Specialized Roles of the Two Pathways for the Synthesis of Mannosylglycerate in Osmoadaptation and Thermoadaptation of Rhodothermus marinus
2004; Elsevier BV; Volume: 279; Issue: 11 Linguagem: Inglês
10.1074/jbc.m312186200
ISSN1083-351X
AutoresNuno Borges, Joey D. Marugg, Nuno Empadinhas, Milton S. da Costa, Helena Santos,
Tópico(s)Enzyme Structure and Function
ResumoRhodothermus marinus responds to fluctuations in the growth temperature and/or salinity by accumulating mannosylglycerate (MG). Two alternative pathways for the synthesis of MG have been identified in this bacterium: a single-step pathway and a two-step pathway. In this work, the genetic and biochemical characterization of the two-step pathway was carried out with the goal of understanding the function of the two pathways and their regulatory mechanisms. Mannosyl-3-phosphoglycerate synthase (MPGS) of the two-step pathway was purified from R. marinus. Sequence information led to the isolation of two contiguous genes, mpgs (encoding MPGS) and mpgp (encoding mannosyl-3-phosphoglycerate phosphatase). The recombinant MPGS had a low specific activity compared with other homologous MPGSs and contained ∼30 additional residues at the C terminus. Truncation of this extension produced a protein with a 10-fold higher specific activity. Moreover, the activity of the complete MPGS was enhanced upon incubation with R. marinus cell extracts, and protease inhibitors abolished activation. Therefore, the C-terminal peptide of MPGS was identified as a regulatory site for short term control of MG synthesis in R. marinus. The control of gene expression by heat and osmotic stress was also studied; the level of mannosylglycerate synthase involved in the single-step pathway was selectively enhanced by heat stress, whereas MPGS was overproduced in response to osmotic stress. The concomitant changes in the level of MG were assessed as well. We conclude that the two alternative pathways for the synthesis of MG are differently regulated at the level of expression to play specific roles in the adaptation of R. marinus to two different types of aggression. This is the only example of pathway multiplicity being rationalized in terms of the need to respond efficiently to distinct environmental stresses. Rhodothermus marinus responds to fluctuations in the growth temperature and/or salinity by accumulating mannosylglycerate (MG). Two alternative pathways for the synthesis of MG have been identified in this bacterium: a single-step pathway and a two-step pathway. In this work, the genetic and biochemical characterization of the two-step pathway was carried out with the goal of understanding the function of the two pathways and their regulatory mechanisms. Mannosyl-3-phosphoglycerate synthase (MPGS) of the two-step pathway was purified from R. marinus. Sequence information led to the isolation of two contiguous genes, mpgs (encoding MPGS) and mpgp (encoding mannosyl-3-phosphoglycerate phosphatase). The recombinant MPGS had a low specific activity compared with other homologous MPGSs and contained ∼30 additional residues at the C terminus. Truncation of this extension produced a protein with a 10-fold higher specific activity. Moreover, the activity of the complete MPGS was enhanced upon incubation with R. marinus cell extracts, and protease inhibitors abolished activation. Therefore, the C-terminal peptide of MPGS was identified as a regulatory site for short term control of MG synthesis in R. marinus. The control of gene expression by heat and osmotic stress was also studied; the level of mannosylglycerate synthase involved in the single-step pathway was selectively enhanced by heat stress, whereas MPGS was overproduced in response to osmotic stress. The concomitant changes in the level of MG were assessed as well. We conclude that the two alternative pathways for the synthesis of MG are differently regulated at the level of expression to play specific roles in the adaptation of R. marinus to two different types of aggression. This is the only example of pathway multiplicity being rationalized in terms of the need to respond efficiently to distinct environmental stresses. Thermophilic and hyperthermophilic organisms, like all organisms living in aqueous environments, are faced with alterations in the water activity to which they must adjust to grow. However, the compatible solutes of these organisms are generally different from those of mesophiles (1Santos H. da Costa M.S. Methods Enzymol. 2001; 334: 302-315Crossref PubMed Scopus (59) Google Scholar). Mannosylglycerate (MG) 1The abbreviations used are: MG, α-mannosylglycerate; Bis/Tris/propane, 2-[bis(hydroxyethyl)amino]-2-(hydroxymethyl)-propane-1,3-diol; MPG, α-mannosyl-3-phosphoglycerate; MPGS, mannosyl-3-phosphoglycerate synthase; MPGP, mannosyl-3-phosphoglycerate phosphatase; MGS, mannosylglycerate synthase. and the derivative compound mannosylglyceramide are the two major osmolytes used by the thermophilic bacterium Rhodothermus marinus to cope with osmotic stress at or below the optimum growth temperature (65 °C). Near the maximum growth temperature (77 °C), however, MG is the only compatible solute detected (2Silva Z. Borges N. Martins L.O. Wait R. da Costa M.S. Santos H. Extremophiles. 1999; 3: 163-172Crossref PubMed Scopus (72) Google Scholar). Therefore, MG appears to be particularly suited to protect cells under conditions of heat aggression. In line with this view, MG is widely distributed among hyperthermophilic archaea and thermophilic bacteria, e.g. the slightly halophilic euryarchaeotes of the genera Pyrococcus and Thermococcus, three species of the genus Archaeoglobus, the crenarchaeote Aeropyrum pernix, and the bacteria Thermus thermophilus and Rubrobacter xylanophilus (3Lamosa P. Martins L.O. da Costa M.S. Santos H. Appl. Environ. Microbiol. 1998; 64: 3591-3598Crossref PubMed Google Scholar, 4Martins L.O. Santos H. Appl. Environ. Microbiol. 1995; 61: 3299-3303Crossref PubMed Google Scholar, 5Gonçalves L.G. Huber R. da Costa M.S. Santos H. FEMS Microbiol. Lett. 2003; 218: 239-244Crossref PubMed Google Scholar, 6Nunes O.C. Manaia C.M. da Costa M.S. Santos H. Appl. Environ. Microbiol. 1995; 61: 2351-2357Crossref PubMed Google Scholar, 7Santos H. da Costa M.S. Environ. Microbiol. 2002; 4: 501-509Crossref PubMed Scopus (205) Google Scholar). The synthesis of sugar-derivative compatible solutes, such as sucrose, trehalose, glucosylglycerol, and galactosylglycerol, most often proceeds via a two-step pathway involving a phosphorylated intermediate (8Curatti L. Folco E. Desplats P. Abratti G. Limones V. Herrera-Estrella L. Salerno G. J. Bacteriol. 1998; 180: 6776-6779Crossref PubMed Google Scholar, 9Giæver H.M. Styrvold O.B. Kaasen I. Strøm A.R. J. Bacteriol. 1988; 170: 2841-2849Crossref PubMed Google Scholar, 10Thomson K.-S. Biochim. Biophys. Acta. 1983; 759: 154-159Crossref Scopus (8) Google Scholar, 11Hagemann M. Effmert U. Kerstan T. Schoor A. Erdmann N. Curr. Microbiol. 2001; 43: 278-283Crossref PubMed Scopus (31) Google Scholar). In each case, the respective sugar nucleotide is combined with the appropriate polyol/sugar phosphate, yielding a phosphorylated intermediate that is subsequently hydrolyzed to form the functional osmolyte. The genes encoding the enzymes involved in the synthase/phosphatase pathways are often found in operon-like structures: trehalose synthesis in Escherichia coli or MG synthesis in Pyrococcus spp. and T. thermophilus are examples of this organization (12Kaasen I. McDougall J. Strøm A.R. Gene (Amst.). 1994; 145: 9-15Crossref PubMed Scopus (99) Google Scholar, 13Empadinhas N. Marugg J.D. Borges N. Santos H. da Costa M.S. J. Biol. Chem. 2001; 276: 43580-43588Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar, 14Empadinhas N. Albuquerque L. Henne A. Santos H. da Costa M.S. Appl. Environ. Microbiol. 2003; 69: 3272-3279Crossref PubMed Scopus (29) Google Scholar). However, in Synechocystis spp. the genes encoding the synthase and the phosphatase for glucosylglycerol synthesis are under the control of distinct promoters (15Hagemann M. Schoor A. Jeanjean R. Zuther E. Joset F. J. Bacteriol. 1997; 179: 1727-1733Crossref PubMed Google Scholar, 16Marin K. Zuther E. Kerstan T. Kunert A. Hagemann M. J. Bacteriol. 1998; 180: 4843-4849Crossref PubMed Google Scholar). R. marinus is the only organism known to have two distinct pathways for the synthesis of MG: the two-step pathway, involving a phosphorylated intermediate, which is also found in Pyrococcus spp., A. pernix, and T. thermophilus, and the single-step pathway that, until recently, appeared to be confined to R. marinus (13Empadinhas N. Marugg J.D. Borges N. Santos H. da Costa M.S. J. Biol. Chem. 2001; 276: 43580-43588Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar, 14Empadinhas N. Albuquerque L. Henne A. Santos H. da Costa M.S. Appl. Environ. Microbiol. 2003; 69: 3272-3279Crossref PubMed Scopus (29) Google Scholar, 17Martins L.O. Empadinhas N. Marugg J.D. Miguel C. Ferreira C. da Costa M.S. Santos H. J. Biol. Chem. 1999; 274: 35407-35414Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). Two sequences recently identified in Griffithsia japonica (GenBank™ accession numbers AY123119 and AF542028) show high homology with the gene encoding mannosylglycerate synthase (MGS) in R. marinus, suggesting the presence of the single-step pathway in this red alga, as well. The presence of two alternative pathways for the synthesis of MG in R. marinus is therefore intriguing. The biochemical and genetic characterization of the single-step pathway was reported earlier by our team (17Martins L.O. Empadinhas N. Marugg J.D. Miguel C. Ferreira C. da Costa M.S. Santos H. J. Biol. Chem. 1999; 274: 35407-35414Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). In the present work, the genetic and biochemical characterization of the two-step pathway was carried out with the goal of understanding the function of the two pathways and their regulatory mechanisms. The mannosyl-3-phosphoglycerate synthase (MPGS) activity that converts GDP-mannose and d-3-phosphoglycerate to mannosyl-3-phosphoglycerate (MPG) was purified from R. marinus cells, and the amino acid sequence of several fragments was used to identify the genes encoding mannosyl-3-phosphoglycerate synthase and mannosyl-3-phosphoglycerate phosphatase (MPGP). These genes were separately cloned and overexpressed in E. coli, the respective recombinant enzymes were characterized in detail, and the regulation of the two pathways was investigated. R. marinus Strain and Growth Conditions—R. marinus strain DSM 4252T (Deutsche Sammlung von Mikroorganismen und Zellkulturen, Braunschweig, Germany) was grown as previously described (18Nunes O.C. Donato M.M. da Costa M.S. Syst. Appl. Microbiol. 1992; 15: 92-97Crossref Scopus (44) Google Scholar). For the evaluation of protein levels by immunoblotting, batch cultures were grown at the optimum NaCl concentration (2% NaCl) at the temperatures 57.5, 65, and 75 °C. Cultures were also grown at the optimum growth temperature (65 °C) in media with different NaCl concentrations (1, 2, 4, and 6%). Preparation of R. marinus Cell Extracts—Cells were harvested during the late-exponential phase of growth. The cell pellet was suspended in 20 mm Tris-HCl, pH 7.6, containing 5 mm MgCl2, DNase I (10 μg/ml), and a mixture of protease inhibitors: phenylmethylsulfonyl fluoride (80 μg/ml), leupeptin (20 μg/ml), and antipain (20 μg/ml). The cells were disrupted in a French press. The amount of protein was estimated by the method of Bradford (19Bradford M.M. Anal. Biochem. 1976; 72: 248-254Crossref PubMed Scopus (217544) Google Scholar). Enzyme Assays—During the purification, MPGS activity at 75 °C was detected with a TLC assay. MPGS activity was determined from the 1H NMR quantification of the product, MPG (13Empadinhas N. Marugg J.D. Borges N. Santos H. da Costa M.S. J. Biol. Chem. 2001; 276: 43580-43588Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). The MPGP activity was assayed by determining the inorganic phosphate released by MPG (13Empadinhas N. Marugg J.D. Borges N. Santos H. da Costa M.S. J. Biol. Chem. 2001; 276: 43580-43588Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). Purification of Native MPGS—The native MPGS was purified by fast protein liquid chromatography (Amersham Biosciences). Ion Exchange Chromatography—The cell extracts were applied to a DEAE-Sepharose column (20 mm Tris-HCl, pH 7.6), and MPGS activity was detected in the flow-through. Active samples were loaded onto a hydroxyapatite column and eluted with a step gradient of potassium phosphate buffer (50–500 mm, pH 7.6). Fractions containing MPGS, eluted at 50 mm, were applied on a Q-Sepharose column (20 mm Tris-HCl, pH 8.0) and eluted with a linear gradient (0–1 m NaCl) in the same buffer. MPGS activity was detected in the fractions eluted at 0.4 m of NaCl. Hydrophobic Chromatography—The active fractions were dialyzed against Tris-HCl (20 mm, pH 7.6) containing 0.5 m ammonium sulfate. The sample was applied to a phenyl-Sepharose column, and elution was carried out with a linear gradient of ammonium sulfate (0.5–0 m). MPGS activity was detected at 0.26 m of (NH4)2SO4. High Resolution Ion Exchange Chromatography—A Mono Q column (20 mm Tris-HCl, pH 6.8) was used; the enzyme eluted at 0.45 m NaCl with a linear gradient (0–1 m NaCl). The active fractions were applied onto a Mono S column (20 mm Tris-HCl, pH 6.5). MPGS activity eluted at 0.35 m NaCl. The amino acid sequences of the N terminus and of two internal fragments were determined (20Edman P. Begg G. Eur. J. Biochem. 1967; 1: 80-91Crossref PubMed Scopus (2426) Google Scholar). Molecular mass was determined by SELDI time-of-flight mass spectrometry (Microchemical Facility, Emory University School of Medicine). Identification of the mpgs and mpgp Genes—Degenerate primers based on the two internal amino acid sequences (FLNQLISYYTGFETE and GEEHIDDMILDDLQVIYH) of the purified native MPGS were designed. A degenerate reverse primer was designed based on a conserved region (REYSE) of known MPGPs (13Empadinhas N. Marugg J.D. Borges N. Santos H. da Costa M.S. J. Biol. Chem. 2001; 276: 43580-43588Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). Chromosomal DNA was isolated as described by Marmur (21Marmur J. J. Mol. Biol. 1961; 3: 208-218Crossref Scopus (9014) Google Scholar). A 315-bp PCR fragment of mpgs was amplified, labeled with digoxigenin (Roche Applied Science) and used as a probe to screen a R. marinus genomic library (17Martins L.O. Empadinhas N. Marugg J.D. Miguel C. Ferreira C. da Costa M.S. Santos H. J. Biol. Chem. 1999; 274: 35407-35414Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). A 2.8-kilobase pair insert containing the entire mpgs gene plus a 42-bp fragment of the mpgp gene was obtained. The entire mpgp gene was obtained using a similar strategy. Cloning and Expression of mpgs, Truncated mpgs, and mpgp Genes— The full-length mpgs and mpgs lacking 99-bp in the C terminus region were amplified and cloned separately in pKK223-3 plasmid (Amersham Biosciences) between the EcoRI and HindIII sites. The mpgp gene was cloned between the EcoRI and PstI sites. Cloning methodology followed standard protocols (22Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual.2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). E. coli BL-21DE cells bearing the constructs were grown at 37 °C in LB medium supplemented with ampicillin (100 μg/ml) to an A600 of 0.8 and induced with 1 mm isopropyl-β-d-thiogalactopyranoside for 6 h. Purification of Recombinant Proteins—E. coli extracts were prepared as described above, but 1 mm dithiothreitol and 2 mm EDTA were added. The proteins were purified from heat-treated cell extracts (10 min at 70 °C) in three chromatographic steps: Q-Sepharose, Resource Q, and Superose 6. The purity of the final protein preparations was evaluated by SDS-PAGE. Characterization of Recombinant Enzymes—The temperature profiles for activity were studied between 40 and 95 °C at pH 8.2. The pH profiles of MPGS and truncated MPGS were determined at 80 °C using 50 mm Bis/Tris/propane-HCl buffer in the pH range 5–10. The effect of pH on MPGP was determined at 80 °C using 50 mm Bis/Tris/propane-HCl buffer in the pH range 5–10 and 50 mm acetate buffer in the pH range 3–5. Long term stability was evaluated from the residual activity of the enzymes (0.84 mg/ml for MPGS, 0.2 mg/ml for truncated MPGS, and 0.5 mg/ml for MPGP) after heat treatment at 80 °C for different time periods. Kinetic parameters (Vmax and Km) were determined under optimum conditions. All of the reactions were done in duplicate. Activation of MPGS by R. marinus Cell Extracts—The full-length MPGS (0.26 mg) was incubated at 72 °C for various periods of time with cell extracts of R. marinus (1.1 mg) grown at 65 °C in medium containing 2% NaCl. MPGS activity was assayed by 1H NMR as previously described. Activation of endogenous MPGS was also determined in control experiments. The activation measurements were also carried out in the presence of a mixture of protease inhibitors purchased from Roche Applied Science. To assess a potential activation of the enzyme by temperature alone, full-length MPGS was incubated at 72 °C with 1.0 mg of a thermostable protein (recombinant rubredoxin of Desulfovibrio gigas (23Lamosa P. Burke A. Peist R. Huber R. Liu M.Y. Silva G. Rodrigues-Pousada C. LeGall J. Maycock C. Santos H. Appl. Environ. Microbiol. 2000; 66: 1974-1979Crossref PubMed Scopus (92) Google Scholar)) to mimic the protein content in the R. marinus extracts. The experiments were repeated three times. Levels of Enzymes and Intracellular Solutes in Response to Heat or Osmotic Stress—The cells were grown to mid-exponential phase at the optimal temperature (65 °C) and NaCl concentration (2%). At this stage, the cultures were transferred to an incubator at 75 °C and grown further for 4 h. Osmotic stress conditions were imposed by the addition of a sterile NaCl solution to obtain a final concentration of 6% NaCl. The cells were grown further for 4 h under osmotic stress. Culture samples were withdrawn after 0, 2, and 4 h under stress conditions and used to determine protein (MGS and MPGS) levels by immunoblotting and to determine intracellular solutes by 1H NMR. The experiments were performed in duplicate. Western Blot Analysis—R. marinus cells were lysed by boiling in Laemmli′s buffer. After electrophoresis on SDS-PAGE, the proteins were transferred onto polyvinylidene fluoride membranes (Millipore) that were incubated overnight with rabbit primary antibodies (anti-MGS or anti-MPGS produced at Eurogentec). Binding was detected with an ECL system (Amersham Biosciences) after treatment with anti-rabbit antibodies conjugated with horseradish peroxidase (Sigma). The signals were quantified by using a program written in MatLab (MathWorks, Inc.). Sequence Analysis of mpgs and mpgp Genes—Two MPGS proteins with the same N-terminal amino acid sequence (MRIEIP) and with molecular masses of 48,795 and 45,584 Da, determined by mass spectrometry, were isolated from R. marinus extracts (Fig. 1). A 3.5-kilobase pair fragment of R. marinus carrying the mpgs (encoding MPGS) contiguous to mpgp (encoding MPGP) genes was isolated from the genomic library (Fig. 2).Fig. 2Nucleotide sequence of the mpg operon and flanking regions in R. marinus. The predicted start codons of mpgs and mpgp genes are indicated in boldtype. The N-terminal and two internal fragments sequenced after purification of native MPGS are underlined. The putative promoter region upstream of mpgs, predicted by using a prokaryotic promoter data base (www.fruitfly.org/seq_tools/promoter.html), is surrounded by a box, and the putative ribosome-binding site is labeled RBS. The deduced amino acid sequence is also shown. The DXD motif common to several glycosyltransferase families is highlighted with black shaded boxes. Conserved motifs of acid phosphatases are showed in gray boxes.View Large Image Figure ViewerDownload Hi-res image Download (PPT) BLAST searches of the protein data bases with the R. marinus MPGS and MPGP amino acid sequences revealed high homology with several proteins (Fig. 3 and the supplemental figure). MPGS had 52% amino acid identity with MPGS from T. thermophilus, about 45% with MPGSs from three Pyrococcus spp., 36% with A. pernix, and 48% with an uncultured crenarchaeote. Given this considerable degree of identity, R. marinus MPGS should be classified in glycosyltransferase family GT55 (afmb.cnrs-mrs.fr/~cazy/CAZY/index.html), which comprises GDP-mannose:α-mannosyltransferases that retain the anomeric configuration of the substrate. Interestingly, the glycosyltransferases involved in the synthesis of other compatible solutes, such as trehalose and sucrose, are also "retaining" enzymes. The R. marinus MPGP had 43% identity with MPGP of T. thermophilus, about 38% with MPGPs of Pyrococcus spp., and 25% with MPGP of A. pernix. Despite the low degree of similarity with other osmolyte-phosphate phosphatases, R. marinus MPGP contains the two conserved motifs (DXDX(T/V)X and GDXXXD) that are characteristic of the phosphohydrolase superfamily designated "DDDD" (25Thaller M.C. Schippa S. Rossolini G.M. Protein Sci. 1998; 7: 1647-1652Crossref PubMed Scopus (135) Google Scholar). Biochemical Characterization of Recombinant MPGS and MPGP—MPGS showed absolute substrate specificity for GDPmannose and d-3-phosphoglycerate. ADP-mannose, UDP-mannose, ADP-glucose, GDP-glucose, UDP-glucose, glycerate, glycerol, d-2-phosphoglycerate, l-glycerol-3-phosphate, 2,3-diphospho-d-glycerate, and phosphoenolpyruvate were examined as potential substrates. The α-configuration of the final product, MPG, was established from the measurement of the coupling constant between the anomeric carbon, C1, and the directly bound proton, H1 (J = 171.8 Hz) (26Bock K. Pedersen C. J. Chem. Soc. Perkin Trans. 1974; II: 293-297Crossref Google Scholar). Under our experimental conditions (5 mm of each substrate), the maximum yield of MPG was only 67%. To investigate whether this incomplete conversion was due to product inhibition, MPGS activity was assessed in the presence of MPG plus GDP (1:1 ratio) or GDP alone. The activity of MPGS was inhibited by 50% at a GDP concentration of 2.5 mm; MPG (2.5 mm) or MG (5 mm) had no effect on the enzyme activity. The presence of Mg2+ was required for MPGS activity. NaCl (or KCl) were inhibitors of the enzyme (Table I). Maximal MPGS activity was observed at 80 °C and pH 7.5 (Fig. 4).Table IKinetic parameters and biochemical properties of recombinant MPGS, truncated MPGS, and MPGP from R. marinusPropertiesMPGSaAll assays were carried out at 80 °C as described under "Experimental Procedures."Truncated MPGSaAll assays were carried out at 80 °C as described under "Experimental Procedures."MPGPaAll assays were carried out at 80 °C as described under "Experimental Procedures."Km (mm)bKm values for MPGS were determined in mixtures containing 0.1-5 mm GDP-mannose and 5 mm 3-PGA, or 5 mm GDP-mannose and 0.1-5 mm 3-PGA. Km value for MPGP was determined in mixtures containing MPG (0.1-2 mm).GDP-man0.503-PGA0.63MPG0.20Vmax (μmol/min.mg)>15156200Mg2+0 mm09015 mm100cExpressed as percentages of the maximum activity.100cExpressed as percentages of the maximum activity.NaClcExpressed as percentages of the maximum activity.50 mm76100150 mm31100KClcExpressed as percentages of the maximum activity.50 mm84100150 mm36100Thermostability at 80 °C half-life (min)407990a All assays were carried out at 80 °C as described under "Experimental Procedures."b Km values for MPGS were determined in mixtures containing 0.1-5 mm GDP-mannose and 5 mm 3-PGA, or 5 mm GDP-mannose and 0.1-5 mm 3-PGA. Km value for MPGP was determined in mixtures containing MPG (0.1-2 mm).c Expressed as percentages of the maximum activity. Open table in a new tab MPGP showed absolute substrate specificity for MPG. Several sugar phosphates, mannose-1-phosphate, mannose-6-phosphate, glucose-1-phosphate, glucose-6-phosphate, glucose-1,6-bisphosphate, fructose-1-phosphate, fructose-6-phosphate, trehalose-6-phosphate, ribose-5-phosphate, GDP, and GMP, were examined as possible substrates for the enzyme. The dephosphorylation of MPG was complete under our experimental conditions, and the activity was not affected by 5 mm MG. MPGP was not strictly dependent on Mg2+, but its presence was required for maximal activity. Maximum activity was reached between 70 and 80 °C and pH 6.0 (Fig. 4). NaCl (or KCl) in the range of 0–150 mm had no effect on MPGP activity. Effect of the C-terminal Extension on MPGS Properties—The final protein preparation of the native MPGS showed two bands on SDS-PAGE (Fig. 1). The molecular mass of the upper band matched the calculated molecular mass of the deduced amino acid sequence of full-length MPGS (427 amino acids), whereas the molecular mass of the lower band corresponded to MPGS lacking 30 amino acids in the C terminus region. Interestingly, the sequence alignment of the full-length MPGS with six homologous proteins showed a C-terminal extension of ∼30 amino acids (Fig. 3). Therefore, we deemed it important to investigate the biochemical features of truncated MPGS. The specific activity of MPGS lacking 33 amino acids in the C-terminal region was 156 μmol/min·mg protein at 80 °C and optimum pH (7.5), which represents a 10-fold increase as compared with the full-length MPGS. The pH profiles for activity were very similar, but the optimal temperature of the truncated MPGS was 10 °C higher than that of the full-length protein (Fig. 4). At 100 °C truncated MPGS retained 85% of its maximum activity, whereas the full-length protein showed no activity. Moreover, the long term stability of the truncated MPGS was much higher; the half-life for inactivation at 80 °C was 79 min to be compared with 40 min determined for the full-length MPGS (Table I). Activation of MPGS by R. marinus Cell Extracts—The dramatic increase observed in the specific activity of MPGS upon deletion of 33 amino acids in the C terminus led us to hypothesize that the activity of MPGS could be regulated in vivo by proteolysis. To obtain experimental evidence to support this hypothesis, the activation profile of full-length MPGS upon incubation with R. marinus cell extracts was studied (Fig. 5). The MPGS activity increased 3-fold upon a 20-min incubation at 72 °C, and this activation was completely abolished by a mixture of protease inhibitors; the extent of activation was calculated after subtracting the activation of endogenous MPGS under the same conditions (1.1-fold). Levels of MGS and MPGS in Response to Heat or Osmotic Stress—The levels of MGS and MPGS were estimated by Western blotting, after confirming that a purified preparation of anti-MGS and anti-MPGS specifically recognized the MGS and MPGS in cell extracts of R. marinus. The measurements were carried out in cells grown under optimum conditions (2% NaCl and 65 °C) and also after imposing upshifts in the salinity of the growth medium to 6% NaCl or in temperature to 75 °C. Constitutive levels of MGS and MPGS were detected under optimum growth conditions, but the amount of MGS, the synthase of the single-step pathway, was selectively enhanced by heat stress (2.5-fold higher after 2 h), whereas the synthase of the two-step pathway, MPGS, was overproduced in response to osmotic stress (3-fold increase in 2 h) (Fig. 6). The effect of these stress conditions on the intracellular levels of compatible solutes was also assessed in the same experiments (Fig. 7). The intracellular pool of MG increased notably during the initial 2-h period of adaptation to osmotic stress. In contrast, mannosylglyceramide was undetectable in cells under optimal conditions but increased primarily during the last hour of osmotic adaptation, reaching concentrations comparable with those of MG.Fig. 7Compatible solute accumulation in R. marinus cells subjected to osmotic or heat stress. Compatible solutes in aliquots of the cell suspensions used in the assays illustrated in Fig. 6 were quantified by 1H NMR in ethanolic cell extracts. Open squares, mannosylglycerate; open circles, mannosylglyceramide.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The levels of MGS and MPGS were also evaluated in separate cell batches grown at different temperatures (57.5, 65, and 75 °C) and NaCl concentrations (1, 2, 4, and 6%). An increase in the growth temperature from 57.5 to 75 °C led to a 2-fold increase of MGS and a 20% decrease of MPGS. Conversely, the level of MPGS increased 3-fold in cells grown in medium containing 6% NaCl, compared with cells grown at optimal salinity (2% NaCl), whereas the level of MGS remained unchanged (data not shown). R. marinus is the only organism known to have two distinct pathways for the synthesis of MG. Herein, the genetic and biochemical characterization of the two-step pathway was achieved to elucidate the role of pathway duplicity in the strategies for osmo- and thermoadaptation of R. marinus. The arrangement of the genes involved in the synthesis of MG in R. marinus is similar to that of T. thermophilus (14Empadinhas N. Albuquerque L. Henne A. Santos H. da Costa M.S. Appl. Environ. Microbiol. 2003; 69: 3272-3279Crossref PubMed Scopus (29) Google Scholar), A. pernix, and Pyrococcus horikoshii (13Empadinhas N. Marugg J.D. Borges N. Santos H. da Costa M.S. J. Biol. Chem. 2001; 276: 43580-43588Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar) insofar as the mpgp gene is found immediately downstream of the mpgs gene. However, the operon-like structure of Pyrococcus spp., comprising the genes encoding the enzymes involved in the synthesis of GDP-mannose from fructose-6-phosphate, was not found in R. marinus. In this respect, the gene organization of R. marinus is most similar to that found in the bacterium T. thermophilus and the crenarchaeote A. pernix. An unrooted phylogenetic tree constructed on the basis of the alignment of the amino acid sequences of known MPGSs predicts the existence of three different clusters (Fig. 8). As expected, the bacterial MPGSs (R. marinus and T. thermophilus) group together, and the archaeal MPGSs from Pyrococcus spp. form a tight cluster. On the other hand, the MPGS of the uncultured crenarchaeote and of A. pernix form distinct branches from the other homologues. It appears, therefore, that the archaeal mpgs genes have more divergent sequences than the bacterial counterparts. However, this preliminary analysis should be regarded with caution for the lack of a larger number of sequences. The observation that recombinant R. marinus MPGS had a specific activity 1 order of magnitude lower than that of the homologous enzymes from P. horikoshii and T. thermophilus (13Empadinhas N. Marugg J.D. Borges N. Santos H. da Costa M.S. J. Biol. Chem. 2001; 276: 43580-43588Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar, 14Empadinhas N. Albuquerque L. Henne A. Santos H. da Costa M.S. Appl. Environ. Microbiol. 2003; 69: 3272-3279Crossref PubMed Scopus (29) Google Scholar) is rather intriguing. Several hypotheses can be put forward to explain this result: (i) The heterologous host used for overproduction may lack the machinery required for correct translation and/or folding. However, the specific MPGS activity was unchanged after an unfolding/refolding procedure with guanidinium chloride, suggesting that MPGS was synthesized in E. coli with the correct folding (data not shown). (ii) R. marinus MPGS could have a much lower intrinsic specific activity than P. horikoshii and T. thermophilus MPGSs and yet be sufficient to support the required synthesis of MG. (iii) Finally, the low activity could be due to a lack of suitable post-translational modifications in E. coli. The hint for this proposition was provided by the purification of two size forms of native MPGS with identical N termini and a molecular mass that differed by ∼30 residues (Fig. 1). This view was further supported by the existence of about 30 extra amino acids in the C terminus of R. marinus MPGS that are not found in other MPGSs (Fig. 3). Moreover, R. marinus MPGS contains an arginine-rich region in the C terminus that could be an adequate proteolytic cleavage site. We therefore hypothesized that a proteolytic cleavage of the C-terminal extension was the underlying cause for activation of the enzyme. Experimental evidence for this idea was the strong enhancement of the catalytic activity (10-fold) achieved by truncation of the C-terminal extension of MPGS. This result led us to question whether proteolytic cleavage could play a role in the activation of the enzyme in vivo. It is worth recalling that several enzymes involved in the synthesis of sugar or sugar-derivative osmolytes are activated by proteolysis. Activation of galactosyl-1-glycerol-3-phosphate synthase from Poterioochromonas malhamensis is attributed to the action of a membrane-bound protease (27Köhle D. Kauss H. Biochim. Biophys. Acta. 1984; 799: 59-67Crossref Scopus (3) Google Scholar). Trehalose-phosphate synthase from baker's yeast, produced as a proprotein with 115 kDa, is activated (approximately 3-fold) by the action of an endogenous protease with concomitant reduction of 20 kDa in size (28Londesborough J. Vuorio O. J. Gen. Microbiol. 1991; 137: 323-330Crossref PubMed Scopus (68) Google Scholar). The homologous synthase from Candida utilis is also activated by the action of a specific protease (29Vicente-Soler J. Arguelles J.C. Gacto M. FEMS Microbiol. Lett. 1991; 82: 157-162Google Scholar). Despite all the evidence we gathered in support for a mechanism of proteolytic activation of R. marinus MPGS, the observation of only the less active form of MPGS (full-length protein) in all the Western blotting assays remains intriguing. One could speculate that, in vivo, the inhibitory effect of the C-terminal extension would be eliminated without implying the actual cleavage of a peptide bond; for example, the interaction with a regulatory protein could lead to the conformational change required to relieve the inhibition of the catalytic domain. Although the fine details of the regulatory mechanism of MPGS remain elusive, we did show that: (i) the C-terminal peptide is implicated in the regulation of the catalytic activity of MPGS and (ii) cell extracts of R. marinus possess the proteolytic machinery to activate the full-length MPGS by cleavage of the C-terminal extension. Several sugar-derivative osmolytes from mesophiles are synthesized predominantly by a two-step pathway involving a phosphorylated intermediate (8Curatti L. Folco E. Desplats P. Abratti G. Limones V. Herrera-Estrella L. Salerno G. J. Bacteriol. 1998; 180: 6776-6779Crossref PubMed Google Scholar, 9Giæver H.M. Styrvold O.B. Kaasen I. Strøm A.R. J. Bacteriol. 1988; 170: 2841-2849Crossref PubMed Google Scholar, 10Thomson K.-S. Biochim. Biophys. Acta. 1983; 759: 154-159Crossref Scopus (8) Google Scholar, 11Hagemann M. Effmert U. Kerstan T. Schoor A. Erdmann N. Curr. Microbiol. 2001; 43: 278-283Crossref PubMed Scopus (31) Google Scholar). In addition to this one, R. marinus has a second pathway for the synthesis of MG, and the reason for pathway duality is explained in the present work. We found that the two-step and single-step pathways have specific roles in the adaptation of R. marinus to different types of stress, namely, osmotic stress and heat stress, respectively. R. marinus cells respond to heat stress by selectively enhancing the levels of MGS (the enzyme of the single-step pathway), whereas osmotic stress had a pronounced effect only in the induction of the synthesis of MPGS (the synthase of the two-step pathway). In eukaryotes, a well known strategy for adaptation to different stressful conditions involves the utilization of isoenzymes, which catalyze the same reaction but are differentially regulated by diverse types of stress. For example, in Saccharomyces cerevisiae, the two glycerol-3-phosphate dehydrogenase isoenzymes involved in the synthesis of glycerol play a dominant role in different stressful situations; one is primarily produced for osmoadaptation, and the other is primarily produced for adaptation to anoxic conditions (30Ansell R. Granath K. Hohmann S. Thevelein J.M. Adler L. EMBO J. 1997; 16: 2179-2187Crossref PubMed Scopus (433) Google Scholar). In bacteria, pathway multiplicity for trehalose synthesis has been reported in a few cases: mycobacteria (31De Smet K.A.L. Weston A. Brown I.N. Young D.B. Robertson B.D. Microbiology. 2000; 146: 199-208Crossref PubMed Scopus (209) Google Scholar), Corynebacterium glutamicum (32Wolf A. Krämer R. Morbach S. Mol. Microbiol. 2003; 49: 1119-1134Crossref PubMed Scopus (152) Google Scholar), and T. thermophilus (33Silva Z. Alarico S. Nobre A. Horlacher R. Marugg J. Boos W. Mingote A.I. da Costa M.S. J. Bacteriol. 2003; 185: 5943-5952Crossref PubMed Scopus (40) Google Scholar). It is assumed that the multiplicity of routes used to synthesize the same metabolite underlines the physiological importance of the final product. Yet the significance of the metabolic multiplicity is generally elusive. In this context, a careful, recent study proposes that among the three routes for the synthesis of trehalose in C. glutamicum, the maltooligosyltrehalose synthase:trehalohydrolase pathway is involved in osmoadaptation, the trehalose synthase pathway is important for trehalose degradation, and the role of the trehalose-6-phosphate synthase/phosphatase pathway remains unclear (32Wolf A. Krämer R. Morbach S. Mol. Microbiol. 2003; 49: 1119-1134Crossref PubMed Scopus (152) Google Scholar). It appears that R. marinus has developed an ingenious strategy to cope with stress imposed by salt or heat. To our knowledge this is the only example of pathway multiplicity being rationalized in terms of the need to respond efficiently to distinct environmental aggressions. This finding also reinforces the view that MG plays a dual role in osmoadaptation and thermoadaptation of R. marinus. In Pyrococcus spp., the level of MG increases dramatically with the NaCl concentration in the growth medium, but MG appears to have a negligible contribution to protection against heat stress (4Martins L.O. Santos H. Appl. Environ. Microbiol. 1995; 61: 3299-3303Crossref PubMed Google Scholar, 13Empadinhas N. Marugg J.D. Borges N. Santos H. da Costa M.S. J. Biol. Chem. 2001; 276: 43580-43588Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). Unlike Pyrococcus spp., R. marinus accumulates MG in response to both osmotic and heat stress; thermoprotection appears to be provided exclusively by MG, whereas MG and mannosylglyceramide contribute to osmoadaptation (2Silva Z. Borges N. Martins L.O. Wait R. da Costa M.S. Santos H. Extremophiles. 1999; 3: 163-172Crossref PubMed Scopus (72) Google Scholar). This conclusion is fully corroborated by the profile of accumulation of these solutes in response to osmotic or heat shock (Fig. 7). In conclusion, the two alternative pathways for the synthesis of MG are differently regulated at the level of expression to play specific roles in the adaptation of R. marinus to two different types of aggression. Moreover, we have shown that the 30-amino acid peptide in the C terminus of MPGS is a regulatory site for short term control of MG synthesis. These results represent an important step toward the full elucidation of mechanisms of osmo- and thermoadaptation in thermophilic organisms, an ambitious goal that can be achieved with the development of genetic tools for manipulation of R. marinus. We are grateful to Prof. Adriano Henriques (Instituto de Tecnologia Química e Biológica, Oeiras, Portugal) for enlightening discussions and helpful advice. We thank Ana Mingote for technical assistance. Download .pdf (.12 MB) Help with pdf files
Referência(s)