BIOLOGY OF REPRODUCTION (2014) 91(5):105, 1–7 Published online before print 17 September 2014. DOI 10.1095/biolreprod.114.121236
Altered Decorin and Smad Expression in Human Fetal Membranes in PPROM1 Casie E. Horgan,3,4 Hailey Roumimper,4,5 Richard Tucker,4 and Beatrice E. Lechner2,4 4
Department of Pediatrics, Women and Infants’ Hospital of Rhode Island, The Warren Alpert Medical School of Brown University, Providence, Rhode Island 5 Brown University, Providence, Rhode Island
Humans with Ehlers-Danlos syndrome, a subtype of which is caused by abnormal decorin expression, are at increased risk of preterm birth due to preterm premature rupture of fetal membranes (PPROM). In the mouse model, the absence of decorin leads to fetal membrane abnormalities, preterm birth, and dysregulation of decorin’s downstream pathway components, including the transcription factor p-Smad-2. However, the role of decorin and p-Smad-2 in idiopathic human PPROM is unknown. Fetal membranes from 20–25 pregnancies per group were obtained as a cross-sectional sample of births at one institution between January 2010 and December 2012. The groups were term, preterm without PPROM, and preterm with PPROM. Immunohistochemical analysis of fetal membranes was performed for decorin and p-Smad-2 using localization and quantification assessment. Decorin expression is developmentally regulated in fetal membranes and is decreased in preterm birth with PPROM compared to preterm birth without PPROM. In preterm with PPROM samples, the presence of infection is associated with significant decorin downregulation compared to preterm with PPROM samples without infection. The preterm with PPROM group exhibited decreased p-Smad-2 staining compared to both the term controls and the preterm-withoutPPROM group. Our findings suggest that dysregulation of decorin and its downstream pathway component p-Smad-2 occurs in fetal membranes during the second trimester in pathological pregnancies, thus supporting a role for decorin and p-Smad-2 in the pathophysiology of fetal membranes and adverse pregnancy outcomes. These findings may lead to the discovery of new targets for the diagnosis and treatment of PPROM. decorin, fetal membranes, PPROM, preterm birth, Smad
INTRODUCTION A leading cause of infant mortality is preterm birth—40% of which is caused by preterm premature rupture of fetal membranes (PPROM) [1, 2]. While various factors, such as presence of infection, a previous history of PPROM, and sociodemographic factors, have been identified as contributors, the pathophysiology of PPROM and thus screening or treatment strategies have remained elusive [3].
MATERIALS AND METHODS 1
Supported by NIGMS P20GM103537 and NICHD 1K08HD054676. 2 Correspondence: Beatrice E. Lechner, Women & Infants Hospital of Rhode Island, 101 Dudley St., Providence, RI 02905. E-mail: [email protected] 3 Current address: Tufts University School of Public Health, Boston, Massachusetts.
Study Patients and Tissue Collection Paraffin-embedded fetal membrane block samples stored in the Women & Infants pathology tissue repository were collected retrospectively from patients who had delivered at Women & Infants Hospital of Rhode Island between January 2010 and December 2012. Fetal membranes from three clinical groups were designated to experimental groups: 1) term births (37 wk gestational age), subsequently referred to as full term; 2) preterm births with no PPROM (36 wk gestational age), subsequently referred to as preterm PPROM; and 3) preterm birth with PPROM (36 wk gestational age), subsequently referred to as preterm þ PPROM. The first 25 samples to meet inclusion criteria in reverse chronological order were selected from each group. Term births (full
Received: 8 May 2014. First decision: 12 June 2014. Accepted: 4 September 2014. Ó 2014 by the Society for the Study of Reproduction, Inc. eISSN: 1529-7268 http://www.biolreprod.org ISSN: 0006-3363
1
Article 105
Downloaded from www.biolreprod.org.
Infants born with Ehlers-Danlos syndrome, an inheritable connective tissue disorder, have an increased risk of preterm birth via PPROM compared to their unaffected siblings [4, 5]. A subset of Ehlers-Danlos syndrome, the progeroid variant, is caused by a lack of posttranslational glycosylation of the small leucine-rich proteoglycan decorin and its homologue, biglycan [4, 5]. Decorin is located in the extracellular matrix and is highly expressed in reproductive tissues [6, 7]. Various roles for decorin have been identified in collagen fibrillogenesis and cell growth [8]. Mice deficient in decorin and biglycan display connective tissue abnormalities resembling Ehlers-Danlos syndrome [9]. We have shown that these mice also display preterm birth and fetal membrane anomalies as well as fetal loss and low birth weight [7, 10]. In addition to its structural role in the extracellular matrix [11, 12], decorin binds transforming growth factor b (TGF-b) and affects the TGF-b signaling pathway (reviewed in Kinsella et al. [13]). In this signaling pathway, the extracellular protein TGF-b binds its membrane receptor. Receptor binding leads to the phosphorylation of the Smad family of transcription factors, which then travel to the nucleus to regulate transcription of various collagens, matrix metalloproteinases (MMPs), and tissue inhibitors of metalloproteinases (TIMPs) [13]. These three groups of proteins play a central role in the pathogenesis of PPROM. For example, increased expression of matrix metalloproteinases in fetal membranes leads to PPROM via the degradation of the extracellular matrix [14, 15], while mutations in TIMPs and collagens increase the risk of PPROM [16]. We have previously shown that the TGF-b-Smad signaling pathway is dysregulated in mouse fetal membranes in the absence of decorin and that recombinant decorin rescues the phenotype [10]. Decorin decreases in fetal membranes at birth in term deliveries [17]. Its homologue, biglycan, plays a role in inflammation [18]. However, although our data in mice suggest a role for decorin in the maintenance of intact fetal membranes, little is known about the role of decorin during human pregnancy, and nothing is known about its role in preterm birth and PPROM. Thus, our objective was to study the regulation of decorin and downstream targets of the TGF-b pathway in human fetal membranes.
ABSTRACT
HORGAN ET AL. then used as a marker of decorin expression for statistical analysis given that they were perceived to be biologically more meaningful than maximum fluorescence values. A representative region of the tissue was selected for photo analysis. All adjustments to contrast were linear and were adjusted uniformly for all images in the data set. For the p-Smad-2 quantitative analysis, slides were assessed for localization, intensity, and pattern of staining by two independent and blinded observers.
term) that functioned as a control group were included if no labor had taken place, there were no microscopic signs of infection or clinical history of chorioamnionitis, and there was no history of prior preterm birth or PPROM. Thus, all term births were scheduled Cesarean sections. For the preterm þ PPROM group, all cases within the study time frame with a history of multiple PPROM were chosen to increase the probability of pathology. Next, of the remaining PPROM cases within the time frame, all cases of PPROM without clinical or microscopic signs of infection were chosen. Finally, the remainder of the samples needed for a total of 25 samples were chosen in reverse chronological order among the remaining cases of PPROM within the study time frame that did display signs of infection. For fetal membranes in the preterm PPROM group, which served as an age-matched control group for the preterm þ PPROM group, all samples chosen were delivered by Cesarean section to avoid exposure to labor. For this group, also, all cases without clinical or microscopic signs of infection were chosen within the study time frame. Next, the remainder of the samples needed for a total of 25 samples were chosen in reverse chronological order among the remaining cases within the study time frame that did display signs of infection. The most common reason for preterm birth in the absence of labor or PPROM is pre-eclampsia. Thus, all samples in the preterm PPROM group came from pregnancies complicated by pre-eclampsia. On microscopic analysis of the groups, the number of samples in the preterm þ PPROM group was reduced to 20 because five samples had to be excluded, as they did not contain any fetal membrane tissue. All samples were assigned a deidentified patient code. The following clinical information on the mother and infant was collected: gestational age, birth weight, gender, parity, mode of delivery, attempted induction, presence of infection (chorioamnionitis), gestational age at PPROM, and history of prior PPROM.
Statistical Analyses Statistical analysis of immunofluorescence across preterm groups was done by ANOVA with Sidak post hoc t-tests. Comparison of immunofluorescence by presence of infection within preterm groups was done by t-test. Analysis controlling for gestational age differences was done by linear regression. To achieve this, we compared decorin levels of both preterm groups to the term group and specified a contrast test between preterm PPROM and preterm þ PPROM, using a regression model with gestational age as a continuous covariate, indicators for each preterm group (with term as reference), and decorin level as the dependent variable. Age was not adjusted for in the developmental regulation experiment since age-dependent decorin expression was the outcome we were evaluating. Thus, in that experiment, age was the independent variable, and decorin fluorescence level was the dependent variable.
RESULTS
Ethical Approval Institutional review board approval was obtained for the study of human tissues. All experiments were conducted in accordance with the Society for the Study of Reproduction’s specific guidelines and standards.
Antibodies The following primary antibodies were used: monoclonal mouse antihuman decorin (R&D Systems); polyclonal rabbit anti-human Smad2, phospho-specific, and ser465/467 (EMD Millipore). Secondary antibody labeling was performed using donkey anti-mouse IgG conjugated to Alexa 488 (Invitrogen) for decorin and goat anti-rabbit IgG conjugated to CY3 (Invitrogen) for p-Smad-2. Mouse IgG and rabbit IgG (Vector Laboratories), respectively, were used as controls.
Histology and Immunohistochemistry Paraffin-embedded fetal membrane blocks for each of the 75 patients (25 per group) were sectioned, mounted on slides, and labeled in a deidentified manner. Slides were baked for 2 h at 658C, then deparaffinized with Richard Allen Scientific Clear-Rite-3 (Thermo Scientific) and rehydrated. Antigen retrieval was then performed using Citrate Buffer Antigen Retriever (SigmaAldrich). Slides were blocked with rabbit, goat, or donkey blocking serum (Sigma-Aldrich) at a 1:20 dilution in PBS (Fisher Scientific) for 1 h at room temperature. Slides were incubated overnight at 48C with primary antibody. The following morning, slides were incubated for 30 min at room temperature with secondary antibody and then mounted using Vectashield with DAPI staining (Vector Laboratories).
Image Analysis Fluorescent microscopy was performed to evaluate the samples using an inverted stage Nikon Eclipse TE2000-E microscope equipped with epifluorescent filters and a Nikon Plan Apo 206 and 406 and a Plan Fluor 106 objective lens with 403 magnification. Images were acquired using a Coolsnap HQ cooled CCD camera (Roper Scientific) and MetaVue software (Molecular Devices). Data sets were typically collected in one sitting to minimize variation. Exposure time and all acquisition settings were held constant for each fluorescence channel. Acquisition settings were chosen to minimize image saturation. For the quantitative analysis of decorin staining, the stromal layers of the fetal membranes were traced using the MetaVue tracing tool, including the fibroblast layers of the amnion and chorion, excluding the epithelial cells, trophoblasts, and any space where the stromal layers had torn from analysis. Average and maximum fluorescence intensity (12-bit scale) were collected for the traced regions, excluding all other regions, using the MetaVue average and maximum fluorescence measurement tool. Average fluorescence values were
2
Article 105
Downloaded from www.biolreprod.org.
Demographic characteristics of the three groups of patients (full term, preterm PPROM, and preterm þ PPROM) are shown in Table 1. Per study design, there were significant differences in gestational age between the full-term group and the two preterm groups. There was also a difference in gestational age between the preterm þ PPROM group and the preterm PPROM group, which was subsequently adjusted for in the data analysis. Of the term samples, none had evidence of infection. In the preterm PPROM group, 4% of the samples showed signs of chorioamnionitis. In the preterm þ PPROM group, however, 60% of samples showed signs of chorioamnionitis. Additionally, 100% of the preterm PPROM group displayed pre-eclampsia, while none of the other groups did. First, we assessed decorin expression in fetal membranes by immunohistochemistry as a function of gestational age. To achieve this, we compared decorin levels of both preterm groups (preterm PPROM and preterm þ PPROM) via regression analysis with age as the independent variable and decorin level as the dependent variable. We observed that decorin staining is developmentally regulated in fetal membranes obtained via preterm birth secondary to either PPROM or pre-eclampsia throughout the course of gestation, with an increase in decorin expression as the pregnancy progresses (P ¼ 0.0225). This developmental regulation was observed regardless of whether the samples were fetal membranes from preterm þ PPROM (Fig. 1A) or from preterm births PPROM (Fig. 1B). At full term (37–39 wk), there is a trend toward an increase in decorin expression to a peak at 38 wk followed by a decrease but no statistically significant difference (Supplemental Figure S1; all Supplemental Data are available online at www.biolreprod.org). Next, immunohistochemical analysis of decorin expression comparing the three groups (full term, preterm þ PPROM, and preterm PPROM) was performed, and the results were quantified. Decorin expression was decreased in the preterm þ PPROM group compared to preterm PPROM fetal membranes (Fig. 2A). This decrease in decorin expression, however, was observed only in preterm þ PPROM fetal membranes in the presence of infection but not in the absence of infection (Fig. 2A). The majority of the decorin staining was located in stromal (fibroblast) layers of the amnion and chorion, with less staining in the epithelial layer of the amnion or the trophoblast layer (schematic of localization in Fig. 2B).
ALTERED DECORIN IN HUMAN PPROM TABLE 1. Birth characteristics of fetal membrane samples, stratified by clinical group. Characteristic
Full term (n ¼ 25)
Preterm PPROM (n ¼ 25)
Preterm þ PPROM (n ¼ 20)
Mean gestational age (SD) Birth weight (mean) Gender (% male) Parity (SD) Mode of delivery (% Cesarean section) Attempted induction (%) Chorioamnionitis (%) Mean gestational age at PPROM (SD) Prior PPROM (%)
38.6 (1.0) 3614 g 52 1.3 (0.6) 100 0 0 — 0
30.3 (2.7) 1400 g 56 0.28 (0.6) 100 44 4 — 0
28.9 (6.2) 1411 g 52 1.2 (1.4) 28 52 60 27.3 (5.8) 52
Pre-eclampsia (%) Severe pre-eclampsia (%) Superimposed severe pre-eclampsia (%) Early-onset pre-eclampsia (%)
0 0 0 0
100 92 4 92
0 0 0 0
FIG. 1. Decorin expression in human fetal membranes during the course of gestation. A) Immunohistochemical staining of decorin increases during the course of gestation in fetal membranes after PPROM between 16 and 34 wk of gestation. B) Immunohistochemical staining of decorin increases during the course of gestation in fetal membranes without PPROM between 26 and 31 wk of gestation. Original magnification 340; bar ¼ 20 lm.
3
Article 105
Downloaded from www.biolreprod.org.
given the difference in group gestational ages (P ¼ 0.0037; Fig. 3A). Next, we measured decorin levels in fetal membranes in the preterm þ PPROM group, dividing this group into those with a history of prior PPROM and those without a history of prior PPROM, for whom this was their first episode of PPROM. There was no difference in decorin levels between the two PPROM groups (data not shown).
Thus, decorin staining in this area of the amnion and chorion but not in the epithelial layer or trophoblast layer was quantified according to lines drawn as demonstrated in Figure 2B. Decorin levels in the term group did not significantly differ from the preterm þ PPROM group. However, significantly decreased levels of decorin were present in the preterm birth þ PPROM group compared to the preterm PPROM group, which was still significant when adjusted for gestational age
HORGAN ET AL.
Inflammation, on the other hand, appears to play a role in the differences in the level of decorin expression between the three groups. In the full-term birth group, there were no samples with infection, as defined by histological analysis of the samples or clinical criteria. In both the preterm þ PPROM and the preterm PPROM groups, there were samples with infection and samples without infection. When infection was adjusted for in the preterm þ PPROM group and the preterm PPROM group via regression, no significant differences were observed between the two preterm groups or between all three groups (full term, preterm PPROM, and preterm þ PPROM). We then compared the preterm þ PPROM samples with infection with the preterm þ PPROM samples without infection and observed significant differences in decorin levels. In preterm þ PPROM samples with infection, decorin was significantly downregulated when compared to preterm þ PPROM samples without infection (P ¼ 0.0026; Fig. 3B).
P-Smad-2 expression and localization in fetal membranes was analyzed using immunohistochemistry. In full-term fetal membranes as well as preterm PPROM fetal membranes, the p-Smad-2 staining was localized almost exclusively to the trophoblast layer. In contrast, the amnion/chorion fibroblast layer displayed very little p-Smad-2 staining in the full-term and preterm PPROM samples. In the trophoblast of the fullterm and preterm PPROM samples, the trophoblast staining was nuclear and consisted of dots of staining. This is the expected localization of p-Smad-2 given its role as a transcription factor. In contrast, the trophoblast layer of the fetal membranes in the preterm þ PPROM group exhibited decreased p-Smad-2 staining compared to the trophoblast layer of both the full-term controls and the preterm PPROM group (Fig. 4). In a manner similar to the full-term and preterm PPROM group, the amnion/chorion fibroblast layer of the preterm þ PPROM group displayed very little p-Smad-2 staining.
FIG. 3. A) Quantitative analysis of immunofluorescence showing a significant in decrease in decorin signal in preterm þ PPROM fetal membranes compared to preterm PPROM (P ¼ 0.0037) and full term and preterm PPROM (P ¼ 0.0091). There is no difference between full-term fetal membranes and the preterm þ PPROM group. Statistical analysis by ANOVA with Sidak post hoc t-test, n ¼ 20–25 per experimental group. Mean 6 SD. Error bars ¼ SD. PPROM ¼ preterm premature rupture of fetal membranes. B) Quantitative analysis of immunofluorescence showing a significant in decrease in decorin signal in preterm þ PPROM fetal membranes with infection compared to preterm þ PPROM without infection (P ¼ 0.0026). Statistical analysis by t-test, n ¼ 20–25 per experimental group. Mean 6 SD. Error bars ¼ SD. PPROM ¼ preterm premature rupture of fetal membranes. * ¼ statistical significance.
4
Article 105
Downloaded from www.biolreprod.org.
FIG. 2. A) Immunohistochemical analysis of decorin expression in fetal membranes at full term, preterm PPROM, and preterm þ PPROM. Decorin expression is decreased in fetal membranes with preterm þ PPROM. This decrease is significant when infection is present. Original magnification 340; bar ¼ 20 lm. B) Schematic of amnion/chorion fibroblast layer used for quantitative analysis of decorin staining, excluding the epithelial layer of amnion and the trophoblast layer.
ALTERED DECORIN IN HUMAN PPROM
DISCUSSION
the assembly of the fetal membrane extracellular matrix [25, 26] as well as signaling pathways [27–30]. The increase in decorin during gestation that we observed, followed by the decrease in decorin during labor [17], may play a role in an increase in fetal membrane tensile strength necessary to maintain pregnancy followed by a decrease in tensile strength that then leads to either PPROM or appropriate rupture of membranes at term. Furthermore, significantly more of the preterm þ PPROM samples displayed signs of inflammation compared to the preterm PPROM and full-term samples. Infection alters important pathways in fetal membranes, such as MMP expression [31, 32]. While not much is known about the role of decorin in inflammation, its homologue, biglycan, signals through toll-like receptors 2 and 4 and is thus a proinflammatory component of the toll-like receptor inflammatory cascade [18, 33]. Decorin displays 55% amino acid sequence homology with biglycan, most of the other amino acids are chemically similar substitutions, and both sport one (decorin) or two (biglycan) glycosaminoglycan chains [34]. Additionally, decorin and biglycan are functionally complementary in a mouse preterm birth model [7] and synergistic in mouse models of other diseases [9, 35]. Thus, decorin in fetal membranes may display similar characteristics to biglycan’s role in inflammation. On the other hand, biglycan and decorin are not uniformly synergistic functionally. In both the reproductive system [10, 22] and other organ systems [9, 36], both proteoglycans also display discrete roles. Also, our data demonstrate a decrease in decorin levels with inflammation, not an increase. Thus, an intriguing future direction is to interrogate the role that decorin plays in inflammatory cascades in fetal membranes and the reproductive system. This is especially important given the dichotomy of roles that decorin would play if it did display
In this study, we demonstrate that decorin is developmentally regulated in human fetal membranes throughout the course of pregnancy, that decorin expression is decreased in human fetal membranes after PPROM, and that this decrease is related to the presence of inflammation but is independent of a prior history of PPROM. Furthermore, we show that fetal membranes display decreased p-Smad-2 expression after PPROM. Much is known about the role of decorin in connective tissue systems, such as skin and bone [9], ligaments [19], tendon [20], and teeth [21]. We have studied the role of decorin during mouse gestation extensively [7, 10, 22]. However, this is the first study linking decorin to adverse outcomes in human pregnancies. Decorin is decreased in fetal membranes after normal-term vaginal birth compared to levels in prelabor Cesarean section samples [23]. This finding, coupled with our observation that decorin increases in fetal membranes during the course of pregnancy but is decreased in PPROM, suggests that decorin is necessary for the maintenance of fetal membrane integrity during pregnancy. Thus, decreased levels of decorin are associated with rupture of membranes at both term birth and preterm birth. The mechanism for this association may be multifaceted. Given decorin’s role in maintaining the mechanical stability of connective tissues such as tendons [24] or skin [9], there may be a direct relationship between decreasing decorin levels and membrane rupture. However, the relationship is likely more complex. For example, the decrease in decorin may be associated with the process of labor since the preterm þ PPROM samples were exposed to labor, while the preterm PPROM and term samples were not. Labor significantly changes the mechanical stability of fetal membranes at term, resulting in decreased tensile strength, and alters 5
Article 105
Downloaded from www.biolreprod.org.
FIG. 4. Immunohistochemical comparison of p-Smad-2 expression in full-term, preterm PPROM, and preterm þ PPROM fetal membranes. There is decreased decorin signal in the preterm þ PPROM group compared to the preterm PPROM group and the full-term group. P-Smad-2 staining was localized more to the trophoblast layer (thin arrows) than the fibroblast layer of the amnion/chorion (thick arrows). Original magnification 340; bar ¼ 20 lm. n ¼ 20–25 per experimental group. PPROM ¼ preterm premature rupture of fetal membranes.
HORGAN ET AL.
methylation in pre-eclamptic fetal membranes compared to full-term fetal membranes [44]. However, there is no data to suggest that pre-eclampsia leads to alterations in decorin levels in the fetal membranes. Thus, given the links between decorin, connective tissue mechanical strength, and PPROM, we hypothesize that it is in fact the fetal membranes with PPROM that have decreased decorin and p-Smad-2 levels as opposed to pre-eclampsia membranes displaying increased levels. In summary, our findings suggest that dysregulation of decorin occurs in fetal membranes during the second trimester in pathological pregnancies, thus supporting a role for decorin and downstream p-Smad-2 in the pathophysiology of fetal membranes and adverse pregnancy outcomes. Future steps might include interrogating decorin’s role in the TGF-b pathway by studying further downstream components and conducting larger studies to allow for regression analysis to control for more clinical confounders.
proinflammatory characteristics similar to biglycan. In that case, the structurally stabilizing and thus pregnancy-lengthincreasing role of decorin would be pitted against the proinflammatory and thus pregnancy-length-decreasing role. Given the breadth of data on decorin’s role in increasing the tensile strength of connective tissues in various organs [37, 38], the decrease in fetal membrane decorin after birth at term [17], our data showing an association between inflammation and a decrease in decorin, and the lack of data on a proinflammatory signaling role, it is more likely that the developmental increase in decorin that we described in this study leads to an increase in fetal membrane tensile strength that contributes to the maintenance of gestation to term until the point within the full-term time time period at which a decrease in decorin is associated with labor, rupture of membranes, and birth. The exact causal and temporal relationship between decorin levels and gestational age within the full-term window of 36–40 wk and their interactions with the processes of labor, rupture of membranes, and birth constitute an important future direction of study. Also, even though the process of fetal membrane rupture displays similarities at term and preterm, there are differences in the physiology of membrane rupture at term versus the pathophysiology of PPROM [39]. Thus, decorin may play a different role in term fetal membranes than earlier during the course of gestation. The absence of decorin results in an alteration of the TGF-b pathway in fetal membranes in the mouse model [10]. Specifically, p-Smad-2 is decreased in decorin-deficient fetal membranes, and p-Smad-2 dysregulation is rescued by recombinant decorin in the fetal membrane cell culture model [10]. Here, we show a similar decrease in p-Smad-2 levels in human fetal membranes after preterm birth with PPROM (preterm þ PPROM group). The decrease in decorin is seen in the fibroblast layer of the amnion/chorion, and the decrease in p-Smad-2 is seen in the trophoblast layer, suggesting that the decreases in the two proteins are spatially and thus functionally discrete. However, because decorin displays much stronger staining in the amnion/chorion layer and p-Smad-2 displays stronger staining in the trophoblast layer, it is conceivable that the techniques utilized were not sensitive enough to measure smaller magnitudes of change in the expression of decorin in trophoblast and p-Smad-2 in amnion/chorion fibroblasts. Thus, while our data do not provide a causal link between the decrease in decorin levels and p-Smad-2 levels in human fetal membranes at PPROM, we suggest that the relationship likely is causal. The TGF-b-p-Smad-2 pathway regulates collagen, MMP, and TIMP production [40, 41], all of which impact the structural integrity of fetal membranes [16, 39, 42]. Thus, the decrease in decorin at preterm birth with PPROM (preterm þ PPROM group) and the associated decrease in a downstream component of the decorin-TGF-b pathway, p-Smad-2, suggests dysregulation of the pathway that leads to stabilization of the fetal membranes. A caveat to these findings is that the preterm PPROM fetal membranes used for age-based comparison are not normal, healthy samples. They are the membranes of pregnancies affected by pre-eclampsia, an endothelial cell disorder in which the intrauterine environment is clearly different from the PPROM intrauterine environment [43]. For obvious reasons, it is not possible to obtain healthy normal preterm fetal membranes. Thus, we recognize that the observed differences in preterm membranes could possibly be attributed to the effects of pre-eclampsia on decorin levels and that the preterm þ PPROM samples, in fact, may represent typical decorin levels at this age. A recent study showed differential
ACKNOWLEDGMENT We thank the Department of Pathology at Women & Infants Hospital for providing the fetal membrane samples.
1. Steer P. The epidemiology of preterm labour. Bjog 2005; 112(suppl 1): 1–3. 2. Practice bulletins No. 139: premature rupture of membranes. Obstet Gynecol 2013; 122:918–930. 3. Romero R, Mazor M, Munoz H, Gomez R, Galasso M, Sherer DM. The preterm labor syndrome. Ann N Y Acad Sci 1994; 734:414–429. 4. Barabas AP. Ehlers-Danlos syndrome: associated with prematurity and premature rupture of foetal membranes; possible increase in incidence. Br Med J 1966; 5515:682–684. 5. Yen JL, Lin SP, Chen MR, Niu DM. Clinical features of Ehlers-Danlos syndrome. J Formos Med Assoc 2006; 105:475–480. 6. Schaefer L, Iozzo RV. Biological functions of the small leucine-rich proteoglycans: from genetics to signal transduction. J Biol Chem 2008; 283:21305–21309. 7. Calmus ML, Macksoud EE, Tucker R, Iozzo RV, Lechner BE. A mouse model of spontaneous preterm birth based on the genetic ablation of biglycan and decorin. Reproduction 2011; 142:183–194. 8. Reed CC, Iozzo RV. The role of decorin in collagen fibrillogenesis and skin homeostasis. Glycoconj J 2002; 19:249–255. 9. Corsi A, Xu T, Chen XD, Boyde A, Liang J, Mankani M, Sommer B, Iozzo RV, Eichstetter I, Robey PG, Bianco P, Young MF. Phenotypic effects of biglycan deficiency are linked to collagen fibril abnormalities, are synergized by decorin deficiency, and mimic Ehlers-Danlos-like changes in bone and other connective tissues. J Bone Miner Res 2002; 17: 1180–1189. 10. Wu Z, Horgan CE, Carr O, Owens RT, Iozzo RV, Lechner BE. Biglycan and decorin differentially regulate signaling in the fetal membranes. Matrix Biol 2014; 35:266–275. 11. Hakkinen L, Strassburger S, Kahari VM, Scott PG, Eichstetter I, Lozzo RV, Larjava H. A role for decorin in the structural organization of periodontal ligament. Lab Invest 2000; 80:1869–1880. 12. Zhang G, Ezura Y, Chervoneva I, Robinson PS, Beason DP, Carine ET, Soslowsky LJ, Iozzo RV, Birk DE. Decorin regulates assembly of collagen fibrils and acquisition of biomechanical properties during tendon development. J Cell Biochem 2006; 98:1436–1449. 13. Kinsella MG, Bressler SL, Wight TN. The regulated synthesis of versican, decorin, and biglycan: extracellular matrix proteoglycans that influence cellular phenotype. Crit Rev Eukaryot Gene Expr 2004; 14:203–234. 14. Parry S, Strauss JF III. Premature rupture of the fetal membranes. N Engl J Med 1998; 338:663–670. 15. Menon R, Fortunato SJ. The role of matrix degrading enzymes and apoptosis in rupture of membranes. J Soc Gynecol Invest 2004; 11: 427–437. 16. Romero R, Friel LA, Velez Edwards DR, Kusanovic JP, Hassan SS, Mazaki-Tovi S, Vaisbuch E, Kim CJ, Erez O, Chaiworapongsa T, Pearce BD, Bartlett J, et al. A genetic association study of maternal and fetal candidate genes that predispose to preterm prelabor rupture of membranes (PROM). Am J Obstet Gynecol 2010; 203:361.e361–e361.e330. 17. Meinert M, Malmstrom A, Tufvesson E, Westergren-Thorsson G, Petersen
6
Article 105
Downloaded from www.biolreprod.org.
REFERENCES
ALTERED DECORIN IN HUMAN PPROM
18.
19. 20.
21.
22. 23.
24.
26.
27. 28. 29.
30.
31.
32.
33.
34.
35.
36. 37.
38. 39.
40.
41. 42.
43.
44.
7
Vadillo-Ortega F. Differential secretion of matrix metalloproteinase-2 and 9 after selective infection with group B streptococci in human fetal membranes. J Soc Gynecol Invest 2006; 13:271–279. Zaga-Clavellina V, Garcia-Lopez G, Flores-Pliego A, Merchant-Larios H, Vadillo-Ortega F. In vitro secretion and activity profiles of matrix metalloproteinases, MMP-9 and MMP-2, in human term extra-placental membranes after exposure to Escherichia coli. Reprod Biol Endocrinol 2011; 9:13. Moreth K, Frey H, Hubo M, Zeng-Brouwers J, Nastase MV, Hsieh LT, Haceni R, Pfeilschifter J, Iozzo RV, Schaefer L. Biglycan-triggered TLR2- and TLR-4-signaling exacerbates the pathophysiology of ischemic acute kidney injury. Matrix Biol 2014; 35:143–151. Fisher LW, Termine JD, Young MF. Deduced protein sequence of bone small proteoglycan I (biglycan) shows homology with proteoglycan II (decorin) and several nonconnective tissue proteins in a variety of species. J Biol Chem 1989; 264:4571–4576. Casar JC, McKechnie BA, Fallon JR, Young MF, Brandan E. Transient up-regulation of biglycan during skeletal muscle regeneration: delayed fiber growth along with decorin increase in biglycan-deficient mice. Dev Biol 2004; 268:358–371. Schonherr E, Sunderkotter C, Schaefer L, Thanos S, Grassel S, Oldberg A, Iozzo RV, Young MF, Kresse H. Decorin deficiency leads to impaired angiogenesis in injured mouse cornea. J Vasc Res 2004; 41:499–508. Ameye L, Aria D, Jepsen K, Oldberg A, Xu T, Young MF. Abnormal collagen fibrils in tendons of biglycan/fibromodulin-deficient mice lead to gait impairment, ectopic ossification, and osteoarthritis. FASEB J 2002; 16:673–680. Chen XD, Shi S, Xu T, Robey PG, Young MF. Age-related osteoporosis in biglycan-deficient mice is related to defects in bone marrow stromal cells. J Bone Miner Res 2002; 17:331–340. Vadillo-Ortega F, Hernandez A, Gonzalez-Avila G, Bermejo L, Iwata K, Strauss JF III. Increased matrix metalloproteinase activity and reduced tissue inhibitor of metalloproteinases-1 levels in amniotic fluids from pregnancies complicated by premature rupture of membranes. Am J Obstet Gynecol 1996; 174:1371–1376. Verrecchia F, Chu ML, Mauviel A. Identification of novel TGF-beta / Smad gene targets in dermal fibroblasts using a combined cDNA microarray/promoter transactivation approach. J Biol Chem 2001; 276: 17058–17062. Kim HS, Luo L, Pflugfelder SC, Li DQ. Doxycycline inhibits TGF-beta1induced MMP-9 via Smad and MAPK pathways in human corneal epithelial cells. Invest Ophthalmol Vis Sci 2005; 46:840–848. Ferrand PE, Parry S, Sammel M, Macones GA, Kuivaniemi H, Romero R, Strauss JF III. A polymorphism in the matrix metalloproteinase-9 promoter is associated with increased risk of preterm premature rupture of membranes in African Americans. Mol Hum Reprod 2002; 8:494–501. Faupel-Badger JM, Fichorova RN, Allred EN, Hecht JL, Dammann O, Leviton A, McElrath TF. Cluster analysis of placental inflammatory proteins can distinguish preeclampsia from preterm labor and premature membrane rupture in singleton deliveries less than 28 weeks of gestation. Am J Reprod Immunol 2011; 66:488–494. Ching T, Song MA, Tiirikainen M, Molnar J, Berry M, Towner D, Garmire LX. Genome-wide hypermethylation coupled with promoter hypomethylation in the chorioamniotic membranes of early onset preeclampsia. Mol Hum Reprod 2014; 20:885–904.
Article 105
Downloaded from www.biolreprod.org.
25.
AC, Laurent C, Uldbjerg N, Eriksen GV. Labour induces increased concentrations of biglycan and hyaluronan in human fetal membranes. Placenta 2007; 28:482–486. Schaefer L, Babelova A, Kiss E, Hausser HJ, Baliova M, Krzyzankova M, Marsche G, Young MF, Mihalik D, Gotte M, Malle E, Schaefer RM, et al. The matrix component biglycan is proinflammatory and signals through Toll-like receptors 4 and 2 in macrophages. J Clin Invest 2005; 115: 2223–2233. Kavanagh E, Ashhurst DE. Distribution of biglycan and decorin in collateral and cruciate ligaments and menisci of the rabbit knee joint. J Histochem Cytochem 2001; 49:877–885. Dunkman AA, Buckley MR, Mienaltowski MJ, Adams SM, Thomas SJ, Kumar A, Beason DP, Iozzo RV, Birk DE, Soslowsky LJ. The injury response of aged tendons in the absence of biglycan and decorin. Matrix Biol 2014; 35:232–238. Haruyama N, Sreenath TL, Suzuki S, Yao X, Wang Z, Wang Y, Honeycutt C, Iozzo RV, Young MF, Kulkarni AB. Genetic evidence for key roles of decorin and biglycan in dentin mineralization. Matrix Biol 2009; 28:129–136. Wu Z, Aron AW, Macksoud EE, Iozzo RV, Hai CM, Lechner BE. Uterine dysfunction in biglycan and decorin deficient mice leads to dystocia during parturition. PLoS One 2012; 7:e29627. Meinert M, Malmstrom A, Tufvesson E, Westergren-Thorsson G, Petersen AC, Laurent C, Uldbjerg N, Eriksen GV. Labour induces increased concentrations of biglycan and hyaluronan in human fetal membranes. Placenta 2007; 28:482–486. Dunkman AA, Buckley MR, Mienaltowski MJ, Adams SM, Thomas SJ, Satchell L, Kumar A, Pathmanathan L, Beason DP, Iozzo RV, Birk DE, Soslowsky LJ. The tendon injury response is influenced by decorin and biglycan. Ann Biomed Eng 2014; 42:619–630. Estrada-Gutierrez G, Zaga V, Gonzalez-Jimenez MA, Beltran-Montoya J, Maida-Claros R, Giono-Cerezo S, Vadillo-Ortega F. Initial characterization of the microenvironment that regulates connective tissue degradation in amniochorion during normal human labor. Matrix Biol 2005; 24: 306–312. Vadillo-Ortega F, Gonzalez-Avila G, Furth EE, Lei H, Muschel RJ, Stetler-Stevenson WG, Strauss JF III. 92-kd type IV collagenase (matrix metalloproteinase-9) activity in human amniochorion increases with labor. Am J Pathol 1995; 146:148–156. Lappas M. NOD1 and NOD2 regulate proinflammatory and prolabor mediators in human fetal membranes and myometrium via nuclear factorkappa B. Biol Reprod 2013; 89:14. Lim R, Barker G, Lappas M. SLIT3 is Increased in supracervical human foetal membranes and in labouring myometrium and regulates proinflammatory mediators. Am J Reprod Immunol 2014; 71:297–311. Chai M, Walker SP, Riley C, Rice GE, Permezel M, Lappas M. Effect of supracervical apposition and spontaneous labour on apoptosis and matrix metalloproteinases in human fetal membranes. Biomed Res Int 2013; 2013:316146. Nhan-Chang CL, Romero R, Tarca AL, Mittal P, Kusanovic JP, Erez O, Mazaki-Tovi S, Chaiworapongsa T, Hotra J, Than NG, Kim JS, Hassan SS, et al. Characterization of the transcriptome of chorioamniotic membranes at the site of rupture in spontaneous labor at term. Am J Obstet Gynecol 2010; 202:462.e1–462.e41. Zaga-Clavellina V, Merchant-Larios H, Garcia-Lopez G, Maida-Claros R,
Altered Decorin and Smad Expression in Human Fetal Membranes in PPROM1 Casie E. Horgan,3,4 Hailey Roumimper,4,5 Richard Tucker,4 and Beatrice E. Lechner2,4 4
Department of Pediatrics, Women and Infants’ Hospital of Rhode Island, The Warren Alpert Medical School of Brown University, Providence, Rhode Island 5 Brown University, Providence, Rhode Island
Humans with Ehlers-Danlos syndrome, a subtype of which is caused by abnormal decorin expression, are at increased risk of preterm birth due to preterm premature rupture of fetal membranes (PPROM). In the mouse model, the absence of decorin leads to fetal membrane abnormalities, preterm birth, and dysregulation of decorin’s downstream pathway components, including the transcription factor p-Smad-2. However, the role of decorin and p-Smad-2 in idiopathic human PPROM is unknown. Fetal membranes from 20–25 pregnancies per group were obtained as a cross-sectional sample of births at one institution between January 2010 and December 2012. The groups were term, preterm without PPROM, and preterm with PPROM. Immunohistochemical analysis of fetal membranes was performed for decorin and p-Smad-2 using localization and quantification assessment. Decorin expression is developmentally regulated in fetal membranes and is decreased in preterm birth with PPROM compared to preterm birth without PPROM. In preterm with PPROM samples, the presence of infection is associated with significant decorin downregulation compared to preterm with PPROM samples without infection. The preterm with PPROM group exhibited decreased p-Smad-2 staining compared to both the term controls and the preterm-withoutPPROM group. Our findings suggest that dysregulation of decorin and its downstream pathway component p-Smad-2 occurs in fetal membranes during the second trimester in pathological pregnancies, thus supporting a role for decorin and p-Smad-2 in the pathophysiology of fetal membranes and adverse pregnancy outcomes. These findings may lead to the discovery of new targets for the diagnosis and treatment of PPROM. decorin, fetal membranes, PPROM, preterm birth, Smad
INTRODUCTION A leading cause of infant mortality is preterm birth—40% of which is caused by preterm premature rupture of fetal membranes (PPROM) [1, 2]. While various factors, such as presence of infection, a previous history of PPROM, and sociodemographic factors, have been identified as contributors, the pathophysiology of PPROM and thus screening or treatment strategies have remained elusive [3].
MATERIALS AND METHODS 1
Supported by NIGMS P20GM103537 and NICHD 1K08HD054676. 2 Correspondence: Beatrice E. Lechner, Women & Infants Hospital of Rhode Island, 101 Dudley St., Providence, RI 02905. E-mail: [email protected] 3 Current address: Tufts University School of Public Health, Boston, Massachusetts.
Study Patients and Tissue Collection Paraffin-embedded fetal membrane block samples stored in the Women & Infants pathology tissue repository were collected retrospectively from patients who had delivered at Women & Infants Hospital of Rhode Island between January 2010 and December 2012. Fetal membranes from three clinical groups were designated to experimental groups: 1) term births (37 wk gestational age), subsequently referred to as full term; 2) preterm births with no PPROM (36 wk gestational age), subsequently referred to as preterm PPROM; and 3) preterm birth with PPROM (36 wk gestational age), subsequently referred to as preterm þ PPROM. The first 25 samples to meet inclusion criteria in reverse chronological order were selected from each group. Term births (full
Received: 8 May 2014. First decision: 12 June 2014. Accepted: 4 September 2014. Ó 2014 by the Society for the Study of Reproduction, Inc. eISSN: 1529-7268 http://www.biolreprod.org ISSN: 0006-3363
1
Article 105
Downloaded from www.biolreprod.org.
Infants born with Ehlers-Danlos syndrome, an inheritable connective tissue disorder, have an increased risk of preterm birth via PPROM compared to their unaffected siblings [4, 5]. A subset of Ehlers-Danlos syndrome, the progeroid variant, is caused by a lack of posttranslational glycosylation of the small leucine-rich proteoglycan decorin and its homologue, biglycan [4, 5]. Decorin is located in the extracellular matrix and is highly expressed in reproductive tissues [6, 7]. Various roles for decorin have been identified in collagen fibrillogenesis and cell growth [8]. Mice deficient in decorin and biglycan display connective tissue abnormalities resembling Ehlers-Danlos syndrome [9]. We have shown that these mice also display preterm birth and fetal membrane anomalies as well as fetal loss and low birth weight [7, 10]. In addition to its structural role in the extracellular matrix [11, 12], decorin binds transforming growth factor b (TGF-b) and affects the TGF-b signaling pathway (reviewed in Kinsella et al. [13]). In this signaling pathway, the extracellular protein TGF-b binds its membrane receptor. Receptor binding leads to the phosphorylation of the Smad family of transcription factors, which then travel to the nucleus to regulate transcription of various collagens, matrix metalloproteinases (MMPs), and tissue inhibitors of metalloproteinases (TIMPs) [13]. These three groups of proteins play a central role in the pathogenesis of PPROM. For example, increased expression of matrix metalloproteinases in fetal membranes leads to PPROM via the degradation of the extracellular matrix [14, 15], while mutations in TIMPs and collagens increase the risk of PPROM [16]. We have previously shown that the TGF-b-Smad signaling pathway is dysregulated in mouse fetal membranes in the absence of decorin and that recombinant decorin rescues the phenotype [10]. Decorin decreases in fetal membranes at birth in term deliveries [17]. Its homologue, biglycan, plays a role in inflammation [18]. However, although our data in mice suggest a role for decorin in the maintenance of intact fetal membranes, little is known about the role of decorin during human pregnancy, and nothing is known about its role in preterm birth and PPROM. Thus, our objective was to study the regulation of decorin and downstream targets of the TGF-b pathway in human fetal membranes.
ABSTRACT
HORGAN ET AL. then used as a marker of decorin expression for statistical analysis given that they were perceived to be biologically more meaningful than maximum fluorescence values. A representative region of the tissue was selected for photo analysis. All adjustments to contrast were linear and were adjusted uniformly for all images in the data set. For the p-Smad-2 quantitative analysis, slides were assessed for localization, intensity, and pattern of staining by two independent and blinded observers.
term) that functioned as a control group were included if no labor had taken place, there were no microscopic signs of infection or clinical history of chorioamnionitis, and there was no history of prior preterm birth or PPROM. Thus, all term births were scheduled Cesarean sections. For the preterm þ PPROM group, all cases within the study time frame with a history of multiple PPROM were chosen to increase the probability of pathology. Next, of the remaining PPROM cases within the time frame, all cases of PPROM without clinical or microscopic signs of infection were chosen. Finally, the remainder of the samples needed for a total of 25 samples were chosen in reverse chronological order among the remaining cases of PPROM within the study time frame that did display signs of infection. For fetal membranes in the preterm PPROM group, which served as an age-matched control group for the preterm þ PPROM group, all samples chosen were delivered by Cesarean section to avoid exposure to labor. For this group, also, all cases without clinical or microscopic signs of infection were chosen within the study time frame. Next, the remainder of the samples needed for a total of 25 samples were chosen in reverse chronological order among the remaining cases within the study time frame that did display signs of infection. The most common reason for preterm birth in the absence of labor or PPROM is pre-eclampsia. Thus, all samples in the preterm PPROM group came from pregnancies complicated by pre-eclampsia. On microscopic analysis of the groups, the number of samples in the preterm þ PPROM group was reduced to 20 because five samples had to be excluded, as they did not contain any fetal membrane tissue. All samples were assigned a deidentified patient code. The following clinical information on the mother and infant was collected: gestational age, birth weight, gender, parity, mode of delivery, attempted induction, presence of infection (chorioamnionitis), gestational age at PPROM, and history of prior PPROM.
Statistical Analyses Statistical analysis of immunofluorescence across preterm groups was done by ANOVA with Sidak post hoc t-tests. Comparison of immunofluorescence by presence of infection within preterm groups was done by t-test. Analysis controlling for gestational age differences was done by linear regression. To achieve this, we compared decorin levels of both preterm groups to the term group and specified a contrast test between preterm PPROM and preterm þ PPROM, using a regression model with gestational age as a continuous covariate, indicators for each preterm group (with term as reference), and decorin level as the dependent variable. Age was not adjusted for in the developmental regulation experiment since age-dependent decorin expression was the outcome we were evaluating. Thus, in that experiment, age was the independent variable, and decorin fluorescence level was the dependent variable.
RESULTS
Ethical Approval Institutional review board approval was obtained for the study of human tissues. All experiments were conducted in accordance with the Society for the Study of Reproduction’s specific guidelines and standards.
Antibodies The following primary antibodies were used: monoclonal mouse antihuman decorin (R&D Systems); polyclonal rabbit anti-human Smad2, phospho-specific, and ser465/467 (EMD Millipore). Secondary antibody labeling was performed using donkey anti-mouse IgG conjugated to Alexa 488 (Invitrogen) for decorin and goat anti-rabbit IgG conjugated to CY3 (Invitrogen) for p-Smad-2. Mouse IgG and rabbit IgG (Vector Laboratories), respectively, were used as controls.
Histology and Immunohistochemistry Paraffin-embedded fetal membrane blocks for each of the 75 patients (25 per group) were sectioned, mounted on slides, and labeled in a deidentified manner. Slides were baked for 2 h at 658C, then deparaffinized with Richard Allen Scientific Clear-Rite-3 (Thermo Scientific) and rehydrated. Antigen retrieval was then performed using Citrate Buffer Antigen Retriever (SigmaAldrich). Slides were blocked with rabbit, goat, or donkey blocking serum (Sigma-Aldrich) at a 1:20 dilution in PBS (Fisher Scientific) for 1 h at room temperature. Slides were incubated overnight at 48C with primary antibody. The following morning, slides were incubated for 30 min at room temperature with secondary antibody and then mounted using Vectashield with DAPI staining (Vector Laboratories).
Image Analysis Fluorescent microscopy was performed to evaluate the samples using an inverted stage Nikon Eclipse TE2000-E microscope equipped with epifluorescent filters and a Nikon Plan Apo 206 and 406 and a Plan Fluor 106 objective lens with 403 magnification. Images were acquired using a Coolsnap HQ cooled CCD camera (Roper Scientific) and MetaVue software (Molecular Devices). Data sets were typically collected in one sitting to minimize variation. Exposure time and all acquisition settings were held constant for each fluorescence channel. Acquisition settings were chosen to minimize image saturation. For the quantitative analysis of decorin staining, the stromal layers of the fetal membranes were traced using the MetaVue tracing tool, including the fibroblast layers of the amnion and chorion, excluding the epithelial cells, trophoblasts, and any space where the stromal layers had torn from analysis. Average and maximum fluorescence intensity (12-bit scale) were collected for the traced regions, excluding all other regions, using the MetaVue average and maximum fluorescence measurement tool. Average fluorescence values were
2
Article 105
Downloaded from www.biolreprod.org.
Demographic characteristics of the three groups of patients (full term, preterm PPROM, and preterm þ PPROM) are shown in Table 1. Per study design, there were significant differences in gestational age between the full-term group and the two preterm groups. There was also a difference in gestational age between the preterm þ PPROM group and the preterm PPROM group, which was subsequently adjusted for in the data analysis. Of the term samples, none had evidence of infection. In the preterm PPROM group, 4% of the samples showed signs of chorioamnionitis. In the preterm þ PPROM group, however, 60% of samples showed signs of chorioamnionitis. Additionally, 100% of the preterm PPROM group displayed pre-eclampsia, while none of the other groups did. First, we assessed decorin expression in fetal membranes by immunohistochemistry as a function of gestational age. To achieve this, we compared decorin levels of both preterm groups (preterm PPROM and preterm þ PPROM) via regression analysis with age as the independent variable and decorin level as the dependent variable. We observed that decorin staining is developmentally regulated in fetal membranes obtained via preterm birth secondary to either PPROM or pre-eclampsia throughout the course of gestation, with an increase in decorin expression as the pregnancy progresses (P ¼ 0.0225). This developmental regulation was observed regardless of whether the samples were fetal membranes from preterm þ PPROM (Fig. 1A) or from preterm births PPROM (Fig. 1B). At full term (37–39 wk), there is a trend toward an increase in decorin expression to a peak at 38 wk followed by a decrease but no statistically significant difference (Supplemental Figure S1; all Supplemental Data are available online at www.biolreprod.org). Next, immunohistochemical analysis of decorin expression comparing the three groups (full term, preterm þ PPROM, and preterm PPROM) was performed, and the results were quantified. Decorin expression was decreased in the preterm þ PPROM group compared to preterm PPROM fetal membranes (Fig. 2A). This decrease in decorin expression, however, was observed only in preterm þ PPROM fetal membranes in the presence of infection but not in the absence of infection (Fig. 2A). The majority of the decorin staining was located in stromal (fibroblast) layers of the amnion and chorion, with less staining in the epithelial layer of the amnion or the trophoblast layer (schematic of localization in Fig. 2B).
ALTERED DECORIN IN HUMAN PPROM TABLE 1. Birth characteristics of fetal membrane samples, stratified by clinical group. Characteristic
Full term (n ¼ 25)
Preterm PPROM (n ¼ 25)
Preterm þ PPROM (n ¼ 20)
Mean gestational age (SD) Birth weight (mean) Gender (% male) Parity (SD) Mode of delivery (% Cesarean section) Attempted induction (%) Chorioamnionitis (%) Mean gestational age at PPROM (SD) Prior PPROM (%)
38.6 (1.0) 3614 g 52 1.3 (0.6) 100 0 0 — 0
30.3 (2.7) 1400 g 56 0.28 (0.6) 100 44 4 — 0
28.9 (6.2) 1411 g 52 1.2 (1.4) 28 52 60 27.3 (5.8) 52
Pre-eclampsia (%) Severe pre-eclampsia (%) Superimposed severe pre-eclampsia (%) Early-onset pre-eclampsia (%)
0 0 0 0
100 92 4 92
0 0 0 0
FIG. 1. Decorin expression in human fetal membranes during the course of gestation. A) Immunohistochemical staining of decorin increases during the course of gestation in fetal membranes after PPROM between 16 and 34 wk of gestation. B) Immunohistochemical staining of decorin increases during the course of gestation in fetal membranes without PPROM between 26 and 31 wk of gestation. Original magnification 340; bar ¼ 20 lm.
3
Article 105
Downloaded from www.biolreprod.org.
given the difference in group gestational ages (P ¼ 0.0037; Fig. 3A). Next, we measured decorin levels in fetal membranes in the preterm þ PPROM group, dividing this group into those with a history of prior PPROM and those without a history of prior PPROM, for whom this was their first episode of PPROM. There was no difference in decorin levels between the two PPROM groups (data not shown).
Thus, decorin staining in this area of the amnion and chorion but not in the epithelial layer or trophoblast layer was quantified according to lines drawn as demonstrated in Figure 2B. Decorin levels in the term group did not significantly differ from the preterm þ PPROM group. However, significantly decreased levels of decorin were present in the preterm birth þ PPROM group compared to the preterm PPROM group, which was still significant when adjusted for gestational age
HORGAN ET AL.
Inflammation, on the other hand, appears to play a role in the differences in the level of decorin expression between the three groups. In the full-term birth group, there were no samples with infection, as defined by histological analysis of the samples or clinical criteria. In both the preterm þ PPROM and the preterm PPROM groups, there were samples with infection and samples without infection. When infection was adjusted for in the preterm þ PPROM group and the preterm PPROM group via regression, no significant differences were observed between the two preterm groups or between all three groups (full term, preterm PPROM, and preterm þ PPROM). We then compared the preterm þ PPROM samples with infection with the preterm þ PPROM samples without infection and observed significant differences in decorin levels. In preterm þ PPROM samples with infection, decorin was significantly downregulated when compared to preterm þ PPROM samples without infection (P ¼ 0.0026; Fig. 3B).
P-Smad-2 expression and localization in fetal membranes was analyzed using immunohistochemistry. In full-term fetal membranes as well as preterm PPROM fetal membranes, the p-Smad-2 staining was localized almost exclusively to the trophoblast layer. In contrast, the amnion/chorion fibroblast layer displayed very little p-Smad-2 staining in the full-term and preterm PPROM samples. In the trophoblast of the fullterm and preterm PPROM samples, the trophoblast staining was nuclear and consisted of dots of staining. This is the expected localization of p-Smad-2 given its role as a transcription factor. In contrast, the trophoblast layer of the fetal membranes in the preterm þ PPROM group exhibited decreased p-Smad-2 staining compared to the trophoblast layer of both the full-term controls and the preterm PPROM group (Fig. 4). In a manner similar to the full-term and preterm PPROM group, the amnion/chorion fibroblast layer of the preterm þ PPROM group displayed very little p-Smad-2 staining.
FIG. 3. A) Quantitative analysis of immunofluorescence showing a significant in decrease in decorin signal in preterm þ PPROM fetal membranes compared to preterm PPROM (P ¼ 0.0037) and full term and preterm PPROM (P ¼ 0.0091). There is no difference between full-term fetal membranes and the preterm þ PPROM group. Statistical analysis by ANOVA with Sidak post hoc t-test, n ¼ 20–25 per experimental group. Mean 6 SD. Error bars ¼ SD. PPROM ¼ preterm premature rupture of fetal membranes. B) Quantitative analysis of immunofluorescence showing a significant in decrease in decorin signal in preterm þ PPROM fetal membranes with infection compared to preterm þ PPROM without infection (P ¼ 0.0026). Statistical analysis by t-test, n ¼ 20–25 per experimental group. Mean 6 SD. Error bars ¼ SD. PPROM ¼ preterm premature rupture of fetal membranes. * ¼ statistical significance.
4
Article 105
Downloaded from www.biolreprod.org.
FIG. 2. A) Immunohistochemical analysis of decorin expression in fetal membranes at full term, preterm PPROM, and preterm þ PPROM. Decorin expression is decreased in fetal membranes with preterm þ PPROM. This decrease is significant when infection is present. Original magnification 340; bar ¼ 20 lm. B) Schematic of amnion/chorion fibroblast layer used for quantitative analysis of decorin staining, excluding the epithelial layer of amnion and the trophoblast layer.
ALTERED DECORIN IN HUMAN PPROM
DISCUSSION
the assembly of the fetal membrane extracellular matrix [25, 26] as well as signaling pathways [27–30]. The increase in decorin during gestation that we observed, followed by the decrease in decorin during labor [17], may play a role in an increase in fetal membrane tensile strength necessary to maintain pregnancy followed by a decrease in tensile strength that then leads to either PPROM or appropriate rupture of membranes at term. Furthermore, significantly more of the preterm þ PPROM samples displayed signs of inflammation compared to the preterm PPROM and full-term samples. Infection alters important pathways in fetal membranes, such as MMP expression [31, 32]. While not much is known about the role of decorin in inflammation, its homologue, biglycan, signals through toll-like receptors 2 and 4 and is thus a proinflammatory component of the toll-like receptor inflammatory cascade [18, 33]. Decorin displays 55% amino acid sequence homology with biglycan, most of the other amino acids are chemically similar substitutions, and both sport one (decorin) or two (biglycan) glycosaminoglycan chains [34]. Additionally, decorin and biglycan are functionally complementary in a mouse preterm birth model [7] and synergistic in mouse models of other diseases [9, 35]. Thus, decorin in fetal membranes may display similar characteristics to biglycan’s role in inflammation. On the other hand, biglycan and decorin are not uniformly synergistic functionally. In both the reproductive system [10, 22] and other organ systems [9, 36], both proteoglycans also display discrete roles. Also, our data demonstrate a decrease in decorin levels with inflammation, not an increase. Thus, an intriguing future direction is to interrogate the role that decorin plays in inflammatory cascades in fetal membranes and the reproductive system. This is especially important given the dichotomy of roles that decorin would play if it did display
In this study, we demonstrate that decorin is developmentally regulated in human fetal membranes throughout the course of pregnancy, that decorin expression is decreased in human fetal membranes after PPROM, and that this decrease is related to the presence of inflammation but is independent of a prior history of PPROM. Furthermore, we show that fetal membranes display decreased p-Smad-2 expression after PPROM. Much is known about the role of decorin in connective tissue systems, such as skin and bone [9], ligaments [19], tendon [20], and teeth [21]. We have studied the role of decorin during mouse gestation extensively [7, 10, 22]. However, this is the first study linking decorin to adverse outcomes in human pregnancies. Decorin is decreased in fetal membranes after normal-term vaginal birth compared to levels in prelabor Cesarean section samples [23]. This finding, coupled with our observation that decorin increases in fetal membranes during the course of pregnancy but is decreased in PPROM, suggests that decorin is necessary for the maintenance of fetal membrane integrity during pregnancy. Thus, decreased levels of decorin are associated with rupture of membranes at both term birth and preterm birth. The mechanism for this association may be multifaceted. Given decorin’s role in maintaining the mechanical stability of connective tissues such as tendons [24] or skin [9], there may be a direct relationship between decreasing decorin levels and membrane rupture. However, the relationship is likely more complex. For example, the decrease in decorin may be associated with the process of labor since the preterm þ PPROM samples were exposed to labor, while the preterm PPROM and term samples were not. Labor significantly changes the mechanical stability of fetal membranes at term, resulting in decreased tensile strength, and alters 5
Article 105
Downloaded from www.biolreprod.org.
FIG. 4. Immunohistochemical comparison of p-Smad-2 expression in full-term, preterm PPROM, and preterm þ PPROM fetal membranes. There is decreased decorin signal in the preterm þ PPROM group compared to the preterm PPROM group and the full-term group. P-Smad-2 staining was localized more to the trophoblast layer (thin arrows) than the fibroblast layer of the amnion/chorion (thick arrows). Original magnification 340; bar ¼ 20 lm. n ¼ 20–25 per experimental group. PPROM ¼ preterm premature rupture of fetal membranes.
HORGAN ET AL.
methylation in pre-eclamptic fetal membranes compared to full-term fetal membranes [44]. However, there is no data to suggest that pre-eclampsia leads to alterations in decorin levels in the fetal membranes. Thus, given the links between decorin, connective tissue mechanical strength, and PPROM, we hypothesize that it is in fact the fetal membranes with PPROM that have decreased decorin and p-Smad-2 levels as opposed to pre-eclampsia membranes displaying increased levels. In summary, our findings suggest that dysregulation of decorin occurs in fetal membranes during the second trimester in pathological pregnancies, thus supporting a role for decorin and downstream p-Smad-2 in the pathophysiology of fetal membranes and adverse pregnancy outcomes. Future steps might include interrogating decorin’s role in the TGF-b pathway by studying further downstream components and conducting larger studies to allow for regression analysis to control for more clinical confounders.
proinflammatory characteristics similar to biglycan. In that case, the structurally stabilizing and thus pregnancy-lengthincreasing role of decorin would be pitted against the proinflammatory and thus pregnancy-length-decreasing role. Given the breadth of data on decorin’s role in increasing the tensile strength of connective tissues in various organs [37, 38], the decrease in fetal membrane decorin after birth at term [17], our data showing an association between inflammation and a decrease in decorin, and the lack of data on a proinflammatory signaling role, it is more likely that the developmental increase in decorin that we described in this study leads to an increase in fetal membrane tensile strength that contributes to the maintenance of gestation to term until the point within the full-term time time period at which a decrease in decorin is associated with labor, rupture of membranes, and birth. The exact causal and temporal relationship between decorin levels and gestational age within the full-term window of 36–40 wk and their interactions with the processes of labor, rupture of membranes, and birth constitute an important future direction of study. Also, even though the process of fetal membrane rupture displays similarities at term and preterm, there are differences in the physiology of membrane rupture at term versus the pathophysiology of PPROM [39]. Thus, decorin may play a different role in term fetal membranes than earlier during the course of gestation. The absence of decorin results in an alteration of the TGF-b pathway in fetal membranes in the mouse model [10]. Specifically, p-Smad-2 is decreased in decorin-deficient fetal membranes, and p-Smad-2 dysregulation is rescued by recombinant decorin in the fetal membrane cell culture model [10]. Here, we show a similar decrease in p-Smad-2 levels in human fetal membranes after preterm birth with PPROM (preterm þ PPROM group). The decrease in decorin is seen in the fibroblast layer of the amnion/chorion, and the decrease in p-Smad-2 is seen in the trophoblast layer, suggesting that the decreases in the two proteins are spatially and thus functionally discrete. However, because decorin displays much stronger staining in the amnion/chorion layer and p-Smad-2 displays stronger staining in the trophoblast layer, it is conceivable that the techniques utilized were not sensitive enough to measure smaller magnitudes of change in the expression of decorin in trophoblast and p-Smad-2 in amnion/chorion fibroblasts. Thus, while our data do not provide a causal link between the decrease in decorin levels and p-Smad-2 levels in human fetal membranes at PPROM, we suggest that the relationship likely is causal. The TGF-b-p-Smad-2 pathway regulates collagen, MMP, and TIMP production [40, 41], all of which impact the structural integrity of fetal membranes [16, 39, 42]. Thus, the decrease in decorin at preterm birth with PPROM (preterm þ PPROM group) and the associated decrease in a downstream component of the decorin-TGF-b pathway, p-Smad-2, suggests dysregulation of the pathway that leads to stabilization of the fetal membranes. A caveat to these findings is that the preterm PPROM fetal membranes used for age-based comparison are not normal, healthy samples. They are the membranes of pregnancies affected by pre-eclampsia, an endothelial cell disorder in which the intrauterine environment is clearly different from the PPROM intrauterine environment [43]. For obvious reasons, it is not possible to obtain healthy normal preterm fetal membranes. Thus, we recognize that the observed differences in preterm membranes could possibly be attributed to the effects of pre-eclampsia on decorin levels and that the preterm þ PPROM samples, in fact, may represent typical decorin levels at this age. A recent study showed differential
ACKNOWLEDGMENT We thank the Department of Pathology at Women & Infants Hospital for providing the fetal membrane samples.
1. Steer P. The epidemiology of preterm labour. Bjog 2005; 112(suppl 1): 1–3. 2. Practice bulletins No. 139: premature rupture of membranes. Obstet Gynecol 2013; 122:918–930. 3. Romero R, Mazor M, Munoz H, Gomez R, Galasso M, Sherer DM. The preterm labor syndrome. Ann N Y Acad Sci 1994; 734:414–429. 4. Barabas AP. Ehlers-Danlos syndrome: associated with prematurity and premature rupture of foetal membranes; possible increase in incidence. Br Med J 1966; 5515:682–684. 5. Yen JL, Lin SP, Chen MR, Niu DM. Clinical features of Ehlers-Danlos syndrome. J Formos Med Assoc 2006; 105:475–480. 6. Schaefer L, Iozzo RV. Biological functions of the small leucine-rich proteoglycans: from genetics to signal transduction. J Biol Chem 2008; 283:21305–21309. 7. Calmus ML, Macksoud EE, Tucker R, Iozzo RV, Lechner BE. A mouse model of spontaneous preterm birth based on the genetic ablation of biglycan and decorin. Reproduction 2011; 142:183–194. 8. Reed CC, Iozzo RV. The role of decorin in collagen fibrillogenesis and skin homeostasis. Glycoconj J 2002; 19:249–255. 9. Corsi A, Xu T, Chen XD, Boyde A, Liang J, Mankani M, Sommer B, Iozzo RV, Eichstetter I, Robey PG, Bianco P, Young MF. Phenotypic effects of biglycan deficiency are linked to collagen fibril abnormalities, are synergized by decorin deficiency, and mimic Ehlers-Danlos-like changes in bone and other connective tissues. J Bone Miner Res 2002; 17: 1180–1189. 10. Wu Z, Horgan CE, Carr O, Owens RT, Iozzo RV, Lechner BE. Biglycan and decorin differentially regulate signaling in the fetal membranes. Matrix Biol 2014; 35:266–275. 11. Hakkinen L, Strassburger S, Kahari VM, Scott PG, Eichstetter I, Lozzo RV, Larjava H. A role for decorin in the structural organization of periodontal ligament. Lab Invest 2000; 80:1869–1880. 12. Zhang G, Ezura Y, Chervoneva I, Robinson PS, Beason DP, Carine ET, Soslowsky LJ, Iozzo RV, Birk DE. Decorin regulates assembly of collagen fibrils and acquisition of biomechanical properties during tendon development. J Cell Biochem 2006; 98:1436–1449. 13. Kinsella MG, Bressler SL, Wight TN. The regulated synthesis of versican, decorin, and biglycan: extracellular matrix proteoglycans that influence cellular phenotype. Crit Rev Eukaryot Gene Expr 2004; 14:203–234. 14. Parry S, Strauss JF III. Premature rupture of the fetal membranes. N Engl J Med 1998; 338:663–670. 15. Menon R, Fortunato SJ. The role of matrix degrading enzymes and apoptosis in rupture of membranes. J Soc Gynecol Invest 2004; 11: 427–437. 16. Romero R, Friel LA, Velez Edwards DR, Kusanovic JP, Hassan SS, Mazaki-Tovi S, Vaisbuch E, Kim CJ, Erez O, Chaiworapongsa T, Pearce BD, Bartlett J, et al. A genetic association study of maternal and fetal candidate genes that predispose to preterm prelabor rupture of membranes (PROM). Am J Obstet Gynecol 2010; 203:361.e361–e361.e330. 17. Meinert M, Malmstrom A, Tufvesson E, Westergren-Thorsson G, Petersen
6
Article 105
Downloaded from www.biolreprod.org.
REFERENCES
ALTERED DECORIN IN HUMAN PPROM
18.
19. 20.
21.
22. 23.
24.
26.
27. 28. 29.
30.
31.
32.
33.
34.
35.
36. 37.
38. 39.
40.
41. 42.
43.
44.
7
Vadillo-Ortega F. Differential secretion of matrix metalloproteinase-2 and 9 after selective infection with group B streptococci in human fetal membranes. J Soc Gynecol Invest 2006; 13:271–279. Zaga-Clavellina V, Garcia-Lopez G, Flores-Pliego A, Merchant-Larios H, Vadillo-Ortega F. In vitro secretion and activity profiles of matrix metalloproteinases, MMP-9 and MMP-2, in human term extra-placental membranes after exposure to Escherichia coli. Reprod Biol Endocrinol 2011; 9:13. Moreth K, Frey H, Hubo M, Zeng-Brouwers J, Nastase MV, Hsieh LT, Haceni R, Pfeilschifter J, Iozzo RV, Schaefer L. Biglycan-triggered TLR2- and TLR-4-signaling exacerbates the pathophysiology of ischemic acute kidney injury. Matrix Biol 2014; 35:143–151. Fisher LW, Termine JD, Young MF. Deduced protein sequence of bone small proteoglycan I (biglycan) shows homology with proteoglycan II (decorin) and several nonconnective tissue proteins in a variety of species. J Biol Chem 1989; 264:4571–4576. Casar JC, McKechnie BA, Fallon JR, Young MF, Brandan E. Transient up-regulation of biglycan during skeletal muscle regeneration: delayed fiber growth along with decorin increase in biglycan-deficient mice. Dev Biol 2004; 268:358–371. Schonherr E, Sunderkotter C, Schaefer L, Thanos S, Grassel S, Oldberg A, Iozzo RV, Young MF, Kresse H. Decorin deficiency leads to impaired angiogenesis in injured mouse cornea. J Vasc Res 2004; 41:499–508. Ameye L, Aria D, Jepsen K, Oldberg A, Xu T, Young MF. Abnormal collagen fibrils in tendons of biglycan/fibromodulin-deficient mice lead to gait impairment, ectopic ossification, and osteoarthritis. FASEB J 2002; 16:673–680. Chen XD, Shi S, Xu T, Robey PG, Young MF. Age-related osteoporosis in biglycan-deficient mice is related to defects in bone marrow stromal cells. J Bone Miner Res 2002; 17:331–340. Vadillo-Ortega F, Hernandez A, Gonzalez-Avila G, Bermejo L, Iwata K, Strauss JF III. Increased matrix metalloproteinase activity and reduced tissue inhibitor of metalloproteinases-1 levels in amniotic fluids from pregnancies complicated by premature rupture of membranes. Am J Obstet Gynecol 1996; 174:1371–1376. Verrecchia F, Chu ML, Mauviel A. Identification of novel TGF-beta / Smad gene targets in dermal fibroblasts using a combined cDNA microarray/promoter transactivation approach. J Biol Chem 2001; 276: 17058–17062. Kim HS, Luo L, Pflugfelder SC, Li DQ. Doxycycline inhibits TGF-beta1induced MMP-9 via Smad and MAPK pathways in human corneal epithelial cells. Invest Ophthalmol Vis Sci 2005; 46:840–848. Ferrand PE, Parry S, Sammel M, Macones GA, Kuivaniemi H, Romero R, Strauss JF III. A polymorphism in the matrix metalloproteinase-9 promoter is associated with increased risk of preterm premature rupture of membranes in African Americans. Mol Hum Reprod 2002; 8:494–501. Faupel-Badger JM, Fichorova RN, Allred EN, Hecht JL, Dammann O, Leviton A, McElrath TF. Cluster analysis of placental inflammatory proteins can distinguish preeclampsia from preterm labor and premature membrane rupture in singleton deliveries less than 28 weeks of gestation. Am J Reprod Immunol 2011; 66:488–494. Ching T, Song MA, Tiirikainen M, Molnar J, Berry M, Towner D, Garmire LX. Genome-wide hypermethylation coupled with promoter hypomethylation in the chorioamniotic membranes of early onset preeclampsia. Mol Hum Reprod 2014; 20:885–904.
Article 105
Downloaded from www.biolreprod.org.
25.
AC, Laurent C, Uldbjerg N, Eriksen GV. Labour induces increased concentrations of biglycan and hyaluronan in human fetal membranes. Placenta 2007; 28:482–486. Schaefer L, Babelova A, Kiss E, Hausser HJ, Baliova M, Krzyzankova M, Marsche G, Young MF, Mihalik D, Gotte M, Malle E, Schaefer RM, et al. The matrix component biglycan is proinflammatory and signals through Toll-like receptors 4 and 2 in macrophages. J Clin Invest 2005; 115: 2223–2233. Kavanagh E, Ashhurst DE. Distribution of biglycan and decorin in collateral and cruciate ligaments and menisci of the rabbit knee joint. J Histochem Cytochem 2001; 49:877–885. Dunkman AA, Buckley MR, Mienaltowski MJ, Adams SM, Thomas SJ, Kumar A, Beason DP, Iozzo RV, Birk DE, Soslowsky LJ. The injury response of aged tendons in the absence of biglycan and decorin. Matrix Biol 2014; 35:232–238. Haruyama N, Sreenath TL, Suzuki S, Yao X, Wang Z, Wang Y, Honeycutt C, Iozzo RV, Young MF, Kulkarni AB. Genetic evidence for key roles of decorin and biglycan in dentin mineralization. Matrix Biol 2009; 28:129–136. Wu Z, Aron AW, Macksoud EE, Iozzo RV, Hai CM, Lechner BE. Uterine dysfunction in biglycan and decorin deficient mice leads to dystocia during parturition. PLoS One 2012; 7:e29627. Meinert M, Malmstrom A, Tufvesson E, Westergren-Thorsson G, Petersen AC, Laurent C, Uldbjerg N, Eriksen GV. Labour induces increased concentrations of biglycan and hyaluronan in human fetal membranes. Placenta 2007; 28:482–486. Dunkman AA, Buckley MR, Mienaltowski MJ, Adams SM, Thomas SJ, Satchell L, Kumar A, Pathmanathan L, Beason DP, Iozzo RV, Birk DE, Soslowsky LJ. The tendon injury response is influenced by decorin and biglycan. Ann Biomed Eng 2014; 42:619–630. Estrada-Gutierrez G, Zaga V, Gonzalez-Jimenez MA, Beltran-Montoya J, Maida-Claros R, Giono-Cerezo S, Vadillo-Ortega F. Initial characterization of the microenvironment that regulates connective tissue degradation in amniochorion during normal human labor. Matrix Biol 2005; 24: 306–312. Vadillo-Ortega F, Gonzalez-Avila G, Furth EE, Lei H, Muschel RJ, Stetler-Stevenson WG, Strauss JF III. 92-kd type IV collagenase (matrix metalloproteinase-9) activity in human amniochorion increases with labor. Am J Pathol 1995; 146:148–156. Lappas M. NOD1 and NOD2 regulate proinflammatory and prolabor mediators in human fetal membranes and myometrium via nuclear factorkappa B. Biol Reprod 2013; 89:14. Lim R, Barker G, Lappas M. SLIT3 is Increased in supracervical human foetal membranes and in labouring myometrium and regulates proinflammatory mediators. Am J Reprod Immunol 2014; 71:297–311. Chai M, Walker SP, Riley C, Rice GE, Permezel M, Lappas M. Effect of supracervical apposition and spontaneous labour on apoptosis and matrix metalloproteinases in human fetal membranes. Biomed Res Int 2013; 2013:316146. Nhan-Chang CL, Romero R, Tarca AL, Mittal P, Kusanovic JP, Erez O, Mazaki-Tovi S, Chaiworapongsa T, Hotra J, Than NG, Kim JS, Hassan SS, et al. Characterization of the transcriptome of chorioamniotic membranes at the site of rupture in spontaneous labor at term. Am J Obstet Gynecol 2010; 202:462.e1–462.e41. Zaga-Clavellina V, Merchant-Larios H, Garcia-Lopez G, Maida-Claros R,
