Abstract Interspecies interactions between bacterial pathogens and the commensal microbiota can influence disease outcome. In the nasal cavities, Staphylococcus epidermidis has been shown to be a determining factor for Staphylococcus aureus colonization and biofilm formation. However, the interaction between S. epidermidis and S. aureus has mainly been described by phenotypic analysis, and little is known about how this interaction modulates gene expression. This study aimed to determine the interactome of nasal S. aureus and S. epidermidis isolates to understand the molecular effect of interaction. After whole-genome sequencing of two nasal staphylococcal isolates, an agar-based RNA sequencing setup was utilized to identify interaction-induced transcriptional alterations in surface-associated populations. Our results revealed differential expression of several virulence genes in both species. We also identified putative non-coding RNAs (ncRNAs) and, interestingly, detected a putative ncRNA transcribed antisense to esp, the serine protease of S. epidermidis, that has previously been shown to inhibit nasal colonization of S. aureus. In our study, the gene encoding Esp and the antisense ncRNA are both downregulated during interaction with S. aureus. Our findings contribute to a better understanding of pathogen physiology in the context of interactions with the commensal microbiota, and may provide targets for future therapeutics. pathogen–commensal interaction, Staphylococcus aureus, Staphylococcus epidermidis, transcriptome, RNAseq, non-coding RNA INTRODUCTION Interactions between pathogenic bacteria and the commensal microbiota can influence disease phenotype or the clinical outcome of the infection (Wigneswaran et al.2016). One example is the gut microbiota providing resistance to pathogen colonization by direct competition for nutrients, though also by activating the host immune system and inducing killing of the invading pathogens (Sassone-Corsi and Raffatellu 2015). Moreover, interactions with commensal bacteria have been shown to directly modulate virulence potential of bacterial pathogens. For example, Enterococcus faecalis has been shown to increase the virulence of Escherichia coli in urinary tract infections (Lavigne et al.2008), while Staphylococcus aureus virulence can be reduced during co-infection with commensal bacterial species (Ngba Essebe et al.2017). In the gut, Bifidobacteria limit infection by enteropathogenic Escherichia coli (Fukuda et al.2011). These and other related examples of interactions between the commensal microbiota and potential pathogens underline the importance of understanding the molecular basis of interactions. Identification and characterization of the bacterial gene products and metabolites that mediate microbe–microbe interactions have revealed an astonishing diversity of signals and molecules being exchanged among microbial species. The molecular basis of microbial interactive processes range from direct cell–cell signaling such as quorum-sensing cross-talk and small molecule transfer (Eberl and Tümmler 2004; Fugère et al.2014), effects induced by release of organic volatile compounds (Briard, Heddergott and Latgé 2016), interspecies transfer of mobile small RNAs (Liu et al.2012; Zhou, Zhou and Chen 2017), to electrical communication among different microbial species (Prindle et al.2015; Humphries et al.2017). It is equally important to determine in which way pathogen–commensal interactions modify the behaviour of the pathogen. For example, application of RNA sequencing to analyse these interactive systems may reveal transcriptomic changes induced by microbe–microbe interactions. The focus of this study is the clinically relevant interaction between pathogenic S. aureus and the commensal S. epidermidis, with the primary reservoirs of both being the nasal cavities. Individuals colonized by S. aureus are at a greater risk of disease development, as these sites can function as infection reservoirs (Moss, Squire and Topley 1948; Wertheim et al.2005). Staphylococcus aureus is the causative agent of many serious infections whereas S. epidermidis is the dominant commensal bacterium in the nasal cavities. Staphylococcus epidermidis appears to be a determining factor for carriage of S. aureus and its production of a serine protease Esp has been reported to inhibit biofilm formation and limit nasal colonization by S. aureus (Iwase et al.2010; Sugimoto et al.2013; Vandecandelaere et al.2014). However, more recent studies have suggested that the inhibitory effect of S. epidermidis might be more complex than initially assessed (Fredheim et al.2015). The insight into the interaction between nasal staphylococci strains has so far been based on phenotypic analyses, with limited knowledge about interactions at the molecular level. The clinical importance of the staphylococcal interaction as well as the unknown molecular basis of this interaction highlights the importance of using molecular methods to characterize the interactome of commensal and pathogen in the nasal cavities. In this study, we use RNA sequencing technology (RNAseq) to obtain molecular level insight into the functional consequences of microbe–microbe interactions in terms of effects on the transcriptome. Non-coding RNAs (ncRNAs) can alter the transcriptional and translational profiles of a range of bacterial species (Gottesman and Storz 2011). Several ncRNAs have been identified in S. aureus with the most well-studied being RNAIII, controlling the agr system that plays a central role in bacterial pathogenesis (Boisset et al.2007; Felden et al.2011; Carroll et al.2016). Even though it has previously been shown that ncRNAs play a role in host–pathogen interaction by modulating the human signaling response (Koeppen et al.2016; Westermann et al.2016), very few studies have focused on the ncRNA pool in response to bacterial interspecies interactions (Miller et al.2017). In this study, we aim to characterize the molecular basis of interactions between S. aureus and S. epidermidis isolated from the nasal cavities of a healthy human. We use an agar plate-based assay to control the interaction and determine changes in gene expression during co-culture by RNAseq. Our analysis of this pathogen–commensal interactive system revealed that expression of several virulence genes is responsive to microbe–microbe interactions. Furthermore, we detect putative ncRNAs implicated in the interaction and identify those that are differentially expressed. MATERIALS AND METHODS Bacterial strains and growth conditions A collection of frozen nasal swabs from Danish adults sampled in a previous study (Andersen et al.2012) was obtained. All swabs were spread onto S. aureus chromID selective plates and blood agar plates and incubated at 37°C to identify staphylococci strains. Staphylococcus epidermidis 8/9A and S. aureus 9B were cultivated in tryptic soy broth (TSB) with shaking (200 rpm) at 37°C overnight (ON). Characterization of staphylococci strains The identification of S. aureus and S. epidermidis was initially done by colony morphology investigation after plating on blood agar. Subsequent confirmation of S. aureus strains and identification of clonal complexes were established by spa typing using Sanger sequencing as previously described (Fode, Stegger and Andersen 2011). Moreover, a PCR strategy targeting the tuf gene was performed to distinguish S. epidermidis from S. aureus strains (Delgado et al.2009). The PCR-based results were confirmed with Gram staining as well as coagulase and catalase tests as previously described (Andersen et al.2012). Staphylococcus epidermidis strains were subjected to a PCR strategy to identify the presence of the serine protease gene esp as previously described (Ikeda et al.2004). The strains were typed using the web-based Multi Locus Sequence Typing (Larsen et al.2012) on the basis of their genome sequence. Biofilm formation assay Semi-quantitative determination of biofilm formation was performed in 96-well tissue culture plates (O’Toole 2011). Single colonies were inoculated in TSB and grown ON at 37°C, 200 rpm. Cell-free supernatants were prepared by passing the ON cultures through 0.2 μm pore size syringe filters. Supernatants and ON cultures were adjusted to an optical density measured at 600 nm (OD600) of 1.0 and 0.1, respectively. The ON cultures were added to individual wells of the 96-well plate to a starting density of OD600 = 0.05 and cell-free supernatants were added onto the cells. The plates were incubated statically at 37°C ON, allowing the cells to form biofilms. Subsequently, OD600 was measured using a microplate reader before the plates were washed with 0.9% sodium chloride solution three times, dried for 60 min at 60°C and stained with 0.1% crystal violet solution. After staining and three washes, the density of adherent stained cells was measured at OD590. The assay was repeated twice and representative results are presented. RNAseq setup Samples for transcriptome sequencing were prepared as follows. ON cultures in TSB were adjusted to OD600 = 1.0, followed by spotting 10 μl of the cultures on top of autoclaved 0.45 μm filters (mixed cellulose ester) placed on tryptic soy agar (TSA) plates. The plates were incubated at 37°C ON. After incubation, the filters were removed from the plates, transferred to individual microfuge tubes and frozen at −80°C prior to RNA isolation. RNA extraction and cDNA library preparation The samples were homogenized in two rounds by FastPrep ® (MP Biomedicals, Santa Ana, California) using 300 mg acid-washed glass beads at speed 6 for 40 s followed by incubation on ice. Total RNA was isolated using the RNeasy Mini Kit (Qiagen, Germany) according to the supplier's instructions. To remove genomic DNA, Ambion®TURBO DNA-free Kit (Invitrogen, Carlsbad, California) was used according to the manual. RNA integrity was determined using the Bioanalyzer 2100 (Agilent Technologies, Santa Clara, California) and samples with RNA integrity number above eight were selected for cDNA library preparation. Four independent biological experiments were performed. Library preparation was performed using the ScriptSeq v2 RNA-Seq Library Preparation Kit (Illumina, San Diego, California) and rRNA removal was performed using Ribo-Zero rRNA Removal Kit (Illumina). Library preparation comprises RNA fragmentation, synthesis of cDNA, terminal tagging of cDNA, amplification and library quality control. cDNA was sequenced as 150 bp paired-end reads on a HiSeq Sequencer (Illumina). Generated reads have been deposited to NCBI Sequence Read Archive (SRA) (BioProject PRJNA421413). Purification of genomic DNA for sequencing of nasal isolates Single colonies of the nasal S. aureus and S. epidermidis strains were inoculated in TSB and allowed to grow ON at 37°C, 200 rpm. The purification of genomic DNA was performed using the Wizard® Genomic DNA Purification Kit (Promega, Madison, Wisconsin) according to the manufacturer's instructions. The genomic libraries were generated using modified (half-volumes of the reagents from the original protocol) KAPA HyperPlus Library Preparation Kit (Roche Molecular Systems, Pleasanton, California). Genomic libraries were sequenced using the MiSeq V2 300 Cycles Reagent Kit (Illumina) as 150 bp paired-end reads. Generated reads have been deposited to NCBI SRA (BioProject PRJNA421413). Genome assembly and annotation Sequencing reads of nasal isolates were trimmed with the seqtk tool v1.2-r94 and assembled into contigs with SPAdes v3.11.0 (‘careful’ mode) (Bankevich et al.2012). Contigs (>200 bp, >7× coverage) were rearranged with the Mauve Contig Mover tool v2.4.0, based on the published reference genome sequence: S. aureus USA300 (NC_007795) and S. epidermidis RP62A (NC_004461) (Rissman et al.2009). Contigs were concatenated using the ‘union’ command from the EMBOSS package v6.6.0 (Rice, Longden and Bleasby 2000) and subsequently annotated with Prokka v1.12 (Seemann 2014). RNAseq differential expression analysis RNAseq reads were trimmed with seqtk and mapped with Bowtie2 v2.3.2 to genomes of respective nasal isolates (‘very-sensitive-local’ mode) (Langmead and Salzberg 2012). Mapped data were further processed with SAMtools v0.1.18 to generate sorted BAM file format (Li et al.2009). BEDTools v2.26.0 was used to calculate read coverage of each annotated feature in the genome (Quinlan and Hall 2010). This was used as input for DESeq2 v1.16.1 to normalize the data and identify differentially expressed features at false discovery rate of 5% (fold change ≥ 1.5; P-adjusted value ≤ 0.05) (Love, Huber and Anders 2014). RNAseq reads were visualized in a strand-specific manner in the Artemis genome browser (Rutherford et al.2000). Identification of ncRNAs RNAseq read coverage for each nucleotide was calculated using BEDtools (Quinlan and Hall 2010). A Python script, named toRNAdo, was written in-house to identify putative ncRNAs based on the nucleotide coverage (https://github.com/pavsaz/toRNAdo). Both putative intergenic ncRNAs and those antisense to coding sequences were detected. The main criteria for ncRNA selection were: (i) normalized nucleotide expression value was above a defined expression threshold; (ii) ncRNA contained a defined expression ‘peak’, with at least a 5-to-1 expression ratio between the highest and lowest points of the ‘peak’; (iii) ncRNA was over 50 bp in size, in order to reduce the number of false positive hits and also due to most ncRNAs being reported above 50 bp in size (Gottesman and Storz 2011). As another measure for improving detection accuracy, only those putative ncRNAs that appeared in all replicates for each RNAseq condition were included. When combining ncRNAs present in multiple samples, a minimum start position and a maximum end position were used to define the new putative ncRNA length. RESULTS Nasal staphylococci strains isolated from healthy Danes To be able to investigate the microbial interactions occurring in the nasal cavities of healthy individuals, we isolated staphylococci strains from nasal swabs of Danish adults (Andersen et al.2012). We investigated a total of 108 nasal swabs, which were chosen because prior microbiome sequencing indicated the presence of staphylococcal strains (Liu et al.2015). We were able to isolate staphylococci strains from 43% of these samples (46/108), giving us a collection of 56 S. aureus and 27 S. epidermidis strains. The bacterial isolation was done by plating, while initial species identification was based on colony morphology and confirmed by Gram staining, PCR, and catalase and coagulase testing as previously described (Andersen et al.2012). Staphylococcus aureus and S. epidermidis were co-isolated from 7% of the nasal swabs (3/46) (Table 1). Table 1. Nasal swab samples where S. aureus and S. epidermidis are co-isolated: overview of the three samples from which both S. aureus and S. epidermidis strains were isolated. Sample Strain name Species (PCR tuf gene) Gram test Catalase test Coagulase test Esp gene (S. epidermidis) A 5A S. aureus G+ C+ Co+ NA 6A S. epidermidis G+ C+ Co− Negative B 9B S. aureus G+ C+ Co+ NA 8/9A S. epidermidis G+ C+ Co− esp C 39A S. aureus G+ C+ Co+ NA 39/40AB S. epidermidis G+ C+ Co− esp Sample Strain name Species (PCR tuf gene) Gram test Catalase test Coagulase test Esp gene (S. epidermidis) A 5A S. aureus G+ C+ Co+ NA 6A S. epidermidis G+ C+ Co− Negative B 9B S. aureus G+ C+ Co+ NA 8/9A S. epidermidis G+ C+ Co− esp C 39A S. aureus G+ C+ Co+ NA 39/40AB S. epidermidis G+ C+ Co− esp G+: Gram-positive; C+: catalase-positive; Co+: coagulase-positive; Co−: coagulase-negative; NA: not applicable. Negative for Esp means that we were unable to detect the esp gene. View Large The S. aureus strains were spa typed and 18 different spa types were identified, with t021 (11% of typed isolates) and t081 (11%) being the most prevalent. The spa-typed isolates were classified into 10 different clonal complexes, with most isolates belonging to the CC30 (21.4% of all isolates) and CC45 (17.9%). As the serine protease Esp has previously been implicated in the interaction between S. aureus and S. epidermidis, we used PCR to identify the esp gene in the S. epidermidis isolates. We detected the esp gene in 92.6% (25/27) of the nasal S. epidermidis strains. Sample B was chosen for detailed study as it included an esp-positive S. epidermidis (strain 8/9A), which is DT-218, and a S. aureus strain (strain 9B), which is ST-59 (Table 1). The strains were whole-genome sequenced and serotyping was based on these sequences. Staphylococcus aureus biofilm formation is not affected by S. epidermidis and vice versa To investigate the interaction phenotype of the nasal S. aureus and S. epidermidis strains, we assayed biofilm formation of the two strains in a microplate setup according to previous protocols (O’Toole 2011). Cells were allowed to form biofilms with the addition of cell-free supernatants and we found that the addition of cell-free supernatant from S. epidermidis did not alter biofilm formation of S. aureus in our setup (Fig. 1). The addition of cell-free S. aureus supernatant to S. epidermidis cells also did not affect biofilm formation properties. The same was true for the TSB medium supplemented with 1% glucose, except that the S. aureus growth was inhibited by addition of its own cell-free supernatant (Fig. S1 in the online supplemental material). Figure 1. View largeDownload slide Staphylococcus aureus biofilm formation is not affected by S. epidermidis supernatant and vice versa. Biofilm formation when grown with (+) and without (−) cell-free supernatants after 22 h of incubation at 37°C. Figure 1. View largeDownload slide Staphylococcus aureus biofilm formation is not affected by S. epidermidis supernatant and vice versa. Biofilm formation when grown with (+) and without (−) cell-free supernatants after 22 h of incubation at 37°C. A simple agar-based RNAseq setup reveals the interactome of S. aureus and S. epidermidis To map the interactome of S. aureus and S. epidermidis nasal isolates, we developed a simple agar-based setup where the strains were incubated either separately (mono-culture) or in close proximity to each other (co-culture) on the same agar plate, with cultures applied on filter paper (Fig. 2). Following the incubation, bacterial colonies were easily removed and prepared for downstream processing by transferring the filters to microfuge tubes. This enables separate processing and data analysis of the different strains and culture conditions. We used the agar-based RNAseq assay to determine the interactome of the nasal S. aureus and S. epidermidis pair when grown in co-culture compared to mono-cultures. Total RNA was isolated from four biological replicates for each condition: mono- vs. co-culture. We obtained from 26.5 to 61.8 million total reads per sample and the average percentage of reads mapped to genomes of respective nasal isolates was 83%. Figure 2. View largeDownload slide Co-culturing of staphylococci strains leads to altered gene expression. (A) Agar-based RNAseq setup used to map staphylococci interactome. Staphylococcus aureus and S. epidermidis are grown in mono- and co-culture and the setup allows for separate downstream processing of samples. (B) The DAVID database (Huang, Sherman and Lempicki 2008) was used to divide the DE genes into functional categories. Pie charts showing the distribution of functional categories of S. aureus and S. epidermidis DE genes. (C) Overview of DE genes. Cut-off: log2 fold change ≥ 1.5 and adjusted P-value > 0.05. Figure 2. View largeDownload slide Co-culturing of staphylococci strains leads to altered gene expression. (A) Agar-based RNAseq setup used to map staphylococci interactome. Staphylococcus aureus and S. epidermidis are grown in mono- and co-culture and the setup allows for separate downstream processing of samples. (B) The DAVID database (Huang, Sherman and Lempicki 2008) was used to divide the DE genes into functional categories. Pie charts showing the distribution of functional categories of S. aureus and S. epidermidis DE genes. (C) Overview of DE genes. Cut-off: log2 fold change ≥ 1.5 and adjusted P-value > 0.05. Staphylococcus aureus showed 10 significantly differentially expressed (DE) genes, which were all upregulated in co-culture with S. epidermidis (Table S1 in the online supplementary material, Fig. 2). Staphylococcus epidermidis showed 50 genes significantly DE in response to co-culture with S. aureus. Of these, 12 genes were upregulated and 38 downregulated (Table S1, Fig. 2). Seventy percent (7 of 10 DE genes) of the S. aureus genes upregulated during co-culture with S. epidermidis are related to the cell membrane and transmembrane transport according to the functional category analysis tool on the DAVID database (Huang, Sherman and Lempicki 2008). The remaining three genes fall into the categories: Biosynthesis of Antibiotics (SAOUHSC_00579), Pyruvate Metabolism (SAOUHSC_00533) and Unknown (SAOUHSC_02726). Two genes of the mnhABCDEFG operon, mnhA and mnhD, which encode subunits of a putative multi-subunit Na+/H+ antiporter (Hiramatsu et al.1998), are significantly upregulated in response to S. epidermidis. Analysis of the S. epidermidis DE genes using the DAVID database revealed that the genes fall into four main functional categories: Metabolic Pathways (20 of 50 DE genes), Infection Related (5/50), Membrane Associated (10/50) and Nucleotide/ATP-binding (15/50) (Fig. 2). Fifty-nine percent of the genes of the Metabolic Pathway category are part of the Biosynthesis of Amino Acids sub-category. Some of the genes of the ‘Infection Related’ category have previously been described as S. epidermidis virulence factors (Otto 2009). For example, it is evident that all genes of a putative operon SE0405-SE0407 of S. epidermidis are highly upregulated in co-culture with S. aureus, with fold-change values ranging from 7.8 to 9.0. This putative operon contains the previously described virulence factors sitA, sitB and sitC (Otto 2009), which encode an iron-regulated ABC-transporter (Cockayne et al.1998). Intriguingly, the Esp serine protease that has been shown to inhibit S. aureus colonization (Iwase et al.2010) is significantly downregulated in co-culture conditions (fold change = −2.3). Repertoire of non-coding RNAs in interacting staphylococci It has been previously shown that staphylococci strains possess a range of ncRNAs involved in gene regulation (Boisset et al.2007). We therefore attempted to identify the expression of ncRNAs in the nasal staphylococcal isolates during mono- and co-culture based on RNAseq data. Using a computational approach we identified 40 putative ncRNAs in the S. aureus nasal isolate and 150 putative ncRNAs in the S. epidermidis nasal isolate (Fig. 3, Materials and methods). Both intergenic and antisense ncRNAs were identified. 67% (100/150) and 63% (25/40) of the S. epidermidis and S. aureus ncRNAs, respectively, were identified in both mono- and co-culture conditions. The average size of ncRNAs is 598 bp for S. aureus and 1164 bp for S. epidermidis (Table S2 in the online supplementary material). Figure 3. View largeDownload slide ncRNA repertoire of nasal staphylococci strains. Venn diagram showing the putative ncRNAs identified in the nasal S. aureus and S. epidermidis strains during mono- and co-culture. Figure 3. View largeDownload slide ncRNA repertoire of nasal staphylococci strains. Venn diagram showing the putative ncRNAs identified in the nasal S. aureus and S. epidermidis strains during mono- and co-culture. By cross-referencing to the Staphylococcal Regulatory RNA Database (Sassi et al.2015), we noted that 13 of the ncRNAs identified in S. aureus in our setup overlap with those previously described in other studies. In fact, two of the ncRNAs identified here have been found to play a role when S. aureus is interacting with Pseudomonas aeruginosa (Miller et al.2017). However, as only a single study has reported on putative ncRNAs in S. epidermidis (Broach, Weiss and Shaw 2016), the overlap between previously described ncRNAs and those identified here is limited. In this study, we are specifically interested in the ncRNAs that show altered expression in response to co-culturing and we therefore performed DE analysis on the ncRNAs. We identified one DE ncRNA in S. aureus and seven DE ncRNAs in S. epidermidis (Table 2), with ncRNA lengths ranging from 365 bp to 4669 bp. As we saw in the gene expression analysis, there is also a tendency for upregulation of ncRNAs in S. aureus and downregulation in S. epidermidis. 75% (6/8) of the DE ncRNAs are encoded antisense to annotated genes and in three cases we see a correlation between the differential expression of the gene and its antisense ncRNA in the RNAseq data. All these examples display downregulation of both protein-coding gene and ncRNA (Fig. 4). One of these cases, ncRNA epi80, is transcribed antisense to the esp gene, which, as mentioned above, has been implicated in the interaction between S. aureus and S. epidermidis (Iwase et al.2010). The two remaining examples are ncRNA epi34, antisense to dltABCD, a previously described virulence factor (Otto 2009) responsible for d-alanylation of teichoic acids (Peschel et al.1999) and implicated in resistance to phage attacks (Fallico et al.2011), and epi116, which is antisense to a co-enzyme A disulphide reductase gene. Figure 4. View largeDownload slide The expression of antisense ncRNA epi80 in mono- and co-culture conditions in the nasal S. epidermidis strain. Paired-end RNAseq reads (blue and green) are shown mapped antisense to the esp gene (grey arrow). Genomic coordinates shown on the x-axis refer to the position in the genome of the nasal S. epidermidis strain. The number of reads mapping antisense to the esp gene is shown on the y-axis. Figure 4. View largeDownload slide The expression of antisense ncRNA epi80 in mono- and co-culture conditions in the nasal S. epidermidis strain. Paired-end RNAseq reads (blue and green) are shown mapped antisense to the esp gene (grey arrow). Genomic coordinates shown on the x-axis refer to the position in the genome of the nasal S. epidermidis strain. The number of reads mapping antisense to the esp gene is shown on the y-axis. Table 2. Co-culturing of staphylococci strains alters ncRNA transcription: overview of differentially expressed ncRNAs in the staphylococci strains. ncRNA Size (bp) Fold change Adj P-value Type Flanking/antisense gene(s)a Fold changeGene Differentially expressed S. aureus ncRNAs au15 457 2.3 1.05E-02 Antisense SAOUHSC_00846 NA Differentially expressed S. epidermidis ncRNAs epi17 1057 −3.2 1.76E-02 Antisense SE0302, rplJ, rplL Ns, Ns, Ns epi34 4669 −3.0 3.83E-02 Antisense SE0623, dltA, dltB, dltC, dltD Ns, −2.1, −2.1, −2.0, −2.4 epi52 638 −2.7 4.57E-02 Intergenic NAb Ns epi80 576 −2.7 1.30E-02 Antisense esp −2.3 epi105 680 −3.9 1.76E-02 Antisense SarZ Ns epi116 2629 −2.8 3.78E-02 Antisense SE2154, SE2155, fda −2.1, Ns, Ns epi135 365 −2.0 4.91E-02 Intergenic SE2336, SE2337 Ns ncRNA Size (bp) Fold change Adj P-value Type Flanking/antisense gene(s)a Fold changeGene Differentially expressed S. aureus ncRNAs au15 457 2.3 1.05E-02 Antisense SAOUHSC_00846 NA Differentially expressed S. epidermidis ncRNAs epi17 1057 −3.2 1.76E-02 Antisense SE0302, rplJ, rplL Ns, Ns, Ns epi34 4669 −3.0 3.83E-02 Antisense SE0623, dltA, dltB, dltC, dltD Ns, −2.1, −2.1, −2.0, −2.4 epi52 638 −2.7 4.57E-02 Intergenic NAb Ns epi80 576 −2.7 1.30E-02 Antisense esp −2.3 epi105 680 −3.9 1.76E-02 Antisense SarZ Ns epi116 2629 −2.8 3.78E-02 Antisense SE2154, SE2155, fda −2.1, Ns, Ns epi135 365 −2.0 4.91E-02 Intergenic SE2336, SE2337 Ns Cut-off: log2 fold change ≥ 1.5 and adjusted P-value > 0.05. Adj P-value: adjusted P-value, Ns: not significantly altered in RNAseq data, NA: not applicable. aAnnotations referring to genes of reference genomes S. aureus USA300 and S. epidermidis RP62a. bNo homolog found in reference genome. Where ncRNAs are antisense to more than one gene, all genes are listed. View Large DISCUSSION Previous studies have shown that virulence phenotypes such as biofilm formation and colonization of the nasal cavities by S. aureus are inhibited by the nasal commensal S. epidermidis (Iwase et al.2010). In this study, we applied RNAseq to obtain molecular insight into the functional consequences regarding effects on the transcriptome of this pathogen–commensal interaction. Our experimental setup involved co-cultivation of S. aureus and S. epidermidis isolates on agar surfaces. Although agar-based models do not accurately mimic the natural habitat, such models nevertheless enable systematic in-depth analysis of the mechanisms and genes that underlie microbe–microbe interactions as well as the associated molecular effects (Frydenlund Michelsen et al.2016). We also note that our in vitro study of the interaction between surface-associated staphylococci populations has some parallels to the bacterial organization found in the nasal cavities where colonizing bacteria are indeed surface-associated and growing in microcolonies (Morawska-Kochman et al.2017). Moreover, our experimental setup gives technical advantages as the downstream processing can be performed separately for each culture, aiding the data analysis. We recognize that bacterial colonies on agar surfaces are composed of clonal cells that can exist in many different physiological states depending on their spatial position. This population heterogeneity will most likely mean that we will not be able to identify subtle changes in gene expression, as these will be masked by noise. Despite these constraints, we were nevertheless able to capture new elements of the interaction response between nasal S. aureus and S. epidermidis isolates. We identified 10 differentially expressed genes in S. aureus and 50 in S. epidermidis when these organisms were co-cultured. It is likely that these interaction-responsive genes represent a subset of a larger transcriptional response that may be detected with the advent of more sensitive transcriptomic techniques, which enable analysis of spatially resolved transcriptomes from surface-associated microbial populations (Heacock-Kang et al.2017). It has previously been shown that exposure to S. epidermidis culture supernatant leads to downregulation of virulence in S. aureus (Otto et al.1999; Iwase et al.2010). In our experimental system, we show that the general response of S. aureus to S. epidermidis co-culturing is upregulation of gene expression, whereas the opposite trend is seen in S. epidermidis, where most altered genes are downregulated. Interestingly, the three-gene operon sitABC, which encodes an iron-regulated ABC-transporter involved in virulence in S. epidermidis (Cockayne et al.1998), is highly upregulated in co-culture with S. aureus. Yet, other S. epidermidis virulence factors are downregulated in co-culture conditions. The dltABCD operon (SE0624-SE0627) and mprF (SE1041), which are responsible for d-alanylation of teichoic acids (Peschel et al.1999) and lysylation of phospholipids, respectively (Peschel et al.2001), are all downregulated in co-culture. Overall, these results indicate that interactions between the nasal staphylococcal strains result in more complex responses than previously described and have the potential to modulate virulence in multiple ways. Adding to the complexity of the interaction-induced transcriptional changes, our analysis of putative ncRNAs in the nasal staphylococcal isolates revealed that ncRNAs also play a part in this pathogen–commensal interaction. ncRNAs have been identified in many different organisms to play a range of regulatory roles during transcription as well as translation (Gottesman and Storz 2011). A common staphylococcal ncRNA pool has been described, but different staphylococcal species also encode unique ncRNAs (Broach, Weiss and Shaw 2016). In our analysis, we find a substantial overlap with previously identified ncRNAs of S. aureus, though a more modest overlap with S. epidermidis ncRNAs. This probably reflects the fact that ncRNAs in S. epidermidis are generally under-studied, and our setup is very different from the single previous study on the subject (Broach, Weiss and Shaw 2016). Analysis of more strains and conditions is required to capture a more complete spectrum of ncRNAs in S. epidermidis. However, in the current setup we identify two ncRNAs in S. aureus that have previously been shown to play a part during interspecies interactions (Miller et al.2017). Despite general interest in ncRNAs and their regulatory roles, not much is known about the role of ncRNAs in connection to microbe–microbe interactions (Miller et al.2017). In our study, we focused on ncRNAs differentially expressed in co-culture conditions and identified eight staphylococcal ncRNAs that play potentially important roles during interaction. Most of these ncRNAs are antisense to protein-encoding genes and in three cases the differentially expressed (antisense) ncRNAs followed the differential expression of their corresponding (sense) protein-encoding genes. For example, we identify a previously undescribed ncRNA (epi80), which is antisense to the esp gene in S. epidermidis. Expression of both the esp gene and the corresponding epi80 antisense ncRNA is downregulated in response to interaction. Previous studies have shown that the Esp serine protease plays a central role in prevention of nasal S. aureus colonization and that the purified protease inhibits biofilm formation and disassembles previously formed biofilms (Iwase et al.2010). Our finding that both esp and the epi80 antisense ncRNA are downregulated when S. epidermidis is co-cultured with S. aureus points towards an ncRNA-based positive regulatory mechanism for esp expression, which could explain why we do not observe an inhibitory effect on S. aureus biofilm grown with supernatant of the nasal S. epidermidis strain. Other studies have found that ncRNAs can have positive effects on gene regulation (Fröhlich and Vogel 2009). For example, antisense RNAs can protect mRNA from degradation by RNase E by masking endonuclease recognition sites (Stazic, Lindell and Steglich 2011). Future experiments will address if the epi80 antisense ncRNA functions in a similar way to regulate expression of esp. Our results, together with the notion that esp is not expressed in all S. epidermidis strains (Fredheim et al.2015), indicate that the observed inhibitory effect on S. aureus biofilm formation might be more complex than previously assumed. We recognize the possibility that our biofilm formation setup would not reveal effects that are exclusively induced during co-culture of the staphylococcal strains. It is possible that clonal variation of the staphylococcal strains and differences in experimental setups are all contributing factors, and further studies are needed to fully determine the factors that control esp expression. Moreover, our data show a self-inflicting inhibition of S. aureus when glucose is added to the medium, indicating difficulties in this experimental setup. The self-inflicting inhibition could be due to external alterations such as changes in pH of the medium. With this study, we provide the first attempt to identify changes in gene expression as a result of interaction between S. aureus and S. epidermidis isolated from the nasal cavities of healthy individuals. We identified clear effects on the transcriptomic profile of both interacting partners and discovered ncRNAs potentially involved in interspecies interactions. Further studies are needed to determine the role of the differentially expressed genes and ncRNAs and to map the interactome in more detail. A comprehensive understanding of how microbe–microbe interactions change the physiology and gene expression of pathogenic bacteria may contribute to the identification of novel targets of interference. SUPPLEMENTARY DATA Supplementary data are available at FEMSLE online. Acknowledgements We acknowledge Marlene Danner Dalgaard and Anna Koza for assistance in the sequencing of staphylococcal isolates. FUNDING The Danish Council for Independent Research (Grant number 6108-00300A) and the Villum Foundation (Grant number VKR023113) supported this work. Conflict of interest. None declared. REFERENCES Andersen PS, Pedersen JK, Fode P et al. Influence of host genetics and environment on nasal carriage of Staphylococcus aureus in Danish middle-aged and elderly twins. J Infect Dis 2012; 206: 1178– 84. Google Scholar CrossRef Search ADS PubMed Bankevich A, Nurk S, Antipov D et al. SPAdes: A new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol 2012; 19: 455– 77. Google Scholar CrossRef Search ADS PubMed Boisset S, Geissmann T, Huntzinger E et al. Staphylococcus aureus RNAIII coordinately represses the synthesis of virulence factors and the transcription regulator Rot by an antisense mechanism. Genes Dev 2007; 21: 1353– 66. Google Scholar CrossRef Search ADS PubMed Briard B, Heddergott C, Latgé J-P. Volatile compounds emitted by Pseudomonas aeruginosa stimulate growth of the fungal pathogen Aspergillus fumigatus. MBio 2016; 7: e00219. Google Scholar CrossRef Search ADS PubMed Broach WH, Weiss A, Shaw LN. Transcriptomic analysis of staphylococcal sRNAs: insights into species-specific adaption and the evolution of pathogenesis. Microb Genom 2016; 2: e000065. Google Scholar PubMed Carroll RK, Weiss A, Broach WH et al. Genome-wide annotation, identification, and global transcriptomic analysis of regulatory or small RNA gene expression in Staphylococcus aureus. MBio 2016; 7: e01990– 15. Google Scholar CrossRef Search ADS PubMed Cockayne A, Hill PJ, Powell NBL et al. Molecular cloning of a 32-kilodalton lipoprotein component of a novel iron-regulated Staphylococcus epidermidis ABC transporter. Infect Immun 1998; 66: 3767– 74. Google Scholar PubMed Delgado S, Arroyo R, Jiménez E et al. Staphylococcus epidermidis strains isolated from breast milk of women suffering infectious mastitis: potential virulence traits and resistance to antibiotics. BMC Microbiol 2009; 9: 82. Google Scholar CrossRef Search ADS PubMed Eberl L, Tümmler B. Pseudomonas aeruginosa and Burkholderia cepacia in cystic fibrosis: genome evolution, interactions and adaptation. Int J Med Microbiol 2004; 294: 123– 31. Google Scholar CrossRef Search ADS PubMed Fallico V, Ross RP, Fitzgerald GF et al. Genetic response to bacteriophage infection in Lactococcus lactis reveals a four-strand approach involving induction of membrane stress proteins, d-alanylation of the cell wall, maintenance of proton motive force, and energy conservation. J Virol 2011; 85: 12032– 42. Google Scholar CrossRef Search ADS PubMed Felden B, Vandenesch F, Bouloc P et al. The Staphylococcus aureus RNome and its commitment to virulence. PLoS Pathog 2011; 7: e1002006. Google Scholar CrossRef Search ADS PubMed Fode P, Stegger M, Andersen PS. Human β-defensin 3 (DEFB103) and its influence on Staphylococcus aureus nasal carriage. Int J Infect Dis 2011; 15: e388– 94. Google Scholar CrossRef Search ADS PubMed Fredheim EGA, Flægstad T, Askarian F et al. Colonisation and interaction between S. epidermidis and S. aureus in the nose and throat of healthy adolescents. Eur J Clin Microbiol Infect Dis 2015; 34: 123– 9. Google Scholar CrossRef Search ADS PubMed Frydenlund Michelsen C, Hossein Khademi SM, Krogh Johansen H et al. Evolution of metabolic divergence in Pseudomonas aeruginosa during long-term infection facilitates a proto-cooperative interspecies interaction. ISME J 2016; 10: 1323– 36. Google Scholar CrossRef Search ADS PubMed Fröhlich KS, Vogel J. Activation of gene expression by small RNA. Curr Opin Microbiol 2009; 12: 674– 82. Google Scholar CrossRef Search ADS PubMed Fugère A, Lalonde Séguin D, Mitchell G et al. Interspecific small molecule interactions between clinical isolates of Pseudomonas aeruginosa and Staphylococcus aureus from adult cystic fibrosis patients. PLoS One 2014; 9: e86705. Google Scholar CrossRef Search ADS PubMed Fukuda S, Toh H, Hase K et al. Bifidobacteria can protect from enteropathogenic infection through production of acetate. Nature 2011; 469: 543– 7. Google Scholar CrossRef Search ADS PubMed Gottesman S, Storz G. Bacterial small RNA regulators: versatile roles and rapidly evolving variations. Cold Spring Harb Perspect Biol 2011; 3: a003798. Google Scholar CrossRef Search ADS PubMed Heacock-Kang Y, Sun Z, Zarzycki-Siek J et al. Spatial transcriptomes within the Pseudomonas aeruginosa biofilm architecture. Mol Microbiol 2017; 106: 976– 85. Google Scholar CrossRef Search ADS PubMed Hiramatsu T, Kodama K, Kuroda T et al. A putative multisubunit Na+/H+ antiporter from Staphylococcus aureus. J Bacteriol 1998; 180: 6642– 8. Google Scholar PubMed Huang DW, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 2008; 4: 44– 57. Google Scholar CrossRef Search ADS Humphries J, Xiong L, Liu J et al. Species-independent attraction to biofilms through electrical signaling. Cell 2017; 168: 200– 209.e12. Google Scholar CrossRef Search ADS PubMed Ikeda Y, Ohara-Nemoto Y, Kimura S et al. PCR-based identification of Staphylococcus epidermidis targeting gseA encoding the glutamic-acid-specific protease. Can J Microbiol 2004; 50: 493– 8. Google Scholar CrossRef Search ADS PubMed Iwase T, Uehara Y, Shinji H et al. Staphylococcus epidermidis Esp inhibits Staphylococcus aureus biofilm formation and nasal colonization. Nature 2010; 465: 346– 9. Google Scholar CrossRef Search ADS PubMed Koeppen K, Hampton TH, Jarek M et al. A novel mechanism of host-pathogen interaction through sRNA in bacterial outer membrane vesicles. PLoS Pathog 2016; 12: e1005672. Google Scholar CrossRef Search ADS PubMed Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods 2012; 9: 357– 9. Google Scholar CrossRef Search ADS PubMed Larsen M V., Cosentino S, Rasmussen S et al. Multilocus sequence typing of total-genome-sequenced bacteria. J Clin Microbiol 2012; 50: 1355– 61. Google Scholar CrossRef Search ADS PubMed Lavigne J-P, Nicolas-Chanoine M-H, Bourg G et al. Virulent synergistic effect between Enterococcus faecalis and Escherichia coli assayed by using the Caenorhabditis elegans model. PLoS One 2008; 3: e3370. Google Scholar CrossRef Search ADS PubMed Li H, Handsaker B, Wysoker A et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 2009; 25: 2078– 9. Google Scholar CrossRef Search ADS PubMed Liu CM, Price LB, Hungate BA et al. Staphylococcus aureus and the ecology of the nasal microbiome. Sci Adv 2015; 1: e1400216. Google Scholar CrossRef Search ADS PubMed Liu H, Wang X, Wang H-D et al. Escherichia coli noncoding RNAs can affect gene expression and physiology of Caenorhabditis elegans. Nat Commun 2012; 3: 1073. Google Scholar CrossRef Search ADS PubMed Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 2014; 15: 550. Google Scholar CrossRef Search ADS PubMed Miller CL, Van Laar TA, Chen T et al. Global transcriptome responses including small RNAs during mixed-species interactions with methicillin-resistant Staphylococcus aureus and Pseudomonas aeruginosa. Microbiologyopen 2017; 6: e00427. Google Scholar CrossRef Search ADS Morawska-Kochman M, Marycz K, Jermakow K et al. The presence of bacterial microcolonies on the maxillary sinus ciliary epithelium in healthy young individuals. PLoS One 2017; 12: e0176776. Google Scholar CrossRef Search ADS PubMed Moss B, Squire J, Topley E. Nose and skin carriage of Staphylococcus aureus in patients receiving penicillin. Lancet 1948; 251: 320– 5. Google Scholar CrossRef Search ADS Ngba Essebe C, Visvikis O, Fines-Guyon M et al. Decrease of Staphylococcus aureus virulence by Helcococcus kunzii in a Caenorhabditis elegans model. Front Cell Infect Microbiol 2017; 7: 77. Google Scholar CrossRef Search ADS PubMed O’Toole GA. Microtiter dish biofilm formation assay. J Vis Exp 2011, DOI: https://doi.org/10.3791/2437. Otto M. Staphylococcus epidermidis—the ‘accidental’ pathogen. Nat Rev Microbiol 2009; 7: 555– 67. Google Scholar CrossRef Search ADS PubMed Otto M, Süßmuth R, Vuong C et al. Inhibition of virulence factor expression in Staphylococcus aureus by the Staphylococcus epidermidis agr pheromone and derivatives. FEBS Lett 1999; 450: 257– 62. Google Scholar CrossRef Search ADS PubMed Peschel A, Jack RW, Otto M et al. Staphylococcus aureus resistance to human defensins and evasion of neutrophil killing via the novel virulence factor MprF is based on modification of membrane lipids with l-lysine. J Exp Med 2001; 193: 1067– 76. Google Scholar CrossRef Search ADS PubMed Peschel A, Otto M, Jack RW et al. Inactivation of the dlt operon in Staphylococcus aureus confers sensitivity to defensins, protegrins, and other antimicrobial peptides. J Biol Chem 1999; 274: 8405– 10. Google Scholar CrossRef Search ADS PubMed Prindle A, Liu J, Asally M et al. Ion channels enable electrical communication in bacterial communities. Nature 2015; 527: 59– 63. Google Scholar CrossRef Search ADS PubMed Quinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 2010; 26: 841– 2. Google Scholar CrossRef Search ADS PubMed Rice P, Longden I, Bleasby A. EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet 2000; 16: 276– 7. Google Scholar CrossRef Search ADS PubMed Rissman AI, Mau B, Biehl BS et al. Reordering contigs of draft genomes using the Mauve Aligner. Bioinformatics 2009; 25: 2071– 3. Google Scholar CrossRef Search ADS PubMed Rutherford K, Parkhill J, Crook J et al. Artemis: sequence visualization and annotation. Bioinformatics 2000; 16: 944– 5. Google Scholar CrossRef Search ADS PubMed Sassi M, Augagneur Y, Mauro T et al. SRD: a Staphylococcus regulatory RNA database. RNA 2015; 21: 1005– 17. Google Scholar CrossRef Search ADS PubMed Sassone-Corsi M, Raffatellu M. No vacancy: how beneficial microbes cooperate with immunity to provide colonization resistance to pathogens. J Immunol 2015; 194: 4081– 7. Google Scholar CrossRef Search ADS PubMed Seemann T. Prokka: rapid prokaryotic genome annotation. Bioinformatics 2014; 30: 2068– 9. Google Scholar CrossRef Search ADS PubMed Stazic D, Lindell D, Steglich C. Antisense RNA protects mRNA from RNase E degradation by RNA-RNA duplex formation during phage infection. Nucleic Acids Res 2011; 39: 4890– 9. Google Scholar CrossRef Search ADS PubMed Sugimoto S, Iwamoto T, Takada K et al. Staphylococcus epidermidis Esp degrades specific proteins associated with Staphylococcus aureus biofilm formation and host-pathogen interaction. J Bacteriol 2013; 195: 1645– 55. Google Scholar CrossRef Search ADS PubMed Vandecandelaere I, Depuydt P, Nelis HJ et al. Protease production by Staphylococcus epidermidis and its effect on Staphylococcus aureus biofilms. Pathog Dis 2014; 70: 321– 31. Google Scholar CrossRef Search ADS PubMed Wertheim HF, Melles DC, Vos MC et al. The role of nasal carriage in Staphylococcus aureus infections. Lancet Infect Dis 2005; 5: 751– 62. Google Scholar CrossRef Search ADS PubMed Westermann AJ, Förstner KU, Amman F et al. Dual RNA-seq unveils noncoding RNA functions in host–pathogen interactions. Nature 2016; 529: 496– 501. Google Scholar CrossRef Search ADS PubMed Wigneswaran V, Amador CI, Jelsbak L et al. Utilization and control of ecological interactions in polymicrobial infections and community-based microbial cell factories. F1000Res 2016; 5: 421. Google Scholar CrossRef Search ADS Zhou G, Zhou Y, Chen X. New insight into inter-kingdom communication: horizontal transfer of mobile small RNAs. Front Microbiol 2017; 8: 768. Google Scholar CrossRef Search ADS PubMed © FEMS 2018. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org
FEMS Microbiology Letters – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera