The ultra-small, obligate parasitic epibiont, TM7x, the first and only current member of the long-elusive Saccharibacteria (formerly the TM7 phylum) phylum to be cultivated, was isolated in co-culture with its bacterial host, Actinomyces odontolyticus subspecies actinosynbacter, XH001. Initial phenotypic characterization of the TM7x-associated XH001 co-culture revealed enhanced biofilm formation in the presence of TM7x compared to XH001 as monoculture. Genomic analysis and previously published transcriptomic profiling of XH001 also revealed the presence of a putative AI-2 quorum sensing (QS) operon, which was highly upregulated upon association of TM7x with XH001. This analysis revealed that the most highly induced gene in XH001 was an lsrB ortholog, which encodes a putative periplasmic binding protein for the auto inducer (AI)-2 QS signaling molecule. Further genomic analyses suggested the lsrB operon in XH001 is a putative hybrid AI-2/ribose transport operon as well as the existence of a luxS ortholog, which encodes the AI-2 synthase. In this study, the potential role of AI-2 QS in the epibiotic-parasitic relationship between XH001 and TM7x in the context of biofilm formation was investigated. A genetic system for XH001 was developed to generate lsrB and luxS gene deletion mutants in XH001. Phenotypic characterization demonstrated that deletion mutations in either lsrB or luxS did not affect XH001's growth dynamic, mono-species biofilm formation capability, nor its ability to associate with TM7x. TM7x association with XH001 induced lsrB gene expression in a luxS-dependent manner. Intriguingly, unlike wild type XH001, which displayed significantly increased biofilm formation upon establishing the epibiotic-parasitic relationship with TM7x, XH001ΔlsrB, and XH001ΔluxS mutants failed to achieve enhanced biofilm formation when associated with TM7x. In conclusion, we demonstrated a significant role for AI-2 QS in modulating dual-species biofilm formation when XH001 and TM7x establish their epibiotic-parasitic relationship.

Despite many examples of obligate epibiotic symbiosis (one organism living on the surface of another) in nature, such an interaction has rarely been observed between two bacteria. Here, we further characterize a newly reported interaction between a human oral obligate parasitic bacterium TM7x (cultivated member of Candidatus Saccharimonas formerly Candidate Phylum TM7), and its basibiont Actinomyces odontolyticus species (XH001), providing a model system to study epiparasitic symbiosis in the domain Bacteria. Detailed microscopic studies indicate that both partners display extensive morphological changes during symbiotic growth. XH001 cells manifested as short rods in monoculture, but displayed elongated and hyphal morphology when physically associated with TM7x. Interestingly, these dramatic morphological changes in XH001 were also induced in oxygen-depleted conditions, even in the absence of TM7x. Targeted quantitative real-time PCR (qRT-PCR) analyses revealed that both the physical association with TM7x as well as oxygen depletion triggered up-regulation of key stress response genes in XH001, and in combination, these conditions act in an additive manner. TM7x and XH001 co-exist with relatively uniform cell morphologies under nutrient-replete conditions. However, upon nutrient depletion, TM7x-associated XH001 displayed a variety of cell morphologies, including swollen cell body, clubbed-ends, and even cell lysis, and a large portion of TM7x cells transformed from ultrasmall cocci into elongated cells. Our study demonstrates a highly dynamic interaction between epibiont TM7x and its basibiont XH001 in response to physical association or environmental cues such as oxygen level and nutritional status, as reflected by their morphological and physiological changes during symbiotic growth.

Bacteriophages (phages) are an integral part of the human oral microbiome. Their roles in modulating bacterial physiology and shaping microbial communities have been discussed but remain understudied due to limited isolation and characterization of oral phage. Here, we report the isolation of LC001, a lytic phage targeting human oral Schaalia odontolytica (formerly known as Actinomyces odontolyticus) strain XH001. We showed that LC001 attached to and infected surface-grown, but not planktonic, XH001 cells, and it displayed remarkable host specificity at the strain level. Whole-genome sequencing of spontaneous LC001-resistant, surface-grown XH001 mutants revealed that the majority of the mutants carry nonsense or frameshift mutations in XH001 gene APY09_05145 (renamed ltg-1), which encodes a putative lytic transglycosylase (LT). The mutants are defective in LC001 binding, as revealed by direct visualization of the significantly reduced attachment of phage particles to the XH001 spontaneous mutants compared that to the wild type. Meanwhile, targeted deletion of ltg-1 produced a mutant that is defective in LC001 binding and resistant to LC001 infection even as surface-grown cells, while complementation of ltg-1 in the mutant background restored the LC001-sensitive phenotype. Intriguingly, similar expression levels of ltg-1 were observed in surface-grown and planktonic XH001, which displayed LC001-binding and nonbinding phenotypes, respectively. Furthermore, the overexpression of ltg-1 failed to confer an LC001-binding and -sensitive phenotype to planktonic XH001. Thus, our data suggested that rather than directly serving as a phage receptor, ltg-1-encoded LT may increase the accessibility of phage receptor, possibly via its enzymatic activity, by cleaving the peptidoglycan structure for better receptor exposure during peptidoglycan remodeling, a function that can be exploited by LC001 to facilitate infection. IMPORTANCE The evidence for the presence of a diverse and abundant phage population in the host-associated oral microbiome came largely from metagenomic analysis or the observation of virus-like particles within saliva/plaque samples, while the isolation of oral phage and investigation of their interaction with bacterial hosts are limited. Here, we report the isolation of LC001, the first lytic phage targeting oral Schaalia odontolytica. Our study suggested that LC001 may exploit the host bacterium-encoded lytic transglycosylase function to gain access to the receptor, thus facilitating its infection.