A species-specific PCR for Lactobacillus iners demonstrates a relative specificity of this species for vaginal colonization

Abstract

The prevalence of Lactobacillus iners is currently difficult to determine since it cannot be propagated on conventional Lactobacillus-selective media. Based on comparative analysis of 16S rRNA gene sequences a simple L. iners-specific PCR method was developed and then applied to screen vaginal, fecal, and saliva samples for this species. Twenty-one (53%) of 40 vaginal specimens were positive, whereas fecal (n=9) and saliva (n=16) samples were PCR-negative for L. iners. These findings support previous molecular and culture-based ecology studies that have indicated L. iners may be relatively more prevalent in the microflora of the human vagina than that of the oral or gastrointestinal cavities.

• Lactobacillus iners
• species-specific PCR
• vaginal colonization

Introduction

Lactobacilli are indigenous to the oral, gastrointestinal, and vaginal cavities and certain strains have been shown to be effective when introduced either orally [1–3] or intravaginally [4–6]. The study of the ecology of Lactobacillus species is important for our greater understanding of the role of these organisms in enhancing health.

Lactobacillus iners was initially isolated from human urine, vaginal discharges, and endometrial and cervical specimens [7]. L. iners differs from other Lactobacillus species in that it cannot be grown on MRS or Rogosa agar [7]. The identification of Lactobacillus species by classical phenotypic methods is difficult, time-consuming, and frequently yields ambiguous results [8], and for L. iners this is complicated by the absence of suitable selective media. An alternative approach is PCR-based identification using species-specific primers designed to variable regions of the 16S rRNA genes [8–10]. To improve our understanding of the ecology of L. iners a simple PCR-based detection methodology resembling the available techniques for other Lactobacillus species is required. Here we report the development, evaluation, and application of just such a technique. Our findings indicate that L. iners is relatively more frequently present in samples from the vagina than from the intestine or oral cavity and support our hypothesis that it may have a significant ecological role within the microflora of the human vagina.

Materials and methods

Bacterial strains

The L. iners type strain CCUG 28746^T was obtained from the culture collection, University of Göteborg, Sweden. L. iners strains BVS011, FB077-03, FB088-01, FB094-05b, FB123-CNA-4, PB2003/087-T2-1, and PB2003/195-T1-1 were obtained from Professor Mario Vaneechoutte (Department of Clinical Chemistry, Microbiology & Immunology, Ghent University Hospital, Belgium). Other bacteria used in this study (Table I) were freshly obtained vaginal isolates, clinical isolates or strains from the culture collection of Prof. J.R. Tagg (Microbiology and Immunology Department, University of Otago, New Zealand). Lactobacilli were propagated on either Lactobacillus-Rogosa agar (Difco, USA) or blood agar (Columbia agar base, Difco) with 5% v/v human blood) with incubation under anaerobic conditions (85% N₂, 10% H₂, and 5% CO₂) at 37°C for 48–72 h.

Table I.  Bacterial strains used in this study.

Strain specification	Strain name	L. iners-specific PCR
Lactobacillus iners	CCUG 28746^T	+
Lactobacillus iners	BVS011	+
Lactobacillus iners	FB077-03	+
Lactobacillus iners	FB088-01	+
Lactobacillus iners	FB094-05b	+
Lactobacillus iners	FB123-CAN-4	+
Lactobacillus iners	PB2003/087-T2-1	+
Lactobacillus iners	PB2003/195-T1-1	+
Lactobacillus brevis	LB4	−
Lactobacillus delbrueckii	LB31	−
Lactobacillus crispatus	ATCC 33199 ^T	−
Lactobacillus crispatus	33820	−
Lactobacillus casei	ATCC 393^T	−
Lactobacillus fermentum	SAV1	−
Lactobacillus fermentum	SAV 3	−
Lactobacillus fermentum	SAV 5	−
Lactobacillus fermentum	ATCC 11739^T	−
Lactobacillus fermentum	ATCC 23271	−
Lactobacillus gasseri	ATCC 33323^T	−
Lactobacillus gasseri	33323	−
Lactobacillus jensenii	ATCC 25258^T	−
Lactobacillus johnsonii	DSM 20553	−
Lactobacillus johnsonii	100-100	−
Lactobacillus johnsonii	NCK 088	−
Lactobacillus plantarum	ATCC 8014	−
Lactobacillus plantarum	ATCC 10012	−
Lactobacillus plantarum	ATCC 14917^T	−
Lactobacillus paracasei	LTH 2570	−
Lactobacillus reuteri	ATCC 23272^T	−
Lactobacillus reuteri	RC-14	−
Lactobacillus reuteri	100-23	−
Lactobacillus reuteri	DSM 20016^T	−
Lactobacillus reuteri	TMW 1.106	−
Lactobacillus rhamnosus	ATCC 7469^T	−
Lactobacillus rhamnosus	GR-1	−
Lactobacillus rhamnosus	GG	−
Lactobacillus salivarius	SAV 6	−
Lactobacillus sakei	23K	−
Enterococcus faecalis	ATCC 19433^T	−
Escherichia coli	ATCC 25922	−
Lactococcus lactis	I6	−
Lactococcus lactis	1404	−
Lactococcus lactis	V1	−
Lactococcus lactis	V5	−
Lactococcus lactis	S1	−
Lactococcus lactis	S5	−
Lactococcus lactis	V63	−
Micrococcus luteus	I1	−
Proteus vulgatus	ATCC 13315	−
Staphylococcus aureus	ATCC 25923	−
Staphylococcus saprophyticus	ATCC 15305^T	−
Streptococcus agalactiae	P3	−
Streptococcus agalactiae	211B	−
Streptococcus anginosus	I3	−
Streptococcus anginosus	SAV4	−
Streptococcus equisimilis	I9	−
Streptococcus pyogenes	I2	−
Streptococcus pyogenes	I5	−
Streptococcus pyogenes	I7	−
Streptococcus pyogenes	I8	−
Streptococcus pyogenes	P2	−
Streptococcus salivarius	SAV 2	−
Streptococcus uberis	I4	−

Vaginal, saliva, and fecal samples

The Otago Ethics Committee approved the study and all participants provided written consent. The vaginal samples (n = 40) were obtained from healthy women attending prenatal clinics. No Gram-stain grading or testing for bacterial vaginosis was performed. High vaginal swabs (TRANSWAB®, England) were obtained from women by a gynecologist, and kept at 4°C until processing within 30 min. The contents of each swab were dispersed in 3 ml brain heart infusion (BHI, Difco) and the resulting suspension was used to obtain total DNA preparations.

For saliva samples, the subjects were asked to provide 1 ml of saliva by spitting into a sterile bottle. The samples were immediately centrifuged at 15 300 g for 5 min, and the pellets (resuspended in 100 µl H₂O) were used to obtain total DNA. InstaGene™ matrix (BioRad Laboratories) was used to obtain total representative DNA from the vaginal and oral specimens, according to the manufacturer's instructions.

Fecal samples were obtained from healthy individuals and kept at 4°C until processing. For DNA extractions, 10 mg of fecal material was washed three times by resuspending in 1.0 ml of phosphate-buffered saline and centrifuging at 14 000 g. The washed pellets were then resuspended in 450 µl of extraction buffer (100 mM Tris-HCl, 40 mM EDTA, pH 9.0) and 50 µl of 10% SDS followed by the addition of 300 mg of glass beads (diameter, 0.1 mm) and 500 µl of buffer-saturated phenol, before vigorous vortex mixing for 30 s using a FastPrep FP 120 (Bio 101, Vista, CA, USA) at a power level of 5.0. After centrifugation at 14 000 g for 5 min, the supernatants were collected. Subsequently, phenol-chloroform extractions were performed, and DNA was obtained by isopropanol precipitation. Finally, the DNA was suspended in 1 ml of TE buffer.

DNA extraction from pure cultures

Extraction from pure cultures was performed using the Qiagen DNAeasy tissue kit (Qiagen, Valencia, CA, USA) according to the manufacturer's instructions, using logarithmic growth phase cells in Todd Hewitt broth (THB, Difco).

Development of species-specific primers for L. iners

Twenty-seven 16S rRNA gene sequences of Lactobacillus species (15 L. iners sequences plus 12 from other lactobacilli identified by BLAST to have sequences that were most closely related to the L. iners consensus sequences) were aligned to facilitate the design of L. iners-specific PCR primers. Alignments and the sequence accession numbers are given in Table II. Alignment was achieved by using the Clustal W alignment method DNAMAN (Lynnon Biosoft, Vaudreuil, Canada) with low gap penalties (0.1 gap open, 0.05 gap extension, 30% delay divergent). The 15 L. iners 16S rRNA sequences were all of those that were available in the public databases. Two regions were identified as specific for L. iners (Table II). The PCR primer pair InersFw (5′-GTC TGC CTT GAA GAT CGG-3′) and InersRev (5′-ACA GTT GAT AGG CAT CAT C-3′) were designed to these regions and synthesized (Invitrogen New Zealand Ltd, Auckland, New Zealand). The chosen primers were based upon positions 42–101 and 162–221 of the sequence AY283264.

Table II.  Alignment of 16S rRNA sequences.

Sequence	Source
5′-G TCT GCC TTG AAG ATC GG-3′	AY283264 (position 42-101); InersFw
GAG TCT GCC TTG AAG ATC GGA GTG	Consensus of 15 L. iners sequences*
--- CT- --- -A- -T- -AT TTG ---	AF243144; L. gasseri strain BLB1b
--- CT- --- -A- -T- -AT TTG ---	AF243151; L. gasseri strain KC36a
--- CTA --- -A- -T- -AT TTG ---	AF243165; L. gasseri strain KC5a
--- CT- --- -A- -T- –T TTG ---	AF371478; uncultured bacterium
--- CT- --- -A- -T- --T TTG ---	AF371479; uncultured bacterium
--- CT- --- -A- -T- -AT TTG ---	AF519171; L. gasseri strain ATCC 33323
--- CT- --T -A- -T- -AT TTG ---	AY094066; Lactobacillus sp. IDSAc
--- CT- --- -A- -T- -AT TTG ---	AY730721; L. gasseri strain 130704-6172
--- CT- --- -A- -T- –T TT- ---	AY862434; Lactobacillus sp. ID920
--- CT- --- -A- -T- -AT TTG ---	M58820; L. gasseri (library: DSM 20243)
--- CT- --- -A- -T- –T TT- ---	M99704; L. acidophilus strain BMF 6Lb6
--- CT- --- -A- -T- -AT TTG ---	AE017198; L. johnsonii strain NCC 533
Sequence	Source
3′-CTA CTA CGG ATA GTT GAC A-5′	InersRev (position 162-221 of AY283264)
ACA GAT GAT GCC TAT CAA CTG TT	Consensus of 15 L. iners sequences*
— CTA –C –A -G- -T- GA- –	Consensus of 12 non-L. iners sequences†

*Consensus of the L. iners sequences AY283264, AY283265, AY283266, AY283267, AY283268, AY23269, AY283270, AY283271, AY283272, AY283273, AY283274, AY283275, AY526083, AY995250 and Y16329.

†Consensus of non-L. iners sequences AF243144, AF243151, AF243165, AF371478, AF371479, AF519171, AY094066, AY730721, AY862434, M58820, M99704, AE017198.

PCR analysis

Species-specific PCR reactions for L. iners comprised 3 µl DNA, 1 U Hot start Taq DNA polymerase (Eppendorf AG, Hamburg, Germany), 5 µl of 10× buffer (Eppendorf), 1 µl of PCR nucleotide mix (Eppendorf), and 1 µl (4 ng) of each primer (Invitrogen). The volume was made up to 50 µl with sterile MQ water. After a 3 min initial denaturation at 95°C, 35 cycles of denaturation at 95°C for 1 min, annealing at 55°C for 1 min, and elongation at 65°C for 1 min were carried out. The primers and conditions for the PCR reactions for eubacterial, Lactobacillus, and L. crispatus rRNA gene sequences have been described previously [10–12] and these were used in this study to confirm the presence of DNA and absence of PCR inhibitors in the samples. Table III lists all the primers used in this study.

Table III.  Primers used in this study.

Primer	Sequence	Reference
InersFw	5′-GTC TGC CTT GAA GAT CGG-3′	This study
InersRev	5′-ACA GTT GAT AGG CAT CAT C-3′	This study
CbsA2F	5′-GTA CCA AGC CAA AGC AAG AC-3′	[12]
CbsA2R	5′-GTT TGA AGC CTT TAC GTA AGT C-3′	[12]
Lac1	5′-AGC AGT AGG GAA TCT TCC A-3′	[23]
Lac2	5′-GC-Clamp-ATT YCA CCG CTA CAC ATG-3′	[23]
16S Fw	5′-GGA CTA HAG GGT ATC TAA T -3′	[11]
16S Re	5′-AGA GTT TGA TCM TGG-3′	[11]

Results

Sensitivity and specificity of the L. iners selective primers

The specificity of the primers for L. iners was evaluated by carrying out PCR using genomic DNA from the various strains of different genera and species listed in Table I. No cross-reactivity was detected with DNA from any other species, including the most closely related lactobacilli (Table I). To determine the sensitivity of the PCR assay, an 18 h THB culture of L. iners was serially diluted in sterile saline and the numbers of colony-forming units (cfu) were assessed by plating on blood agar. Known numbers of L. iners were then added to clinical samples that had been previously found to be PCR-negative for L. iners before DNA extraction. The PCR lower limit of detection was estimated to be 50 cfu, as addition of this number of cells (10 µl of diluted culture) to clinical specimens (50 µl) led to their conversion from PCR-negative to PCR-positive.

PCR analysis of clinical samples

L. iners was detected in 21 (53%) of the 40 tested vaginal specimens, but not in any of the fecal or saliva specimens. Figure 1 shows four of nine representative vaginal specimens exhibiting PCR-positive reactions using the L. iners-specific primers. Amplicons from selected samples were DNA-sequenced and this confirmed the identity of the products to be L. iners. All the samples were PCR-positive for eubacterial DNA. All the vaginal and saliva samples and five of the nine fecal samples were PCR-positive for Lactobacillus DNA. Fourteen of the 40 vaginal samples, one saliva sample, but none of the fecal specimens contained detectable L. crispatus DNA.

Figure 1.  Lactobacillus iners-specific PCR assay of vaginal specimens. Lanes 1–9, PCR products of vaginal specimen DNA amplified from subjects 1–9; lane 10, negative control; lane 11, positive control using L. iners DNA; lane M, GeneRuler™ DNA ladder.

Discussion

The acquisition of PCR products for DNA from eubacteria, Lactobacillus, and L. crispatus established that the samples contained amplifiable DNA and also that detection frequencies for other Lactobacillus species were in the normal range for these specimens [7], [13–17]. The relatively frequent detection of L. iners in the vaginal specimens from healthy women indicates that this species could be a significant inhabitant of the normal microbiota of this site. All the tested saliva and fecal samples were PCR-negative for L. iners. Several previous studies [7], [13–17] have also indicated that this species inhabits the vagina, whereas molecular ecological studies of the gastrointestinal [18] and oral ecosystems [19–21] have failed to detected L. iners. The rectum, which is considered to be a primary source of lactobacilli for vaginal microflora maintenance, seems relatively seldom to harbor L. iners (4%) when compared with other Lactobacillus species such as L. crispatus (43%), L. jensenii (23%), and L. gasseri (18%) [22]. On the basis of our present findings using a sensitive and specific PCR-detection method, and previous ecological studies [18–21], it seems that L. iners may indeed have relatively specific affinity for colonization of the human vagina. The primers InersFw and InersRev can be used in many different settings and in different nucleic acid-based ecological applications. The clinical relevance of the changes in vaginal microbiota as it shifts from healthy to dysbiosis (bacterial vaginosis) is not fully understood. L. iners could be associated with early disease, relapse or recovery as either a final protective species or a disease facilitator.

Acknowledgements

Mr Alqumber is supported by a scholarship provided by the Saudi Arabian Ministry of Higher Education.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.