Staphylococcus Pasteuri (BCVME2) Resident In Buffalo Cervical Vaginal Mucus (CVM) - A Potential Source of Estrus-Specic Sex Pheromone(s)

There are microbes resident in the reproductive tract, some of which could be pathogenic while a few others would, perhaps, play important roles in protecting the reproductive tract from infections. Volatile compounds are known to play role as sex pheromones that attract the males for coitus during estrus or heat. It is likely that these compounds themselves are secondary metabolites of the bacterial ora resident in the vagina. In order to substantiate this hypothesis, bacteria were isolated from cervico-vaginal mucus (CVM) of buffalo during various phases of the oestrous cycle and identied, based on morphological, biochemical and molecular characteristics, as Bacillus during preestrus as well as diestrus, and Staphylococcus during all phases of the oestrous cycle. But, the populations of Staphylococcus differed between different phases of the oestrous cycle, the predominant forms being S. warneri (BCVMPE1_1) during preestrus, S. pastueri (BCVME2) during oestrus and S. epidermis (BCVMDE3) during diestrus. Mice, when used as sensors, eciently differentiated the oestrus-specic S. pastueri (BCVME2) from the others. HS-GC-MS analysis showed that S. pastueri (BCVME2) produces key volatile compounds viz., acetic, propanoic, isobutyric, butyric, isovaleric and valeric acids. In addition, it is evidenced that S. pasteuri (BCVME2) volatiles inuence the sexual behaviours such as ehmen and mounting of the bull. Thus, the paper reports that S. pasteuri (BCVME2) is the potential source of vaginal pheromone(s) during oestrus in buffalo.


Introduction
The cervico-vaginal mucus (CVM) microbial community is associated with reproductive health and fertility in mammals. For example, the cervical secretions are known to play crucial roles in the reproductive success of cows (Adnane et al. 2018). CVM of the healthy cow contains aerobic as well as anaerobic microbiota, especially those belonging to Enterobacteriaceae family, and also Staphylococcus sp., Streptococcus sp., Enterococcus sp. and Lactobacillus sp. (Otero et al. 1999; Wang et al. 2013). The CVM microbiota are in uenced by hormonal, immunological and nutritional factors, and even the antibiotics and hormones that are used as drugs for treatment of the host (Otero et al. 2006). The Research so far has revealed that the culture-independent method using next-generation sequencing (NGS) is appropriate to nd the diversity of CVM microbiome with particular attention to the buffalo oestrous cycle (Srinivasan et al. 2019). Proteobacteria, Actinobacteria, Firmicutes, Bacteroidetes and Tenericutes are the most common CVM of buffalo. Among these ve families, Firmicutes has been found to be highly abundant during oestrus compared to the other phases (Srinivasan et al. 2019 Volatile compounds such as acetic acid and propanoic acid have been identi ed in cow vaginal mucus (Sankar and Archunan 2011) and oleic acid from buffalo vaginal secretion (Karthikeyan and Archunan 2013) during oestrus, which act as female sex pheromones. Interestingly, acetic acid has been found in buffalo urine, with a substantial increase in concentration during estrus relative to other phases of the oestrous cycle (Muniasamy 2016). The bacterial populations present in the CVM are reported to potentially affect the sexual attraction of ewes (Ungerfeld and Silva 2005). The report suggests that fatty acids may act as pheromones as well as estrus indicators in buffaloes and that these microbes might be responsible for the speci c odour released at oestrus.
On the basis of substantial evidences, buffalo CVM has been shown to produce estrus-speci c volatiles that affect the reproductive behaviour of male buffalo during mating. However, the source of the buffalo CVM pheromone compounds is not well established. Evidence for the impact of CVM bacteria and their role in production of volatiles in the context of male attraction towards female buffalo is inadequate. The present study was therefore intended to i) demonstrate the bacterial community in buffalo CVM through a culture-based method during different phases of the oestrous cycle, and (ii) screen the oestrus-speci c bacteria using mice as model sensor system, iii) nd if bacterial secretory substance contains volatiles, and iv) evaluate bull behaviour in response to estrus-speci c bacterial secretory substance of CVM in order to con rm that the very bacterial volatiles are the pheromones that elicit speci c premating behaviours in the males.

Animal and sample collection
The healthy female buffaloes (n = 12) and phases of oestrous cycle were monitored as per the protocol of Rajanarayanan

Isolation and characterization of bacteria
For the isolation of bacteria, the CVM was collected from buffaloes representing the three phases of the oestrous cycle. CVM, 0.5 mL, was homogenized in 4.5 mL of sterile distilled water. An aliquot of 0.1 mL of the homogenate was serially diluted and plated in Nutrient-MacConkey-De Man, Rogosa and Sharpe-and Mannitol salt-agar by standard plate count method (Cassoli et al. 2016). Plates were incubated at 37°C for 24-48 hr. Following the incubation, the bacterial colonies were counted and expressed by log values (Log CFU/mL i.e. Log10 (CFU / (dilution factor x aliquot). Morphologically distinct colonies alone were picked and further processed. Primary characterization was carried out based on colony morphology which included colour, size and shape. Further, the isolates were subjected to Gram staining and biochemical tests for catalase, oxidase, sugar fermentation and haemolysis.
Extraction of genomic DNA and PCR ampli cation of 16S rRNA gene Genomic DNA of bacterial isolates was extracted based on the method of Vingataramin and Frost (2015). The quality of gDNA was evaluated on 1.0% agarose gel. Fragment of 16S rRNA gene was ampli ed by PCR with the primers 27F AGAGTTTGATCMTGGCTCAG and 1492R CGGTTACCTTGTTACGACTT. PCR reactions were executed in a total volume of 30 µL containing 15 µL 2x PCR premix, 2 µL template DNA, 0.5 µL each of the primer and 12 µL sterile distilled water. The cycling parameters used were as follows: initial denaturation at 94°C for 3 min, followed by 35 cycles of denaturation at 94°C for 30 sec, annealing at 50°C for 30 sec and extension at 72°C for 90 sec, and then a nal elongation at 72°C for 7 min (Coombs and Franco, 2003). Euro ns Genomics India Pvt. Ltd (Bangalore, India) conducted the 16S rRNA gene sequencing.

Sequencing and phylogenetic analysis of 16S rRNA gene
The amplicons were puri ed using pre-sequencing kit and sequenced on ABI 3730xl automated sequencer (Perkin Elmer Applied Biosystems, Foster City, CA, USA) and ABI Prism BigDye Terminator Cycle Sequencing Ready Reaction Kit when AmpliTaq® DNA polymerase was used as per protocols recommended by the manufacturer.
The dataset of 16S rRNA gene sequences was prepared for BLAST search. All individual representative sequences were performed at NCBI (Altschul et al. 1990). Sequences were obtained from BLAST searches based on E-value where identity scores differed depending on the size of the gene family. Nucleotide compositional parameters were predicted from BioEdit 7.0.4 (Hall 2011). Besides the BLAST results, 16S rRNA gene sequences of the most closely related taxa were retrieved from the GenBank database and aligned using the ClustalW tool implemented in MEGA X (Kumar et al. 2018). MEGA X was used to reconstruct the phylogenetic tree adopting neighbour-joining (NJ) method using bootstrap values based on 1000 replications (Kim et

Bull behavioural observation towards CVM bacterial volatile samples
Non-estrus female buffaloes (n = 6) and bulls (n = 6) were selected for the behaviour studies. The supernatant of bacterial culture, estrus CVM, non-estrus CVM and LB broth were subjected to bull behaviour analysis. Each sample (5 mL) was individually soaked in cotton and rubbed on the vulvar region of non-estrus buffalo (i.e. dummy cow). The bulls were allowed to sniff the test animals, and behaviours were observed for 30 min (Karthikeyan and Archunan 2013). The ehmen and mounting behaviours exhibited by the bull in response to the test samples were recorded.

Statistical analysis
Analysis of variance (ANOVA) was carried out for behaviour data analysis by using SPSS (V.2.3) package, and images were plotted in GraphPad Prism (V.8.0.2).

Isolation of bacteria from buffalo CVM
The bacteria isolated from CVM during preestrus, estrus and diestrus were present as 38 CFU/mL, 51 CFU/mL and 36 CFU/mL, respectively as determined adopting serial dilution method. Thus, cumulatively as many as 125 bacterial colonies were observed covering all three phases. Amongst them, 6 isolates were chosen, based on morphological characters such as smoothness and colour (white or creamy yellow), for higher level analyses. Biochemical characterization such as Gram staining, sugar fermentation tests (glucose, fructose, maltose and sucrose) and haemolytic assay showed (Table S1). The bacteria found in oestrus CVM are Gram positive, non-haemolytic, and fermentable sugars, including glucose, fructose, maltose, and sucrose, with acid production. Construction of the phylogenetic tree Pairwise 16S rRNA gene sequence similarities were determined using BLASTn analysis. Based on the identity (98% and above from the query coverage) the species was selected. The variations in nucleotide composition were identi ed for CVM bacterial isolates (in percentage as well as numbers). G + C base pair was found to be high, above 52% (Table S2). NCBI-BLAST was practiced to identify isolates having > 98% similar sequence at inter-and intra-speci c levels. The result showed that the tree could be grouped into two major clusters. The rst cluster consists of various species of Staphylococcus genus; the second cluster consists of Bacillus species as represented in phylogenetic tree analysis (Fig. 1A). The molecular taxonomic characterization showing the differential expression in respect to three phases of oestrous cycle is presented in Fig. 1B. Discrimination of the bacterial volatiles from CVM adopting mice behavioural assay There were signi cant differences among the bacterial volatiles across different phases of the oestrous cycle. The behavioural assessment of the mice revealed greater attraction towards estrus-speci c bacteria than the non-estrus bacterial sample (Fig. 2). The estrus-speci c bacterial volatiles was found to be most attractive as mice visited them frequently and spent more time than the non-estrus bacterial volatiles (p < 0.001). The behavioural study further expounded that the estrus S. pasteuri (BCVME2) most attracted the mice.

Identi cation of volatiles produced by bacteria as revealed in headspace GC-MS
The GC-MS analysis of cell free supernatant of estrus-speci c bacteria isolated from CVM revealed the nature of volatiles (Fig. 3). Six volatiles viz., acetic acid, propanoic acid, isobutyric acid, butyric acid, isovaleric acid and valeric acid are produced by S. pasteuri (Table 1). Acetic acid appeared to be at a higher peak intensity among bacterial volatiles, followed by isovaleric acid, valeric acid, isobutyric acid, butyric acid, and propanoic acid. Bull behaviour conditioned by the volatiles of S. pasteuri The bulls were attracted to the maximum towards intact estrus CVM followed by supernatant of S. pasteuri, LB broth, and non-estrus CVM. The frequency of ehmen response exhibited by bull was highest to the intact estrus CVM (5.83 ± 1.16) as well as S. pasteuri secretory substance (4.66 ± 1.21). By contrast, bull showed signi cantly lesser attraction to non-estrus CVM (0.67 ± 1.03) and no response at all to LB broth. Likewise, the mounting behaviour exhibited by bull was remarkably higher towards intact estrus CVM (4.83 ± 1.94) and S. pasteuri secretory substance (4.5 ± 1.64). However, bull showed less response towards non-estrus CVM (0. 5 ± 1.22) and no response at all towards LB broth. However, the ehmen and mounting behaviours observed during exposure to estrus CVM sample and S. pasteuri secretory substance are almost similar in response. Details of ehmen and mounting behaviours of bull are given in Fig. 4.

Discussion
Our previous study showed differences in the CVM bacterial community with respect to various phases of the oestrous cycle (Srinivasan et al. 2019), in which the Firmicutes phylum was more abundant during the oestrus phase. Probably, expression of certain speci c bacterial genes is higher during estrus phase than the other phases. The present investigation revealed expression of speci c gene(s) in S. pasteuri during the estrus phase. The available information indicates that the CVM bacterial strains get in uenced during estrus in buffalo. It has also been reported that the vaginal bacterial population is affected by the cyclical variation during the oestrous cycle (Otero et al. 1999;Otero et al. 2000).
In the present study, S. pasteuri were incidentally observed in buffalo CVM during estrus. Similar studies have been reported in various animals. Association of vaginal Simonsiella spp in reference to estrus phase has been documented in lions and leopards (Callealta et al. 2018). It is interesting to note that Streptococcus and Staphylococcus are the most dominant bacteria in healthy cows during the reproductive cycle (Amin et al. 1996). Staphylococcus (70%) and Escherichia (15%) were observed more frequently during natural and induced estrus of ewes (Orihuela et al. 2019) and all the isolates had a high sensitivity to cipro oxacin antimicrobials that evaluated vaginitis regulation (Mohammed et al. 2017). The present nding supports the previous report of the presence of Staphylococcus pasteuri, which comes under the phylum Firmicutes, and is abundant during estrus (Srinivasan et al. 2019).
In addition, bacteria such as Bacillus spp., Staphylococcus spp., and Streptococcus spp. The mice system was used to screen the volatile compound samples. Among 7 samples tested with mice, S. pasteuri secretory substance exhibited to be more attractive than the other cultures. The results are consistent with an earlier report that mice are capable of discriminating the estrus and non-estrus samples. A recent study provides supportive evidence that mice are highly sensitive to volatiles produced by the microbes (Peixoto et al. 2018). Truly, the mice study helped to pick out the bacteria potentially responsible for volatiles production.
GC-MS analysis indicated six volatiles, acetic acid, propanoic acid, isobutyric acid, butyric acid, isovaleric acid and valeric acid, as produced by S. pasteuri. It is known that short-chain fatty acids such as acetic-, propanoic-and butyric acid, are relatively abundant due to bacterial fermentation in colonic lumen which regulate entero-pathogenic colonization by host immunomodulation in human (Smith et al. 2013;Wrigley 2004). A recent study shows that acetic, valeric, caproic and myristoleic acids are at signi cantly higher levels in the milk of estrus cow (Zebari et  Since estrus-speci c S. pasteuri produces volatiles, we focused on testing the role of S. pasteuri secretory substance in bull behaviours which con rmed the estrus-speci city and its use as a reliable estrus indicator. Interestingly, bull exhibited reproductive behaviours such as ehmen and mounting, which are important in assessing sexual desire of bull as response to S. pasteuri secretory substance. The present study also corroborates previous ndings that the estrus-speci c faeces pheromone has a signi cant effect on ehmen and mounting behaviour in buffalo ). It has been reported that bulls exhibit this reproductive behavioural patterns in response to the presence of a mixture of volatiles in the CVM of estrus cow (Karthikeyan and Archunan 2013; Rajanarayanan and Archunan 2004; Sankar and Archunan 2004). Vaginal pheromones have been reported to be acetic acid, propanoic acid, and 1iodoundane, in cows (Sankar and Archunan 2004) and oleic acid as an estrus indicator in buffalo (Karthikeyan and Archunan 2013) to stimulate ehmen and mounting behaviour response in bulls. The present ndings suggest that the volatile compounds of bacterial origin may act as attractants of the bull and induce the olfactory system to elicit ehmen and mounting behaviours in order to maintain the health and management during reproduction. Since S. pasteuri is capable of secreting the volatiles viz., acetic, propanoic, isobutyric, butyric, isovaleric, and valeric acids, they can be considered as major bacterial volatiles that induce the buffalo bull's sexual behaviour. For instance, acetic acid, 2-butanone and oleic acid have been shown to improve Zebu bull sexual behaviour and total sperm production (Mondal et al., 2019). Thus, this study reports the presence in the buffalo vaginal uid of volatile compounds secreted by CVM bacteria which facilitate the expression of male reproductive behaviours, which in turn suggests that these volatiles are putative pheromone compounds produced during oestrus in buffalo.
Further studies are needed to nd the mechanisms associated with the speci c effects of S. pasteuri volatiles. Characterization of protein pro le of S. pasteuri from buffalo CVM during estrus is required to understand the host-bacterial interaction and their role in chemo-signal communication that would enlighten the reproduction management in buffaloes. Chromatogram of bacterial secretory substance. A. S. pasteuri and B. LB broth.