| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pregnancy |
by Whole Fetal Membranes Depend on Initial Interactions of Amnion or Choriodecidua with Lipopolysaccharides or Group B Streptococci1
Biomedical Research Branch,3
Direction of Research,4 Obstetrics
Gynecology Branch,5 and
Neonatology Branch,6 Instituto Nacional de Perinatologia, Mexico City 11000, Mexico
Experimental Medicine Department,7 School of Medicine, Universidad Nacional Autonoma de Mexico, Mexico City 04510, Mexico
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
by fetal membranes stimulated with group B streptococci (GBS) and lipopolysaccharide (LPS). The aim was to evaluate the initial response of full-thickness membranes to the microbial insult using an in vitro experimental model that allowed testing of the individual contributions of amnion and choriodecidua to stimulation. Full-thickness membranes were obtained after delivery by elective cesarean section from women at 37–40 wk of gestation without evidence of active labor. The membranes were mounted in Transwell devices, physically separating the upper and lower chambers. The LPS (500 ng/ml) or GBS (1 x 106 colony-forming units/ml) was added to either the amniotic or choriodecidual surface, and accumulation of IL-1ß and TNF were measured in both compartments using a specific ELISA. Fetal membranes followed different patterns of secretion of proinflammatory cytokines that depended on the side to which the stimulus was added or the nature of the stimulus itself. The TNF was secreted by amnion and choriodecidua in the presence of LPS or GBS, and stimulation with GBS induced a greater synthesis of IL-1ß than did stimulation with LPS. Choriodecidual tissue was more responsive than amniotic tissue, and this response tended to be higher even when the stimulation was only on the amniotic side. However, the amnion plays an active role in recognizing LPS or GBS, contributing a significant amount of TNF. Thus, cooperative and bidirectional communications occur between amnion and choriodecidua in response to bacterial products, which include intermembranous cytokine traffic and signaling between tissues.
cytokines, parturition, placenta, pregnancy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
[7], and matrix metalloproteinases [8]. The resulting microenvironment results in variable amounts of uterotonic compounds and extracellular matrix-degrading enzymes that may act on both fetal membranes and cervix [9], leading to either PROM or preterm delivery. Although we do not know the specific mechanisms that lead to PROM or preterm delivery, rupture of the membranes will develop if selective degradation of the amniochorionic extracellular matrix is the main event [10–12].
Several reports have evaluated human fetal membrane responses to entire bacteria and their products by using in vitro models that resemble intrauterine infection. Decidual cells are capable of synthesizing at least interleukin (IL)-1ß and tumor necrosis factor (TNF) as well as IL-6, IL-8, and IL-10 in response to infection-like conditions [5, 13– 15], and they have IL-1, IL-6, TNF-I, and TNF-II receptors [16]. Chorionic cells are also capable of responding to bacterial products secreting at least IL-1ß, IL-6, IL-8, and IL-10 [17, 18], and they express IL-1ß and IL-6 receptors [19]. Local macrophages express IL-6, TNF-I, and TNF-II receptors [16]. Epithelial amnion cells can also react to infection-emulating conditions by secreting IL-6 and IL-8 [20]. This reveals the existence of complex autocrine and paracrine interactions between amniochorionic cells when exposed to intrauterine infections [21]. However, few experimental studies using entire membranes have been conducted to study the integral reaction of the amniochorion to bacterial products by trying to emulate the natural conditions of bacteria and host interactions. Enough clinical information is available to support the idea that intrauterine infections arise from initial bacterial growths in the lower genital tract that overcome natural, nonspecific defense barriers in the cervix, making the choriodecidua the next site of microbial invasion [2]. At this stage, the membranes must act as a mechanical barrier and deploy an inflammatory response to control the infection. We have evidence from experimental models that choriodecidual infection can follow a chronic course or may be controlled without any further complications [9]. Although pregnancy is severely compromised when intraamniotic infection occurs, further knowledge regarding the mechanisms of interaction between different microorganisms and fetal membranes is central to the understanding of how infection may end in PROM or preterm labor.
To understand the interactions between choriodecidua and amnion in the presence of infectious agents, we have developed an ex vivo experimental model in which we emulate the anatomical relationships of the fetal membranes, maintaining the selective barrier function of the membranes (making possible selective stimulation of either the choriodecidua or amnion with LPS or GBS) and measuring the compartmentalized secretions of IL-1ß and TNF. Characterization of the initial response of amnion or choriodecidua resulting in secretion of these cytokines is addressed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This project was approved by the Internal Review Board of Instituto Nacional de Perinatologia in Mexico City (register no. 212250-06101). Ten fetal membranes were obtained after delivery by elective cesarean section with written informed consent. Five of the membranes were used for stimulation with LPS, and the other five were used for GBS experiments. Women at 37–40 wk of gestation without evidence of active labor or clinical or microbiological signs of chorioamnionitis or lower genital tract infections were included. General microbiological analyses were conducted on the placenta and fetal membranes immediately after delivery by rolling a sterile swab across a randomly selected area and performing standard microbiological procedures.
The membranes were transported to the laboratory in sterile Dulbecco modified Eagle medium (DMEM; Gibco BRL, Bethesda, MD) and rinsed in sterile Hanks balanced salt solution (Gibco BRL) to remove adherent blood clots. Membranes were manually cut into disks (diameter, 18 mm) and were held with silicone rubber rings in the upper chamber of a Transwell system (Costar, New York, NY) in which the original polyethylene terephthalate membrane had been removed. In this model, the choriodecidua faces the upper chamber, and the amnion faces the lower chamber, making it possible to test the two compartments independently (Fig. 1).
|
One milliliter of DMEM supplemented with 10% fetal calf serum (FCS), 1 mM sodium pyruvate, and 1x antibiotic-antimycotic solution (penicillin, 100 U/ml; streptomycin, 100 µg/ml; amphotericin B, 0.25 µg/ ml; Gibco BRL) was added to each chamber. The mounted explant was then placed in a 12-well tissue-culture plate (Costar) and incubated under 5% CO2 in 95% air at 37°C.
Another set of studies was performed simultaneously in which amnion and choriodecidua were manually separated and cut into disks (diameter, 12 mm) using a biopsy punch. Two pieces of choriodecidua or amnion were placed in each well of a 24-well tissue-culture plate with 1 ml of DMEM supplemented as described above.
Tissue Viability
Viability of membranes in the two-chamber model and in separated amnion or choriodecidua was determined by a colorimetric assay using tetrazolium salts added to the culture medium (Boehringer Mannheim, Germany). The assay is based on cleavage by metabolically active cells of the yellow tetrazolium salt XTT to form an orange formazan dye [22]. The assay was performed every 24 h of culture over 5 days.
Verification of the existence of a "two-chamber" system in which the amniochorion acts as a true selective barrier was evaluated by measuring the transepithelial electrical resistance (TER) of membranes as a measure of their physical integrity during experimental manipulation [23]. Briefly, randomly chosen membranes that were in culture in the Transwell model were taken out at 24, 48, 72, and 96 h under every experimental condition and immediately placed as a flat sheet between two Lucite chambers (filled with DMEM) of an Ussing chamber (UNAM, Mexico). A Millipore acetate filter (diameter, 13 mm) was used as a support for the membrane. A current pulse of 20 µA was applied, and the voltage deflection was monitored in the vicinity of the membrane by two silver electrodes. Electrical resistance was measured in at least 10 different points for each membrane, and the contributions of the filter, solutions, and electrodes were subtracted. This information was interpreted as the actual integrity of the membranes at different incubation times. At least five different membranes were evaluated in each experiment.
Microscopy
Membranes were analyzed by standard histology techniques using hematoxylin-and-eosin staining to verify the anatomical integrity of the tissues following incubation.
Stimulation of Membranes in Culture
Explants were preincubated for 48 h in medium containing FCS to stabilize the tissues after manipulation [24]. After this time, the medium was changed to DMEM with 0.2% lactalbumin hydrolysate (Gibco BRL), and all experiments were carried out in this medium after 24 h. The explants were stimulated with 500 ng/ml of LPS from Escherichia coli 055: B5 (Sigma, St Louis, MO) [24] or coincubated with 1 x 106 colony-forming units/ml of Streptococcus agalactiae [5], serotype III (9710860) isolated from vaginal exudates. Potentially toxic effects of bacterial overgrowth were prevented by the presence of antibiotics in the media. Media from either the amnion- or the choriodecidua-facing chamber were collected after 24 h of incubation.
Each experiment with explants mounted in the Transwell system included the following four conditions in triplicate for each membrane: 1) control membranes in which only medium was added to the compartments, 2) GBS or LPS were added simultaneously to both compartments, 3) GBS or LPS were only added to the choriodecidual side, and 4) GBS or LPS were only added to the compartment in contact with the amnion. The media on either side were collected at 24 h and stored frozen at –70°C until assayed.
Separated amniotic or choriodecidual tissues were stimulated under equivalent conditions. The concentrations of protein in all samples were estimated using the method of Bradford [25].
Cytokine Assays
The IL-1ß and TNF concentrations were quantified by ELISA according to standard methods [24]. Monoclonal antibodies against IL-1ß or TNF (R&D Systems, Minneapolis, MN) were used as the capture antibodies, and polyclonal biotinylated antibodies against both cytokines were used as detection antibodies (R&D Systems). Standard curves were developed using human recombinant IL-1ß (R&D Systems) or human recombinant TNF (R&D Systems). The IL-1ß ELISA had a sensitivity of 15.6 pg and was linear in the range from 20 to 1000 pg/ml. The TNF ELISA had a sensitivity of 3.91 pg/ml and was linear in the range from 5.0 to 500 pg/m. Both intra- and interassay coefficients of variation were less than 5%. A rigorous quality-control program, including external and internal standards for both cytokines, is followed in our laboratory.
Statistical Analysis
Comparisons between groups were performed using the Kruskal-Wallis one-way analysis of variance on rank tests, and P < 0.05 was considered to be significant. All values are mean ± SD unless otherwise stated.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
/cm2) did not show a significant decline during incubation with either LPS (53.38 ± 13.56 /cm2) or GBS (48.59 ± 10.0 /cm2). Cytokine production was normalized as ng/cm2 of full-thickness membranes.
We evaluated IL-1ß and TNF secretion into the culture media after stimulation with 500 ng/ml of LPS for 24 h. Stimulation of any side of the membranes with LPS using the Transwell model induced a minimum increase of 22-fold in the IL-1ß concentration in both amniotic and choriodecidual compartments compared with the corresponding controls (n = 5, P < 0.05) (Table 1). However, IL-1ß reached twice the concentration in the choriodecidual compartment compared with the amniotic compartment when the choriodecidua was directly stimulated either alone or simultaneously with the amnion (n = 5, P < 0.001) (Table 1). When choriodecidua and amnion were cultured and assayed individually, only the choriodecidua secreted IL-1ß in response to LPS (n = 5, P < 0.05) (Fig. 2a).
|
|
Similar results were obtained for TNF secretion after stimulation with LPS, except that the minimum increase in secretion was 35-fold above the basal conditions (n = 5, P < 0.05) (Table 1). When amnion and choriodecidua were manually stripped and cultured, they secreted TNF in response to LPS; however, the choriodecidua secreted threefold more cytokine than the amnion (n = 5, P < 0.01) (Fig. 2b).
When the membranes were stimulated with GBS, the IL-1ß concentration increased in the choriodecidual compartment to an average level 30-fold above the basal concentrations and only when GBS was added directly to the choriodecidual compartment (n = 5, P < 0.05) (Table 2). Adding GBS simultaneously to amnion and choriodecidua resulted in increased secretion of IL-1ß into the choriodecidual compartment, but this never reached the response observed when the choriodecidua was stimulated alone (n = 5, P < 0.05) (Table 2). The nonresponsive behavior of the amnion was confirmed using separated membranes. In those experiments, only the choriodecidua was capable of secreting IL-1ß (n = 5, P < 0.05) (Fig. 3a). However, the maximum secretion of IL-1ß by the isolated choriodecidua was only a fraction of that secreted when the membranes were cultured together (Fig. 3a).
|
|
Secretion of TNF was induced by GBS, and higher concentrations were found in the choriodecidual compartment when this portion of the membranes was stimulated directly (n = 5, P < 0.05) (Table 2). However, concentrations never reached the maximum responses observed with LPS stimulation. When choriodecidua or amnion were cultured in separate wells and stimulated with GBS, the production of TNF was significantly higher in the amnion (n = 5, P < 0.05) (Fig. 3b).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
found in the choriodecidual or amniotic compartments after stimulation with LPS or GBS were selectively secreted by these membranes. We selected these cytokines as functional markers of membrane responses to LPS or GBS, both because other reports have shown that entire membranes can express these cytokines on stimulation with bacterial products and because they are key modulators of the inflammatory response. Previous reports on the amniochorionic response to infectious elements have focused on the effects on isolated cells [17, 18] or entire membranes in which no effort was made to preserve the in vivo spatial relations of amnion and choriodecidua as a separating barrier between the maternal/fetal compartments,[24, 27]. To our knowledge, in only two published studies have entire membranes been evaluated using a system such as ours, and these studies have demonstrated that PGs are synthesized in the amniotic compartment following stimulation of the choriodecidual side [28, 29].
The amniotic membrane and the choriodecidua showed selective and distinct TNF and IL-1ß secretion patterns after stimulation with LPS and GBS. Some of these effects appear to be tissue-specific responses, but others can only be explained by cooperative interaction between the amnion and choriodecidua. In general, the choriodecidua was the more reactive tissue, secreting both IL-1ß and TNF, and the amnion was only capable of secreting TNF. On the other hand, stimulation with LPS elicited high secretion of TNF by both amnion and choriodecidua, whereas GBS elicited a higher secretion of IL-1ß by the choriodecidua alone. This supports the concept that the fetal membranes have polarity both for stimulus recognition and for cytokine secretion.
Stimulation with LPS resulted in accumulation of IL-1ß and TNF in both compartments, independently if the compound was only added to the amnion or to the choriodecidua. Secretion of IL-1ß to the amniotic compartment was demonstrable only when choriodecidua and amnion were together, in contrast to experiments in which both tissues were mechanically separated and stimulated, and only the choriodecidua secreted IL-1ß. These results are in agreement with previous results demonstrating that the mRNA for this cytokine is only expressed in the choriodecidua [19]. Therefore, the presence of IL-1ß in the amniotic compartment during our experiments can only be explained by the diffusion of choriodecidua-secreted IL-1ß through the fetal membranes. This possibility has been addressed by Kent et al. [30], who demonstrated the transmembrane diffusion of several cytokines. Although we know little about the cellular biology of cytokine transport across tissues, we know that these signals can modify paracellular transport, a phenomenon that has been explored in several epithelial tissues and that may explain the extensive diffusion of cytokines in our model. On the other hand, stimulation of IL-1ß secretion by the choriodecidua after amniotic stimulation with LPS implies that this molecule can travel across the amnion to reach the choriodecidua or that there occurs an initial recognition of LPS by amnion and transduction of the signal to the choriodecidua. The chemical nature of LPS makes it a nondiffusible compound in the amniochorion, and to our knowledge, no experimental evidence suggests LPS transmembrane transport [28, 31]. Hence, the only mechanism to explain the amnion-dependent choriodecidual secretion of IL-1ß is an intramembranous signaling pathway that starts in the amnion.
The release of large amounts of TNF by membranes under LPS stimulation results from summative amniochorion secretion, because both tissues can independently secrete this cytokine, as shown by the experiments using separated amnion or choriodecidua. The choriodecidual compartment accumulated more TNF than the amniotic compartment, but the presence of TNF in the choriodecidual compartment after amniotic stimulation can also be explained by diffusion of the cytokine or activation of choriodecidual secretion of TNF by a hitherto-unidentified, intramembranous signaling pathway.
In contrast to the effects of LPS, stimulation of either amnion or choriodecidua with GBS resulted in IL-1ß secretion only by the choriodecidua. We found no evidence of cooperative communication between the fetal membranes, and GBS was better than LPS for stimulating the choriodecidual secretion of IL-1ß. These findings support the idea that the choriodecidua is the only source for IL-1ß secretion and that it contains the elements for recognition of GBS. However, as demonstrated previously, IL-1ß can reach the amniotic compartment by a poorly characterized, intramembranous trafficking pathway [19, 28, 31]. On the other hand, both amnion and choriodecidua can cooperate in recognizing GBS, leading to TNF secretion as well as IL-1ß secretion. The choriodecidua was the main source for TNF even when the amnion face received the primary stimulus, reinforcing the hypothesis of an intramembranous signaling pathway. However, a puzzling situation emerged when the secretions of both IL-1ß and TNF were analyzed under simultaneous stimulation of choriodecidua and amnion. We observed no response in the amniotic compartment and lower secretions of both cytokines in the choriodecidual compartment relative to the maximum secretion induced with GBS. This also contrasts with the capacities of the amnion and choriodecidua when they were tested separately, at which time the amnion secreted appreciable amounts of TNF. It is possible that counterregulatory signals were exchanged between tissues, resulting in a state of pseudoanergy to GBS. The possibility that this situation occurs in vivo is highly improbable but deserves further investigation.
Our results add to the information about the widely recognized secretion of proinflammatory cytokines in amniotic fluids from women with intraamniotic infections [5], and they point toward a cellular origin of these compounds (Fig. 4). Because the choriodecidua appears to be the major source of cytokines in response to infection, further research concerning the choriodecidua/amnion interactions resulting in cytokine release is needed. The inflammatory-response genetic background may explain the variable response of membranes to infection; as demonstrated recently [32], membranes carrying the "hyper-responsive" gene polymorphisms of IL-1ß may secrete greater amounts of this cytokine on equivalent stimulation with LPS compared to membranes carrying the more common gene polymorphism. This may account for the observed difference in the basal secretion of this cytokine in our experiments.
|
Also of interest is the mechanism of cytokine transmembrane passage. In the present study, we found evidence that the fetal membranes can react differently to the arrival of microorganisms depending on the primary route of contact (amnion or choriodecidua) or the nature of the stimuli (entire organism or soluble products). Our results also suggest the existence of a cooperative interrelation between both membranes and the existence of at least two different mechanisms for membrane activation in response to microorganisms. One depends on the interaction between soluble products, such as LPS. This compound may exert a whole response in chorioamnion, independently of the stimulated side. Toll-like receptors (TLRs) are essential for the induction of innate immune response and were identified early on as receptors for LPS-induced signal transduction. The effective TLR4 activation by LPS requires the interaction of LPS with CD14 and the accessory protein MD-2 [33]. Although to our knowledge no reports of TLR expression in fetal membranes have appeared, some reports have demonstrated TLR2 and TLR4 proteins in human placenta of normal-term deliveries [34]. Furthermore, other authors have demonstrated up-regulation of TLR4 expression in villous Hofbauer cells of preterm placentas and amniotic fluid with chorioamnionitis [35, 36], stressing their possible role in the initial recognition of LPS-containing bacteria.
Recent evidence of an amnion-derived mediator able to activate chorion under infection-related conditions has been described by Ognjanovic et al. [37]. They showed that pre-B-cell colony-enhancing factor is a cytokine that is expressed constitutively by the human fetal membranes during pregnancy but is up-regulated during chorioamnionitis.
The second mechanism can only be explained by invoking the recognition of GBS by immune cells located in the choriodecidual side to initiate an inflammatory response. In addition, our experiments suggest the existence of a noncharacterized, intermembranous signaling network that permits the coordinate and bidirectional activation of both amnion and choriodecidua.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Felipe Vadillo-Ortega, Instituto Nacional de Perinatologia, Montes Urales 800, Lomas de Virreyes, Mexico D.F. 11000, Mexico. FAX: 52 5 5520 0034; felipe.vadillo{at}uia.mx
Received: 17 February 2004.
First decision: 5 March 2004.
Accepted: 7 June 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) release from cultured human fetal membranes in response to endotoxic lipopolysaccharide mirrors amniotic fluid concentrations. Am J Obstet Gynecol 1996 174:1855-1862[CrossRef][Medline] and IL-10 regulate lipopolysaccharide-stimulated human choriodecidual cytokine and prostaglandin E2 production. J Immunol 2003 170:158-166 and interleukin-1ß. Mol Human Reprod 2003 9:1-5This article has been cited by other articles:
|
|
 |
|
  C. Oner, F. Schatz, G. Kizilay, W. Murk, L. F. Buchwalder, U. A. Kayisli, A. Arici, and C. J. Lockwood Progestin-Inflammatory Cytokine Interactions Affect Matrix Metalloproteinase-1 and -3 Expression in Term Decidual Cells: Implications for Treatment of Chorioamnionitis-Induced Preterm Delivery J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 252 - 259. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Chevillard, A. Derjuga, D. Devost, H. H Zingg, and V. Blank Identification of interleukin-1{beta} regulated genes in uterine smooth muscle cells Reproduction, December 1, 2007; 134(6): 811 - 822. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Garcia-Lopez, F. Vadillo-Ortega, H. Merchant-Larios, R. Maida-Claros, M. Osorio, D. Soriano-Becerril, H. Flores-Herrera, J. Beltran-Montoya, Y. Garfias-Becerra, and V. Zaga-Clavellina Evidence of in vitro differential secretion of 72 and 92 kDa type IV collagenases after selective exposure to lipopolysaccharide in human fetal membranes Mol. Hum. Reprod., June 1, 2007; 13(6): 409 - 418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-J. Leroy, E. Dallot, I. Czerkiewicz, T. Schmitz, and M. Breuiller-Fouche Inflammation of Choriodecidua Induces Tumor Necrosis Factor Alpha-Mediated Apoptosis of Human Myometrial Cells Biol Reprod, May 1, 2007; 76(5): 769 - 776. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Lockwood, F. Arcuri, P. Toti, C. D. Felice, G. Krikun, S. Guller, L. F. Buchwalder, and F. Schatz Tumor Necrosis Factor-{alpha} and Interleukin-1{beta} Regulate Interleukin-8 Expression in Third Trimester Decidual Cells: Implications for the Genesis of Chorioamnionitis Am. J. Pathol., October 1, 2006; 169(4): 1294 - 1302. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Zaga-Clavellina, H. Merchant-Larios, G. Garcia-Lopez, R. Maida-Claros, and F. Vadillo-Oretega Differential Secretion of Matrix Metalloproteinase-2 and -9 After Selective Infection With Group B Streptococci in Human Fetal Membranes Reproductive Sciences, May 1, 2006; 13(4): 271 - 279. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
