Characterization of culturable vaginal Lactobacillus species among women with and without bacterial vaginosis from the United States and India: a cross-sectional study

Abstract

Lactobacillus species play an integral part in the health of the vaginal microbiota. We compared vaginal Lactobacillus species in women from India and the USA with and without bacterial vaginosis (BV). Between July 2009 and November 2010, a cross-sectional study was conducted among 40 women attending a women’s health clinic in Mysore, India, and a sexually transmitted diseases clinic in San Francisco, USA. Women were diagnosed with BV using Amsel’s criteria and the Nugent score. Lactobacillus 16S rDNA was sequenced to speciate the cultured isolates. Ten Indian and 10 US women without BV were compared with an equal number of women with BV. Lactobacilli were isolated from all healthy women, but from only 10 % of Indian and 50 % of US women with BV. 16S rDNA from 164 Lactobacillus colonies was sequenced from healthy women (126 colonies) and women with BV (38 colonies). Seven cultivable Lactobacillus species were isolated from 11 Indian women and nine species from 15 US women. The majority of Lactobacillus species among Indian women were L. crispatus (25 .0%), L. jensenii (25.0 %) and L. reuteri (16.7 %). Among US women, L. crispatus (32.0 %), L. jensenii (20.0 %) and L. coleohominis (12.0 %) predominated. L. jensenii and L. crispatus dominated the vaginal flora of healthy Indian and US women. Indian women appeared to have a higher percentage of obligate heterofermentative species, suggesting the need for a larger degree of metabolic flexibility and a more challenging vaginal environment.

Introduction

It is well accepted that Lactobacillus species are a critical component of the vaginal microflora of healthy women. These Gram-positive rods have been shown to have a protective effect against overgrowth by pathogenic micro-organisms (Thomas, 1928). While studies have demonstrated that the vaginal microflora is dominated by four Lactobacillus species, L. crispatus, L. jensenii, L. gasseri and L. iners, there is substantial heterogeneity among different human populations (Pavlova et al., 2002).

Studies in the USA, Europe and Japan have shown that women are predominantly colonized by obligate homofermentative lactobacilli that produce only lactic acid (Giorgi et al., 1987; Antonio et al., 1999; Song et al., 1999; Vásquez et al., 2002; Martín et al., 2008). Pavlova et al. (2002) found that the vaginal microbiota was dominated by L. crispatus, L. jensenii and L. gasseri in seven countries, while the microbiota of women from several low- and middle-income countries more often included heterofermentative Lactobacillus species such as L. vaginalis, L. fermentum and L. rhamnosus. Other studies have also shown greater numbers of heterofermentative vaginal Lactobacillus among women from various developing countries (Ocana et al., 1999; Jin et al., 2007; Martinez et al., 2008; Garg et al., 2009; Damelin et al., 2011; Kazi et al., 2012), compared with women in the USA and Europe (Antonio et al., 1999; Vásquez et al., 2002). The aim of this study was to characterize vaginal lactobacilli species in US and Indian women with and without bacterial vaginosis (BV).

Methods

Between July 2009 and November 2010, participants were recruited from a reproductive health clinic in Mysore, India, and a sexually transmitted diseases clinic in San Francisco, USA. Study eligibility criteria included age 18–35 years, sexually active and the ability to provide informed consent. Women who were pregnant, menstruating or with a diagnosis of any reproductive tract infections except BV, who had used antibiotics in the prior 30 days or who had a history of taking probiotics were excluded. The study was approved by the institutional review boards of the University of California, Berkeley, and the Public Health Research Institute of India (PHRII).

All participants underwent a physical examination and biological specimens were collected to detect reproductive tract infections. The diagnosis of BV was initially based on the criteria of Amsel et al. (1983). Three vaginal swab specimens were obtained from the posterior fornix of the vagina. The first swab was used to measure the vaginal pH, with the swab smeared onto a microscope slide and then placed in a tube containing four drops of normal saline for wet-mount preparation. The remaining two swabs were transferred to BBL Port-A-Cul tubes (Becton Dickinson) and stored at −20 °C until further use. Microscope slides were Gram stained to obtain the Nugent score (Nugent et al., 1991). The vaginal flora was defined as ‘healthy’ if the score was 0–3; ‘intermediate’ at 4–6; and ‘BV’ at ≥7. Only women diagnosed as ‘healthy’ or ‘BV’ were included in the analyses. Lactobacilli were characterized by type of glucose fermentation as obligate homofermentative, facultative heterofermentative or obligate heterofermentative species according to the review of taxonomic information gathered by Felis & Dellaglio (2007).

In India, the two swabs were transported to the PHRII laboratory. The swabs were plated on Rogosa and sheep blood agar (HiMedia Laboratories) within 12 h of collection, and incubated for 24–48 h at 37 °C in 5 % CO₂ anaerobic jars containing AnaeroPacks (Mitsubishi). In the USA, the swabs were cultured on Rogosa (BD Diagnostics) and Columbia blood agar plates (Hardy Diagnostics) and incubated as described above. Only plates containing 10–200 c.f.u. were further analysed. Up to 10 colonies were randomly selected from each plate. Gram-positive rods were further evaluated.

A modified version of the bead-beating method was used to extract bacterial DNA for PCR analysis (Rantakokko-Jalava & Jalava, 2002). Lactobacillus 16S rDNA PCR amplification was carried out according to a reported protocol with the primers published previously (Martinez-Freijo et al., 1998). Escherichia coli ATCC 25922 DNA was used as a positive control for the 16S rDNA. PCR products were purified using an AxyPrep kit (Axygen) or GeneJET gel extraction kit (Fermentas) in India and using ExoSAP-IT (USB) in the USA, and sequenced on a 3730 DNA analyser (Applied Biosystems) at the DNA Sequencing Facility of the University of California, Berkeley, USA, and at SciGenom or Eurofins Genomics, India. Genus and species were determined by the criteria of 98 % sequence identity for species and 95 % sequence identity for genus (Bäckhed et al., 2005).

Data analysis.

Data were analysed using sas version 9.3 (SAS Institute). A Fisher’s exact test was performed to test the relationship between BV status and the presence of any lactobacilli and specific Lactobacillus species. Two-tailed P values <0.05 were considered statistically significant. Only samples with confirmed blast identity ≥95 % and match length ≥500 bp were included in the analysis. Three BV-positive vaginal samples from India were excluded because they did not meet these criteria.

Results and Discussion

Vaginal samples were obtained from 40 reproductive-age women. Twenty women were healthy (10 Indian, 10 US) and 20 had BV (10 Indian, 10 US). All 20 US women reported inconsistent condom use, and none used hormonal contraceptives. Enrolled US participants were seen at the clinic to screen for eligibility to participate in a larger research study, and were not regular patients of a sexually transmitted diseases clinic with symptoms or complaints. A majority of the Indian women reported tubal ligation for birth control (n = 16; 80 %) and the remainder used no contraceptive methods. The primary reason for Indian participants to attend the reproductive health clinic was a routine health check-up.

Of the 40 women, 37 (17 Indian and 20 US) had Lactobacillus 16S rDNA with a sequence match length of ≥500 bp and blast identity ≥95 %. Of these 37 women, 26 (70 %; 11 Indian and 15 US) had cultivable Lactobacillus colonies. Lactobacilli were isolated from all healthy Indian and US women, but from only one Indian and five US women with BV. Lactobacillus 16S rDNA was successfully sequenced for 164 isolates. Unsurprisingly, 126 isolates were from healthy women (58 isolates from India and 68 from the USA) and only 38 isolates were from women with BV (10 isolates from India and 28 from the USA). Overall, 76.8 % of cultivable Lactobacillus species were obtained from women with a Nugent score of 0–3, indicating normal vaginal microflora, and 23.2 % were obtained from women with a Nugent score of 7–10, indicating BV.

Seven Lactobacillus species were isolated from 11 of the 20 Indian women and nine species were isolated from 15 of the 20 US women. Of the 14 cultivable Lactobacillus species, seven (50 %) were from Indian women and nine (64 %) were from US women (Table 1). The predominant Lactobacillus species among Indian women were L. crispatus, L. jensenii and L. reuteri; while L. crispatus, L. jensenii and L. coleohominis were most commonly found among US women (Table 1).

Table 1. Distribution of colonizing Lactobacillus species by country and BV status.

Only species that were cultivable in the study are presented.

Lactobacillus species	Indian women		US women
	Healthy	BV positive	Healthy	BV positive
L. acidophilus	1	–	–	–
L. coleohominis	–	–	1	2
L. crispatus	3	–	7	1
L. fermentum	1	–	–	–
L. gasseri	–	–	–	1
L. iners	–	–	2	–
L. jensenii	3	–	4	1
L. johnsonii	–	–	1	1
L. mucosae	1	–	–	–
L. reuteri	1	1	–	–
L. rhamnosus	–	–	2	–
L. ruminis	–	–	–	1
L. salivarius	1	–	–	–
L. vaginalis	–	–	1	–

Among the 26 colonized women, nine had more than one Lactobacillus species. L. acidophilus, L. fermentum, L. iners, L. mucosae, L. rhamnosus, L. salivarius and L. vaginalis were only isolated from women with normal microbiota, whereas the other predominant Lactobacillus species (L. coleohominis, L. crispatus, L. jensenii, L. johnsonii and L. reuteri) were isolated from women both with and without BV. Colonization by any lactobacilli, and in particular L. crispatus, was significantly associated with a normal vaginal microbiota, while L. reuteri and L. coleohominis were significantly associated with BV (Table 2). A majority of the Lactobacillus species among Indian women were heterofermentative (Table 3).

Table 2. Distribution of cultivable Lactobacillus species isolates by BV status among US and Indian women.

Percentages represent column totals.

	Healthy women, n (%)	BV-positive women, n (%)	Total, n (%)	P value
Any Lactobacillus
Absent	86 (40.6)	122 (76.3)	208 (55.9)	0.0001
Present	126 (59.4)	38 (23.8)	164 (44.1)
L. acidophilus
Absent	124 (98.4)	38 (100)	162 (98.8)	1.000
Present	2 (1.6)	0 (0)	2 (1.2)
L. coleohominis
Absent	125 (99.2)	21 (55.3)	146 (89.0)	<0.0001
Present	1 (0.79)	17 (44.7)	18 (10.9)
L. crispatus
Absent	67 (53.2)	37 (97.4)	104 (63.4)	<0.0001
Present	59 (46.8)	1 (2.7)	60 (36.6)
L. fermentum
Absent	125 (99.2)	38 (100)	163 (99.4)	1.000
Present	1 (0.8)	0 (0)	1 (0.61)
L. gasseri
Absent	126 (100)	36 (94.7)	162 (98.8)	0.052
Present	0 (0)	2 (5.26)	2 (1.22)
L. iners
Absent	124 (98.4)	38 (100)	162 (98.8)	1.000
Present	2 (1.59)	0 (0)	2 (1.22)
L. jensenii
Absent	100 (79.4)	35 (92.1)	135 (82.3)	0.089
Present	26 (20.6)	3 (7.9)	29 (17.7)
L. johnsonii
Absent	122 (96.8)	35 (92.1)	157 (95.7)	0.354
Present	4 (3.2)	3 (7.9)	7 (4.3)
L. mucosae
Absent	121 (96.0)	38 (100)	159 (96.9)	0.590
Present	5 (3.97)	0 (0)	5 (3.05)
L. reuteri
Absent	116 (92.1)	28 (73.7)	144 (87.8)	0.008
Present	10 (7.94)	10 (26.3)	20 (12.2)
L. rhamnosus
Absent	124 (98.4)	38 (100)	162 (98.8)	1.000
Present	2 (1.59)	0 (0)	2 (1.22)
L. ruminis
Absent	126 (100)	36 (94.7)	162 (98.8)	0.052
Present	0 (0)	2 (5.26)	2 (1.22)
L. salivarius
Absent	119 (94.4)	38 (100)	157 (95.7)	0.202
Present	7 (5.56)	0 (0)	7 (4.27)
L. vaginalis
Absent	121 (96.0)	38 (100)	159 (96.9)	0.590
Present	5 (3.97)	0 (0)	5 (3.06)

Table 3. Percentage of cultivable Lactobacillus species classified by glucose fermentation among women with and without BV in India and the USA.

Glucose fermentation	BV positive, n (%)		BV negative, n (%)
	India	USA	India	USA
Facultative heterofermentative	–	20 (71.4)	15 (26.8)	14 (20.6)
Obligate heterofermentative	10 (100.0)	–	16 (28.6)	5 (7.4)
Obligate homofermentative	–	8 (28.6)	25 (44.6)	49 (72.1)

Our study shows that the vaginal flora of healthy Indian and US women is dominated by two cultivable Lactobacillus species (L. crispatus and L. jensenii), similar to the findings of Pavlova et al. (2002). These species have been suggested to be beneficial in maintaining a normal vaginal microflora and preventing sexually transmitted infections (Antonio et al., 1999; Pavlova et al., 2002; Antonio et al., 2005; Damelin et al., 2011). Our findings, however, differ from those of an Indian study by Garg et al. (2009), who found that L. reuteri, L. fermentum and L. salivarius were the main species in the vaginal flora of women in Delhi, northern India. These differences suggest that vaginal Lactobacillus species may vary in subpopulations of India.

Our findings reveal that despite differences in ethnicity and geographical setting, healthy women share similar vaginal Lactobacillus species (especially L. crispatus and L. jensenii), but that the vaginal ecology is altered when women develop BV. The finding that the vaginal microbiome was more heavily weighted toward heterofermentative species among the Indian women in our sample is consistent with the literature, but deserves more attention (Gonzalez et al., 2011).

There are several limitations to this study. First, because we used culture, only species that could be cultured were identified. Selective media such as Rogosa or MRS agar do not support the growth of certain species such as L. iners (Fredricks et al., 2005; Vasquez et al., 2005). Cultivation-independent methods have found that L. iners is one of the most frequently isolated organisms from the vagina of healthy women (Vasquez et al., 2002; Anukam et al., 2006; Zhou et al., 2007). Our negative result on isolation of L. iners with blood agar in samples from Indian women does not rule out the possibility that this species was not present in the population, because small colonies may have been easily overlooked. Secondly, the differences in the diversity and distribution of Lactobacillus species may have resulted from participants being in different phases of their menstrual cycle. Finally, findings from our study cannot be generalized to other populations because of the small sample size, non-probability sample and limited statistical power.

Despite these limitations, our study fills a gap in the literature since little is known about the vaginal flora of Indian women. These data suggest that more needs to be known about the microbial diversity of vaginal flora of women in different parts of the world and how it affects vaginal health. Larger studies using genotyping methods and longitudinal study designs are needed to better characterize this important vaginal defence.

Abbreviations:

BV

bacterial vaginosis

PHRII

Public Health Research Institute of India