2018 年 6 巻 2 号 p. 68-72
Aim: Vaginal environment is an important factor in predicting threatened abortion and premature rupture of membranes (PROM) in early pregnancy. Vaginal human β-defensin-2 (HBD-2), a main antimicrobial peptide associated with innate immunity, plays multiple roles to protect the lower genital tract from microbes. This study aimed to assess the usefulness of vaginal HBD-2 as a biochemical marker for monitoring the vaginal environment in pregnant women.
Methods: Twenty-eight pregnant women were enrolled in this study. Vaginal secretion samples were collected to determine Nugent scores, and vaginal washing fluid (4 ml saline) was used to measure HBD-2 levels by ELISA.
Results: In the first trimester, vaginal HBD-2 levels were significantly lower in patients with bacterial vaginosis (BV) (42.1±19.8 pg/ml) compared to those without BV (350.1±68.7 pg/ml, P<0.05), and in patients with PROM (57.2±31.9 pg/ml) compared to those without PROM (375.9±69.4 pg/ml, P<0.05).
Conclusions: Low vaginal HBD-2 levels during early pregnancy suggest a poor vaginal environment and the possibility of developing PROM.
Bacterial vaginosis (BV) in early pregnancy is a known risk factor for preterm delivery.1,2) Antibiotic prophylaxis reportedly does not improve maternal or perinatal outcomes.3) Assessment of the vaginal condition in early pregnancy is important for predicting perinatal outcome, but currently, there is no biochemical marker that precisely reflects the vaginal environment. The Nugent score, which is calculated based on microscopic quantification of Lactobacillus,4) is used as a standard measure to evaluate BV, although its measurement reproducibility is low.
The innate immune system plays a role in protecting the female reproductive tract from bacterial, fungal, and viral infections. Antimicrobial peptides are small cationic peptides, many of which have positively charged residues and amphipathic structures, and play multiple roles in the innate immune system.5,6) Although the precise mechanism of antimicrobial action remains to be elucidated, antimicrobial peptides are known to create pores on target membranes to exert antimicrobial activity.
Humans and other mammals have antimicrobial peptides belonging to two main families: defensins and cathelicidins.7) Defensins are divided into three main subfamilies (alpha, beta, and theta), with varying lengths of peptide segments between cysteine residues connected by disulfide bonds.7) Human beta-defensins (HBDs) activate the adaptive immune system through chemotaxis of T lymphocytes and dendritic cells to the infection site.8) There are three types of HBDs (HBD-1, -2, and -3) expressed in the vaginal epithelium, placenta, amniotic epithelium, and decidua.11) HBD-1 is constitutively expressed in epithelial cells, whereas expressions of HBD-2 and HBD-3 are induced by stimulation from infection.9,10) HBD-2 attracts memory T-cells and immature dendritic cells, while HBD-3 is chemotactic for macrophages. (Jung S, Mysliwy J. Antimicrob Agents Chemother. 2011; 55(3): 954–60) High levels of HBD-2 in amniotic fluid have been reported in cases of preterm delivery.12) In addition, a significant decrease in the level of HBD-2 was observed in patients with BV during pregnancy.13) However, few reports have examined vaginal HBD-2 levels in cases of preterm delivery with premature rupture of membranes (PROM).
The composition of the vaginal flora is known to change in pregnancy so as to prevent uterine infection.14) In this study, we examined changes in HBD-2 levels in vaginal secretions of pregnant women as a biochemical marker for evaluating the vaginal environment involved in preterm delivery and PROM.
Twenty-eight pregnant women who visited NTT West Matsuyama Hospital were enrolled in this study in 2015. Demographic and clinical data are summarized in Table 1. Written informed consent was obtained from all participants prior to enrollment. The study protocol was approved by the ethics committee of Ehime University Hospital.
| Characteristics | Value |
|---|---|
| Age (years), mean±SD (range) | 30.1±4.5 (22–39) |
| 20–29 | 11 (39%) |
| 30–39 | 17 (61%) |
| Gravidity, median (range) | 1 (0–4) |
| 0 | 9 (32%) |
| ≥1 | 19 (68%) |
| Parity, median (range) | 1 (0–2) |
| 0 | 13 (46%) |
| ≥1 | 15 (54%) |
| Nugent score, first trimester | |
| 0–3 (normal) | 24 (85.7%) |
| 4–6 (intermediate) | 1 (3.6%) |
| 7–10 (BV) | 3 (10.7%) |
Vaginal secretion samples were collected using cotton swabs for Nugent score calculation in the first trimester (8–11 weeks of gestation) and third trimester (28–31 weeks of gestation). Vaginal washing fluid was (4 ml saline) was used for microbe analysis. A portion of each sample was centrifuged at 3,000 rpm for 10 min at 4ºC, and the supernatant was stored at −80ºC until the time of HBD-2 assay.
Nugent scores and microbiological analysisGram-stained vaginal smears of vaginal secretions were examined at a magnification of ×1,000 with an oil immersion lens. The presence of large Gram-positive rods (Lactobacillus spp. morphotypes), small Gram-variable rods (Gardnerella morphotypes), and curved Gram-negative rods (Mobiluncus spp. morphotypes) was quantified according to the Nugent criteria. Composite scores were calculated: scores 0–3 were considered normal, and scores 4–6 and 7–10 indicated an intermediate state and BV, respectively.
The presence of Ureaplasma urealyticum, Mycoplasma hominis, Mycoplasma genitalium, and Chlamydia trachomatis was analyzed by polymerase chain reaction (PCR). Oligonucleotide primers for PCR reaction were designed; the forward and reverse primers are listed in Table 2. PCR was carried out using Emerald Amp PCR Master Mix (TaKaRa Bio, Shiga, Japan) in a thermal cycler (Chromo4 System: Bio-Rad Laboratories, Hercules, CA, USA). The reaction profile was as follows: initial denaturation at 94°C for 120 s, followed by 35 cycles at 98°C for 10 s, 54°C for 30 s, and 72°C for 45 s, and final extension at 72°C for 420 s.
| Primer | Sequence (5’–3’) |
|---|---|
| Ureaplasma urealyticum U-5F | CAATCTGCTCGTGAAGTATTAC |
| Ureaplasma urealyticum U-4R | ACGACGTCCATAAGCAACT |
| Mycoplasma gentalium MgPaW1F | AAGTGGAGCGATCATTACTAAC |
| Mycoplasma gentalium MgPaR1R | CCGTTGTTATCATACCTTCTGA |
| Mycoplasma hominis F-P | GACACTAGCAAACTAGAGTTAG |
| Mycoplasma hominis R | CACCATCTGTCACTCTGTTAACCTC |
| First PCR Chlamydia trachomatis | TTGCGA TCCTTGCACCACTT |
| Chlamydia trachomatis | GCTCGAGACCATTTAACTCC |
| Nested PCR Chlamydia trachomatis | GGTGACTTTGTTTTCGACCG |
| Chlamydia trachomatis | CTCCAATGTAGGGAGTGAAC |
The presence of Candida species in the vaginal flora was determined by microbial cultivation and microscopic analysis.
HBD-2 ELISAHBD-2 levels in vaginal fluid were determined by ELISA (HBD2 ELISA kit: Phoenix Pharmaceuticals, Inc., Burlingame, CA, USA) according to the manufacturer’s instructions. Light absorbance was measured at 450 nm with a microplate reader (iMark: Bio-Rad Laboratories). All samples were measured in triplicate.
Statistical analysisData were expressed as mean±SE. The parameters were analyzed by one-way ANOVA with the Bonferroni post hoc test. Significant differences in reverse transcriptase-PCR results were detected using the Mann–Whitney U test. P values <0.05 were considered statistically significant.
Participant characteristics are summarized in Table 1. In the first trimester, 85.7% of participants had normal Nugent scores (normal group), and 10.7% had BV (BV group). Vaginal HBD-2 levels were significantly higher in the normal group (350.1±68.7 pg/ml) than in the BV group (42.1±19.8 pg/ml, P<0.05; Table 3). Vaginal HBD-2 levels were significantly increased in the presence of Candida (P<0.05), but not affected by the presence of U. urealyticum (Table 4).
| Nugent score | n | Age (years, mean±SD) | HBD-2 (pg/ml, mean±SE) |
|---|---|---|---|
| First trimester | |||
| 0–3 (normal) | 24 | 30.4±4.4 | 350.1±68.7* |
| 4–6 (intermediate) | 1 | 22 | 720.9 |
| 7–10 (BV) | 3 | 30.7±4.0 | 42.1±19.8* |
| Third trimester | |||
| 0–3 (normal) | 28 | 30.1±4.5 | 128.1±22.3 |
| 4–6 (intermediate) | 0 | ||
| 7–10 (BV) | 0 |
| Group | n | HBD-2 (pg/ml, mean±SE) |
|---|---|---|
| Candida positive | 8 | 598.2±143.6* |
| negative | 20 | 223.2±52.8* |
| Mycoplasma hominis positive | 0 | |
| negative | 28 | 330.4±63.2 |
| Mycoplasma genitalium positive | 0 | |
| negative | 28 | 330.4±63.2 |
| Ureaplasma urealyticum positive | 15 | 406.7±105.4 |
| negative | 13 | 242.2±56.4 |
| Chlamydia trachomatis positive | 1 | 838.7 |
| negative | 27 | 311.5±62.6 |
None of the participants had preterm birth, although five participants required hospitalization and treatment with intravenous ritodrine hydrochloride for 14–60 days to avoid preterm delivery. HBD-2 levels in vaginal secretion did not differ between participants who received treatment and those who did not (249.0±125.3 and 348.0±72.6 pg/ml, respectively).
Four participants developed PROM (PROM group); three were in the normal group, and one was in the BV group. Vaginal HBD-2 levels in the first trimester were significantly lower in the PROM group (57.2±31.9 pg/ml) compared to the non-PROM group (375.9±69.4 pg/ml, P<0.05; Figure 1). In the third trimester, no significant difference in HBD-2 levels was observed between the two groups (PROM, 78.3±33.8 pg/ml; non-PROM, 141.0±26.4 pg/ml).
Levels of human β-defensin-2 (HBD-2) in vaginal fluid from women with and without premature rupture of membranes (PROM).
HBD-2 levels in the first trimester were significantly decreased in patients with PROM (PROM group: 57.2±31.9 pg/ml; non-PROM group: 375.9±69.4 pg/ml). In the third trimester, no significant difference was observed between the two groups (PROM group: 78.3±33.8 pg/ml; non-PROM group: 141.0±26.4 pg/ml). Data are presented as mean±SE. * P<0.05.
In pregnant women, BV confers high risks of preterm delivery and PROM. A healthy vaginal flora is important to avoid these conditions and to have a good perinatal outcome.1,2) In previous studies, antibiotics treatment for BV during pregnancy was not associated with perinatal outcomes.3) A useful biochemical marker that reflects changes in vaginal flora in early pregnancy is needed in order to predict preterm delivery and PROM and thereby improve perinatal outcomes.
Beta (β)-defensins, a subfamily of antimicrobial peptides, are produced by epithelial cells of the amniotic membrane, chorionic membrane, and vaginal wall. Vaginal β-defensins form the first line of defense against vaginal infection, and play multiple roles in the innate immune system, including immunocyte chemotaxis, cytokine generation, and wound healing.4,6) In 2007, Soto et al.12) reported that amniotic HBD-2 concentrations were increased in patients with intra-amniotic inflammation. Gardnerella vaginalis and Candida albicans have also been reported to induce up-regulation of HBD-2 expression in chorioamniotic membranes.15,16) While colonic/gingival, corneal, and respiratory HBD-2 expressions are induced by stimulation from infections with various microbes and inflammation,9,10,17,18,19) some studies have shown that vaginal HBD-2 expression is not induced in BV patients.13,20,21) In the present study, HBD-2 levels were clearly decreased in patients with BV in the first trimester. Moreover, vaginal HBD-2 levels were not affected by the presence of M. hominis, M. genitalium, U. urealyticum, or C. trachomatis. On the other hand, vaginal candidiasis significantly increased vaginal HBD-2 levels. The difference between Candida and other microbes is the ability of the former to coexist with Lactobacillus. In the present study, seven of eight patients with Candida infection had normal Nugent scores. These results suggest that vaginal HBD-2 levels might be regulated by the presence of Lactobacillus. In other words, the presence of not only microbes but also Lactobacillus might be important in maintaining the vaginal environment.
Lactobacillus levels are known to be increased in the vaginal flora of healthy pregnant women.14) On the other hand, levels of Atopobium, Prevotella, Sneathia, and Gardnerella, which are involved in the onset of BV, are decreased in pregnant women. The composition of the vaginal flora changes during pregnancy so as to prevent reproductive tract infection. Interestingly, in the present study, vaginal HBD-2 levels were decreased during late pregnancy. Vaginal HBD-2 is suspected to play an important role against the onset of BV early in pregnancy; however, many other factors may be involved in the defense mechanism against reproductive tract infection late in pregnancy. Unique immunological conditions exist during pregnancy. Monitoring the vaginal flora and levels of antibacterial peptides is thus important to assess antibacterial activity during pregnancy.
In the present study, all Nugent scores in the third trimester were within the normal range. Although five participants required hospitalization and treatment with intravenous ritodrine hydrochloride to avoid preterm delivery, no preterm delivery or PROM was observed. Meanwhile, vaginal HBD-2 levels in the first trimester were significantly lower in participants who later developed PROM. The management of patients with low HBD-2 levels during the first trimester might help prevent preterm delivery and PROM. Yet, the effect of interventions on perinatal outcome still remains clear. A larger sample size and evaluation of severe cases of threatened premature delivery will be necessary to clarify the implication of low HBD-2 levels in early pregnancy.
In conclusion, low vaginal HBD-2 levels during early pregnancy suggest the presence of a poor vaginal environment, as well as the possibility of developing PROM. Further investigation will be needed to evaluate the effect of HBD-2 on perinatal outcomes.
We thank all staff members who were involved in this study. We especially thank Sawa Yamamoto for her assistance in molecular biological analysis, and Hitoshi Miyamoto and Shinji Soneda for their assistance in bacteriological analysis.
None.
