Pyruvate-Converting Activity in the Spores of the Microsporidian Genus Paranosema (Antonospora) Viacheslav V

Pyruvate-converting activity in the spores of the microsporidian genus Paranosema (Antonospora) Viacheslav V. Dolgikh & Ruslan I. Al-Shekhadat

Laboratory of Microbiological Control, All-Russian Institute for Plant Protection, St Petersburg, Russia

Correspondence: Viacheslav V. Dolgikh, Abstract Laboratory of Microbiological Control, All- Downloaded from https://academic.oup.com/femsle/article/259/1/142/453567 by guest on 03 October 2021 Russian Institute for Plant Protection, 3 Microsporidia, a large group of fungi-related protozoa with an obligate intracel- Podbelskogo Rd., Pushkin, St. Petersburg lular lifestyle, are characterized by a drastically reduced cell machinery and a 196608, Russia. Tel.: 17 812 4705110; fax: unique metabolism. These parasites possess genes encoding glycolysis components 17 812 4705110; e-mail: and glycerol-phosphate shuttle, but lack typical mitochondria, Krebs cycle, [email protected] respiratory chain and pyruvate-converting enzymes, except for two subunits of

the E1 enzyme of the pyruvate dehydrogenase complex. This study demonstrates Received 8 December 2005; revised 28 March that in spite of the above, destroyed spores of the microsporidian Paranosema 2006; accepted 28 March 2006. (Antonospora) grylli and P. locustae deplete pyruvate content in the incubation First published online May 2006. medium. This activity is sensitive to heat, proportionally distributed between the doi:10.1111/j.1574-6968.2006.00259.x soluble and the insoluble fractions and does not depend on additional ions or cofactors. Editor: Claire Remacle

Keywords microsporidia; spores; pyruvate metabolism; enzyme activity.

the mitosome, rather than in the accumulation of glycerol- Introduction 3-phosphate as an end product of glycolysis. Thus, all trioses Microsporidia make up a large group of intracellular formed in glycolysis should be finally converted into pyr- eukaryotic parasites infecting a wide range of animals and uvate. At the same time, according to the data from the some protozoan species. Several genera are of medical human parasite Encephalitozoon cuniculi genome project, significance since they infect humans, mostly immunodefi- microsporidia lack any pyruvate-converting enzymes except cient patients. As a result of their intracellular lifestyle, these two subunits of the E1 component of the pyruvate dehydro- fungi-related parasites have lost lysosomes, peroxisomes, genase complex (PDH). Neither E2 nor E3 enzymes of PDH typical mitochondria, a classical Golgi complex, many were found in the microsporidial genome (Katinka et al., metabolic pathways and reduced their genome down to the 2001). The presence of genes encoding both E1 PDH minimal level possible (Katinka et al., 2001). Microsporidia subunits was also demonstrated for the distantly related to possess genes encoding enzymes of the Embden–Meyerhof E. cuniculi microsporidia Paranosema (Antonospora) locus- pathway, but lack tricarboxylic acid cycle and respiratory tae (Fast & Keeling, 2001). Recently, participation of micro- chain components. Since glycerol-3-P dehydrogenase (G- sporidial PDH E1 in the further metabolizing of pyruvate 3-PDH) has been shown to be a single potential enzyme was supported by the demonstration of expression of both reoxidizing NADH generated in glycolysis (Dolgikh et al., subunits in P. locustae spores (Williams & Keeling, 2005). 1997; Katinka et al., 2001), glycerol-3-phosphate was con- In this study, we have shown that destroyed spores of the sidered as an end product of anaerobic catabolism. However, microsporidian Paranosema (Antonospora) grylli deplete in addition to G-3-PDH, the second (mitochondrial) com- pyruvate content in the incubation medium. This activity ponent of glycerol-phosphate shuttle was found in micro- was sensitive to heat, present in soluble and insoluble sporidial genome (Katinka et al., 2001). Furthermore, fractions and did not require any additional ions or cofac- mitochondria-derived cryptic organelles called ‘mitosomes’ tors. However, dialysis of spore proteins followed by pre- were described in the microsporidian Trachipleistophora cipitation with 80% ammonium sulphate caused enzyme hominis (Williams et al., 2002). This suggests that G- inactivation. Broken spores of the closely related species 3-PDH participates in the transportation of electrons into P. locustae demonstrate a similar rate of pyruvate-converting

c 2006 Federation of European Microbiological Societies FEMS Microbiol Lett 259 (2006) 142–146 Published by Blackwell Publishing Ltd. All rights reserved Pyruvate-converting activity in Paranosema (Antonospora) 143 activity in both soluble and insoluble fractions. These at 4 1C and 1.4 mL of supernatant was loaded onto the findings suggest the presence of a functional protein in column. Equilibration of the column and elution was microsporidial spores. The reaction catalysed by this enzyme carried out with TS. An eluted volume corresponding to could be oxidative decarboxylation of pyruvate with free the unconfined space of the column (2 mL) was discarded acetate production. and the next 1.5 mL was collected for specific activity determination. Materials and methods For dialysis and precipitation with ammonium sulphate (SA), 0.5 mL of P. grylli spore pellet was destroyed in 6 mL of Microsporidial spores 50 mM Tris-HCl pH 8.0 containing 0.5 mM PMSF, 0.5 mM EDTA-Na2, 0.5 mM 2-ME and centrifuged at 300 000 g for To avoid bacterial contamination, Paranosema grylli and 20 min. The supernatant was dialyzed against 1 L of 10 mM Paranosema locustae spores were isolated from freshly pre- Tris-HCl pH 8.0 containing 50 mM NaCl, 0.05 mM PMSF, Downloaded from https://academic.oup.com/femsle/article/259/1/142/453567 by guest on 03 October 2021 pared fat bodies of artificially infected crickets Gryllus 0.1 mM EDTA-Na2, 0.1 mM 2-ME overnight at 4 1C. Dia- bimaculatus and locusts Locusta migratoria. Mature spores lyzed proteins were precipitated by addition of small por- were purified as described previously (Seleznev et al., 1995; tions of dry SA (Serva, Germany) up to 80% on the ice bath Dolgikh et al., 1997). The quality of spore purification was with gentle stirring and pH control. The suspension was checked under a light microscope. Purified spores were stored for 4 days at 4 1C and centrifuged at 3000 g for destroyed by shaking with 2.5 mm glass balls (Bio-Rad) on 20 min. The pellet was resuspended in 20 mM Tris-Cl pH a Vortex shaker at 4 1C for 30 min in Buffer A (50 mM Tris- 7.8, 80% SA with the volume brought up to 1/10 of that of Cl pH 8.0) containing 0.5 mM phenylmethanesulfonyl the SA suspension and pyruvate-converting activity was fluoride (PMSF), 0.5 mM EDTA-Na2, 0.5 mM 2-mercap- determined in the pellet and in the supernatant. The toethanol (2-ME) or in Tris-sucrose solution (TS, 50 mM calculated concentration of SA in activity assay medium Tris-Cl pH 8.0, 0.25 M sucrose). The suspension of broken was 2% in the case of pellet and 20% in the case of spores was centrifuged as mentioned in Results and pellets supernatant. Protein concentrations were determined by were resuspended in the same solution, with the volume the Bradford method (Bradford, 1976). brought up to that of the final supernatant.

Pyruvate-converting activity assay Results and discussion Spore homogenate, supernatant or resuspended pellet were For the first experiment, freshly isolated spores of the mixed with an equal volume of 50 mM potassium phosphate microsporidian P. grylli and P. locustae were destroyed in Buffer A and pyruvate-converting activity was measured in buffer (pH 7.5) containing 10 mM MgCl2, 0.4 mM pyru- the spore homogenate. It was determined that incubation of vate-Na2 (Serva, Germany) to obtain a final concentration of 0.2 mM pyruvate. The buffer used for protein extraction or spore proteins in the presence of 0.2 mM pyruvate (final centrifuge pellet resuspension was added in the control concentration) for 16 h at room temperature caused the disappearance of about 0.4 mmol of substrate per 100 mLof tubes. Reaction mixtures were incubated for 16 h (over- 9 night) at room temperature or for 1 h at 30 1C, boiled for spore pellet (1.5 10 spores). To obtain further informa- 10 min with subsequent sedimentation of denatured materi- tion, P. grylli spores were broken in the isotonic TS solution al by centrifugation. Pyruvate content in the supernatant and centrifuged at 3000 g for 20 min. Some supernatant and 1 was measured by enzymatic method at a wavelength of the pellet resuspended in TS were incubated at 100 C for 340 nm using lactic dehydrogenase (LDH) activity. The 10 min before addition to the activity assay medium. This assay mixture comprised 50 mM Tris-Cl pH 7.5, 1.3 mM heat treatment completely inactivated the putative enzyme NADH (Reanal, Hungary), 3 U mL1 LDH (ICN, CA) and in the supernatant and in the pellet. Total and specific (per the aliquot of supernatant. The differential molar extinction mg protein) activities were found to be comparable in both of NADH and NAD at 340 nm equalled 6220 M1 cm1. fractions (Table 1). To investigate the distribution of pyruvate-converting activity in the cell, P. grylli spores destroyed in TS were Heating, gel filtration, dialysis and ammonium centrifuged in series at 100 g for 10 min, at 3000 g for 20 min sulphate precipitation of P. grylli spore proteins and finally at 300 000 g for 20 min. Analysis of centrifuge The heating of samples was carried out by boiling for 10 min fractions showed once more the sedimentation of about half before the activity assay. The desalting of the spore extract of the total activity at 3000 g for 20 min. However, another was performed on a Sephadex G-25 column 1.5 5cm part of the activity was not pelleted even by high-speed (Pharmacia), according to the manufacturer’s instructions. ultracentrifugation (Table 2), suggesting that the enzyme is Spores broken in TS were centrifuged at 13 000 g for 30 min soluble. A similar distribution of pyruvate-converting

Table 1. Pyruvate-converting activity in Paranosema grylli spore homogenate after centrifugation at 3000 g for 20 min ActivityÃ Protein Speciﬁc activity % of total Activity after Fraction (nmol mL1) (mg mL1) (nmol mg1 protein) activityw inactivation at 100 1C Pellet 194 1.4 140 41 0 Supernatant 283 2.8 100 59 0

ÃIncubation for 16 h at room temperature. wThe volumes of ﬁnal supernatant and resuspended pellets were 1.2 mL.

Table 2. Distribution of Paranosema grylli pyruvate-converting activity in the fractions after centrifugation Activity (nmol per 109 spores) % of total activityÃ Speciﬁc activity (nmol mg protein1)

1h301C16h221C1h301C16h221C1h301C16h221C Downloaded from https://academic.oup.com/femsle/article/259/1/142/453567 by guest on 03 October 2021 w 100 g 10 min, crude debris 27 83 34 29 ND ND 3000 g 20 min, pellet 9 65 12 23 23 163 300 000 g 20 min, pellet 3 20 4 7 4 30 300 000 g 20 min, supernatant 39 115 50 41 32 96

ÃThe volumes of ﬁnal supernatant and pellets resuspended in Tris-sucrose solution were 1 mL. wProtein concentration was not determined in the crude debris.

Table 3. Distribution of Antonospora locustae pyruvate-converting activity in the fractions after centrifugation ActivityÃ (nmole per 109 spores) % of total activityw 100 g 10 min pellet (crude debris) 81 27 3000 g 10 min pellet 31 11 15 000 g 10 min pellet 7 2 300 000 g 30 min pellet 43 14 300 000 g 30 min supernatant 137 46

ÃIncubation for 16 h at room temperature. wThe volumes of ﬁnal supernatant and pellets resuspended in Tris-sucrose solution were 1 mL.

Table 4. Pyruvate-converting activity of Paranosema grylli spore proteins after gel filtration on Sephadex G-25 ActivityÃ Protein Specific activity Volume Total activity Yield of (nmol mL1) (mg mL1) (nmol mg protein1) (mL) (nmol) activity (%) Initial supernatant 201 1.88 107 1.2 240 100 Gel-filtration fraction 123 1.1 112 1.5 185 77

ÃIncubation for 16 h at room temperature. activity was observed in P. locustae spores destroyed in TS enzyme may be more or less loosely bound with spore and centrifuged at similar rates (Table 3). membranes. For example, some membrane components, Since both species demonstrated the presence of about such as quinones or cytochromes, could serve to accept 30% of the total activity in crude debris, the ‘insoluble’ electrons generated in the reaction of oxidative decarboxyla- characteristic of the pyruvate-converting enzyme could be tion of pyruvate. explained by its difficult extraction from destroyed spores. Desalting of P. grylli spore proteins by gel-filtration on a The structure of the microsporidian spore is known to be Sephadex G-25 column was performed to prevent the very complicated, and soluble protein may be included influence of any endogenous spore compounds on pyru- within some membrane compartment. If this compartment vate-converting activity. Spores destroyed in TS were cen- were sufficiently large to be ruptured during spore homo- trifuged at 13 000 g for 20 min and the supernatant was genization with 2.5 mm glass balls, soluble enzyme may be passed through the column equilibrated with the same released into the medium. Transportation of substrate into buffer. As shown in Table 4, specific pyruvate-converting such a compartment might be provided by a protein similar activities (per mg of protein) in the starting supernatant and to the putative pyruvate transporter, whose mRNA tran- after desalting were indistinguishable. This result suggests scripts were recently found in Antonospora locustae spores that the assay mixture components [0.2 mM pyruvate,

(Williams & Keeling, 2005). At the same time, the putative 25 mM potassium phosphate (pH 7.5), 5 mM MgCl2] were

c 2006 Federation of European Microbiological Societies FEMS Microbiol Lett 259 (2006) 142–146 Published by Blackwell Publishing Ltd. All rights reserved Pyruvate-converting activity in Paranosema (Antonospora) 145 sufﬁcient for metabolization of pyruvate. Moreover, the pyruvate-converting enzyme is functional without any com- desalted protein fraction exhibited 1.4-fold higher activity pounds added, but may be inactivated by dialysis and SA when the 5 mM MgCl2 was not included in the reaction precipitation, a tight association of the microsporidian’s medium (not shown). Thus, the pyruvate-converting activ- putative apoenzyme and ThDP-Me21 complex is more ity in microsporidian spores appears to be independent of probable. An independence of catalytic activity of any any additional compounds. At the same time, desalting of cofactors added was demonstrated in the course of spore proteins by dialysis followed by their precipitation L. plantarum pyruvate oxidase puriﬁcation. However, this with 80% SA resulted in practically complete inactivation of bacterial enzyme was reversibly inactivated by precipitation the pyruvate-converting enzyme. with SA (Sedewitz et al., 1984) owing to dissociation of The most plausible pyruvate-converting enzyme in mi- Me21-ThDP complex and apoenzyme. crosporidia should consist of two E1a and two E1b PDH As P. grylli and P. locustae spores metabolize pyruvate in a subunits, providing oxidative decarboxylation of substrate. very simple incubation medium, the question of the puta- Downloaded from https://academic.oup.com/femsle/article/259/1/142/453567 by guest on 03 October 2021

Separate expression of Bacillus stearothermophilus E1a and tive electron acceptor is the weak point in the hypothesis E1b PDH in Escherichia coli with subsequent mixing of both suggesting oxidative decarboxylation of pyruvate in micro- subunits resulted in the assembly of an active a2b2 hetero- sporidia. Besides, we did not observe any influence of tetrameric enzyme (Lessard & Perham, 1994). Most parts of 0.2 mM pyruvate on the reduction of potassium ferricyanide both subunits were found as soluble forms. The reaction (an artificial electron acceptor) in the presence of P. locustae catalysed by this enzyme may be described by two equations: spore homogenate (unpublished data). Thus, some other mechanisms of pyruvate metabolizing in microsporidia E1-ThDP þ pyruvate cannot be ruled out. In spite of the absence of any ð1Þ ! E1-a-hydroxyethyl-ThDP þ CO2 pyruvate-converting enzymes in the microsporidia E. cuniculi genome, except two PDH subunits, the functions of about half of the predicted proteins remain unknown E1-a-hydroxyethyl-ThDP þ DCPIP ð2Þ (Katinka et al., 2001). ! E -ThDP þ acetate þ DCPIPH 1 2 In order to understand pyruvate-metabolizing machinery where ThDP is thiamine diphosphate and DCPIP (2,6- of these parasites, it is necessary to perform further experi- dichlorophenoindophenol) is an artificial electron (proton) ments involving reversible inactivation and re-activation of acceptor. However, under native intracellular conditions, E1 the enzyme by its incubation with potential cofactors 21 is normally associated with E2 and E3 components of the (Me , ThDP, FAD), to investigate pyruvate-converting PDH complex, and lipoic acid of E2 is the acceptor of both activity in the presence of electron acceptors (DCPIP), to a-hydroxyethyl and electrons. Thus, the putative a2b2 detect whether pyruvate metabolizing entails acetate pro- enzyme of microsporidia should be functionally closer to duction and to study heterologous gene expression of PDH pyruvate:quinone oxidoreductase (EC 1.2.2.2) or pyruvate subunits, with subsequent functional analysis and immuno- oxidase (EC 1.2.3.3) catalysing the oxidative decarboxyla- localization. tion of pyruvate to acetate and CO2 in bacterial cells. Pyruvate:quinone oxidoreductase from E. coli and Coryne- bacterium glutamicum are homotetrameric flavoproteins Acknowledgements interacting with the inner surface of the membrane (Wil- liams & Hager, 1966; Schreiner & Eikmanns, 2005) and We acknowledge financial support from the Russian Foun- using quinones and DCPIP, correspondingly, as physiologi- dation for Basic Research (Grant no. 05-04-49616). cal and artificial electron acceptors. Pyruvate oxydase from lactic acid bacteria Lactobacillus delbrueckii and L. plantarum catalyse phosphate- and oxygen-dependent formation of References acetyl-phosphate, CO and hydrogen peroxide. This enzyme 2 Blake RII, O’Brien TA, Gennis RB & Hager LP (1982) Role of the is also a homotetrameric flavoprotein mainly found in the divalent metal cation in the pyruvate oxidase reaction. J Biol soluble fraction (Sedewitz et al., 1984). Chem 257: 9605–9611. All pyruvate decarboxylating enzymes require ThDP and Bradford M (1976) A rapid and sensitive method for the 21 a divalent metal cation (Me ) for catalytic activity. ThDP is quantitation of protein utilizing the principle of protein-dye 21 bound with apoenzyme but the true cofactor is the Me - binding. Anal Biochem 72: 248–254. ThDP complex because the only role of the metal is to bind Dolgikh VV, Sokolova JJ & Issi IV (1997) Activities of enzymes of to the cofactor (Blake et al., 1982). Such a complex may be carbohydrate and energy metabolism of the spores of the loosely (E. coli pyruvate oxidase) or tightly (L. plantarum microsporidian, Nosema grylli. J Eukaryot Microbiol 44: pyruvate oxidase) bound with apoenzyme. As P. grylli 246–249.

Fast NM & Keeling PJ (2001) Alpha and beta subunits of pyruvate from Lactobacillus plantarum. J Bacteriol 160: dehydrogenase E1 from the microsporidian Nosema locustae: 273–278. mitochondrion-derived carbon metabolism in microsporidia. Seleznev KV, Issi IV, Dolgikh VV, Belostotskaya GB, Antonova OA Mol Biochem Parasitol 117: 201–209. & Sokolova JJ (1995) Fractionation of different life cycle Katinka MD, Duprat S, Cornillot E, et al. (2001) Genome stages of microsporidia Nosema grylli from crickets sequence and gene compaction of the eukaryote parasite Gryllus bimaculatus by centrifugation in Percoll density Encephalitozoon cuniculi. Nature 414: 450–453. gradient for biochemical research. J Eukaryot Microbiol Lessard IAD & Perham RN (1994) Expression in Escherichia coli 42: 288–292. of genes encoding the E1a and E1b subunits of the pyruvate Williams FR & Hager LP (1966) Crystalline ﬂavin pyruvate dehydrogenase complex of Bacillus stearothermophilus and oxidase from Escherichia coli. I. Isolation and properties of the

assembly of a functional E1 component (a2b2) in vitro. J Biol ﬂavoprotein. Arch Biochem Biophys 116: 168–176.

Chem 269: 10378–10383. Williams BA & Keeling PJ (2005) Microsporidian mitochondrial Downloaded from https://academic.oup.com/femsle/article/259/1/142/453567 by guest on 03 October 2021 Schreiner ME & Eikmanns BJ (2005) Pyruvate:quinine proteins: expression in Antonospora locustae spores and oxidoreductase from Corynebacterium glutamicum: identification of genes coding for two further proteins. purification and biochemical characterization. J Bacteriol 187: J Eukaryot Microbiol 52: 271–276. 862–871. Williams BAP, Hirt RP, Lucocq JM & Embley TM (2002) Sedewitz B, Schleifer KH & Gotz¨ F (1984) Purification Amitochondrial remnant in the microsporidian and biochemical characterization of pyruvate oxidase Trachipleistophora hominis. Nature 418: 865–869.