Comparison of the diversity of the vaginal microbiota in HIV-infected and HIV-uninfected women with or without bacterial vaginosis.
Journal: 2008/November - Journal of Infectious Diseases
ISSN: 0022-1899
Abstract:
BACKGROUND
Whether human immunodeficiency virus (HIV) infection is associated with a change in the diversity of genital microbiota in women was investigated.
METHODS
Amplicon length heterogeneity polymerase chain reaction (LH-PCR) analysis and pyrosequencing of the 16S ribosomal RNA gene were used to analyze the diversity of the microbiota in HIV-positive (HIV(+)) and HIV-negative (HIV(-)) women with or without bacterial vaginosis (BV).
RESULTS
LH-PCR analysis revealed significantly more microbiota diversity in BV-positive (BV(+)) women than in BV-negative (BV(-)) women, but no significant difference was noted between HIV(+) women and HIV(-) women. Pyrosequencing revealed that Lactobacillus organisms constituted a median of 96% of the bacteria in BV(-) women. BV(+) women had a significantly higher number of taxa found at>> or =1% of the total genital microbiota (median, 11 taxa). Common taxa in BV(+) women were Prevotella, Megasphaera, Gardnerella, Coriobacterineae, Lachnospira, and Sneathia. There was a trend (P = .07) toward the presence of a higher number of taxa in HIV(+)BV(+) women than in HIV(-)BV(+) women. Propionibacterineae, Citrobacter, and Anaerococcus were the taxa found only in HIV(+) women (P < .05).
CONCLUSIONS
The present study demonstrated that both LH-PCR analysis and pyrosequencing differentiated microbiota in BV(+) women from that in BV(-) women and that pyrosequencing indicated a trend toward increased diversity in BV(+)HIV(+) women, suggesting that HIV infection is associated with changes in the diversity of genital microbiota.
Relations:
Content
Citations
(60)
References
(32)
Grants
(4K+)
Diseases
(2)
Drugs
(1)
Chemicals
(2)
Organisms
(3)
Processes
(1)
Anatomy
(1)
Affiliates
(1)
Similar articles
Articles by the same authors
Discussion board
J Infect Dis 198(8): 1131-1140

Comparison of the Diversity of the Vaginal Microbiota in HIV-infected and -uninfected women with and without Bacterial Vaginosis

Background

This study investigated whether HIV-infection is associated with a change in diversity of genital microbiota in women.

Methods

Amplicon length heterogeneity PCR (LH-PCR) and pyrosequencing of the 16S rRNA gene were used to analyze diversity of the microbiota from HIV-positive (HIV+) and HIV-negative (HIV-) women with or without bacterial vaginosis (BV).

Results

LH-PCR analysis showed significantly more diversity in BV-positive (BV+) women than in BV-negative (BV-) women, but no significant difference between HIV+ women and HIV- women. Pyrosequencing revealed that Lactobacillus constituted a median of 96% of the bacteria in BV- women. BV+ women had a significantly higher number of taxa found at ≥ 1% of the microbiota (median of 11). Common taxa in BV were Prevotella, Megasphaera, Gardnerella, Coriobacterineae, Lachnospira, and Sneathia. There was a trend (p=0.07) toward a higher number of taxa in HIV+BV+ compared to HIV-BV+ women. Propionibacterineae, Citrobacter and Anaerococcus were found only in HIV+ women (p<0.05).

Conclusions

This study showed that both LH-PCR and pyrosequencing differentiated BV+ from BV- microbiota and that pyrosequencing indicated a trend toward increased diversity in BV+HIV+ suggesting that HIV-infection is associated with changes in diversity of genital microbiota.

Introduction

The lower genital tract of women can be colonized by many types of bacteria. In some women the genital microbiota consists predominantly of Lactobacillus sp.. In other women, a common condition called bacterial vaginosis (BV) exists where Lactobacillus is not the principal bacteria type, but instead diverse and variable mixtures of other bacteria are present [1, 2].

The types of genital microbiota that are present in women have been associated with acquisition or expression of several sexually transmitted infections (STIs). For example, BV is associated with a higher risk of HIV infection [3-6] and HIV-infected women with BV have higher levels of HIV in genital secretions [7-9]. BV is also associated with increased susceptibility to infection with herpes simplex virus type 2, gonorrhea, Trichomonas vaginalis and Chlamydia trachomatis [10-13].

Essentially all studies of associations between STIs and genital microbiota have used clinical-based methods to identify BV, such as the Amsel criteria, or gram stains of bacteria in mucosal secretions, such as the Nugent criteria. The Amsel criteria assesses vaginal pH, presence of bacteria-coated epithelial cells (clue cells), release of amine odor upon addition of KOH and vaginal fluid consistency while the Nugent criteria evaluates the morphology and gram reactivity of bacteria. While these two methods are very effective for diagnosis of BV, the Amsel criteria provides no information on the types or diversity of genital microbiota present while gram stains provide limited information in this regard.

Recent studies used cloning and sequencing of the 16S rRNA gene for the identification of genital microbiota and confirmed that many bacteria that were previously identified by culture, such as Gardnerella vaginalis and Lactobacillus sp., are present at high levels in women with or without BV respectively. However, these studies also showed that bacteria that were previously unidentified or difficult to culture make up a substantial fraction of the BV microbiota in many women [14-17]. These molecular studies also confirmed that the types of bacteria and diversity of bacteria present in BV can be very different between individuals. For example, Fredricks et al. [14] observed that in nine women with BV, nine different taxa each contributed more than 10% of the total genital microbiota while 27 different taxa were found at 1% or more of the total microbiota in at least one of the women. In contrast, in seven out of eight BV-negative women, only Lactobacillus was found to make up more than 10% of the microbiota. Hyman et al. [15] found at least seven different taxa as the predominant bacterium in the genital microbiota from 10 women where Lactobacillus was not the principal type.

The lower genital tract microbiota in HIV+ women has not been studied with molecular techniques capable of identifying the diversity and/or relative proportions of bacteria types present in a culture-independent manner. We hypothesized that analysis of genital microbiota using these types of methods could reveal previously unknown relationships between HIV and microbiota, especially considering that BV is a range of different constellations of bacteria rather than one fixed set of bacteria. The goal of this study therefore, was to compare the diversity of genital flora in HIV+ and HIV- women using LH-PCR and pyrosequencing, methods useful for assessing diversity of microbiota in a culture-independent manner [18-22]. Because women with BV are known to have much more diverse microbiota than women without BV, both BV-positive and BV-negative women were included in the HIV+ and HIV- groups.

Materials and Methods

Patients and sample acquisition

All patients analyzed in this study were from the Women's Interagency HIV Study (WIHS). WIHS is a longitudinal, multi-center cohort study of HIV-infected and uninfected women followed at six clinical sites in the United States. Informed consent was obtained from all participants. A detailed interview, physical and gynecologic examination, and laboratory monitoring were performed at the time of sample donation [Barkan]. Genital tract samples were collected by cericovaginal lavage (CVL) that was performed by irrigation of the cervix with 10 mL of nonbacteriostatic sterile saline, followed by aspiration from the posterior fornix. CVL was held on ice until processing, which occurred within 6 h of collection. CVL was gently vortexed to evenly distribute cells before storage at -70 C.

CVL samples were obtained from 21 women divided into four groups based on HIV-seropositivity (HIV+ or HIV-) and Nugent gram stain analysis for BV. BV-positive (BV+) had scores of 7-10 while BV-negative (BV-) had scores of 0-3. Three groups each had five subjects (HIV+ BV+: HIV- BV+; and HIV- BV-) while the HIV+ BV- group had six subjects. The median ages for these groups were 39 for HIV+ BV+ (range 32-49), 39 for HIV+ BV- (33-44), 39 for HIV- BV+ (39-45) and 35 for HIV- BV- (25-45). The median CD4 counts and plasma viral loads of the HIV+ groups were 327 CD4+ cells/mm3 (range 0-555) and 15,000 HIV RNA copies/ml (range 7,500-210,000) for the HIV+BV+ group and 375 CD4+ cells/mm3 (range 14-726) and 42,000 HIV RNA copies/ml (range 6,400-120,000) for HIV+BV-. All of the women were in good general health, none of the women were undergoing current antibiotic treatment and none had current infection/colonization with Trichomonas or yeast (determined by wet mount and KOH). Chlamydia and GC were not assessed since the prevalence in this cohort was 1% and <1% respectively [23].

DNA isolation

DNA was isolated as described elsewhere [11]. In brief, bacteria from CVL samples were pelleted by centrifugation and treated with a lysis buffer containing lysosyme for 20 min at room temperature. Bacteria were further disrupted by addition of SDS and proteinase K for 30 min at 37 C. Lysates were mixed with phenol/chloroform and centrifuged. DNA was precipitated by incubation with absolute ethanol, with glycogen added as a carrier.

Length Heterogeneity PCR (LH-PCR)

For LH-PCR fingerprinting of the 16S RNA gene [18], extracted DNA (10 ng) was amplified by PCR by using a fluorescently labeled forward primer L27F (5′-[6FAM] AGAGTTTGATCCTGGCTCA G-3′) and unlabeled reverse primer 355R′ (5′-GCTGCCTCCCGTAGGAGT-3′). Both primers are broad range primers for bacteria [24]. The reactions were performed using 20 μl (final volume) mixtures containing 1 X PCR buffer, 0.01% bovine serum albumin, 2.5 mM MgCl2, 0.5 mM of each deoxynucleoside triphosphate, 0.2 μM of each primer, and 2 U of TaqGold DNA polymerase (Applied Biosystems, Foster City, CA). The initial denaturation step was 95°C for 11 min, followed by 30 cycles of denaturation at 95°C for 30 sec, annealing at 48°C for 30 sec, and extension at 72°C for 2 min, followed by a final extension step that consisted of 72°C for 35 min. LH-PCR samples were stored at 4°C in the dark until used for fingerprinting. To ensure these conditions would not over-amplify minor components, several dilutions of the original extracted DNA were amplified and run on LH-PCR to check that amplification was in the linear range for each sample. LH-PCR of negative controls determined that there was no detectable contamination products from the reagents used in the process. The LH-PCR products were separated on a SCE9610 capillary fluorescent sequencer (Spectrumedix LLC, State College, PA) and analyzed with GenoSpectrum™ software (Version 2.01). Negative and positive controls were analyzed for each run and gave appropriate results. The software package deconvolves the fluorescence data into electropherograms where the peaks of the electropherograms represent PCR amplicons of different length in base pairs (bp) and identify different genus/species/strains or Operational Taxanomic Units (OTUs) of microflora. The LH-PCR fingerprinting data were then analyzed using a custom PERL script (Interleave 1.0, BioSpherex LLC) that combines data from several runs, interleaves the various profiles, normalizes the data, calculates the averages for each amplicon size and determines diversity indices. The normalized peak areas were calculated by dividing an individual peak area by the total peak area in that profile. LH-PCR fingerprint patterns (i.e. presence or absence of certain amplicon peaks) were expressed as stacked histograms of the LH-PCR amplicon normalized abundances and analyzed by visual inspection and principle coordinate analysis (PCO). PCO was used to give a measure of the similarities between the cases directly, rather than the variables. The Multivariate Statistical Package (MVSP) (Kovach Computing Services, Wales, UK) was used for PCO analysis.

Multitag Pyrosequencing and Phylogenetic Analysis

Pyrosequencing and its use for determining diversity of microbial communities by sequencing of the 16S rRNA gene have been previously described [21, 22]. We developed and used a novel multiplexed pyrosequencing method by generating a set of twelve primers that each contained either the 27F or 355R primers (see above) that were tagged on the 5 prime end with a 4 base “barcode.” PCR was performed on individual patient samples using the unique barcoded primers and then 10 to 12 samples were pooled and ligated to the PCR linkers used in the emulsion step of pyrosequencing. All samples were amplified for 30 cycles as described above for the LH–PCR. Pyrosequencing of the amplified, tagged DNA was performed by 454 Life Sciences (Branford, CT) with 10 to 12 separately tagged samples included in a single slot. The data from each well was “deconvoluted” by sorting the sequences into bins based on the barcodes and the taxa in the samples was normalized by the total number of reads from each barcode.

We used custom PERL scripts to automatically sort the sequences based on the barcodes, search against the rRNA database, identify the reads and exclude sequences that had multiple tags which indicated chimeric sequences. We used the Bayesian Classifier provided by the Ribosomal Database II Project (RDP 9) that uses a posteriori probability to identify query sequence based on the occurrence of 7 base words in their rRNA database [25]. We then downloaded the annotations for each sequence and then used a PERL script to tabulate the taxa as a percentage of the total community in each sample. Sequences were aligned using Clustal X [26] and Neighbor Joining trees were constructed using PAUP software (Swofford, D. L., Phylogenetic Analysis Using Parsimony, 4th ed. Sinauer Associates, Sunderland, MA).

Patients and sample acquisition

All patients analyzed in this study were from the Women's Interagency HIV Study (WIHS). WIHS is a longitudinal, multi-center cohort study of HIV-infected and uninfected women followed at six clinical sites in the United States. Informed consent was obtained from all participants. A detailed interview, physical and gynecologic examination, and laboratory monitoring were performed at the time of sample donation [Barkan]. Genital tract samples were collected by cericovaginal lavage (CVL) that was performed by irrigation of the cervix with 10 mL of nonbacteriostatic sterile saline, followed by aspiration from the posterior fornix. CVL was held on ice until processing, which occurred within 6 h of collection. CVL was gently vortexed to evenly distribute cells before storage at -70 C.

CVL samples were obtained from 21 women divided into four groups based on HIV-seropositivity (HIV+ or HIV-) and Nugent gram stain analysis for BV. BV-positive (BV+) had scores of 7-10 while BV-negative (BV-) had scores of 0-3. Three groups each had five subjects (HIV+ BV+: HIV- BV+; and HIV- BV-) while the HIV+ BV- group had six subjects. The median ages for these groups were 39 for HIV+ BV+ (range 32-49), 39 for HIV+ BV- (33-44), 39 for HIV- BV+ (39-45) and 35 for HIV- BV- (25-45). The median CD4 counts and plasma viral loads of the HIV+ groups were 327 CD4+ cells/mm3 (range 0-555) and 15,000 HIV RNA copies/ml (range 7,500-210,000) for the HIV+BV+ group and 375 CD4+ cells/mm3 (range 14-726) and 42,000 HIV RNA copies/ml (range 6,400-120,000) for HIV+BV-. All of the women were in good general health, none of the women were undergoing current antibiotic treatment and none had current infection/colonization with Trichomonas or yeast (determined by wet mount and KOH). Chlamydia and GC were not assessed since the prevalence in this cohort was 1% and <1% respectively [23].

DNA isolation

DNA was isolated as described elsewhere [11]. In brief, bacteria from CVL samples were pelleted by centrifugation and treated with a lysis buffer containing lysosyme for 20 min at room temperature. Bacteria were further disrupted by addition of SDS and proteinase K for 30 min at 37 C. Lysates were mixed with phenol/chloroform and centrifuged. DNA was precipitated by incubation with absolute ethanol, with glycogen added as a carrier.

Length Heterogeneity PCR (LH-PCR)

For LH-PCR fingerprinting of the 16S RNA gene [18], extracted DNA (10 ng) was amplified by PCR by using a fluorescently labeled forward primer L27F (5′-[6FAM] AGAGTTTGATCCTGGCTCA G-3′) and unlabeled reverse primer 355R′ (5′-GCTGCCTCCCGTAGGAGT-3′). Both primers are broad range primers for bacteria [24]. The reactions were performed using 20 μl (final volume) mixtures containing 1 X PCR buffer, 0.01% bovine serum albumin, 2.5 mM MgCl2, 0.5 mM of each deoxynucleoside triphosphate, 0.2 μM of each primer, and 2 U of TaqGold DNA polymerase (Applied Biosystems, Foster City, CA). The initial denaturation step was 95°C for 11 min, followed by 30 cycles of denaturation at 95°C for 30 sec, annealing at 48°C for 30 sec, and extension at 72°C for 2 min, followed by a final extension step that consisted of 72°C for 35 min. LH-PCR samples were stored at 4°C in the dark until used for fingerprinting. To ensure these conditions would not over-amplify minor components, several dilutions of the original extracted DNA were amplified and run on LH-PCR to check that amplification was in the linear range for each sample. LH-PCR of negative controls determined that there was no detectable contamination products from the reagents used in the process. The LH-PCR products were separated on a SCE9610 capillary fluorescent sequencer (Spectrumedix LLC, State College, PA) and analyzed with GenoSpectrum™ software (Version 2.01). Negative and positive controls were analyzed for each run and gave appropriate results. The software package deconvolves the fluorescence data into electropherograms where the peaks of the electropherograms represent PCR amplicons of different length in base pairs (bp) and identify different genus/species/strains or Operational Taxanomic Units (OTUs) of microflora. The LH-PCR fingerprinting data were then analyzed using a custom PERL script (Interleave 1.0, BioSpherex LLC) that combines data from several runs, interleaves the various profiles, normalizes the data, calculates the averages for each amplicon size and determines diversity indices. The normalized peak areas were calculated by dividing an individual peak area by the total peak area in that profile. LH-PCR fingerprint patterns (i.e. presence or absence of certain amplicon peaks) were expressed as stacked histograms of the LH-PCR amplicon normalized abundances and analyzed by visual inspection and principle coordinate analysis (PCO). PCO was used to give a measure of the similarities between the cases directly, rather than the variables. The Multivariate Statistical Package (MVSP) (Kovach Computing Services, Wales, UK) was used for PCO analysis.

Multitag Pyrosequencing and Phylogenetic Analysis

Pyrosequencing and its use for determining diversity of microbial communities by sequencing of the 16S rRNA gene have been previously described [21, 22]. We developed and used a novel multiplexed pyrosequencing method by generating a set of twelve primers that each contained either the 27F or 355R primers (see above) that were tagged on the 5 prime end with a 4 base “barcode.” PCR was performed on individual patient samples using the unique barcoded primers and then 10 to 12 samples were pooled and ligated to the PCR linkers used in the emulsion step of pyrosequencing. All samples were amplified for 30 cycles as described above for the LH–PCR. Pyrosequencing of the amplified, tagged DNA was performed by 454 Life Sciences (Branford, CT) with 10 to 12 separately tagged samples included in a single slot. The data from each well was “deconvoluted” by sorting the sequences into bins based on the barcodes and the taxa in the samples was normalized by the total number of reads from each barcode.

We used custom PERL scripts to automatically sort the sequences based on the barcodes, search against the rRNA database, identify the reads and exclude sequences that had multiple tags which indicated chimeric sequences. We used the Bayesian Classifier provided by the Ribosomal Database II Project (RDP 9) that uses a posteriori probability to identify query sequence based on the occurrence of 7 base words in their rRNA database [25]. We then downloaded the annotations for each sequence and then used a PERL script to tabulate the taxa as a percentage of the total community in each sample. Sequences were aligned using Clustal X [26] and Neighbor Joining trees were constructed using PAUP software (Swofford, D. L., Phylogenetic Analysis Using Parsimony, 4th ed. Sinauer Associates, Sunderland, MA).

Length Heterogeneity PCR (LH-PCR)

For LH-PCR fingerprinting of the 16S RNA gene [18], extracted DNA (10 ng) was amplified by PCR by using a fluorescently labeled forward primer L27F (5′-[6FAM] AGAGTTTGATCCTGGCTCA G-3′) and unlabeled reverse primer 355R′ (5′-GCTGCCTCCCGTAGGAGT-3′). Both primers are broad range primers for bacteria [24]. The reactions were performed using 20 μl (final volume) mixtures containing 1 X PCR buffer, 0.01% bovine serum albumin, 2.5 mM MgCl2, 0.5 mM of each deoxynucleoside triphosphate, 0.2 μM of each primer, and 2 U of TaqGold DNA polymerase (Applied Biosystems, Foster City, CA). The initial denaturation step was 95°C for 11 min, followed by 30 cycles of denaturation at 95°C for 30 sec, annealing at 48°C for 30 sec, and extension at 72°C for 2 min, followed by a final extension step that consisted of 72°C for 35 min. LH-PCR samples were stored at 4°C in the dark until used for fingerprinting. To ensure these conditions would not over-amplify minor components, several dilutions of the original extracted DNA were amplified and run on LH-PCR to check that amplification was in the linear range for each sample. LH-PCR of negative controls determined that there was no detectable contamination products from the reagents used in the process. The LH-PCR products were separated on a SCE9610 capillary fluorescent sequencer (Spectrumedix LLC, State College, PA) and analyzed with GenoSpectrum™ software (Version 2.01). Negative and positive controls were analyzed for each run and gave appropriate results. The software package deconvolves the fluorescence data into electropherograms where the peaks of the electropherograms represent PCR amplicons of different length in base pairs (bp) and identify different genus/species/strains or Operational Taxanomic Units (OTUs) of microflora. The LH-PCR fingerprinting data were then analyzed using a custom PERL script (Interleave 1.0, BioSpherex LLC) that combines data from several runs, interleaves the various profiles, normalizes the data, calculates the averages for each amplicon size and determines diversity indices. The normalized peak areas were calculated by dividing an individual peak area by the total peak area in that profile. LH-PCR fingerprint patterns (i.e. presence or absence of certain amplicon peaks) were expressed as stacked histograms of the LH-PCR amplicon normalized abundances and analyzed by visual inspection and principle coordinate analysis (PCO). PCO was used to give a measure of the similarities between the cases directly, rather than the variables. The Multivariate Statistical Package (MVSP) (Kovach Computing Services, Wales, UK) was used for PCO analysis.

Multitag Pyrosequencing and Phylogenetic Analysis

Pyrosequencing and its use for determining diversity of microbial communities by sequencing of the 16S rRNA gene have been previously described [21, 22]. We developed and used a novel multiplexed pyrosequencing method by generating a set of twelve primers that each contained either the 27F or 355R primers (see above) that were tagged on the 5 prime end with a 4 base “barcode.” PCR was performed on individual patient samples using the unique barcoded primers and then 10 to 12 samples were pooled and ligated to the PCR linkers used in the emulsion step of pyrosequencing. All samples were amplified for 30 cycles as described above for the LH–PCR. Pyrosequencing of the amplified, tagged DNA was performed by 454 Life Sciences (Branford, CT) with 10 to 12 separately tagged samples included in a single slot. The data from each well was “deconvoluted” by sorting the sequences into bins based on the barcodes and the taxa in the samples was normalized by the total number of reads from each barcode.

We used custom PERL scripts to automatically sort the sequences based on the barcodes, search against the rRNA database, identify the reads and exclude sequences that had multiple tags which indicated chimeric sequences. We used the Bayesian Classifier provided by the Ribosomal Database II Project (RDP 9) that uses a posteriori probability to identify query sequence based on the occurrence of 7 base words in their rRNA database [25]. We then downloaded the annotations for each sequence and then used a PERL script to tabulate the taxa as a percentage of the total community in each sample. Sequences were aligned using Clustal X [26] and Neighbor Joining trees were constructed using PAUP software (Swofford, D. L., Phylogenetic Analysis Using Parsimony, 4th ed. Sinauer Associates, Sunderland, MA).

Results

LH-PCR analysis of genital microbiota in HIV+ and HIV- women

To determine the diversity of genital microbiota in subjects in the four groups (HIV+BV+; HIV+BV-; HIV-BV+; and HIV-BV-), duplicate PCRs were performed for each of the samples from 21 subjects and amplified products were analyzed for length. Figure 1 shows representative LH-PCR “fingerprints” of microbiota collected from subjects in each group. One to five peaks (median of 1) were observed in samples from the 11 BV- women (Fig. 1a and 1b). We define these peaks as ‘operational taxonomic units’ (OTUs) as it is unknown whether they represent a single genus, species, or strain. In contrast, analysis of microbiota in samples from the 10 BV+ subjects (Fig. 1c and 1d) resulted in 7 to 14 OTUs (median of 11, p=0.002 compared to BV-, Mann Whitney test).

An external file that holds a picture, illustration, etc.
Object name is nihms161100f1a.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f1b.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f1c.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f1d.jpg

LH-PCR plots of 16S rRNA amplified products from genital microbiota. Representative LH-PCR plots of three subjects from each group. A. HIV+BV-; B. HIV-BV-; C. HIV+BV+; D. HIV-BV+.

The five HIV+BV+ samples had a median of 11 OTUs (range 2 to 13) while the five HIV-BV+ samples had a median of 9.5 OTUs (range 7-14) (not significant). The six HIV+BV-samples had a median of 2 OTUs (range 1 to 3) while the five HIV-BV- samples had a median of 1 OTUs (range 1-5) (not significant).

Twenty-three different OTUs could be differentiated in the 21 subjects based on the differential mobilities of the amplified products (Fig. 2). When BV-positive and BV-negative samples were compared, women with BV had much greater diversity, with 21 different OTUs detected compared to 9 OTUs in BV-negative women. HIV-negative women had 18 OTUs while HIV-positive women had 20 OTUs. Two of the OTUs (362.6 and 372.3) were found at high levels in many of the 11 BV-negative subjects and several of these subjects had only one of the two OTUs present. These two PCR products were sequenced from several of the samples and corresponded to L. crispatus and L. johnsonii according to Ribosomal database 9 search.

An external file that holds a picture, illustration, etc.
Object name is nihms161100f2.jpg

The relative abundance of operational taxonomic units (OTUs). The relative abundance of the OTUs from each sample are presented as stacked histograms and color coded for the presence of vaginosis and HIV infection. Light blue are OTUs from HIV negative samples without vaginosis; dark blue are OTUs from HIV positive samples without vaginosis; light red are OTUs from HIV negative samples with vaginosis; and dark red are OTUs from HIV positive samples with vaginosis.

The LH-PCR data was analyzed by principle component analysis (PCO) where each point represents one sample and where the data are projected into a multidimensional plot onto two or three principle components using classic Eigen analysis (Fig. 3). The PCO plot shows overlap of subjects from the two BV+ groups indicating a trend of similarity in diversity in BV+ positive samples. The plots of the two BV- groups overlapped to a large extent suggesting that the diversity of these samples was very similar. BV+ and BV- subjects had little overlap indicating very different diversity between the BV+ and BV- groups.

An external file that holds a picture, illustration, etc.
Object name is nihms161100f3.jpg

Principal coordinate analysis of LH-PCR. Data from the LH-PCR was plotted by classic Eigen analysis. Diamonds=BV-negative, HIV-negative; Squares=BV-positive, HIV-negative; Triangles=BV-negative, HIV-positive; Circles=BV-positive, HIV-positive.

Pyrosequence analysis of genital microbiota in HIV+ and HIV- women

We also performed pyrosequencing of the samples in order to identify the bacteria present in genital microbiota, identify minor constituents of the microbiota, and determine the diversity of the microbiota. Pyrosequencing of samples from the 21 women resulted in 36,724 sequences with a median length of 249 base pairs (range 156 to 314, mean±SD of 250±13). Ribosomal database 9 search of the sequences revealed 137 distinct taxa of bacteria. The 10 samples from BV-positive women contained a total of 115 different taxa while the 11 BV-negative samples had 61 taxa. The median number (and range) of taxa in BV+ and BV- samples was 33 (19-46) and 12 (8-20) respectively (p=0.0003, Mann-Whitney). The types and relative proportions of taxa observed in representative subjects from each of the four groups are shown in figure 4a. The 5 samples from HIV+BV+ women contained a total of 103 different taxa while the 5 HIV-BV+ samples had 51 taxa with the median numbers in the two groups of 41 (range of 29-46) and 29 (19-36) respectively (0.1167, Mann-Whitney).

An external file that holds a picture, illustration, etc.
Object name is nihms161100f4a.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f4b.jpg

Relative abundance of taxa determined by pyrosequencing. A. The relative abundance of taxa from one representative individual in each of the four groups is shown. Taxa comprising <0.3% of the sequences are not shown for clarity. B. The relative abundance of taxa comprising ≥1% of the total averaged from all five women in the HIV+BV+ group and the HIV-BV+ group. The category “other” represents the proportion of sequences from taxa that averaged less than 1%.

Taxa found at ≥ 1% of total microbiota by pyrosequencing

We analyzed the taxa found at ≥ 1% of the total community under the a priori assumption that the most abundant taxa are the ones that contribute significantly to the functionality of the community. A total of 35 different taxa were found at ≥ 1% of the total microbiota in at least one of the 21 women (Table 1). The 10 BV+ samples contained 33 different taxa that were found at ≥1% in at least one sample while the 11 BV- samples contained 10 different taxa found at ≥1%. The median number of taxa found at ≥1% in the 10 BV+ women was 11 (range 8-18) while the median number in the 11 BV- samples was 2 (range 1-4) (p<0.0001, Mann-Whitney). The taxa most commonly found at ≥1% in BV+ samples included Prevotella, Megasphaera, Gardnerella, Coriobacterineae, Lachnospira, Sneathia and Actinomycineae (Table 1 and Fig. 5). Lactobacillus was found in all 11 BV- samples and in 5 of the BV+ samples at ≥1%. Comparison of the Lactobacillus sequences showed that this group was highly diverse. For example, 18 different clusters, or OTUs, of Lactobacillus sequences were found in the HIV+BV- women (Fig. 5) although most of the Lactobacilli were restricted to five different clusters.

An external file that holds a picture, illustration, etc.
Object name is nihms161100f5a.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f5b.jpg

Phylogenetic Analysis of Bacteria from Two Groups. Shown are the phylogenetic trees for the BV+HIV+ (A) and BV-HIV+ (B) groups. The total number of reads in the group is shown and taxa found at >200 reads are in bold.

Table 1

Number of Subjects with Taxa at ≥1% of Microbiota in at least one subject
Genusfound at ≥1%found at ≥10%
BV+BV-BV+BV-
Lactobacillus51111
Prevotella1028
Megasphaera10
Gardnerella9331
Coriobacterineaea911
Lachnospira827
Sneathia85
Actinomycineaeb8
Allisonella5
Micromonas5
Propionibacterineaec4121
Dialister4
Peptoniphilus3
Porphyromonas3
Comamonas22
Citrobacter21
Acinetobacter21
Mycoplasma21
Rhodobaca21
Proteiniphilum, Anaerotruncus and Corynebacterineae were found in two BV+ subjects at ≥1% Acetanaerobacterium, Achromobacter, Aerococcus, Anaerococcus, Faecalibacterium, Micrococcineae, Peptostreptococcus, Pseudomonas, Staphylococcus, Stenotrophomonas, and Xylanibacter were each found in one BV+ subject at ≥1%
Pelomonas and Sphingomonas were each found in one BV- subject at ≥1%
Atopobium and Eggerthella,
Predominantly Mobiluncus,
Predominantly Propionibacterium,
Predominantly Porynebacterium

The 5 HIV+BV+ women had a median of 14 taxa (range 10-18) found at ≥1% in contrast to the 5 HIV-BV+ samples that had a median of 10 (range 8-14) taxa (p=0.07, Mann-Whitney). Averages of taxa at ≥ 1% in the HIV+BV+ and HIV-BV+ groups are shown in Figure 4b. The 6 HIV+BV- samples had a median of 2 taxa (range 1-3) found at ≥1% while the 5 HIV-BV-samples had a median of 2 taxa (range 1-4) (p=0.93, Mann-Whitney).

Taxa found at ≥ 10% of total microbiota by pyrosequencing

A total of 10 different taxa were found at ≥ 10% of the total microbiota in at least one of the 21 women (Table 1). The 10 BV+ samples contained 9 different taxa that were found at ≥10% in at least one sample while the BV- samples contained 3 different taxa. Lactobacillus constituted at least 80% of the taxa in all BV- women (median of 96%). The median number of taxa found at ≥10% in the 10 BV+ women was 3 (range 1-4) while the median number in the 11 BV- samples was 1 (range 1-2) (p<0.0014, Mann-Whitney). The 5 HIV+BV+ samples had a median of 2 taxa (range 1-3) found at ≥10% while the 5 HIV-BV+ samples had a median of 3 (range 2-4) taxa (p=0.14, Mann-Whitney).

Taxa found only in HIV+ women by pyrosequencing

Four taxa were found in two or more of the HIV+BV+ women but none of the HIV-BV+ women. Thus, Propionibacterineae was found in all five of the HIV+BV+ women and in four of these women at >1% of the flora, but was not detected in any HIV-BV+ women (p=0.002, chi square analysis). Interestingly, Propionibacterineae was also detected in one of the HIV+BV-women but in none of the HIV-BV- women. Anaerococcus and Citrobacter were found in three of the HIV+BV+ women (p=0.04) while Acinetobacter was found in two of the HIV+BV+ women (p=0.11). No taxa were found only in HIV-BV+ women when compared with HIV+BV+ women.

LH-PCR analysis of genital microbiota in HIV+ and HIV- women

To determine the diversity of genital microbiota in subjects in the four groups (HIV+BV+; HIV+BV-; HIV-BV+; and HIV-BV-), duplicate PCRs were performed for each of the samples from 21 subjects and amplified products were analyzed for length. Figure 1 shows representative LH-PCR “fingerprints” of microbiota collected from subjects in each group. One to five peaks (median of 1) were observed in samples from the 11 BV- women (Fig. 1a and 1b). We define these peaks as ‘operational taxonomic units’ (OTUs) as it is unknown whether they represent a single genus, species, or strain. In contrast, analysis of microbiota in samples from the 10 BV+ subjects (Fig. 1c and 1d) resulted in 7 to 14 OTUs (median of 11, p=0.002 compared to BV-, Mann Whitney test).

An external file that holds a picture, illustration, etc.
Object name is nihms161100f1a.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f1b.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f1c.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f1d.jpg

LH-PCR plots of 16S rRNA amplified products from genital microbiota. Representative LH-PCR plots of three subjects from each group. A. HIV+BV-; B. HIV-BV-; C. HIV+BV+; D. HIV-BV+.

The five HIV+BV+ samples had a median of 11 OTUs (range 2 to 13) while the five HIV-BV+ samples had a median of 9.5 OTUs (range 7-14) (not significant). The six HIV+BV-samples had a median of 2 OTUs (range 1 to 3) while the five HIV-BV- samples had a median of 1 OTUs (range 1-5) (not significant).

Twenty-three different OTUs could be differentiated in the 21 subjects based on the differential mobilities of the amplified products (Fig. 2). When BV-positive and BV-negative samples were compared, women with BV had much greater diversity, with 21 different OTUs detected compared to 9 OTUs in BV-negative women. HIV-negative women had 18 OTUs while HIV-positive women had 20 OTUs. Two of the OTUs (362.6 and 372.3) were found at high levels in many of the 11 BV-negative subjects and several of these subjects had only one of the two OTUs present. These two PCR products were sequenced from several of the samples and corresponded to L. crispatus and L. johnsonii according to Ribosomal database 9 search.

An external file that holds a picture, illustration, etc.
Object name is nihms161100f2.jpg

The relative abundance of operational taxonomic units (OTUs). The relative abundance of the OTUs from each sample are presented as stacked histograms and color coded for the presence of vaginosis and HIV infection. Light blue are OTUs from HIV negative samples without vaginosis; dark blue are OTUs from HIV positive samples without vaginosis; light red are OTUs from HIV negative samples with vaginosis; and dark red are OTUs from HIV positive samples with vaginosis.

The LH-PCR data was analyzed by principle component analysis (PCO) where each point represents one sample and where the data are projected into a multidimensional plot onto two or three principle components using classic Eigen analysis (Fig. 3). The PCO plot shows overlap of subjects from the two BV+ groups indicating a trend of similarity in diversity in BV+ positive samples. The plots of the two BV- groups overlapped to a large extent suggesting that the diversity of these samples was very similar. BV+ and BV- subjects had little overlap indicating very different diversity between the BV+ and BV- groups.

An external file that holds a picture, illustration, etc.
Object name is nihms161100f3.jpg

Principal coordinate analysis of LH-PCR. Data from the LH-PCR was plotted by classic Eigen analysis. Diamonds=BV-negative, HIV-negative; Squares=BV-positive, HIV-negative; Triangles=BV-negative, HIV-positive; Circles=BV-positive, HIV-positive.

Pyrosequence analysis of genital microbiota in HIV+ and HIV- women

We also performed pyrosequencing of the samples in order to identify the bacteria present in genital microbiota, identify minor constituents of the microbiota, and determine the diversity of the microbiota. Pyrosequencing of samples from the 21 women resulted in 36,724 sequences with a median length of 249 base pairs (range 156 to 314, mean±SD of 250±13). Ribosomal database 9 search of the sequences revealed 137 distinct taxa of bacteria. The 10 samples from BV-positive women contained a total of 115 different taxa while the 11 BV-negative samples had 61 taxa. The median number (and range) of taxa in BV+ and BV- samples was 33 (19-46) and 12 (8-20) respectively (p=0.0003, Mann-Whitney). The types and relative proportions of taxa observed in representative subjects from each of the four groups are shown in figure 4a. The 5 samples from HIV+BV+ women contained a total of 103 different taxa while the 5 HIV-BV+ samples had 51 taxa with the median numbers in the two groups of 41 (range of 29-46) and 29 (19-36) respectively (0.1167, Mann-Whitney).

An external file that holds a picture, illustration, etc.
Object name is nihms161100f4a.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f4b.jpg

Relative abundance of taxa determined by pyrosequencing. A. The relative abundance of taxa from one representative individual in each of the four groups is shown. Taxa comprising <0.3% of the sequences are not shown for clarity. B. The relative abundance of taxa comprising ≥1% of the total averaged from all five women in the HIV+BV+ group and the HIV-BV+ group. The category “other” represents the proportion of sequences from taxa that averaged less than 1%.

Taxa found at ≥ 1% of total microbiota by pyrosequencing

We analyzed the taxa found at ≥ 1% of the total community under the a priori assumption that the most abundant taxa are the ones that contribute significantly to the functionality of the community. A total of 35 different taxa were found at ≥ 1% of the total microbiota in at least one of the 21 women (Table 1). The 10 BV+ samples contained 33 different taxa that were found at ≥1% in at least one sample while the 11 BV- samples contained 10 different taxa found at ≥1%. The median number of taxa found at ≥1% in the 10 BV+ women was 11 (range 8-18) while the median number in the 11 BV- samples was 2 (range 1-4) (p<0.0001, Mann-Whitney). The taxa most commonly found at ≥1% in BV+ samples included Prevotella, Megasphaera, Gardnerella, Coriobacterineae, Lachnospira, Sneathia and Actinomycineae (Table 1 and Fig. 5). Lactobacillus was found in all 11 BV- samples and in 5 of the BV+ samples at ≥1%. Comparison of the Lactobacillus sequences showed that this group was highly diverse. For example, 18 different clusters, or OTUs, of Lactobacillus sequences were found in the HIV+BV- women (Fig. 5) although most of the Lactobacilli were restricted to five different clusters.

An external file that holds a picture, illustration, etc.
Object name is nihms161100f5a.jpg
An external file that holds a picture, illustration, etc.
Object name is nihms161100f5b.jpg

Phylogenetic Analysis of Bacteria from Two Groups. Shown are the phylogenetic trees for the BV+HIV+ (A) and BV-HIV+ (B) groups. The total number of reads in the group is shown and taxa found at >200 reads are in bold.

Table 1

Number of Subjects with Taxa at ≥1% of Microbiota in at least one subject
Genusfound at ≥1%found at ≥10%
BV+BV-BV+BV-
Lactobacillus51111
Prevotella1028
Megasphaera10
Gardnerella9331
Coriobacterineaea911
Lachnospira827
Sneathia85
Actinomycineaeb8
Allisonella5
Micromonas5
Propionibacterineaec4121
Dialister4
Peptoniphilus3
Porphyromonas3
Comamonas22
Citrobacter21
Acinetobacter21
Mycoplasma21
Rhodobaca21
Proteiniphilum, Anaerotruncus and Corynebacterineae were found in two BV+ subjects at ≥1% Acetanaerobacterium, Achromobacter, Aerococcus, Anaerococcus, Faecalibacterium, Micrococcineae, Peptostreptococcus, Pseudomonas, Staphylococcus, Stenotrophomonas, and Xylanibacter were each found in one BV+ subject at ≥1%
Pelomonas and Sphingomonas were each found in one BV- subject at ≥1%
Atopobium and Eggerthella,
Predominantly Mobiluncus,
Predominantly Propionibacterium,
Predominantly Porynebacterium

The 5 HIV+BV+ women had a median of 14 taxa (range 10-18) found at ≥1% in contrast to the 5 HIV-BV+ samples that had a median of 10 (range 8-14) taxa (p=0.07, Mann-Whitney). Averages of taxa at ≥ 1% in the HIV+BV+ and HIV-BV+ groups are shown in Figure 4b. The 6 HIV+BV- samples had a median of 2 taxa (range 1-3) found at ≥1% while the 5 HIV-BV-samples had a median of 2 taxa (range 1-4) (p=0.93, Mann-Whitney).

Taxa found at ≥ 10% of total microbiota by pyrosequencing

A total of 10 different taxa were found at ≥ 10% of the total microbiota in at least one of the 21 women (Table 1). The 10 BV+ samples contained 9 different taxa that were found at ≥10% in at least one sample while the BV- samples contained 3 different taxa. Lactobacillus constituted at least 80% of the taxa in all BV- women (median of 96%). The median number of taxa found at ≥10% in the 10 BV+ women was 3 (range 1-4) while the median number in the 11 BV- samples was 1 (range 1-2) (p<0.0014, Mann-Whitney). The 5 HIV+BV+ samples had a median of 2 taxa (range 1-3) found at ≥10% while the 5 HIV-BV+ samples had a median of 3 (range 2-4) taxa (p=0.14, Mann-Whitney).

Taxa found only in HIV+ women by pyrosequencing

Four taxa were found in two or more of the HIV+BV+ women but none of the HIV-BV+ women. Thus, Propionibacterineae was found in all five of the HIV+BV+ women and in four of these women at >1% of the flora, but was not detected in any HIV-BV+ women (p=0.002, chi square analysis). Interestingly, Propionibacterineae was also detected in one of the HIV+BV-women but in none of the HIV-BV- women. Anaerococcus and Citrobacter were found in three of the HIV+BV+ women (p=0.04) while Acinetobacter was found in two of the HIV+BV+ women (p=0.11). No taxa were found only in HIV-BV+ women when compared with HIV+BV+ women.

Discussion

The main goal of this study was to compare the diversity of the lower genital tract microbiota in HIV+ and HIV- women and is, to our knowledge, the first study to compare microbial diversity in these groups using culture-independent methods. Analysis of the sequences obtained by pyrosequencing showed that there was a trend (p=0.07) towards higher microbial diversity in HIV+BV+ when the number of taxa found at ≥1% of the microbiota in this group was compared with HIV-BV+. Additionally, three different taxa, Propionibacterineae, Anaerococcus and Citrobacter were found only in HIV+ women and this difference was statistically significant. While the total number of taxa found in HIV+BV+ women was higher than in HIV-BV+ women, this difference was not significant. Taken together, sequence comparisons showed higher microbial diversity in HIV+BV+ women compared to HIV-BV+. Sequence analysis did not show any differences between HIV+BV- women and HIV-BV- women. In contrast to the sequence analysis, LH-PCR did not distinguish any significant differences in microbial diversity between HIV+ and HIV- women.

Several studies have found that HIV-infected women with BV have higher levels of HIV in genital secretions [7-9] and levels of specific bacteria have been associated with HIV levels [7]. It is possible that changes in the diversity of microbiota associated with HIV infection could in turn be associated with changes in HIV expression in the genital tract. Changes in microbial diversity could also affect either severity or recurrance of BV which could in turn affect pregnancy in HIV-infected women.

While we found a trend towards higher microbial diversity in HIV+ women, an underlying cause for this association is not yet known. Since HIV infection suppresses immunity, it is possible that reduced immunity could play a role in the higher diversity. Candida vaginitis is more frequent and gynecologic diseases are more severe in HIV infected women suggesting that immunity in the lower genital tract that normally restricts growth of microbes is reduced in HIV-infected women [23, 27-29]. An influence of HIV on the incidence of BV was not previously noted [23, 30].

Another goal of this study was to determine the utility of molecular techniques, LH-PCR and pyrosequencing, for analyzing female genital microbiota. Our data confirmed numerous previous studies [1, 14, 15, 31] showing that the microbiota in BV+ women is much more diverse than in BV- women with lactobacilli constituting most of the bacteria in the latter. Thus, both pyrosequencing and LH-PCR can differentiate these clinically important conditions, while only pyrosequencing was capable of showing the more subtle differences between HIV+ and HIV-. LH-PCR has the additional limitation of not providing identification of the organisms. Pyrosequencing confirmed that BV microbiota can be highly variable between women. Thus, although some bacteria (e.g. Prevotella, Megasphaera, Gardnerella and Lachnospira) were found in essentially all BV+ women at ≥1% of microbiota, there were 11 taxa found in only one of the BV+ women at these levels and 8 taxa found in 2 women (Table 1).

While LH-PCR has been used to analyze diversity of intestinal microbiota [32], there are no reports for genital microbiota. Recently, pyrosequencing was used to analyze genital microbiota in six samples pooled from HIV-negative pregnant women [33]. As in our study, high levels of lactobacilli, Prevotella, Bifidobacterium and Veillonella were found although it was not clear how many individuals were pooled in samples. The gynecologic health of the women was not defined in contrast to our study and the read lengths of the amplified products averaged 100 bp; less than half the size of the median reads in our study (249). Longer read lengths and use of 16S rRNA Variable Region 1 and 2 in our study likely results in higher accuracy of identification.

A possible limitation of our study is a potential bias introduced during PCR amplification so that the proportion of reads may not represent the actual proportions of organisms in the women. However, both methods showed high levels of diversity in individuals with BV compared to no BV providing confirmation of utility. In conclusion, this study provides evidence that HIV infection is associated with increased diversity of the lower genital tract microbiota which could have pathogenic consequences for HIV sexual transmission.

Acknowledgments

Data in this manuscript were collected by the Women's Interagency HIV Study (WIHS) Collaborative Study Group with centers (Principal Investigators) at New York City/Bronx Consortium (Kathryn Anastos); Brooklyn, NY (Howard Minkoff); Washington DC Metropolitan Consortium (Mary Young); The Connie Wofsy Study Consortium of Northern California (Ruth Greenblatt); Los Angeles County/Southern California Consortium (Alexandra Levine); Chicago Consortium (Mardge Cohen); Data Coordinating Center (Stephen Gange). The WIHS is funded by the National Institute of Allergy and Infectious Diseases with supplemental funding from the National Cancer Institute, and the National Institute on Drug Abuse(UO1-AI-35004, UO1-AI-31834, UO1-AI-34994, UO1-AI-34989, UO1-AI-34993, and UO1-AI-42590). Funding is also provided by the National Institute of Child Health and Human Development (grant UO1-HD-23632) and the National Center for Research Resources (grants MO1-RR-00071, MO1-RR-00079, MO1-RR-00083).

Department of Immunology/Microbiology, Rush University Medical Center, Chicago, Illinois
Department of Medicine, Rush University Medical Center, Chicago, Illinois
Center for Metabiome Analysis and Department of Environmental Science and Policy, George Mason University, Manassas, Virginia
Ruth M Rothstein CORE Center/Stroger Hospital of Cook County, Chicago, Illinois
Address Correspondence to: Dr. Gregory T. Spear, Department of Immunology/Microbiology, Rush University Medical Center, 1653 W. Congress Pkwy., Chicago, IL 60612, Phone: 312-942-2083, Fax:312-942-2808, ude.hsur@raepsg

Abstract

Background

This study investigated whether HIV-infection is associated with a change in diversity of genital microbiota in women.

Methods

Amplicon length heterogeneity PCR (LH-PCR) and pyrosequencing of the 16S rRNA gene were used to analyze diversity of the microbiota from HIV-positive (HIV+) and HIV-negative (HIV-) women with or without bacterial vaginosis (BV).

Results

LH-PCR analysis showed significantly more diversity in BV-positive (BV+) women than in BV-negative (BV-) women, but no significant difference between HIV+ women and HIV- women. Pyrosequencing revealed that Lactobacillus constituted a median of 96% of the bacteria in BV- women. BV+ women had a significantly higher number of taxa found at ≥ 1% of the microbiota (median of 11). Common taxa in BV were Prevotella, Megasphaera, Gardnerella, Coriobacterineae, Lachnospira, and Sneathia. There was a trend (p=0.07) toward a higher number of taxa in HIV+BV+ compared to HIV-BV+ women. Propionibacterineae, Citrobacter and Anaerococcus were found only in HIV+ women (p<0.05).

Conclusions

This study showed that both LH-PCR and pyrosequencing differentiated BV+ from BV- microbiota and that pyrosequencing indicated a trend toward increased diversity in BV+HIV+ suggesting that HIV-infection is associated with changes in diversity of genital microbiota.

Keywords: HIV infection, Microbiota, Bacteria, Pyrosequencing, Diversity, Bacterial Vaginosis, Lactobacillus
Abstract

Footnotes

the authors do not have a commercial or other association that might pose a conflict of interest.

Financial support for this study was provided by NIH Grant P01HD40539. The Chicago WIHS is funded by NIAID UO134993.

This work has not been presented in abstract form or at meetings

Footnotes

References

  • 1. Hillier SLDiagnostic microbiology of bacterial vaginosis. Am J Obstet Gynecol. 1993;169:455–9.[PubMed][Google Scholar]
  • 2. Koumans EH, Sternberg M, Bruce C, et al The prevalence of bacterial vaginosis in the United States, 2001-2004; associations with symptoms, sexual behaviors, and reproductive health. Sex Transm Dis. 2007;34:864–9.[PubMed][Google Scholar]
  • 3. Cohen CR, Duerr A, Pruithithada N, et al Bacterial vaginosis and HIV seroprevalence among female commercial sex workers in Chiang Mai, Thailand. Aids. 1995;9:1093–7.[PubMed][Google Scholar]
  • 4. Sewankambo N, Gray RH, Wawer MJ, et al HIV-1 infection associated with abnormal vaginal flora morphology and bacterial vaginosis [see comments] [published erratum appears in Lancet 1997 Oct 4;350(9083):1036] Lancet. 1997;350:546–50.[PubMed][Google Scholar]
  • 5. Taha TE, Hoover DR, Dallabetta GA, et al Bacterial vaginosis and disturbances of vaginal flora: association with increased acquisition of HIV. Aids. 1998;12:1699–706.[PubMed][Google Scholar]
  • 6. Taha TE, Gray RH, Kumwenda NI, et al HIV infection and disturbances of vaginal flora during pregnancy. J Acquir Immune Defic Syndr Hum Retrovirol. 1999;20:52–9.[PubMed][Google Scholar]
  • 7. Sha BE, Zariffard MR, Wang QJ, et al Female genital-tract HIV load correlates inversely with Lactobacillus species but positively with bacterial vaginosis and Mycoplasma hominis. J Infect Dis. 2005;191:25–32.[PubMed][Google Scholar]
  • 8. Coleman JS, Hitti J, Bukusi EA, et al Infectious correlates of HIV-1 shedding in the female upper and lower genital tracts. Aids. 2007;21:755–9.[PubMed][Google Scholar]
  • 9. Cu-Uvin S, Hogan JW, Caliendo AM, Harwell J, Mayer KH, Carpenter CCAssociation between bacterial vaginosis and expression of human immunodeficiency virus type 1 RNA in the female genital tract. Clin Infect Dis. 2001;33:894–6.[PubMed][Google Scholar]
  • 10. Martin HL, Richardson BA, Nyange PM, et al Vaginal lactobacilli, microbial flora, and risk of human immunodeficiency virus type 1 and sexually transmitted disease acquisition. J Infect Dis. 1999;180:1863–8.[PubMed][Google Scholar]
  • 11. Cherpes TL, Meyn LA, Krohn MA, Lurie JG, Hillier SLAssociation between acquisition of herpes simplex virus type 2 in women and bacterial vaginosis. Clin Infect Dis. 2003;37:319–25.[PubMed][Google Scholar]
  • 12. Wiesenfeld HC, Hillier SL, Krohn MA, Landers DV, Sweet RLBacterial vaginosis is a strong predictor of Neisseria gonorrhoeae and Chlamydia trachomatis infection. Clin Infect Dis. 2003;36:663–8.[PubMed][Google Scholar]
  • 13. Kaul R, Nagelkerke NJ, Kimani J, et al Prevalent herpes simplex virus type 2 infection is associated with altered vaginal flora and an increased susceptibility to multiple sexually transmitted infections. J Infect Dis. 2007;196:1692–7.[PubMed][Google Scholar]
  • 14. Fredricks DN, Fiedler TL, Marrazzo JMMolecular identification of bacteria associated with bacterial vaginosis. N Engl J Med. 2005;353:1899–911.[PubMed][Google Scholar]
  • 15. Hyman RW, Fukushima M, Diamond L, Kumm J, Giudice LC, Davis RWMicrobes on the human vaginal epithelium. Proc Natl Acad Sci U S A. 2005;102:7952–7.[Google Scholar]
  • 16. Verhelst R, Verstraelen H, Claeys G, et al Cloning of 16S rRNA genes amplified from normal and disturbed vaginal microflora suggests a strong association between Atopobium vaginae, Gardnerella vaginalis and bacterial vaginosis. BMC Microbiol. 2004;4:16.[Google Scholar]
  • 17. Ferris MJ, Masztal A, Aldridge KE, Fortenberry JD, Fidel PL, Jr, Martin DHAssociation of Atopobium vaginae, a recently described metronidazole resistant anaerobe, with bacterial vaginosis. BMC Infect Dis. 2004;4:5.[Google Scholar]
  • 18. Ritchie NJ, Schutter ME, Dick RP, Myrold DDUse of length heterogeneity PCR and fatty acid methyl ester profiles to characterize microbial communities in soil. Appl Environ Microbiol. 2000;66:1668–75.[Google Scholar]
  • 19. Bernhard AE, Colbert D, McManus J, Field KGMicrobial community dynamics based on 16S rRNA gene profiles in a Pacific Northwest estuary and its tributaries. FEMS Microbiol Ecol. 2005;52:115–28.[PubMed][Google Scholar]
  • 20. Rogers GB, Hart CA, Mason JR, Hughes M, Walshaw MJ, Bruce KDBacterial diversity in cases of lung infection in cystic fibrosis patients: 16S ribosomal DNA (rDNA) length heterogeneity PCR and 16S rDNA terminal restriction fragment length polymorphism profiling. J Clin Microbiol. 2003;41:3548–58.[Google Scholar]
  • 21. Margulies M, Egholm M, Altman WE, et al Genome sequencing in microfabricated high-density picolitre reactors. Nature. 2005 doi: 10.1038/nature03959.] [[Google Scholar]
  • 22. Sogin ML, Morrison HG, Huber JA, et al Microbial diversity in the deep sea and the underexplored “rare biosphere” Proc Natl Acad Sci U S A. 2006;103:12115–20.[Google Scholar]
  • 23. Greenblatt RM, Bacchetti P, Barkan S, et al Lower genital tract infections among HIV-infected and high-risk uninfected women: findings of the Women's Interagency HIV Study (WIHS) Sex Transm Dis. 1999;26:143–51.[PubMed][Google Scholar]
  • 24. Lane DJ. 16S/23S rRNA Sequencing. In: Goodfellow ESaM., editor. Nucleic Acid Techniques in Bacterial Systematics. West Sussex, England: John Wiley &amp; Sons Ltd; 1991. pp. 115–175. [PubMed]
  • 25. Cole JR, Chai B, Farris RJ, et al The Ribosomal Database Project (RDP-II): sequences and tools for high-throughput rRNA analysis. Nucleic Acids Research. 2005;33:294–296.[Google Scholar]
  • 26. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DGThe CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 1997;25:4876–82.[Google Scholar]
  • 27. Sobel JDVulvovaginal candidiasis: a comparison of HIV-positive and -negative women. Int J STD AIDS. 2002;13:358–62.[PubMed][Google Scholar]
  • 28. Minkoff HL, Eisenberger-Matityahu D, Feldman J, Burk R, Clarke LPrevalence and incidence of gynecologic disorders among women infected with human immunodeficiency virus. Am J Obstet Gynecol. 1999;180:824–36.[PubMed][Google Scholar]
  • 29. Ohmit SE, Sobel JD, Schuman P, et al Longitudinal study of mucosal Candida species colonization and candidiasis among human immunodeficiency virus (HIV)-seropositive and at-risk HIV-seronegative women. J Infect Dis. 2003;188:118–27.[PubMed][Google Scholar]
  • 30. Watts DH, Springer G, Minkoff H, et al The occurrence of vaginal infections among HIV-infected and high-risk HIV-uninfected women: longitudinal findings of the women's interagency HIV study. J Acquir Immune Defic Syndr. 2006;43:161–8.[PubMed][Google Scholar]
  • 31. Fredricks DN, Fiedler TL, Thomas KK, Oakley BB, Marrazzo JMTargeted PCR for detection of vaginal bacteria associated with bacterial vaginosis. J Clin Microbiol. 2007;45:3270–6.[Google Scholar]
  • 32. Komanduri S, Gillevet PM, Sikaroodi M, Mutlu E, Keshavarzian ADysbiosis in pouchitis: evidence of unique microfloral patterns in pouch inflammation. Clin Gastroenterol Hepatol. 2007;5:352–60.[PubMed][Google Scholar]
  • 33. Sundquist A, Bigdeli S, Jalili R, et al Bacterial flora-typing with targeted, chip-based Pyrosequencing. BMC Microbiol. 2007;7:108.[Google Scholar]
Collaboration tool especially designed for Life Science professionals.Drag-and-drop any entity to your messages.