ABSTRACT
DNA microarrays were used to examine the transcriptional response of Pseudomonas aeruginosa to anaerobiosis and nitrate. In response to anaerobic growth, 691 transcripts were differentially expressed. Comparisons of P. aeruginosa grown aerobically in the presence or the absence of nitrate showed differential expression of greater than 900 transcripts.
Pseudomonas aeruginosa is capable of anaerobic growth by anaerobic respiration with nitrate, nitrite, or nitrous oxide as the terminal electron acceptor (8, 9) or by generating ATP from arginine catabolism (16, 26). Biofilms display hypoxic gradients, and biofilm formation is enhanced under oxygen limitation (7, 28, 31). P. aeruginosa grows as a biofilm in the anoxic environment of the lower airway mucus plugs in cystic fibrosis patients (7, 28).
Except with genes involved in denitrification (1, 29) and several other genes (12, 14, 15, 20), little is known about anaerobic gene expression in P. aeruginosa. The present study used microarrays to identify genes differentially expressed by P. aeruginosa in response to anaerobiosis and nitrate.
Differential gene expression in response to anaerobic growth.
Growth curves for P. aeruginosa PAO1 cultured aerobically or anaerobically were generated to establish the RNA sampling points (27) (supplemental Fig. A at http://www.urmc.rochester.edu/smd/mbi/bhi/ ). Total RNA from three independent P. aeruginosa PAO1 aerobic or anaerobic cultures was isolated and processed as previously described (27). RNA integrity was assessed by reverse transcriptase PCR (RT-PCR) using primers specific for pilA, and purity was confirmed by PCR. Processing of RNA, microarray data generation, analysis, and validation by quantitative RT-PCR were performed as previously described (27) (supplemental Table C at http://www.urmc.rochester.edu/smd/mbi/bhi/ ). All RT-PCR data were normalized using PA4232, as the expression does not change under the conditions examined (our data and reference 27).
The expression of total transcripts (72% to 80%) was comparable to that seen in other P. aeruginosa microarray studies (19, 27). A total of 691 transcriptional changes, representing approximately 12% of the genome, resulted in statistically significantly different levels of expression in response to anaerobic growth, with 245 transcripts up-regulated and 446 transcripts down-regulated (supplemental Table A at http://www.urmc.rochester.edu/smd/mbi/bhi/ ). Those transcripts (n = 153) demonstrating a threefold or higher change are listed in Table 1 and grouped into functional categories (supplemental Fig. B at http://www.urmc.rochester.edu/smd/mbi/bhi/ ).
Differentially expressed transcripts between cultures grown aerobically with nitrate and cultures grown anaerobically with nitratea
We found numerous genes with expression patterns consistent with anaerobic growth and previous reports, such as the repression of napBAD, napF (17, 32), hcnAC (20), flgB, flgE, flgI, flgL, fliC, fliD, fleS, fleR, fliE, fliF, fliM, flhA, and flhF (10) and increased expression of hemF (22) (supplemental Table A at http://www.urmc.rochester.edu/smd/mbi/bhi/ ).
Many genes involved in quorum sensing (lasR, lasA, lasB, rhlR, rhlI, rhlA, and mvfR) were repressed under oxygen limitation. Consistent with reduced mvfR expression, transcripts involved in the biosynthesis of the Pseudomonas quinolone signal (pqsB and pqsE) and anthranilate synthase components I and II (phnAB) (4) were decreased.
Genes involved in cytochrome c maturation, ccmB, ccmC, ccmE, and ccmF, were up-regulated under anaerobic conditions, which is consistent with observations of Escherichia coli (24). Additionally, transcript levels for PA5491 (a probable cytochrome) were increased, suggesting that this previously uncharacterized cytochrome may play a role in anaerobic respiration. Transcript levels for several other putative cytochromes (PA0918, PA1555, PA1556, PA2266, PA2482, PA3331, PA4571, and PA4619) were repressed, suggesting that they may not be required for anaerobic respiration.
Our data implicated many novel genes in anaerobic growth. There were 284 transcripts classified as genes encoding hypothetical proteins differentially expressed under anaerobic conditions. Several genes which play a role in virulence (PA0930) (21) or biofilm formation (PA2128, PA2129, and PA2130) (25) were induced during anaerobic growth.
Differential gene expression in response to nitrate.
In contrast to previous studies (32), no significant changes were observed (Table 1) for most of the genes involved in denitrification (nar, nir, nos, and nor), suggesting that nitrate may induce their expression. This is supported by the capacity of P. aeruginosa for aerobic denitrification (5) and aerobic Nir activity when nitrate is available (1, 13). Many denitrification genes are influenced by the presence of an N-oxide (2). To investigate this, microarray analysis using RNA from cultures grown aerobically in the presence or the absence of nitrate was performed. Nearly 18% of the genome (919 transcripts; 415 transcripts induced and 504 repressed) exhibited differential expression in response to nitrate (supplemental Table B at http://www.urmc.rochester.edu/smd/mbi/bhi/ ). The 266 genes demonstrating a threefold or greater change are listed in Table 2. Functional categories are shown in supplemental Fig. C at http://www.urmc.rochester.edu/smd/mbi/bhi/ .
Differentially expressed transcripts between cultures grown aerobically without nitrate and cultures grown aerobically with nitratea
The transcription of narI was up-regulated by the presence of nitrate, while narG, narH, and narJ were not found to be statistically differentially expressed. napB, napA, napD, nosRDFYL, and nirS exhibited increased expression in the presence of nitrate. Other genes encoding proteins either implicated (PA0513, PA0514, PA0516, PA0518, and PA0521) or known to be involved (nirL, nirM, nirN, and nirF) in the processing of respiratory system components were up-regulated in response to nitrate. Our results indicate that nitrate alone is sufficient to induce the expression of many enzymes involved in denitrification regardless of the presence or absence of oxygen and explain the apparent lack of differential expression of some of these genes in our anaerobic experiments.
The expression of a number of genes involved in the production of virulence factors of P. aeruginosa were influenced by the addition of nitrate. For example, while mexAB and rhlAB expression were repressed by nitrate, mexF was up-regulated. Transcription of a recently described chemotaxis cluster (PA0174-0179) found to be required for optimal chemotaxis (6) and aerotaxis (11) was induced by the nitrate. The R-type pyocins (PA0614 to PA0646) (18) were induced by nitrate. We also observed differential expression of 306 transcripts which currently do not have defined functions. Importantly, a couple of transcripts (PA0459 and PA5167) have been previously found to be required for lung infection (21).
Consistent with previous reports for Escherichia coli (23), Shewanella oneidensis, and Bacillus subtilis (3, 30), we found that P. aeruginosa significantly changes its transcriptional profile in the absence of oxygen or in the presence of nitrate. It should be noted that our results are biased towards using nitrate as the terminal electron acceptor, and using other nitric oxides or arginine may affect other genes.
Our results provide a global view of oxygen-regulated gene expression in P. aeruginosa and illustrate the complex regulation of anaerobic metabolism in this organism. Changes in genes encoding virulence factors and quorum-sensing components implicate altered pathogenic pathways during anaerobic growth. Our identification of a substantial number of genes encoding proteins of unknown function should contribute to further annotation of the genome and provide impetus for further research on the role of these genes in P. aeruginosa physiology and metabolism.
ACKNOWLEDGMENTS
We thank A. Brooks, K. Miller, L. Ascroft, and K. Wahowski at the Microarray Core Facility in The Functional Genomics Center at the University of Rochester for technical support and assistance with the quantitative RT-PCR and Cystic Fibrosis Foundation Therapeutics, Inc., for subsidizing the P. aeruginosa Affymetrix GeneChip arrays.
This work was supported by grants IGLEWS00V0 and IGLEWS03FG0 to B.H.I., L.P., and C.G.H. from Cystic Fibrosis Foundation Therapeutics and grant R37AI33713 to B.H.I from the NIH. M.J.F is supported by an NIH fellowship (F32AI056825).
FOOTNOTES
- Received 15 October 2004.
- Returned for modification 24 November 2004.
- Accepted 2 February 2005.
- Copyright © 2005 American Society for Microbiology
