| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Obstetrics and Gynecology (B.C.T., M.S.M.) and Cell Biology (C.G.), University of Texas Southwestern Medical Center, Dallas, Texas 75390-9032; and Baylor College of Medicine (S.M.M.), Houston, Texas 77030
Address all correspondence and requests for reprints to: Mala Mahendroo, Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, Texas 75390-9032. E-mail: mala.mahendroo{at}utsouthwestern.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-reductase type 1 deficient mouse, which has an elevated local progesterone concentration, also has aberrant claudin 1 and 2 expressions, fails to form claudin 1-enriched TJs, and lacks normal expression of genes involved in epithelial terminal differentiation. These data suggest that changes in permeability barrier properties during cervical ripening are, in part, negatively regulated by progesterone, and that dynamic changes in barrier properties of the cervix occur during pregnancy and parturition. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Throughout pregnancy, the cervical epithelia must proliferate for sufficient growth and remodeling in preparation for parturition (4, 5, 6), maintain fluid balance during softening and ripening via regulated expression of aquaporin water channels (7), protect the cervical stroma and upper reproductive tract against invasion of microbes, and maintain a permeability barrier that regulates paracellular transport of solutes through the apical junctional complex (AJC) (8). During parturition and PP recovery, the vulnerable state of the ripe, open cervix must require increased regulatory mechanisms to ensure that the cervical epithelia can continue to maintain critical functions.
Epithelial cell sheets form a barrier that separates the internal environment from the external environment to maintain tissue homeostasis (9). For these cellular sheets to function as barriers, the intercellular route must be sealed, though individual cells within these sheets continue to move or rearrange as they undergo proliferation and differentiation. Cell-cell junctions termed the AJC mediate intercellular adhesion in mammalian epithelia. The most apical AJC component, the tight junction (TJ), is composed of transmembrane proteins that physically interact with opposing partners on adjacent cells, providing a mechanical link between two plasma membranes and establishing an effective paracellular barrier. TJs are dynamic structures that are disassembled and reassembled in a variety of epithelial remodeling processes (7, 8, 9, 10).
TJs are composed of integral membrane proteins such as occludin and claudins, and associated peripheral membrane proteins, such as zona occludens 1 and 2 (ZO-1 and ZO-2, respectively) (10, 11). Claudin proteins, of which there are 24 family members, vary in their tissue distribution and expression level. The composition of claudin proteins in a given cell type and tissue determines the "tightness" of the TJ and, therefore, the degree of sealing of the permeability barrier. Changes in the composition of junctional proteins in a given cell type can alter the barrier properties of that tissue. For example, mice deficient in claudin-1 die 1 d after birth from dehydration due to a loss of an intact epidermal barrier, despite the presence of claudin-4 positive TJs in the epithelia (12).
The mouse cervical epithelia are primarily stratified epithelia in contrast to the human, which contains both stratified and columnar epithelia (1, 13). The cervical stratified squamous epithelium is a continuously renewing tissue composed of cells at different stages of differentiation. In response to steroid hormones, the nonpregnant cervical stratified mucosal epithelium undergoes a cyclic change in proliferation and differentiation of cells, as well as mucosal composition and secretion (14). Upon leaving the basal layer, cells of stratified tissues down-regulate expression of specific keratins, and as the differentiating cells move into the upper layers, they induce expression of different keratins (15). As these cells progress further along the differentiation pathway, they express proteins involved in formation of the insoluble cornified envelope such as small-proline rich proteins (Sprr)-1b, involucrin, loricrin, and protease inhibitors that protect barrier proteins from degradation [e.g. serine protease inhibitor Kazal type 5 (Spink5)] (15, 16).
Barrier function within stratified epithelia requires both a cornified envelope with lipid lamellae in and between terminally differentiating epithelia, as well as tissue and cell-specific expression of TJ proteins within specific layers of the epithelium (12, 17). Recent evidence suggests that claudins play a role in both formation of an epidermal permeability barrier as well as in epithelial differentiation. Mice overexpressing claudin 6 in the suprabasal layers of the epidermis die shortly after birth due to barrier dysfunction that leads to transepidermal water loss and dehydration (17, 18). Loss of epidermal barrier function in these animals results in the aberrant expression of epidermal differentiation markers and the formation of fragile, defective cornified envelopes. Thus, the two systems for regulation of a permeability barrier within stratified epithelia, TJ composition and epithelial cell differentiation, are interregulated. Changes in the composition of TJ proteins and the pattern of epithelial cell differentiation based on expression of specific markers have not been evaluated in the cervix during pregnancy and parturition.
Epithelial cells lining the cervical canal are associated with each other by TJs, which regulate the passage of ions and molecules through the paracellular pathway. Steroid hormones, in part, regulate the paracellular transport of fluids in the human cervix, which is thought to influence the amount and composition of cervical mucus that is required for normal cervical function (19, 20). A decline in paracellular transport in postmenopausal women is associated with a reduction in cervical secretions and increased vaginal dryness (21). Both claudin 4 and occludin are expressed in human cervical epithelia, and a change in phosphorylation status of occludin regulates paracellular permeability (22). The presence and function of TJ proteins in control of paracellular transport within cervical and uterine epithelia has been described in the nonpregnant human and rat as well as during early pregnancy in the rat (19, 20, 22).
The current study explores the processes by which the cervical epithelial barrier is maintained, and perhaps enhanced during parturition and PP. Based on other studies that suggest permeability barrier function is modulated by TJ composition as well as differentiation (17, 18), the focus of the current study was to evaluate the expression and localization of TJ proteins, as well as markers of terminal differentiation in the cervix during pregnancy and parturition in normal mice and steroid 5-reductase deficient mice, in which 70% of pregnant mothers fail to deliver live young due to a defect in cervical ripening. Junction proteins and differentiation markers that have previously been described in the nonpregnant cervix (20, 22, 24, 25) or those identified as up or down-regulated genes by Affymetrix microarray analysis (Affymetrix, Inc., Santa Clara, CA) of pregnant cervix (data not shown) determined the choice of genes evaluated. These studies suggest that hormonally regulated changes in junctional protein composition, cell specific localization, as well as increased terminal differentiation of stratified epithelia occur during parturition and/or PP.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-Reductase type 1 Srd5a1 –/– deficient mice were generated and genotyped as described previously (25). Females housed overnight with males were checked at midday for vaginal plugs. Plug day was counted as d 0, and birth generally occurred on the morning of d 19 in wild-type (WT) mice but not Srd5a1 –/– mice who fail to deliver their young. In all cases, samples collected from WT females on d 19 were after birth had occurred and are labeled according to the hours PP (2, 10, or 24 h PP). Cervical tissues were routinely collected at midday for each desired time point. One exception is samples collected on gestation d 18. Samples labeled as d 18 were collected at midday, while samples collected after 1800 h were labeled as late d 18 or d 18.75. This distinction is important as many changes in gene expression are accelerated in the final hours before birth in the mouse. All studies were conducted upon approval by the University of Texas Southwestern Medical Center Institutional Animal Care and Research Advisory Committee.
Animals treated with progesterone were administered time-release progesterone pellets (50 mg pellet, 21-d release; Innovative Research of America, Sarasota, FL) on d 17. As previously described, this dose prevents onset of labor in WT mice (25). To confirm efficacy of progesterone treatment, progesterone levels were quantified in serum by RIA after chromatographic separation of steroids on Sephadex LH-20 columns, and the average progesterone concentration was determined to be 123 ± 27 ng/ml. The intraassay coefficient for this experiment was 6.4%. Cervices were collected from these animals on gestation d 18 as well as mice treated with a placebo pellet. Dr. David Hess performed steroid measurements at the Oregon Regional Primate Research Center (Beaverton, OR).
RNA measurements and quantitative real-time PCR
Total RNA was extracted from frozen mouse tissue using RNA Stat 60 (Tel-Test "B" Inc., Friendswood, TX). Subsequently, total RNA was treated with DNAse I to remove any genomic DNA using DNA-free (Ambion Inc., Austin, TX). cDNA synthesis was performed using 2 µg total RNA in a 100 µl volume (TaqMan cDNA synthesis kit; Applied Biosystems, Foster City, CA). Quantitative real-time PCR was performed using SYBR Green and a PRISM7900HT Sequence Detection System (Applied Biosystems). Aliquots (20 ng) of cDNA were used for each quantitative PCR reaction, and each reaction was run in triplicate. Relative gene expression between experimental groups was determined using the dd CT method, as described in User Bulletin No. 2 (Applied Biosystems). Cyclophilin B was used as the normalizer housekeeping gene.
Cervices of specific genotype and day of gestation were analyzed individually, and data are presented as the average relative gene expression ± SE of the means. Sample size for each experiment is indicated in the figure or figure legend.
Protein blotting
Mouse cervical extracts were prepared by homogenization in buffer (150 mM NaCl; 10 mM Tris, pH 7.2; 0.1% SDS; 1% Triton X-100; 1% deoxycholate; 5 mM EDTA) containing 1% protease inhibitor (catalog no. 2714; Sigma, St. Louis, MO). Thirty-two micrograms of protein was diluted in 2x Lammeli buffer (Bio-Rad, Hercules, CA), boiled for 5 min, and run on a 15% reducing Tris-HCl gel (Bio-Rad) along with protein size standards (Precision Plus Protein Kaleidoscope, catalog no. 161-0375; Bio-Rad). Proteins were transferred to a charged membrane (Immobilon-P; Millipore Corp., Bedford, MA). Nonspecific antibody binding was blocked by an overnight incubation with Tris-buffered saline with Tween 20 (TBST) [10 mm Tris (pH7.5), 150 mM NaCl, and 0.05% Tween 20] containing 3% nonfat dry milk. Blots were then incubated for 2 h with the primary antibody (1:1000) in blocking solution, washed in TBST, incubated with horseradish peroxidase-labeled antirabbit IgG (1:10,000) (catalog no. 711-036-152; Jackson Immunoresearch Laboratories, West Grove, PA) for 45 min, and washed again in TBST. Chemiluminescence detection was performed using the enhanced chemiluminescence detection system (Ambion, Inc.). Claudin 1 antibody was purchased from Zymed (San Francisco, CA; catalog no.71-7800). Polyclonal rabbit antimouse claudin 2 and 4 antibodies were a kind gift from Dr. Mikio Furuse (Kyoto University, Kyoto, Japan). A rabbit polyclonal anti-GRP78 antibody (catalog no. SPA-826; Stressgen Bioreagents, Victoria, Canada) was used as a loading control. This experiment was conducted using protein extracts from at least three animals per genotype and time point.
Immunofluorescence
Freshly excised cervices were embedded in OCT compound (Tissue Tek; Bayer Corp., Elhart, IN), frozen immediately in liquid nitrogen, and 8-µm sections were cut. Air-dried tissue sections were fixed for 10 min in ice-cold acetone. Nonspecific binding was blocked using 1.5% normal donkey serum (Jackson Immunoresearch Laboratories) in 1% BSA (Pierce Chemical, Rockford, IL), 0.1% gelatin from cold water fish skin (Sigma), 0.1% Triton X-100 (Sigma) in 0.1% PBS for 1 h at room temperature. Sections were incubated with primary antibody in 1% BSA in PBS for 1 h at room temperature. Primary antibodies include rabbit anti-claudin 1 and rabbit anti-claudin 2. Samples were washed three times for 5 min in PBS followed by 30 min at room temperature with fluorescein isothiocyanate conjugated donkey antirabbit antibody (1:200) (Jackson Immunoresearch Laboratories) in 1% BSA in PBS. Sections were washed, and coverslips were mounted with Prolong Gold antifade reagent (Invitrogen, Eugene, OR). Shown are representative photos from one of three cervices analyzed for each genotype/time point. One or two sections were analyzed for each animal.
Electron microscopy
Cervices were initially fixed in 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer. After a subsequent fixation in 1% osmium tetroxide in the same buffer, samples were dehydrated in ascending concentrations of ethanol, and infiltrated and embedded via a transitional solvent, propylene oxide, in Embed 812 epoxy resin (EMS, Hatfield, PA). Sections, 80-nm in thickness, were cut, picked up on copper grids, and stained with lead citrate and uranyl acetate before being examined in a JEOL 1200EX transmission electron microscope (JEOL USA, Peabody, MA). Shown are representative photos from one of three cervices analyzed for each genotype/time point. One or two sections were analyzed for each animal.
Statistics
For the data of Figs. 1
and 6, the hypothesis of equality of the means at several observation times was examined using ANOVA. If the ANOVA is statistically significant, then individual differences of observation times to the control time (18.75) are evaluated using Dunnett’s test for these pairwise multiple comparisons. The assumption of equality of variances among the days is examined using Levene’s test, with a Welch’s adjustment if this provides evidence of significant difference among variances. If the raw data do not pass Levene’s test, as in Fig. 6, then a logarithmic transformation was performed, and ANOVA was applied to the transformed data. Mean differences (and 95% confidence intervals) to the control are presented for each outcome. For Fig. 2, the experiment is conducted in a factorial design of two factors each at two levels: genotype (WT and knockout) and time (d 15 and 18.75). A two-factor ANOVA with interaction is used to evaluate these effects. In Figs. 7 and 8, differences in expression between two groups (e.g. comparison of WT and Srd5a1–/– samples) were determined using the Student’s t test for normally distributed data or the Mann-Whitney rank sum for data not normally distributed. P < 0.05 are considered statistically significant.
|
|
|
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Claudin 1 was expressed at a similar level from gestation d 10 onwards until late on gestation d 18, when there is a sharp increase in expression. Claudin 1 expression remains elevated 2 h and 24 h after birth. Claudin 2 was expressed at gestation d 10, and transcript abundance gradually declined through the rest of pregnancy, with a loss of expression by d 18.75 and in the PP period. Claudin 4 expression is constant throughout pregnancy from gestation d 10 until d 18. An increase in claudin 4 transcripts is detected in the PP period (2 h PP). Occludin transcripts are expressed during the latter half of gestation at a constant level and do not increase until after birth. Compared with d 18.75, the TJ associated proteins, ZO-1 and ZO-2, were expressed at a similar or lower level throughout the second half of gestation but significantly increased after birth.
The mRNA expression of claudin 1, claudin 2, and occludin was further analyzed in a cervical ripening defective mouse model, the steroid 5-reductase type 1 null mouse (Srd5a1–/–) (25). The absence of the Srd5a1 gene results in inadequate local progesterone metabolism, leading to elevated tissue progesterone concentration, failure of cervical ripening, and birth of young at term (26). Using real-time PCR, the expression of TJ genes was measured on d 15 and d 18.75 (Fig. 2). Claudin 1 mRNA expression is increased on late d 18 compared with d 15 in WT cervix. Although similar to WT on d 15, the expression of claudin 1 failed to increase in the Srd5a1–/– on gestation d 18.75, compared with the d 18.75 WT. In contrast, the expression of claudin 2 was markedly increased in Srd5a1–/– cervices at gestation d 15 and 18.75 in comparison to WT. Occludin expression did not change from d 15–18.75 in WT or Srd5a1–/– cervices. On gestation d 15, occludin expression was lower in the Srd5a1–/– cervix compared with WT; however, on d 18.75, the expression was similar. ZO-1 and ZO-2 expression in the cervical ripening defective Srd5a1–/– was similar to WT (data not shown).
To determine whether changes in mRNA expression for TJ proteins result in changes in protein expression, protein blotting was performed using cervices from WT mice at gestation d 15, d 18.75, 10 h PP, and d 18.75 cervix from the 5-reductase type 1 deficient mouse (Fig. 3). Compared with d 15, the protein expression for claudin 1 is increased on d 18.75 and 10 h PP in WT mice, but not in the d 18.75 Srd5a1–/– mouse, which is similar to the pattern of claudin 1 mRNA expression. Claudin 4 protein expression does not vary appreciably between time points. No difference in claudin 4 protein levels was detected between d 18.75 WT and 5-reductase type 1 deficient cervix. Protein blotting of claudin 2 using a polyclonal rabbit antimouse claudin 2 antibody was unsuccessful in our hands.
|
and 5). On d 15, claudin 1 is expressed within all epithelial layers and localized primarily to intracellular spaces, with little staining in the plasma membrane (Fig. 4). The cellular localization of claudin 1 is changed by late gestation d 18 (Fig. 4). In the most differentiated apical epithelia, claudin 1 is primarily localized to the epithelial plasma membranes in a typical lattice structure that is characteristic of TJs, and in the less differentiated epithelial layers, claudin 1 is primarily located intracellularly. In the 5-reductase type 1 deficient cervix on gestation d 15, claudin 1 localization is similar to d 15 WT cervix (Fig. 4). However, by late d 18, claudin 1 localization remains primarily intracellular, with little or no junctional staining in the plasma membrane of apical epithelia, as was observed in the WT cervix at this time point (Fig. 4).
|
|
). At d 15 and d 18, claudin 2 appears to be diffusely localized to the plasma membrane of all epithelial layers as well as distributed within the cytoplasm (Fig. 5). The claudin 2 plasma membrane staining is dissimilar to the junctional lattice type staining observed for claudin 1 in Fig. 4. The presence of immunodetectable claudin 2 in the late d 18 cervix, despite the decline in transcripts (Fig. 2), suggests that the protein has a relatively long half-life, similar to other junctional proteins (27, 28). Similar to WT, on gestation d 15 and d 18, cervices from the Srd5a1–/– mice express claudin 2 throughout the epithelia in both the intracellular space and diffusely along the plasma membrane.
Recent evidence supports a role of claudin proteins in regulation of epithelial cell differentiation, another important aspect to formation of a functional epithelial permeability barrier (12, 17). To understand better changes in epithelial differentiation during pregnancy and parturition, the mRNA expression of marker genes expressed by terminally differentiated cervical epithelia [keratin 16 (Kert 1–16), Sprr1b] was measured through the latter half of pregnancy (Fig. 6). In addition, other proteins associated with epithelial differentiation or function and identified as up-regulated genes in gestation d 18 cervices by Affymetrix microarray analysis were evaluated. These included keratin differentiation-associated protein (Krtdap) and fatty acid-binding protein 5 (keratin lipid-binding protein) (Fabp5) (29, 30, 31, 32). Expression of all four genes is low to undetectable until gestation d 18, when they increase severalfold. Maximal expression for all genes occurs 2 h after birth, and then expression falls by 24 h PP. Transcriptional activation of these genes is rapid because all four genes have low expression on gestation d 18 and high expression on d 18.75 (unpublished observation). This gene expression data indicate an activation of epithelial terminal differentiation pathways during cervical ripening that coincides with the timing of incorporation of claudin 1 protein within TJs in these terminally differentiated cells.
The observations in the current study, in which the pattern of claudin 1 and 2 expression as well as the localization of claudin 1 protein at the end of pregnancy differs in the 5-reductase type 1 null mice compared with WT suggests that the cervical epithelia do not function normally in these mice. The observation by other investigators that overexpression of claudin 6 in the suprabasal layer of the epidermis leads to a defect in terminal differentiation of stratified epithelia (17, 18) along with our observation that genes expressed by terminally differentiated stratified epithelia are down regulated in cervices of gestation d 18 Srd5a1–/– compared with d 18 WT cervices (Fig. 6 and unpublished microarray observations) led us to evaluate gene expression of Kert 1–16, Krtdap, Fabp5, Sprr1b, and Spink5 and lipocalin (Lcn).
Quantitative real-time PCR measurements suggest a significant reduction in expression of Kert 1–16, Krtdap, in late d 18 cervices of 5-reductase type 1 deficient mice compared with WT controls (n = 9–10 cervices/genotype) (Fig. 7). Additionally, Lcn2 expression is elevated 9-fold in d 18.75 cervices of Srd5a1–/– compared with WT. Changes in expression of Fabp5, Sprr1b, and Spink5 were not statistically significant.
The increase in transcription of marker genes for terminal differentiation of squamous epithelia in the WT cervix coincides with the decrease in concentrations of progesterone in serum required for initiation of parturition. To test the role of progesterone signaling pathways in regulation of epithelial differentiation, progesterone was administered to WT mice on gestation d 17 and cervices collected on gestation d 18.75 (Fig. 8). mRNA expression of Krtdap, but not Krt 1–16, Fabp5, Sprr1b, Spink5, or Lcn2, was reduced in progesterone treated compared with WT control cervices.
Aberrant expression of late epithelial differentiation markers and altered claudin 1 and 2 expression in the cervical ripening defective Srd5a1–/– may result in structural abnormalities of TJs. Electron microscopy was used to assess junctional complexes in ultrathin sections of gestation d 15 and d 18 cervices from WT and Srd5a1–/– cervices. This included assessment of the apical TJ, the lateral desmosomes, and the hemi-desmosomes located at the basement membrane. No structural differences in the overall junctional complex were noted between WT and Srd5a1–/– at gestation d 15 (data not shown) and d 18 (Fig. 9).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several important insights occur from or are highlighted in this study. Alterations in expression of specific claudins may perturb formation of an epithelial permeability barrier in cervical mucosal stratified epithelia. Although junctional complexes appeared structurally normal by electron microscopy, and expression of junctional proteins such as occludin, claudin 4, ZO-1, and ZO-2 in the late gestation d 18 Srd5a1–/– cervix was also normal, alterations in claudin 1 and claudin 2 expression resulted in reduced localization of claudin 1 protein within tight junctional complexes. The lack of correlation with junction structure and function is exemplified in the claudin 1 deficient mouse, in which junction function is compromised in the absence of claudin 1, yet junctional complexes have normal ultrastructure (12). In addition to the protein composition of junctional complexes, the abundance of claudin proteins also influences function. For example, transgenic mice that overexpress claudin 6 in its endogenous location in the suprabasal layer of the epidermis have a disruption of cell differentiation and epidermal permeability barrier dysfunction (17, 18).
Claudins are directly involved not only in formation of a TJ, but the choice of claudin protein also regulates the barrier capability and ion or size selectivity of the junction. The canine kidney epithelial cell line, MDCK I, which expresses claudins 1 and 4, has a much greater barrier capability compared with the MDCK II cell line, which expresses claudin 2 in addition to claudins 1 and 4 (33). This finding suggests that claudin 2 decreases the tightness of individual TJ strands. It is possible that cervical claudin 2 expression during pregnancy results in a weakened TJ complex, and/or changes the ion or size selectivity of the epithelial barrier relative to the TJ complex that is formed upon increased claudin 1 expression at term. A stronger, more selective barrier capability during parturition and post-term may better protect the remodeled, distensible cervix until it returns back into a rigid, closed structure. In addition, a claudin 1-enriched barrier may direct appropriate trafficking of ions and solutes required for optimal cervical mucus composition and the increased mucus production that is observed at term. Interestingly, transcripts for claudin 2 are not expressed in the nonpregnant cervix at any stage of the estrous cycle in contrast to claudins 1 and 4, suggesting a specific function of claudin 2 during pregnancy (data not shown). Future studies will be required to understand the functional differences between claudins 1 and 2 within the cervix and to quantitate changes in barrier capability in cervical epithelial cells containing claudin 1 or claudin 2 alone or in combination.
TJs are dynamic structures that can be rapidly remodeled in response to numerous stimuli (e.g. proinflammatory cytokines, oxidative stress, and pathogens) under physiological conditions such as epithelial morphogenesis or cell migration, as well as pathophysiological conditions such as inflammatory bowel disease (9, 34). Because assembly and disassembly of AJC occurs rapidly (as little as 1 h) and the half-life of junctional proteins is up to 12 h, the down regulation of junctional proteins may not account for the rapid disassembly of epithelial junctions. Instead, there is growing evidence that remodeling of the AJC is mediated by internalization and recycling of junctional proteins via endocytic pathways. Phosphorylation of claudin 4 and occludin has recently been described to cause disassembly of TJ proteins (22, 35). The mechanisms regulating protein trafficking of junctional proteins to and from apical junctions in the cervix during pregnancy and parturition remain to be elucidated.
Throughout pregnancy, increased proliferation and reduced apoptosis of epithelia and stroma result in growth of tissue with an increase in circumference of the cervical lumen, which is necessary for remodeling of the cervix (4, 5, 6). During cervical ripening, in addition to continued proliferation, a pathway of epithelial cell differentiation is initiated and coincides with the formation of claudin 1-enriched junctional complexes and, subsequently, an intact epithelial permeability barrier. We speculate that the pregnancy specific expression of claudin 2 may play a regulatory role in suppression of epithelial cell differentiation. As the cells progress along the differentiation pathway and claudin 2 gene expression declines, the cells gain expression of proteins involved in the eventual formation of a cornified envelope as well as proteins that serve to protect the permeability barrier such as the serine protease inhibitor, lympho-epithelial Kazal-type related inhibitor (Spink5) (16, 36). Cervices of the Srd5a1–/– model that fail to initiate the differentiation pathway, continue to express claudin 2 transcripts, have reduced expression of claudin 1 and some genes expressed by the apical differentiated epithelia (Fig. 7) as well as elevated expression of the gene encoding Lcn2, a marker for dysregulated keratinocyte differentiation in human skin (37). The hormones progesterone, estrogen, and relaxin, in part, regulate epithelial cell proliferation during pregnancy (4, 5, 6). The differentiation pathway may also be in part hormonally regulated and require a reduction in local progesterone concentration, as suggested by the reduced expression of Krt-16 and Krtdap in the Srd5a1–/– model and the ability of progesterone administration to block expression of Krtdap in WT mice. We speculate that induction of the differentiation pathway occurs very late on d 18 because there was great variability in gene expression of this group of genes dependant on the timing of tissue collection. This resulted in variability in induction of differentiation gene markers and may account for the lack of significance for some genes analyzed in Figs. 7 and 8. The terminal differentiation of epithelia at parturition is consistent with the observation that mice are in the estrous phase of the cycle within hours after birth (38).
Finally, these studies highlight the important regulatory role that cervical epithelia play during pregnancy, parturition, and PP recovery of the cervix back to the nonpregnant state. During pregnancy the epithelia must proliferate for sufficient tissue growth and remodeling (5, 6, 23), maintain fluid balance during softening and ripening by temporal and tissue-specific regulation of aquaporin water channels (7), and initiate surveillance mechanisms to protect the integrity of the stromal matrix. In addition to these responsibilities, our studies in the mouse reveal that during pregnancy, a unique pattern of TJ protein expression is maintained until the accelerated phase of cervical ripening on gestation d 18. Furthermore, during cervical ripening, pathways that alter the epithelial permeability barrier are activated and maintained into the PP period. Barrier formation may or may not be required for cervical ripening per se; however, further studies are required to determine if claudin 1-enriched TJs and the presence of terminally differentiated mucosal stratified epithelia play a regulatory role in the final stages of cervical ripening and dilatation. We speculate that this process is necessary to ensure that the epithelia can maintain their critical functions and provide additional protection against invasion of pathogens until the cervix sufficiently remodels back into a rigid, closed structure after birth. These studies enhance our understanding of cervical epithelial biology as well as reveal new physiological scenarios in which claudin-based barriers play an important role. Future studies are required to understand mechanisms regulating changes in cellular localization of junctional proteins during cervical ripening, and the effects of these changes on epithelial cell differentiation and proliferation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: B.C.T., S.M.M., C.G., and M.S.M. have nothing to declare.
First Published Online November 30, 2006
Abbreviations: AJC, Apical junctional complex; Fabp5, fatty acid-binding protein 5 (keratin lipid-binding protein); Kert 1–16, keratin 16; Krtdap, keratin differentiation-associated protein; Lcn, lipocalin; PP, postpartum; Spink5, serine protease inhibitor Kazal type 5; Sprr, small-proline rich proteins; TBST, Tris-buffered saline with Tween 20; TJ, tight junction; WT, wild type; ZO, zona occludens.
Received June 22, 2006.
Accepted for publication November 16, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-reducase type 1 knockout mice is due to impaired cervical ripening. Mol Endocrinol 13:981–992
-triggered internalization of tight-junction proteins in inflammatory bowel disease. Sci STKE 316:pe1This article has been cited by other articles:
 |
|
 |
|
  B. C. Timmons and M. Mahendroo Processes Regulating Cervical Ripening Differ From Cervical Dilation and Postpartum Repair: Insights From Gene Expression Studies Reproductive Sciences, December 1, 2007; 14(8_suppl): 53 - 62. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
0.05) when compared with late d 18.
indicates significance (P 
