All bacterial taxa require purine nucleotides to accomplish essential metabolic processes such as the synthesis of DNA, RNA, and protein. Several pathways for de novo purine synthesis and salvage converge to satisfy this requirement in bacteria and thereby, support survival in niche environments that are limited in purine bioavailability, as reported for Borrelia [1], Erwinia [2], Lactococcus [3], and Helicobacter [4]. Phosphoribosyl pyrophosphate is utilized to generate purines, and the biosynthetic pathway bifurcates at the point of inosine monophosphate (IMP) to produce adenine monophosphate or guanine monophosphate (GMP) as precursors for deoxyribonucleotides. The latter requires guaA-encoded GMP synthetase, which provides Guanosine-5′-triphosphate for essential cellular processes.

The resident microbiota of the human microbiome inhabits diverse niches that vary in nutrient availability [5]. One such niche is the “urogenital tract”, a term which refers to an organ system that encompasses the urinary tract as well as the anatomical sites/organs of reproduction, which may affect the microbial load in urine, as reviewed elsewhere in the context of the human microbiome [6]. Different tissues of the live human host vary in relative concentrations of purines which are influenced by external factors such as diet [7]. In human urine (HU), the purine guanine is a precursor of uric acid and potential biomarker of disease [8]. Bacteria of the commensal microbiome persist in the urogenital tract through metabolic flexibility [6]; however, the extent to which purines such as guanine are utilized remains largely unclear. Among the estimated 50 genera of the human urogenital microbiota [6], several including Streptococcus persist as commensals but can also act as opportunistic pathogens. We explored the role of guaA in Streptococcus survival and colonization in the urinary tract as a part of the “urogenital tract” niche.

We analyzed the growth of wild-type (WT) S. agalactiae 834 compared to a guaA-deficient mutant (Supplementary Table S1) in HU pooled from several individuals. We also compared the growth of these strains in synthetic human urine, which revealed a striking attenuation of guaA- S. agalactiae in both conditions (Fig. 1a, b). Parallel growth assays using Todd Hewitt Broth showed no general growth defect of the guaA mutant (Fig. 1c). Owing to its contribution to GMP synthesis, we were able to chemically complement the guaA− mutation by supplying exogenous guanine, confirming that guaA is essential to support bacterial survival and growth in environments where guanine levels are limited (Fig. 1d). Consistent with our findings, guaA supports survival of E. coli in urine, which has been associated with a shift to commensalism for this gram-negative organism [9,10,11]. We note the genomic arrangement of guaA and guaB (encoding IMP dehydrogenase that catalyzes the step of GMP synthesis that precedes that of guaA) in Streptococcus differs markedly from E. coli (Fig. S1) and Staphylococcus in which control systems of these genes are well-defined [12, 13]. Transcriptional analyses showed modest repression of Streptococcus guaA and guaB in media conditions of exogenous guanine (Fig. 1e), despite the distinct genomic arrangement; consistent with findings in E. coli in which guaA is not induced as a milieu-specific response following culture in urine [11]. Together, these observations hint at convergent systems of guanine-dependent control of guaA regardless of spatial gene arrangement disparities among members of the commensal urogenital tract microbiota. Interestingly, some strains of uropathogenic Escherichia coli upregulate guaA only during active UTI but not during growth in pooled HU ex vivo [11]; these findings suggest that there may be value in examining the transcriptional activity of the gua genes in the context of active S. agalactiae UTI in the host environment.

Fig. 1: guaA supports S. agalactiae growth in urine and colonization of the mouse bladder.
figure 1

a Growth of WT and guaA-deficient S. agalactiae were compared in human urine (HU), b synthetic human urine (SHU), and c THB with guaA supplied in trans to complement the mutation (ΔguaA::guaA). d WT and ΔguaA S. agalactiae grown in the absence (Ctrl) or presence of 0.5 mM guanine-HCl (+Gua), compared in the same growth conditions at the 24 h timepoint. e Expression (fold change) of guaA and guaB was quantified from WT S. agalactiae in SHU ± 0.5 mM guanine. Data shown are pooled results of three to five independent assays, each using different batches of HU and SHU; bars show means ± SEM, compared using area under the curve and Student’s t tests. The number of bacteria recovered from the (f) urine, and (g) bladders of female C57BL/6 mice at 24 h postinoculation. Data shown are from three independent experiments, each with ≥7 mice per group. Data were compared using Mann–Whitney U test (***p < 0.001). The bars represent medians, displayed with interquartile ranges.

Full size image

Analysis of the role of guaA in bacterial colonization in vivo revealed a marked attenuation in survival of guaA− Streptococcus compared to the WT in urine of mice that were experimentally infected by transurethral delivery of bacteria to the bladder (Fig. 1f). A similar attenuation was detected in the bladders 24 h after infection (Fig. 1g), illuminating a major role for guaA in mediating Streptococcus survival and colonization in the urinary tract. Bacterial adaptation to this niche may represent an evolutionary strategy toward commensalism [14]; the finding of guaA’s contribution to S. agalactiae survival and colonization in the urinary tract may offer new explanation of how other gram-positive bacteria survive and persist in the urinary tract, given a high degree of conservation in the sequence of guaA among clinical isolates (Supplementary Table S2). Additional mechanisms that underpin bacterial survival in urine include acquisition of iron, malic acid metabolism, and tolerance to D-serine [15]. Notably, Streptococcus guaA also contributes to bacterial survival in other body niches, including blood [16]; this is relevant given that S. agalactiae can spread haematogenously to cause pyelonephritis [17] and can also cause urosepsis [18]. A recent clinical study reported that urine was the second most common source for S. agalactiae bacteraemia [19]. In this context, it is noteworthy that mammalian plasma contains very low concentrations of purines and purine nucleosides, which can vary during states of physiological stress and disease [20].

In Firmicutes, including S. agalactiae, guaA is a hotspot for mobile insertion elements [21, 22], offering potential for genomic rearrangement at this locus. We analyzed the genomic locations of guaA in streptococci and other commensal bacteria of the human urogenital tract [23] to reveal substantial cross-phyla diversity among Firmicutes, Bacteroidetes, Actinobacteria, and Proteobacteria, and an organizational structure that is distinct from the guaAB operon characteristic of many enteric bacteria and staphylococci (Fig. 2) [12, 13]. Despite relatively conserved amino acid sequences among the Firmicutes (62–89% homology) analyzed in this study, guaA genetic organization as well as sequence diversity is more apparent comparing between other phyla (Figs. 2 and  S1). Additional elements of gene regulation and gene transfer are also apparent, including the presence of guanine riboswitch sequences that effect transcriptional activation and insertion sequences that reflect the hotspot for integration.

Fig. 2: Cross-phyla genetic organization and sequence diversity of guaA among commensal bacterial genera of the human urogenital tract microbiota.
figure 2

a Genomic arrangement of guaA and guaB homologs in different bacterial species that colonize the urogenital tract niche [23], using reference genomes of representative members of selected genera for which sequence data are available. Where guaA (red) and guaB (green) are separated, the number of bases between the genes is indicated by dashed lines. In S. agalactiae 834, IS1381 insertion element is located at the 3′ end of guaA, which is often co-located divergently from a GntR-type regulator in streptococci. Percentage identity to S. agalactiae guaA is shown in white text. †bacterial species for which some members are uropathogenic and cause UTI [31]. b In staphylococci, guaA and guaB are adjacent to purine salvage genes xpt and pbuX, and are controlled by an upstream guanine-sensing riboswitch [12].

Full size image

The discovery that guaA is critical for survival of and colonization by bacteria in urine aligns closely to studies of nutritional immunity that show metabolic gene products can be crucial to microbial fitness in a host. The ability of microbes to persist and proliferate in host niches depends on essential nutrient provision, which can stem from the host or be synthesized de novo. A widely studied example is iron acquisition and the counteraction of iron limitation by the host to restrict microbial growth, which might be harnessed to medicinally target bacteria [24]. Only certain nutrients are limiting in urine [10] and guanine is one of these for different types of bacteria. More broadly, previous studies that have identified selective inhibitors of guanine-related metabolic pathways have used such compounds to treat infection [25]. Interestingly, a prior study demonstrated that infection of the mouse bladder with S. agalactiae 834 engages the innate immune response to generate an inflammatory, chemotactic, and regulatory cytokine milieu [26]. Thus, at least in mice, the host significantly activates immune defenses in response to this bacterium, which was isolated from the urine of an asymptomatic adult [27, 28]. Some patterns of cytokine production in mice with acute UTI parallel immune responses that occur in humans with UTI, as reviewed elsewhere [29]; however, reconciling the asymptomatic nature of S. agalactiae 834 as a clinical isolate with stimulation of innate immune responses in mice will require further study.

Some bacteria, such as E. coli can behave as a commensal organisms in some host niches such as the gut, but not in other host niches such as the urinary tract (where the bacteria cause inflammation and persists in the absence of antibiotic treatment) and the vagina [30]. Further understanding of S. agalactiae as a commensal of the human urogenital tract and an opportunistic pathogen will require analysis of the bacteria’s requirements for survival in different host niches. It will be of interest to study the genetic organization of guaA and other genes (for guanine salvage), the regulatory mechanisms of their activities in relevant models of infection, and characterize the role of metabolic gene products, including guanine as potential targets for microbial control. A proposed pathway for guanine metabolism in S. agalactiae is illustrated in Supplementary Fig. S2. Work on other commensal microbes should examine guanine utilization in host niches, given the diverse environmental and host tissue habitats that these microbes encounter.