http://orcid.org/0000-0002-7257-1051
Figures
Abstract
Moraxella catarrhalis is a human-adapted, opportunistic bacterial pathogen of the respiratory mucosa. Although asymptomatic colonization of the nasopharynx is common, M. catarrhalis can ascend into the middle ear, where it is a prevalent causative agent of otitis media in children, or enter the lower respiratory tract, where it is associated with acute exacerbations of chronic obstructive pulmonary disease in adults. Phase variation is the high frequency, random, reversible switching of gene expression that allows bacteria to adapt to different host microenvironments and evade host defences, and is most commonly mediated by simple DNA sequence repeats. Bioinformatic analysis of five closed M. catarrhalis genomes identified 17 unique simple DNA sequence repeat tracts that were variable between strains, indicating the potential to mediate phase variable expression of the associated genes. Assays designed to assess simple sequence repeat variation under conditions mimicking host infection demonstrated that phase variation of uspA1 (ubiquitous surface protein A1) from high to low expression occurs over 72 hours of biofilm passage, while phase variation of uspA2 (ubiquitous surface protein A2) to high expression variants occurs during repeated exposure to human serum, as measured by mRNA levels. We also identify and confirm the variable expression of two novel phase variable genes encoding a Type III DNA methyltransferase (modO), and a conserved hypothetical permease (MC25239_RS00020). These data reveal the repertoire of phase variable genes mediated by simple sequence repeats in M. catarrhalis and demonstrate that modulation of expression under conditions mimicking human infection is attributed to changes in simple sequence repeat length.
Citation: Tan A, Blakeway LV, Taha, Yang Y, Zhou Y, Atack JM, et al. (2020) Moraxella catarrhalis phase-variable loci show differences in expression during conditions relevant to disease. PLoS ONE 15(6): e0234306. https://doi.org/10.1371/journal.pone.0234306
Editor: Sean Reid, Ross University School of Medicine, DOMINICA
Received: September 29, 2019; Accepted: May 22, 2020; Published: June 18, 2020
Copyright: © 2020 Tan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This work was supported by the Australian National Health and Medical Research Council (NHMRC) [Career Development Fellowship to KLS; Project Grant 1099279 to KLS, JMA], the Garnett Passe and Rodney Williams Memorial Foundation [Fellowship to AT and Grant in Aid Supplementation to KLS and JMA], and an Australian Government Research Training Program Scholarship to LVB. The funders had no role in study design, data collection and analysis, or preparation of the manuscript. https://www.nhmrc.gov.au/funding https://gprwmf.org.au/ https://www.education.gov.au/research-training-program
Competing interests: The authors have declared that no competing interests exist.
Introduction
Moraxella catarrhalis is a Gram negative, human-adapted, opportunistic bacterial pathogen of the respiratory tract. While commonly isolated from the nasopharynx as an asymptomatic colonizer [1], M. catarrhalis is also a prevalent aetiological agent of otitis media (OM) in children and exacerbations of chronic obstructive pulmonary disease (COPD) in the elderly [2]. OM is the most common bacterial infectious disease of childhood, and is particularly prevalent in children under five in Oceania [3], with Indigenous Australian children among the most severely affected [4]. Approximately 20% of children suffer recurring infections [5], and associated complications such as repeated tympanic membrane perforation lead to acute or chronic hearing loss [6, 7]. COPD is the fourth most common cause of death worldwide [8] and repeated exacerbations due to bacterial infections lead to progressive loss of lung function and dramatically increase the risk of mortality [9]. M. catarrhalis infection accounts for approximately 20% of cases of OM [10] and 10% of exacerbations of COPD [11]. Despite the significant burden of M. catarrhalis associated disease, no proposed vaccine candidates have progressed to clinical trial [12]. The development of a vaccine against M. catarrhalis has been hindered by the lack of a suitable animal model, identification of correlates of protection, and identification of vaccine candidates that are immunogenic, conserved, and stably expressed [12]. It is this last feature that this study primarily addresses, as a number of respiratory pathogens possess a highly mutable genome, which contributes to their virulence and complicates selection of stably expressed vaccine candidates.
Phase variation is the high frequency, random, reversible switching of gene expression that allows bacteria to adapt to different host microenvironments and evade host defences, and is often mediated by simple DNA simple sequence repeats (SSRs) [13]. Switching rates of phase variable genes are approximately one million times more frequent than the background rate of mutations (i.e., phase variation switching occurs at a rate of 10−2 to 10−5 compared with point mutation rates of ∼10−8 to 10−11 for the spontaneous acquisition of antibiotic resistance) [13]. In both cases, mutations occur independently of their utility, and the environment selects for variants that facilitate bacterial growth and/or survival. To date, six loci containing SSRs have been identified in M. catarrhalis; these include four genes encoding outer membrane proteins (mid/hag, uspA1 and uspA2, or the mutually exclusive uspA2H variant), and two encoding cytoplasmic localized Type III DNA methyltransferases (modM and modN) [reviewed in 14].
UspA1 is involved in biofilm formation [15], and adherence to multiple epithelial cell types [16, 17] and extracellular matrix components [18, 19]. Reversible gain or loss of repeat units in a G(n) SSR tract located upstream of the uspA1 open reading frame (ORF) alters transcription of the gene, resulting in a graded switching of expression (high-to-low) that alters the ability of M. catarrhalis to adhere to epithelial cells in vitro [20]. Similarly, phase variation of uspA2 occurs at the transcriptional level through variation in length of a 5′-AGAT(n)-3′ tetranucleotide SSR upstream of the uspA2 ORF, and the length of the uspA2 repeat tract affects resistance of M. catarrhalis to complement mediated killing [21]. In the other identified phase variable genes in M. catarrhalis, SSRs are located within the ORF close to the 5′ end and cause ‘on-off’ switching of gene expression. mid/hag is involved in serum resistance and binding to vibronectin [21], and agglutination of M. catarrhalis cells, red blood cells and immunoglobulin D binding [22, 23]. Phase variation of mid/hag is mediated by a G(n) SSR, with on-off switching altering auto-aggregation and adherence [23, 24]. The uspA2H ORF contains a A(n) SSR that alters auto-aggregation, serum resistance, and adherence phenotypes [25]. We have previously shown that 5′-CAAC(n)-3′ tetranucleotide SSRs are responsible for phase variation of both the modM and modN genes [26], and differential methylation of the genome caused by ModM phase variation epigenetically regulates the switching of expression of multiple genes in a phasevarion [27, 28], further complicating the identification of stably expressed genes in M. catarrhalis. Further to this, we have also identified a third potentially phase variable Type III DNA methyltransferase, modO, which contains a 5′-CAACG(n)-3′ pentanucleotide repeat tract upstream of its ORF [29]. However, analysis of modO expression has not been described.
Comparison with other human-restricted respiratory tract mucosal pathogens suggests that M. catarrhalis contains relatively few phase variable genes. For example, Neisseria meningitidis contains as many as 83 putatively phase variable genes [30–32], including genes that encode outer membrane proteins, e.g., porA [33] and opc [34], genes involved in lipooligosaccharide (LOS) synthesis [35], and the mod methyltransferases [36, 37]. In Haemophilus influenzae, at least nineteen predicted phase variable genes have been identified [38], including LOS synthesis [39–41], outer membrane proteins Hia [42] and HMW [43], as well the modA methyltransferase that is also found in Neisseria species [44–47]. Many of the studied phase variable genes are virulence factors and/or vaccine candidates [48, 49]. Whilst phase variable candidates can be used in vaccines (for example, NadA in Bexsero [50–52]), there is the possibility of vaccine evasion due to phase variation. However, it may be possible to use phase variable proteins in vaccines in combination with other stably expressed determinants, or if the phase-variable candidate(s) are required at key stages during colonisation or disease.
In order to better understand phase variable gene expression, here we report the repertoire of putatively phase variable genes in M. catarrhalis that are associated with SSRs and investigate the expression of three previously identified and three novel phase variable gene candidates. We also examine the frequency of phase variation and the role of these phase variable systems in conditions mimicking human infection.
Materials and methods
Serum was isolated from human blood from healthy volunteers with informed written consent (in accordance with the guidelines and approval of the Griffith University Human Ethics Committee (HREC 2012/798)).
Bioinformatic identification of SSRs
Sequences of five closed M. catarrhalis genomes (BBH18 [53], 25239 [27], 25240 [54], FDAARGOS_213 (Accession NZ_CP020400), and CCRI-195ME [55]) analysed in this study were acquired from GenBank. All possible combinations of repeats of between one and nine repeating units were formulated and the five closed M. catarrhalis genomes were searched for these sequences, as described previously [56]. Data generated in these analyses were moved into spreadsheets for manual curation, and comparison of repeat lengths in a specific genes between strains was performed using an alignment of the five genomes (aligned using Geneious version 10.1.3 (http://www.geneious.com) [57] with the mauve plug-in [58]). Positive hits were omitted from further analysis if the SSR was found in less than three of the five genomes, varied due to SNP mutations rather than insertion/deletion of a repeat unit, or was located in regions of high frequency recombination (e.g. phage or transposon associated ORFs).
Bacterial strain and growth conditions
M. catarrhalis strains used in this study include CCRI-195ME [55], American Type Culture Collection (ATCC; Manassas, VA, USA) strains ATCC 25239 and ATCC 23246, and derivatives thereof (Table 2). M. catarrhalis strains were grown on brain heart infusion (BHI) agar (Oxoid, Basingstoke, UK) at 37°C with 5% CO2, or in BHI broth (Oxoid, Basingstoke, UK) at 37°C with orbital shaking at 200 rpm.
Fragment length analysis
Fragment length analysis was performed with strains 195ME and 25239 when measuring the length of repeat tracts in the uspA1, uspA2, mid/hag, gor (glutathione disulphide reductase), and hyp (hypothetical permease MC25239_RS00020), and with strain 23246 for the modO gene, as previously described [27, 59]. Briefly, 100–300 bp regions spanning SSRs were amplified with 6-Carboxyfluorescein (6-FAM) or Hexachlorofluorescein (HEX) labelled primers (Integrated DNA Technologies) (S1 Table) using GoTaq polymerase (Promega) as per manufacturer’s instructions. The size and relative quantity of each fluorescently labelled amplicon was measured using a 3130xl Genetic Analyser and GeneScan (Applied Biosystems, Grand Island, NY, USA), and electropherogram traces were visualized using Peakscanner version 1.0 (Applied Biosystems, Grand Island, NY, USA). Selection of M. catarrhalis subpopulations containing predominantly one repeat tract length for a given gene was carried out by repeated rounds of fragment length analysis and subculturing, as previously described [37].
Quantitative real time PCR (qRT-PCR)
Overnight plate cultures of M. catarrhalis were standardised to an optical density of OD600 = 0.1 in 20 mL BHI broth and grown for an additional 3.5 hours. 4 mL of RNAprotect Bacteria Reagent (Qiagen) was added to 2mL of bacterial culture, and RNA was extracted using the RNeasy Mini Kit (Qiagen) enzymatic lysis protocol as per manufacturer’s instructions. RNA was subsequently treated with DNaseI (NEB) and the absence of contaminating DNA was confirmed by PCR using GoTaq polymerase (Promega) and copB primers (S1 Table). cDNA was prepared using ProtoScript II Reverse Transcriptase (NEB), with random Primer 6 (NEB) and 1 μg RNA. 2.5 ng of cDNA was used as the template in 20 μl qRT-PCR reactions with SsoAdvanced Universal SYBR Green Supermix (Biorad). qRT-PCR reactions were performed in triplicate using a CFX96 Real-Time PCR Detection System (Biorad), and primers for these reactions are listed in S1 Table. The copB gene was used as an endogenous reference to normalize the results obtained with the phase variable genes.
Biofilm formation assays
M. catarrhalis strains were grown overnight on BHI agar, resuspended in BHI broth, and standardized to OD600 = 1. Cells were diluted 1:10 into chemically defined media [60], and 1 mL aliquots were dispensed into wells of a 24-well plate in triplicate (final concentration of approximately 107 CFU/mL of M. catarrhalis). Plates were incubated at 37°C with orbital shaking at 100 rpm for 24 hours. Media and planktonic cells were then aspirated from wells, and adherent cells were scraped from plates and resuspended in fresh BHI broth. 100 μl of each suspension was used to inoculate 900 μl of freshly prepared chemically defined media in a new 24-well plate, with passaging performed for three consecutive days.
Serum survival assays
Serum was isolated from human blood from volunteers, using Bio-One Vacuette serum separator tubes (Greiner) and processed as per manufacturer’s instructions. M. catarrhalis strains were grown overnight on BHI agar, standardized to OD600 = 0.05 in 20 mL BHI broth, and grown for a further 3.5 hours to mid-log phase. Cultures were equalised to OD600 = 1.0 in fresh BHI broth, three 10-fold serial dilutions were performed, and 10 μl of this suspension was inoculated into 90 μl serum in triplicate in a 96-well plate (final concentration of approximately 104 CFU/mL of M. catarrhalis). Due to differences in innate serum resistance, strain CCRI-195ME was incubated in 90% serum, and strain 25239 was incubated in 10% and 20% serum. Plates were incubated at 37°C, 5% CO2, and after 24 hours of growth, 10 μl was transferred to a new plate of serum, with passaging performed for 3 consecutive days.
Results
Bioinformatic identification of phase variable repeat tracts in M. catarrhalis
The closed genomes of five M. catarrhalis strains (BBH18, 25239, 25240, FDAARGOS_213 and CCRI-195ME) were analysed to identify SSRs and genes associated with them. Only SSRs containing greater than 7 mononucleotide (e.g., G(n)), 3 dinucleotide (e.g., AT(n)), or 2 trinucleotide, tetranucleotide, or pentanucleotide (e.g., TAA(n), CAAC(n), CAACG(n)), repeat units were included, as these represent the minimum number of repeat units likely to allow phase variation [38]. To compare repeat tracts from the five genomes, the location of repeat tracts relative to reading frames (upstream, within or downstream) was noted, and the repeat-associated genes were considered the same ORF if the BLASTn reported P value was less than 10−3. The dataset was further curated by review of repeat sequences to ensure tracts from different genomes were grouped under common designations (e.g. the repeat unit in the site-specific DNA methyltransferase modM was listed as AACC and ACCA in different genomes, but generates the same overall repeat tract sequence). At this stage, repeat tracts were removed from the analysis if all five genomes contained the same number of repeats in the same relative position, as this indicated that the repeats are not phase variable. After this process, 212 putative phase variable repeat tracts were identified across the five genomes.
To allow visual inspection of repeat regions, genomes were aligned with Mauve in Geneious. From this, repeat tract duplications were removed (i.e., if the same repeat tract had been associated with the genes upstream and downstream of it) and any missing data from specific strains was added (e.g. repeats may not have been included in the dataset if repeat numbers fell below set thresholds in a genome for that strain). From these data, a shortlist of repeat tracts that were likely to be phase variable was formed based on whether there was sufficient data available to assess if a region was phase variable, and whether the variations in repeat numbers looked to be genuine. In the first case, repeat tracts were eliminated from further consideration if the associated ORF was not present in at least three of the five genomes, or if the repeat tracts were present in regions with markers of high frequency recombination (e.g. phage or transposon related reading frames). In the latter case, repeat sequences were examined, and eliminated from further consideration if repeat tract numbers varied due to single nucleotide polymorphisms in the sequence rather than insertion/deletion of a repeat unit. For single nucleotide repeats, if variability was only seen in one genome or maximum repeat length was at the low end of minimal repeat tracts previously demonstrated to phase vary (e.g. 7 or 8 repeats seen only), tracts were also omitted from further consideration. These analyses left 17 putative phase variable repeat tracts (Table 1). All four genes that have been previously reported to be phase variable (mid/hag, modM, uspA1 and uspA2) were identified, providing suitable validation for our methodology and thresholds. Three loci (restriction endonuclease subunit M, glutathione disulphide reductase (gor), and a hypothetical permease MC25239_RS00020 (hyp)) contained upstream repeat tracts which we considered as highly likely to mediate phase variation, as repeat tract lengths varied substantially between the five genomes. However, the restriction endonuclease subunit M was previously found to be a truncated ORF located in a recombinational hotspot for DNA restriction-modification systems and was excluded from further analysis [29]. Although the Type III DNA methyltransferase gene, modO (Table 1), was not identified in our bioinformatic analyses as there are no closed RB2/3 lineage genomes, it was included based on our previous work [29].
Differences in numbers of repeat units in phase variable repeat tracts correlate with mRNA transcript level differences for some genes in M. catarrhalis 25239, CCRI-195ME and 23246
Six loci were selected to characterise phase variation in detail–the three previously identified phase variable loci (mid/hag, uspA1, uspA2) and three novel loci with a high likelihood of being phase variable (hyp, gor and modO). The length of the DNA repeat tracts in these genes, and the percentage of each repeat length in a population was determined by GeneScan fragment length analysis of SSR amplicons. Initial screening of groups of 12 pooled M. catarrhalis 25239 and 195ME single colonies revealed the presence of multiple uspA1, uspA2, mid/hag, and hyp amplicons of varying size, corresponding to natural variation in the number of repeat units in each gene’s SSR tract (Fig 1A–1D). Screening of M. catarrhalis 23246 colonies similarly demonstrated that natural variation in repeat tract length occurs in the modO gene (Fig 1F). No variation in gor glutathione disulphide reductase amplicon size was observed in M. catarrhalis 25239 or 195ME, suggesting that gor is not phase variable under standard laboratory conditions, or the frequency of gor phase variation is too low for SSR variants to be detected with our method (Fig 1E). Through successive passaging and fragment length analysis of single colonies derived from these initially mixed populations, uspA1, uspA2, hyp and modO SSR variants were isolated that were highly enriched for repeat tracts of a single length (Table 2). Although the poly-G tract of mid/hag showed natural variation (Fig 1D), populations with enriched repeat tract lengths could not be successfully isolated after several rounds of enrichment.
Fragment length analysis electropherograms of A) uspA1, B) uspA2, C) hypothetical permease MC25239_RS00020, D) mid/hag, E) glutathione disulphide reductase gene gor, and F) modO for strains as indicated above. The repeat unit for each gene is shown in the top right corner of each panel. Peaks are labelled individually with the number of repeat units present in each amplicon. Scale bars indicate amplicon length in base pairs.
To determine whether variations in SSR tract length correspond to differences in detectable mRNA transcript levels, we selected variants with defined SSR lengths (as determined by fragment length analysis) and compared the mRNA levels of uspA1, uspA2, hyp and modO by qRT-PCR. For uspA1 in strain 195ME, changes in SSR resulted in substantial changes in mRNA levels, with G(n) tract lengths containing 8, 9 or 10 G residues corresponding to low mRNA, and 11 or 12 G residues corresponding to high mRNA (21.31 fold difference between 10 G residues and 12 G residues; Fig 2A). uspA1 expression was not investigated in M. catarrhalis 25239, as this strain contains nonsense mutations in the uspA1 gene. uspA2 shows a maximum difference in mRNA levels of 3.37-fold between 17 and 19 5′-AGAT(n)-3′ repeat units in M. catarrhalis 25239. However, only a modest difference in expression is observed between 12–16 5′-AGAT(n)-3′ repeat units in strain 195ME (Fig 2B), and isolation of variants with longer repeat tracts may be required for maximal differences in expression to be observed in this strain. The number of 5′-GTTC(n)-3′ repeats in the hyp repeat tract minimally affected MC25239_RS00020 mRNA levels, with a 1.43 fold and 1.2 fold difference in expression observed in strains 195ME and 25239, respectively (Fig 2B). However, only a limited number of hyp SSR variants could be isolated in this study and further analysis using a broader range of repeat variants is required to adequately assess hyp phase variation. In contrast, mRNA levels of modO in M. catarrhalis 23246 was clearly correlated with repeat number; low levels of modO mRNA was observed when 9 or 11 5′-CAACG(n)-3′ repeats are present, while 11.81–20.66 greater mRNA was observed when 10 5′-CAACG(n)-3′ repeats were present (Fig 2D).
Relative differences in mRNA levels measured by qRT-PCR for A) uspA1, B) uspA2, C) hypothetical permease MC25239_RS00020, and D) modO. Fold-change is calculated relative to the variant with the lowest expression (set as 1.00), and values are shown above the bars. P-values were calculated using a two-tailed Student’s t-test comparing mRNA levels relative to the variant with the lowest mRNA (set at 1.00). *, P < 0.05 **, P ≤ 0.01, ***, P ≤ 0.001.
Expression of the adhesins encoded by uspA1 and mid/hag are selected against during biofilm formation
Phase variable expression of mid/hag, modM, uspA1, uspA2, gor and hyp were next analysed in conditions mimicking stages of M. catarrhalis human infection. Biofilms are multicellular communities of bacteria that are frequently seen in colonisation of hosts and that are often responsible for increased antibiotic resistance and persistent infections [61, 62]. Bacterial cells have a series of surface adhesins that aid biofilm formation [63] and M. catarrhalis has been identified in biofilms in tube otorrhea [64] and middle ear effusions [65]. Therefore, we passaged strains under biofilm forming conditions to identify whether this selected for strains with enriched phase variants. For this work, we started with three separate M. catarrhalis 195ME populations with different proportions of SSRs in uspA1 (i.e., with starting populations enriched for low expression (G10 in the SSR) or high expression (G11 or G12)). Regardless of the initial repeat tract length, after three rounds of selection of biofilm cells all resultant populations are significantly skewed to repeat tract lengths correlating to low levels of uspA1 expression (G10 residues) (Figs 3A and S1). Expression of mid/hag is also selected against in biofilm selection, with a distinct shift in SSR length that results in an on to off switch of mRNA expression (Figs 3B and S1). No consistent shift of greater than 10% of the population was observed for modM, uspA2, gor, or hyp repeat tract lengths during biofilm selection these samples (S1 Fig).
Fragment length analysis of M. catarrhalis 195ME populations passaged in biofilm formation assays for 3 consecutive days. Each graph shows a different starting population, enriched for (A) 10, 11 or 12 repeats in uspA1, or (B) 6 repeats in mid/hag. Stacked bars indicate the proportion of each repeat length found in pre-passaging (0 hrs) and post-passaging (72 hrs) populations. For uspA1, bar colour indicates the relative mRNA level correlated with each repeat length: black, high mRNA level; grey, low mRNA level; white, unknown mRNA level (uspA1 13 repeat variant only). For mid/hag bar colour indicates expression phase: black, on; white, off. Assays were carried out in biological triplicate, and proportions averaged. See S1 Fig for individual assays.
High uspA2 expressing variants are selected for during growth in human serum
Complement‐mediated killing is an important aspect of the human innate immune response, with increased levels of complement factors present during OM and exacerbations of COPD, Serum resistance is considered a key virulence factor of M. catarrhalis [66], and UspA2 is involved in serum resistance in many strains [67]. Therefore, serum killing assays were performed with M. catarrhalis strains 25239 and 195ME inoculated into human serum and grown for 24 hours, and serially passaged for three days to determine if serum effected selection of mid/hag, modM, uspA1, uspA2, gor, hyp expression. Selection for high level expression of uspA2 occurred in both strain 195E and 25239 after three passages in human serum (Figs 4 and S2). Specifically, starting with populations enriched for low or mid uspA2 expression (i.e., uspA2 SSR variants with 12–13 A(n) repeats for strain 195ME, and 18–20 repeats for strain 25239), serum selection resulted in populations enriched for SSRs consistent with high level expression (195ME enriched for tract lengths of 14–15 A(n) and strain 25239 was enriched for SSR tract lengths of 15–17 A(n)). No consistent, significant or substantial shift was seen in SSR tract length for other putative phase variable determinants in this assay (S2 Fig).
Fragment length analysis of M. catarrhalis 195ME and 25239 populations passaged in serum for 3 consecutive days. Each graph shows a different starting population, enriched for A) 12 or 13 repeats in uspA2 in strain 195ME, or B) 18, 19, or 20 repeats in strain 25239. Stacked bars indicate the proportion of each repeat length found in pre-passaging (0 hrs) and post-passaging (24 or 72 hrs) populations. Bar colour indicates the relative expression level correlated with each repeat length: black, high expression; dark grey, mid expression; light grey, low expression; white, unknown expression level. Assays were carried out in biological triplicate, and proportions averaged. See S2 Fig for individual assays.
Discussion
Phase variation is the high frequency reversible switching of gene expression, which can provide a selective advantage for bacteria as expression of determinants can be temporarily altered without changes to the coding sequence of the gene [13]. As a result, phase variation allows cells to evade specific antibodies and generate population heterogeneity. In our bioinformatic analysis of five M. catarrhalis genomes, we identified 17 unique SSRs with the potential to mediate phase variable expression of the associated genes. Of those, we observed natural variation in SSR in the genomes of M. catarrhalis for five genes. We found that altered SSR tract lengths modulates levels of mRNA transcript detectable by qRT-PCR, and that phase variation of three major outer membrane proteins occurs due to selection pressure exerted under biologically relevant conditions.
Our work details that phase-variable expression of the uspA1, uspA2, and mid/hag genes is associated with changes in SSR repeat tract lengths, confirming earlier findings for uspA1 [20], uspA2 [21], and mid/hag [23]. In addition to being phase-variably expressed, uspA1 and uspA2 show significant sequence variation (antigenic variation) between strains due to rearrangement, deletion, and duplication of modular domains [20, 23, 68]. This variability is concentrated in surface-exposed regions, while their membrane-embedded regions are more conserved [68]. Antigenic variation is a feature of many outer-membrane proteins in various host adapted pathogens and aids in immune-evasion [49]. In addition, two further variants of UspA2 have been described: the hybrid UspA1-UspA2 protein known as UspA2H [16] and UspA2V [69]. The consequence of combining both phase and antigenic variability in a single gene/gene product is a high degree of strain to strain variability. Furthermore, when sequence variability is located in genomic regions near SSRs, this variability may also complicate the analysis of expression. For example, while it is reported that in strains O12E and O35E uspA1 repeat tracts of G(9) and G(10) cause low and high level expression, respectively [20], we found that in strain 195ME high uspA1 mRNA levels are associated with G(11) and G(12) tract lengths. Sequence analysis of the uspA1 gene in these three strains shows this difference is likely due to the presence of an additional adenine residue upstream of the repeat region in strain 195ME. Similarly, while expression of the uspA2 locus has been reported to correlate with increasing numbers of 5′-AGAT(n)-3′ repeats in strain O12E, with maximal expression seen with 18 5′-AGAT(n)-3′ repeats [21], we found that 14–15 and 16–17 5′-AGAT(n)-3′ repeat units correspond to the highest levels of mRNA transcripts in strains 195ME and 25239, respectively. This emphasises the effect that variability between genomes may have on expression levels, and should serve as a caveat for future studies. Given the phase variability of the UspA1, UspA2 and Mid/Hag proteins and necessity of determining expression status on a case-by-case basis, these proteins would not necessarily be considered to be good vaccine candidates. However, these proteins are major outer membrane components of M. catarrhalis and comprise most of the surface projections from the cell [22]. They also mediate key virulence pathways, including biofilm formation [15] and adherence to multiple epithelial cell types [16, 17] and extracellular matrix components [18, 19] for UspA1; serum resistance and binding to vibronectin [21] and agglutination of M. catarrhalis cells, red blood cells and immunoglobulin D [22, 23] for Mid/Hag. In this case, it may still be possible to use these determinants as vaccine candidates if they remain expressed over the course of an infection for targeting by the immune system. However, our work showed that expression of uspA1 and mid/hag is not stable over long periods under conditions mimicking human infection, with expression decreasing over time in biofilms on an abiotic surface. A recent report also found that mid/hag expression is turned off in clinical samples during infection [24]. This suggests that Mid/Hag in particular is not a good vaccine candidate, as its expression is not required for persistent colonisation. In contrast, expression of uspA2 increases over time in serum and may provide a more consistent vaccine target during infection. It is interesting to note that there is no variation of mid/hag and uspA1 during exposure to human serum, as compared to the biofilm passaging, whereas uspA2 shows variation under exposure to serum but not during biofilm passaging. These data suggest that whilst individual cells in a population may phase vary at random, the overall population maintains a relatively stable proportion of variants unless a selective force is present and a difference in fitness between variants exists.
This study demonstrated that the modO gene is phase variable due to 5ʹ-CAACG(n)-3ʹ repeats upstream of its ORF, and that there is a 20-fold difference in detectable transcript of modO depending on repeat tract size. The confirmation of the phase variable nature of modO brings the number of phase variable DNA methyltransferases in M. catarrhalis to three (modM, modN and modO), and transcriptomic or proteomic analysis will be required to confirm whether ModO regulates a phasevarion in M. catarrhalis. Investigation of how these epigenetic regulators interact with other virulence determinants of M. catarrhalis may provide valuable insight into the pathogenicity of this species.
The length of the SSR tract associated with the MC25239_RS00020 hypothetical permease clearly shows variation between strains and within individual strains that have been serially passaged, however substantial differences in transcription levels were not observed under the conditions tested. Although a large range of hyp SSR repeats were observed between strains (6–27 5'-GTTC(n)-3' repeats), only a few hyp SSR variants were isolated. Greater differences in expression may be observed upon isolation of the full range of SSR lengths, and this warrants further investigation. In the case of the glutathione sulphide reductase gene, gor, inter-genome variation in SSR tract length does not appear to be correlated with phase variation as no natural variation in gor repeat length was observed within an individual strain. However, future studies of gor using a strain with a higher number of repeats within the 5'-GACTGTTT(n)-3' SSR tract, such as the 17 repeat tract in strain FDAARGOS_213, may allow phase variation to be observed at a measurable frequency. Further investigations into these two loci are needed to fully elucidate if these genes are phase-variably expressed, and determine their role in M. catarrhalis pathobiology.
Relative to other respiratory tract pathogens, M. catarrhalis does not appear to have as large a number of phase variable genes that vary by SSRs. However, the phase variable genes identified in this study may not represent the full repertoire of phase variable genes in M. catarrhalis due to several limitations of our analysis. Firstly, screening was restricted to only the five closed genomes available at the time of analysis (necessitated as short-read assemblies often split SSRs between contigs, preventing complete analysis of SSR length). In addition, all of the five closed genomes analysed are from the M. catarrhalis RB1 lineage, and consequently, phase variable genes specific to the RB2/3 lineage may remain to be identified–as we found with modO which was identified in our previous work [29]. Phase variable genes may have been omitted from analysis due to our strict criteria. For example, repeat tracts were eliminated from further consideration if they and the associated reading frames were not present in a majority of the genomes, or if only minimal repeat variation was seen (particularly with polynucleotide SSR). These exclusions do not reflect on the potential variability of these tracts, but on the diversity of genomes used. In addition, phase variation can also occur via other mechanisms, for example genome inversion as seen with fimS in E. coli [70], or site-specific recombination as observed for Type I restriction-modification systems [71]. However, identification of genes that switch expression by these mechanisms is complicated, due to their mediating motifs often being short and highly variable in sequence, placement, and orientation, and was not within the scope of this study.
Overall, we define the repertoire of simple sequence repeats and putatively phase variable genes in M. catarrhalis. In addition, we show that phase variation of several genes occurs during growth and in conditions that mimic human infection which contributes to our understanding of M. catarrhalis pathogenesis and will aid in future selection of candidate vaccine antigens.
Supporting information
S1 Fig. Analysis of DNA repeat tract lengths of putative phase variable genes during biofilm passaging.
Fragment length analysis of M. catarrhalis 195ME populations passaged in biofilm formation assays for 3 consecutive days. Each graph includes three different starting populations, enriched for 10, 11 or 12 repeats in uspA1 (Sample 1, 2, or 3, respectively). Assays were carried out in triplicate, and each circle indicates a separate repeat (closed circle is at 0 h; open circle is at 72 h). The bar represents the mean, and error bars represent ±1 standard deviation. A two-tailed Student’s t-test was used to compare time 0 h vs 72 h (*, P < 0.05 **, P ≤ 0.01, ***, P ≤ 0.001).
https://doi.org/10.1371/journal.pone.0234306.s001
(PDF)
S2 Fig. Analysis of DNA repeat tract lengths of putative phase variable genes during serum passaging.
Fragment length analysis of M. catarrhalis 195ME and 25239 populations passaged in serum for 3 consecutive days. Each graph includes five different starting populations, enriched for 12 or 13 repeats in uspA2 in strain 195ME (Sample 1 or 2, respectively) or 18, 19, or 20 repeats in uspA2 in strain 25239 (Sample 1 or 2, or 3, respectively). Assays were carried out in triplicate, and each circle indicates a separate repeat (closed circle is at 0 h; open circle is at 72 h). The bar represents the mean, and error bars represent ±1 standard deviation. A two-tailed Student’s t-test was used to compare time 0 h vs 72 h (*, P < 0.05 **, P ≤ 0.01, ***, P ≤ 0.001).
https://doi.org/10.1371/journal.pone.0234306.s002
(PDF)
References
- 1. Vaneechoutte M, Verschraegen G, Claeys G, Weise B, Van den Abeele AM. Respiratory tract carrier rates of Moraxella (Branhamella) catarrhalis in adults and children and interpretation of the isolation of M. catarrhalis from sputum. J Clin Microbiol. 1990;28(12):2674–80. pmid:2126266
- 2. Murphy TF, Parameswaran GI. Moraxella catarrhalis, a human respiratory tract pathogen. Clin Infect Dis. 2009;49(1):124–31.
- 3. Monasta L, Ronfani L, Marchetti F, Montico M, Vecchi Brumatti L, Bavcar A, et al. Burden of disease caused by otitis media: systematic review and global estimates. PLoS One. 2012;7(4):e36226. pmid:22558393
- 4. Boswell JB, Nienhuys TG. Onset of otitis media in the first eight weeks of life in aboriginal and non-aboriginal Australian infants. Ann Otol Rhinol Laryngol. 1995;104(7):542–9. pmid:7598367
- 5. Pichichero ME. Recurrent and persistent otitis media. Pediatr Infect Dis J. 2000;19(9):911–6. pmid:11001126
- 6. Watson DS, Clapin M. Ear health of aboriginal primary school children in the Eastern Goldfields Region of Western Australia. Aust J Public Health. 1992;16(1):26–30. pmid:1627706
- 7. Lehmann D, Weeks S, Jacoby P, Elsbury D, Finucane J, Stokes A, et al. Absent otoacoustic emissions predict otitis media in young Aboriginal children: a birth cohort study in Aboriginal and non-Aboriginal children in an arid zone of Western Australia. BMC Pediatr. 2008;8:32. pmid:18755038
- 8.
World Health Organization. The top 10 causes of death (Fact sheet No. 310) 2012 [updated May 2014]. Available from: http://www.who.int/mediacentre/factsheets/fs310/en/.
- 9. Suissa S, Dell'Aniello S, Ernst P. Long-term natural history of chronic obstructive pulmonary disease: severe exacerbations and mortality. Thorax. 2012;67(11):957–63. pmid:22684094
- 10. Pettigrew MM, Alderson MR, Bakaletz LO, Barenkamp SJ, Hakansson AP, Mason KM, et al. Panel 6: Vaccines. Otolaryngol Head Neck Surg. 2017;156(4_suppl):S76–S87. pmid:28372533
- 11. Murphy TF, Brauer AL, Grant BJ, Sethi S. Moraxella catarrhalis in chronic obstructive pulmonary disease: burden of disease and immune response. Am J Respir Crit Care Med. 2005;172(2):195–9. pmid:15805178
- 12. Perez AC, Murphy TF. A Moraxella catarrhalis vaccine to protect against otitis media and exacerbations of COPD: An update on current progress and challenges. Hum Vaccin Immunother. 2017;13(10):2322–31. pmid:28853985
- 13. Moxon R, Bayliss C, Hood D. Bacterial contingency loci: the role of simple sequence DNA repeats in bacterial adaptation. Annu Rev Genet. 2006;40:307–33. pmid:17094739
- 14. Blakeway LV, Tan A, Peak IRA, Seib KL. Virulence determinants of Moraxella catarrhalis: distribution and considerations for vaccine development. Microbiology. 2017.
- 15. Pearson MM, Laurence CA, Guinn SE, Hansen EJ. Biofilm formation by Moraxella catarrhalis in vitro: roles of the UspA1 adhesin and the Hag hemagglutinin. Infect Immun. 2006;74(3):1588–96. pmid:16495530
- 16. Lafontaine ER, Cope LD, Aebi C, Latimer JL, McCracken GH Jr., Hansen EJ. The UspA1 protein and a second type of UspA2 protein mediate adherence of Moraxella catarrhalis to human epithelial cells in vitro. J Bacteriol. 2000;182(5):1364–73. pmid:10671460
- 17. Aebi C, Lafontaine ER, Cope LD, Latimer JL, Lumbley SL, McCracken GH Jr., et al. Phenotypic effect of isogenic uspA1 and uspA2 mutations on Moraxella catarrhalis 035E. Infect Immun. 1998;66(7):3113–9. pmid:9632574
- 18. Tan TT, Nordstrom T, Forsgren A, Riesbeck K. The respiratory pathogen Moraxella catarrhalis adheres to epithelial cells by interacting with fibronectin through ubiquitous surface proteins A1 and A2. J Infect Dis. 2005;192(6):1029–38. pmid:16107956
- 19. Tan TT, Forsgren A, Riesbeck K. The respiratory pathogen Moraxella catarrhalis binds to laminin via ubiquitous surface proteins A1 and A2. J Infect Dis. 2006;194(4):493–7. pmid:16845633
- 20. Lafontaine ER, Wagner NJ, Hansen EJ. Expression of the Moraxella catarrhalis UspA1 protein undergoes phase variation and is regulated at the transcriptional level. J Bacteriol. 2001;183(5):1540–51. pmid:11160084
- 21. Attia AS, Hansen EJ. A conserved tetranucleotide repeat is necessary for wild-type expression of the Moraxella catarrhalis UspA2 protein. J Bacteriol. 2006;188(22):7840–52. pmid:16963572
- 22. Pearson MM, Lafontaine ER, Wagner NJ, St Geme JW 3rd, Hansen EJ. A hag mutant of Moraxella catarrhalis strain O35E is deficient in hemagglutination, autoagglutination, and immunoglobulin D-binding activities. Infect Immun. 2002;70(8):4523–33. pmid:12117964
- 23. Mollenkvist A, Nordstrom T, Hallden C, Christensen JJ, Forsgren A, Riesbeck K. The Moraxella catarrhalis immunoglobulin D-binding protein MID has conserved sequences and is regulated by a mechanism corresponding to phase variation. J Bacteriol. 2003;185(7):2285–95. pmid:12644500
- 24. Murphy TF, Brauer AL, Pettigrew MM, LaFontaine ER, Tettelin H. Persistence of Moraxella catarrhalis in chronic obstructive pulmonary disease and regulation of the Hag/MID adhesin. J Infect Dis. 2019;219(9):1448–55. pmid:30496439
- 25. Wang W, Pearson MM, Attia AS, Blick RJ, Hansen EJ. A UspA2H-negative variant of Moraxella catarrhalis strain O46E has a deletion in a homopolymeric nucleotide repeat common to uspA2H genes. Infect Immun. 2007;75(4):2035–45. pmid:17220316
- 26. Seib KL, Peak IR, Jennings MP. Phase variable restriction-modification systems in Moraxella catarrhalis. FEMS Immunol Med Microbiol. 2002;32(2):159–65. pmid:11821238
- 27. Blakeway LV, Power PM, Jen FE, Worboys SR, Boitano M, Clark TA, et al. ModM DNA methyltransferase methylome analysis reveals a potential role for Moraxella catarrhalis phasevarions in otitis media. Faseb j. 2014;28(12):5197–207. pmid:25183669
- 28. Blakeway LV, Tan A, Jurcisek JA, Bakaletz LO, Atack JM, Peak IR, et al. The Moraxella catarrhalis phase-variable DNA methyltransferase ModM3 is an epigenetic regulator that affects bacterial survival in an in vivo model of otitis media. BMC Microbiol. 2019;19(1):276. pmid:31818247
- 29. Blakeway LV, Tan A, Lappan R, Ariff A, Pickering JL, Peacock CS, et al. Moraxella catarrhalis restriction-modification systems are associated with phylogenetic lineage and disease. Genome Biol Evol. 2018.
- 30. Martin P, van de Ven T, Mouchel N, Jeffries AC, Hood DW, Moxon ER. Experimentally revised repertoire of putative contingency loci in Neisseria meningitidis strain MC58: evidence for a novel mechanism of phase variation. Molecular Microbiology. 2003;50(1):245–57. pmid:14507378
- 31. Saunders NJ, Jeffries AC, Peden JF, Hood DW, Tettelin H, Rappuoli R, et al. Repeat-associated phase variable genes in the complete genome sequence of Neisseria meningitidis strain MC58. Mol Microbiol. 2000;37(1):207–15. pmid:10931317
- 32. Snyder LAS, Butcher SA, Saunders NJ. Comparative whole-genome analyses reveal over 100 putative phase-variable genes in the pathogenic Neisseria spp. Microbiology. 2001;147(8):2321–32.
- 33. van der Ende A, Hopman CT, Dankert J. Multiple mechanisms of phase variation of PorA in Neisseria meningitidis. Infect Immun. 2000;68(12):6685–90. pmid:11083782
- 34. Sarkari J, Pandit N, Moxon ER, Achtman M. Variable expression of the Opc outer membrane protein in Neisseria meningitidis is caused by size variation of a promoter containing poly-cytidine. Mol Microbiol. 1994;13(2):207–17. pmid:7984102
- 35. Berrington AW, Tan YC, Srikhanta Y, Kuipers B, van der Ley P, Peak IR, et al. Phase variation in meningococcal lipooligosaccharide biosynthesis genes. FEMS Immunol Med Microbiol. 2002;34(4):267–75. pmid:12443826
- 36. Srikhanta YN, Dowideit SJ, Edwards JL, Falsetta ML, Wu HJ, Harrison OB, et al. Phasevarions mediate random switching of gene expression in pathogenic Neisseria. PLoS Pathog. 2009;5(4):e1000400. pmid:19390608
- 37. Seib KL, Pigozzi E, Muzzi A, Gawthorne JA, Delany I, Jennings MP, et al. A novel epigenetic regulator associated with the hypervirulent Neisseria meningitidis clonal complex 41/44. Faseb j. 2011;25(10):3622–33. pmid:21680891
- 38. Power PM, Sweetman WA, Gallacher NJ, Woodhall MR, Kumar GA, Moxon ER, et al. Simple sequence repeats in Haemophilus influenzae. Infect Genet Evol. 2009;9(2):216–28. pmid:19095084
- 39. Fox KL, Atack JM, Srikhanta YN, Eckert A, Novotny LA, Bakaletz LO, et al. Selection for phase variation of LOS biosynthetic genes frequently occurs in progression of non-typeable Haemophilus influenzae infection from the nasopharynx to the middle ear of human patients. PLoS One. 2014;9(2):e90505. pmid:24587383
- 40. Poole J, Foster E, Chaloner K, Hunt J, Jennings MP, Bair T, et al. Analysis of Nontypeable Haemophilus influenzae phase-variable genes during experimental human nasopharyngeal colonization. The Journal of Infectious Diseases. 2013;208(5):720–7. pmid:23715658
- 41. Hood DW, Deadman ME, Jennings MP, Bisercic M, Fleischmann RD, Venter JC, et al. DNA repeats identify novel virulence genes in Haemophilus influenzae. Proc Natl Acad Sci U S A. 1996;93(20):11121–5. pmid:8855319
- 42. Atack JM, Winter LE, Jurcisek JA, Bakaletz LO, Barenkamp SJ, Jennings MP. Selection and counterselection of Hia expression reveals a key role for phase-variable expression of Hia in infection caused by Nontypeable Haemophilus influenzae. J Infect Dis. 2015;212(4):645–53. pmid:25712964
- 43. Dawid S, Barenkamp SJ, St Geme JW 3rd. Variation in expression of the Haemophilus influenzae HMW adhesins: a prokaryotic system reminiscent of eukaryotes. Proc Natl Acad Sci U S A. 1999;96(3):1077–82. pmid:9927696
- 44. Srikhanta YN, Maguire TL, Stacey KJ, Grimmond SM, Jennings MP. The phasevarion: a genetic system controlling coordinated, random switching of expression of multiple genes. Proc Natl Acad Sci U S A. 2005;102(15):5547–51. pmid:15802471
- 45. Bayliss CD, Callaghan MJ, Moxon ER. High allelic diversity in the methyltransferase gene of a phase variable type III restriction-modification system has implications for the fitness of Haemophilus influenzae. Nucleic Acids Research. 2006;34(14):4046–59. pmid:16914439
- 46. De Bolle X, Bayliss CD, Field D, Van De Ven T, Saunders NJ, Hood DW, et al. The length of a tetranucleotide repeat tract in Haemophilus influenzae determines the phase variation rate of a gene with homology to type III DNA methyltransferases. Molecular Microbiology. 2000;35(1):211–22. pmid:10632891
- 47. Fox KL, Dowideit SJ, Erwin AL, Srikhanta YN, Smith AL, Jennings MP. Haemophilus influenzae phasevarions have evolved from type III DNA restriction systems into epigenetic regulators of gene expression. Nucleic Acids Research. 2007;35(15):5242–52. pmid:17675301
- 48. Tan A, Atack JM, Jennings MP, Seib KL. The capricious nature of bacterial pathogens: Phasevarions and vaccine development. Front Immunol. 2016;7:586. pmid:28018352
- 49. Phillips ZN, Tram G, Seib KL, Atack JM. Phase-variable bacterial loci: how bacteria gamble to maximise fitness in changing environments. Biochem Soc Trans. 2019.
- 50. Liguori A, Malito E, Lo Surdo P, Fagnocchi L, Cantini F, Haag AF, et al. Molecular basis of ligand-dependent regulation of NadR, the transcriptional repressor of meningococcal virulence factor NadA. PLoS Pathogens. 2016;12(4):e1005557. pmid:27105075
- 51. Martin P, Makepeace K, Hill SA, Hood DW, Moxon ER. Microsatellite instability regulates transcription factor binding and gene expression. Proceedings of the National Academy of Sciences of the United States of America. 2005;102(10):3800–4. pmid:15728391
- 52. Metruccio MM, Pigozzi E, Roncarati D, Berlanda Scorza F, Norais N, Hill SA, et al. A novel phase variation mechanism in the meningococcus driven by a ligand-responsive repressor and differential spacing of distal promoter elements. PLoS Pathog. 2009;5(12):e1000710. pmid:20041170
- 53. de Vries SP, van Hijum SA, Schueler W, Riesbeck K, Hays JP, Hermans PW, et al. Genome analysis of Moraxella catarrhalis strain BBH18, [corrected] a human respiratory tract pathogen. J Bacteriol. 2010;192(14):3574–83. pmid:20453089
- 54. Daligault HE, Davenport KW, Minogue TD, Bishop-Lilly KA, Bruce DC, Chain PS, et al. Complete genome assembly of a quality control reference isolate, Moraxella catarrhalis strain ATCC 25240. Genome Announc. 2014;2(5).
- 55. Tan A, Blakeway LV, Bakaletz LO, Boitano M, Clark TA, Korlach J, et al. Complete genome sequence of Moraxella catarrhalis strain CCRI-195ME, isolated from the middle ear. Genome Announc. 2017;5(21).
- 56. Atack JM, Yang Y, Seib KL, Zhou Y, Jennings MP. A survey of Type III restriction-modification systems reveals numerous, novel epigenetic regulators controlling phase-variable regulons; phasevarions. Nucleic Acids Res. 2018;46(7):3532–42. pmid:29554328
- 57. Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, et al. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics. 2012;28(12):1647–9. pmid:22543367
- 58. Darling AE, Mau B, Perna NT. progressiveMauve: multiple genome alignment with gene gain, loss and rearrangement. PLoS One. 2010;5(6):e11147. pmid:20593022
- 59. Jen FE, Seib KL, Tan A. Screening DNA repeat tracts of phase variable genes by fragment analysis. Methods Mol Biol. 2019;1969:93–104. pmid:30877671
- 60. Tan A, Li WS, Verderosa AD, Blakeway LV, T DM, Totsika M, et al. Moraxella catarrhalis NucM is an entry nuclease involved in extracellular DNA and RNA degradation, cell competence and biofilm scaffolding. Sci Rep. 2019;9(1):2579. pmid:30796312
- 61. Costerton JW, Stewart PS, Greenberg EP. Bacterial biofilms: A common cause of persistent infections. Science. 1999;284(5418):1318–22. pmid:10334980
- 62. Post JC, Stoodley P, Hall–Stoodley L, Ehrlich GD. The role of biofilms in otolaryngologic infections. Current Opinion in Otolaryngology & Head and Neck Surgery. 2004;12(3):185–90.
- 63. Berne C, Ducret A, Hardy GG, Brun YV. Adhesins involved in attachment to abiotic surfaces by Gram-negative bacteria. Microbiology spectrum. 2015;3(4): pmid:26350310
- 64. Idicula WK, Jurcisek JA, Cass ND, Ali S, Goodman SD, Elmaraghy CA, et al. Identification of biofilms in post-tympanostomy tube otorrhea. The Laryngoscope. 2016;126(8):1946–51. pmid:27426942
- 65. Thornton RB, Wiertsema SP, Kirkham L-AS, Rigby PJ, Vijayasekaran S, Coates HL, et al. Neutrophil extracellular traps and bacterial biofilms in middle ear effusion of children with recurrent acute otitis media–a potential treatment target. PLOS ONE. 2013;8(2):e53837. pmid:23393551
- 66. Wirth T, Morelli G, Kusecek B, van Belkum A, van der Schee C, Meyer A, et al. The rise and spread of a new pathogen: seroresistant Moraxella catarrhalis. Genome research. 2007;17(11):1647–56. pmid:17895425
- 67. Attia AS, Ram S, Rice PA, Hansen EJ. Binding of vitronectin by the Moraxella catarrhalis UspA2 protein interferes with late stages of the complement cascade. Infection and Immunity. 2006;74(3):1597–611. pmid:16495531
- 68. Brooks MJ, Sedillo JL, Wagner N, Laurence CA, Wang W, Attia AS, et al. Modular arrangement of allelic variants explains the divergence in Moraxella catarrhalis UspA protein function. Infect Immun. 2008;76(11):5330–40. pmid:18678659
- 69. Hill DJ, Whittles C, Virji M. A novel group of Moraxella catarrhalis UspA proteins mediates cellular adhesion via CEACAMs and vitronectin. PLoS One. 2012;7(9):e45452. pmid:23049802
- 70. Gally DL, Leathart J, Blomfield IC. Interaction of FimB and FimE with the fim switch that controls the phase variation of type 1 fimbriae in Escherichia coli K-12. Molecular Microbiology. 1996;21(4):725–38. pmid:8878036
- 71. Manso AS, Chai MH, Atack JM, Furi L, De Ste Croix M, Haigh R, et al. A random six-phase switch regulates pneumococcal virulence via global epigenetic changes. Nat Commun. 2014;5:5055. pmid:25268848