A novel adaptation of aldolase regulates virulence in Streptococcus pyogenes
2006; Springer Nature; Volume: 25; Issue: 22 Linguagem: Inglês
10.1038/sj.emboj.7601393
ISSN1460-2075
AutoresJennifer A. Loughman, Michael G. Caparon,
Tópico(s)Neonatal and Maternal Infections
ResumoArticle26 October 2006free access A novel adaptation of aldolase regulates virulence in Streptococcus pyogenes Jennifer A Loughman Jennifer A Loughman Department of Molecular Microbiology, Washington University School of Medicine, St Louis, MO, USA Search for more papers by this author Michael G Caparon Corresponding Author Michael G Caparon Department of Molecular Microbiology, Washington University School of Medicine, St Louis, MO, USA Search for more papers by this author Jennifer A Loughman Jennifer A Loughman Department of Molecular Microbiology, Washington University School of Medicine, St Louis, MO, USA Search for more papers by this author Michael G Caparon Corresponding Author Michael G Caparon Department of Molecular Microbiology, Washington University School of Medicine, St Louis, MO, USA Search for more papers by this author Author Information Jennifer A Loughman1 and Michael G Caparon 1 1Department of Molecular Microbiology, Washington University School of Medicine, St Louis, MO, USA *Corresponding author. Department of Molecular Microbiology, Washington University School of Medicine, Campus Box 8230, 660 S. Euclid Ave., St Louis, MO 63110-1093, USA. Tel.: +1 314 362 1485; Fax: +1 314 362 3203; E-mail: [email protected] The EMBO Journal (2006)25:5414-5422https://doi.org/10.1038/sj.emboj.7601393 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Regulation of virulence factor expression is critical for pathogenic microorganisms that must sense and adapt to a dynamic host environment; yet, the signal transduction pathways that enable this process are generally poorly understood. Here, we identify LacD.1 as a global regulator of virulence factor expression in the versatile human pathogen, Streptococcus pyogenes. LacD.1 is derived from a class I tagatose-1,6-bisphosphate aldolase homologous to those involved in lactose and galactose metabolism in related prokaryotes. However, regulation of transcription by LacD.1 is not dependent on this enzymatic activity or the canonical catabolite repression pathway, but likely does require substrate recognition. Our results suggest that LacD.1 has been adapted as a metabolic sensor, and raise the possibility that regulation of gene expression by metabolic enzymes may be a novel mechanism by which Gram-positive bacteria, including S. pyogenes, coordinate multiple environmental cues, allowing essential transcription programs to be coupled with perceived nutritional status. Introduction In eukaryotes, metabolic enzymes of the glycolytic pathway have been adapted to function as regulators of gene transcription. Examples include lactate dehydrogenase, glyceraldehyde-3 phosphate dehydrogenase, hexokinase and enolase, each of which can be translocated into the nucleus to act as components of transcriptional regulatory complexes. Typically, these enzymes function as metabolic sensors to couple cellular metabolism to gene regulation (for a review see Kim and Dang, 2005). How these enzymes have evolved into regulators is not understood. Given that many metabolic pathways are evolutionarily ancient, comparisons to prokaryotic organisms would be highly informative. However, possible adaptation of prokaryotic metabolic enzymes for transcription regulation has not been well studied. This paucity of information is surprising given the intimate relationship between the metabolisms of prokaryotes and their environment. Many pathogenic bacteria couple the sensing of specific nutrients with the regulation of genes that encode virulence proteins important for the interaction with host cells. Often, these nutritional cues are used to distinguish different temporal stages of the infection process. For example, the Gram-positive pathogen Streptococcus pyogenes (group A streptococcus), the causative agent of numerous human diseases ranging from pharyngitis and impetigo to the often fatal necrotizing fasciitis and septicemia (for a review see Cunningham, 2000), expresses numerous virulence genes in a highly ordered temporal pattern (Virtaneva et al, 2005) that is influenced by nutrient availability (Chaussee et al, 1997; Podbielski and Leonard, 1998; Smoot et al, 2001). Temporal ordering is likely of critical importance to S. pyogenes pathogenesis, since numerous global transcription regulators that contribute to this process have been identified, and mutation of these typically results in large alterations of virulence (for a review see Kreikemeyer et al, 2003). Thus, understanding how the regulatory network interrogates metabolic cues will be important for understanding the contribution of gene regulation to pathogenesis. As a lactic acid bacterium, S. pyogenes relies exclusively on glycolysis for production of energy, suggesting that there are mechanisms to promote communication between metabolism and the virulence regulatory network. Many bacterial species monitor the activity of metabolic enzymes to control the hierarchical utilization of available sugars, a process known as carbon catabolite repression (CCR). In Bacillus subtilis, a sensory pathway that monitors the concentration of the glycolytic intermediate fructose-1,6-bisphosphate controls a transcription repressor known as CcpA. Greater than 80% of all genes regulated by glucose in B. subtilis are regulated by CcpA (Moreno et al, 2001). Although CcpA homologs have been identified in many Gram-positive bacteria, there are important differences in the mechanisms of CCR. Furthermore, the role of CcpA may be more limited in the global pathways of carbon source regulation by Gram-positive pathogens. For example, in Listeria monocytogenes, a homolog of CcpA mediates catabolite control of some genes, but not carbon source regulation of virulence genes (Behari and Youngman, 1998). In contrast, an analysis of a streptococcal pathogen, S. pneumoniae, has shown that while CcpA contributes to virulence, it is not a global regulator of CCR (Iyer et al, 2005). Thus, the pathways linking metabolism, CCR and virulence are not understood. Our approach to understand how virulence and metabolism are connected in S. pyogenes has been to focus on a single virulence gene known to be highly expressed in vivo, and whose expression is sensitive to multiple nutritional cues during growth in vitro. In this regard, the gene encoding the secreted SpeB cysteine protease (speB) has been a useful model. While its specific role in virulence is unclear, it has been shown that speB is highly expressed in tissue during infection (Graham et al, 2002; Cho and Caparon, 2005; Virtaneva et al, 2005; Loughman and Caparon, 2006). In addition, transcription of speB is responsive to a number of cues during in vitro culture that are linked to nutrient availability, including bacterial cell density, carbon source depletion, pH, chloride and growth phase (Chaussee et al, 1997; Podbielski and Leonard, 1998; Lyon et al, 2001; Smoot et al, 2001; Loughman and Caparon, 2006). Global profiling of the transcriptome in response to several of these cues revealed a large cohort of genes co-regulated with speB (Loughman and Caparon, 2006). Furthermore, the expression pattern of this group showed a significant correlation with that of speB in vivo (Loughman and Caparon, 2006). Thus, further analyses of the regulation of speB will likely reveal regulatory pathways with direct relevance to pathogenesis. The observation that there are a large number of genes that are coordinately regulated with speB, in response to multiple signals suggested that there is a common pathway for sensing and responding to these cues. Therefore, a search was conducted for a gene that would regulate speB in response to carbon source availability, pH and chloride. This analysis resulted in the identification of LacD.1, a putative class I tagatose-1,6-bisphosphate aldolase similar to enzymes involved in galactose metabolism in Gram-positive bacteria. These data suggest that like eukaryotes, prokaryotic organisms have also adapted enzymes of central metabolism to roles in gene transcription. Further analyses of LacD.1 will provide insight into the pathogenesis of Gram-positive bacterial infections, and the molecular mechanisms used to adapt classical metabolic enzymes for management of a broad range of regulatory programs. Results LacD.1 is a negative regulator of SpeB activity Known regulators of speB include RopB (also known as Rgg), a DNA-binding protein (Neely et al, 2003) that is an essential activator of speB transcription (Lyon et al, 1998; Chaussee et al, 1999). However, RopB does not appear to directly sense environmental cues (Neely et al, 2003; Loughman and Caparon, 2006). To search for additional regulatory factors, we screened for negative regulators of SpeB by examining colonies of a transposon mutant library on protease indicator media that had been buffered to pH 7.5. Since expression of speB requires a pH of less than 6.5 (Loughman and Caparon, 2006), any mutant expressing protease activity would arise from the inactivation of a negative regulator. Of approximately 1000 colonies analyzed, three expressed protease activity. These mutants were then tested on protease indicator plates containing 0.15 M NaCl. A single mutant with aberrant expression in response to both pH and NaCl was identified that contained a transposon insertion in an open reading frame (spy1704) that encodes a putative tagatose-1,6-bisphosphate aldolase annotated as LacD.1 (Figure 1A). In Gram-positive bacteria, LacD is part of the tagatose 6-phosphate pathway of lactose catabolism (van Rooijen et al, 1991; Rosey and Stewart, 1992). Figure 1.LacD.1 is a negative regulator of SpeB activity. (A) Organization of the LacD.1 locus is shown. Arrows indicate the positions of the open reading frames. Gene names and genetic loci are indicated above and below the arrows, respectively, and are based on the genome of strain SF370 (Ferretti et al, 2001). The position of a transposon insertion in lacD.1 is indicated by the triangle. (B) A montage of colonies patched onto protease indicator media is shown. Protease activity is apparent as a zone of clearance surrounding bacterial growth. As indicated at the left, media was unmodified (Unmod.), buffered (pH 7.5) or supplemented with 0.15 M NaCl (NaCl). As noted above the figure, the behavior of HSC5 (WT) and mutant JL151 (LacD.1−) is shown in the absence of plasmid (none); pJL154, containing lacD.1 (pLacD.1), or vector alone (vector). (C) Secreted protease activity for the indicated stains grown in various modified media is shown. Activity is reported as the percentage of WT activity in unmodified media. Data represent the mean and standard deviation of three independent experiments with samples analyzed in triplicate. An asterisk denotes a significant difference from WT/vector grown in the same medium (P<0.001). Download figure Download PowerPoint The importance of lacD.1 was confirmed by the analysis of a LacD.1− mutant (JL151) containing an in-frame deletion allele. This mutant recapitulated the original mutant phenotype as follows: in contrast to the wild-type strain (Figure 1B; WT/none), the LacD.1− mutant produced protease under modified conditions of pH and salt (Figure 1B; LacD.1−/none). Activity was derepressed under the modified conditions to levels equivalent to WT under optimal conditions (Figure 1C; compare WT to LacD.1−). Introduction of a plasmid containing lacD.1 complemented the mutant (Figure 1B; compare WT/vector to LacD.1−/pLacD.1) and restored sensitivity to the modified media (Figure 1C). Consistent with a role as a negative regulator, overexpression of LacD.1 in the WT resulted in significant repression of SpeB under normally permissive conditions (Figure 1B; compare WT/vector to WT/pLacD.1) to levels about 65% less than WT alone (Figure 1C; P<0.001). The more modest reduction of activity in the complemented mutant (approximately 30%) compared to WT (Figure 1C; LacD.1−/pLacD.1) is likely also due to overexpression of LacD.1. Taken together, these data establish that LacD.1 is a negative regulator of SpeB activity that participates in pathways responsive to both pH and NaCl. LacD.1 mediates coordinate regulation by pH and NaCl Analysis of transcription using real-time RT–PCR revealed that while the abundance of the speB transcript decreased in modified medium for WT, only a modest decrease was observed for the LacD.1− mutant when compared to optimal medium (Table I), indicating that the mutant phenotype is due to derepression of speB transcription. A selection of other messages co-regulated with speB (Loughman and Caparon, 2006) was also probed. This panel of genes are either up- or downregulated greater than two-fold by pH and/or NaCl in WT, but were all less than two-fold different under these conditions in the mutant (Table I). In contrast, one gene (ntpK) whose expression is altered in response to NaCl, but not pH in WT, has a wild-type expression pattern in the mutant (Table I). These data demonstrate that there are at least two NaCl-sensitive pathways and implicate LacD.1 as a regulator of a subset of genes that are responsive to both pH and NaCl. Table 1. LacD.1 mediates coordinate regulation by pH and NaCl in S. pyogenes Locusa Gene Putative function WTb LacD.1−b pH NaCl pH NaCl Spy2039 speB Cysteine protease −664.5±0.1 −567.7±0.1 −9.7±0.1 −4.1±0.1 Spy1738 manL Mannose PTS II 6.7±1.8 3.7±0.9 1.7±0.3 1.8±0.2 Spy0167 slo Streptolysin O precursor 3.8±1.0 5.1±2.6 1.4±0.2 1.1±0.2 Spy0745 sagH Streptolysin S biogenesis 2.50±0.6 2.8±0.3 1.5±0.1 1.2±0.2 Spy1915 salA Lantibiotic precursor −6.9±0.1 −2.8±0.3 −1.7±0.1 −1.4±0.1 Spy1815 scrA Sucrose PTS II −5.1±0.1 −2.0±0.3 −1.2±0.1 1.3±0.1 Spy0149 ntpK Sodium ATPase 1.0±0.5 6.4±3.1 −1.5±0.1 4.0±0.1 a Genomic loci, gene names, and putative function are based on current annotation of the genome of S. pyogenes SF370 (Genbank AE004092). b Fold-change in transcript abundance for cultures grown to early stationary phase in buffered (pH) or salt supplemented-media (NaCl) versus cultures grown in unmodified medium as determined by real-time RT–PCR. HSC5 and JL151 are WT and LacD.1−, respectively. Data represent the mean and standard deviation of three independent experiments analyzed in triplicate. LacD.1 contributes to CCR The involvement of LacD.1 in global regulatory pathways that coordinate virulence factor expression and carbohydrate metabolism (Loughman and Caparon, 2006) suggested that it might also mediate carbon source-specific regulation. To test the possible involvement of LacD.1 in CCR, the effect of glucose on speB transcription was tested. For WT, glucose repressed speB transcription greater than 50-fold (Figure 2A; speB, WT) similar to pH and NaCl (Figure 2B), but failed to repress transcription in the LacD.1− mutant (Figure 2A), whose expression levels were similar to those observed with pH and NaCl (Figure 2B). Figure 2.Carbohydrates influence protease regulation through CcpA dependent and independent pathways. (A) Transcript abundance of speB and ropB for bacteria grown in medium supplemented with 0.15 M glucose relative to unmodified medium is shown, as determined by real-time RT–PCR. Compared are HSC5 (WT), JL151 (LacD.1−), and JL245 (CcpA−). Data represent the mean and standard deviation of three independent experiments analyzed in triplicate. (B) A montage of colonies patched onto protease indicator medium plates is shown. Strains are as described for (A), and media are as described in Figure 1. (C) Quantitative protease activity produced by the indicated strains following growth in the various carbohydrate-supplemented media is shown. Activity is reported as the percentage of activity for each strain in unmodified medium. Data represent the mean and standard deviation of three independent experiments, with samples analyzed in triplicate. An asterisk denotes a significant difference from WT grown in the same medium (P<0.001). Abbreviations: Gro, glycerol; Glc, glucose; Lac, lactose; Gal, galactose; Fru, fructose; Mal, maltose; Man, mannose; and Scr, sucrose. Download figure Download PowerPoint LacD.1 can function independently of CcpA Our inspection of the known speB regulatory region (Neely et al, 2003) revealed the presence of a candidate binding site for CcpA about 600 bases upstream of the speB start codon (data not shown). This raised the question as to whether LacD.1 pathway depends on CcpA. When compared, glucose also did not repress speB transcription in the CcpA− mutant (Figure 2A), and did not repress protease activity on indicator plates (Figure 2B; CcpA−/Glc). Expression of RopB was unaffected by the addition of glucose to the medium (Figure 2A; ropB) and there was also no obvious difference in growth between WT and the two mutants (data not shown). However, unlike the LacD.1− mutant, the CcpA− mutant retained its ability to respond to both pH and NaCl (Figure 2B), suggesting that the pathways may function independently. To further explore this, protease expression was compared in response to a panel of carbohydrates. While the CcpA− mutant retained the ability to repress speB in response to several sugars (e.g. galactose, fructose; Figure 2C), the LacD.1− mutant was insensitive to all of the sugars tested (Figure 2C). These data suggest (i) that both CcpA and LacD.1 contribute to carbon source regulation; (ii) that LacD.1, but not CcpA, has a role in gene regulation in response to multiple signals; and (iii) that the LacD.1 pathway can function in the absence of CcpA, indicating that the regulators may influence gene expression through independent pathways. Integration of LacD.1 and RopB circuits contributes to temporal regulation Ectopic expression of ropB earlier in the growth cycle does not alter the temporal kinetics of speB expression (Neely et al, 2003). This suggests that a negative regulator, like LacD.1, restricts RopB activation of speB transcription until the appropriate signals are detected. This model predicts that constitutive expression of ropB in the absence of LacD.1 should permit protease production early in the growth cycle. To test this, a plasmid containing a constitutively expressed allele of ropB was introduced into the LacD.1− mutant. This now allowed significant protease production at early time points (2–4 h), when activity is normally undetectable in WT (Figure 3; compare WT to LacD.1− (pRopB), P<0.03). This level of activity was about 50% of that observed at later time points, indicating that multiple factors are involved in regulation. Nevertheless, these data support a model for temporal regulation that requires activation by RopB in the context of derepression by LacD.1. Also, lacD.1 itself is not regulated by growth phase (data not shown), suggesting that it must process a signal for derepression to occur. Figure 3.Temporal regulation requires cooperation of RopB and LacD.1. (A) Secreted protease activity of the indicated strains as a percentage of WT activity in 8 h supernatant fluids is shown. Compared are HSC5 (WT), and JL151 (LacD.1−) in the presence or absence of a plasmid, pJL77, expressing ropB (pRopB). E64 is an inhibitor of SpeB cysteine protease activity. Data represent the mean and standard deviation for three independent experiments, with samples analyzed in triplicate. (B) RopB and LacD.1 form a complex. Shown is an immunoblot developed with an αFLAG antibody of untreated lysates (L) or lysates immunoprecipitated with an αHA antiserum (IP) from strains expressing various combinations of RopB-HA from a plasmid (pJL77) and LacD.1-FLAG from the chromosome, as indicated at the top of the figure. These strains included JL320 (LacR.1−), JL418 (LacR.1−, LacD.1-FLAG+), JL423 (LacR.1−, RopB-HA+) and JL425 (LacR.1−, LacD.1-FLAG+, RopB-HA+). The band corresponding to LacD.1-FLAG is indicated with a shaded arrow and a band corresponding to an unknown streptococcal protein that crossreacts with the αFLAG antibody, but does not co-immnoprecipitate with RopB, is indicated with an open arrow. Molecular mass of selected standards is shown on the left and is in kilodaltons. Download figure Download PowerPoint To evaluate possible protein–protein interactions involving RopB and LacD.1, a co-immunoprecipitation strategy was employed using a strain (JL320) that overexpresses lacD.1 due to the deletion of the gene encoding the LacR.1 repressor (Loughman and Caparon, submitted). This strain represses SpeB expression to an even greater extent than that obtained by ectopic overexpression of LacD.1 in WT, which should promote any RopB–LacD.1 interactions that contribute to repression. The chromosomal lacD.1 locus in JL320 was further modified to express LacD.1 with a C-terminal FLAG tag, which was readily detectable in immunoblots of whole-cell lysates (Figure 3B, lane 3). Introduction of a plasmid expressing RopB-HA, followed by treatment with an αHA antiserum, resulted in the co-immunoprecipitation of LacD.1-FLAG (Figure 3B, lane 8). Co-immunoprecipitation of LacD.1-FLAG was specific since it was not observed in the absence of RopB-HA (Figure 3B; lanes 2, 4 and 6) and because RopB-HA did not co-immunoprecipitate a streptococcal protein that fortuitously cross-reacted with the αFLAG antibody (Figure 3B, lanes 6 and 8). In the reciprocal experiment, LacD.1-FLAG co-immunoprecipitated RopB-HA using an αFLAG antibody (data not shown). These data indicate that under conditions promoting repression of SpeB expression, RopB and LacD.1 can be found in an interacting complex. LacD.1 is a unique regulatory aldolase The tagatose 6-phosphate pathway has been best characterized in Staphalococcus aureus, Streptococcus mutans and Lactococcus lacits, where the enzymes are encoded in a seven gene operon, lacABCDEFG (Rosey et al, 1991; van Rooijen et al, 1991; Rosey and Stewart, 1992). However, the Lac.1 locus deviates considerably from this structure. It lacks lacG, encoding phospho-β-galactosidase, and the gene that should encode LacC.1 is truncated and contains several frameshift mutations. This structure is also conserved in all available S. pyogenes sequenced genomes. However, there is a second Lac locus in S. pyogenes (Lac.2) that closely resembles the canonical operon, encoding an aldolase, LacD.2, which is highly homologous (73% identical, 86% similar) to LacD.1 at the amino-acid level (Figure 4A). To evaluate whether LacD.1 was unique, a LacD.2− mutant was analyzed and was found to possess a WT pattern of expression, repressing SpeB in response to pH and NaCl (Figure 4B; compare WT/none to LacD.2−/none) indistinguishable from WT (Figure 4C). Furthermore, LacD.2 could neither complement a LacD.1− mutant (Figure 4B; compare LacD.1−/pLacD.2 to LacD.1−/pLacD.1) nor repress SpeB expression in WT when overexpressed (Figure 4B; compare WT/pLacD.2 to WT/pLacD.1). The LacD proteins were expressed to equivalent levels in both the WT and LacD.1− strain as assessed by Western blotting of whole-cell extracts (data not shown). Thus, while LacD.1 and LacD.2 are closely related, only LacD.1 can function in SpeB regulation. Figure 4.LacD.1 is distinct from LacD.2. (A) Organization of the LacD.2 locus. Arrows indicate the positions of the open reading frames. Gene names and genetic loci and are indicated above and below the figure, respectively, and are based on strain SF370 (Ferretti et al, 2001). (B) A montage of colonies patched onto protease indicator media is shown, as described for Figure 1. Strains analyzed include: HSC5 (WT), JL151 (LacD.1−), and JL251 (LacD.2−) in the absence of plasmid (none), pJL154 expressing lacD.1 (pLacD.1), or pJL156, expressing lacD.2 (pLacD.2). (C) Secreted protease activity in supernatant fluids derived from the indicated strains is shown. Media are as described for Figure 1 and the strains as described for (B). Activity is reported as the percentage versus the WT strain grown in unmodified medium. Data represent the mean of at least three independent experiments, with samples analyzed in triplicate. An asterisk denotes a significant difference from WT/none grown in the same medium (P 98% decrease in aldolase activity (Table II; LacD.1K204A). However, LacD.1K204A retained an ability to complement the regulatory defect of the LacD.1− mutant (Figure 5A; LacD.1−: compare LacD.1 allele/K204A to LacD.1 allele/WT) and repress SpeB activity in WT (Figure 5A; WT: compare LacD.1 allele/K204A to LacD.1 allele/WT). There were no significant differences in SpeB expression between LacD.1 and LacD.1K204A expressed in WT and in LacD.1− backgrounds, respectively (Figure 5C). Thus, aldolase activity is dispensable for LacD.1 regulatory function. Figure 5.Mutation of catalytic and non-catalytic residues uncouples aldolase activity from regulatory activity. (A) A montage of colonies patched onto protease indicator media is shown, as described for Figure 1. Strains include HSC5 (WT) and JL151 (LacD.1−), containing pJL154 expressing lacD.1 (WT), derivatives expressing the indicated mutant proteins (LacD.1 allele), or the vector alone (vector). Media are as described for Figure 1. (B) Secreted protease activity from JL151 (LacD.1−) expressing the various plasmids described in (A) is shown. Media are as described for Figure 1. (C) The LacD.1 overexpression phenotype in HSC5 (WT) is shown in the presence of vector or plasmids expressing unmodified (WT) or mutant LacD.1 proteins, as described for (A). For (B, C), activity is reported as the percentage of activity obtained for WT grown in unmodified medium. Data represent the mean of three independent experiments, with samples analyzed in triplicate. An asterisk denotes a significant difference from LacD.1-/vector (B) or WT/vector (C) grown in the same medium (P<0.05). Download figure Download PowerPoint Table 2. Summary of protein enzymatic and regulatory functions Aldolasea Mutationa Aldolase activityb Regulation competentc Putative functiond LacD.2 None 2.370±0.39 − NA LacD.1 None 2.638±0.44 + NA K204A (K229) 0.043±0.03 + Forms Schiff-base with C2 K124A (K146) 0.095±0.04 − H-bonds with C4 OH− E162A (E187) ND − H-bonds with C4 OH−, H+ acceptor A249F (G272) ND − H-bonds with C1 PO4− G276F (G302) ND −/+ H-bonds with C1 PO4− R277A (R303) ND + H-bonds with C1 PO4− a The indicated aldolase contained no mutations (none) or was modified at specific positions as indicated. For LacD.1, the identities of the corresponding residues in AldoA (St-Jean et al, 2005) are indicated by parentheses. b Activity of purified protein was determined by assay of fructose-1,6-bisphosphate cleavage as described in Materials and methods. Data are presented in units/mg and represent the mean of at least three independent experiments analyzed in triplicate. Activity of AldoA in the assay was 11.59±1.92 U/mg. ND, not determined. c Competence (+) indicates that regulatory activity was not significantly different from WT; (−), that activity was not significantly different from vector alone; (−/+), that activity was significantly different from both WT and vector alone (P<0.05), as assessed by pH and NaCl regulation upon expression in a LacD.1− mutant, and overexpression in WT. d Function of the indicated residue in LacD.1 is predicted as noted based on biochemical and structural analyses of AldoA (St-Jean et al, 2005). NA, not applicable. Substrate binding is implicated for regulatory function The active site of class I aldolases is highly conserved across kingdoms (Lorentzen et al, 2004). Further analysis of LacD.1 using structural data of other class I aldolases revealed several conserved residues likely to interact with the C4 hydroxyl and C1 phosphate of the bisphosphate substrate (see Table II). Mutation of C4 hydroxyl-interacting residues (Table II; K124A, E162A) produced LacD.1 mutants that were completely defective in ability to complement a LacD.1− mutant (Figure 5A; LacD.1−: compare LacD.1 allele/K124A and E162A to vector), expressed protease activities identical to the mutant alone (Figure 5B), and could not repress activity when overexpressed in WT (Figure 5C). Based on a prior study (St-Jean et al, 2005), mutation of C1 phosphate-interacting residues utilized replacement with a bulky residue (Table II). One of these (R277A) was fully functional in regulation (Table II). However, the other two were either partially (G276F) or completely (A249F) defective for complementation (Figure 5A), protease activity (Figure 5B) or r
Referência(s)