Cultivable Microbiome Approach Applied to Cervical Cancer Exploration

Abstract

Traditional microbiological methodology is valuable and essential for microbiota composition description and microbe role assignations at different anatomical sites, including cervical and vaginal tissues; that, combined with molecular biology strategies and modern identification approaches, could give a better perspective of the microbiome under different circumstances. This pilot work aimed to describe the differences in microbiota composition in non-cancer women and women with cervical cancer through a culturomics approach combining culture techniques with Vitek mass spectrometry and 16S rDNA sequencing. To determine the possible differences, diverse statistical, diversity, and multivariate analyses were applied; the results indicated a different microbiota composition between non-cancer women and cervical cancer patients. The Firmicutes phylum dominated the non-cancer (NC) group, whereas the cervical cancer (CC) group was characterized by the predominance of Firmicutes and Proteobacteria phyla; there was a depletion of lactic acid bacteria, an increase in the diversity of anaerobes, and opportunistic and non-typical human microbiota isolates were present. In this context, we hypothesize and propose a model in which microbial composition and dynamics may be essential for maintaining the balance in the cervical microenvironment or can be pro-oncogenesis microenvironmental mediators in a process called Ying-Yang or have a protagonist/antagonist microbiota role.

Keywords: aerobic and anaerobic cultures; cervical cancer; cervical epithelium; culturomics; microbiology.

Figure 1

Cervicovaginal microbiota detected in cytology samples taken from non-cancer and cervical cancer patients.

Figure 2

Differences in relative phyla abundance between non-cancer and cervical cancer groups. (A) The box plots compare the mean number of isolations obtained for NC and CC populations; despite the difference in the identity of isolates for each group, the * p-value higher than 0.05 was considered insignificant. (B) Cervical microbiota was sorted in phyla taxa; the average relative abundances of each phylum were represented on the bar chart as an accumulative percentage, and to determine the predilection of each phylum for a group of patients, non-cancer or with cervical cancer, Fisher’s exact test was used (** p < 0.05). Firmicutes and Fusobacteria phyla were the most abundant in both groups; meanwhile, Proteobacteria and the other phyla were more significant in the CC group.

Figure 3

Microbiota diversity comparison between non-cancer and cervical cancer groups of women. The graphs show the results for alpha and beta diversities. (A) Alpha diversity profiles. (A1) show that non-cancer microbiota (NC blue line) have more specific genus 1 dominance, whereas microbiota from the CC group (Red line) are more diverse. The upper square shows the diversity increase attributable to treatment (CC + T green line) and cancer Bethesda stages (CC1 purple line, CC2 pink line, CC3 yellow line and CC4 aqua line). (A2) Diversity indexes, the table show there is more 2 richness and 3 abundance in the number of genera on CC (red value), meanwhile NC has more 1 dominance (blue value), however, both groups have similar 4 evenness or transitory microbiota presence, the graph under the table exemplify pictorially the diversity indexes meaning for NC (left) (dominance index) and CC (right) (richness and abundance indexes), using different microbial morphologies examples (different color and shapes). (B) Beta diversity indicates the changes in the genera composition in proportions of 0.41 according to the Whittaker index and rarefaction curves, NC (blue line) and CC (red line) with 95% confidence intervals (green lines); the graph at right is useful to illustrate that NC (blue) and CC (red) microecosystems are distinct in microbiota populations, despite that initially they shared some core genera.

Figure 4

Relevant genera of the microbiota identified in non-cancer and cervical cancer groups. (A) PCA plot shows that the relevant genera in microbiota from the non-cancer group are Lactobacillus, Gardnerella, and Bifidobacterium; meanwhile, Streptococcus, Candida, and Bacteroides are more relevant for the CC group, although, Staphylococcus, Corynebacterium, Escherichia, Peptoniphulus, Prevotella, and Enterococcus are relevant to both groups. (B) Hierarchical clustering analysis for both populations generate seven clusters (A–G) corresponding to isolation frequency correlations among microbiota genera. Clusters A–B had Lactobacillus dominance, mainly from non-cancer women (NC); meanwhile, clusters C–G dominated CC women samples. The more relevant bacteria for each cluster were included in blocks of color; we observed four principal characteristics: the dominance of Lactobacillus genera in the NC group, the constant presence of opportunistic and transitory genera, the possible influence of some specific genus on CC microbiota, and the permanence or prevalence of other specific genera in both groups, named here for illustrative purposes as “scavengers”.

Figure 5

Distribution of cervical microbiota in study populations. The image shows the transformation process from normal cervical epithelium to invasive carcinoma and the microbiota composition associated with each stage. Initially, the “Eubiosis” condition on healthy epithelium was characterized by the dominance of Lactobacillus spp. and other lactic acid bacteria. However, the depletion of these bacteria caused a dysbiosis condition characterized by increased diversity of opportunistic and transitory microbiota established during cervical disease progression. We proposed the * “Anaerobiosis” status in cancerous cells because of a noticeable increase in anaerobic bacteria. Here, it is proposed that the role of Staphylococcus, Escherichia, and Corynebacterium (SEC) added to Peptoniphulus, Prevotella, and Enterococcus genera, together denominated as “Scavengers”, have constant participation in non-cancer and cancer populations. The transitional stage could be an opportunity for a therapeutic approach with probiotics. ** Shows the possible areas for implementing mechanistic studies to understand and improve the model.