- Email: [email protected]
isocitrate lyase, in Euglena gracilis
Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452 www.elsevier.com/locate/cbpb
Molecular characterization of a bifunctional glyoxylate cycle enzyme, malate synthase/isocitrate lyase, in Euglena gracilis Masami Nakazawaa,*, Tomomi Minamia, Koji Teramuraa, Shohei Kumamotoa, Sayaka Hanatoa, Shigeo Takenakab, Mitsuhiro Uedaa, Hiroshi Inuia, Yoshihisa Nakanoa, Kazutaka Miyatakea a
Department of Applied Biological Chemistry, Osaka Prefecture University, Sakai, Osaka 599-8531, Japan b Department of Veterinary Science, Osaka Prefecture University, Sakai, Osaka 599-8531, Japan Received 26 February 2005; received in revised form 9 May 2005; accepted 10 May 2005 Available online 17 June 2005
Abstract Euglena gracilis induced glyoxylate cycle enzymes when ethanol was fed as a sole carbon source. We purified, cloned and characterized a bifunctional glyoxylate cycle enzyme from E. gracilis (EgGCE). This enzyme consists of an N-terminal malate synthase (MS) domain fused to a C-terminal isocitrate lyase (ICL) domain in a single polypeptide chain. This domain order is inverted compared to the bifunctional glyoxylate cycle enzyme in Caenorhabditis elegans, an N-terminal ICL domain fused to a C-terminal MS domain. Purified EgGCE catalyzed the sequential ICL and MS reactions. ICL activity of purified EgGCE increased in the existence of acetyl-CoA at a concentration of micromolar order. We discussed the physiological roles of the bifunctional glyoxylate cycle enzyme in these organisms as well as its molecular evolution. D 2005 Elsevier Inc. All rights reserved. Keywords: Bifunctional enzyme; Glyoxylate cycle; Euglena gracilis; Inverted order; Malate synthase; Isocitrate lyase; Ace operon; Gene fusion; Acetyl-CoA; Mitochondrial localization
1. Introduction The glyoxylate cycle is a metabolic pathway which is very important for organisms to synthesize carbohydrates from C2 compounds, as first proposed by Kornberg and Krebs (1957). This cycle has two key specific enzymes; isocitrate lyase (ICL; EC 4.1.3.1) and malate synthase (MS; EC 2.3.3.9). ICL catalyzes the cleavage of d-isocitrate to glyoxylate and succinate, and glyoxylate formed by the ICL reaction is condensed with acetyl-CoA to produce l-malate by the action of MS. These two enzymes ensure the bypass of two of the decarboxylation steps of the tricarboxylic acid (TCA) cycle in the synthesis of succinate. Thus, the Abbreviations: MS, Malate synthase; ICL, Isocitrate lyase; EgGCE, Euglena gracilis bifunctional glyoxylate cycle enzyme; CeGCP, Caenorhabditis elegans bifunctional glyoxylate cycle protein. * Corresponding author. Tel.: +81 72 254 9468; fax: +81 72 254 9467. E-mail address: [email protected] (M. Nakazawa). 1096-4959/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpc.2005.05.006
glyoxylate cycle is very important especially under carbon limiting conditions. In certain higher plants, the glyoxylate cycle has been reported to play a pivotal role in the synthesis of carbohydrates from storage lipids during seedling (Eastmond and Graham, 2001). Isocitrate lyases have been found in a wide range of species including bacteria (Kornberg, 1966), archaea (Serrano et al., 1998), yeast (Taylor et al., 1996), fungi (Lorenz and Fink, 2001) and higher plants (Eastmond and Graham, 2001). ICLs commonly consist of four identical subunits in either prokaryotes or eukaryotes. However, the subunit molecular mass is significantly different between bacterial (approximately 47 kDa) and eukaryotic (60 – 64 kDa) ICLs. On the basis of the amino acid sequence features, MSs have been divided into two major families, isoforms A (MSA) and G (MSG). MSA with a molecular mass of about 65 kDa occurs in bacteria (Kornberg, 1966), yeast (Hartig et al., 1992), fungi (Lorenz and Fink, 2001) and higher plants (Eastmond and Graham, 2001), whereas MSG with around
446
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
80 kDa has been found only in bacteria. Certain bacteria, such as Escherichia coli, express both the two types of MSs (Molina et al., 1994). Liu et al. (1997, 1995) found that ICL and MS are encoded by a single gene and expressed as a single bifunctional polypeptide in nematode Caenorhabditis elegans. The polypeptide, called bifunctional glyoxylate cycle protein (GCP), consists of two separate domains for ICL and MS, and the ICL domain locates in the amino-terminal side of the polypeptide. The physiological functions of the GCP during development in nematode have been well studied; however, the catalytic properties of the unique enzyme have not been reported. It has been reported that Euglena gracilis, a unicellular protist containing chloroplasts, has both ICL and MS activities, and these enzyme activities are greatly enhanced when ethanol is fed as a sole carbon source (Inui et al., 1992). In the present paper, we have purified a bifunctional enzyme having the ICL and MS activities from E. gracilis grown on ethanol, and cloned cDNA encoding this bifunctional enzyme. We reported that the bifunctional glyoxylate enzyme found in E. gracilis (EgGCE), as well as the GCP in C. elegans, consists of two functional components, the ICL and MS domains, but that the ICL domain is found in the carboxy-terminal side in the EgGCE, in contrast to the nematode enzyme. In addition, it is also shown that the ICL reaction is activated in the presence of acetyl-CoA, a substrate of the MS reaction, in EgGCE; Vmax increases whereas Km for Mg2 + –isocitrate complex decreases in the presence of acetyl-CoA.
2. Materials and methods 2.1. Organism and culture Euglena gracilis SM-ZK (Oda et al., 1982), a nonphotosynthetic mutant derived from strain Z by treatment with streptomycin, was cultured in Cramer –Myers medium (Cramer and Myers, 1952), supplemented with ethanol at 85 mM as a sole carbon source, with aeration at 27 -C for 4 days. 2.2. Enzyme purification All operations during the enzyme purification were conducted at 4 -C. E. gracilis cells grown on ethanol (about 10 g wet basis) were harvested, washed and suspended in buffer A (10 mM potassium phosphate buffer, pH 7.0, containing 1 mM dithiothreitol, 1 mM MgCl2, 1 mM phenylmethylsulfonyl fluoride and 10 AM leupeptin). The cells were disrupted by sonication (10 kHz, 1 min 8) and centrifuged at 20,000 g for 30 min to obtain a crude enzyme solution. The solution was further centrifuged at 100,000 g for 1 h. Ammonium sulfate was added to the
supernatant to the 40% saturation and the solution was centrifuged at 18,000 g for 20 min. The concentrations of ammonium sulfate in the supernatant increased to 45% saturation after centrifugation, and the solution was centrifuged at 18,000 g for 20 min to collect the precipitate. The precipitate was dissolved in buffer A and dialyzed to remove ammonium sulfate. The enzyme solution was applied onto a DEAE – Sepharose FF column (1.5 11 cm), which had been equilibrated with the buffer A. The column was washed with 75 mL of the same buffer and the enzyme was eluted with 300 mL of a linear concentration gradient (10 to 400 mM) of potassium phosphate in the buffer. Active fractions were collected and applied onto a Phenyl – Sepharose column (1.3 4.5 cm), which had been equilibrated with buffer B (50 mM potassium phosphate buffer, pH 7.0, containing 1 mM dithiothreitol, 1 mM MgCl2, 1 mM phenylmethylsulfonyl fluoride and 10 AM leupeptin). The column was washed with 30 mL of buffer B containing 40% ethyleneglycol, and the enzyme was eluted by increasing the concentration of ethyleneglycol to 55% in the buffer. Active fractions were collected and dialyzed against the buffer A, and applied onto a hydroxyapatite column (1.7 5 cm) pre-equilibrated with buffer A. The column was washed with 50 mL of buffer A, and the enzyme was eluted with 200 mL of a linear concentration gradient (10 to 500 mM) of potassium phosphate in the buffer. Active fractions were combined and concentrated by membrane filtration with a Centricon 30 filter (Millipore, MA, USA). The enzyme solution (2 mL) was chromatographed on a Hi-Load Superdex column (Amersham Biosciences, NJ, USA) pre-equilibrated with buffer B containing 0.1 M KCl, using a fast protein liquid chromatography system (Amersham Biosciences). The active fractions obtained were combined and used as the purified preparation of EgGCE. The protein content was determined according to Bradford (1976) with bovine serum albumin as a standard. 2.3. Enzyme assays ICL reaction, the cleavage reaction of isocitrate to succinate and glyoxylate, was assayed by measuring the formation of glyoxylate – phenylhydrazone at 334 nm (Malhotra and Srivastava, 1982) at 30 -C. The reaction mixture (1 mL) contained 100 mM potassium phosphate buffer, pH 6.5, 30 mM MgCl2, 2 mM dithiothreitol, 4 mM phenylhydrazine –HCl, 30 mM sodium isocitrate, and the enzyme. MS activity was assayed by measuring the acetyl-CoA degradation (Cook, 1970) at 30 -C. The reaction mixture contained 100 mM Tris – HCl buffer, pH 8.0, 150 AM acetylCoA, 10 mM MgCl2, 1 mM glyoxylate, and the enzyme in a total volume of 1 mL. The activity was determined by following the decrease in absorbance at 232 nm owing to the degradation of acetyl-CoA.
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
The overall reaction of EgGCE, namely, the synthesis of malate from isocitrate and acetyl-CoA by the sequential actions of the ICL and MS domains, was conducted in a reaction mixture (1 mL), containing 100 mM potassium phosphate buffer, pH 6.5, 150 AM acetyl-CoA, 30 mM MgCl2, 2 mM dithiothreitol, sodium isocitrate, and the enzyme, at 30 -C. The concentrations of sodium isocitrate were changed in each reaction condition and are described in Fig. 3A.
447
Ten micrograms of protein was subjected to SDS-PAGE (7% gel) according to Laemmli (1970).
System Model 310 (Applied Biosystems). The E. gracilis cDNA library was constructed using EZAP II EcoR I/CIAP treated vector and Gigapack III gold packaging extracts (Stratagene, CA, USA). Sequential plaque screening of the E. gracilis cDNA library was carried out using the PCR fragments labeled with alkaline phosphatase with a commercial kit (Gene Imagesi AlkPhos Directi Labeling and Detection Kit; Amersham Biosciences) as the probe. Hybridization was done with the probe at 55 -C for 16 h, and the hybridized probe on the membrane was detected by a chemiluminescent method with CDP-Stari (Amersham Biosciences) as a substrate. The longest cDNA clones isolated from the positive plaque were sequenced and analyzed by BLAST and CLUSTAL W programs on the GenomeNet WWW server.
2.5. Peptide mapping
2.7. Kinetic analysis
The purified enzyme was partially cleaved by lysylendopeptidase or V8 protease on the PVDF membrane. Digested peptides were separated using high performance liquid chromatography and applied to the protein sequencer (Model 491-1; Applied Biosystems, CA, USA).
The Km values were determined using a doublereciprocal Lineweaver – Burk plot. Concentration of the Mg2 + –isocitrate complex was calculated from the following equilibrium; K 0 = [Mg 2 + -isocitrate] / [Mg 2 + ] [isocitrate] = 1350 M 1 (Giachetti et al., 1988).
2.6. Cloning
2.8. Determination of pH optimum on MS and ICL activities
Total RNA was extracted from E. gracilis SM-ZK according to the method of Shirzadegan et al. (1991), and then poly (A) mRNA was purified using OligotexidT30 (TAKARA BIO, Shiga, Japan). Poly (A) mRNA (3 Ag) was reverse-transcribed with SuperScripti II RNase H Reverse Transcriptase (Invitrogen, CA, USA) using oligo (dT)12 – 18 primer. Degenerate oligonucleotide primers were designed according to the amino acid sequences AMAPSGKNV and FHAMGGM which were archived by peptide mapping. Sense primer used was GCCATGGCCCCNTCCGGNAARAAYGT, and antisense primer used was CATICCICCCATIGCRTGRAA. The polymerase chain reaction (PCR) was run for 45 cycles (95 -C for 1 min, 56 -C for 1 min, 72 -C for 2 min) using Ready-to-Goi PCR beads (Amersham Biosciences). The PCR product (approximately 650 bp) was sequenced by the dideoxy chain termination method using a DNA Sequencing
To determine the pH optimum for MS and ICL activities, assays were performed using 0.1 M potassium phosphate buffer from pH 4.5 to 7.0, 0.1 M Tris – HCl buffer from pH 7.0 to 9.0.
2.4. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
3. Results and discussion 3.1. Purification of bifunctional glyoxylate cycle enzyme from E. gracilis Typical purification steps of the bifunctional glyoxylate cycle enzyme (EgGCE) are summarized in Table 1. The bifunctional enzyme was purified about 20-fold over the crude extract prepared from Euglena cells grown on ethanol. In all the steps during the purification, the MS activity co-fractionated with the ICL activity, and the ratios
Table 1 Purification steps of the bifunctional glyoxylate cycle enzyme from E. gracilis Purification step
Protein (mg)
Total activity
1490 827 215 43.9 20.1 10.0 3.14
408 361 168 162 80.9 40.5 17.7
MS
Specific activity MS
(Amol/min/mg protein)
(%)
51.9 46.6 26.0 20.0 10.1 5.78 2.17
0.27 0.44 0.78 3.69 4.03 4.03 5.19
100 88.5 41.4 39.7 19.8 9.90 4.30
(Amol/min) 1. 2. 3. 4. 5. 6. 7.
Crude extract 100,000 g supernatant (NH4)2SO4 fractionation DEAE-Sepharose Phenyl-Sepharose Hydroxyapatite Hi-Load Superdex
Yield
ICL
ICL
0.035 0.056 0.121 0.461 0.528 0.576 0.636
MS
ICL
100 89.8 50.1 38.5 19.5 11.1 4.20
448
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
of the MS activity versus the ICL activity remained constant. The purified preparation of EgGCE revealed a single protein band with a molecular mass of 110 kDa upon SDSpolyacrylamide gel electrophoresis. In contrast, the molecular mass of EgGCE was estimated to be 420 kDa by HiLoad Superdex gel filtration. It is thus suggested that EgGCE consists of four identical subunits. 3.2. Isolation and characterization of EgGCE cDNA EgGCE cDNA was first screened by RT-PCR using degenerate oligonucleotide primers, which were based on the amino acid sequence derived from peptide mapping of purified EgGCE. The PCR product (650 bp) obtained was sequenced and used as a hybridization probe to obtain the full-length of EgGCE cDNA from the Euglena cDNA library. The positive clone containing the longest insert was used in the analysis which follows. EgGCE cDNA was 4032 bp length (the nucleotide sequence data is in the DDBJ/ EMBL/GenBank nucleotide sequence databases with the accession number AB067657), and had a spliced leader sequence that was commonly found at the 5V-end of a vast majority of cytoplasmic mRNAs in Euglena (Tessier et al., 1991). Poly (A) sequence was also found at the 3V-end of the cDNA. These confirm that the full length of EgGCE cDNA was obtained. The EgGCE cDNA encoded a polypeptide composed of 1165 amino acid residues (Fig. 1). The calculated molecular weight of the polypeptide was 129,749; this value is slightly higher than the subunit molecular weight of EgGCE estimated by SDS-polyacrylamide gel electrophoresis (about 110,000). Results of database search suggested that the amino acid sequence of the N-terminal side of EgGCE (amino acid position: 1 –519) resembled the sequence of E. coli MS isoform A (EcMSA; 36% positional identity), cotton MS (GhMS; 32%), and the MS domain of CeGCP (35%). The N-terminal amino acid sequence of EgGCE showed a lower similarity to E. coli MS isoform G (EcMSG; 12%). However, the amino acid residues which had been proposed to be important for the catalytic function of MS by X-ray crystallography of EcMSG, were conserved in EgGCE (R187 and D475) as well as EcMSA and other malate synthases (Fig. 1A). It was thus revealed that EgGCE contains an N-terminal domain having the catalytic activity
of MS, in contrast to CeGCP in which MS domain is found at the C-terminal side. The C-terminal side of EgGCE (amino acid position: 597– 1165) showed structural similarity to prokaryotic and eukaryotic ICLs (20% and 13%, respectively, and positional identities to E. coli and cotton enzymes). In addition, multiple alignment showed not only active site residues (E743 and C778) but also residues that were important in stabilizing the dioxyanion of aci-carboxylate of succinate (S933 and S935), were conserved in EgGCE (Fig. 1B). In general, prokaryotic and eukaryotic ICLs show the difference in molecular mass (47 kDa and 60 – 64 kDa, respectively), and this difference is due to polypeptide insertion in eukaryotic ICLs (residues 274– 373 in the cotton enzyme). Although Euglena is a eukaryote, the common inserted sequence in eukaryotic ICLs was not found in EgGCE, as well as in CeGCP. However, in contrast to CeGCP, a new type of inserted sequence was found in EgGCE (amino acid position: 947– 1067). Database search of the inserted sequence of EgGCE showed no significant similarity to any other proteins. The location of this insertion in EgGCE corresponds to helix a17A which consists of TIM barrel structure near from ICL active site of A. nidulans ICL (Britton et al., 2000). ICL domain of EgGCE is thought to be a new family of ICL. The central region of EgGCE from residue A519 to S596 showed no similarity to either MS and ICL. In addition, the amino acid sequence of this region also showed no similarity to other proteins in the public database. Sequence data obtained in this paper show that EgGCE, a homotetrameric protein, has two separate domains, ICL (Cterminal side) and MS (N-terminal side), in each subunit. Homology search and multiple alignment show that EgGCE has all of the conserved residues that are considered important for the catalysis of both ICL and MS (Britton et al., 2000; Anstrom et al., 2003; Sharma et al., 2000; Howard et al., 2000), suggesting that each domain of EgGCE can function independently. We propose that EgGCE evolved by linking with ICL and MS by gene fusion. Previously, we have purified a protein with a subunit molecular mass of 116 kDa as malate synthase from ethanol-grown Euglena cells, and reported that this enzyme (which is identical to EgGCE in this report) is found in mitochondria (but not in glyoxysomes) when analyzed immunohistochemically (Ono et al., 2003). This unique
Fig. 1. Multiple sequence alignments of E. gracilis bifunctional glyoxylate cycle enzyme and other glyoxylate cycle enzymes. (A) Alignment of E. gracilis bifunctional glyoxylate cycle enzyme and various malate synthases. Asterisks indicate residues proposed to be important in catalysis, Arg-187 and Asp-475 (numbered according to E. gracilis GCE). Boxes indicate conserved residues among more than four sequences including E. gracilis GCE; black showed identical residues, gray showed similar residues. Abbreviations: EgGCE, E. gracilis bifunctional glyoxylate cycle enzyme (this study; BAC06572); CeGCP, C. elegans bifunctional glyoxylate cycle protein (AAA85857); EcMSA, E. coli malate synthase A (AceB; NP_418438); GhMS, Gossypium hirsutum (cotton) malate synthase (P17432); EcMSG, E. coli malate synthase G (GlcB; CAA52639). (B) Alignment of E. gracilis bifunctional glyoxylate cycle enzyme and isocitrate lyases. Asterisks indicate residues proposed to be important in catalysis, Glu-743, Cys-778, Ser-933, and Ser-935 (numbered according to E. gracilis GCE). Abbreviations: EgGCE, E. gracilis bifunctional glyoxylate cycle enzyme (this study; AB067657); CeGCP, C. elegans bifunctional glyoxylate cycle protein (AAA85857); EcICL, E. coli isocitrate lyase (AceA; AAC43109); GhICL, Gossypium hirsutum (cotton) isocitrate lyase (P17069); AnICL, Aspergillus nidulans FGSC A4 isocitrate lyase (EAA62727).
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
449
450
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
intracellular localization of EgGCE is supported by a computational prediction indicating that this enzyme has an N-terminal amino acid sequence targeting to mitochondria (PSORT programs; http://psort.nibb.ac.jp/). In contrast, Woodcock and Merrett (1978) reported to purify malate synthase with a subunit molecular mass of 175 kDa from Euglena cells grown on acetate. This malate synthase seems not to be identical to EgGCE and could be localized in glyoxysomes (but not in mitochondria), because it was reported that the malate synthase activity was detected in both the mitochondrial and microbody fractions (Collins and Merrett, 1975; Graves et al., 1972). C. elegans GCP is suggested to be localized in glyoxysome using immunological detection (Liu et al., 1995), comparing to mitochondrial localization of EgGCE. 3.3. Deduced ancient origin of bifunctional glyoxylate cycle enzyme in E. gracilis and C. elegans EgGCE consists of an amino terminal MS domain and carboxyl terminal ICL domain connected by a central linker region. This domain order was inverted to C. elegans bifunctional GCP (Fig. 2A). EgGCE showed low homology to other glyoxylate cycle enzymes; the MS domain showed 35 –40% identity to MSA from g-proteobacterium and the ICL domain showed 20 –25% identity to the ICL from proteobacterium and nematode. In contrast, C. elegans GCP showed very high amino acid homology to bacterial glyoxylate cycle enzymes; the MS domain showed about 60% identity to a- and g-proteobacterial MSA and the ICL domain showed about 70% identity to a-proteobacterial ICL. Genes encoding ICL and MS proteins reside in the same operon named ace in certain bacteria, including E. coli. We found two different gene orders of bacterial ace operon on
the bacterial genome database and COG database (Tatusov et al., 1997) (Fig. 2B). One operon structure is aceB (gene encoding MSA) –aceA (gene encoding ICL). This type of operon structure is often observed in the bacterial genome database; for example, E. coli (aceB locates 30-bp upstream of aceA) and Vibrio vulnificus YJ016 (68 bp). Another type of operon gene order is aceA (ICL)– aceB (MSA). We found only three bacterial species having this operon structure in the bacterial genome database; Streptomyces coelicolor A3(2) (aceA locates 101 bp upstream of aceB), Streptomyces avermitilis MA-4680 (145 bp) and Caulobacter crescentus (343 bp). Difference of domain structure and phylogeny between E. gracilis GCE and C. elegans GCP might arise through origins that have a different operon structure of ancient genes. Our analyses showed that gene fusion between MS and ICL had occurred at different times in the ancestor of E. gracilis and that of C. elegans. In E. gracilis and C. elegans, most of precursor RNAs are processed by trans-splicing at the 5Vends of genes by adding a specific trans-spliced leader (Tessier et al., 1991; Blumenthal, 1995). In addition, up to 15% of the genes are organized in polycistronic operons in C. elegans (Blumenthal et al., 2002). Polycistronic precursor RNAs are also processed by trans-splicing. Gene fusion events for making bifunctional enzymes may relate to trans-splicing and polycistronic transcription. Stechmann and Cavalier-Smith (2002) analyzed a derived gene fusion between dihydrofolate reductase (DHFR) and thymidylate synthase (TS) in order to locate the root of the eukaryote tree. They assumed the genes fused just once and were never secondarily split or laterally transferred within eukaryotes. Take our findings into account, gene fusion events possibly occurred more than once in some cases. 3.4. Catalytic properties of EgGCE In EgGCE, pH optima for the ICL activity and MS activity were observed to be 6.5, 7.5 –8.5, respectively. Kinetic profile in the ICL reaction for Mg2 + –isocitrate complex showed Michaelis – Menten kinetics, and the apparent Km value for the substrate was estimated to be 7.6 mM. In the MS reaction, Michaelis– Menten kinetics was also observed for both acetyl-CoA and glyoxylate, and the apparent Km values for these substrates were 25 and 40 AM, respectively. We determined the EgGCE overall reaction, sequential ICL and MS reactions. We measured the isocitrate-dependent acetyl-CoA degradation. The reaction scheme is as follows:
Fig. 2. Deduced models of the structure and the origin of E. gracilis bifunctional glyoxylate cycle enzyme. (A) Comparison of the structures between two bifunctional glyoxylate cycle enzymes, EgGCE and CeGCP. (B) Genomic organization of bacterial malate synthases and isocitrate lyases in ace operon. Black arrows show aceB gene that encodes malate synthase A. White arrows show aceA gene that encodes isocitrate lyase. The figure is not drawn to scale.
Isocitrate Y Glyoxylate þ SuccinateðReaction1; ICLÞ Glyoxylate þ Acetyl CoA þ H2 O Y Malate þ CoAðReaction2; MSÞ:
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
Fig. 3. EgGCE overall reaction and ICL activities. (A) Comparison between EgGCE overall reaction activity and ICL activity at the different concentrations of isocitrate. White bars show ICL activity. The reaction mixture for measuring ICL activity contained 100 mM potassium phosphate buffer, pH 6.5, 30 mM MgCl2, 2 mM dithiothreitol, 4 mM phenylhydrazine – HCl, indicated concentration of sodium isocitrate, and the enzyme in the total volume of 1 mL. Black bars show EgGCE overall reaction activity. The reaction mixture for measuring the overall reaction activity contained 100 mM potassium phosphate buffer, pH 6.5, 30 mM MgCl2, 2 mM dithiothreitol, indicated concentration of sodium isocitrate, 150 AM acetyl-CoA and the enzyme in the total volume of 1 mL. Each experiment was triplicated. Error bars show TSD. (B) Effect of acetyl-CoA on the ICL activity of EgGCE. ICL activities were measured containing various concentrations of acetyl-CoA in the reaction mixture. Cleavage reaction of isocitrate to succinate and glyoxylate was assayed by measuring the formation of glyoxylate – phenylhydrazone at 334 nm. Each experiment was triplicated. Error bars show TSD. (C) Lineweaver – Burk plots for the effects of acetyl-CoA on the ICL activity. Isocitrate concentrations were changed in the ICL reaction mixtures. Concentrations of acetyl-CoA were 150 AM. The definition of each symbol is the following; open circle, ICL activity without acetyl-CoA; closed square, ICL activity with 150 AM acetyl-CoA. Each experiment was triplicated, and a typical result is shown.
451
Overall reaction activity should be lower than the ICL activity if there are no special effects in these reactions. The overall reaction mixture contained 100 mM potassium phosphate buffer, pH 6.5, 30 mM MgCl2, 2 mM dithiothreitol, 150 AM acetyl-CoA, sodium isocitrate, and the enzyme in the total volume of 1 mL. The concentrations of sodium isocitrate in the reaction mixture were adjusted to 1, 7 and 30 mM. These isocitrate concentrations are correspond to Mg2 + – isocitrate complex concentrations of 0.975, 6.78 and 25.6 mM, respectively. Overall reaction activity was lower than the ICL activity in the isocitrate concentration of 30 mM. However, the specific activities of the overall reaction were higher than that of the ICL reactions in the isocitrate conditions of 1 and 7 mM. In order to determine the reason why the overall reaction activity of EgGCE was higher than its ICL activity, we added acetyl-CoA, which was the substrate for the MS reaction, into the ICL reaction mixture (Fig. 3B). The concentration of sodium isocitrate was adjusted to 1 mM. ICL activity measured in the presence of 150 AM acetyl-CoA was increased more than four times compared to the activity measured in the absence of acetyl-CoA. In addition, the ICL activity with acetylCoA showed almost the same activity as the overall reaction activity in these experimental conditions (Fig. 3A and 3B). Acetyl-CoA also affects the isocitrate lyase activity at the concentration of 5 AM or above. It is considered that acetyl-CoA also increases the ICL activity in vivo. Acetyl-CoA did not act as an allosteric effector of the ICL activity of EgGCE (Fig. 3C). The apparent Km value for the Mg2 + – isocitrate of ICL with 150 AM acetyl-CoA was 2500 AM. There has been no research concerning whether CoA derivatives increase ICL activity. In fact, acetyl-CoA did not affect the activity of ICL from Bacillus stearothermiphilus (Fluka) at the concentration of 150 AM (data not shown). EgGCE shows the specific catalytic property of ICL, namely, its activity is increased by acetyl-CoA. Not only the deduced amino acid sequence but also the catalytic features are different from other ICLs. We need further study to determine the mechanisms of how acetyl-CoA regulates ICL.
4. Conclusions In this paper we reported that the bifunctional glyoxylate cycle enzyme in E. gracilis had special catalytic features, i.e., its ICL activity was increased by acetyl-CoA. This property was beneficial to sequential ICL and MS reactions under physiological conditions. We also report that EgGCE had an inverted domain structure compared to C. elegans bifunctional GCP. This is an important finding to understand the ancient formation of bifunctional enzymes.
452
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
References Anstrom, D.M., Kallio, K., Remington, S.J., 2003. Structure of the Escherichia coli malate synthase G: pyruvate:acetyl-coenzyme A ˚ resolution. Protein Sci. 12, abortive ternary complex at 1.95 A 1822 – 1832. Blumenthal, T., 1995. Trans-splicing and polycistronic transcription in Caenorhabditis elegans. Trends Genet. 11, 132 – 136. Blumenthal, T., Evans, D., Link, C.D., Guffanti, A., Lawson, D., Thierry Mieg, J., Thierry Mieg, D., Chiu, W.L., Duke, K., Kiraly, M., Kim, S.K., 2002. A global analysis of Caenorhabditis elegans operons. Nature 417, 851 – 854. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248 – 254. Britton, K., Langridge, S., Baker, P.J., Weeradechapon, K., Sedelnikova, S.E., De Lucas, J.R., Rice, D.W., Turner, G., 2000. The crystal structure and active site location of isocitrate lyase from the fungus Aspergillus nidulans. Struct. Fold Des. 8, 349 – 362. Collins, N., Merrett, M.J., 1975. Microbody-marker enzymes during transition from phototrophic to organotrophic growth in Euglena. Plant Physiol. 55, 1018 – 1022. Cook, J.R., 1970. Properties of partially purified malate synthase for Euglena gracilis. J. Protozool. 17, 232 – 235. Cramer, M., Myers, J., 1952. Growth and photosynthetic characteristics of Euglena gracilis. Arch. Mikrobiol. 17, 384 – 402. Eastmond, P.J., Graham, I.A., 2001. Re-examining the role of the glyoxylate cycle in oilseeds. Trends Plant Sci. 6, 72 – 78. Giachetti, E., Pinzauti, G., Bonaccorsi, R., Vanni, P., 1988. Isocitrate lyase from Pinus pinea. Characterization of its true substrate and the action of magnesium ions. Eur. J. Biochem. 172, 85 – 91. Graves, L.B.J., Trelease, R.N., Grill, A., Becker, W.M., 1972. Localization of glyoxylate cycle enzymes in glyoxysomes in Euglena. J. Protozool. 19, 527 – 532. Hartig, A., Simon, M.M., Schuster, T., Daugherty, J.R., Yoo, H.S., Cooper, T.G., 1992. Differentially regulated malate synthase genes participate in carbon and nitrogen metabolism of S. cerevisiae. Nucleic Acids Res. 20, 5677 – 5686. Howard, B.R., Endrizzi, J.A., Remington, S.J., 2000. Crystal structure of Escherichia coli malate synthase G complexed with magnesium and ˚ resolution: mechanistic implications. Biochemistry glyoxylate at 2.0 A 39, 3156 – 3168. Inui, H., Miyatake, K., Nakano, Y., Kitaoka, S., 1992. Synthesis of reserved polysaccharide from wax esters accumulated as the result of anaerobic energy generation in Euglena gracilis returned from anaerobic to aerobic conditions. Int. J. Biochem. 24, 799 – 803. Kornberg, H.L., Krebs, H.A., 1957. Synthesis of cell constituents from C2units by a modified tricarboxylic acid cycle. Nature 179, 988 – 991. Kornberg, H.L., 1966. The role and control of the glyoxylate cycle in Escherichia coli. Biochem. J. 99, 1 – 11.
Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680 – 685. Liu, F., Thatcher, J.D., Barral, J.M., Epstein, H.F., 1995. Bifunctional glyoxylate cycle protein of Caenorhabditis elegans: a developmentally regulated protein of intestine and muscle. Dev. Biol. 169, 399 – 414. Liu, F., Thatcher, J.D., Epstein, H.F., 1997. Induction of glyoxylate cycle expression in Caenorhabditis elegans: a fasting response throughout larval development. Biochemistry 36, 255 – 260. Lorenz, M.C., Fink, G.R., 2001. The glyoxylate cycle is required for fungal virulence. Nature 412, 83 – 86. Malhotra, O.P., Srivastava, P.K., 1982. Isolation and characterization of isocitrate lyase of castor endosperm. Arch. Biochem. Biophys. 214, 164 – 171. Molina, I., Pellicer, M.T., Badia, J., Aguilar, J., Baldoma, L., 1994. Molecular characterization of Escherichia coli malate synthase G. Differentiation with the malate synthase A isoenzyme. Eur. J. Biochem. 224, 541 – 548. Oda, Y., Nakano, Y., Kitaoka, S., 1982. Utilization and toxicity of exogenous amino acid in Euglena gracilis. J. Gen. Microbiol. 128, 853 – 858. Ono, K., Kondo, M., Osafune, T., Miyatake, K., Inui, H., Kitaoka, S., Nishimura, M., Nakano, Y., 2003. Presence of glyoxylate cycle enzymes in the mitochondria of Euglena gracilis. J. Eukaryot. Microbiol. 50, 92 – 96. Serrano, J.A., Camacho, M., Bonete, M.J., 1998. Operation of glyoxylate cycle in halophilic archaea: presence of malate synthase and isocitrate lyase in Haloferax volcanii. FEBS Lett. 434, 13 – 16. Sharma, V., Sharma, S., Hoener zu Bentrup, K., McKinney, J.D., Russell, D.G., Jacobs, W.R.J., Sacchettini, J.C., 2000. Structure of isocitrate lyase, a persistence factor of Mycobacterium tuberculosis. Nat. Struct. Biol. 7, 663 – 668. Shirzadegan, M., Christie, P., Seemann, J.R., 1991. An efficient method for isolation of RNA from tissue cultured plant cells. Nucleic Acids Res. 19, 6055. Stechmann, A., Cavalier-Smith, T., 2002. Rooting the eukaryote tree by using a derived gene fusion. Science 297, 89 – 91. Tatusov, R.L., Koonin, E.V., Lipman, D.J., 1997. A genomic perspective on protein families. Science 278, 631 – 637. Taylor, K.M., Kaplan, C.P., Gao, X., Baker, A., 1996. Localization and targeting of isocitrate lyases in Saccharomyces cerevisiae. Biochem. J. 319, 255 – 262. Tessier, L.H., Keller, M., Chan, R.L., Fournier, R., Weil, J.H., Imbault, P., 1991. Short leader sequences may be transferred from small RNAs to pre-mature mRNAs by trans-splicing in Euglena. EMBO J. 10, 2621 – 2625. Woodcock, E., Merrett, M.J., 1978. Purification and immunochemical characterization of malate synthase from Euglena gracilis. Biochem. J. 173, 95 – 101.
Molecular characterization of a bifunctional glyoxylate cycle enzyme, malate synthase/isocitrate lyase, in Euglena gracilis Masami Nakazawaa,*, Tomomi Minamia, Koji Teramuraa, Shohei Kumamotoa, Sayaka Hanatoa, Shigeo Takenakab, Mitsuhiro Uedaa, Hiroshi Inuia, Yoshihisa Nakanoa, Kazutaka Miyatakea a
Department of Applied Biological Chemistry, Osaka Prefecture University, Sakai, Osaka 599-8531, Japan b Department of Veterinary Science, Osaka Prefecture University, Sakai, Osaka 599-8531, Japan Received 26 February 2005; received in revised form 9 May 2005; accepted 10 May 2005 Available online 17 June 2005
Abstract Euglena gracilis induced glyoxylate cycle enzymes when ethanol was fed as a sole carbon source. We purified, cloned and characterized a bifunctional glyoxylate cycle enzyme from E. gracilis (EgGCE). This enzyme consists of an N-terminal malate synthase (MS) domain fused to a C-terminal isocitrate lyase (ICL) domain in a single polypeptide chain. This domain order is inverted compared to the bifunctional glyoxylate cycle enzyme in Caenorhabditis elegans, an N-terminal ICL domain fused to a C-terminal MS domain. Purified EgGCE catalyzed the sequential ICL and MS reactions. ICL activity of purified EgGCE increased in the existence of acetyl-CoA at a concentration of micromolar order. We discussed the physiological roles of the bifunctional glyoxylate cycle enzyme in these organisms as well as its molecular evolution. D 2005 Elsevier Inc. All rights reserved. Keywords: Bifunctional enzyme; Glyoxylate cycle; Euglena gracilis; Inverted order; Malate synthase; Isocitrate lyase; Ace operon; Gene fusion; Acetyl-CoA; Mitochondrial localization
1. Introduction The glyoxylate cycle is a metabolic pathway which is very important for organisms to synthesize carbohydrates from C2 compounds, as first proposed by Kornberg and Krebs (1957). This cycle has two key specific enzymes; isocitrate lyase (ICL; EC 4.1.3.1) and malate synthase (MS; EC 2.3.3.9). ICL catalyzes the cleavage of d-isocitrate to glyoxylate and succinate, and glyoxylate formed by the ICL reaction is condensed with acetyl-CoA to produce l-malate by the action of MS. These two enzymes ensure the bypass of two of the decarboxylation steps of the tricarboxylic acid (TCA) cycle in the synthesis of succinate. Thus, the Abbreviations: MS, Malate synthase; ICL, Isocitrate lyase; EgGCE, Euglena gracilis bifunctional glyoxylate cycle enzyme; CeGCP, Caenorhabditis elegans bifunctional glyoxylate cycle protein. * Corresponding author. Tel.: +81 72 254 9468; fax: +81 72 254 9467. E-mail address: [email protected] (M. Nakazawa). 1096-4959/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpc.2005.05.006
glyoxylate cycle is very important especially under carbon limiting conditions. In certain higher plants, the glyoxylate cycle has been reported to play a pivotal role in the synthesis of carbohydrates from storage lipids during seedling (Eastmond and Graham, 2001). Isocitrate lyases have been found in a wide range of species including bacteria (Kornberg, 1966), archaea (Serrano et al., 1998), yeast (Taylor et al., 1996), fungi (Lorenz and Fink, 2001) and higher plants (Eastmond and Graham, 2001). ICLs commonly consist of four identical subunits in either prokaryotes or eukaryotes. However, the subunit molecular mass is significantly different between bacterial (approximately 47 kDa) and eukaryotic (60 – 64 kDa) ICLs. On the basis of the amino acid sequence features, MSs have been divided into two major families, isoforms A (MSA) and G (MSG). MSA with a molecular mass of about 65 kDa occurs in bacteria (Kornberg, 1966), yeast (Hartig et al., 1992), fungi (Lorenz and Fink, 2001) and higher plants (Eastmond and Graham, 2001), whereas MSG with around
446
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
80 kDa has been found only in bacteria. Certain bacteria, such as Escherichia coli, express both the two types of MSs (Molina et al., 1994). Liu et al. (1997, 1995) found that ICL and MS are encoded by a single gene and expressed as a single bifunctional polypeptide in nematode Caenorhabditis elegans. The polypeptide, called bifunctional glyoxylate cycle protein (GCP), consists of two separate domains for ICL and MS, and the ICL domain locates in the amino-terminal side of the polypeptide. The physiological functions of the GCP during development in nematode have been well studied; however, the catalytic properties of the unique enzyme have not been reported. It has been reported that Euglena gracilis, a unicellular protist containing chloroplasts, has both ICL and MS activities, and these enzyme activities are greatly enhanced when ethanol is fed as a sole carbon source (Inui et al., 1992). In the present paper, we have purified a bifunctional enzyme having the ICL and MS activities from E. gracilis grown on ethanol, and cloned cDNA encoding this bifunctional enzyme. We reported that the bifunctional glyoxylate enzyme found in E. gracilis (EgGCE), as well as the GCP in C. elegans, consists of two functional components, the ICL and MS domains, but that the ICL domain is found in the carboxy-terminal side in the EgGCE, in contrast to the nematode enzyme. In addition, it is also shown that the ICL reaction is activated in the presence of acetyl-CoA, a substrate of the MS reaction, in EgGCE; Vmax increases whereas Km for Mg2 + –isocitrate complex decreases in the presence of acetyl-CoA.
2. Materials and methods 2.1. Organism and culture Euglena gracilis SM-ZK (Oda et al., 1982), a nonphotosynthetic mutant derived from strain Z by treatment with streptomycin, was cultured in Cramer –Myers medium (Cramer and Myers, 1952), supplemented with ethanol at 85 mM as a sole carbon source, with aeration at 27 -C for 4 days. 2.2. Enzyme purification All operations during the enzyme purification were conducted at 4 -C. E. gracilis cells grown on ethanol (about 10 g wet basis) were harvested, washed and suspended in buffer A (10 mM potassium phosphate buffer, pH 7.0, containing 1 mM dithiothreitol, 1 mM MgCl2, 1 mM phenylmethylsulfonyl fluoride and 10 AM leupeptin). The cells were disrupted by sonication (10 kHz, 1 min 8) and centrifuged at 20,000 g for 30 min to obtain a crude enzyme solution. The solution was further centrifuged at 100,000 g for 1 h. Ammonium sulfate was added to the
supernatant to the 40% saturation and the solution was centrifuged at 18,000 g for 20 min. The concentrations of ammonium sulfate in the supernatant increased to 45% saturation after centrifugation, and the solution was centrifuged at 18,000 g for 20 min to collect the precipitate. The precipitate was dissolved in buffer A and dialyzed to remove ammonium sulfate. The enzyme solution was applied onto a DEAE – Sepharose FF column (1.5 11 cm), which had been equilibrated with the buffer A. The column was washed with 75 mL of the same buffer and the enzyme was eluted with 300 mL of a linear concentration gradient (10 to 400 mM) of potassium phosphate in the buffer. Active fractions were collected and applied onto a Phenyl – Sepharose column (1.3 4.5 cm), which had been equilibrated with buffer B (50 mM potassium phosphate buffer, pH 7.0, containing 1 mM dithiothreitol, 1 mM MgCl2, 1 mM phenylmethylsulfonyl fluoride and 10 AM leupeptin). The column was washed with 30 mL of buffer B containing 40% ethyleneglycol, and the enzyme was eluted by increasing the concentration of ethyleneglycol to 55% in the buffer. Active fractions were collected and dialyzed against the buffer A, and applied onto a hydroxyapatite column (1.7 5 cm) pre-equilibrated with buffer A. The column was washed with 50 mL of buffer A, and the enzyme was eluted with 200 mL of a linear concentration gradient (10 to 500 mM) of potassium phosphate in the buffer. Active fractions were combined and concentrated by membrane filtration with a Centricon 30 filter (Millipore, MA, USA). The enzyme solution (2 mL) was chromatographed on a Hi-Load Superdex column (Amersham Biosciences, NJ, USA) pre-equilibrated with buffer B containing 0.1 M KCl, using a fast protein liquid chromatography system (Amersham Biosciences). The active fractions obtained were combined and used as the purified preparation of EgGCE. The protein content was determined according to Bradford (1976) with bovine serum albumin as a standard. 2.3. Enzyme assays ICL reaction, the cleavage reaction of isocitrate to succinate and glyoxylate, was assayed by measuring the formation of glyoxylate – phenylhydrazone at 334 nm (Malhotra and Srivastava, 1982) at 30 -C. The reaction mixture (1 mL) contained 100 mM potassium phosphate buffer, pH 6.5, 30 mM MgCl2, 2 mM dithiothreitol, 4 mM phenylhydrazine –HCl, 30 mM sodium isocitrate, and the enzyme. MS activity was assayed by measuring the acetyl-CoA degradation (Cook, 1970) at 30 -C. The reaction mixture contained 100 mM Tris – HCl buffer, pH 8.0, 150 AM acetylCoA, 10 mM MgCl2, 1 mM glyoxylate, and the enzyme in a total volume of 1 mL. The activity was determined by following the decrease in absorbance at 232 nm owing to the degradation of acetyl-CoA.
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
The overall reaction of EgGCE, namely, the synthesis of malate from isocitrate and acetyl-CoA by the sequential actions of the ICL and MS domains, was conducted in a reaction mixture (1 mL), containing 100 mM potassium phosphate buffer, pH 6.5, 150 AM acetyl-CoA, 30 mM MgCl2, 2 mM dithiothreitol, sodium isocitrate, and the enzyme, at 30 -C. The concentrations of sodium isocitrate were changed in each reaction condition and are described in Fig. 3A.
447
Ten micrograms of protein was subjected to SDS-PAGE (7% gel) according to Laemmli (1970).
System Model 310 (Applied Biosystems). The E. gracilis cDNA library was constructed using EZAP II EcoR I/CIAP treated vector and Gigapack III gold packaging extracts (Stratagene, CA, USA). Sequential plaque screening of the E. gracilis cDNA library was carried out using the PCR fragments labeled with alkaline phosphatase with a commercial kit (Gene Imagesi AlkPhos Directi Labeling and Detection Kit; Amersham Biosciences) as the probe. Hybridization was done with the probe at 55 -C for 16 h, and the hybridized probe on the membrane was detected by a chemiluminescent method with CDP-Stari (Amersham Biosciences) as a substrate. The longest cDNA clones isolated from the positive plaque were sequenced and analyzed by BLAST and CLUSTAL W programs on the GenomeNet WWW server.
2.5. Peptide mapping
2.7. Kinetic analysis
The purified enzyme was partially cleaved by lysylendopeptidase or V8 protease on the PVDF membrane. Digested peptides were separated using high performance liquid chromatography and applied to the protein sequencer (Model 491-1; Applied Biosystems, CA, USA).
The Km values were determined using a doublereciprocal Lineweaver – Burk plot. Concentration of the Mg2 + –isocitrate complex was calculated from the following equilibrium; K 0 = [Mg 2 + -isocitrate] / [Mg 2 + ] [isocitrate] = 1350 M 1 (Giachetti et al., 1988).
2.6. Cloning
2.8. Determination of pH optimum on MS and ICL activities
Total RNA was extracted from E. gracilis SM-ZK according to the method of Shirzadegan et al. (1991), and then poly (A) mRNA was purified using OligotexidT30
To determine the pH optimum for MS and ICL activities, assays were performed using 0.1 M potassium phosphate buffer from pH 4.5 to 7.0, 0.1 M Tris – HCl buffer from pH 7.0 to 9.0.
2.4. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
3. Results and discussion 3.1. Purification of bifunctional glyoxylate cycle enzyme from E. gracilis Typical purification steps of the bifunctional glyoxylate cycle enzyme (EgGCE) are summarized in Table 1. The bifunctional enzyme was purified about 20-fold over the crude extract prepared from Euglena cells grown on ethanol. In all the steps during the purification, the MS activity co-fractionated with the ICL activity, and the ratios
Table 1 Purification steps of the bifunctional glyoxylate cycle enzyme from E. gracilis Purification step
Protein (mg)
Total activity
1490 827 215 43.9 20.1 10.0 3.14
408 361 168 162 80.9 40.5 17.7
MS
Specific activity MS
(Amol/min/mg protein)
(%)
51.9 46.6 26.0 20.0 10.1 5.78 2.17
0.27 0.44 0.78 3.69 4.03 4.03 5.19
100 88.5 41.4 39.7 19.8 9.90 4.30
(Amol/min) 1. 2. 3. 4. 5. 6. 7.
Crude extract 100,000 g supernatant (NH4)2SO4 fractionation DEAE-Sepharose Phenyl-Sepharose Hydroxyapatite Hi-Load Superdex
Yield
ICL
ICL
0.035 0.056 0.121 0.461 0.528 0.576 0.636
MS
ICL
100 89.8 50.1 38.5 19.5 11.1 4.20
448
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
of the MS activity versus the ICL activity remained constant. The purified preparation of EgGCE revealed a single protein band with a molecular mass of 110 kDa upon SDSpolyacrylamide gel electrophoresis. In contrast, the molecular mass of EgGCE was estimated to be 420 kDa by HiLoad Superdex gel filtration. It is thus suggested that EgGCE consists of four identical subunits. 3.2. Isolation and characterization of EgGCE cDNA EgGCE cDNA was first screened by RT-PCR using degenerate oligonucleotide primers, which were based on the amino acid sequence derived from peptide mapping of purified EgGCE. The PCR product (650 bp) obtained was sequenced and used as a hybridization probe to obtain the full-length of EgGCE cDNA from the Euglena cDNA library. The positive clone containing the longest insert was used in the analysis which follows. EgGCE cDNA was 4032 bp length (the nucleotide sequence data is in the DDBJ/ EMBL/GenBank nucleotide sequence databases with the accession number AB067657), and had a spliced leader sequence that was commonly found at the 5V-end of a vast majority of cytoplasmic mRNAs in Euglena (Tessier et al., 1991). Poly (A) sequence was also found at the 3V-end of the cDNA. These confirm that the full length of EgGCE cDNA was obtained. The EgGCE cDNA encoded a polypeptide composed of 1165 amino acid residues (Fig. 1). The calculated molecular weight of the polypeptide was 129,749; this value is slightly higher than the subunit molecular weight of EgGCE estimated by SDS-polyacrylamide gel electrophoresis (about 110,000). Results of database search suggested that the amino acid sequence of the N-terminal side of EgGCE (amino acid position: 1 –519) resembled the sequence of E. coli MS isoform A (EcMSA; 36% positional identity), cotton MS (GhMS; 32%), and the MS domain of CeGCP (35%). The N-terminal amino acid sequence of EgGCE showed a lower similarity to E. coli MS isoform G (EcMSG; 12%). However, the amino acid residues which had been proposed to be important for the catalytic function of MS by X-ray crystallography of EcMSG, were conserved in EgGCE (R187 and D475) as well as EcMSA and other malate synthases (Fig. 1A). It was thus revealed that EgGCE contains an N-terminal domain having the catalytic activity
of MS, in contrast to CeGCP in which MS domain is found at the C-terminal side. The C-terminal side of EgGCE (amino acid position: 597– 1165) showed structural similarity to prokaryotic and eukaryotic ICLs (20% and 13%, respectively, and positional identities to E. coli and cotton enzymes). In addition, multiple alignment showed not only active site residues (E743 and C778) but also residues that were important in stabilizing the dioxyanion of aci-carboxylate of succinate (S933 and S935), were conserved in EgGCE (Fig. 1B). In general, prokaryotic and eukaryotic ICLs show the difference in molecular mass (47 kDa and 60 – 64 kDa, respectively), and this difference is due to polypeptide insertion in eukaryotic ICLs (residues 274– 373 in the cotton enzyme). Although Euglena is a eukaryote, the common inserted sequence in eukaryotic ICLs was not found in EgGCE, as well as in CeGCP. However, in contrast to CeGCP, a new type of inserted sequence was found in EgGCE (amino acid position: 947– 1067). Database search of the inserted sequence of EgGCE showed no significant similarity to any other proteins. The location of this insertion in EgGCE corresponds to helix a17A which consists of TIM barrel structure near from ICL active site of A. nidulans ICL (Britton et al., 2000). ICL domain of EgGCE is thought to be a new family of ICL. The central region of EgGCE from residue A519 to S596 showed no similarity to either MS and ICL. In addition, the amino acid sequence of this region also showed no similarity to other proteins in the public database. Sequence data obtained in this paper show that EgGCE, a homotetrameric protein, has two separate domains, ICL (Cterminal side) and MS (N-terminal side), in each subunit. Homology search and multiple alignment show that EgGCE has all of the conserved residues that are considered important for the catalysis of both ICL and MS (Britton et al., 2000; Anstrom et al., 2003; Sharma et al., 2000; Howard et al., 2000), suggesting that each domain of EgGCE can function independently. We propose that EgGCE evolved by linking with ICL and MS by gene fusion. Previously, we have purified a protein with a subunit molecular mass of 116 kDa as malate synthase from ethanol-grown Euglena cells, and reported that this enzyme (which is identical to EgGCE in this report) is found in mitochondria (but not in glyoxysomes) when analyzed immunohistochemically (Ono et al., 2003). This unique
Fig. 1. Multiple sequence alignments of E. gracilis bifunctional glyoxylate cycle enzyme and other glyoxylate cycle enzymes. (A) Alignment of E. gracilis bifunctional glyoxylate cycle enzyme and various malate synthases. Asterisks indicate residues proposed to be important in catalysis, Arg-187 and Asp-475 (numbered according to E. gracilis GCE). Boxes indicate conserved residues among more than four sequences including E. gracilis GCE; black showed identical residues, gray showed similar residues. Abbreviations: EgGCE, E. gracilis bifunctional glyoxylate cycle enzyme (this study; BAC06572); CeGCP, C. elegans bifunctional glyoxylate cycle protein (AAA85857); EcMSA, E. coli malate synthase A (AceB; NP_418438); GhMS, Gossypium hirsutum (cotton) malate synthase (P17432); EcMSG, E. coli malate synthase G (GlcB; CAA52639). (B) Alignment of E. gracilis bifunctional glyoxylate cycle enzyme and isocitrate lyases. Asterisks indicate residues proposed to be important in catalysis, Glu-743, Cys-778, Ser-933, and Ser-935 (numbered according to E. gracilis GCE). Abbreviations: EgGCE, E. gracilis bifunctional glyoxylate cycle enzyme (this study; AB067657); CeGCP, C. elegans bifunctional glyoxylate cycle protein (AAA85857); EcICL, E. coli isocitrate lyase (AceA; AAC43109); GhICL, Gossypium hirsutum (cotton) isocitrate lyase (P17069); AnICL, Aspergillus nidulans FGSC A4 isocitrate lyase (EAA62727).
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
449
450
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
intracellular localization of EgGCE is supported by a computational prediction indicating that this enzyme has an N-terminal amino acid sequence targeting to mitochondria (PSORT programs; http://psort.nibb.ac.jp/). In contrast, Woodcock and Merrett (1978) reported to purify malate synthase with a subunit molecular mass of 175 kDa from Euglena cells grown on acetate. This malate synthase seems not to be identical to EgGCE and could be localized in glyoxysomes (but not in mitochondria), because it was reported that the malate synthase activity was detected in both the mitochondrial and microbody fractions (Collins and Merrett, 1975; Graves et al., 1972). C. elegans GCP is suggested to be localized in glyoxysome using immunological detection (Liu et al., 1995), comparing to mitochondrial localization of EgGCE. 3.3. Deduced ancient origin of bifunctional glyoxylate cycle enzyme in E. gracilis and C. elegans EgGCE consists of an amino terminal MS domain and carboxyl terminal ICL domain connected by a central linker region. This domain order was inverted to C. elegans bifunctional GCP (Fig. 2A). EgGCE showed low homology to other glyoxylate cycle enzymes; the MS domain showed 35 –40% identity to MSA from g-proteobacterium and the ICL domain showed 20 –25% identity to the ICL from proteobacterium and nematode. In contrast, C. elegans GCP showed very high amino acid homology to bacterial glyoxylate cycle enzymes; the MS domain showed about 60% identity to a- and g-proteobacterial MSA and the ICL domain showed about 70% identity to a-proteobacterial ICL. Genes encoding ICL and MS proteins reside in the same operon named ace in certain bacteria, including E. coli. We found two different gene orders of bacterial ace operon on
the bacterial genome database and COG database (Tatusov et al., 1997) (Fig. 2B). One operon structure is aceB (gene encoding MSA) –aceA (gene encoding ICL). This type of operon structure is often observed in the bacterial genome database; for example, E. coli (aceB locates 30-bp upstream of aceA) and Vibrio vulnificus YJ016 (68 bp). Another type of operon gene order is aceA (ICL)– aceB (MSA). We found only three bacterial species having this operon structure in the bacterial genome database; Streptomyces coelicolor A3(2) (aceA locates 101 bp upstream of aceB), Streptomyces avermitilis MA-4680 (145 bp) and Caulobacter crescentus (343 bp). Difference of domain structure and phylogeny between E. gracilis GCE and C. elegans GCP might arise through origins that have a different operon structure of ancient genes. Our analyses showed that gene fusion between MS and ICL had occurred at different times in the ancestor of E. gracilis and that of C. elegans. In E. gracilis and C. elegans, most of precursor RNAs are processed by trans-splicing at the 5Vends of genes by adding a specific trans-spliced leader (Tessier et al., 1991; Blumenthal, 1995). In addition, up to 15% of the genes are organized in polycistronic operons in C. elegans (Blumenthal et al., 2002). Polycistronic precursor RNAs are also processed by trans-splicing. Gene fusion events for making bifunctional enzymes may relate to trans-splicing and polycistronic transcription. Stechmann and Cavalier-Smith (2002) analyzed a derived gene fusion between dihydrofolate reductase (DHFR) and thymidylate synthase (TS) in order to locate the root of the eukaryote tree. They assumed the genes fused just once and were never secondarily split or laterally transferred within eukaryotes. Take our findings into account, gene fusion events possibly occurred more than once in some cases. 3.4. Catalytic properties of EgGCE In EgGCE, pH optima for the ICL activity and MS activity were observed to be 6.5, 7.5 –8.5, respectively. Kinetic profile in the ICL reaction for Mg2 + –isocitrate complex showed Michaelis – Menten kinetics, and the apparent Km value for the substrate was estimated to be 7.6 mM. In the MS reaction, Michaelis– Menten kinetics was also observed for both acetyl-CoA and glyoxylate, and the apparent Km values for these substrates were 25 and 40 AM, respectively. We determined the EgGCE overall reaction, sequential ICL and MS reactions. We measured the isocitrate-dependent acetyl-CoA degradation. The reaction scheme is as follows:
Fig. 2. Deduced models of the structure and the origin of E. gracilis bifunctional glyoxylate cycle enzyme. (A) Comparison of the structures between two bifunctional glyoxylate cycle enzymes, EgGCE and CeGCP. (B) Genomic organization of bacterial malate synthases and isocitrate lyases in ace operon. Black arrows show aceB gene that encodes malate synthase A. White arrows show aceA gene that encodes isocitrate lyase. The figure is not drawn to scale.
Isocitrate Y Glyoxylate þ SuccinateðReaction1; ICLÞ Glyoxylate þ Acetyl CoA þ H2 O Y Malate þ CoAðReaction2; MSÞ:
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
Fig. 3. EgGCE overall reaction and ICL activities. (A) Comparison between EgGCE overall reaction activity and ICL activity at the different concentrations of isocitrate. White bars show ICL activity. The reaction mixture for measuring ICL activity contained 100 mM potassium phosphate buffer, pH 6.5, 30 mM MgCl2, 2 mM dithiothreitol, 4 mM phenylhydrazine – HCl, indicated concentration of sodium isocitrate, and the enzyme in the total volume of 1 mL. Black bars show EgGCE overall reaction activity. The reaction mixture for measuring the overall reaction activity contained 100 mM potassium phosphate buffer, pH 6.5, 30 mM MgCl2, 2 mM dithiothreitol, indicated concentration of sodium isocitrate, 150 AM acetyl-CoA and the enzyme in the total volume of 1 mL. Each experiment was triplicated. Error bars show TSD. (B) Effect of acetyl-CoA on the ICL activity of EgGCE. ICL activities were measured containing various concentrations of acetyl-CoA in the reaction mixture. Cleavage reaction of isocitrate to succinate and glyoxylate was assayed by measuring the formation of glyoxylate – phenylhydrazone at 334 nm. Each experiment was triplicated. Error bars show TSD. (C) Lineweaver – Burk plots for the effects of acetyl-CoA on the ICL activity. Isocitrate concentrations were changed in the ICL reaction mixtures. Concentrations of acetyl-CoA were 150 AM. The definition of each symbol is the following; open circle, ICL activity without acetyl-CoA; closed square, ICL activity with 150 AM acetyl-CoA. Each experiment was triplicated, and a typical result is shown.
451
Overall reaction activity should be lower than the ICL activity if there are no special effects in these reactions. The overall reaction mixture contained 100 mM potassium phosphate buffer, pH 6.5, 30 mM MgCl2, 2 mM dithiothreitol, 150 AM acetyl-CoA, sodium isocitrate, and the enzyme in the total volume of 1 mL. The concentrations of sodium isocitrate in the reaction mixture were adjusted to 1, 7 and 30 mM. These isocitrate concentrations are correspond to Mg2 + – isocitrate complex concentrations of 0.975, 6.78 and 25.6 mM, respectively. Overall reaction activity was lower than the ICL activity in the isocitrate concentration of 30 mM. However, the specific activities of the overall reaction were higher than that of the ICL reactions in the isocitrate conditions of 1 and 7 mM. In order to determine the reason why the overall reaction activity of EgGCE was higher than its ICL activity, we added acetyl-CoA, which was the substrate for the MS reaction, into the ICL reaction mixture (Fig. 3B). The concentration of sodium isocitrate was adjusted to 1 mM. ICL activity measured in the presence of 150 AM acetyl-CoA was increased more than four times compared to the activity measured in the absence of acetyl-CoA. In addition, the ICL activity with acetylCoA showed almost the same activity as the overall reaction activity in these experimental conditions (Fig. 3A and 3B). Acetyl-CoA also affects the isocitrate lyase activity at the concentration of 5 AM or above. It is considered that acetyl-CoA also increases the ICL activity in vivo. Acetyl-CoA did not act as an allosteric effector of the ICL activity of EgGCE (Fig. 3C). The apparent Km value for the Mg2 + – isocitrate of ICL with 150 AM acetyl-CoA was 2500 AM. There has been no research concerning whether CoA derivatives increase ICL activity. In fact, acetyl-CoA did not affect the activity of ICL from Bacillus stearothermiphilus (Fluka) at the concentration of 150 AM (data not shown). EgGCE shows the specific catalytic property of ICL, namely, its activity is increased by acetyl-CoA. Not only the deduced amino acid sequence but also the catalytic features are different from other ICLs. We need further study to determine the mechanisms of how acetyl-CoA regulates ICL.
4. Conclusions In this paper we reported that the bifunctional glyoxylate cycle enzyme in E. gracilis had special catalytic features, i.e., its ICL activity was increased by acetyl-CoA. This property was beneficial to sequential ICL and MS reactions under physiological conditions. We also report that EgGCE had an inverted domain structure compared to C. elegans bifunctional GCP. This is an important finding to understand the ancient formation of bifunctional enzymes.
452
M. Nakazawa et al. / Comparative Biochemistry and Physiology, Part B 141 (2005) 445 – 452
References Anstrom, D.M., Kallio, K., Remington, S.J., 2003. Structure of the Escherichia coli malate synthase G: pyruvate:acetyl-coenzyme A ˚ resolution. Protein Sci. 12, abortive ternary complex at 1.95 A 1822 – 1832. Blumenthal, T., 1995. Trans-splicing and polycistronic transcription in Caenorhabditis elegans. Trends Genet. 11, 132 – 136. Blumenthal, T., Evans, D., Link, C.D., Guffanti, A., Lawson, D., Thierry Mieg, J., Thierry Mieg, D., Chiu, W.L., Duke, K., Kiraly, M., Kim, S.K., 2002. A global analysis of Caenorhabditis elegans operons. Nature 417, 851 – 854. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248 – 254. Britton, K., Langridge, S., Baker, P.J., Weeradechapon, K., Sedelnikova, S.E., De Lucas, J.R., Rice, D.W., Turner, G., 2000. The crystal structure and active site location of isocitrate lyase from the fungus Aspergillus nidulans. Struct. Fold Des. 8, 349 – 362. Collins, N., Merrett, M.J., 1975. Microbody-marker enzymes during transition from phototrophic to organotrophic growth in Euglena. Plant Physiol. 55, 1018 – 1022. Cook, J.R., 1970. Properties of partially purified malate synthase for Euglena gracilis. J. Protozool. 17, 232 – 235. Cramer, M., Myers, J., 1952. Growth and photosynthetic characteristics of Euglena gracilis. Arch. Mikrobiol. 17, 384 – 402. Eastmond, P.J., Graham, I.A., 2001. Re-examining the role of the glyoxylate cycle in oilseeds. Trends Plant Sci. 6, 72 – 78. Giachetti, E., Pinzauti, G., Bonaccorsi, R., Vanni, P., 1988. Isocitrate lyase from Pinus pinea. Characterization of its true substrate and the action of magnesium ions. Eur. J. Biochem. 172, 85 – 91. Graves, L.B.J., Trelease, R.N., Grill, A., Becker, W.M., 1972. Localization of glyoxylate cycle enzymes in glyoxysomes in Euglena. J. Protozool. 19, 527 – 532. Hartig, A., Simon, M.M., Schuster, T., Daugherty, J.R., Yoo, H.S., Cooper, T.G., 1992. Differentially regulated malate synthase genes participate in carbon and nitrogen metabolism of S. cerevisiae. Nucleic Acids Res. 20, 5677 – 5686. Howard, B.R., Endrizzi, J.A., Remington, S.J., 2000. Crystal structure of Escherichia coli malate synthase G complexed with magnesium and ˚ resolution: mechanistic implications. Biochemistry glyoxylate at 2.0 A 39, 3156 – 3168. Inui, H., Miyatake, K., Nakano, Y., Kitaoka, S., 1992. Synthesis of reserved polysaccharide from wax esters accumulated as the result of anaerobic energy generation in Euglena gracilis returned from anaerobic to aerobic conditions. Int. J. Biochem. 24, 799 – 803. Kornberg, H.L., Krebs, H.A., 1957. Synthesis of cell constituents from C2units by a modified tricarboxylic acid cycle. Nature 179, 988 – 991. Kornberg, H.L., 1966. The role and control of the glyoxylate cycle in Escherichia coli. Biochem. J. 99, 1 – 11.
Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680 – 685. Liu, F., Thatcher, J.D., Barral, J.M., Epstein, H.F., 1995. Bifunctional glyoxylate cycle protein of Caenorhabditis elegans: a developmentally regulated protein of intestine and muscle. Dev. Biol. 169, 399 – 414. Liu, F., Thatcher, J.D., Epstein, H.F., 1997. Induction of glyoxylate cycle expression in Caenorhabditis elegans: a fasting response throughout larval development. Biochemistry 36, 255 – 260. Lorenz, M.C., Fink, G.R., 2001. The glyoxylate cycle is required for fungal virulence. Nature 412, 83 – 86. Malhotra, O.P., Srivastava, P.K., 1982. Isolation and characterization of isocitrate lyase of castor endosperm. Arch. Biochem. Biophys. 214, 164 – 171. Molina, I., Pellicer, M.T., Badia, J., Aguilar, J., Baldoma, L., 1994. Molecular characterization of Escherichia coli malate synthase G. Differentiation with the malate synthase A isoenzyme. Eur. J. Biochem. 224, 541 – 548. Oda, Y., Nakano, Y., Kitaoka, S., 1982. Utilization and toxicity of exogenous amino acid in Euglena gracilis. J. Gen. Microbiol. 128, 853 – 858. Ono, K., Kondo, M., Osafune, T., Miyatake, K., Inui, H., Kitaoka, S., Nishimura, M., Nakano, Y., 2003. Presence of glyoxylate cycle enzymes in the mitochondria of Euglena gracilis. J. Eukaryot. Microbiol. 50, 92 – 96. Serrano, J.A., Camacho, M., Bonete, M.J., 1998. Operation of glyoxylate cycle in halophilic archaea: presence of malate synthase and isocitrate lyase in Haloferax volcanii. FEBS Lett. 434, 13 – 16. Sharma, V., Sharma, S., Hoener zu Bentrup, K., McKinney, J.D., Russell, D.G., Jacobs, W.R.J., Sacchettini, J.C., 2000. Structure of isocitrate lyase, a persistence factor of Mycobacterium tuberculosis. Nat. Struct. Biol. 7, 663 – 668. Shirzadegan, M., Christie, P., Seemann, J.R., 1991. An efficient method for isolation of RNA from tissue cultured plant cells. Nucleic Acids Res. 19, 6055. Stechmann, A., Cavalier-Smith, T., 2002. Rooting the eukaryote tree by using a derived gene fusion. Science 297, 89 – 91. Tatusov, R.L., Koonin, E.V., Lipman, D.J., 1997. A genomic perspective on protein families. Science 278, 631 – 637. Taylor, K.M., Kaplan, C.P., Gao, X., Baker, A., 1996. Localization and targeting of isocitrate lyases in Saccharomyces cerevisiae. Biochem. J. 319, 255 – 262. Tessier, L.H., Keller, M., Chan, R.L., Fournier, R., Weil, J.H., Imbault, P., 1991. Short leader sequences may be transferred from small RNAs to pre-mature mRNAs by trans-splicing in Euglena. EMBO J. 10, 2621 – 2625. Woodcock, E., Merrett, M.J., 1978. Purification and immunochemical characterization of malate synthase from Euglena gracilis. Biochem. J. 173, 95 – 101.