1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

The oxidative decarboxylation of coproporphyrinogen III catalysed by an oxygen-dependent oxidase (HemF) and an oxygen-independent dehydrogenase (HemN) is one of the key regulatory points of haem biosynthesis in Pseudomonas aeruginosa. To investigate the oxygen-dependent regulation of hemF and hemN, the corresponding genes were cloned from the P. aeruginosa chromosome. Recognition sequences for the Fnr-type transcriptional regulator Anr were detected −44.5 bp from the 5′ end of the hemF mRNA transcript and at an optimal distance of −41.5 bp with respect to the transcriptional start of hemN. An approximately 10-fold anaerobic induction of hemN gene expression was mediated by the dual action of Anr and a second Fnr-type regulator, Dnr. Regulation by both proteins required the Anr recognition sequence. Surprisingly, aerobic expression of hemN was dependent only on Anr. An anr mutant did not contain detectable amounts of hemN mRNA and accumulated coproporphyrin III both aerobically and anaerobically, indicating the importance of HemN for aerobic and anaerobic haem formation. Mutation of hemN and hemF did not abolish aerobic or anaerobic growth, indicating the existence of an additional HemN-type enzyme, which was termed HemZ. Expression of hemF was induced approximately 20-fold during anaerobic growth and, as was found for hemN, both Anr and Dnr were required for anaerobic induction. Paradoxically, oxygen is necessary for HemF catalysis, suggesting the existence of an additional physiological function for the P. aeruginosa HemF protein.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

The tetrapyrrole requirement of a bacterial cell depends on the environmental conditions and, as a consequence, on the choice of the most effective mode of energy generation. Electron transfer associated with respiratory processes requires haem biosynthesis for cytochrome function, while low-energy-yielding fermentative processes lead to a reduction in haem formation. Oxygen tension was found to be one of the most effective parameters regulating bacterial haem formation. (Doss and Philipp-Dornston, 1971; Jacobs et al., 1972, 1973; Philipp-Dornston and Doss, 1973). In the denitrifying soil bacterium Pseudomonas aeruginosa clear differences in haem content between aerobic and anaerobic denitrifying or arginine fermenting growth were observed (Doss and Philipp-Dornston, 1971; Jacobs et al., 1972; 1973). Formation of the general precursor molecule for all tetrapyrroles, 5-aminolaevulinic acid (ALA), and the oxidative decarboxylation of coproporphyrinogen III are the regulated steps of P. aeruginosa haem biosynthesis (Doss and Philipp-Dornston, 1971; Jacobs et al., 1972; 1973; Philipp-Dornston and Doss, 1973; Jahn et al., 1992; Hungerer et al., 1995a,b). Oxidative decarboxylation of coproporphyrinogen III to form protoporphyrinogen IX is catalysed in bacteria by two different enzymes (Tait, 1969; 1972). Coproporphyrinogen III oxidase encoded by the hemF gene requires molecular oxygen for catalysis (Tait., 1969; 1972; Seehra et al., 1982; Xu et al., 1992; Xu and Elliott, 1993; Troup et al., 1994). During anaerobic growth, a second oxygen-independent enzyme encoded by the hemN gene is required (Tait, 1969, 1972; Seehra et al., 1982; Coomber et al., 1992; Xu et al., 1992; Xu and Elliott, 1994; Troup et al., 1995). It has been suggested that this enzyme, which is probably S-adenosylmethionine dependent, should be called coproporpyhrinogen III dehydrogenase (Tait, 1972).

Both the hemF and hemN genes have been cloned from a variety of bacterial sources (Coomber et al., 1992; Xu and Elliott, 1993; 1994; Troup et al., 1994; 1995; Fleischmann et al., 1995; Zeilstra-Ryalls and Kaplan, 1995; de Gier et al., 1996; Homuth et al., 1996; Hippler et al., 1997; Tomb et al., 1997; Lieb et al., 1998; accession nos D90901, D90904, D90912 and Z81368). Transcription of the Escherichia coli hemF gene has been shown to be induced by oxygen stress and the growth phase (Mukhopadhyay and Schellhorn, 1997), while the E. coli hemN gene is induced under anaerobic growth conditions and its expression is dependent on the presence of iron (Troup et al., 1995). Helicobacter pylori and Bacillus subtilis possess two hemN genes and no hemF gene (Hippler et al., 1997; Tomb et al., 1997). No obvious oxygen regulation was observed for the investigated B. subtilis hemN gene (Hippler et al., 1997). In Alcaligenes eutrophus, the hemN gene has a potential Fnr binding site in its 5′ region, and hemN has been found to be essential for anaerobic growth (Lieb et al., 1998). In E. coli, hemN is essential for the formation of active nitrite reductase under anaerobic conditions (Tyson et al., 1997). The observed differences in the occurrence, number and regulation of genes involved in coproporphyrinogen III oxidation reflect the importance of this key regulatory point for the species-specific adjustment of haem biosynthesis to different physiological conditions.

Here, we describe the cloning and sequencing of the P. aeruginosa hemF and hemN genes and the analysis of their regulation by oxygen tension via the two transcription factors Anr and Dnr. The redox response regulator Anr contains the conserved cysteine residues involved in the formation of the proposed 4Fe/4S cluster of the E. coli homologue Fnr, which is responsible for redox sensing (Sawers, 1991; Zimmermann et al., 1991; Khoroshilova et al., 1995; Green et al., 1996; Lazazzera et al., 1996). Anr uses a DNA binding site (TTGAT … ATCAA) similar to the Fnr site (Haas et al., 1992; Winteler and Haas, 1996). The amino acid sequence of Dnr deduced from the recently cloned gene, while similar to Anr, lacks these conserved cysteine residues, excluding its role as a direct redox sensor (Arai et al., 1995).

However, the overall amino acid sequence and especially the potential helix–turn–helix motif for DNA binding show a high degree of similarity between Anr and Dnr, suggesting the recognition and binding of similiar DNA sequences. Recently, the anaerobic induction of various genes involved in denitrification (nirS, norC) has been demonstrated via an Anr–Dnr regulatory cascade (Arai et al., 1997). In this cascade, changes in oxygen tension are sensed by Anr, which then activates the expression of the dnr gene. Dnr is directly responsible for inducing the expression of the target genes of the denitrification pathway (Arai et al., 1997). As cytochromes are integral parts of the denitrification machinery, a potential anaerobic co-regulation of haem biosynthesis via Anr and Dnr activity was investigated. In this report, we provide evidence that indicates that both Anr and Dnr function directly in regulating anaerobic expression of the P. aeruginosa hemF and hemN genes.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

Cloning and DNA sequence determination of genomic fragments encoding the P. aeruginosa hemF and hemN genes

Previous physiological investigations have demonstrated the presence of a regulated step in P. aeruginosa haem biosynthesis at the level of coproporphyrinogen III conversion to protoporphyrinogen IX (Doss and Philipp-Dornston, 1971; Jacobs et al., 1972; 1973; Philipp-Dornston and Doss, 1973). To elucidate the role of oxygen tension at the level of gene expression, the hemF gene encoding the oxygen-dependent coproporphyrinogen III oxidase and the hemN gene encoding the oxygen-independent coproporphyrinogen III dehydrogenase were cloned from a P. aeruginosa gene library by functional aerobic complementation of a S. typhimurium hemF hemN double mutant (TE3006) (Xu et al., 1992; Troup et al., 1994; 1995). Three clones, each containing the same 4202 bp BamHI fragment, which only complemented the Salmonella double mutant during aerobic growth, were obtained. Complete DNA sequence determination of the fragment (database accession no. X85015) revealed the presence of three complete open reading frames (ORFs) (Fig. 1). The protein deduced from the first ORF (position 1031–54 of the BamHI fragment) had 59% identity at the amino acid level with an E. coli soluble quinone oxidoreductase encoded by the qor gene (Thorn et al., 1995). The protein deduced from the second ORF transcribed in the opposite direction to qor (position 1222–2139) showed 70% amino acid sequence identity to the product of the E. coli hemF gene (Troup et al., 1994). Owing to its high amino acid similarity to other known HemF proteins and its capacity to complement the S. typhimurium hemF hemN mutant exclusively under aerobic conditions, the cloned P. aeruginosa gene was named hemF. The protein encoded by the third ORF (position 2197–3021) had 52% amino acid sequence identity to the product of the E. coli aroE gene encoding shikimate 5-dehydrogenase (Anton and Coggins, 1988). The hemF and aroE genes are transcribed in the same orientation and separated by only 57 bp. The short distance and the absence of obvious promoter elements between both genes could indicate that they are co-transcribed. The proposed structure of all three ORFs (qor, hemF and aroE ) and the deduced molecular masses of the encoded proteins were experimentally confirmed by in vivo protein synthesis experiments in E. coli using T7 RNA polymerase-driven systems as described previously (data not shown; Troup et al., 1994; 1995).


Figure 1. . Genomic arrangement of the P. aeruginosa hemF (A) and hemN genes (B). The orientation of the genes found on the cloned genomic fragments containing hemF (for oxygen-dependent coproporphyrinogen III oxidase), qor (for NADPH-dependent quinone oxidoreductase), aroE (for shikimate 5-dehydrogenase), hemN (for oxygen-independent coproporphyrinogen III dehydrogenase), orfX (unknown function) and anr (for the redox response regulator Anr) are outlined. The location and DNA sequences of the detected Anr binding sites are indicated. The localization and structure of various reporter gene fusions of hemF, qor and hemN with lacZ are shown. Construction of the various fusions is described in Experimental procedures.

Download figure to PowerPoint

Six clones complementing the S. typhimurium hemF hemN double mutant under aerobic and anaerobic conditions were obtained, indicating the presence of the hemN gene encoding the oxygen-independent coproporphyrinogen III dehydrogenase. Four of the six clones contained the same approximately 4300 bp EcoRI fragment, which was partially sequenced (deposited in the database under the accession no. X97981). DNA sequences were obtained that were part of an ORF (orfX ) already cloned by Haas and colleagues during examination of the genomic region flanking the anr gene (position 2607–2140 of X97981; Zimmermann et al., 1991). Downstream from orfX, an ORF (position 709–2031) encoding a protein with strong amino acid identity to HemN proteins (50% amino acid identity to E. coli HemN) deduced from previously cloned hemN genes was found. Therefore, the cloned gene will be referred to as P. aeruginosa hemN (Fig. 1). The molecular masses of the proteins encoded by orfX and hemN were determined experimentally by in vivo protein synthesis experiments in E. coli as described above (data not shown).

Mapping of the hemF and hemN gene to the P. aeruginosa chromosome

Physical mapping of the hemF and hemN genes to the P. aeruginosa PAO1 chromosome was performed by a method described previously (Römling et al., 1989). As shown in 2Fig. 2A, the SpeI fragment AC and the DpnI fragment E contained DNA sequences corresponding to hemF, while the SpeI fragment R and the DpnI fragment C contained DNA sequences identical to hemN (Fig. 2B). In both cases, the two identified fragments map to the same area of the chromosome corresponding to 14′ on the new genomic map and 28′ on the old genomic map for hemF and 57′–59′ on the new and 83′–85′ on the old map for hemN (Holloway et al., 1994).


Figure 2. . Physical mapping of the hemF and hemN genes to the P. aeruginosa chromosome. A and B. Digests of the P. aeruginosa PAO1 chromosome with the restriction enzymes SpeI (lanes a2 and b2) and DpnI (lanes c2 and d2) and lambda DNA digested with HindIII as size standard (lanes a1, b1, c1 and d1). The genomic DNA in (A) was probed with the labelled hemF gene and in (B) with the labelled hemN gene. Lambda DNA standards were visualized with labelled lambda DNA (Römling et al., 1989). The hemF probe detected the fragments SpeI AC and DpnI E, while the hemN probe detected the fragments SpeI R and DpnI C according to the physical map of Holloway et al. (1994).

Download figure to PowerPoint

The physiological consequences of hemF and hemN mutation and the evidence for a second HemN-type enzyme called HemZ

P. aeruginosa mutants carrying insertions in hemF (PAO9101) or hemN (PAO9102) and a hemF hemN double mutant (PAO9103) were constructed and tested for their aerobic and anaerobic growth characteristics. Surprisingly, mutation of both genes (PAO9103) did not abolish aerobic and anaerobic respiratory growth, indicating the existence of an additional oxygen-independent HemN-type enzyme, termed HemZ. The presence of two HemN-type enzymes has been described for Bacillus subtilis, Helicobacter pylori and Rhodobacter sphaeroides (Zeilstra-Ryalls and Kaplan, 1995; Hippler et al., 1997; Tomb et al., 1997). Similar to our findings, mutation of the B. subtilis hemN did not result in any obvious growth phenotype, as a second HemN-type enzyme HemZ complements the defect. While HemN and HemZ possess a high degree of amino acid sequence identity, HemF has no structural relationship to either oxygen-independent enzyme.

Notably, the hemF hemN double mutant PAO9103 grown in the presence of ALA accumulated high amounts of coproporphyrin III under both aerobic and anaerobic growth conditions, indicating a significant block in haem biosynthesis at the level of coproporphyrinogen III conversion (Table 1). As expected, hemF (PAO9101) and hemN (PAO9102) single mutations did not exhibit any obvious growth defects (data not shown). Interestingly, the hemN mutant accumulated coproporphyrin III under both aerobic and anaerobic growth conditions, indicating the importance of HemN for both aerobic and anaerobic coproporpyhrinogen III conversion.

Table 1.  . Comparison of the intracellular coproporphyrin III levels between wild-type P. aeruginosa (PAO1), the anr deletion mutant (PAO6261), the hemF mutant (PAO9101) and the hemN mutant (PAO9102). a. Methyl esters of the porphyrins were formed, separated via thin-layer chromatography and quantified as outlined in Experimental procedures.Thumbnail image of

Determination of the 5′ ends of the hemF and hemN mRNAs and identification of potential promoter elements

The 5′ regions of the P. aeruginosa hemF and hemN genes (Fig. 3) were analysed for potential transcription start sites using the primer extension technique (Boorstein and Craig, 1989; Troup et al., 1994; Hungerer et al., 1995a). As shown in 3Fig. 3A, the 5′ end of the hemF mRNA was located 56 bp upstream from the translational start codon. A DNA sequence (TTGAC-GGCC-GTCAA) matching the Anr consensus binding site (Anr box) and centred 44.5 bp upstream from the transcriptional start site was detected (Jayaraman et al., 1988; 1989; Bell et al., 1990; Haas et al., 1992). As is often found for Anr-regulated genes, no obvious −35 region of a σ70-dependent promoter was detected, while a potential −10 region was observed downstream of the Anr box (Winteler and Haas, 1996).


Figure 3. . Determination of the 5′ end of the hemF (A) and hemN (B) mRNA. Primer extension experiments were performed using total cellular RNA prepared from wild-type P. aeruginosa PAO1 [lanes 1, 2 and 5 in (A); lanes 1, 2, 5 and 6 in (B)] and an anr deletion mutant PAO6261 [lanes 3, 4, 6 and 7 in (A); lanes 3, 4, 7 and 8 in (B)] grown under aerobic (lanes 1–4 in A and B) or anaerobic denitrifying conditions [lanes 5–7 in (A) and lanes 5–8 in (B)]. The amounts of cellular RNA used were 5 μg [lanes 2, 4, 6 and 8 in (B)], 10 μg [lanes 2, 4, 5 and 6 in (A)], 50 μg [lanes 1, 3, 5 and 7 in (B)] and 100 μg [lanes 1, 3 and 7 in (A)]. The position of the band corresponding to the hemF and hemN transcripts is indicated by arrows. DNA sequencing reactions using the same primers used for the extension experiments were performed and separated on the same sequencing gel. The lanes are labelled according to the sequencing reactions. Note that, for the hemF and hemN experiments with the anr mutant, the same RNA preparation was used. Promoter regions of hemF (C) and hemN (D). The positions of the determined transcriptional start sites are indicated by arrows. Potential Anr/Dnr binding sites (Anr/Dnr) and −10 regions (−10) are shown.

Download figure to PowerPoint

The transcriptional start site for the hemN gene was mapped to a position 29 bp upstream from the translational start site (Fig. 3B, lanes 1 and 5). A perfect Anr box (TTGAT-ACAA-ATCAA) was found exactly 41.5 bp upstream from the transcriptional start site of hemN. No obvious −35 region of a σ70-dependent promoter was detected, while a potential −10 region was observed downstream of the Anr box. In both primer extension experiments, weak non-specific signals were observed. Most of these signals were also present in the lanes without specific transcripts. Control experiments using different primers confirmed that these weak signals were indeed non-specific (data not shown).

To investigate the influence of the redox response regulator Anr on the observed utilization of the hemF and hemN promoters, RNA prepared from an aerobically and anaerobically grown anr mutant was compared with RNA prepared from wild-type cells. No obvious differences in the intensity of the signals were observed for hemF under aerobic conditions between the wild type and the anr mutant (Fig. 3A, lanes 1 and 3). Under anaerobic conditions, a clear induction of hemF was observed (Fig. 3A, lane 1 using 100 μg of RNA and lane 5 using 10 μg of RNA). This induction was dependent on Anr (Fig. 3A, lane 7 using 100 μg of RNA). These results indicate an anaerobic anr-dependent induction of hemF. Surprisingly, aerobic hemN transcription was totally abolished in the anr mutant (Fig. 3B, lanes 1 and 3). Only a slight anaerobic increase in the hemN signal was detectable using the primer extension technique (Fig. 3B, lane 1 using 50 μg of RNA and lane 5 using 10 μg of RNA). Again, mutation of anr abolished hemN transcription (Fig. 3B, lanes 5 and 7). Note that, for the hemF and hemN experiments, the same RNA preparations were used. These results indicate an anr-dependent aerobic and anaerobic transcription of hemN.

Mutation of the P. aeruginosa anr gene drastically reduces aerobic and anaerobic coproporphyrinogen III conversion

As both cloned genes were found to be regulated by the redox response regulator Anr, we investigated the physiological consequences of an anr mutation on haem formation. For this purpose, biosynthetic intermediates including coproporphyrin III from various strains were isolated and quantified. As shown in Table 1, the anr mutant accumulated 260-fold more coproporphyrin III under aerobic and 8.2-fold more under anaerobic conditions compared with wild-type P. aeruginosa. Interestingly, the wild-type strain PAO1 accumulated slightly more intermediates under anaerobic compared with aerobic growth conditions. This observation is in good agreement with data from earlier investigations, which showed that haem biosynthesis was induced when P. aeruginosa was growing under anaerobic denitrifying conditions (Hungerer et al., 1995a). A plausible explanation as to why the anr mutant accumulated more coproporphyrin III under aerobic compared with anaerobic growth conditions could be provided by our previous findings, which indicate that Anr is required for anaerobic hemA induction and thereby regulates the overall flow of the haem precursor ALA into the pathway (Hungerer et al., 1995a). In the anr mutant under anaerobic growth conditions, the induction of haem biosynthesis at the initial step of ALA formation does not occur, less metabolic flow into the haem biosynthetic pathway results and, consequently, less coproporphyrin III is formed and accumulated. The intracellular accumulation of coproporphyrin III in the anr mutant grown aerobically leads to excretion of the porphyrins into the medium. These results clearly indicate the importance of Anr for the biosynthesis of the coproporphyrinogen III-using enzymes.

Anaerobic induction of P. aeruginosa hemN requires the dual action of Anr and Dnr, while aerobic expression depends solely on the presence of Anr

For a detailed analysis of the redox-dependent transcription of hemN, two fragments of different lengths containing the potential hemN promoter region and including the Anr binding site were fused to lacZ. Comparison of the β-galactosidase activity obtained from wild-type strains carrying the two constructs revealed a 7.2-fold anaerobic induction for the short promoter fragment (phemNlacZ563–818) and a 10.8-fold anaerobic induction for the long fragment (phemNlacZ176–1184). Control reactions using a promoterless construct (pQF50) in P. aeruginosa wild type (PAO1) showed low β-galactosidase activity, which was independent of the growth conditions (Table 2). Analysis of the hemN–lacZ fusions in an anr deletion strain (PAO6261) in the presence and in the absence of oxygen revealed only low background activity similar to the control reaction using the promoterless lacZ gene (Table 2). These results demonstrate the importance of Anr for aerobic and anaerobic hemN expression (Table 2). Overexpression of the anr gene in the anr deletion strain led to a drastic increase in aerobic hemN expression to the level attained during anaerobic growth, while the level of anaerobic induction was restored to the wild-type level (Table 2). Similar results were obtained for the overexpression of anr in wild-type cells carrying a hemNlacZ fusion (Table 2). A related observation has been noted for the anaerobically induced E. coli pfl gene in which the presence of the fnr gene on a multicopy plasmid led to the induction under aerobic and anaerobic conditions (G. Sawers, personal communication).

Table 2.  . The influence of oxygen and the redox response regulators Anr and Dnr on the expression of P. aeruginosa hemN. a. All values are the result of three independent experiments performed in triplicate. The β-galactosidase activities are given in Miller units and are calculated as outlined in Experimental procedures.b. The anr and dnr genes were cloned in a vector compatible with the lacZ fusion vector pQF50 and co-transformed where indicated. Vector without insert served as a control.Thumbnail image of

Mutation of the Anr binding site by changing the TTGAT-ACAA-ATCAA sequence to CGCGT-ACAA-ATCAA (phemNlacZ563–818ΔANR) drastically reduced anaerobic as well as aerobic hemN expression in wild-type cells and in the anr mutant (Table 2). These results clearly indicate that Anr is not only necessary for anaerobic transcription of hemN but also that Anr is active under aerobic conditions and is absolutely required for the aerobic expression of hemN. Further evidence for an aerobic function of Anr was provided by the finding that RNA prepared from aerobically grown P. aeruginosa wild-type cells gave rise to a clear primer extension signal with a hemN specific primer, while no signals were observed with RNA prepared from the anr deletion mutant (Fig. 3B). The same RNA preparations from the aerobically grown anr mutant did, however, deliver signals in primer extension experiments corresponding to the 5′ end of the hemF mRNA (Fig. 3A, lane 3). These findings are in agreement with the results of a recent investigation of nitrite reductase expression in P. aeruginosa PAO1 in which an aerobic requirement of Anr was identified (Ka et al., 1997). Earlier investigations also uncovered an aerobic activity of the E. coli Fnr protein (Sawers et al., 1988).

Recently, the anaerobic induction of various denitrification genes in P. aeruginosa via Anr was found to depend on the activity of a second regulator named Dnr (Arai et al., 1997). In the Anr–Dnr regulatory cascade, anaerobic conditions were necessary for an Anr-dependent activation of dnr transcription. Consequently, Dnr and not Anr activated the transcription of various target genes coding for enzymes of the denitrification process (Arai et al., 1997). To test the influence of Dnr on hemN transcription, reporter gene fusions were tested in the P. aeruginosa dnr mutant RM536. Mutation of dnr significantly reduced the observed anaerobic induction of hemN (Table 2). However, aerobic hemN transcription remained unaffected by the dnr mutation. First, these results indicate that Dnr is required for anaerobic hemN induction. Secondly, from the fact that mutation of dnr did not affect aerobic hemN expression, we conclude that only Anr is required for aerobic hemN expression.

To investigate the relationship between Anr and Dnr during anaerobic hemN activation, cross-complementation experiments were performed. For this purpose, anaerobic hemN expression was measured in a dnr mutant expressing anr in trans and vice versa (Table 2). Surprisingly, no clear cross-complementation was observed. In the case of an Anr–Dnr cascade (Anr induces dnr, Dnr induces target gene), overexpression of dnr should overcome the anr mutation. However, only a small proportion (4–7%) of the Anr-induced activity of the hemN promoter was restored in the presence of dnr in trans, while the presence of anr in the dnr mutant led to the total loss of the observed residual anaerobic hemN expression. From these results, we conclude that the presence of both Anr and Dnr is required for effective anaerobic hemN induction. Surprisingly, the dnr gene in trans in the dnr mutant only partially restored anaerobic hemN expression. The levels of both transcription factors normally present in wild-type cells was clearly disturbed by the overexpression of Dnr, suggesting that both proteins are required at specific stoichiometry for the optimal expression of hemN. Aerobic overexpression of dnr in the dnr mutant and in wild-type cells led to enhanced hemN transcription (Table 2). The failure of Anr to activate dnr under aerobic conditions explains why Dnr is not normally involved in the aerobic hemN expression (Arai et al., 1997). However, overexpression of Dnr under aerobic conditions restored the dual action of Anr and Dnr. The observed Anr–Dnr induction proceeds via the described Anr binding site, as mutation of the Anr binding site totally abolished all observed effects (Table 2).

The P. aeruginosa hemF gene is induced under anaerobic conditions by the dual action of Anr and Dnr

Two promoter fragments of different lengths containing the 5′ region of the hemF gene were fused to lacZ. Expression studies using wild-type P. aeruginosa strain PAO1 revealed an approximately 20-fold induction of hemF expression for both the long and the short promoter fragments (phemFlacZ890–1397 and phemFlacZ974–1284) under anaerobic conditions (Table 3). This result was unexpected, as the enzyme encoded by hemF, coproporphyrinogen III oxidase, requires molecular oxygen for catalysis. Moreover, P. aeruginosa hemF did not complement the S. typhimurium hemF hemN double mutant under anaerobic conditions, although it was capable of complementation aerobically. Expression studies using the anr and dnr mutants demonstrated that both proteins are required for the observed anaerobic induction (Table 3). Co-expression of the anr gene in trans complemented the negative effect of the anr deletion mutant and restored anaerobic induction, and multicopy dnr almost completely restored the effects of a dnr mutant. As was demonstrated for hemN expression, anaerobic expression of hemF in an anr mutant was not restored by the expression of dnr in trans, and anr in trans did not complement a dnr mutation. These results indicate the dual induction of hemF by Anr and Dnr under anaerobic growth conditions. No obvious influence of an anr or a dnr mutation on the aerobic expression of hemF was observed. In agreement with these findings, primer extension experiments revealed the presence of hemF mRNA in an anr mutant under aerobic conditions (Fig. 3, lane 3).

Table 3.  . The influence of oxygen and the redox response regulators Anr and Dnr on the expression of P. aeruginosa hemF. a. All values are the result of three independent experiments performed in triplicate. The β-galactosidase activities are given in Miller units and are calculated as outlined in Experimental procedures.b. The anr and dnr genes were cloned in a vector compatible with the lacZ fusion vector pQF50 and co-transformed where indicated. Vector without insert served as a control.Thumbnail image of

Remarkably, overexpression of either anr or dnr led to an aerobic induction of hemF expression, again demonstrating the aerobic activity of both proteins. Mutation of the Anr binding half-site (TTGAC to CGCGC, phemFlacZ974–1284ΔANR) drastically reduced the observed anaerobic induction in a wild-type genetic background. This result demonstrates that this sequence is important for Anr–Dnr-dependent expression of hemF.

The interpretation of the data obtained for hemN and hemF in combination with previous data is summarized in the model in Fig. 4. Anaerobic induction of P. aeruginosa hemF and hemN requires the induction of dnr by Anr and the subsequent dual action of Anr and Dnr at the Anr binding site of both genes. Aerobic hemN expression is only dependent on Anr, as dnr is not activated by Anr under aerobic conditions. To investigate the biochemical nature of the dual action of both regulators, we are in the process of establishing an in vitro system using recombinant Anr and Dnr.


Figure 4. . Model for the co-ordinated redox-dependent expression of P. aeruginosa hemF and hemN by the combined anaerobic activities of Anr and Dnr and the aerobic activation of hemN by Anr. +, induction of the formation of the indicated protein.

Download figure to PowerPoint

Why is hemF encoding an oxygen-dependent enzyme induced under oxygen-limiting conditions?

Like HEM13 (hemF analogue) from Saccharomyces cerevisiae, the P. aeruginosa hemF gene was induced under conditions of oxygen limitation (Zagorec et al., 1988). As the oxidative decarboxylation of coproporphyrinogen III is the limiting step of haem biosynthesis in yeast under conditions of low oxygen, it was proposed that the cells increased the concentration of the enzyme, which possesses a high affinity for oxygen, to maintain haem biosynthesis (Zagorec et al., 1988). As bacteria already contain active HemN and HemZ under aerobic and anaerobic growth conditions, they are not limited in coproporphyrinogen III oxidase activity. As outlined in the Introduction, E. coli hemF was found to be subject to regulation by different cellular and environmental parameters, including growth phase and oxygen stress (Mukhopadhyay and Schellhorn, 1997). Expression of hemF in bacteria could be used as a second regulatory point of haem biosynthesis for adaptation to changing environmental conditions. As shown for the yeast enzyme, P. aeruginosa HemF has a high affinity for oxygen (< 0.1 μM) (Zagorec et al., 1988). Induction of the enzyme during the switch from aerobic to anaerobic growth would enhance the consumption of residual oxygen in the cell and help to protect oxygen-sensitive anaerobic functions expressed during the transition.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

Bacterial strains and growth conditions

The P. aeruginosa strains PAO1 (wild type), the anr deletion mutant PAO6261 and the dnr deletion strain RM536 have been described previously (Ye et al., 1995; Arai et al., 1997). The hemF mutant PAO9101 and the hemN mutant PAO9102 were generated by the insertion of plasmids that do not replicate freely in P. aeruginosa into the chromosomal copy of the respective gene via homologous recombination. For this purpose, a partial DNA sequence of hemF corresponding to nucleotides 63–801 of the coding region was amplified via polymerase chain reaction (PCR). The primers used are hemFPaXhoI (5′-CGCCCTCGAGGCGGAGGACG-3′) and hemFPaBamHIrev (5′-CATCAGGATCCATTCGCTACGGCCGC-3′), which contained artificial restriction sites that were used to clone the digested PCR product into the appropriate BamHI and XhoI sites of pBluescript-II SK+ (Stratagene) to generate pARP1. A partial hemN sequence corresponding to nucleotides 209–1009 was amplified using the primers hemNPaPstI (5′-GCGCCTGCAGCAAGGTGATC-3′) and hemNPaPvuIrev (5′-CGATCTGGCCGATCGCCGAGAC-3′). The PCR product was digested with PstI and PvuI and cloned into pBR322, which had been cut with the appropriate enzymes, to generate pARP2. Both plasmids were transformed via electroporation into PAO1. Recombinant strains containing the integrated plasmid were selected on Pseudomonas isolation agar (Difco) containing the appropriate antibiotic (200 μg ml−1 tetracycline and 500 μg ml−1 carbenicillin). For the construction of the hemF hemN double mutant, pARP1 (hemF ) was transformed into hemN mutant PAO9102 with selection for carbenicillin and tetracycline on plates containing 50 μg ml−1 haemin. The E. coli strain DH5α was used for routine cloning procedures (Sambrook et al., 1989).

The S. typhimurium mutant TE3006 [env-53 hemN704::Mud-J(b)hemF707::Tn10d-Tet]was used in complementation experiments for the cloning of the hemF and hemN genes from P. aeruginosa (Xu et al., 1992). Media for aerobic and oxygen-limiting conditions have been specified previously (Hungerer et al., 1995a). Plasmid maintenance was ensured by the addition of 50 μg ml−1 tetracycline in E. coli and 300 μg ml−1 carbenicillin in P. aeruginosa. The S. typhimurium mutant was grown with the addition of 5 μg ml−1 tetracycline, 50 μg ml−1 kanamycin and 50 μg ml−1 haemin on minimal medium containing 40 mM glycerol.

Plasmid construction and DNA sequencing

All DNA manipulations were carried out as described previously (Sambrook et al., 1989). The plasmids carrying the different promoter fragments fused to the lacZ gene in the plasmid pQF50 (Farinha and Kropinski, 1990) were constructed as follows. Two primers, each with artificial BamHI sites, were designed that hybridized between positions 974 and 1000 (5′-CGCGGGATCCGAGGACTTCGGGGCCGC-3′) and 1284–1257 (5′-GGCGCGGATCCGGTCTTGCAGGTCGAGC-3′) of the hemF-containing genomic fragment. These were used for PCR amplification of a portion of the hemF gene. The resulting 298 bp fragment was digested with BamHI and ligated in the BamHI site of plasmid pQF50 (Farinha and Kropinski, 1990) to form the plasmid phemFlacZ974–1284. In a second PCR experiment, primers containing artificial NcoI sites spanning the region from position 890 to 916 (5′-GGTAGTAGGTGTCCATGGAGTTCAGGC-3′) and from 1397 to 1373 (5′-CCCTTCTCGATCCATGGGCCGTCGC-3′) were used to produce a 508 bp fragment. The PCR products were digested with NcoI and ligated in pQF50 digested with NcoI to form the plasmid phemFlacZ890–1284. To measure hemN promoter activity, two primers with artificial NcoI sites from position 563–587 (5′-CTGGCCCATGGGGCATTGACGGACG-3′) and 818–792 (5′-GGGCCCATG-

GCTTCGTGGAACTGCACG-3′) were used in PCR to amplify the regulatory region of hemN. The 260 bp product was digested with NcoI and ligated in pQF50 to form the plasmid phemNlacZ563–818. One fusion of the hemN promoter to lacZ was constructed by inserting the NcoI–SalI fragment (1010 bp) from pBKhemN (see below) containing 534 bp of the 5′ regulatory region and the first 476 bp of the hemN gene into the NcoI–SalI sites of pQF50 creating the operon fusion phemNlacZ176–1184. The Anr-boxes in the promoter region of hemF and the hemN gene were mutated via site-directed mutagenesis (5′-TTGAC-GGCC-GTCAA-3′ to 5′-CGCGC-GGCC-GTCAA-3′, phemFlacZ974–1284ΔANR; 5′-TTGAT-ACAA-ATCAA-3′ to 5′-CGCGT-ACAA-ATCAA-3′, phemNlacZ563–818ΔANR) using Amersham's site-directed mutagenesis kit. The orientation and integrity of the constructs were confirmed by complete DNA sequence determination. The plasmids were introduced into P. aeruginosa by electroporation according to the protocol described previously (Diver et al., 1990). Plasmids pHA411 and pHA541Ω containing the anr and dnr genes were stably maintained in the presence of pQF50-derived constructs as described previously (Arai et al., 1997).

Cloning of the P. aeruginosa hemF and hemN genes by complementation of a haem-auxotrophic S. typhimurium mutant

The haem-deficient hemF hemN double mutant of S. typhimurium TE3006 (Xu et al., 1992) was transformed via electroporation with a P. aeruginosa genomic library prepared in pACYC184 (Sawers, 1991; Troup et al., 1995). Complementing plasmids were selected by plating the transformed bacteria on minimal medium M9 (40 mM glycerol, 10 mM KNO3) followed by incubation under aerobic and anaerobic conditions. The hemN-containing plasmids were distinguished from hemF-containing plasmids by their capacity to complement the Salmonella mutant under aerobic and anaerobic conditions. A 4202 bp BamHI fragment was liberated from the P. aeruginosa hemF-containing clone in pACYC184 and ligated into pBluescript SK+ to generate pBKhemF for DNA sequence determination. From the pACYC184 derivative containing hemN, an approximately 4300 bp EcoRI fragment was cloned into pBluescriptSK+ to generate pBKhemN. The DNA sequence of both inserts was determined completely by standard procedures using Sequenase II from Amersham.

Mapping of the 5′ ends of the hemF and hemN mRNAs

Total cellular RNA was prepared from a 20 ml culture of aerobically and anaerobically grown PAO1 and PAO6261 (shifted to anaerobic conditions) cultures as outlined elsewhere (Hungerer et al., 1995a,b). The 5′ ends of mRNAs encoded by hemF and hemN were mapped by the primer extension technique (Boorstein and Craig, 1989) with oligonucleotides complementary to position 1271–1250 (5′-TCTTGCAGGTCGA-

GCAGGTAGG-3′) for hemF and to position 788–769 (5′-CAGCCTATCCTCCACATTGC-3′) for hemN that had been labelled at their 5′ ends with T4 polynucleotide kinase and [γ-32P]-ATP. Primer extensions were performed as outlined previously (Troup et al., 1994; 1995; Hungerer et al., 1995a,b) with the indicated amounts of total cellular RNA.

Physical mapping of the hemF and hemN gene on the P. aeruginosa chromosome

Genomic mapping of both genes was performed as described in detail previously (Römling et al., 1989; Hungerer et al., 1995a,[20]b).

Enzyme assay

β-Galactosidase specific activities were determined in cultures of exponentially growing cells and expressed in Miller units (Sambrook et al., 1989). Aerobic growth conditions were achieved by vigorous shaking of the 25 ml cultures (300 r.p.m. min−1) in 250 ml baffled flasks. Anaerobic growth conditions were achieved as described in detail by Hoffmann et al. (1998). Values of β-galactosidase activity reported are the averages of at least three independent experiments performed in triplicate. The copy number of the lacZ fusion plasmid pQF50 was determined for each experiment by quantitative PCR. Moreover, the low-level lacZ expression detected using the vector pQF50 served as a second background control. The copy number for the experiments was in the range of 12–15 plasmids per cell. These differences were reflected independently by the β-galactosidase activities measured for the vector pQF50. β-Galactosidase activity values were corrected for the small variations in plasmid copy number.

Tetrapyrrole analysis

P. aeruginosa strains were grown aerobically and anaerobically on minimal media as described previously (Hungerer et al., 1995a,b). Harvested cells were sonicated, lyophilized (Jahn et al., 1991) and, subsequently, the total cellular extracts were incubated overnight with methanol–sulphuric acid (95:5, v/v). The esterification mixture was extracted with chloroform, washed twice with distilled water and subsequently dried with sodium sulphate. The mixture was filtered, and chloroform was removed under vacuum. Dried porphyrin methyl esters were redissolved in chloroform and separated via silica gel thin-layer chromatography. The sheets were prerun with chloroform–methanol (120:20, v/v), and samples were separated in petroleum ether (40–60°)–diethyl ether (3:1, v/v). When the front had reached the edge, sheets were run in benzene–ethyl acetate–methanol (85:13.5:1.5, by volume). The different porphyrin methyl esters were eluted from the sheets with chloroform and subsequently quantified spectrophotometrically (Doss, 1974).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

We thank G. Sawers (Norwich, UK) for the gift of the P. aeruginosa genomic library in pACYC184, many helpful discussions of our results and critical reading of the manuscript. We are indebted to D. Haas (Université de Lausanne, Switzerland) for the gift of wild-type strain P. aeruginosa PAO1 and the anr mutant. We thank A.M. Kropinski (Queen's University, Kingston, Canada) for the gift of the lacZ fusion vector pQF50. We thank R.K. Thauer (Max-Planck-Institut, Marburg, Germany) for continuous support. We thank M. Jahn and N. Frankenberg (Universität Freiburg) for critical reading of the manuscript. This work was supported by grants from the Universität Freiburg, the Deutsche Forschungsgemeinschaft (grant 1363/2-1, JA 203/3-3), the Sonderforschungsbereich 395, the Graduiertenkolleg Enzymchemie of the Philipps-Universität Marburg, Fonds der Chemischen Industrie and the Max-Planck-Gesellschaft.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  • 1
    Arai, H., Igarashi, Y., Kodama, T. (1995) Expression of the nir and nor genes for denitrification of Pseudomonas aeruginosa requires a novel CRP/FNR-related transcriptional regulator, DNR, in addition to ANR. FEBS Lett 371: 7376.
  • 2
    Arai, H., Kodama, T., Igarashi, Y. (1997) Cascade regulation of the two CRP/FNR-related transcriptional regulators (ANR and DNR) and the denitrification enzymes in Pseudomonas aeruginosa. Mol Microbiol 25: 11411148.
  • 3
    Anton, I.A. & Coggins, J.R. (1988) Sequencing and overexpression of the Escherichia coli aroE gene encoding shikimate dehydrogenase. Biochem J 249: 319326.
  • 4
    Bell, A.I., Cole, J.A., Busby, S.J.W. (1990) Molecular genetic analysis of an FNR-dependent anaerobically inducible Escherichia coli promoter. Mol Microbiol 4: 17531763.
  • 5
    Boorstein, W.R. & Craig, E.A. (1989) Primer extension analysis of RNA. Methods Enzymol 180: 247369.
  • 6
    Coomber, S.A., Jones, R.M., Jordan, P.M., Hunter, C.N. (1992) A putative anaerobic coproporphyrinogen III oxidase in Rhodobacter sphaeroides. I. Molecular cloning, transposon mutagenesis and sequence analysis of the gene. Mol Microbiol 6: 31553169.
  • 7
    Diver, J.M., Byran, L.E., Sokol, P.A. (1990) Transformation of Pseudomonas aeruginosa by electroporation. Anal Biochem 189: 7579.
  • 8
    Doss, M.O. (1974) Porphyrins and porphyrin precursors. In Clinical Biochemistry. Principles and Methods. Curtius, H., and Roth, M. (eds). New York: de Gruyter.
  • 9
    Doss, M.O. & Philipp-Dornston, W.K. (1971) Porphyrin and heme biosynthesis from endogenous and exogenous δ-aminolevulinic acid in Escherichia coli, Pseudomonas aeruginosa, and Achromobacter metalcaligenes. Hoppe-Seyler's Z Physiol Chem 352: 725733.
  • 10
    Farinha, M.A. & Kropinski, A.M. (1990) Construction of broad-host-range plasmid vectors for easy visible selection and analysis of promoters. J Bacteriol 172: 34963499.
  • 11
    Fleischmann, R.D., Adams, M.D., White, O., Clayton, R.A., Kirkness, E.F., Kerlavage, A.R., et al (1995) Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269: 496512.
  • 12
    De Gier, J.W., Schepper, M., Reijnders, W.N.M., Van Dyck, S.J., Slotboom, D.J., Warne, A., et al (1996) Structural and functional analysis of aa3-type and cbb3-type cytochrome c oxidases of Paracoccus denitrificans reveals significant differences in proton-pump design. Mol Microbiol 20: 12471260.
  • 13
    Green, J., Bennett, B., Jordan, P., Ralph, E.T., Thomson, A.J., Guest, J.R. (1996) Reconstitution of the (4Fe-4S) cluster in FNR and demonstration of the aerobic-anaerobic transcription switch in vitro. Biochem J 316: 887892.
  • 14
    Haas, D., Gamper, M., Zimmermann, A. (1992) Anaerobic control in Pseudomonas aeruginosa, In Pseudomonas, Molecular Biology and Biotechnology. Galli, E., Silver, S., and Witholt, B. (eds). Washington, DC: American Society for Microbiology Press, pp. 177187.
  • 15
    Hippler, B., Homuth, G., Hoffmann, T., Hungerer, C., Schumann, W., Jahn, D. (1997) Characterization of Bacillus subtilis hemN. J Bacteriol 179: 71817185.
  • 16
    Hoffmann, T., Frankenberg, N., Marino, M., Jahn, D. (1998) Ammonification in Bacillus subtilis utilizing dissimilatory nitrite reductase is dependent on resDE. J Bacteriol 180: 186189.
  • 17
    Holloway, B.W., Römling, U., Tümmler, B. (1994) Genomic mapping of Pseudomonas aeruginosa PAO. Microbiology 140: 29072929.
  • 18
    Homuth, G., Heinemann, M., Zuber, U., Schumann, W. (1996) The genes lepA and hemN form a bicistronic operon in Bacillus subtilis. Microbiology 142: 16411649.
  • 19
    Hungerer, C., Troup, B., Römling, U., Jahn, D. (1995a) Regulation of the hemA gene during 5-aminolevulinic acid formation in Pseudomonas aeruginosa. J Bacteriol 177: 14351443.
  • 20
    Hungerer, C., Troup, B., Römling, U., Jahn, D. (1995b) Cloning, mapping and characterization of Pseudomonas aeruginosa heml gene. Mol Gen Genet 248: 375380.
  • 21
    Jacobs, N.J., Jacobs, J.M., Morgan, J.R. (1972) Comparative effect of oxygen and nitrate on protoporphyrin and heme synthesis from D-aminolevulinic acid in bacterial cultures. J Bacteriol 112: 14441445.
  • 22
    Jacobs, N.J., Jacobs, J.M., Mills, B.A. (1973) Role of oxygen in the late steps of heme synthesis in Pseudomonads and Escherichia coli. Enzyme 16: 5056.
  • 23
    Jahn, D., Michelsen, U., Söll, D. (1991) Two glutamyl-tRNA reductase activities in Escherichia coli. J Biol Chem 266: 25422548.
  • 24
    Jahn, D., Verkamp, E., Söll, D. (1992) Glutamyl-transfer RNA: a precursor of heme and chlorophyll biosynthesis. Trends Biochem Sci 17: 215218.
  • 25
    Jayaraman, P.-S., Gaston, K.L., Cole, J.A., Busby, S.J.W. (1988) The nirB promoter of Escherichia coli: location of the nucleotide sequences essential for regulation by oxygen, the FNR protein and nitrite. Mol Microbiol 2: 527530.
  • 26
    Jayaraman, P.S., Cole, J.A., Busby, S.J.W. (1989) Mutational analysis of the nucleotide sequence at the FNR-dependent nirB promoter in Escherichia coli. Nucleic Acids Res 17: 135145.
  • 27
    Ka, J.O., Urbance, J., Ye, R.W., Ahn, T.Y., Tiedje, J.M. (1997) Diversity of oxygen and N-oxide regulation of nitrite reductases in denitrifying bacteria. FEMS Microbiol Lett 156: 5560.
  • 28
    Khoroshilova, N., Beinert, H., Kiley, P. (1995) Association of a polynuclear iron-sulfur center with a mutant FNR protein enhances DNA binding. Proc Natl Acad Sci USA 92: 24992503.
  • 29
    Lazazzera, B.A., Beinert, H., Khoroshilova, N., Kennedy, M.C., Kiley, P.J. (1996) DNA binding and dimerization of the Fe-S-containing FNR protein from Escherichia coli are regulated by oxygen. J Biol Chem 271: 27622768.
  • 30
    Lieb, C., Siddiqui, R.A., Hippler, B., Jahn, D., Friedrich, B. (1998) The Alcaligenes eutrophus hemN gene encoding the oxygen-independent coproporphyrinogen III oxidase is required for heme biosynthesis during anaerobic growth. Arch Microbiol 169: 5260.
  • 31
    Mukhopadhyay, S. & Schellhorn, H. (1997) Identification and characterization of hydrogen peroxide-sensitive mutants of Escherichia coli: Genes that require OxyR for expression. J Bacteriol 179: 330338.
  • 32
    Philipp-Dornston, W.K. & Doss, M.O. (1973) Comparison of porphyrin and heme in various heterotrophic bacteria. Enzyme 16: 5764.
  • 33
    Römling, U., Grothues, D., Bautsch, W., Tümmler, B. (1989) A physical genome map of Pseudomonas aeruginosa PAO. EMBO J 8: 31993206.
  • 34
    Sambrook, J., Fritsch, E.F., Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • 35
    Sawers, R.G. (1991) Identification and molecular characterization of a transcriptional regulator from Pseudomonas aeruginosa PAO1 exhibiting structural and functional similarity to the FNR protein of Escherichia coli. Mol Microbiol 5: 14691481.
  • 36
    Sawers, R.G., Zehelein, F., Böck, A. (1988) Two-dimensional gel electrophoretic analysis of Escherichia coli proteins: influence of various anaerobic growth conditions and the fnr gene product on cellular protein composition. Arch Microbiol 149: 240244.
  • 37
    Seehra, J.S., Jordan, P.M., Akhtar, M. (1982) Anaerobic and aerobic coproporphyrinogen III oxidases of Rhodopseudomonas sphaeroides. Biochem J 269: 709718.
  • 38
    Tait, G.H. (1969) Coproporphyrinogenase activity in extracts from Rhodopseudomonas sphaeroides. Biochem Biophys Res Commun 37: 116122.
  • 39
    Tait, G.H. (1972) Coproporphyrinogenase activities in extracts of Rhodopseudomonas sphaeroides and Chromatium strain D. Biochem J 128: 11591169.
  • 40
    Thorn, J.M., Barton, J.D., Dixon, N.E., Ollis, D.L., Edwards, K.J. (1995) Crystal structure of Escherichia coli Qor quinone oxidoreductase complexed with NADPH. J Mol Biol 249: 785799.
  • 41
    Tomb, J.F., White, O., Kerlavage, A.R., Clayton, R.A, Sutton, G.G., Fleischmann, R.D., et al (1997) The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 386: 539547.
  • 42
    Troup, B., Jahn, M., Hungerer, C., Jahn, D. (1994) Isolation of the hemF operon containing the gene for the Escherichia coli aerobic coproporphyrinogen III oxidase by in vivo complementation of a yeast HEM13 mutant. J Bacteriol 176: 673680.
  • 43
    Troup, B., Hungerer, C., Jahn, D. (1995) Cloning and characterization of the Escherichia coli hemN gene encoding the oxygen-independent coproporphyrinogen III oxidase. J Bacteriol 177: 33263331.
  • 44
    Tyson, K., Metheringham, R., Griffiths, L., Cole, J. (1997) Characterization of Escherichia coli K-12 mutants defective in formate-dependent nitrite reduction: essential roles for hemN and the menFDBCE operon. Arch Microbiol 168: 403411.
  • 45
    Winteler, H.V. & Haas, D. (1996) The homologous regulators ANR of Pseudomonas aeruginosa and FNR of Escherichia coli have overlapping but distinct specificities for anaerobically inducible promoters. Microbiology 142: 685693.
  • 46
    Xu, K. & Elliott, T. (1993) An oxygen-dependent coproporphyrinogen oxidase encoded by the hemF gene of Salmonella typhimurium. J Bacteriol 175: 49904998.
  • 47
    Xu, K. & Elliott, T. (1994) Cloning, DNA sequence, and complementation analysis of the Salmonella typhimurium hemN gene encoding a putative oxygen-independent coproporphyrinogen III oxidase. J Bacteriol 176: 31963203.
  • 48
    Xu, K., Delling, J., Elliott, T. (1992) The genes required for heme synthesis in Salmonella typhimurium include those encoding alternative functions for aerobic and anaerobic coproporphyrinogen oxidation. J Bacteriol 174: 39533963.
  • 49
    Ye, R.W., Haas, D., Ka, J.-O., Krishnapillai, V., Zimmermann, A., Baird, C., et al (1995) Anaerobic activation of the entire denitrification pathway in Pseudomonas aeruginosa requires Anr, an analog of Fnr. J Bacteriol 177: 36063609.
  • 50
    Zagorec, M., Buhler, J.M., Treich, I., Keng, T., Guarente, L., Labbe-Bois, R. (1988) Isolation, sequence, and regulation by oxygen of yeast HEM13 gene coding for coproporphyrinogen oxidase. J Biol Chem 263: 97189724.
  • 51
    Zeilstra-Ryalls, J.H. & Kaplan, S. (1995) Regulation of the 5-aminolevulinic acid synthesis in Rhodobacter sphaeroides 2.4.1. genetic basis of mutant H-5 auxotrophy. J Bacteriol 177: 27602768.
  • 52
    Zimmermann, A., Reimann, C., Galimand, M., Haas, D. (1991) Anaerobic growth and cyanide synthesis of Pseudomonas aeruginosa depend on anr, a regulatory gene homologous with fnr of Escherichia coli. Mol Microbiol 5: 14831490.