The research team, led by David Broderick with Arne Koefoed, Steffen Albrecht, Richard Douglas and Kristi Biswas, grew eight clinically relevant bacterial species under carefully optimised conditions and sequenced their transcriptomes across four experimental states: planktonic growth, untreated established biofilm, biofilm treated with doxycycline, and biofilm treated with Zoono GermFree24, an aqueous non-alcoholic biocidal hand sanitiser in common use. The panel spanned methicillin-susceptible and methicillin-resistant Staphylococcus aureus, Pseudomonas aeruginosa, Staphylococcus epidermidis, Moraxella catarrhalis, Streptococcus pneumoniae, and two commensal taxa, Corynebacterium accolens and Dolosigranulum pigrum, which are more often reported to suppress pathogenicity than to build biofilms themselves. Doxycycline was applied at 0.27 micrograms per millilitre, a concentration chosen to reflect reported sinonasal mucus levels after oral administration, while Zoono exposure durations ranged from 20 seconds to 5 minutes depending on species, calibrated in preliminary experiments to produce measurable but non-complete reductions in viability.

Technically, the study sequenced 32 RNA samples, one for each species-condition combination, on an Illumina NovaSeq X Plus platform after library preparation with a ribosomal RNA depletion kit. Samples yielded a median of 85.1 million fragments, with an average alignment rate of 73.1 percent, indicating robust transcriptome coverage. Because no biological replication was available within species-condition combinations, the authors deliberately avoided formal differential expression testing and false-discovery-rate inference. Instead, they ranked genes within each species by the magnitude of transcriptional change and retained the top 10 percent as fold-change-prioritised candidates, an exploratory strategy designed to capture the strongest responses within each organism for downstream cross-species comparison rather than to claim statistical significance.

The cross-species comparison rested on ortholog clustering. Protein sequences predicted with Prodigal were clustered using MMseqs2 at a minimum threshold of 30 percent sequence identity and 50 percent coverage, a compromise chosen to detect related protein families across phylogenetically diverse airway bacteria despite substantial sequence divergence. Sensitivity analyses at stricter thresholds of 40 percent identity with 60 percent coverage and 50 percent identity with 70 percent coverage confirmed that the broad conclusions were not an artefact of a single clustering scheme. Functional annotation drew on Swiss-Prot and Pfam databases, with genes grouped into broad biological processes covering metabolism and energy, transport, information processing, regulation and stress, cell envelope and membrane functions, and poorly characterised other functions. Directional agreement, defined as the proportion of species sharing the majority direction of change within an ortholog cluster, was then quantified and compared against a binomial null expectation based on random assignment of transcriptional direction.

The clustering analysis identified 1,503 ortholog clusters shared across multiple species, while the fold-change approach defined 6,991 candidate genes within species. Only 875 genes were supported by both approaches, forming a high-confidence subset with evidence of both strong expression change and cross-species conservation. Gene-level conservation proved restricted: just 29 clusters were detected across all eight taxa, and under the primary clustering threshold 823 raw clusters were present in at least five species while 248 were shared across all eight. Increasing stringency to 50 percent identity and 70 percent coverage reduced these figures to 277 and 48 clusters respectively, yet 138 candidate-associated clusters remained linked to clusters present in at least five species even at the strictest threshold, showing that cross-species structure survives a range of orthology definitions.

When the researchers examined the transition from planktonic to biofilm growth, they found widespread but generally modest transcriptional changes across all species, touching metabolism and energy, transport, regulation and stress, and information processing. The relative contribution of these functional categories varied substantially between species. Corynebacterium accolens showed enrichment in transport and cell envelope-associated functions, while Moraxella catarrhalis and Staphylococcus epidermidis upregulated regulation and stress pathways. Pseudomonas aeruginosa, Dolosigranulum pigrum and Streptococcus pneumoniae displayed more distributed responses without a single dominant category, and the two S. aureus strains were similarly variable. Conserved gene-level responses were limited and primarily associated with transport and metabolic functions, with even these shared clusters showing modest effect sizes and incomplete directional agreement.

Perturbation of established biofilms told a very different story. Responses to both doxycycline and Zoono were far more heterogeneous than biofilm formation responses, with substantial variation in magnitude and direction across species. Corynebacterium accolens showed decreases in transport and membrane-associated functions, Streptococcus pneumoniae exhibited transport increases alongside stress-pathway reductions, Pseudomonas aeruginosa reduced information processing activity, and Staphylococcus epidermidis increased cell envelope and membrane responses. Doxycycline, consistent with its mechanism as a tetracycline-class antibiotic that binds the bacterial 30S ribosomal subunit and inhibits protein synthesis, was more frequently associated with changes in information processing functions, whereas Zoono induced broader but less consistently directed shifts across functional categories. Despite these differences in emphasis, both perturbations converged in their lack of cross-species conservation.

The quantitative heart of the study lies in the directional agreement analysis, and the numbers are sobering. Mean directional agreement differed only slightly from values expected under random direction assignment: 70.3 percent observed versus 67.2 percent expected for biofilm formation, 69.0 percent versus 67.2 percent for doxycycline, and 66.0 percent versus 67.2 percent for Zoono. Strongly conserved transcriptional responses were uncommon despite shared functional themes, and most ortholog clusters fell within intermediate agreement ranges. Even broadly shared orthologs, such as those associated with ATP-dependent helicases and membrane transport systems, were rarely regulated uniformly across species, reinforcing the conclusion that individual candidate genes should be read as indicators of broader functional trends rather than definitive markers of conserved responses.

The authors interpret this split through a conceptual framework of functional constraint and regulatory flexibility. Biofilm formation represents a relatively constrained adaptive state in which shared environmental pressures, including surface attachment, matrix production and nutrient limitation, narrow the range of viable adaptive states and drive partial convergence at the level of broad functional processes. Perturbation relaxes those constraints, allowing species-specific regulatory networks to resolve stress through multiple organism-specific strategies, including efflux modulation, altered membrane composition, stress-response regulator activation and metabolic state changes. This balance aligns with ecological models of microbial functional redundancy, in which similar community-level functions can be maintained despite variation in the taxa, genes or regulatory pathways responsible.

The implications for treating chronic biofilm-associated infections are significant. Because biofilm establishment involves partially conserved functional adaptation, broadly targetable pathways may exist for disrupting biofilm formation across pathogens. But the species-specific nature of stress responses suggests that broadly effective interventions against established biofilms will be harder to find, and the absence of a conserved transcriptional response to perturbation raises the possibility that sublethal or transient antimicrobial exposures, common in clinical settings, may promote heterogeneous adaptive states within microbial communities. In polymicrobial airway biofilms, where neighbouring taxa influence architecture, metabolic activity and antimicrobial susceptibility, mixed communities may contain multiple partially independent stress-response programs that resist disruption through any single conserved molecular target. The authors caution that their findings derive from monoculture biofilms without biological replication, detailed biofilm architecture validation or equivalent pharmacodynamic exposures across species, and they call for future work integrating transcriptomics with proteomic, metabolomic and phenotypic measurements, extending to defined mixed-species biofilms and ex vivo or in vivo systems, and pairing comparative transcriptomics with spatial imaging and species-resolved abundance measurements to determine whether the conserved functional responses identified here are maintained, amplified or suppressed within real polymicrobial communities.

Subject of Research: Comparative transcriptomics of biofilm formation and antimicrobial perturbation across eight airway bacterial species

Article Title: Comparative RNA-Seq reveals conserved and species-specific biofilm-associated pathways across clinically relevant airway bacteria

Article References: Broderick, D., Koefoed, A., Albrecht, S., Douglas, R., & Biswas, K. (2026). Comparative RNA-Seq reveals conserved and species-specific biofilm-associated pathways across clinically relevant airway bacteria. Frontiers in Bacteriology, 5, Article 1892414. https://doi.org/10.3389/fbrio.2026.1892414

Image Credits: AI Generated

DOI: 10.3389/fbrio.2026.1892414

Keywords: biofilm, RNA sequencing, airway pathogens, comparative transcriptomics, antimicrobial perturbation, chronic rhinosinusitis, Comparative, RNA-Seq, reveals, conserved, species-specific, biofilm-associated