Selective Peptide Defence Landscapes: Ecological Distribution of Bacteriocin Defence Systems in Oral Streptococci

1. Introduction

Bacteriocins are ribosomally synthesised antimicrobial peptides or proteins produced by bacteria to inhibit competing microorganisms occupying the same ecological niche. Since the landmark review of Tagg, Dajani and Wannamaker, bacteriocins have been recognised as important mediators of interference competition among closely related bacteria and as major determinants of microbial population dynamics [1]. Subsequent studies have demonstrated remarkable structural and functional diversity among bacteriocins, particularly within the lantibiotic family, and have revealed their important contributions to microbial colonisation, niche establishment and community stability [2,3,4].

Among the best-characterised examples are the salivaricins produced by oral isolates of Streptococcus salivarius. These lantibiotic peptides have attracted considerable attention because of their roles in maintaining oral microbial homeostasis and because bacteriocin-producing S. salivarius strains, particularly K12 and M18, have been successfully developed as oral probiotics [6,7,8,9]. More broadly, studies of bacteriocin-producing streptococci have provided important insights into microbial competition, ecological fitness and probiotic selection. Hols et al. further emphasised that bacteriocins produced by commensal organisms such as S. salivarius can contribute to local microbial competition and may be harnessed to reshape endogenous microbiota, reinforcing the relevance of bacteriocin activity to community-level ecological organisation [5].

The ecological consequences of bacteriocin production depend not only on inhibitory peptides themselves but also on the distribution of peptide defence systems within neighbouring microorganisms. Bacteriocin-producing organisms typically possess dedicated peptide defence systems that protect them from their own inhibitory peptides, while related bacteria may acquire peptide-specific defence systems through a variety of evolutionary processes. Consequently, susceptibility to bacteriocins is frequently highly selective rather than reflecting generalised resistance to antimicrobial peptides. Many bacteriocins show greatest activity against phylogenetically related organisms; among streptococci, inhibitory spectra commonly extend most strongly to other streptococcal species, making the distribution of selective susceptibility especially relevant to oral streptococcal ecology.

In parallel with advances in bacteriocin biology, increasing attention has been directed towards the mechanisms by which bacteria resist host and microbial antimicrobial peptides. Numerous studies have identified cell-envelope modifications, membrane remodelling, peptide sequestration, efflux systems and regulatory networks that contribute to antimicrobial peptide resistance [10,11,12]. These investigations have substantially advanced understanding of host–pathogen interactions and antimicrobial resistance. Nevertheless, they have focused primarily on resistance mechanisms operating within individual organisms rather than on the ecological distribution of peptide defence systems across microbial communities.

Recent genomic investigations have further highlighted close links between bacteriocin-associated loci, competence pathways and adaptive responses in oral streptococci. Comparative analyses have revealed considerable diversity in bacteriocin loci, frequent horizontal transfer of peptide-associated genes and coordinated regulation of bacteriocin production with other ecological traits. Recent genomic and transcriptomic studies of Streptococcus anginosus have identified coordinated expression of bacteriocin-associated genes, competence systems and virulence-associated determinants, emphasising the integration of antimicrobial systems within broader adaptive networks [13].

The biological significance of bacteriocin-associated loci may extend beyond antimicrobial activity. Salivaricin peptides have been shown to function as intra- and interspecies signalling molecules, indicating roles in microbial communication as well as inhibition [14]. Furthermore, components of the Streptococcus pyogenes sal locus contribute to virulence-associated phenotypes, including intracellular survival and resistance to host antimicrobial mechanisms, suggesting that peptide defence systems may also be recruited for host adaptation [15]. Subsequent studies demonstrated that expression of the sal locus is controlled by multiple promoters and responds to environmental signals, consistent with these loci functioning as adaptive regulatory modules rather than constitutively expressed antimicrobial systems [16].

A recurring observation emerging from studies of oral streptococci is that peptide susceptibility patterns are frequently highly specific. Some organisms exhibit resistance to individual bacteriocins while remaining susceptible to closely related inhibitory peptides, suggesting that peptide defence systems are distributed selectively within microbial populations. Despite numerous investigations of bacteriocin production and immunity, comparatively little attention has been directed towards the ecological significance of these selective susceptibility patterns.

The origins of comparative streptococcal antagonism testing were explicitly linked to both bacteriocin production and susceptibility. Tagg and Bannister developed a “fingerprinting” scheme [17] in which streptococci were characterised according to their production of bacteriocin-like inhibitory substances (P typing) and their sensitivity to inhibitory activity produced by a defined set of standard producer strains (S typing). Nine producer strains with differing inhibitory spectra were selected for S typing, recognising susceptibility as an informative characteristic of the test organism as well as a means of detecting producer activity. Subsequent applications of deferred-antagonism testing increasingly emphasised the detection and characterisation of bacteriocin production, while the susceptibility component came to function predominantly as an indicator of producer activity. The present study revisits the earlier S-typing principle and asks whether comparative susceptibility profiles themselves constitute reproducible biological phenotypes with ecological significance. The study design and relationship between comparative susceptibility profiling and the SPDL framework are summarised in Figure 1.

The ecological significance of susceptibility was subsequently examined under natural conditions by Tompkins and Tagg [18]. They identified individuals who consistently harboured exceptionally high proportions of bacteriocin-producing S. salivarius and found that the accompanying Gram-positive alpha-haemolytic cocci contained significantly higher proportions of organisms resistant to the inhibitory activity of the resident producer strains than did organisms recovered from control individuals. The authors interpreted these findings as consistent with selection of a relatively bacteriocin-resistant accompanying microbiota by active bacteriocin production in situ. Importantly, their study also retained the comparative susceptibility dimension of the earlier fingerprinting approach: inhibitory spectra were assessed against a defined panel of ten indicator organisms, and reproducible differences in inhibition-zone widths were recognised as part of the susceptibility phenotype. Their panel included FF22 and strain 71-679 (M4, T-pattern 4), providing an early experimental precedent for treating selective susceptibility patterns as ecologically informative rather than merely as a means of detecting producer activity.

In the present study, we compare susceptibility phenotypes of representative Streptococcus anginosus, Streptococcus mitis and Streptococcus pyogenes isolates using a panel of well-characterised bacteriocin-producing streptococci, including Streptococcus pyogenes FF22, representing diverse lantibiotic and non-lantibiotic inhibitory systems. On the basis of these comparative phenotypic observations, together with published genomic and ecological evidence, we propose the concept of Selective Peptide Defence Landscapes (SPDLs). We suggest that consideration of the ecological distribution of peptide defence systems, alongside bacteriocin production itself, provides a useful framework for understanding microbial competition, coexistence and community organisation within the oral microbiome. We use the term Selective Peptide Defence Landscape (SPDL) to describe the characteristic pattern of susceptibility and resistance exhibited by an indicator organism across a defined repertoire of peptide-mediated challenges. The SPDL concept extends the historical S-typing principle from a strain-discrimination tool to an ecological framework for interpreting peptide susceptibility.

2. Materials and Methods

2.1. Bacterial Strains

Representative isolates of Streptococcus anginosus, Streptococcus mitis and Streptococcus pyogenes were examined for susceptibility to a panel of ten bacteriocin-like inhibitory substance (BLIS)-producing streptococci selected to represent diverse peptide-mediated inhibitory systems. The final producer panel comprised Streptococcus salivarius K12 (salivaricins A2 and B), S. salivarius M18 (salivaricins A2, 9, MPS and M), S. pyogenes strain 71–679 (Ind 5; M4, T-pattern 4; SalA1-associated), S. pyogenes emm86 (salivaricin A-associated activity), Streptococcus mitis SK648 (salivaricin B-associated activity), S. pyogenes FF22 (Ind 2; streptococcin A-FF22), S. pyogenes M76 (streptin), S. pyogenes strain 71–724 (M57; SA-M57), S. salivarius strain 36 (broad-spectrum BLIS activity), and S. salivarius strain 9 (salivaricin 9-associated activity).

Producer strains were selected as historically well-characterised reference strains representing diverse streptococcal peptide-mediated inhibitory systems rather than as a comprehensive survey of bacteriocin-producing streptococci. The panel was retained as a comparative phenotypic reference set, allowing susceptibility patterns to be examined across three indicator groups: S. anginosus, S. mitis and S. pyogenes. Characteristics of the producer strains, together with their principal inhibitory peptides and supporting references, are summarised in Supplementary Table S1.

2.2. Deferred Antagonism Assay

BLIS activity was assessed using the deferred antagonism method of Tagg and Bannister [17], with minor modifications. The present implementation follows the same producer-versus-indicator logic used in the subsequent ecological study of bacteriocinogenic S. salivarius by Tompkins and Tagg [18], in which susceptibility to a defined panel of indicator organisms was used to examine the ecological consequences of naturally occurring bacteriocin production. Producer strains were grown as diametric streaks on Columbia sheep blood agar and incubated at 37 °C in 5% CO₂ for 18 h. Following incubation, producer growth was removed and the agar surface sterilised by exposure to chloroform vapour. Indicator strains were then streaked perpendicular to the original producer growth and the plates reincubated under identical conditions.

Zones of inhibition were scored semi-quantitatively as absent (−), weak (+), moderate (++), or strong (+++). Representative assays were repeated on separate occasions to confirm reproducibility of the observed susceptibility phenotypes.

2.3. Comparative Analysis

The susceptibility matrix generated from the deferred antagonism assays comprised ten producer strains tested against 46 indicators: nine historical reference indicators (I1–I9), 12 S. anginosus isolates (A1–A12), 12 S. mitis isolates (M1–M12), and 13 S. pyogenes isolates (P1–P13). The matrix was used to compare qualitative susceptibility phenotypes among the three species groups. The objective of the analysis was not to quantify bacteriocin potency, but to identify reproducible patterns of selective susceptibility across distinct inhibitory systems.

These comparative phenotypic observations were subsequently interpreted in the context of published information concerning bacteriocin-associated loci, peptide defence systems and streptococcal ecology to develop the conceptual framework presented in Figure 2.

2.4. Generative AI Use

ChatGPT (OpenAI; GPT-5.6) was used during manuscript preparation for language editing, conceptual organisation and critical review of the manuscript, including identification of literature for subsequent verification. It was not used to generate or alter experimental data, perform the susceptibility analysis, or make decisions about the reported experimental results. All AI-assisted output was reviewed and edited by the author, who takes full responsibility for the content of this publication.

3. Results

3.1. Comparative Susceptibility Patterns Support the Selective Peptide Defence Landscapes Concept

The comparative susceptibility matrix is presented in Figure 3. Clear and reproducible differences in peptide susceptibility were observed among the three species examined. Rather than exhibiting uniform susceptibility or resistance to all inhibitory systems, isolates displayed reproducible combinations of resistance and susceptibility that were characteristic of individual peptide defence systems.

The S. anginosus isolates showed a particularly distinctive phenotype. Across the final producer panel, susceptibility was strongly dependent on the inhibitory system rather than on producer species alone. SalA-associated producers, including S. pyogenes strain 71–679 and emm86, produced little or no inhibition of the S. anginosus isolates, whereas the same isolates remained susceptible to several alternative systems, most notably salivaricin B-associated activity and streptococcin A-FF22. Susceptibility to streptin and SA-M57 varied among isolates but was generally greater than that observed for the SalA-associated producers.

The S. mitis isolates, in contrast, displayed broader susceptibility across the producer panel. Particularly strong inhibition was observed with S. salivarius M18, whose multiple lantibiotic repertoire includes salivaricins A2, 9, MPS and M. This pattern is consistent with the broader inhibitory repertoire of M18, although the relative contribution of individual peptides was not resolved in the present assays.

The S. pyogenes indicator strains displayed a third susceptibility configuration. Several isolates were susceptible to multiple heterologous producer systems while retaining resistance to particular homologous or closely associated inhibitory phenotypes. The resulting pattern differed from both the selective resistance observed among S. anginosus isolates and the broader susceptibility characteristic of the S. mitis group.

Collectively, these observations demonstrate that peptide susceptibility among oral streptococci is highly selective and cannot be explained solely by phylogenetic relatedness. The observed susceptibility phenotypes were reproducible across replicate deferred antagonism assays. No isolate exhibited complete resistance or complete susceptibility to the full panel of inhibitory systems, emphasising the highly selective nature of the observed peptide defence phenotypes.

3.2. Salivaricin A Resistance is Not Accompanied by Broad Peptide Resistance

Comparison of the susceptibility matrix revealed that resistance to salivaricin A did not predict resistance to other peptide systems.

The S. anginosus isolates showed markedly reduced susceptibility to inhibition by SalA-producing strains while remaining susceptible to salivaricin B, streptococcin A-FF22 and, in many cases, streptin. Likewise, susceptibility patterns differed substantially among the inhibitory systems examined despite similarities in producer species.

These findings demonstrate that resistance to one peptide system was not associated with broad resistance to the diverse inhibitory systems examined.

3.3. Dual Lantibiotic Production Broadens Inhibitory Activity

The inhibitory spectrum produced by S. salivarius M18 differed from that observed for strains producing individual lantibiotics.

Although both K12 and M18 possess salivaricin A-related systems, M18 consistently showed a broader inhibitory spectrum than K12 against the indicator panel. This observation is consistent with the additional lantibiotic repertoire of M18, although the individual contributions of salivaricins A2, 9, MPS and M to the observed inhibition were not determined.

The broad-spectrum BLIS-producing S. salivarius strain 36 inhibited both S. anginosus and S. mitis isolates, illustrating the considerable variation in inhibitory breadth among the producer strains examined.

3.4. Comparative Susceptibility Patterns Support a Selective Peptide Defence Framework

When considered collectively, the observed susceptibility patterns were inconsistent with a simple producer-versus-sensitive model of bacteriocin ecology.

Instead, the results revealed reproducible combinations of peptide susceptibility and resistance that differed between bacterial species and between individual inhibitory systems. Rather than reflecting broad resistance to antimicrobial peptides, these observations support the existence of selective peptide defence landscapes likely to influence microbial competition, colonisation and ecological persistence within the oral microbiome.

4. Discussion

The present findings suggest that the ecological consequences of bacteriocin production are determined not only by where inhibitory peptides are produced, but also by where corresponding peptide defence phenotypes are distributed within the competing community. Across the ten-producer, 46-indicator matrix, closely related streptococcal groups displayed distinct and reproducible combinations of susceptibility and resistance. The resulting patterns were not consistent with a simple continuum from bacteriocin-sensitive to bacteriocin-resistant organisms. Instead, susceptibility was dependent on the particular inhibitory system encountered, providing the empirical basis for the Selective Peptide Defence Landscapes (SPDL) concept.

This interpretation has an important historical precedent. Tompkins and Tagg [18] examined naturally occurring bacteriocinogenic S. salivarius populations and found that individuals harbouring very high proportions of bacteriocin-producing strains also carried significantly higher proportions of bacteriocin-resistant Gram-positive alpha-haemolytic cocci than did control individuals. They interpreted these observations as consistent with selection of a relatively resistant accompanying microbiota by bacteriocin production in situ. Their study also retained the susceptibility dimension of the earlier P/S fingerprinting scheme, using a defined indicator panel to characterise inhibitory spectra. The present study extends that early ecological observation by examining susceptibility to multiple peptide systems across three streptococcal species groups and by treating the resulting patterns as structured ecological phenotypes.

The significance of the present study therefore lies less in the discovery of an individual bacteriocin or immunity mechanism than in the integration of observations that have traditionally been considered separately. Bacteriocin studies commonly ask which organisms produce an inhibitory peptide, while studies of antimicrobial-peptide resistance ask how a target organism avoids inhibition. The SPDL framework brings these perspectives together by asking how the distribution of peptide-specific susceptibility and defence phenotypes shapes the pattern of potential interactions within a microbial community.

The S. anginosus phenotype provides a particularly clear example. These isolates showed markedly reduced susceptibility to inhibition by the SalA-associated producer strains represented in the present panel while remaining susceptible to several alternative systems, including salivaricin B-associated activity and streptococcin A-FF22. Such a phenotype is not readily described as general bacteriocin resistance. Rather, it implies selective compatibility between the target population and particular peptide-mediated challenges. Whether this selectivity reflects immunity proteins, receptor or cell-envelope differences, regulatory states, or combinations of these mechanisms cannot be resolved from the present phenotypic data alone.

An additional, testable explanation for some apparently susceptible phenotypes concerns the physiological state of the target cell. For some bacteriocins, susceptibility is strongly growth-phase and energy-state dependent: bacteriocin J46, for example, was bactericidal toward log-phase cells but ineffective against stationary or energy-depleted cells, and its activity required both a transmembrane pH gradient and membrane potential [28]. Nisin susceptibility has likewise been shown to vary markedly with growth phase [29]. Thus, slower growth or reduced energetic activity could diminish susceptibility to bacteriocins whose action depends on active cellular physiology or an electrochemical gradient, without constituting genetically encoded peptide immunity. This possibility is particularly relevant when interpreting deferred-antagonism phenotypes, which integrate peptide production, diffusion and the physiological state of the indicator during the assay. It should, however, be regarded as a mechanistic hypothesis rather than an explanation established by the present study, because lantibiotics do not share a single mechanism of action. Salivaricin B, for example, has been shown to interfere with cell-wall biosynthesis without detectable pore formation or membrane-potential dissipation [30].

The contrasting behaviour of the S. mitis group reinforces this interpretation. Its broader susceptibility across the producer panel indicates a different position within the comparative peptide-defence landscape. Likewise, S. pyogenes displayed a third reproducible configuration rather than simply occupying an intermediate position between S. anginosus and S. mitis. The important observation is therefore the configuration of responses across the producer panel, rather than the absolute susceptibility of any single isolate or species.

The ecological interpretation is also compatible with earlier observations that bacteriocin activity can influence bacterial interactions without necessarily producing immediate killing. Studies of group A streptococci have demonstrated bacteriocin-associated inhibition with bacteriostatic effects under some conditions [19]. Separately, components of salivaricin-associated systems have been shown to participate in intra- and interspecies signalling [14]. Such observations broaden the possible ecological consequences of peptide-mediated interactions: selective susceptibility may influence persistence, competitive exclusion or relative abundance without requiring wholesale elimination of the target population.

A further consideration is that bacteriocin-associated loci and peptide defence determinants need not be perfectly coupled to detectable peptide production. The occurrence of salivaricin-associated loci in organisms for which corresponding inhibitory activity is not readily demonstrated raises the possibility of ‘immunity without production’—that is, persistence of a defence determinant after loss, attenuation or regulatory silencing of the associated antimicrobial phenotype. The present study does not establish this mechanism genetically, but the selective susceptibility patterns provide a phenotypic context in which such uncoupling becomes ecologically relevant.

One possible explanation for retention of immunity determinants after loss or attenuation of lantibiotic production is that the defence function may be retained through cross-protection against other cationic antimicrobial peptides encountered in the host. This possibility is particularly relevant to the Streptococcus pyogenes sal locus, where conservation and altered sequence characteristics of the SalY immunity component have been proposed to reflect adaptation towards protection from host-derived antimicrobial peptides [15]. Thus, a locus initially associated with interbacterial lantibiotic competition could potentially acquire or retain value through protection against both microbial and host cationic peptides. This hypothesis provides one possible evolutionary explanation for the persistence of immunity-associated determinants in the apparent absence of a corresponding production phenotype, but requires direct functional testing.

Taken together, these observations provide a framework for connecting several otherwise separate features of streptococcal biology: diversity of salivaricin loci, mobility of bacteriocin-associated genetic elements, variation in peptide production, peptide-specific immunity, and the occurrence of bacteriocin-associated loci in organisms that do not necessarily display the corresponding production phenotype. Viewed through an SPDL framework, these are not isolated curiosities but components of a dynamic ecological system in which the outcome of a peptide-mediated interaction depends on both the producer’s inhibitory repertoire and the target community’s distribution of compatible defence phenotypes.

Historically, bacteriocin research has concentrated predominantly on the ecology of inhibition—who produces which peptide and which organisms are inhibited. The present findings suggest that an equally informative question is the ecology of defence: which organisms can withstand which peptide challenges, and how are those selective defence phenotypes distributed across a microbial community? The transition from P/S fingerprinting to SPDL retains the empirical value of susceptibility testing while changing its interpretive level from strain identification towards ecological organisation.

The present study has several limitations. The conclusions are based primarily on comparative phenotypic observations, supported by published information concerning bacteriocin-associated loci and peptide defence systems, rather than direct genomic or molecular analysis of the indicator isolates. In particular, the S. anginosus isolates examined phenotypically in the present study were not genome-sequenced, and their observed resistance patterns therefore cannot be directly attributed to specific salivaricin immunity determinants. Conversely, the comparative genomic observations cited here establish the occurrence of bacteriocin-associated determinants but do not by themselves demonstrate their expression or functional activity. The specific determinants responsible for individual susceptibility phenotypes therefore remain unresolved. In addition, the producer panel was intentionally selected as a historically informative reference set and should not be regarded as a comprehensive representation of the diversity of streptococcal bacteriocins. Finally, deferred antagonism assays provide comparative phenotypes under defined laboratory conditions and cannot by themselves establish the strength, persistence or ecological importance of individual interactions in vivo.

These limitations also define the next experimental opportunities. Genome-resolved analysis of producer and indicator strains could test whether particular susceptibility phenotypes correlate with cognate immunity genes, receptor determinants, cell-envelope features or regulatory loci. Experimental reconstruction of individual producer-defence combinations could then distinguish true peptide-specific immunity from broader physiological resistance. Longitudinal or community-level studies would further test whether the patterns observed in vitro correspond to differential persistence or abundance in the oral microbiome. Thus, SPDL is proposed not as a replacement for mechanistic bacteriocin research, but as a framework for connecting molecular defence determinants with their distribution and potential ecological consequences.

5. Conclusions

The present study demonstrates that peptide susceptibility among oral streptococci is highly selective and is consistent with the ecological distribution of peptide-specific defence systems rather than generalised resistance to antimicrobial peptides. Comparative phenotypic analysis of representative S. anginosus, S. mitis and S. pyogenes isolates identified reproducible combinations of susceptibility and resistance across a diverse producer panel.

These observations provide the basis for the concept of Selective Peptide Defence Landscapes (SPDLs), an ecological framework that complements traditional studies of bacteriocin production by considering the ecological distribution of peptide defence systems within microbial communities. We suggest that understanding which organisms possess specific peptide defence systems may be as important as identifying which organisms produce inhibitory peptides when seeking to explain microbial competition, coexistence and community organisation.

Although the SPDL framework has been developed from studies of oral streptococci, the underlying ecological principles are unlikely to be confined to this group of organisms. Similar relationships may occur wherever microorganisms compete using highly specific inhibitory molecules, including other bacteriocin families, bacterially derived antimicrobial peptides and potentially other classes of specialised microbial metabolites. Future investigations integrating comparative phenotypic analyses with genomics, transcriptomics and functional characterisation of peptide defence systems should determine how widely these ecological principles apply.

Historically, bacteriocin research has focused primarily on the ecology of inhibition rather than the ecology of defence. The present work suggests that equal attention should now be directed towards the ecology of peptide defence. We anticipate that mapping the ecological distribution of peptide defence systems will become an increasingly informative approach for understanding the organisation, resilience and stability of microbial communities and will provide a valuable framework for investigating the ecological organisation of diverse microbial communities.