000 X g) for 75 min in a Sorvall SV-288 vertical rotor, a glycosome enriched fraction was recovered from the 42/54% sucrose interface. The sucrose concentration of the recovered fraction was first obtained, using an Abbe refracto- meter (American Optical) and after adding 17.5 ml Percoll (Pharmacia); sufficient 100 mM HEPES, pH 7.2, and 1.0 M sucrose were added to produce final concentrations of

221

10 mM and 250 mM respectively in a total volume of 39.5 ml. Centrifugation for 2.0 h at 18 000 rpm (42 500 X g) in a SS-34 rotor resulted in two closely separated bands in the upper third of the Percoll gradient. The lower band (G-L) proved to be enriched in glycosomes as judged by the increase in GPD activity (see Table I). For initial experiments trypanosomes were surface labelled using the fluorescamine-13-cyclodex- trin complex as described previously [16].

Toluene treatment of isolated glycosomes. The method employed was modified from one first described as a means of studying the in situ properties of normally latent mitochondrial enzymes [17]. To a 0.5 ml suspension of isolated glycosomes (4-5 mg protein) contained in 250 mM sucrose was added 2.0 ml PEG buffer (6% polyethylene- glycol, mol. wt. 6 000; 0.05% bovine serum albumin; 1.0 mM EDTA; 2.0 mM HEPES; 50 mM TRIS; 200 mM sucrose, pH 7.4). To this suspension ofglycosomes was added 44 lal of toluene (Malinckrodt-nanograde), giving a final concentration of 1.76% toluene, and after constant agitation for 2 min at 4°C, the glycosomes were sedimented by centrifugation at 18000 rpm for 2.5 min using the SS-34 rotor.

The small toluene layer was discarded and the underlying aqueous supernatant and pellet retained. The glycosome pellet was gently resuspended in PEG buffer and after further centrifugation the pellet finally suspended in 1.0 ml PEG buffer. The protocol for studying the effect of varying concentrations of KC1 on enzyme activities in toluene treated and normal glycosomes is detailed in Fig. 1.

Phospholipase exposure of isolated glycosomes. In order to investigate the role of membrane phospholipid on the latency/activity of glycosomal enzymes, intact glyco- somes were incubated with purified phospholipases. To isolated glycosomes (4-5 mg protein) in 1.0 ml buffered sucrose (250 mM sucrose; 10 mM HEPES, pH 7.4) were added 50 btl 200 mM Tris-acetate, pH 7.8, 24 lal 100 mM calcium acetate; 120 lal 5% BSA (fatty acid free, Sigma) and finally either 10 U porcine pancreas phospholipase A2 (Sigma, 833 U mg -1 protein), 10 U cobra venom phospholipase A2 (Sigma 1 200 U mg -~ protein) or 15 U Bacillus aureus phospholipase C (Sigma, 1 100 U mg -~ protein). The glycosome suspension was then incubated at 30°C and, at the intervals indicated on Figs. 2 and 3, 300 lal aliquots removed and further lipolytic activity stopped by mixing with 60 lal I0 mM EGTA. Tubes containing aliquots were all kept on ice until required for glycosomal enzyme assays. As a control a suspension of glycosomes was incubated as described, except for the omission of any phospholipase.

The procedure for reactivating adenylate kinase after exposure to phospholipase A2

is described in Fig. 4. As a visual check of phospholipase action, normal and treated glycosomes, obtain-

ed after a 30 min incubation, were extracted with CHC13/CH3OH (2:1), the solvent phase removed, reduced in volume and streaked onto a silica gel plate (Brinkman). Plates were developed using a solvent mixture of CHCi3/CH3OH/23% NH4OH (65:25:5), dried and visualized using a chromic acid spray [18]. Standards, obtained

222

from Sigma Chemical Co. containing dipalmitoyl phosphatidyl choline (and its lyso derivative) and palmitic acid were used for comparison.

Release of adenylate kinase from isolated glycosomes. To a suspension ofglycosomes in buffered sucrose was added one of a range of detergents, Triton X-100, Lubrol W (both from Sigma Chemical Co.) or Zwittergent 3-12 (Calbiochem-Hoechst)to a final concentration of 0.5% and centrifuged at 42 500 X g for 30 min. The supernatant fraction and pellet were removed and each assayed for adenylate kinase activity. Glycosomes were also sonicated [3] or exposed to phospholipase A2 or phospholipase C for 30 min (see above), centrifuged as before and the pellet and supernatant fractions assayed for adenylate kinase.

Enzyme assays. The glycosomal enzymes GPD (NAD ÷ linked), hexokinase and adenylate kinase as well as mitochondrial GPD (DCPIP linked) were assayed as previously stated [3]. The assay of acid phosphatase, using ct-glycerophosphate as substrate, has also been described previously [15]. The surface membrane Ca2*-ATP - ase is to be described in detail elsewhere: briefly, the enzyme assay used involves a Ca2÷-CDTA buffer (100 nmol free Ca 2÷) with 1 mM ATP (vanadate free, Sigma) and 30 mM glycylglycine, pH 8.0. After incubation for 20-30 min at 30°C liberated PO4 was measured as described in a preceding paper [19].

RESULTS

Isolation of glycosomes. The method described for the isolation of glycosomes resulted in a 12-fold purification and 31% recovery as judged by GPD activity for fraction G-1 (Table I). This compares favorably with a similar procedure described recently by Opperdoes et al. [20]. From the present results it is evident that there is little detectable contamination with mitochondria (GPD-DCPIP linked), flagella pocket membrane (acid phosphatase) or surface membrane (relative fluorescence, Ca2+-ATPase). The distribution of these markers in T. rhodesiense with the exception of Ca:÷-ATPase (which is to be described in detail elsewhere) has been previously established [15]. Surface membrane is a particularly serious source of potential contamination, the presence of an attached microtubular system resulting in an unusually high equilibrium density, similar to that of glycosomes [15].

Adenylate kinase, whilst predominantly glycosomal is also found in the cell sap and mitochondrion [3]. In this study, 28% of the enzyme was recovered in fraction G-L (glycosomes) being 9.5 times enriched (specific activity 12.8 U mg -1 protein).

Effect of toluene treatment on glycosomes. The use of toluene treatment as a means of studying the in situ properties of glycosomal core enzymes was evaluated. It is evident from the results of Table II that for the three glycosomal enzymes studied, treatment caused neither loss of activity nor release of enzyme, all remaining fully sedimentable.

223

TABLE I

Distribution of subcellular components between those fractions obtained during the purification of

glycosomes from T. rhodesiense

Enzyme Specific

activity

(mU mg -~ protein)

% distribution for each fraction

H N Pt St 52/45 G-L

Protein 107 + 161 29 15 56 6.5 3

GPD (NAD linked) 225 + 34 15 39 46 39 34

GPD (DCPIP linked) 115 + 22 21 37 42 9 N.D.

Ca2*-ATPase 15 + 6 28 58 14 12 N.D.

Acid phosphatase 34 + 12 33 14 53 7 N.D.

Relative fluorescence (%) 1002 31 27 42 9 0.7

H = homogenate; N = nuclei/cell debris; Pt = pellet; St = supernate both obtained after 10 min (5 100 × g)

centrifugation. The 52/45 and G-L fractions were obtained after discontinuous sucrose gradient and

Percoll gradient centrifugation, respectively (see Methods). Results based on four experiments with

standard deviation given in parentheses. 1 unit of enzyme activity = 1 mol product formed per min at 30°C.

N.D., none detected.

TABLE II

Comparison of some properties of toluene treated and normal glycosomes

Enzyme Latency (o~ )a Sedimentability (%)b Km (mM)C

Normal Treated Normal Treated Free Bound

a-glycerophosphate 96 0 100 95

dehydrogenase

(NAD-linked) Hexokinase 97 0 98 91

Adenylate kinase 0 0 d 100 100

0.26 (0.07) 0.065 (0.012)

Free enzyme refers to activity released after exposure to 200 mM KCI, and bound to that associated with

toluene treated glycosomes. Final concentration of 1.5 mM KCI for both assays.

a Latency values expressed as the difference between activity assayed in the presence of 250 mM sucrose

with (total) and without (free) 0.1% Triton X-100, as a percentage of the total activity. b Sedimentability determined after centrifugation at 10000 rpm for 12 min, using an SS-34 rotor.

c Km values calculated using direct linear plots [21] of velocity as a function of substrate concentration.

Standard deviation given in parentheses, based on four determinations.

d Exhibits approximately 20% increase in activity.

224

There was a slight but consistent increase (15-20%) in adenylate kinase activity which may reflect the proposed membrane association of this enzyme. Of more significance was the complete loss of latency observed for both GPD and hexokinase, indicating

that toluene treatment was permitting unrestricted access ofsubstrates and co-factors to these still bound enzymes. The K m values of free (released) G P D as compared to enzyme still bound to toluene treated glycosomes, shown in Table II, is 4 times greater, indicating that in situ bound to the glycosome G P D can respond more effectively to reduced substrate concentrations.

The stimulatory effect of CI- on T. rhodesiense G P D was previously described [3]. The influence of KC1 on G P D using both normal and toluene treated glycosomes was

compared and, as would be expected from the increased permeability of toluene treated glycosomes, G P D was activated by KC1 (Fig. la), the effect being most apparent up to 20-25 mM KCI. With untreated glycosomes G P D activity remained virtually unchanged over the entire KC1 concentration range. From Fig. lb it is

apparent that KCI also effects the release of GPD from toluene treated glycosomes. Noticeable however, is the lack of identity between the plot for release and that for activation. Whereas activation of G P D was most pronounced between 0-25 mM KC1, very little enzyme was released over this same concentration range. Above 25 mM KC1

G P D release increased rapidly, whilst the rate of enzyme activation decreased. Adenylate kinase was neither activated nor released over the entire range of KCI

concentrations used above, regardless of whether normal or toluene treated glyco- somes were used.

The effect of phospholipase on glycosomal enzyme activity. The data presented in Fig. 2 shows that exposure of glycosomes to phospholipase A2 results in a rapid loss of

.=_

o Q.

._u

l.q 0 2 5 5 0 7 5 1~X:)

loc

@

5c 'y, g

KCI mM

B

0 2~ 50 75 100

Fig. 1. The effect of KCI on the GPD of normal and toluene treated glycosomes. (A) Increase in GPD

activity for toluene treated (~" *) and normal (A A) glycosomes. (B) Release of GPD from toluene

treated (g ") and normal (e e) glycosomes as a function of KCI concentration. Enzyme release was

determined by mixing glycosomes (500-750/ag), suspended in buffered sucrose, with varying amounts of 500

mM KC1 and 1.0 M sucrose in a final volume of 0.5 ml, to an osmolarity equivalent to 250 mM sucrose. After

centrifugation of the glycosome suspension at 33 000 × g for 30 min, an aliquot of the supernatant solution

was used for enzyme determinations. For all assays of GPD a constant concentration of 10 mM KC1 was maintained.

225

100

v

50

._1

i I I i 0 15 30 45

Time ( ra in)

100

>

5c

a

i I I I 0 15 30 45

Time ( min )

Fig. 2. The loss of GPD latency on exposure of glycosomes to phospholipase A 2. Purified glycosomes were incubated with porcine pancreas phospholipase A2 and aliquots removed at the time intervals indicated (A A) as described in the Methods. For comparison glycosomes were incubated without adding

phospholipase (* k). Latency is as defined in Table II.

Fig. 3. The loss of adenylate kinase activity on exposure of glycosomes to phospholipase A2. Purified glycosomes were incubated with ( o - - o ) and without (u u) porcine pancreas phospholipase A2,

aliquots being removed at the time intervals indicated as described in Methods.

latency for GPD, with no loss of activity. In complete contrast adenylate kinase activity decreased rapidly (Fig. 3), the loss of activity appearing to be related specific- ally to membrane phospholipid hydrolysis, rather than a non-specific detergent induced effect due to the products of phospholipase digestion. The effect was progres- sive with time and the inclusion of bovine serum albumin in the assays should have removed any surface active end products [ 16]. Adenylate kinase activity also decreas- ed after exposure to phospholipase C, though the effect was less pronounced being only 64% and 43% that of controls after 10 and 30 min, respectively.

Visual inspection of thin-layer chromatography plates revealed an increase in components comigrating with lysophospholipid (lysophosphatidyl choline standard) for lipid extracts from phospholipase A2-exposed glycosomes as compared to normal glycosomes.

Glycosomal adenylate kinase has an absolute requirement for phosphatidyl choline. In order to confirm the existence of an essential phospholipid requirement for adenylate kinase, indicated by the results shown in Fig. 3, an attempt was made to restore enzyme activity to phospholipase A2 exposed glycosomes using sonically dispersed phospholipid. The exact protocol followed and results obtained are shown in Fig. 4. Restoration of adenylate kinase activity to glycosomes previously exposed to porcine pancreas phospholipase A was almost 100% when dipalmitoyl phosphatidyl choline was used, occurring rapidly above 300 Ixg ml -~ lipid. This rapid increase points to

226

100

o ~c

>_-

I~lgJ I I I 100 200 300 400

/Jg Phospholipid/ml

OI 500

Fig. 4. Phospholipid dependent restoration of adenylate kinase activity for glycosomes previously exposed to phospholipase A 2. Glycosomes were first incubated with 6.25 U porcine pancrease (A A) or cobra venom ( , , ) phospholipase A2 for 30 min at 30°C (see Methods for buffered incubation mixture) then aliquots added to increasing concentrations of sonically dispersed dipalmitoyl phosphatidyl choline. Glycosomes exposed to porcine pancreas phospholipase A2 as above then mixed with sonically dispersed dipalmitoyl phosphatidyl ethanolamine (o o). A stock solution of phospholipid (5 mg ml -l) was prepared in 250 mM sucrose, 10 mM HEPES, pH 7.2, by sonication (Braunsonic 1510) at 300W for 3 separate, 3 rain periods. For phosphatidyl ethanolamine 0.05% Triton X-100 was used in some instances to aid dispersion.

enzyme reac t iva t ion being a coopera t ive process , a c o m m o n feature of the b ind ing

i so therms ob ta ined for the assoc ia t ion o f a m p h i p h i l i c l igands (e.g. phospho l ip id ) with

p ro te in [22]. A Hill p lot o f the da t a f rom Fig. 4 gives a value for n H, = 1.56 ( r 2 = 0.97)

which tends to impl ica te a coopera t ive process. More precise i n fo rma t ion using

pur i f ied enzyme would be requi red before a t t empt ing to es tabl ish the mechan i sm

involved in enzyme react iva t ion . F o r g lycosomes exposed to cob ra venom phospho l i -

pase A2 only 55% of the or ig inal act ivi ty was res tored.

The need for p h o s p h o l i p i d was specific, since a range of detergents bo th non- ionic

(Tr i ton X-100, Lubro l w) and zwit ter ionic (Zwi t te rgent 3-12, 10) were wi thout effect

in res tor ing activity. In add i t ion , reac t iva t ion was only observed with phospha t idy i

chol ine , phospha t idy l e thano lamine as can be seen f rom Fig. 4 having no effect even if

d i spersed in the presence of 0.05% of Tr i ton X-100.

Release of adenylate kinase in an enzymatically active form. Previous a t t empts [3] to

d issocia te adenyla te kinase f rom glycosomes using Tr i ton X-100 or son ica t ion were

once more to ta l ly unsuccessful , all o f the act ivi ty being recovered in the pellet af ter

cent r i fugat ion . By using the zwit ter ionic detergent Zwit tergent 3-12 it was poss ible to

affect release of 30% of the to ta l act iv i ty (38% of that recovered) . Inc lus ion of

phospha t idy l chol ine p r o d u c e d a mode ra t e increase to 42% the to ta l act ivi ty released.

227

There was no demonstrable release of adenylate kinase after exposure of glycosomes to either phospholipase C or phospholipase A2, even if up to 350 lag ml -~ sonically dispersed dipalmitoyl phosphatidyl choline were added to the supernatant fraction after centrifugation of the exposed glycosomes. None of the above treatments released

any of the GPD.

DISCUSSION

The most significant finding of the present investigation is the clear demonstration of the intimate nature of the membrane association between adenylate kinase and the glycosomal membrane. This is a totally unique finding, for in all other cells this enzyme is free, either in the cell sap or the mitochondrial inter-membrane space [23]. Phospholipids have been observed to exert a profound influence on the activity of a number of membrane associated enzymes. In some instances the in situ expression of activity is constrained [24-26], exposure to phospholipases or detergent relieving the

effect with a consequent increase in activity. For other membrane associated enzymes, most notably cytochrome oxidase [27]

and a number of ATPases, catalytic activity is wholly or in part dependent on the presence of phospholipid. Thus a recent report concerning the Ca2++Mg2+-ATPase of human platelets [28] describes the loss of activity occurring after exposure to phos- pholipases C or A. Activity could be restored using phosphatidyl serine or phosphat- idyl inositol but not with phosphatidyl ethanolamine, phosphatidyl choline or sphingomyelin. This specificity with regard to the type of phospholipid required has been extensively investigated for a number of membrane bound enzymes and trans- port proteins, particularly the sarcoplasmic reticulum ATPase. In this latter instance it has been found that the nature of the phospholipid polar head group influences Ca 2÷ transport and the fatty acid side chains the ATPase function [30].

Unfortunately little is known concerning the regulatory effect of phospholipid on membrane enzyme function for any of the parasitic protozoa. In an earlier report of a surface membrane Ca2+-ATPase in Entamoeba histolytica [ 19], exhaustive attempts to establish a regulatory role for lipid, including phospholipase exposure, produced no supporting evidence. The glycosomal adenylate kinase described here appears to be the first instance where a direct dependence on a specific phospholipid has been established for any enzyme, in a protozoan parasite. The efficacy of phosphatidyl choline in restoring adenylate kinase activity to phospholipase treated glycosomes is compatible with the fact that together with phosphatidyl ethanolamine, these are the only two phospholipids present in T. brucei glycosomes [20]. This requirement ap- pears to be specific, since phosphatidyl ethanolamine is totally ineffective in restoring

activity. Based upon the results of both this and a previous investigation [3] the possible

intra-organellar disposition of GPD and adenylate kinase as glycosomal core and

228

membrane enzymes respectively is depicted schematically in Fig. 5. From a considera-

t ion o f tile latency and use of trypsin and suramin as membrane probes [3] plus the C1-

release data, G P D is shown bound to the core of the glycosome (possibly by analogy

with peroxisomes, 14, to some non-catalytic f ramework protein) through weak ionic linkages. Such a proposal is in conflict with earlier findings of Opperdoes and Nwaya

[31] where G P D was purpor ted to be associated with the glycosomal membrane. The

present evidence is unequivocal , even after toluene treatment G P D remained fully

sedimentable following subsequent exposure to either 0.25% Tri ton X-100 or Zwitter-

gent 3-12. If similarly treated glycosomes are exposed to KCI, release of G P D is

readily observed, whereas none occurs for intact glycosomes. The findings of a more recent investigation, that there is no evidence of any interaction between the multi-en-

zyme complex and glycosomal phosphol ipid [32], is fully compatible with the results presented in this study.

The fact that activation of G P D occurs at C1- concentrat ions below those permit-

ting any substantial release of enzyme from toluene treated glycosomes, suggests that

any significant conformat ional changes affecting activity are occurr ing whilst the

enzyme remains sequestered within the glycosome. Whether this might be of any

relevance to the in situ modula t ion o f core enzyme activities is purely speculative at

present. It has, however, been firmly established that changes in the electrostatic environment of bound enzymes can exert a regulatory effect [33].

Fig. 5 depicts adenylate kinase as an integral membrane protein with a por t ion

~,~-'ie~ .--.4 Phospholipid

- ~ - - C . = - ~ . ~ , Glycosome ~e--- ~ ~ATP i ~ I G P O

Membrane Cytosol

Adenylate kinase --~ PhosphatJdyl choline

•1• Translocation

~ '+ Glycosome core

Phosphofructokinase

Fig. 5. Proposed disposition of adenylate kinase and (NAD ÷ linked) ct-glycerophosphate dehydrogenase within the glycosome of T. rhodesiense. Adenylate kinase is shown in the glycosomal membrane, a portion penetrating the lipid bilayer in a region enriched in phosphatidyl choline. It is postulated that the enzyme could regulate intra-glycosomal [ATP/ADP], possibly involving translocation across the glycosomal membrane. The GPD is depicted bound to the core of the glycosome by weak ionic interactions, and possibly in a functionally constrained form. The other glycolytic enzymes, phosphofructokinase and hexokinase are also shown similarly bound to the core (see [3]).

229

penetrating the lipid bilayer of the glycosomal membrane in a region enriched in phosphatidyl choline. Possibly the enzyme completely spans the membrane, however there is at present no supporting evidence. The intimate membrane association of this enzyme suggests, by analogy with other lipid dependent membrane enzymes, some regulatory/transport function. This could, as indicated in Fig. 5, involve the control of intraglycosomal A T P / A D P levels, thereby exerting a modulating effect on those glycolytic sequences within the glycosome. An important constituent of the inner mitochondrial membrane is an A D P / A T P counter-transport protein known to have an essential phospholipid requirement [34]. Since it is possible to release adenylate kinase in an active form, this should permit the use of liposome reconstituted enzyme

to identify any associated A D P / A T P transport function.

ACKNOWLEDGEMENTS

The author is deeply indebted to Dr. Arba Ager, Jr. and the staff of The Rane Laboratory, University of Miami for supplying T. rhodesiensae-infected rat blood.

REFERENCES

1 Opperdoes, F.R. and Borst, P. (1977) Localization of nine glycolytic enzymes in a microbody like

organelle in Trypanosoma brucei. FEBS Lett. 80, 360-364. 2 Taylor, M.B., Berghausen, H., Heyworth, P., Mersenger, N., Rees, L.J. and Gutteridge, W.E. (1980)

Subcellular localization of some glycolytic enzymes in parasitic flagellated protozoa. Int. J. Biochem.

11, 117-120. 3 McLaughlin, J. (1981) Association of adenylate kinase with the glycosome of Trypanosoma rhode-

siense. Biochem. Int. 2, 345-353. 4 Hammond, D.J., Gutteridge, W.E. and Opperdoes, F.R. (1981) A novel location for two enzymes

of de novo pyrimidine biosynthesis in Trypanosomes and Leishmania. FEBS Lett. 128, 27-29. 5 Opperdoes, F.R., Markos, A. and Steiger, R.F. (1981) Localization of malate dehydrogenase,

adenylate kinase and glycolytic enzymes in glycosomes and the threonine pathway in the mitochon-

drion of cultured procyclics of Trypanosoma brucei. Mol. Biochem. Parasitol. 4, 291-310. 6 Opperdoes, F.R. and Cottem, D. (1982) Involvement of the glycosome of Trypanosoma brucei in

carbon dioxide fixation. FEBS Lett. 143, 60-64. 7 Broman, K., Knuffer, A.L., Ropars, M. and Deshusses, J. (1983) Occurrence and role of phosphoe-

nolpyruvate carboxykinase in procyclic Trypanosoma brucei brucei glycosomes. Mol. Biochem.

Parasitol. 8, 79-87. 8 Opperdoes, F.R. (1984) Localization of the initial steps in alkoxyphospholipid biosynthesis in

glycosomes (microbodies) of Trypanosoma brucei. FEBS Lett. 169, 35-39. 9 Reddy, M.K., Hollenberg, P.F. and Reddy, J.K. (1980) Partial purification and immunoreactivity of

80 000 molecular weight polypeptide associated with peroxisome proliferation in rat liver. Biochem.

J. 188, 731-740. 10 Hajra, A.K., Burke, C.L. and Jones, C.L. (1979) Subcellular localization of acyl coenzyme A:

dihydroxyacetone phosphate acyltransferase in rat liver peroxisomes (microbodies). J. Biol. Chem.

254, 10896-10900. 11 Jones, C.L. and Hajra, A. (1984) Solubilization and partial characterization of dihydroxyacetone-

phosphate acyl transferase from guinea pig liver. Arch. Biochem. Biophys. 226, 155-165.

230

12 Fujiki, Y., Fowler, S., Shio, H., Hubbard, A.L. and Lazarow, P.B. (1982) Polypeptide and phospholi- pid composition of the membrane of rat liver peroxisomes: comparison with endoplasmic reticulum and mitochondrial membranes. J. Cell. Biol. 93, 103-110.

13 Lazarow, P.D., Robbi, M., Fujiki, Y. and Wong, L. (1982) Biogenesis ofperoxisomal proteins in vivo and in vitro. Ann. N.Y. Acad. Sci. 386, 285-297.

14 Watanabe, T., Suga, T. and Hayashi, H. (1977) Studies on peroxisome. VIII. Evidence for framework protein of the cores of rat liver peroxisomes. J. Biochem. (Tokyo) 82,607-609.

15 McLaughlin, J. (1982) Subcellular distribution of particle-associated antigens in Trypanosoma rhode- siense. J. Immunol. 128, 2656-2663.

16 McLaughlin, J. (1984) Evidence for lipid-protein interactions in the attachment of antigens to a low-density membrane fraction isolated from Trypanosoma rhodesiense. Infect. Immun. 43,294-301.

17 Matlib, M.A., Shannon, W.A. and Srere, P.A. (1979) Measurement of matrix enzyme activity in situ in isolated mitochondria made permeable with toluene. Methods Enzymol. 46,544-550.

18 Kates, M. (1972) Techniques oflipidology. In: Laboratory Techniques in Biochemistry and Molecular Biology (Work, T.S. and Work, E., eds.), pp. 269-610. North Holland, Amsterdam.

19 McLaughlin, J. and Muller, M. (1981) A calcium regulated adenosine triphosphatase in Entamoeba histolytica. Mol. Biochem. Parasitol. 3,369-379.

20 Opperdoes, F.R., Baudhuin, P., Coppens, I., De Roe, C., Edwards, S.W., Weijers, P.J. and Misset, O. (1984) Purification, morphometric analysis and characterization of the glycosome (microbodies) of the protozoan hemoflagellate Trypanosoma brucei. J. Cell. Biol. 98, 1178-1184.

21 Cornish-Bowden, A. (1979) Fundamentals of Enzyme Kinetics. Butterworths, London. 22 Tanford, C. (1973) The Hydrophobic Effect. John Wiley and Sons, New York. 23 Noda, L. (1973) Adenylate kinase. In: The Enzymes (Boyer, P., ed.), Vol. 8A, pp. 279-305. Academic

Press, New York.

24 Dipple, I., Elliot, K.R.F. and Houslay, M.D. (1978) Detergents modify the form of Aarhenius plots of 5'-nucleotidase activity. FEBS Lett. 89, 153-157.

25 Zakim, D., Goldenberg, J. and Vessey, D.A. (1973) Regulation of microsomal enzymes by phospboli- pids. VII. Abnormal enzyme-lipid interactions in liver microsomes from Gunn rats. Biochim. Biophys. Acta 297,497-502.

26 Sandermann, H. (1980) Regulation of membrane enzymes by lipids. Biochim. Biophys. Acta 515, 209-237.

27 Robinson, N.C. (1982) Specificity and binding affinity of phospholipid to the high affinity cardiolipin sites of beef heart cytochrome c oxidase. Biochemistry 21, 184-188.

28 De Metz, M., Lebret, M., Enouf, J. and Levy-Toledano, S. (1984) The phospbolipid requirement of the (Ca2*-Mg2+)-ATPase from human platelets. Biochim. Biophys. Acta 770, 159-165.

29 Malathi, P. (1983) Liposome reconstitution: application in cell physiology. In: Liposomes (Ostro, M.J., ed.), pp. 125-144, Marcel Dekker, New York.

30 Bennett, J.P., McGill, K.A. and Warren, G.B. (1980) The role of lipids in the functioning of a membrane protein: the sarcoplasmic reticulum calcium pump. In: Current Topics in Membranes and Transport (Bonner, F. and Kleinzeller, A., eds.), pp. 127-164, Academic Press, New York.

31 Opperdoes, F.R. and Nwaya, M. (1980) Sub-organellar localization of glycolytic enzymes in the glycosome of Trypanosoma brucei. In: The Host-Invader Interplay (Van der Bossche, H., ed.), pp. 683-686, Elsevier, Amsterdam.

32 Hart, D.T., Misset, O., Edwards, S.W. and Opperdoes, F.R. (1984) A comparison of the glycosomes (microbodies) isolated from Trypanosoma brueei bloodstream form and cultured procyclic trypomas- tigotes. Mol. Biochem. Parasitol. 12, 25-36.

33 Douzou, P. and Maurel, P. (1977) Ionic control of biochemical reactions. Trends Biochem. Sci. 2, 14-17.

34 Kramer, R. and Klingenberg, M. (1980) Enhancement of reconstituted ADP/ATP exchange activity by phosphatidyl ethanolamine and by ionic phospholipids. FEBS Lett. 119, 257-260.