Bacteria are ubiquitous and diverse organisms. While the vast majority of bacteria live harmlessly in the environment, a select few are able to colonize and cause disease in humans and other animals. In order to adapt to growth in the human body, bacterial pathogens have evolved ways to evade detection by the immune system. In a variety of pathogens, specific instances of large outer membrane proteins being lost relative to their non-pathogenic relatives has been observed. It is possible that this is due to selective pressure to lose outer membrane proteins for the purpose of immune evasion. Here, we look at the genome dynamics of outer membrane proteins in members of the genus Pseudomonas and the family Bacteriaceae. Our analysis shows that across closely related species, the percent of the genome encoding for outer membrane proteins is stable, with the exception of Shigella, which has significantly less outer membrane than the closely related Escherichia coli. Since Shigella was the only bacterial pathogen that we looked at that lost outer membrane proteins relative to other closely related bacteria, we wondered whether pathogens transcriptionally repress outer membrane proteins during infection. By analyzing transcriptome datasets from Shigela flexneri, enterotoxigenic E. coli, and Pseudomonas aeruginosa, we saw that all three pathogens had a bias towards repressing outer membrane proteins.
In order to colonize their host and establish infection, pathogenic bacteria must have mechanisms to evade or withstand the host immune system. To this end, bacterial pathogens have evolved a variety of strategies. Some invading bacteria have been shown to alter their membrane surface charge to repel cationic antimicrobial peptides1. Other bacteria are known to modify their capsular polysaccharides in such a way that they resemble host glycans2. Despite the immune evasion strategies just mentioned, and many others not mentioned3,4, the host is often able to generate an immune response to invading bacteria. In many cases, the host immune system is responding to outer membrane proteins that are exposed on the bacterial surface5–9.
Interestingly, it has been observed that pathogenic bacteria undergo gene loss at a faster rate than closely related non-pathogenic bacteria10–14. In various pathogens, the loss of genes encoding for outer membrane proteins that their non-pathogenic relatives still express has been documented15–17. Presumably, it is possible that this gene loss may be related to a selective pressure on pathogens to lose outer membrane proteins for immune evasion. Despite many specific examples of pathogens losing genes for outer membrane proteins, whether large scale genome loss of outer membrane proteins is a strategy employed by bacteria for immune evasion has not been studied.
In this project, we will look at the genome dynamics of outer membrane proteins amongst select pathogenic and non-pathogenic gram-negative bacteria. Specifically, we will focus on members of the genus Pseudomonas and the family Enterobacteriaceae. For the purpose of our analysis, we will utilize the subcellular localization tool PSORTb18. In addition to genome dynamics, we will look at the transcriptional adaptation of pathogens to infection, specifically focusing on how it relates to outer membrane proteins.
Prediction of Subcellular localization:
Predictions of the subcellular localization for proteins in bacterial genome were obtained from the PsortB database. For organisms that were not in the database, annotated genomes were downloaded from genbank in fasta format and run through PsortB18 to generate tables of predicted localization. Pseudomonas aeruginosa genome annotations were obtained from Pseudomonas.com.
Determination of genomic outer membrane protein percentage in different bacterial genera:
Genome analysis was performed using a script (script 1) that iterates through folders containing PsortB files, and calculates the number of outer membrane, extracellular, and total proteins. Comparative analysis was performed within members of Bacteriaceae family: E. coli, Shigella spp, Salmonella spp, and Enterobacter spp; and within members of the Pseudomonas genus: Pseudomonas aeruginosa, Pseudomonas fluorescens and Pseudomonas putida. In total, Psortb data from 107 E. coli, 14 Shigella, 23 Salmonella, 15 Enterobacter, and 44 Pseudomonas genomes were analyzed.
Analysis of outer membrane protein frequencies from gene expression data:
To analyze how outer membrane protein expression changes during infection, published RNA- seq and microarray data comparing Shigella flexneri21, enterotoxigenic E. col22i, and Pseudomonas aureginosa27gene expression during in vitro growth to growth in infection models were obtained. Genes were considered significantly and differentially regulated under infection according to the dataset authors’ own criteria. Transcriptome analysis was performed using a Python script (script 2) that creates two separate lists containing identifiers (either systematic gene name, or locus tag) for all of the downregulated and upregulated genes. Then, the script goes through the genbank files of the organism of interest, and writes two separate fasta files, one containing only the downregulated genes and one containing only the upregulated genes. Once obtained, the fasta files for upregulated and downregulated genes were uploaded to Psortb and outer membrane proteins were analyzed by Script 1, as previously described.28
Amongst select members of Enterobacteriaceae — E. coli, Salmonella, and Enterobacter — the proportion of total genes encoding proteins that are predicted to be localized to the outer membrane is between 2.09% and 2.16 % of all genes (Figure 1A). However, in Shigella, a genus belonging to the same family, the percentage of genes encoding proteins that are predicted to be localized to the outer membrane is significantly lower at 1.80% (Figure 1A). This suggests that relative to other pathogenic and non-pathogenic members of Enterobacteriaceae, Shigella, has likely lost a comparatively higher percentage of outer membrane proteins.
Shigella is closely related to E. coli. In clinical settings, infections caused by Shigella species and pathogenic E. coli strains are often indistinguishable19, and while typically considered to be its own genus, it has long been suggested that Shigella should actually belong to the species E. coli20. Because of these similarities, we wanted to look at how the prevalence of outer membrane proteins in Shigella relates to that of other pathogenic E. coli. Interestingly, even compared to other pathogenic E. coli strains, Shigella has a substantially lower proportion of outer membrane proteins (Figure 1B).
We suspect that part of the reason that evidence for a change in outer membrane protein frequency could be seen in Shigella, but not other pathogenic E. coli is due to their lifestyles. Most pathogenic E. coli can live in multiple environments. Many have the ability to colonize various different animals, while others will asymptomatically colonize one part of the human body, but symptomatically colonize another. This varies greatly from Shigella, which has no known environmental reservoir, and will only colonize the colon of humans and closely related non-human primates. Past work has suggested that due to its restricted niche, Shigella is undergoing gene loss at a faster rate than other pathogenic E. coli13. Our data is consistent with this finding and seems to imply that not only has Shigella lost genes relative to non-pathogenic E. coli, but that it has lost genes in the outer membrane at a higher rate than some of the genes localized to other parts of the cell.
Besides gene loss, another mechanism by which pathogens can suppress the amount of outer membrane proteins available for immune recognition is by repressing them transcriptionally during infection. For bacteria such as certain pathogenic E. coli strains that have multiple niches, a broad host range, or both, it is possible that repression of transcription is more favorable to gene loss, because of the flexibility it allows for. To asses this possibility, we analyzed gene expression data from Shigella21 and enterotoxigenic E. coli (ETEC)22 during infection. When we predicted the localization of genes that were differentially regulated during Shigella infection, we saw that for most cell envelope genes, more of the genes were up regulated during infection than down regulated (Figure 2A). However, for the outer membrane proteins, slightly more genes were downregulated than upregulated (Figure 2A). Considering that across the whole dataset, 175 more genes were up regulated than down regulated, we binned the genes into up and down regulated genes and looked at outer membrane proteins as a percentage of each group (Figure 3). When we analyzed the data in this way, we saw that out of all the down regulated genes, 2.19% encoded outer membrane proteins. On the other hand, out of the up regulated genes, only 1.87% percent were outer membrane. The entire genome had 1.94% outer membrane proteins. Together, these data suggest that amongst down regulated genes, there is a slight bias for them to encode outer membrane proteins. When analyzed in the same way, expression data from ETEC showed a similar pattern. More genes encoding proteins that localize to the cytoplasmic membrane and extracellular environment were up regulated than down regulated during infection (Figure 2B). However, more genes encoding proteins that localize to the periplasm and outer membrane were down regulated. In this dataset, 131 more genes were up regulated than down regulated. Taking this into consideration, when the data was binned into up and down regulated genes, it was seen that of the down regulated genes 3.37% were outer membrane proteins, while of the up regulated genes, only 2.11% were. This is compared to 2.19% across the whole ETEC genome.
Because of the genomic and transcriptomic changes that we saw in the genes encoding for outer membrane proteins of gastrointestinal enteric pathogens, we wondered whether similar changes could be seen in other pathogenic and non-pathogenic bacteria. For this analysis, we decided to look at members of the genus Pseudomonas.
Pseudomonas aeruginosa is a well-studied opportunistic pathogen that is considered to have one of the largest bacterial genomes23. Its large genome composed of an unusually high number of regulatory genes, which in turn enables it to engage in complex lifestyles capable of adapting to various environmental challenges. Along with being a nosocomial opportunistic pathogen, responsible for many infections of immunocompromised patients24, it is also found living environmentally in soil25 and water. On the other hand, other species of the Pseudomonas genus, such as P. putida, P. fluorescens, and P. syringae, are mostly found in the environment. These organisms, while sometimes able to cause infection in immunocompromised patients, do so much less frequently than P. aeruginosa26. When we performed a comparative genomic analysis on various strains of P. aeruginosa and other Pseudomonas species, we saw that P. aeruginosa had a slightly higher percentage of predicted outer membrane proteins than its less pathogenic relatives (Figure 4). It is tempting to speculate that this higher number of genes encoding outer membrane proteins, contributes to the ability of P. aeruginosa to successfully grow in a large number of distinct environments.
Given the large number genes encoding for outer membrane proteins, we hypothesized that during infection many of these genes would be downregulated during infection in order to avoid host immunity. Transcriptome data27 was analyzed to test this hypothesis. When looking at the sub cellular localization of differentially expressed genes, most locations showed that more genes were upregulated than downregulated (Figure 5A). However, this trend was not observed for genes encoding outer membrane and extracellular proteins. A similar amount of outer membranes were up and downregulated. However, since 198 more genes were upregulated than downregulated during infection, outer membrane proteins comprised a substantially larger percentage of the downregulated genes (4.5%) than the upregulated genes (3.3%) (Figure 5B). This suggests that during infection, there is a bias for P. aeruginosa to repress genes encoding for outer membrane proteins.
1. Bacterial Evasion of Host Antimicrobial Peptide Defenses. Available at: https://www-ncbi-nlm-nih-gov.ezproxy.lib.utexas.edu/pmc/articles/PMC4804471/. (Accessed: 28th April 2019)
2. Masquerading microbial pathogens: Capsular polysaccharides mimic host-tissue molecules. Available at: https://www-ncbi-nlm-nih-gov.ezproxy.lib.utexas.edu/pmc/articles/PMC4120193/. (Accessed: 28th April 2019)
3. The complex interactions of bacterial pathogens and host defenses – ScienceDirect. Available at: https://www-sciencedirect-com.ezproxy.lib.utexas.edu/science/article/pii/S1369527413000234?via%3Dihub. (Accessed: 28th April 2019)
4. Bacterial Immune Evasion through Manipulation of Host Inhibitory Immune Signaling. Available at: https://www-ncbi-nlm-nih-gov.ezproxy.lib.utexas.edu/pmc/articles/PMC4351076/. (Accessed: 28th April 2019)
5. Serum antibody response to capsular polysaccharide, outer membrane, and lipooligosaccharide in children with invasive Haemophilus influenzae type b… – PubMed – NCBI. Available at: https://www-ncbi-nlm-nih-gov.ezproxy.lib.utexas.edu/pubmed/3501433. (Accessed: 28th April 2019)
6. Antibody response to outer membrane protein of nontypeable Haemophilus influenzae in otitis-prone children. – PubMed – NCBI. Available at: https://www-ncbi-nlm-nih-gov.ezproxy.lib.utexas.edu/pubmed/8429433. (Accessed: 28th April 2019)
7. Characterization of the Humoral Immune Response to Chlamydia Outer Membrane Protein 2 in Chlamydial Infection | Clinical and Vaccine Immunology. Available at: https://cvi-asm-org.ezproxy.lib.utexas.edu/content/10/1/103.long. (Accessed: 28th April 2019)
8. Helminen, M. E. et al. Human Immune Response against Outer Membrane Proteins of Moraxella (Branhamella) catarrhalis Determined by Immunoblotting and Enzyme Immunoassay. CLIN DIAGN LAB IMMUNOL 2, 5 (1995).
9. Identification of the Immunogenic Outer Membrane Protein A Antigen of Haemophilus parasuis by a Proteomics Approach and Passive Immunization with Monoclonal Antibodies in Mice. Available at: https://www-ncbi-nlm-nih-gov.ezproxy.lib.utexas.edu/pmc/articles/PMC3187029/. (Accessed: 28th April 2019)
10. Antivirulence Genes: Insights into Pathogen Evolution through Gene Loss. Available at: https://www-ncbi-nlm-nih-gov.ezproxy.lib.utexas.edu/pmc/articles/PMC3497401/. (Accessed: 28th April 2019)
11. Genes Lost and Genes Found: Evolution of Bacterial Pathogenesis and Symbiosis | Science. Available at: https://science.sciencemag.org/content/292/5519/1096?casa_token=EOtF6QQPM4gAAAAA:7-4_yBl356qYxihO_cEaKGevK37O-ciGg3RdlNit_hm_-kI3PVBXuh_J0Y4ZC5V2P5ei917C0Qz5Cxw. (Accessed: 28th April 2019)
12. Microbial Minimalism: Genome Reduction in Bacterial Pathogens – ScienceDirect. Available at: https://www-sciencedirect-com.ezproxy.lib.utexas.edu/science/article/pii/S0092867402006657. (Accessed: 28th April 2019)
13. Reduced selection leads to accelerated gene loss in Shigella | Genome Biology | Full Text. Available at: https://genomebiology-biomedcentral-com.ezproxy.lib.utexas.edu/articles/10.1186/gb-2007-8-8-r164. (Accessed: 28th April 2019)
14. Why Might Bacterial Pathogens Have Small Genomes?: Trends in Ecology & Evolution. Available at: https://www.cell.com/trends/ecology-evolution/fulltext/S0169-5347(17)30237-9. (Accessed: 28th April 2019)
15. Lawrence, J. G. Common themes in the genome strategies of pathogens. Curr. Opin. Genet. Dev. 15, 584–588 (2005).
16. Parkhill, J. et al. Comparative analysis of the genome sequences of Bordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica. Nat. Genet. 35, 32–40 (2003).
17. Nakata, N. et al. The absence of a surface protease, OmpT, determines the intercellular spreading ability of Shigella: the relationship between the ompT and kcpA loci. Mol. Microbiol. 9, 459–468 (1993).
18. PSORTb 3.0: improved protein subcellular localization prediction with refined localization subcategories and predictive capabilities for all prokaryotes. Available at: https://www-ncbi-nlm-nih-gov.ezproxy.lib.utexas.edu/pmc/articles/PMC2887053/. (Accessed: 28th April 2019)
19. Khot, P. D. & Fisher, M. A. Novel Approach for Differentiating Shigella Species and Escherichia coli by Matrix-Assisted Laser Desorption Ionization–Time of Flight Mass Spectrometry. J. Clin. Microbiol. 51, 3711–3716 (2013).
20. Lan, R. & Reeves, P. R. Escherichia coli in disguise: molecular origins of Shigella. Microbes Infect. 4, 1125–1132 (2002).
21. Lucchini, S., Liu, H., Jin, Q., Hinton, J. C. D. & Yu, J. Transcriptional Adaptation of Shigella flexneri during Infection of Macrophages and Epithelial Cells: Insights into the Strategies of a Cytosolic Bacterial Pathogen. Infect. Immun. 73, 88–102 (2005).
22. Crofts, A. A. et al. Enterotoxigenic E. coli virulence gene regulation in human infections. Proc. Natl. Acad. Sci. 115, E8968–E8976 (2018).
23. Freschi, L. et al. The Pseudomonas aeruginosa Pan-Genome Provides New Insights on Its Population Structure, Horizontal Gene Transfer, and Pathogenicity. Genome Biol. Evol. 11, 109–120 (2019).
24. Moradali, M. F., Ghods, S. & Rehm, B. H. A. Pseudomonas aeruginosa Lifestyle: A Paradigm for Adaptation, Survival, and Persistence. Front. Cell. Infect. Microbiol. 7, (2017).
25. Kung, V. L., Ozer, E. A. & Hauser, A. R. The accessory genome of Pseudomonas aeruginosa. Microbiol. Mol. Biol. Rev. MMBR 74, 621–641 (2010).
26. Kittinger, C. et al. Antibiotic Resistance Patterns of Pseudomonas spp. Isolated from the River Danube. Front. Microbiol. 7, (2016).
27. Cornforth, D. M. et al. Pseudomonas aeruginosa transcriptome during human infection. Proc. Natl. Acad. Sci. 115, E5125–E5134 (2018).
28. Hata, H. et al. Phylogenetics of family Enterobacteriaceae and proposal to reclassify Escherichia hermannii and Salmonella subterranea as Atlantibacter hermannii and Atlantibacter subterranea gen. nov., comb. nov. Microbiol. Immunol. 60, 303–311 (2016).