SEARCH

SEARCH BY CITATION

Keywords:

  • Choriodecidual inflammatory response;
  • preterm birth;
  • Ureaplasma urealyticum

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Ureaplasma urealyticum is the bacterial species most often connected with preterm birth, although it often colonises the amniotic fluid without any adverse effects. The induction of preterm labour seems to depend on whether the bacteria produce an inflammatory reaction. In vitro stimulation of choriodecidual tissue with high amounts of U.urealyticum or with lipopolysaccharide induced a qualitatively similar inflammatory response detected by the production of tumour necrosis factor alpha, followed by secretion of anti-inflammatory cytokine interleukin-10 and of prostaglandin E2. Lower quantities of bacteria failed to induce any response.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In the majority of cases, spontaneous preterm birth is caused by intrauterine inflammation following bacterial infection.1,2 The infection is usually subclinical, without any signs of infection before the onset of labour or the preterm prelabour rupture of membranes (PPROM). Although the presence of bacteria can often be shown in the amniotic fluid, signs of an inflammatory reaction, especially increased concentration of pro-inflammatory cytokine interleukin-6 (IL-6), predict better the onset of preterm labour.

Choriodecidua is the interface separating the mother and the fetus. Maternal cells in the decidua have a special immunologic role in preventing the mother from rejecting the semi-allogenic fetus. Immunosuppression in the decidua is maintained by several mechanisms.3 The immunosuppressive state in decidua is also likely to prevent a weak bacterial antigenic stimulus from inducing a vigorous inflammatory reaction. This would be beneficial to the continuation of the pregnancy, as such antigenic stimulus which does not threaten the wellbeing of a fetus or the mother can be ignored instead of inducing inflammatory reaction leading to preterm delivery.

Ureaplasma urealyticum is the bacterial species most often found in amniotic fluid.2 It can be found in up to 12% of the amniotic fluid samples during the second trimester, although it causes preterm birth in less than 25% of cases while most of pregnancies proceed normally to term.4,5 Apparently due to the immunosuppressive properties of choriodecidua and the low pathogenicity of U. urealyticum, it can colonise the choriodecidua and the amniotic fluid without causing inflammation and consequently induction of preterm birth. However, U. urealyticum is also the species most frequently associated with preterm labour and PPROM, comprising more than half of the positive isolates in amniotic fluid. It is known to be able to produce an inflammatory reaction in a dose-dependent manner in various experimental set-ups.

Our interest was to study the inflammatory response induced in choriodecidua by U. urealyticum and observe whether the response differs from that produced by lipopolysaccharide (LPS) of Escherichia coli. We also wanted to find out if the quantity of Ureaplasma is the factor that determines whether an inflammatory reaction will be induced. The study was performed by treating human choriodecidual explants with different concentrations of inactivated U. urealyticum and LPS and measuring the production of the pro-inflammatory cytokine tumour necrosis factor α (TNF-α), the anti-inflammatory cytokine interleukin-10 (IL-10) and prostaglandin E2 (PGE2) from the culture medium. As Ureaplasma infection is often silent, we wanted to study if U. urealyticum induces a more pronounced anti-inflammatory IL-10 response than LPS. We chose prostaglandin as the end-point product because it is believed to be the principal initiator of labour.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The study was approved by the Ethical Committee of the Hospital District of South West Finland.

The Ureaplasma antigen was prepared by culturing U. urealyticum serotype standard strain 8 (T960; ATCC) at 37°C in 1.5 l of Ureaplasma broth medium containing 22.5 g/l of trypticase soy broth, 16.5% horse serum, 7.5% of 25% fresh yeast extract, 0.36% urea, 380 000 units/l penicillin G and phenol red. Cells were harvested in late log phase by centrifugation at 30 000 ×g for 90 minutes at 4°C. The pellet was washed three times by resuspension in PBS and centrifuged as above for 30 minutes. After the final wash, the Ureaplasma was resuspended in a total volume of 2 ml of phosphate buffered saline (PBS). The number of colour changing units (CCU) of the concentrated suspension was determined in duplicate by ten-fold titration in Ureaplasma broth. The remaining U. urealyticum suspension was heat killed by incubation in a waterbath at 56°C for 20 minutes. Complete killing was assured by incubating 25 μl of the suspension on Ureaplasma agar as well as incubation in Ureaplasma broth without urea supplement and subsequent subculture on agar on days 1, 3 and 7. The latter procedure was used as the heat-killed suspension of U. urealyticum produced a prompt colour change in Ureaplasma broth because of urease activity. The U. urealyticum antigen was stored at −70°C in 0.1 ml aliquots until used.

For the choriodecidual explant culture, fetal membranes from five healthy women undergoing term elective caesarean section with intact membranes and without labour were obtained. Immediately after placental delivery, an approximately 150-cm2 section of the fetal membranes was cut out. It was then carefully rinsed with sterile 0.9% saline solution to remove residual blood. The amnion was manually removed. Round choriodecidual explants with a diameter of 6 mm were excised with a biopsy punch. Explants were pooled and randomly distributed into six-well plates, with six explants per well. Each well contained 4 ml of Iscove’s modified dulbecco’s medium (IMDM) (Gibco BRL, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (BioWhittaker, Walkersville, MD, USA), Hepes (Gibco BRL), β-mercaptoethanol (Gibco BRL), L-glutamate (BioWhittaker) and gentamicin (Biological Industries, Kibbutz Beit Haemek, Israel). All experiments were performed in duplicate wells. The explants were allowed to equilibrate overnight at 37°C in a humidified atmosphere of 5% CO2/95% air. The following day, the medium was replaced. Ureaplasma antigen, corresponding to final concentrations of 102, 104 or 106 CCU/ml, or 5 micrograms per millilitre LPS (E. coli serotype O127:B8 from Sigma [St Louis, MO, USA]) was added to the wells with the explants. Wells without added stimulant served as a negative control to rule out pre-existing inflammation or other spontaneous release of cytokines. After 6 hours, 500 μl of medium from each well was removed and centrifuged. Ketoprofen (20 micrograms per millilitre supernatant) (Sanofi-Aventis, Paris, France) was added to the supernatant to inhibit further PGE2 production and stored at −70°C until analysed. The rest of the medium was collected after 24 hours and treated in a similar manner. The production of the cytokines and prostaglandin was standardised to the area of the explants in each well.

TNF-α and IL-10 were measured using enzyme-linked immunosorbent assay from CLB Sanquin (Amsterdam, The Netherlands) and PGE2 by enzyme immunoassay (EIA) from R&D Systems (Minneapolis, MN, USA), according to the supplier’s instructions. Detection levels were 15 pg/ml for both TNF-α and IL-10 and 65 pg/ml for PGE2.

The statistical analyses were performed using SAS System for Windows (version 9.1; SAS Institute Inc., Cary, NC, USA). The differences between the experiments and time points in the concentrations of TNF-α, IL-10 and PGE2 were compared using linear mixed model. The fetal membrane was used as a random effect in mixed models because the two experiments from each membrane were interrelated. Due to positively skewed distribution, concentration values were logarithmically transformed for statistical analysis. Tukey’s method was used to adjust P values for multiple comparisons. P < 0.05 was considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The results from the experiments are summarised in Figure 1.

image

Figure 1. A summary of the stimulation experiments. On the x-axis are presented both time points (6 and 24 hours) from each experiment (medium [M], three different concentrations of Ureaplasma antigen, which correspond to the amount of CCUs per millilitre of Ureaplasma culture, and LPS 5 micrograms per millilitre). The box represents the 25th and 75th percentiles of the concentration values, with the median marked as a line inside the box. Minimum and maximum values are shown with whiskers. The symbol ‘*’ marks the bars where there is a significant statistical difference (P < 0.05) in the concentration, when compared with only the medium at the same time point. The abbreviation ‘d.l.’ refers to the detection limit of enzyme-linked immunosorbent assay.

Download figure to PowerPoint

The amount of Ureaplasma antigen corresponding to 102 and 104 CCU/ml culture did not stimulate the explants to produce TNF-α either at 6 or 24 hours. With 106 CCU/ml of the antigen, there was a significant rise in the cytokine concentration, which peaked at 6 hours and was on the decline after 24 hours. The same pattern was observed with LPS. The induction of IL-10 production also required 106 CCU/ml of the antigen. IL-10 response to 106 CCU/ml of Ureaplasma antigen and to LPS was similar, with a concentration that increased between 6 and 24 hours.

After 6 hours, only LPS produced a statistically significant increase in the concentration of PGE2 compared with the medium. After 24 hours, PGE2 production with both 106 CCU/ml of Ureaplasma antigen and LPS was significantly higher than with merely the medium. The PGE2 concentration increased between 6 and 24 hours. Using 102 and 104 CCU/ml of Ureaplasma antigen or the medium, there was no increase in the concentration of PGE2 between the time points.

In this study, the reaction of TNF-α to the stimulation was the fastest, followed by the responses of IL-10 and PGE2. This is the expected order for these cytokines to be secreted after an inflammatory stimulus. The equilibrium between TNF-α and IL-10 was similar irrespective of whether the explants were stimulated with Ureaplasma antigen or LPS, indicating no bias towards anti-inflammatory reactivity with Ureaplasma.

A high amount of Ureaplasma antigen was needed to generate an inflammatory response in choriodecidua. While there was a threshold for the amount of Ureaplasma antigen needed to induce an inflammatory reaction, we could not demonstrate a straightforward dose dependency in the response to U. urealyticum because we were not able to increase the concentration of Ureaplasma antigen in the cultures. The response was qualitatively similar to the one observed with LPS. Quantitative comparison with LPS cannot be made, as the amounts of LPS and Ureaplasma antigen are not comparable.

The stimulation experiments were performed in duplicates using five membranes from different individuals. There was only slight variation between the two parallel experiments from one membrane. As expected, there was quantitative variation in the results between the tissues from different individuals. Despite this variation, the changes in cytokine concentrations between the time points followed the same pattern in each individual.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

U. urealyticum is by far the most common intrauterine bacterial species during pregnancy2 and thus a major contributor to spontaneous preterm birth. We wanted to study local maternal inflammatory response in human decidua to U. urealyticum and compare it with the response to LPS from E. coli, which is known to induce a vigorous response. As a model, we used cultured tissue explants from human postdelivery choriodecidua obtained from healthy women who had undergone a caesarean delivery without labour or premature rupture of fetal membranes or other reason to suspect pre-existing inflammation. The explants mainly consist of maternal decidua harbouring cells of immune system and fetal chorion, which is mostly connective and trophoblast tissue. The inflammatory reaction in choriodecidua is likely to be of maternal origin. The tissue explant model has the advantage that it imitates the in vivo function of an organ with many types of responsive cells, not just reactivity of a single isolated cell type. The explants are also treated minimally, contrary to experiments requiring cell extraction and isolation. This ensures that the response is less susceptible to confounding reactions from pre-treatment.

IL-6 is the pro-inflammatory cytokine most often used as a marker for intrauterine inflammation. High concentrations of IL-6 in amniotic fluid and fetal blood are known to predict the onset of labour. However, for our experimental set-up, IL-6 is too slow to react and not sufficiently specific for inflammation induced by bacterial antigenic stimulation. For this purpose, TNF-α is the most suitable cytokine because it is produced by macrophages as immediate pro-inflammatory response and also has a short half-life. The production of anti-inflammatory cytokine IL-10 is induced in macrophages by TNF-α as a counterbalance but is also directly stimulated by the inflammatory antigen.

U. urealyticum can persist for a long time in amniotic fluid without any response. This could be due to the low pathogenicity of this bacterial species, small amounts of U. urealyticum may not be capable of inducing inflammation in an environment with powerful anti-inflammatory capacity such as decidua. The other possibility is that the response induced by U. urealyticum is qualitatively different from the response induced by more virulent bacteria. Our results seem to indicate that the first hypothesis is more probable because a large amount of U. urealyticum antigen was needed to achieve a measurable reaction, whereas lower concentrations failed to induce any response. Moreover, excessively strong defence mechanisms against bacteria in decidua could compromise the pregnancy if a few bacteria or bacterial antigens were to reach the uterus from the lower genital tract. Thus, it is logical that there would be a response threshold for bacterial stimulus. In addition, the response observed with a high concentration of U. urealyticum was qualitatively similar to that of LPS, with no apparent difference in the relative production of pro-inflammatory TNF-α and anti-inflammatory IL-10, indicating no bias towards anti-inflammatory reactivity by U. urealyticum.

Contrary to U. urealyticum, virulent bacteria such as E. coli seldom colonise amniotic fluid without clinical symptoms. Instead, they usually rapidly lead to symptomatic chorioamnionitis and preterm delivery with fetal and neonatal infections. U. urealyticum and other species in the class mollicutes differ from other bacteria in that they lack a cell wall with specific antigenic properties or LPS. The outer plasma membrane of U. urealyticum has a number of distinct lipoproteins anchored to it, which serve as antigens. Also in the choriodecidua, Mycoplasma antigens are mainly recognised by Toll-like receptor (TLR)-2, as LPS is recognised by TLR-4.6 The ligation of an antigen to either of the receptors leads to the activation of an identical reaction cascade in the cells of the innate immune system. The strength of the host response may vary depending on the capability of the antigen to stimulate the immune system. Our experiment shows that despite the different antigenic properties and recognition pathways, the response to U. urealyticum in term choriodecidua does not qualitatively differ from that to LPS.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The study was financially supported by grants from The Finnish Cultural Foundation, The Paulo Foundation and The Turku University Foundation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Goldenberg RL, Hauth JC, Andrews WW. Intrauterine infection and preterm delivery. N Engl J Med 2000;342:15007.
  • 2
    Maxwell NC, Davies PL, Kotecha S. Antenatal infection and inflammation: what’s new? Curr Opin Infect Dis 2006;19:2538.
  • 3
    Mellor AL, Munn DH. Immunology at the maternal-fetal interface: lessons for T cell tolerance and suppression. Annu Rev Immunol 2000;18:36791.
  • 4
    Gerber S, Vial Y, Hohlfeld P, Witkin SS. Detection of Ureaplasma urealyticum in second-trimester amniotic fluid by polymerase chain reaction correlates with subsequent preterm labor and delivery. J Infect Dis 2003;187:51821.
  • 5
    Perni SC, Vardhana S, Korneeva I, Tuttle SL, Paraskevas LR, Chasen ST, et al. Mycoplasma hominis and Ureaplasma urealyticum in midtrimester amniotic fluid: association with amniotic fluid cytokine levels and pregnancy outcome. Am J Obstet Gynecol 2004;191:13826.
  • 6
    Kim YM, Romero R, Shim SS, Kim EC, Yoon BH. Toll-like receptor-2 and -4 in the chorioamniotic membranes in spontaneous labor at term and in preterm parturition that are associated with chorioamnionitis. Am J Obstet Gynecol 2004;191:134655.