| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
Clinical Sciences at South Bristol,3 Division of Obstetrics and Gynaecology, Henry Wellcome Laboratories for Integrative Neuroscience and Endocrinology, Dorothy Hodgkin Building, University of Bristol, Bristol BS1 3NY, United Kingdom
Department of Pathology,4 Bristol Royal Infirmary, University of Bristol, Bristol BS2 8HW, United Kingdom
Department of Biochemistry,5 School of Medical Sciences, University of Bristol, Bristol BS8 1TD, United Kingdom
ABSTRACT
RHO GTP-binding proteins are important regulators of actin-myosin interactions in uterine smooth muscle cells. Active (GTP-bound) RHOA binds to RHO-associated protein kinase (ROCK1), which inhibits the myosin-binding subunit (PPP1R12A) of myosin light chain phosphatase, leading to calcium-independent increases in myosin light chain phosphorylation and tension, which are termed "calcium sensitization." The RHO effector protein kinase N (PKN1) also increases calcium sensitization by phosphorylating the protein kinase C (PRKCB)-dependent protein CPI-17 (PPP1R14A) to inhibit the PPP1c subunit of myosin phosphatase. Moreover, other RHO proteins, such as RHOB, RHOD, and their effectors (DIAPH1 and DIAPH2), may modulate PKN1/ ROCK1 signaling to effect changes in myosin phosphatase activity and myosin light chain phosphorylation. The increases in contractile activity observed in term and preterm labor may be due to an increase in RHO activity and/or changes in RHO-related proteins. We found that the RHOA and RHOB mRNA levels in the myometrium were increased in pregnancy, although the expression levels of the RHOA and RHOB proteins did not change with pregnancy or labor. GTP-bound RHOA was increased in pregnancy, and this increase was significant in spontaneous preterm labor myometrium. PKN1 expression and PPP1R14A phosphorylation were dramatically increased in the pregnant myometrium. We also observed increases in DIAPH1 expression in spontaneous term and preterm labor myometrial tissues. The present study shows that human pregnancy is characterized by increases in PKN1 expression and PPP1R14A phosphorylation in the myometrium. Moreover, increases in GTP-bound RHOA and DIAPH1 expression may contribute to the increase in uterine activity in idiopathic preterm labor.
DIAPH1, female reproductive tract, oxytocin, parturition, PKN1, RHO, uterus
Preterm birth is the most important cause of perinatal mortality and morbidity worldwide. Spontaneous uterine contractions leading to preterm labor affect around 30% of all patients [1, 2]. Effective management of this important clinical problem is currently hampered by the inability of existing tocolytic agents to arrest safely uterine activity for more than 48 h [3].
Uterine smooth muscle contraction is primarily regulated by the calcium-dependent myosin light chain kinase (MYLK) [4]. Increases in intracellular calcium ([Ca2+]i) activate MYLK, which phosphorylates the regulatory myosin light chain (MYL), thereby enhancing myosin ATPase activity to cause an increase in tension [5–7]. Myosin phosphatase can reverse MYL phosphorylation to induce a state of relaxation. However, during agonist-induced contraction at constant [Ca2+]i, concurrent inhibition of myosin phosphatase leads to increases in MYL phosphorylation and tension [8, 9]. These calcium-independent increases in myosin phosphorylation and tension are termed ‘calcium sensitization' [10].
The RHO family of small GTPases encompasses molecular switches that act as important regulators of the actin cytoskeleton and stress fiber formation. GTP-bound RHOA binds to RHO-associated kinase (ROCK1), which phosphorylates the myosin-binding subunit (PPP1R12A) of myosin phosphatase, thereby inhibiting its activity [11, 12]. Activated ROCK can also phosphorylate MYL at Thr18 and Ser19. Both actions lead to calcium sensitization [13, 14]. Myosin phosphatase is also regulated at the level of the protein phosphatase subunit (PPP1c) by the protein kinase C (PRKCB) inhibitory protein PPP1R14A (CPI-17) [15, 16]. Moreover, other RHO family members, e.g., the endosomal GTPases RHOB and RHOD, can also regulate actin-myosin interactions.
Although RHOB has 83% homology to RHOA, it is differentially regulated and prenylated [17]. Activated RHOB is known to target the serine-threonine kinase PKN1 (a RHO effector [18, 19]) to endosomes, where it regulates EGF receptor trafficking [20, 21]. Furthermore, RHOB stimulates actin filament assembly and regulates endosome trafficking through the effector protein DIAPH1 [22]. Recent genomic studies have demonstrated upregulation of RHOB and RAB7 GTPases during pregnancy in the Guinea pig [23]. In neuronal cells, RHOB rather than RHOA is the primary regulator of MYL phosphorylation [24].
RHOD is usually localized to early endosomes and the plasma membrane [25]. RHOD can disrupt endosomal trafficking by targeting a novel splice variant of the effector protein DIAPH2 to promote actin polymerization and to stabilize endosome alignment along actin filaments [26]. Conversely, the expression of activated RHOD reverses lysophosphatidic acid-and RHOA-mediated stress fiber formation in 3T3 fibroblasts [27].
A number of RHO effector proteins regulate actin-myosin interactions by binding to actin. The mammalian diaphanous-related formin proteins DIAPH1 and DIAPH2 are RHO effectors that promote the polymerization, nucleation, and branching of actin filaments by interacting with profilin [28–30]. GTP-bound RHOA activates DIAPH1 to transform ROCK1-induced actin fibers into stress fibers [31, 32]. Activated diaphanous proteins can also activate ROCK1 independently, and they serve as cross-regulatory links between RHO and tyrosine kinase signaling pathways [32, 33].
We and others have previously demonstrated the presence of RHOA and its effectors, ROCK1 and ROCK2, in the pregnant human myometrium [34–36]. Chronic exposure to U44619 (a thromboxane analogue) stimulates a RHO-mediated increase in p160ROCK1 protein expression, which is blocked by receptor antagonism and C3-exotoxin inhibition [37]. We have also recently demonstrated pregnancy-related upregulation of the expression of RND2 and RND3 RHO-binding proteins, together with a loss of PPP1R12A phosphorylation in the human myometrium [38]. Transfection of RND2 and RND3 cDNAs has been shown to result in a dramatic loss of actin stress fibers in cultured myometrial cells [38].
We hypothesize that an increase in RHO activity leads to the premature myometrial activity observed in spontaneous preterm labor. The purpose of the present study was to determine the relative expression levels of the RHOA, RHOB, and RHOD GTP-binding proteins, as well as the levels of the previously uncharacterized RHO effector proteins PKN1, DIAPH1, and DIAPH2 in myometrial tissues during pregnancy and labor. We studied the immunohistochemical localization of these proteins within uterine tissues and cultured human myometrial cells, and we measured the activities of RHOA and RHOB in pregnant myometrium using GTP pulldown assays.
The following primary antibodies were used: ARHGDIA (RHOGDI, sc 360), RHOA (sc 418), RHOB (sc 8048), and PKN1 (PRK1, sc 7969) from Santa Cruz Biotechnology (Santa Cruz, CA); and RHOB (BL927), DIAPH1 (BL719), and DIAPH2 (BL722) from Bethyl Laboratories (Montgomery TX). RHOD was produced in-house (Dr. Harry Mellor). The TUBA1 (
-tubulin, T5168) mouse monoclonal antibody was obtained from Sigma, (Poole, UK). The following secondary HRP-conjugated polyclonal antibodies were used: rabbit anti-goat, pig anti-rabbit, and rabbit anti-mouse from DAKO (Cambridge, UK). Alexa Fluor 546 goat anti-rabbit, Alexa Fluor 546 goat anti-mouse, and Alexa Fluor 488 goat anti-rabbit were purchased from Molecular Probes (Eugene, OR).
Blotto A was purchased from Santa Cruz Biotechnology, RTU Vectastain Quick Universal Kit from Vector Laboratories (Peterborough, UK), Optimax buffer from BioGenex (Berkshire, UK), RNeasy Mini RNA extraction kit and QIAshredder were obtained from Qiagen (Crawley UK), Superscript II First Strand Synthesis System cDNA kit from Invitrogen (Paisley, UK), SYBR Green detection from Dynamo Finnzymes (Espoo, Finland), and PVDF membranes and protein kaleidoscope molecular weight markers were acquired from Bio-Rad Laboratories (Hemel Hempstead, UK). ECL Plus and hyperfilm were purchased from Amersham Biosciences (Buckinghamshire, UK). Oleoyl-L--lysophosphatidic acid sodium salt (LPA) was obtained from Sigma (St. Louis, MO). GST-tagged fusion protein (GST-RBD), 100x GTP
S (10 mM) and 100x GDP (10 mM) were purchased from Upstate Biotechnology (Lake Placid, NY). The phosphatase inhibitor cocktail (P5726) and protease inhibitor cocktail were obtained from Sigma.
Human Myometrial Tissue Collection
Myometrial tissue was obtained from nonpregnant (NP) premenopausal women who were undergoing hysterectomy for benign gynecological disorders, and from pregnant women who were not in labor (NIL), at term (37 to 40 weeks of gestation) who were undergoing cesarean section with the following indications: maternal request, breech presentation, previous caesarean section or placenta previa. Myometrial tissue was excised from the upper border of the uterine incision, taking care to exclude serosal, decidual or scar tissue. Myometrial tissue was also obtained from women in spontaneous labor at term (SL) and spontaneous idiopathic preterm labor (less than 37 weeks of gestation) (SPT) with no evidence of infection. Labor was defined as regular uterine contractions and cervical dilatation greater than 4 cm. Tissues were snap-frozen and stored in liquid nitrogen. This study was approved by the United Bristol Healthcare Trust Research Ethics Committee. All patients gave written informed consent.
Myometrial samples were harvested into RNA later and snap frozen in liquid nitrogen. The samples were then homogenized using a pestle and mortar and QIA shredder. RNA was extracted from the tissues using the Oligotex TM RNA Mini kit (Qiagen), and treated with RNase-free DNase according to the manufacturer's instructions. RNA concentration and quality were assessed in an Eppendorf Biophotometer.
Real-time quantitative PCR (RT-qPCR) was used to confirm the relative expression levels of the mRNAs of the selected GTP-binding proteins. Reverse transcription was carried out using 0.2 µg of random hexamer primers, 10 mM dNTP mix, and 5 µg RNA, which was first denatured by incubation at 65°C for 5 min, followed by reverse transcription with Superscript II at 42°C for 50 min. The reactions were terminated by incubation for 15 mins at 70°C, and RNase H was added to each tube.
RT-qPCR primers were designed using the DNAStar primer software and synthesized by MWG Biotech (Germany). RPII (RNA polymerase II) was used as an endogenous control gene. The following primers were used: for RHOA, forward 5'-CCGGCGCGAAGAGGCTGGACT-3' and reverse 5'-GCACATACACCTCTGGGAACT-3'; for RHOB, forward 5'-GGTCCCCTGAGCATGCTTTTCTGA-3' and reverse 5'-GCCACACTCCCGCGCCAATCTC-3'; and for RPII, forward 5'-CTGGAACGGTGAAGGTGACA-3' and reverse 5'-AAGGGACTTCCTGTAAACAATGCA-3'.
Five microliters of the resultant target tissue cDNA was used as a template for each 50-µl PCR reaction. All the PCR reactions were performed in triplicate. Four-fold serial dilutions of HeLa cDNA were used to construct standard curves for the endogenous control RPII and the target genes RHOA and RHOB.
RT-qPCR was performed using SYBR Green detection in the Opticon 2 system (Bio-Rad), with the following cycling parameters: 42 cycles of denaturation at 95°C for 10 s, annealing at 55°C for 10 s, and extension at 72°C for 10 min, followed by a final extension at 70°C for 10 min. After amplification, dissociation curve analysis (55°C to 100°C in steps of 0.4°C, with a hold time of 1 s) was performed, to check the products obtained. The products were sequenced to confirm the identities of the genes investigated. The baseline for all the reactions on each plate was corrected with RT negative and water controls. For each target gene investigated, a cycle profile was conducted for the myometrial tissue and HeLa cell control cDNA, to determine the exponential phase of the reaction. The threshold was manually adjusted within the exponential phase of amplification to obtain maximum efficiency. The cycle numbers for the log linear phase of the reaction were plotted against a standard curve of serially diluted HeLa cDNA samples. The cycle threshold method was used to calculate the relative expression levels. The amounts of the target genes (RHOA and RHOB) and endogenous gene (RPII) in each sample were determined from the standard curves. The amount of RHOA or RHOB was divided by the amount of RPII to obtain the normalized level of RHOA or RHOB.
Myometrial samples were homogenized in ice-cold RIPA lysis buffer using a mechanical homogenizer, as described previously [38]. Proteins were separated from nuclei and nondisrupted cells by centrifugation (10 000 x g for 1 h at 4°C). The supernatant was stored at –80°C until use.
Western Blotting and Densitometry
Homogenate proteins (100 µg) were separated by SDS-PAGE and transferred to PVDF membranes using a semidry blotting system. Membrane proteins were blocked with Blotto A before Western blotting with a primary antibody and a compatible secondary HRP-conjugated antibody. Immunoblots of different patient samples were performed under conditions in which the intensities of the bands obtained were proportional to the protein concentrations and ECL reactions. Repeat immunoblots carried out on different days gave similar results. All samples were analyzed in duplicate and normalized to their respective tubulin (TUBA1) signals. The level of TUBA1 expression did not change among the different groups. Densitometric analysis was performed using the Image J software (Wright Cell Imaging Facility, Toronto, Canada).
Human Uterine Smooth Muscle (HUSM) Cell Culture and Immunofluorescence
Myometrial cells were isolated from myometrial tissues by limited enzyme digest, as described by Casey et al. [39], and grown in culture, as described by Phaneuf et al. [40]. Subconfluent myometrial cells were trypsinized and seeded onto coverslips, fixed in 4% (w:v) paraformaldehyde, permeabilized with 0.2% Triton X-100, and treated with fresh sodium borohydride. Fixed myometrial cells were stained for endogenous protein expression using anti-RHOA (1:100), anti-RHOB (1:50), anti-RHOD (1:100), anti-PKN1 (1:50), anti-DIAPH1, and anti-DIAPH2 (1:100) antibodies. Secondary Alexa Fluor-conjugated antibodies were used to visualize the relevant antigens. Filamentous actin was detected using Alexa 488-conjugated phalloidin.
RHO Pulldown Assays of Cell Extracts and Tissue Homogenates
Human myometrial cells were cultured to 80% confluency in T25 flasks, and the growth medium was replaced with serum-free medium (SFM) 24 h prior to use. The performance of the assay was tested with GTPS and GDP loading controls according to the manufacturer's instructions [41]. Briefly, 10 µl of 0.5 M EDTA were added to 0.5 ml of myometrial cell extract, followed by the addition of 100 µM GTPS to the positive control tubes, 1 mM GDP to the negative control tubes, and SFM to the basal activity tubes. The tubes were incubated at 30°C for 30 min and the reactions were terminated by adding 60 mM MgCl2 and placing the tubes on ice.
For the RHO stimulation experiments, we used a modification of the method developed by Ren et al. [42]. HUSM cells were incubated in SFM for 24 h, and then treated with 30 µM LPA for reaction times that ranged from 0 to 5 min. Following stimulation, the supernatants were removed and the reactions were terminated by washing with ice-cold Tris-buffered saline (TBS; 50 mM Tris base [pH 7.6], 140 mM NaCl) before incubation in lysis buffer (50 mM Tris-HCl [pH 7.2], 1% Triton X-100, 0.1% SDS, 500 mM NaCl, 10 mM MgCl2, protease and phosphatase inhibitor cocktails diluted 1:100). The cell extracts were centrifuged to remove debris, and equal volumes of the supernatants were incubated with 30 µg of GST-RBD (gluthathione S-transferase tagged rhotekin binding domain) bound to Sepharose beads (Upstate Biotechnology), at 4°C for 45 min. Aliquots were removed for assessment of total RHO as loading control. For the assessment of RHO activity in tissue samples, 3 mg of homogenate proteins were incubated with 30 µg of GST-RBD to measure GTP-bound RHO. Total RHO was measured in parallel in 100 µg of homogenate proteins per sample. The data were expressed as the ratio of GTP-RHO to total RHO. Following sedimentation by centrifugation, the samples were washed four times with ice-cold lysis buffer. The bound proteins were eluted with Laemmli buffer that contained 40 mM DTT, and boiled for 10 min. Equal volumes of the bound proteins were separated by SDS-PAGE and Western blotting was performed with the anti-RHOA (1:1000) and anti-RHOB (1:1000) antibodies.
Immunohistochemistry was carried out on paraffin-embedded tissue sections, as described previously [38]. The following primary antibodies were used: anti-RHOA mouse monoclonal antibody (1:200), anti-RHOB mouse monoclonal antibody (1:400), anti-RHOD rabbit polyclonal antibody (1:100), anti-PKN1 goat polyclonal antibody (1:100), and anti-DIAPH1 and anti-DIAPH2 rabbit polyclonal antibodies (1:1000). Frozen sections were fixed in 2% (wt/vol) paraformaldehyde, and permeabilized with 0.3% Triton X-100 before overnight incubation with the primary antibody. The antigens were then visualized using fluorescent Alexa Fluor 546-conjugated goat anti-rabbit, Alexa Fluor 546-conjugated goat anti-mouse, and Alexa Fluor 488-conjugated goat anti-rabbit secondary antibodies. Actin stress fibers were visualized using Alexa Fluor 488-conjugated phalloidin. Liver, brain, and placental tissues were used as appropriate positive controls for protein expression (data not shown). Appropriate isotype and preimmune serum controls were included for all the primary antibodies used. The primary antibody was omitted from the staining protocol as a negative control for the secondary antibodies used.
Fluorescent images were captured on a Leica TCS-NT camera attached to a DM/RBE epifluorescent microscope using a Plan APO x63/1.36 oil immersion lens (Leica, Heidelberg Germany), as previously described by Gampel et al. [20]. Fluorophores were excited using the 405-nm line of a diode laser (for DAPI staining) and using the 488 nm and 546 nm lines of a Kr-Ar laser, respectively, for Cy2 and Cy3 bands. There was no detectable bleed of fluorescence from one channel to the other in the single-stained samples. Microscope settings were adjusted so that the imaging parameters were kept constant for both the red and green channels. Series of images were taken at 0.3-µm intervals through the Z plane of the section and were processed to form a projected image. The captured images were processed with Adobe Photoshop 7.0 (Adobe Systems, San Jose, CA). The image brightness and contrast was altered only using a linear operation (gamma curves not manipulated).
The data were analyzed using the GraphPad Prism software (Hearne Scientific Software, Dublin, Ireland). RHO protein signals were not normally distributed and the data were log-transformed before analysis by ANOVA with Tukeys post-hoc analysis. Statistical significance was determined at P < 0.05.
Upregulation of RHOA and RHOB mRNAs in Pregnant Myometrium
We investigated the expression levels of RHOA and RHOB GTPase mRNAs in nonpregnant (n = 5) and pregnant (n = 5) samples, using RNA polymerase II as a housekeeping gene. Analysis of the RT-qPCR curves revealed a two- to three-fold increase in RHOA (P < 0.02) and RHOB mRNAs (P < 0.04) in the pregnant tissues relative to the nonpregnant tissues (Fig. 1). Melting curve analysis confirmed the formation of a single product as the result of PCR amplification with the RHOA and RHOB primers. No product was formed after PCR amplification of the RT negative and cDNA negative (water) controls.
|
RHOA, RHOB and RHOD Protein Expression in Pregnant Versus Nonpregnant Human Myometrium
In light of the observed increases in RHOA and RHOB mRNA expression in pregnant myometrium, we used immunoblotting to determine the relative expression levels of the RHO proteins in the nonpregnant (NP), pregnant not in labor (NIL), spontaneous labor at term (SL), and spontaneous preterm labor (SPT) groups. The level of TUBA1 (tubulin) was used to ensure equal protein loading across all four groups [38]. The anti-RHOA and anti-RHOB antibodies detected bands at 25 kDa and 24 kDa, respectively, in both the nonpregnant and pregnant tissues (Fig. 2). There were no differences in the relative expression levels of RHOA and RHOB among the four groups.
|
We also measured RHOD expression in the four groups. The anti-RHOD antibody detected a band of approximately 26 kDa (Fig. 2). There was a significant reduction in RHOD expression in the spontaneous labor myometrium relative to the nonpregnant myometrium (P < 0.001). However, we failed to detect any significant differences in RHOD expression between the nonlabor and labor groups.
Myometrial GTP-Bound RHOA Is Increased in Spontaneous Preterm Labor
To determine the significance of the RHOA and RHOB mRNA and protein expression levels during pregnancy, we measured RHO activity in the myometrium using a RHO pulldown assay. GTPases cycle between an inactive GDP-bound and an active GTP conformation. GTP-bound RHO activates effector proteins, such as ROCK1, by binding to specific domains, e.g., the RTKN (Rhotekin)-binding domain. Thus, RHO activity is a function of the ratio of GTP-bound RHO to total RHOA. Our validation experiments (Fig. 3, A and B) demonstrate that GTPS loading of myometrial cell extracts produces increases in GTP-bound RHOA and RHOB, whereas GDP loading results in a marked reduction in GTP-RHO compared to the control cells. LPA stimulation of myometrial cells produced increases in GTP-RHOA and GTP-RHOB at 1 min and 5 min (Fig. 3, A and B).
|
To assess possible changes in RHO activity in myometrial tissues during pregnancy, we normalized the ratio of RHO-GTP to total RHO in the pregnant groups to those of the nonpregnant group. There was an apparent increase in GTP-bound RHOA in all the pregnant groups relative to the nonpregnant group, and this was significant in the spontaneous preterm labor group (n = 5; P < 0.05) (Fig. 3C). There were no significant differences in GTP-bound RHOA among the three pregnant groups. In contrast to RHOA, there were no significant changes in GTP-bound RHOB among the four groups examined (data not presented).
Expression of RHO Proteins and Their Effectors in Uterine Tissues
In light of this increase in RHOA activity in spontaneous preterm labor, we investigated the presence of known RHO effectors DIAPH1, DIAPH2, and PKN1 in human myometrium in comparison to other uterine tissues. RHOA and RHOB were both highly expressed in the myometrial, placental, and decidual tissues examined (Fig. 4). PKN1, DIAPH1, and DIAPH2 were expressed in the myometrial tissues and other tissues examined. DIAPH2 was not present in the cord blood or decidua parietalis. RHOB was not detected in cord blood.
|
Immunohistochemical Localization of RHOA, RHOB, RHOD DIAPH1, DIAPH2, and PKN1 in Uterine Tissue Sections
The presence of RHO proteins and their effectors in myometrial tissues was confirmed by immunohistochemistry (Figs. 5 and 6). We observed positive cytoplasmic and membrane staining with marked perinuclear enhancement of RHOA and RHOB in all the nonpregnant and pregnant uterine sections examined (Figs. 5 and 6). Endometrial stromal glands were also positive for the three RHO GTPases examined (Fig. 5). The effector proteins DIAPH1 and DIAPH2 showed intense nuclear and cytoplasmic staining reactions in both nonpregnant and pregnant myometrial fibers (Fig. 6). Interestingly we observed a marked increase in the cytoplasmic immunoreactivity of PKN1 in pregnant compared to nonpregnant myometrial fibers (Fig. 6, B and C). There was strong DIAPH1 staining of the cytoplasm of nonpregnant endometrial glands and stroma, whereas the other two effectors proteins, DIAPH2 and PKN1, stained weakly positive (Fig. 6, A, D, and G). An antibody that recognizes Desmin, which is a smooth muscle specific protein, was used to confirm the identity of the myometrial cells. Positive control experiments confirmed that RHOB was present in the brain [24] (data not shown). No staining was observed when the primary antibody was omitted or substituted with the appropriate isotype or preimmune serum control (Fig. 6, J and K).
|
|
This is the first study to demonstrate the presence of RHOB, RHOD, DIAPH1, DIAPH2, and PKN1 in human myometrium, and to confirm by immunofluorescence the intracellular localizations of these proteins in freshly dispersed cultured human myometrial cells (Fig. 7). RHOB and RHOD showed punctate perinuclear and endoplasmic localizations within the myometrial cells [20, 26, 43] (Fig. 7). DIAPH1 showed a diffuse nuclear and cytoplasmic distribution, in contrast to DIAPH2, which demonstrated an intense perinuclear and cytoplasmic distribution (Fig. 8). PKN1 showed a well-defined punctate cytoplasmic distribution [21]. As expected, pregnant myometrial tissue samples from which the myocytes were enzymatically dispersed also demonstrated positive staining for the proteins examined.
|
|
DIAPH1 Expression is Upregulated in Term and Preterm Labor Myometrium
Recent evidence suggests that DIAPH1 cooperates with ROCK1 in RHO-induced actin reorganization [30, 32]. We used immunoblotting to determine the relative expression levels of DIAPH1 and DIAPH2 in nonpregnant and pregnant myometrial tissues. The anti-DIAPH1 and anti-DIAPH2 antibodies detected bands at 200 kDa and 150 kDa, respectively. Despite considerable interpatient variability, we observed an increase in the expression of DIAPH1 in spontaneous labor (P < 0.05) and spontaneous preterm labor (P < 0.05) tissues as compared to labor tissues (Fig. 9). There were no significant differences in DIAPH2 expression among the four groups.
|
Upregulation of PKN1 Expression in Pregnant Human Myometrium Is Associated with Increased PPP1R14A-Thr 38 Phosphorylation
The PKN family of serine/threonine kinases was initially discovered as a family of potent RHOA effectors [19, 44]. PKN1 binds to RHOA via two distinct domains. The HR1a domain binds exclusively to GTP-RHOA, while its HRIb domain is capable of binding to both GDP and GTP-bound RHOA [21]. Recent evidence implicates the C-terminal region of PKN1 as essential for its activation by RHOA [45]. Pathways of calcium sensitization involve not only RHOA-ROCK1, but also the protein kinase C (PRKCB1)-dependent protein PPP1R14A (formerly known as CPI-17) [15, 16, 46]. PKN1 can phosphorylate PPP1R14A at Thr38 in vitro, thereby increasing its inhibitory effect on the myosin phosphatase PPP1c catalytic subunit [46, 47].
The anti-PKN1 antibody detected a band of 125 kDa in the pregnant samples. We observed dramatic upregulation of PKN1 expression in the pregnant tissues compared to the nonpregnant tissues, i.e., not in labor vs. nonpregnant, P < 0.001; spontaneous labor vs. nonpregnant, P < 0.001; and spontaneous preterm labor vs. nonpregnant, P < 0.001 (Fig. 10). There were no significant differences in PKN1 expression among the pregnant groups.
|
We also observed a corresponding increase in Thr38-phosphorylated PPP1R14A in all the pregnant samples relative to the nonpregnant samples (Fig. 10). There were no differences in PPP1R14A phosphorylation among the pregnant samples. Functional phosphorylation of PPP1R14A was induced in freshly dispersed myometrial cells by the contractile agonist oxytocin (Fig. 10C). Similar results were obtained with carbachol (results not shown).
The present study demonstrates for the first time the presence of RHOB, RHOD, and their effector proteins, PKN1, DIAPH1, and DIAPH2, in human myometrium. The activities of the RHO proteins are regulated by their levels of expression, subcellular localizations, and phosphorylation and reflected in the proportion of RHO protein that is bound to GTP [48]. Our pulldown data demonstrate for the first time that GTP-bound RHOA increases in pregnancy and is significantly elevated in spontaneous preterm labor. GTP-bound RHOB remains unchanged during pregnancy and labor. This implies that RHOA is an important modulator of uterine activity during preterm labor. This is consistent with a previous suggestion that ROCK1 inhibition with Y27632 leads to lower delivery rates in PGF2- and lipopolysaccharide-induced preterm labor in mice [49]. It seems timely to consider the development of RHOA-ROCK1 inhibitors, such as Y27632, HA1077, and H-1125, as potential tocolytic agents in preterm labor.
Our data demonstrate a decrease in RHOD protein expression in the spontaneous labor (SL) group relative to the nonpregnant (NP) group. Transfection of activated RHOD cDNA into cultured HeLa, BHK, and NIH 3T3 cells resulted in the disappearance of RHOA-induced actin stress fibers and focal adhesion complexes that contained vinculin and paxillin [25]. Furthermore, RHOD WT cDNA has been shown to inhibit the formation of LPA- and RHOA-induced stress fibers in NIH 3T3 fibroblasts [27]. Thus, a reduction in RHOD expression in the myometrium may potentiate the effect of RHOA in labor.
The RHO effector proteins DIAPH1, DIAPH2, and PKN1 are strongly expressed in myometrial tissues. DIAPH1 and DIAPH2 are proline-rich members of the formin homology family, the proteins of which bind to the G-actin binding protein profilin to cause increased actin polymerization [30, 31]. These proteins are primarily regulated by autoinhibition and activated by GTP-bound RHOA [31, 50]. They have been implicated in a wide range of biological activities, including the regulation of ovarian function [51]. RHO-activated DIAPH1 can organize ROCK1-induced thin actin filaments into contractile stress fibers [32, 52]. Therefore, the elevated expression of DIAPH1 in the term and preterm labor myometrial tissues, as shown in the present study, suggests potentiation of RHOA activity through DIAPH1-ROCK co-operation.
A remarkable finding of the present study is the significant increase of the RHOA/RHOB effector PKN1 in pregnant tissues. To our knowledge, this is the first description of PKN1 expression in the myometrial tissue of any species. RHOA activation of PKN1 can result in the regulation of a range of cellular events, encompassing E-cadherin cell adhesion in keratinocytes [53], insulin-stimulated glucose transport [54], and the regulation of meiotic maturation and embryonic cycles in starfish oocytes [55]. However, PKN1 can be activated by other GTPases. For example, PKN1 can be targeted to late endosomes by RHOB where it regulates the actin cytoskeleton in endosome trafficking [21].
It has been suggested that an increase in PPP1R14A phosphorylation in pregnant compared to nonpregnant myometrial tissues may mediate a gestation-induced increase in calcium sensitization [38, 56]. PKN1 can independently phosphorylate PPP1R14A in vitro [47]. Using point mutations of PPP1R14A, Hamaguchi et al. [47] have mapped the site of PKN1 phosphorylation on PPP1R14A to Thr38, and have further suggested that the level of potency of PKN1 phosphorylation on PPP1R14A is similar to that of ROCK1 [57]. Therefore, the significant upregulation of PKN1 expression and the increase in PPP1R14A phosphorylation during pregnancy strongly indicate that the PKN1/PPP1R14A pathway can co-operate with RHOA-ROCK1 to regulate the MYL/MYL-P equilibrium in the pregnant myometrium.
Previously, we have suggested that increases in the RND2 and RND3 proteins mediate a loss of ROCK1-induced inhibition of the myosin phosphatase PPP1R12A subunit in the pregnant myometrium [38]. We believe that the activity of myosin phosphatase is influenced by two counterbalancing pathways: RHO-ROCK1 inhibition of the PPP1R12A subunit and PKN1-regulated phosphorylation of PPP1R14A (see Fig. 11). Inhibition of the former pathway may be important in regulating quiescence during gestation and the increase in PKN1 expression may be a mechanism to prepare the uterus for the contractile demands of labor.
|
It should be borne in mind that other kinases, such as ROCK1 [57, 58], integrin-linked kinase [59], zip kinase [60–62], and p21-activated kinases [63], can potentially regulate PPP1R14A activity by phosphorylation at Thr38 (Fig. 11). Our previous data, which demonstrate increased expression and activity of p21-activated kinases in the myometrium during pregnancy, suggest that this kinase is also implicated in PPP1R14A regulation in the pregnant myometrium [34]. Further research is needed to determine the effects of multiple kinases on the activity of PPP1R14A in the pregnant uterus.
In conclusion, we provide evidence for the presence of two new RHO proteins, RHOB and RHOD, in human myometrium. Our data also demonstrate that the levels of the effector proteins PKN1 and DIAPH1 are increased in the pregnant myometrium and are likely to be involved in the regulation of actin-myosin interactions during pregnancy and labor. The striking increase in PKN1 expression is paralleled by an increase in PPP1R14A phosphorylation in the pregnant myometrium. We have also observed an increase in GTP-bound RHOA in spontaneous preterm labor myometrium. Increased RHOA activation coupled with elevated DIAPH1 expression in the myometrium could be a physiopathological mechanism for preterm labor. Further research is needed to clarify the roles of RHO-related GTPases and their effector proteins in pregnancy maintenance and the onset of labor.
ACKNOWLEDGMENTS
We are grateful to the staff of St Michael's Hospital (Bristol) for help with the collection of samples. We thank Dr. Claire Perks for assistance with the RHO pulldown assay and Dr. S. Pellegrin for comments during the preparation of this manuscript. We are indebted to Dr. M. Taggart (University of Manchester) for assistance with tissue immunofluorescence. This research is integrated in the SAFE Network of Excellence and in the European Preterm Labour Group.
FOOTNOTES
1J.L. is a recipient of a WellBeing of Women (Royal College of Obstetricians and Gynaecologists) Research Training Fellowship. 
Correspondence: 2FAX: 44 117 928 5290; e-mail: A.LopezBernal{at}bristol.ac.uk
Received: 14 November 2006.
First decision: 30 November 2006.
Accepted: 2 February 2007.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  F. Dabertrand, N. Fritz, J. Mironneau, N. Macrez, and J.-L. Morel Role of RYR3 splice variants in calcium signaling in mouse nonpregnant and pregnant myometrium Am J Physiol Cell Physiol, September 1, 2007; 293(3): C848 - C854. [Abstract] [Full Text] [PDF]
|
|
| |||||||||
