The lower genital tract microbiota in relation to cytokine‐, SLPI‐ and endotoxin levels: application of checkerboard DNA‐DNA hybridization (CDH)

Abstract

In the present study the lower genital tract microbiota in asymptomatic fertile women (n=34) was identified and quantified by culturing vaginal secretions. Also, vaginal and cervical samples were analyzed by a semiquantitative checkerboard DNA-DNA hybridization technique (CDH) based on genomic probes prepared from 13 bacterial species (Bacteroides ureolyticus, Escherichia coli, Fusobacterium nucleatum, Gardnerella vaginalis, Mobiluncus curtisii ss curtisii, Prevotella bivia, Prevotella disiens, Prevotella melaninogenica, Atopobium vaginae, Lactobacillus iners, Staphylococcus aureus ss aureus, Streptococcus anginosus, and Streptococcus agalactiae). The bacterial species found by either culture or CDH were correlated with proinflammatory cytokines (IL-1 alpha, IL-1 beta, IL-6, IL-8), secretory leukocyte protease inhibitor (SLPI), and endotoxin in the cervicovaginal samples. Grading the women into healthy, intermediate, or bacterial vaginosis (BV) as based on Gram staining of vaginal smears, the viable counts of lactobacilli (L. gasseri) and of streptococci-staphylococci combined were highest in the intermediate group. In BV, particularly the high concentrations of Actinomyces urogenitalis, Atopobium vaginae, and Peptoniphilus harei were noted (>or=10(11) per ml). The total viable counts correlated with both cervical IL-1 alpha and IL-1 beta. A strong negative correlation was observed between L. iners and total viable counts, G. vaginalis, or cervical IL-1 alpha, while it correlated positively with SLPI. Analysis of vaginal and cervical samples from 26 out of the 34 women by CDH showed that anaerobic bacteria were more frequently detected by CDH compared to culture. By this method, A. vaginae correlated with G. vaginalis, and L. iners with S. aureus. With regard to cytokines, B. ureolyticus correlated with both cervical and vaginal IL-1 alpha as well as with cervical IL-8, while F. nucleatum, S. agalactiae, S. anginosus, or S. aureus correlated with vaginal IL-1 alpha. Furthermore, all Gram-negative bacteria taken together, as measured by CDH, correlated with vaginal endotoxin and inversely with vaginal SLPI. The significance of the results is discussed. In summary, mapping of the identity and quantity of vaginal bacterial species and their association with locally produced host innate immune factors will help in defining various types of abnormal vaginal microbiota, developing new ways of assessing the risk of ascending subclinical infections, and in treating them. CDH appears to be a suitable tool for future analyses of large numbers of clinical samples with an extended number of bacterial probes.