It has been reported that the female cervix maintains communities of microbial species, which have a symbiotic relationship with the host (1). It has been demonstrated that the female cervix is colonized by diverse microbiota, which serve a crucial role in cervicovaginal health (1). However, it has been indicated that the type of organisms present is dependent on the prevailing environmental conditions and host factors (1,2). Over the last decade, it has been indicated that the development and introduction of molecular-based technology has provided novel information regarding the composition of the vaginal-cervical flora, as well as the abnormal colonization of the genital tract by pathogens (2). The aforementioned findings may help to elucidate the microbiome of the genital tract in healthy women and in human papilloma virus (HPV)-dependent carcinogenesis.

The majority of previous studies have indicated that the cervical/vaginal microbial flora has a prevalence of Lactobacillus species, which produce lactic acids that maintain an acidic environment and may inhibit pathogenic growth (2–8). Specifically, lactic acid and other related acidic compounds have been reported to inhibit bacterial growth associated with bacterial vaginosis (BV), as well as viral infections (3). In addition, lactic acid has been recognized as a component of the immune defence system, as it has been demonstrated to potentiate the production of protective proinflammatory cytokines by vaginal epithelial cells, to promote the activation of T helper 17 lymphocytes, to stimulate dendritic cell maturation and induce interferon production (1). Over 120 Lactobacillus species have been identified and over 20 species have been identified in the vagina (2).

The dominant microorganisms in the vagina of a healthy woman during puberty have been reported to be from the Lactobacillus genus, including L. acidophilus, L. fermentum, L. plantarum, L. brevis, L. jenseni, L. casei, L. catenaforme, L. delbrueckii and L. salivarius (1,2). It is considered that in the majority of cases, vaginal inflammation is not caused by novel microorganisms introduced from the outside, but instead by a disturbance to the proportion and number of microorganisms already existing in the vagina (4). Each bacterium has been reported as a potential etiological factor of inflammation, including Lactobacillus (4). It has been reported that anaerobic bacteria have a significantly increased pathogenic potential compared with aerobic bacteria (5,6). Factors, which have been reported to potentially disturb vaginal biocenosis include the following hormonal changes: Pregnancy, puberty, menopause and hormonal contraception, particularly using low doses of estrogen; vaginal sterilization following chemo- or antibiotic therapy; surgical conditions such as vaginoplasty, erosions, poor genitalia hygiene, including the vulva and the vagina, and lack of a regimented sex life, i.e. frequent, unprotected, lack of a regular sex life (4–6). Large variation in the number and composition of bacteria has been reported among women, and among time intervals for one woman (5,6). Therefore, attempting to evaluate the vaginal microbiome is extremely difficult.

A healthy cervicovaginal microenvironment has been reported to be characterised by high levels of different species of Lactobacillus, including the predominant L. crispatus, L. iners, L. jensenii and L. gasseri. Other species may occur occasionally (2,6). In rare cases, it has been reported that the cervix can be colonized by the same two or four Lactobacillus species (4). The aforementioned cases have been demonstrated to be dependent on a number of genetic and environmental factors, including nationality, diet and age (2,6). A number of previous studies have indicated that, in a significant proportion of healthy women, the Lactobacillus in the vagina may be replaced by other lactic acid-producing bacteria, including Atopobium vaginae, Megaspharea and Leptotrichia species (3). It has been indicated that an abnormal vaginal microenvironment may be caused by sexually transmitted infection (3). It has been reported that trichomoniasis may be caused by colonization with a microorganism not commonly identified within vaginal colonies, including Streptococcus pneumoniae, Haemophilus influenzae and Listeria monocytogenes (6,7). In addition, it has been reported that an abnormal microenvironment may be caused by an invasion of an alternative organism, which is a component of the normal vaginal flora, including Escherichia coli (3). Bacterial vaginosis has been reported as a disorder characterised by a decrease in the quality or quantity of Lactobacillus and the growth of Mycoplasma hominis, Gardnerella vaginalis, Mobiluncus species, Neisseria gonorrhea, Trichomonas vaginalis, Chlamydia trachomatis and Prevotella species (4,5). A previous meta-analysis reported a positive association between cervical HPV infection and BV (6,7). The reverse phenomenon of HPV as a risk factor for BV invasion has also been described (8).

It has been demonstrated that HPV is considered a principal factor responsible for the development of cervical cancer (8). However, it has been suggested that HPV infection alone does not cause cervical carcinogenesis and that other factors, such as prolonged oral contraceptive or smoking, may be involved in the disease progression (8). To determine whether the vaginal microflora is affected by one of these factors, the present study investigated the association between cervical and vaginal infections, and pre-cancerous lesions of the cervix.

It has been indicated that the introduction of Next-Generation Sequencing (NGS) into research allowed for the identification of specific ecological niches for microorganisms within living organisms. This was possible due to the amplification and parallel sequencing of gene fragments, which were highly conserved among microorganisms, the most common being the 16S rRNA subunit, as well as RAD51 recombinase and inhibin subunit a (12). Highly conserved regions of the 16S rRNA gene (V1-V9) have been reported to allow for phylogenetic and taxonomic characterisation of the analysed microbial communities (12). In the present study, the V4 hypervariable regions of 16S rRNA were used, which allowed for the identification of 3 CSTs of HSIL HPV(+), LSIL HPV(+) and the healthy group HPV(−).

In addition, >70 classes of bacteria were identified and the analysis of their frequencies demonstrated that cervical swabs of flora obtained from the volunteers and women diagnosed with LSIL HPV(+) were predominantly composed of the Bacilli class. The presence of Gammaproteobacteria and Actinobacterium classes in patients with HSIL HPV(+) were also detected. Further analysis revealed no cluster formation within the Gammaproteobacteria class and, therefore, this class was excluded from the present study.

Further analysis focussed only on the Actinobacterium and Bacilli classes of bacterial species. Regardless of the type of diagnosis, the bacterial species within the analysed classes only formed two clusters. It was observed that swabs from healthy women were characterised by an increased level of Lactobacillus crispatus, Lactobacillus iners and Lactobacillus taiwanensis. However, Gardnerella vaginalis and Lactobacillus acidophilus were absent. In the CST patients diagnosed with LSIL HPV(+) the predominant types of bacteria were Lactobacillus acidophilus and Lactobacillus iners, and there was no presence of Lactobacillus crispatus detected. The CST cervical swabs obtained from women with HSIL HPV(+) were rich in Gardnerella vaginalis and Lactobacillus acidophilus, while Lactobacillus taiwanensis, Lactobacillus iners and Lactobacillus crispatus were not detected. However, similar levels of Lactobacillus acidophilus were identified in the control and LSIL HPV(+) groups.

In the majority of healthy women in the present study, Lactobacillus crispatus, Lactobacillus iners and Lactobacillus taiwanensis were the dominant cervical microbiota. According to previous studies, these species were also reported to be dominant in the vaginal microbiota of Asian women (6,13). The women with the highest risk of HSIL HPV(+) indicated low levels of Lactobacillus crispatus, Lactobacillus iners and Lactobacillus taiwanensis. It has been reported that hydrogen peroxide-producing Lactobacilli are present in 96% of women with a normal vaginal bacterial community (4). In addition, species in the Lactobacillus genus have been reported to maintain a low pH by producing lactic acid (4). It has been reported that Lactobacillus crispatus was associated with a low vaginal pH compared with Lactobacillus iners, suggesting that these two species differ in ecological function (14). The cervical swabs obtained from healthy women consisted mainly of Lactobacillus bacteria. Due to their presence, the physiological pH value (3.6–4.5) was achieved by fermentation of epithelial glycogen. In addition, it has been reported that Lactobacillus bacteria have a high adherence potential, which allows them to adhere tightly to the vaginal epithelium and cover its surface, protecting the vagina from colonisation by pathogenic microorganisms (15–17). A number of Lactobacillus species can form biofilms and produce bacteriocins and hydrogen peroxide, which together have been reported to inhibit the development of undesirable anaerobes in the vagina (15,16). Lactobacillus iners has been previously reported to become a predominant part of the microbial community in the vaginal microbiota transition between abnormal and normal states (17). HPV infection can alter the mucosal metabolism and host immunity, and can induce changes in the vaginal microbiota (18).

It has been reported that microbe-induced inflammation can contribute to cervical cancer by stimulating the production of specific cytokines and chemokines, which promote proliferation and/or inhibit apoptosis (18,19). Microbiota have been demonstrated to increase, as well as decrease susceptibility to HPV infection (18,19). An association between vaginal microbiota and HPV infection was previously described by Gao et al (19). They observed a significant presence of Lactobacillus species, including Lactobacillus gallinarum, Lactobacillus iners and Lactobacillus gasseri in all women as well as Lactobacillus gasseri and Gardnerella vaginalis in HPV(+) women. The reduced population of the Lactobacillus species and the presence of Fusobacteria species in HPV(+) patients was also observed by Lee et al (6). The present study demonstrated that bacterial communities in the cervix are more complex than previously thought. The analyses suggest an association between HPV infection and decreased abundance of Lactobacillus species and increased abundance of Gammaproteobacteria anaerobes. The aforementioned results are similar to those reported by Lee et al (6) and Dareng et al (8). The changes to the ‘core microbiome’ may be associated with changes in human health and risk of exposure to HPV infection and cervical cancer development.

In summary, analysis of the association between the occurrence of bacterial species and the histopathological diagnosis in the analysed population revealed that the vaginal microbiota may be clustered into two groups. Cluster 1 was predominantly affected by Lactobacillus crispatus, Lactobacillus iners and Lactobacillus taiwanensis and Cluster 2 predominated by Gardnerella vaginalis. The frequency of Lactobacillus acidophilus was identical in the clustered groups. The proposed mechanism of a cervical cancer formation may start with a sexually transmitted carcinogenic HPV infection (Fig. 1).

The fastest rate of HPV clearance may occur in women with Cluster 1 bacteria as the dominant microbiota as >50% of all infections were cleared within a year. The lowest clearance rate may occur in women classified with Cluster 2 dominant microbiota. In the aforementioned patients the chance of HSIL(+) development gradually increases, representing the growth of a clonal high-grade lesion up to a number of years following HPV infection. It was recently demonstrated that Gardnerella vaginalis, a dominant Cluster 2 bacterium, was able to adhere to and displace precoated protective Lactobacilli from vaginal epithelial cells, while other BV-associated anaerobes, including Atopobium vaginae, were less virulent (20).

The findings of the present study demonstrated a possible interaction between bacterial flora and HPV infection, as well as an association between this interaction and clinical cervical neoplasia. It was observed that bacterial dysbiosis, characterized by a predominance of Gardnerella vaginalis and a concomitant paucity of Lactobacillus crispatus, Lactobacillus iners and Lactobacillus taiwanensis, may be an HPV-dependent cofactor for cervical neoplasia development. However, without continuous observation it is difficult to confirm that microbiota dysbiosis contributes to HPV infection and carcinogenesis. Future researches are required to confirm the results of 16S rRNA sequencing and determine that microflora dysbiosis may be associated with HPV-induced cervical carcinogenicity.