| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, UK
2 School of Biology, Georgia Institute of Technology, 310 Ferst Drive, Atlanta, GA 30332-0230, USA
Correspondence
Stephen Spiro
stephen.spiro{at}biology.gatech.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Present address: John Innes Centre, Colney, Norwich NR4 7UH, UK.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Zumft, 1997). A recurring theme amongst denitrifying organisms is the regulation of expression of the genes encoding the nitrite and NO reductases by an NO-responsive transcription factor, either an FNR/CRP family member or a sigma-54-dependent enhancer binding protein (Zumft, 2002). In Paracoccus denitrificans, the FNR/CRP family member NNR (nitrite reductase and nitric oxide reductase regulator) activates transcription of the nir and nor operons, which encode the respiratory nitrite and NO reductases, respectively (Van Spanning et al., 1995, 1997). A number of lines of evidence indicate that NNR is responsive in vivo to NO and can be activated by physiological (e.g. nitrate and nitrite) and non-physiological (e.g. nitroprusside) sources of NO and NO+ (Hutchings et al., 2000; Van Spanning et al., 1999). Thus, the enzymes that synthesize and consume NO in the denitrification pathway are co-ordinately activated by NO, and this is thought to be a strategy to prevent the accumulation of toxic concentrations of NO (Baker et al., 1998; Zumft, 2002). Proteins related to NNR that are similarly NO-responsive have been characterized: NnrR from Rhodobacter sphaeroides (Kwiatkowski & Shapleigh, 1996); DnrD from Pseudomonas stutzeri (Vollack et al., 1999); and DNR from Pseudomonas aeruginosa (Hasegawa et al., 1998). NnrR from Bradyrhizobium japonicum may also be a sensor of NO (Mesa et al., 2003). However, these NO-responsive regulators do not fall into a single well-defined clade within the wider FNR/CRP family (Korner et al., 2003; Mesa et al., 2003; Vollack et al., 1999), and alignments of the sequences of these proteins reveal very few conserved residues that might point towards a mechanism of signal recognition. The two groupings of NNR-like proteins correspond to the proteins that activate expression of genes encoding a cytochrome cd1 or a copper-containing nitrite reductase (Mesa et al., 2003). It is not known whether these phylogenetic groupings reflect mechanistic differences, and evidence for the biochemical basis of activation of NNR and related proteins by NO has remained elusive. Site-directed mutagenesis of the genes encoding NNR and NnrR (from R. sphaeroides) has implicated several residues in their function, but in no case has a mechanistic explanation been forthcoming (Hutchings et al., 2000; Laratta & Shapleigh, 2003).
We have previously shown that NNR is active and signal-responsive in Escherichia coli (Hutchings et al., 2000), an organism that does not have a denitrification pathway but that can reduce nitrate to ammonia during anaerobic growth. In this heterologous system, NNR activity can be activated by nitroprusside, a convenient source of NO+, or by nitrate and nitrite, which very likely act as sources of NO produced as a by-product of respiratory metabolism (Hutchings et al., 2000). Others have also shown that NO is produced during nitrate and nitrite respiration in E. coli, and it has recently been suggested that the respiratory nitrite reductase Nrf is the major source (Corker & Poole, 2003; Ji & Hollocher, 1988; Van Doorslaer et al., 2003). Activation of NNR in the heterologous background is most easily explained by a model in which NO, or a related molecule, interacts directly with the protein, since the participation of other proteins in the signalling mechanism would require those proteins to be conserved in E. coli. NNR activity cannot be detected in aerobic cultures of E. coli (Hutchings et al., 2000), which makes good physiological sense, given that enzymes of the denitrification pathway are only required and expressed in anaerobically growing cultures (Baker et al., 1998). The absence of NNR activity in an aerobic culture suggests either that the protein is directly inactivated by oxygen (or some other signal of aerobic metabolism), or that the molecule that activates NNR is absent or unstable in aerobically growing cells. In this paper, we show that NNR is rapidly inactivated following a shift to aerobic growth conditions, in a way that is consistent with direct inactivation of the protein. Further, we report the isolation of mutant nnr alleles that encode NNR proteins that have significant activity in anaerobic cultures in the absence of NO, but remain inactive in aerobic cultures. The properties of these proteins also demonstrate that NNR activity is inhibited in vivo by a signal of aerobic metabolism, and therefore that NNR is a dual sensor of both oxygen and NO. A haem-based mechanism would provide a means for NNR to sense both NO and oxygen, and we provide preliminary evidence to suggest that NNR activity requires haem.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
lacX74 galU galK rpsL (ara–leu) (tyrR–fnr–rac–trg)] was used as the host for assays of NNR activity (Hutchings et al., 2000). For experiments using a single-copy NNR reporter fusion, JRG1728 was lysogenized with a lambda phage (a gift from David Lee, University of Birmingham, UK) carrying a consensus FNR binding site (FF), located 41·5 bp upstream of the transcription start site of the melR promoter fused to lacZ (Li et al., 1998). Strain DH10B [mcrA (mrr hsdRMS mcrBC) 
80dlacZM15 lacX74 deoR recA endA araD (ara leu) galU galK rpsL nupG] was used for routine DNA manipulations, and XL1-Red [endA gyrA thi hsdR supE relA lac mutD mutS mutT : : Tn10] from Stratagene was used for random mutagenesis. H500 (hemA : : kan) was a gift from K. Ito (University of Kyoto, Japan) and C. Wandersman (University of Paris, France). The hemA : : kan mutation was transferred to JRG1728 by transduction with bacteriophage P1. The plasmids used were pRW2A/FF (from Steve Busby), containing the multiple-copy FF–melR reporter fusion; pBAD24 (Guzman et al., 1995) for controlled expression of nnr; and pNNR, a pUC18 derivative which expresses nnr from the lac promoter (Hutchings et al., 2000). The rich medium for bacterial growth was L broth (10 g tryptone l–1, 5 g yeast extract l–1 and 5 g NaCl l–1), supplemented with 0·5 % (w/v) glucose for anaerobic growth. The minimal medium was either M9 (Miller, 1992) or a mineral salts medium (Spencer & Guest, 1973), supplemented with glucose, maltose and arabinose, as indicated. Eosin methylene blue (EMB) lactose medium (Miller, 1992) was used to screen for nnr alleles with altered phenotypes. Anaerobic cultures were grown in filled standing bottles, conditions which for E. coli are physiologically equivalent to strict anoxia (Constantinidou et al., 2006). NO was added to cultures from a saturated solution (2 mM) using a gas-tight syringe.
DNA manipulations.
For cloning into pBAD24, nnr and its mutant alleles were amplified from pNNR (or the corresponding derivative) by PCR, with a 5' primer that introduced a KpnI site immediately downstream of the ATG codon of nnr. The genes were cloned into the KpnI and XbaI sites of pBAD24 (Guzman et al., 1995). The KpnI site is immediately downstream of an ATG codon in the vector, so this procedure introduced valine and proline codons immediately after the initiating codon of nnr. Since the genes are active in vivo, there is no reason to believe that this change influences NNR activity.
Mutagenesis of the nnr gene.
The cloned nnr gene on pNNR was transformed into E. coli XL1-Red. Transformants were pooled, inoculated into L broth and grown overnight. The culture was inoculated into fresh L broth and grown overnight again and this procedure was repeated for a third culture. Then plasmid DNA was prepared, and the insert containing nnr was purified and subcloned into pUC18. The resulting plasmid DNA was transformed into JRG1728 containing the chromosomal FF–melR–lacZ fusion and colonies were screened for gain-of-function phenotypes on EMB plates incubated aerobically. Loss-of-function nnr alleles were isolated on EMB plates supplemented with 50 mM nitrate. Site-directed mutagenesis of pNNR was by a PCR-based method, as previously described by Hutchings et al. (2000).
Analytical procedures.
-Galactosidase was assayed according to the method of Miller (1992). Western blotting with a polyclonal anti-NNR antiserum was carried out as previously described by Hutchings et al. (2000).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). In an E. coli reporter system, NNR could likewise be activated by nitroprusside, and by nitrate and nitrite, which very likely act as sources of traces of NO made as a by-product of respiratory metabolism (Corker & Poole, 2003; Hutchings et al., 2000). We were interested to establish the range of compounds that might activate NNR in vivo, so have tested aqueous NO, along with S-nitrosoglutathione (GSNO), which is a potential source of NO as well as being a major sink for NO in vivo. Apparent NNR activity was assayed in a strain expressing nnr from a multi-copy plasmid and with an NNR-regulated promoter (FF–melR) fused to lacZ on a lambda prophage. In this system, -galactosidase activity provides an indirect measure of transcription activation by NNR. Cultures were grown anaerobically in defined media to exponential phase and then exposed to inducer for 2 h before assaying -galactosidase. Under these conditions, 200 µM nitrite was an efficient inducer of NNR activity (909 units of -galactosidase, compared to <20 units in control cultures not exposed to any inducer). High levels of NNR activity were also seen when 100 µM nitroprusside (2440 units), 50 mM nitrate (427 units), 50 µM GSNO (2857 units), or 1 µM aqueous NO (911 units) were used as inducers.
NNR activation does not require de novo protein synthesis
We were interested to explore whether NNR synthesized under non-activating conditions can be activated by subsequent exposure to NO, in other words, whether activation requires de novo synthesis of NNR polypeptide. We cloned the nnr gene into pBAD24, which allows activation of nnr expression by addition of arabinose to growth media, and repression by addition of glucose (Guzman et al., 1995). When the nnr clone in pBAD24 was introduced into the reporter strain, we found that expression of -galactosidase became dependent on arabinose, in an anaerobic culture supplemented with an inducer of NNR (Fig. 1a). Western blots confirmed that NNR abundance in the cell was regulated by arabinose in these experiments, and was below the detection limit in cultures amended with glucose (data not shown). We then grew the reporter strain in an aerobic culture containing arabinose, conditions which allow the synthesis of NNR in an inactive form. In mid-exponential phase, cultures were washed thoroughly, then resuspended in medium containing glucose (to repress further nnr expression) and nitrate as an inducer of NNR activity. Following the switch in growth conditions, NNR was rapidly activated, as indicated by the immediate synthesis of -galactosidase (Fig. 1b). The activity of the lacZ reporter levelled off after approximately one doubling of the culture, presumably because active NNR was diluted out of the cells as a consequence of growth. Accordingly, in the presence of arabinose (to allow continued nnr expression), -galactosidase accumulated at the same rate following the shift in growth conditions, but did not show the same levelling off (Fig. 1b). The results of this experiment clearly demonstrate that inactive NNR can be activated, with no requirement for new protein synthesis.
|
1 h lag before any loss of NNR activity was detectable and thereafter, NNR activity declined at a rate close to that predicted, assuming inactivation of the protein and loss of -galactosidase activity by dilution in the growing culture (Fig. 2a). Thus, removal of the activating signal (in this case nitrate) did not lead to instantaneous and rapid inactivation of NNR. The loss of NNR activity in the later part of the experiment presumably reflected turnover of the activating signal, since inclusion of nitrate in the medium for this experiment allowed NNR activity to continue to increase throughout the time period (not shown).
|
-galactosidase activity declined rapidly at a rate close to that predicted, assuming immediate inactivation of NNR and dilution of -galactosidase as a consequence of cell growth (Fig. 2b
). Thus, it seems clear that NNR was inactivated following a shift to aerobic growth conditions, which can be explained either by direct inactivation of NNR by oxygen, or by rapid removal of the activating signal in the presence of oxygen. The results of the aerobic-to-anaerobic shift experiment also imply that the inactivation of NNR is fully reversible.
Comparing the results of the two experiments, it is clear that NNR is inactivated rapidly following a transition to aerobic conditions, but not under anaerobic conditions in the absence of nitrate. E. coli is capable of metabolizing NO under both aerobic and anaerobic conditions, using its flavohaemoglobin and flavorubredoxin, respectively (Gardner & Gardner, 2002). E. coli cells grown anaerobically and exposed to NO consume NO at similar rates in the presence and absence of oxygen (Gardner & Gardner, 2002). Thus, the difference in the apparent rates of inactivation of NNR under aerobic and anaerobic conditions is unlikely to reflect a difference in the rates of NO turnover.
Isolation of nnr alleles encoding gain-of-function proteins
The only residue in the signalling domain that is completely conserved amongst NNR and its presumed orthologues is Phe-82. We have previously shown this residue to be essential for normal NNR activity (Hutchings et al., 2000). Our attention has therefore focussed on this region of the NNR protein and we have begun to substitute residues close to Phe-82 in the primary structure. The first substitution made was of Arg-80 by alanine, with surprising consequences. The R80A variant of NNR showed significantly increased activity in aerobic cultures and greatly increased activity in anaerobic cultures in the absence of sources of NO (Table 1). Indeed, this protein showed a high level of activity in unamended minimal medium in anaerobic cultures, conditions under which the wild-type protein was almost completely inactive (Table 1). Thus, the R80A variant of NNR had become independent of the activating signal (NO) and, by analogy with other FNR/CRP family members (see below), we designate this phenotype NNR*. In aerobic cultures, NNR R80A (when expressed from the lac promoter in pUC18) had an activity >20-fold higher than the basal level of activity seen with the wild-type protein, but nevertheless significantly lower than the activity seen in anaerobic cultures (Table 1).
|
-galactosidase activity were sought by screening on EMB and X-Gal indicator media. We were able to isolate loss-of-function mutants (some of which are described below) from this screening at a frequency of 0·4 %, confirming the effectiveness of the mutagenesis. From a total of 4000 colonies, one was isolated with a slightly more intense Lac+ phenotype on an aerobically incubated EMB indicator plate, and a fully Lac+ phenotype on an anaerobically incubated X-Gal indicator plate (with a subsequent incubation in the presence of oxygen to allow development of the blue pigment). Sequence analysis of this clone indicated that it too had a mutation in the codon corresponding to Arg-80, which was substituted with histidine in the encoded protein. The R80H protein had a qualitatively similar NNR* phenotype to that of NNR R80A (Table 1). Like NNR R80A, the R80H protein was essentially independent of NO for activation in anaerobic cultures, and had significant activity in aerobic cultures. Western blotting (not shown) confirmed that NNR* proteins were present in the cell at a similar abundance to that of the wild-type protein when expressed from the same vector. Thus the phenotype cannot be explained by (and is not consistent with) a non-specific stabilization of the protein.
Effect of NNR abundance on activity
Partially active variants of transcription factors like NNR can show artefactually high activities in in vivo assays dependent on reporter fusions, especially if the protein is overexpressed (Kerby et al., 2003). The reason is, presumably, that an overexpressed protein with partial activity (i.e. where the equilibrium is shifted towards the inactive species) may nevertheless fully occupy a single-copy promoter, and therefore drive high levels of transcription. This effect has been observed for other FNR/CRP family members, including FNR itself and CooA (Kerby et al., 2003; Moore & Kiley, 2001). Similarly, the significantly increased activity of NNR* proteins (under aerobic conditions, and anaerobic conditions in the absence of NO) may reflect only a small increase in specific activity. The ideal solution is to assay DNA binding in vitro, but we have not so far been able to identify suitable conditions for detecting activity in purified NNR. Another approach is to measure activity in vivo, under conditions where the transcription factor is expressed at a low level and so is limiting for promoter activity (Kerby et al., 2003; Moore & Kiley, 2001). We therefore utilized regulated expression in pBAD24 to further characterize the activity of the R80A and R80H variants of NNR. NNR activity was measured in cultures grown in minimal medium containing maltose as the carbon and energy source (to avoid catabolite repression of the araBAD promoter) and supplemented with 0·2 % arabinose. At this concentration of arabinose, NNR activity was submaximal (Fig. 1a), presumably because the protein abundance in the cell was limiting for promoter occupancy. Under these assay conditions, wild-type NNR showed a smaller response to nitroprusside and no response to nitrate (Table 1), presumably reflecting the lower abundance of the protein in the cell and the fact that nitrate respiration generates very low concentrations of NO. The NNR* proteins still had significant activity under anaerobic conditions in the absence of NO, but had very low activity in aerobic cultures (Table 1). This result implies that the NO-independent activity of the star proteins did not arise from a small increase in specific activity that was magnified by overexpression from the lac promoter. When the star proteins were expressed at levels limiting for gene expression, they had the same activity as the wild-type protein in aerobic cultures, but sevenfold (R80A) and 16-fold (R80H) higher activity (compared to the wild-type protein) in cultures grown anaerobically in the absence of an activating signal. Thus, the mutation has a specific effect on the NO signalling mechanism, rather than a general effect on NNR activity that would be manifested under both aerobic and anaerobic conditions (for example by increasing the strength of the interaction with RNA polymerase).
Structural model of NNR
To gain further insight into the possible role of Arg-80, we modelled the structure of NNR, using the known structures of CRP (Fig. 3) and CooA (not shown) as the templates. In both cases, Arg-80 was predicted to be in the signalling domain of NNR, lying on the surface of a cleft between this domain and the long C-helix that forms the interface between monomers (Fig. 3). Comparison of the two models (active and inactive conformations) suggests that there may be conformational changes in both the C-helix and the region of the protein surrounding Arg-80. His-134 in the C-helix is the residue closest to Arg-80 in both models. Interestingly, in alignments of all of the known NO-sensing orthologues of NNR, arginine is uniquely found at position 80 (NNR numbering) in NNR, and His-134 is substituted by arginine in every other protein (Fig. 3). Thus, in NNR, Arg-80 and His-134 are predicted to be physically close in the tertiary structure, and these residues co-vary in NNR compared to its relatives. This is suggestive of a mechanistic relationship between Arg-80 and His-134. We tested the importance of His-134 by substituting it with arginine and found that the altered protein was inactive (data not shown). Thus, the substitution of histidine for arginine at this position of the C-helix of NNR orthologues may be mechanistically significant.
|
The fact that NNR is active in E. coli implies a productive interaction with the heterologous E. coli RNA polymerase. That interaction is presumably equivalent to the well-characterized interactions between the E. coli FNR and CRP proteins and RNA polymerase, which rely on surface-exposed activating regions (ARs) on the transcriptional activators (Busby & Ebright, 1999). We have previously speculated that activation by NNR at Class II promoters (where the activator binding site is centred 41·5 bp upstream of the transcription start site) requires an AR equivalent to the AR3 of CRP and FNR (Busby & Ebright, 1999; Hutchings et al., 2000; Lamberg et al., 2002). AR3 in CRP is a surface loop and its activity requires Glu-58 (Busby & Ebright, 1999); the conserved Glu-87 in FNR also has a role in AR3 activity (Lamberg et al., 2002). To test whether NNR activity in E. coli required a functional AR3, we substituted the corresponding glutamate (at position 70) with alanine and found that the altered protein was inactive in our in vivo assay (data not shown). The phenotype is consistent with a role for Glu-70 of NNR as an AR3 residue.
Haem-dependence of NNR activity
The pattern of activity of NNR is superficially similar to the haem-based sensor CooA, which is activated by carbon monoxide (CO) and is inactive in aerobic cultures (Thorsteinsson et al., 2001). A haem-based mechanism would provide a plausible explanation for the observation that NNR can be activated by NO and apparently inactivated by oxygen. To test this idea, we transduced a hemA mutation into our reporter strain and assayed NNR activity in cultures grown in the absence and presence of exogenous 5-aminolaevulinic acid (ALA), which restores haem biosynthesis to hemA strains. In the absence of ALA, NNR was significantly less responsive to nitrite, nitroprusside and aqueous NO, but high levels of activity were restored by addition of ALA to the growth medium (Table 2). Cultures continued to grow in the 2 h following addition of the inducing compound, indicating that loss of NNR activity was not a simple consequence of sensitivity of the hemA mutant to the inducing compound. Loss of the response to nitrite was to be expected in the hemA mutant, given that the generation of NO from nitrite probably requires the activity of haem-containing enzymes. However, the loss of the response to NO and nitroprusside can only be explained in terms of a direct or indirect requirement of NNR for haem. NNR surprisingly showed high activity in cultures of the hemA mutant grown without ALA and in the presence of nitrate (Table 2). These cultures were capable of reducing nitrate to nitrite (not shown), which is a haem-dependent pathway. There is evidence for a hemA-independent mechanism for haem biosynthesis in E. coli and Salmonella typhimurium (Elliott & Roth, 1989; Jahn et al., 1991), and we speculate that anaerobic growth on nitrate may stimulate the activity of this second pathway.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Here, we have shown that NNR can be activated in anaerobic cultures with micromolar concentrations of aqueous NO, metabolic and chemical sources of nitric oxide (nitrate, nitrite and GSNO), and the nitrosylating agent sodium nitroprusside (SNP). This spectrum of compounds that can activate NNR in vivo may provide some clues as to the activation mechanism. NO per se is not a nitrosylating agent and so cannot, for example, nitrosylate thiols unless an oxidizing agent (oxygen or a transition metal ion) is present (Poole & Hughes, 2000). NO can however react with haem and non-haem iron centres, and with Fe–S clusters. GSNO is transported into the cell through a peptide permease, whereupon it can modify intracellular targets, either by rather slow homolytic decomposition to NO or by the heterolytic transfer of NO+ (De Groote et al., 1995). GSNO reacts with haemoglobin to form nitrosyl haemoglobin, in which NO is bound to the iron of the ferrous haem (Spencer et al., 2000). Nitroprusside does not directly release NO (except through photolytic decomposition), but is an effective nitrosylating agent (Poole & Hughes, 2000), and is a potent activator of soluble guanylate cyclase, a haem-based NO sensor (Martin et al., 2003). The fact that NO activates NNR in the absence of oxygen argues against thiol nitrosylation as the activating mechanism, and the sole cysteine of NNR is partially dispensable for activity (Hutchings et al., 2000). The dispensability of the cysteine residue also argues against modification of an Fe–S cluster as the activation mechanism, and we have shown that NNR retains activity in an icsS mutant that has a defect in Fe–S cluster assembly (N. Shearer & S. Spiro, unpublished observations). All of the compounds that activate NNR are consistent with a haem-based mechanism for NO sensing. The partial loss of NNR activity in a hemA mutant is further evidence for a haem-based sensing mechanism. Biochemical efforts to demonstrate that purified NNR contains or can bind haem have so far proved unsuccessful, so we cannot exclude the possibility that the requirement for haem is indirect. In other words, there may be a haem-containing sensor protein (analogous to the eukaryotic soluble guanylate cyclase) that generates a secondary signal that activates NNR, although this model cannot easily be reconciled with the phenotypes of NNR* proteins. It has been reported that the NO-sensing DnrD protein of Ps. stutzeri purifies with a bound haem, but a specific function cannot be attributed to the haem (Vollack & Zumft, 2001). Cultures of P. denitrificans which have been depleted for haem (by treatment with the ALA analogue laevulinic acid) synthesize the apo-form of the periplasmic cytochrome cd1 nitrite reductase (Page & Ferguson, 1990). Thus, the NNR-activated nirS gene can continue to be transcribed in cells that are depleted for haem. One possible explanation is that the imposed haem limitation (which allows growth to continue for at least one generation) does not deplete haem sufficiently to abolish activity of the cytoplasmic NNR protein, but does prevent export of haem to the periplasm for assembly of the nitrite reductase. Also, expression of the P. denitrificans nitrite reductase is not fully dependent on NNR (Van Spanning et al., 1999). In Ps. stutzeri, the NO-responsiveness of the NNR orthologue DnrD is not abolished by a mutation that prevents biosynthesis of haem d1 (Vollack & Zumft, 2001), implying that haem d1 is not required for DnrD activity. This is consistent with the fact that NNR is active in E. coli, an organism that does not make haem d1.
We have sought signal-independent variants of NNR (NNR*) in the hope that characterization of such proteins might shed further light on the regulatory mechanism. NNR* proteins with substitutions of Arg-80 were isolated; these proteins appear to have a perturbed signal-recognition mechanism, such that they are independent of the activating signal (NO). As far as we can tell, the two proteins substituted at position 80 (with alanine and histidine) have similar properties, despite the very different physicochemical characteristics of the two amino acids. The NNR* proteins remain sensitive to aerobic growth conditions. NNR is inactive during anaerobic growth in the absence of NO and during aerobic growth. If lack of activity under aerobic growth conditions reflects the rapid metabolism of the activating signal (most likely NO), then the two inactive states of NNR would be equivalent, in that they both result simply from the absence of NO. If this were the case, gain-of-function mutations should have similar effects under aerobic and anaerobic (in the absence of NO) conditions. This is not the case, since the star mutations selectively enhance the activity of NNR under anaerobic conditions (Table 1). Thus, it is likely that NNR is directly inactivated by O2 or perhaps by an indirect signal of aerobic metabolism. Rapid inactivation of NNR is consistent with the observation that nitrite reductase mRNA abundance decays to the basal level within 1 h of switching to aerobic growth in cultures of P. denitrificans (Baumann et al., 1996).
FNR* proteins have been isolated and have proved crucial in establishing the mechanism of oxygen sensing. FNR is inactivated by oxygen and FNR* proteins, with increased activity in the presence of oxygen, have mutations that stabilize an oxygen-sensitive Fe–S cluster (Bates et al., 2000) or increase the tendency of the protein to dimerize in the presence of oxygen (Lazazzera et al., 1993; Moore & Kiley, 2001). CooA from Rhodospirillum rubrum is another FNR/CRP family member which contains a haem and is activated by CO binding to the haem. Like NNR, CooA is inactive in anaerobic cultures in the absence of its signal (CO) and is inactive in aerobic cultures (Thorsteinsson et al., 2001). CO-independent CooA variants (CooA*) have been isolated, some of which have substitutions in the C-helix of the protein that forms the major monomer–monomer interface (Kerby et al., 2003). Introduction of a novel haem ligand in the C-helix also generates a CooA* phenotype (Youn et al., 2002). For both CooA and CRP there are also signal-independent variants that have substitutions in a hinge region connecting the effector-binding and DNA-binding domains; these mutations presumably lock the protein in an active conformation (Aono et al., 1997). In an effort to understand the basis for the star phenotypes of the NNR R80A and R80H proteins, we have modelled the structure of NNR by homology-based methods, using the structures of CooA in the inactive conformation (Lanzilotta et al., 2000) and CRP in the active conformation (Passner et al., 2000) as templates. In the NNR model, Arg-80 is predicted to be in the effector-binding domain, with its side chain projecting towards the C-helix at the monomer–monomer interface, where it approaches to within 5·5 Å of a histidine residue at position 134 (Fig. 3). Interestingly, Arg-80 is not conserved amongst orthologues of NNR; indeed, NNR appears to be unique in the wider FNR/CRP family in having an arginine at this position. His-134 is also not conserved amongst NNR orthologues, and occurs rarely at this position in FNR/CRP family members. Thus, given the phenotypes of the R80A and R80H proteins, we are drawn to the conclusion that NNR orthologues that sense NO may not have a common mechanism for so doing. The equivalent position to Arg-80 in CRP is occupied by an aspartate residue, which is salt-bridged to Arg-123 in the C-helix (at the position equivalent to His-134 of NNR). Arg-123 of CRP is one of the cAMP ligands, and substitutions at this position have a severe effect on CRP activity (Moore et al., 1992). Several residues in the C-helix of CooA are close to the haem, and the L116K substitution in the C-helix is thought to generate a novel axial ligand to the haem (Youn et al., 2002). Leu-116 of CooA is three residues away from His-134 of NNR in a sequence alignment and is displaced by three positions from His-134 in helical wheel projections (of the known structure of CooA and the modelled structure of NNR). Thus, taken together, the evidence from related systems is consistent with the idea that Arg-80 and/or His-134 approach closely to a site in NNR in which signals (NO and O2) are sensed, or that substitutions at Arg-80 perturb the structural rearrangements around the C-helix that are presumably required for activation (Kerby et al., 2003; Moore & Kiley, 2001).
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Baker, S. C., Ferguson, S. J., Ludwig, B., Page, M. D., Richter, O. M. & van Spanning, R. J. (1998). Molecular genetics of the genus Paracoccus: metabolically versatile bacteria with bioenergetic flexibility. Microbiol Mol Biol Rev 62, 1046–1078.
Bates, D. M., Popescu, C. V., Khoroshilova, N., Vogt, K., Beinert, H., Munck, E. & Kiley, P. J. (2000). Substitution of leucine 28 with histidine in the Escherichia coli transcription factor FNR results in increased stability of the [4Fe-4S](2+) cluster to oxygen. J Biol Chem 275, 6234–6240.
Bates, P. A., Kelley, L. A., MacCallum, R. M. & Sternberg, M. J. (2001). Enhancement of protein modeling by human intervention in applying the automatic programs 3D-JIGSAW and 3D-PSSM. Proteins 39–46.
Baumann, B., Snozzi, M., Zehnder, A. & Van Der Meer, J. (1996). Dynamics of denitrification activity of Paracoccus denitrificans in continuous culture during aerobic-anaerobic changes. J Bacteriol 178, 4367–4374.
Busby, S. & Ebright, R. H. (1999). Transcription activation by catabolite activator protein (CAP). J Mol Biol 293, 199–213.[CrossRef][Medline]
Constantinidou, C., Hobman, J. L., Griffiths, L., Patel, M. D., Penn, C. W., Cole, J. A. & Overton, T. W. (2006). A reassessment of the FNR regulon and transcriptomic analysis of the effects of nitrate, nitrite, NarXL and NarQP as Escherichia coli K-12 adapts from aerobic to anaerobic growth. J Biol Chem 281, 4802–4815.
Corker, H. & Poole, R. K. (2003). Nitric oxide formation by Escherichia coli: dependence on nitrite reductase, the NO-sensing regulator Fnr, and flavohemoglobin Hmp. J Biol Chem 278, 31584–31592.
De Groote, M. A., Granger, D., Xu, Y., Campbell, G., Prince, R. & Fang, F. C. (1995). Genetic and redox determinants of nitric oxide cytotoxicity in a Salmonella typhimurium model. Proc Natl Acad Sci U S A 92, 6399–6403.
Elliott, T. & Roth, J. R. (1989). Heme-deficient mutants of Salmonella typhimurium: two genes required for ALA synthesis. Mol Gen Genet 216, 303–314.[CrossRef][Medline]
Gardner, A. M. & Gardner, P. R. (2002). Flavohemoglobin detoxifies nitric oxide in aerobic, but not anaerobic, Escherichia coli. Evidence for a novel inducible anaerobic nitric oxide-scavenging activity. J Biol Chem 277, 8166–8171.
Guzman, L. M., Belin, D., Carson, M. J. & Beckwith, J. (1995). Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J Bacteriol 177, 4121–4130.
Hasegawa, N., Arai, H. & Igarashi, Y. (1998). Activation of a consensus FNR-dependent promoter by DNR of Pseudomonas aeruginosa in response to nitrite. FEMS Microbiol Lett 166, 213–217.[CrossRef][Medline]
Hutchings, M. I., Shearer, N., Wastell, S., van Spanning, R. J. & Spiro, S. (2000). Heterologous NNR-mediated nitric oxide signaling in Escherichia coli. J Bacteriol 182, 6434–6439.
Jahn, D., Michelsen, U. & Soll, D. (1991). Two glutamyl-tRNA reductase activities in Escherichia coli. J Biol Chem 266, 2542–2548.
Ji, X. B. & Hollocher, T. C. (1988). Reduction of nitrite to nitric oxide by enteric bacteria. Biochem Biophys Res Commun 157, 106–108.[CrossRef][Medline]
Kerby, R. L., Youn, H., Thorsteinsson, M. V. & Roberts, G. P. (2003). Repositioning about the dimer interface of the transcription regulator CooA: a major signal transduction pathway between the effector and DNA-binding domains. J Mol Biol 325, 809–823.[CrossRef][Medline]
Korner, H., Sofia, H. J. & Zumft, W. G. (2003). Phylogeny of the bacterial superfamily of Crp-Fnr transcription regulators: exploiting the metabolic spectrum by controlling alternative gene programs. FEMS Microbiol Rev 27, 559–592.[CrossRef][Medline]
Kwiatkowski, A. V. & Shapleigh, J. P. (1996). Requirement of nitric oxide for induction of genes whose products are involved in nitric oxide metabolism in Rhodobacter sphaeroides 2.4.3. J Biol Chem 271, 24382–24388.
Lamberg, K. E., Luther, C., Weber, K. D. & Kiley, P. J. (2002). Characterization of activating region 3 from Escherichia coli FNR. J Mol Biol 315, 275–283.[CrossRef][Medline]
Lanzilotta, W. N., Schuller, D. J., Thorsteinsson, M. V., Kerby, R. L., Roberts, G. P. & Poulos, T. L. (2000). Structure of the CO sensing transcription activator CooA. Nat Struct Biol 7, 876–880.[CrossRef][Medline]
Laratta, W. P. & Shapleigh, J. P. (2003). Site-directed mutagenesis of NnrR: a transcriptional regulator of nitrite and nitric oxide reductase in Rhodobacter sphaeroides. FEMS Microbiol Lett 229, 173–178.[CrossRef][Medline]
Lazazzera, B. A., Bates, D. M. & Kiley, P. J. (1993). The activity of the Escherichia coli transcription factor FNR is regulated by a change in oligomeric state. Genes Dev 7, 1993–2005.
Li, B., Wing, H., Lee, D., Wu, H. C. & Busby, S. (1998). Transcription activation by Escherichia coli FNR protein: similarities to, and differences from, the CRP paradigm. Nucleic Acids Res 26, 2075–2081.
Martin, E., Sharina, I., Kots, A. & Murad, F. (2003). A constitutively activated mutant of human soluble guanylyl cyclase (sGC): implication for the mechanism of sGC activation. Proc Natl Acad Sci U S A 100, 9208–9213.
Mesa, S., Bedmar, E. J., Chanfon, A., Hennecke, H. & Fischer, H.-M. (2003). Bradyrhizobium japonicum NnrR, a denitrification regulator, expands the FixLJ-FixK2 regulatory cascade. J Bacteriol 185, 3978–3982.
Miller, J. H. (1992). A Short Course in Bacterial Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Moore, L. J. & Kiley, P. J. (2001). Characterization of the dimerization domain in the FNR transcription factor. J Biol Chem 276, 45744–45750.
Moore, J., Kantorow, M., Vanderzwaag, D. & McKenney, K. (1992). Escherichia coli cyclic AMP receptor protein mutants provide evidence for ligand contacts important in activation. J Bacteriol 174, 8030–8035.
Notredame, C., Higgins, D. G. & Heringa, J. (2000). T-Coffee: a novel method for fast and accurate multiple sequence alignment. J Mol Biol 302, 205–217.[CrossRef][Medline]
Page, M. D. & Ferguson, S. J. (1990). Apo forms of cytochrome c550 and cytochrome cd1 are translocated to the periplasm of Paracoccus denitrificans in the absence of haem incorporation caused either mutation or inhibition of haem synthesis. Mol Microbiol 4, 1181–1192.[CrossRef][Medline]
Passner, J. M., Schultz, S. C. & Steitz, T. A. (2000). Modeling the cAMP-induced allosteric transition using the crystal structure of CAP-cAMP at 2·1 A resolution. J Mol Biol 304, 847–859.[CrossRef][Medline]
Poole, R. K. & Hughes, M. N. (2000). New functions for the ancient globin family: bacterial responses to nitric oxide and nitrosative stress. Mol Microbiol 36, 775–783.[CrossRef][Medline]
Spencer, M. E. & Guest, J. R. (1973). Isolation and properties of fumarate reductase mutants of Escherichia coli. J Bacteriol 114, 563–570.
Spencer, N. Y., Zeng, H., Patel, R. P. & Hogg, N. (2000). Reaction of S-nitrosoglutathione with the heme group of deoxyhemoglobin. J Biol Chem 275, 36562–36567.
Thorsteinsson, M. V., Kerby, R. L., Youn, H., Conrad, M., Serate, J., Staples, C. R. & Roberts, G. P. (2001). Redox-mediated transcriptional activation in a CooA variant. J Biol Chem 276, 26807–26813.
Van Doorslaer, S., Dewilde, S., Kiger, L., Nistor, S. V., Goovaerts, E., Marden, M. C. & Moens, L. (2003). Nitric oxide binding properties of neuroglobin. A characterization by EPR and flash photolysis. J Biol Chem 278, 4919–4925.
Van Spanning, R. J., De Boer, A. P., Reijnders, W. N., Spiro, S., Westerhoff, H. V., Stouthamer, A. H. & Van der Oost, J. (1995). Nitrite and nitric oxide reduction in Paracoccus denitrificans is under the control of NNR, a regulatory protein that belongs to the FNR family of transcriptional activators. FEBS Lett 360, 151–154.[CrossRef][Medline]
Van Spanning, R. J., De Boer, A. P., Reijnders, W. N., Westerhoff, H. V., Stouthamer, A. H. & Van Der Oost, J. (1997). FnrP and NNR of Paracoccus denitrificans are both members of the FNR family of transcriptional activators but have distinct roles in respiratory adaptation in response to oxygen limitation. Mol Microbiol 23, 893–907.[CrossRef][Medline]
Van Spanning, R. J., Houben, E., Reijnders, W. N., Spiro, S., Westerhoff, H. V. & Saunders, N. (1999). Nitric oxide is a signal for NNR-mediated transcription activation in Paracoccus denitrificans. J Bacteriol 181, 4129–4132.
Vollack, K. U. & Zumft, W. G. (2001). Nitric oxide signaling and transcriptional control of denitrification genes in Pseudomonas stutzeri. J Bacteriol 183, 2516–2526.
Vollack, K. U., Hartig, E., Korner, H. & Zumft, W. G. (1999). Multiple transcription factors of the FNR family in denitrifying Pseudomonas stutzeri: characterization of four fnr-like genes, regulatory responses and cognate metabolic processes. Mol Microbiol 31, 1681–1694.[CrossRef][Medline]
Watmough, N. J., Butland, G., Cheesman, M. R., Moir, J. W., Richardson, D. J. & Spiro, S. (1999). Nitric oxide in bacteria: synthesis and consumption. Biochim Biophys Acta 1411, 456–474.[Medline]
Youn, H., Kerby, R. L., Thorsteinsson, M. V., Clark, R. W., Burstyn, J. N. & Roberts, G. P. (2002). Analysis of the L116K variant of CooA, the heme-containing CO sensor, suggests the presence of an unusual heme ligand resulting in novel activity. J Biol Chem 277, 33616–33623.
Zumft, W. G. (1997). Cell biology and molecular basis of denitrification. Microbiol Mol Biol Rev 61, 533–616.[Abstract]
Zumft, W. G. (2002). Nitric oxide signaling and NO dependent transcriptional control in bacterial denitrification by members of the FNR-CRP regulator family. J Mol Microbiol Biotechnol 4, 277–286.[Medline]
Received 21 December 2005;
revised 31 January 2006;
accepted 2 February 2006.
This article has been cited by other articles:
|
|
 |
|
  N. P. Tucker, B. D'Autreaux, F. K. Yousafzai, S. A. Fairhurst, S. Spiro, and R. Dixon Analysis of the Nitric Oxide-sensing Non-heme Iron Center in the NorR Regulatory Protein J. Biol. Chem., January 11, 2008; 283(2): 908 - 918. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Bushart and S. J. Roux Conserved Features of Germination and Polarized Cell Growth: A Few Insights from a Pollen-Fern Spore Comparison Ann. Bot., January 1, 2007; 99(1): 9 - 17. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||
