The expression of Abl interactor 2 in leiomyoma and myometrium and regulation by GnRH analogue and transforming growth factor-β

Abstract

BACKGROUND: Abelson (Abl) interactor 2 (Abi-2) has been considered as a key regulator of cell/tissue structural organization and is differentially expressed in leiomyomas. The objective of this study was to evaluate the expression of Abi-2 in leiomyoma/myometrium during the menstrual cycle and following GnRH analogue (GnRHa) therapy, as well as regulation by transforming growth factor (TGF)-β1 in leiomyoma and myometrial smooth muscle cells (LSMC and MSMC). METHODS: We used real-time PCR, Western blotting and immunohistochemistry to determine the expression of Abi-2 in paired leiomyoma and myometrium (n = 27) from proliferative (n = 8) and secretory (n = 12) phases of the menstrual cycle and from patients who received GnRHa therapy (n = 7). Time-dependent action of TGF-β1 (2.5 ng/ml) and GnRHa (0.1 µM) on Abi-2 expression was determined in LSMC and MSMC. RESULTS: Leiomyomas express elevated levels of Abi-2 as compared with myometrium from the proliferative but not the secretory phase of the menstrual cycle, with a significant reduction following GnRHa therapy (P < 0.05). Western blotting showed a similar trend in Abi-2 protein expression in leiomyoma/myometrial tissue extracts, which was immunolocalized in LSMC and MSMC, connective tissue fibroblasts and arterial walls. The expression of Abi-2 in LSMC and MSMC was increased by TGF-β1 (2.5 ng/ml) and was inhibited by GnRHa (0.1 µM) in a time- and cell-dependent manner, and pretreatment with Smad3 SiRNA and U0126, an MEK-1/2 inhibitor, respectively, reversed their actions. CONCLUSION: Based on the menstrual cycle-dependent expression, the influence of GnRHa therapy, and regulation by TGF-β in LSMC/MSMC, we conclude that Abi-2 may have a key regulatory function in leiomyomas cellular/tissue structural organization during growth and regression.

Introduction

The Abelson (Abl) interactors (Abi) are among a group of proteins that interact with Abl tyrosine kinase, a non-receptor tyrosine kinase whose activation results in the regulation of cytoskeleton function, cell growth and apoptosis in response to a variety of biological stimuli (Pendergast, 2002; Van Etten, 2003; Hernandez et al., 2004). Abi exists as multiple forms in mammalian cells, with Abi-1 and Abi-2 sharing an overall identity of 69% (Dai and Pendergast, 1995; Shi et al., 1995; Dai et al., 1998; Fan et al., 2003; Tani et al., 2003; Lin et al., 2004). Abi is linked to the Abl tyrosine kinase signalling pathway that regulates Abl-mediated transformation, kinase activity and signal transduction from Ras to Rac, small GTP-binding proteins (Shi et al., 1995; Scita et al., 1999; Tani et al., 2003). At the cellular level, Abi proteins are localized to sites of actin polymerization, which is a dynamic event affecting cell morphogenesis and migration (Courtney et al., 2000; Stradal et al., 2001). Abi-1 and Abi-2 have been found to inhibit NIH-3T3 cell transformation, epidermal growth factor (EGF) receptor-mediated cell proliferation and Rac-dependent membrane ruffling in response to platelet-derived growth factor (PDGF) (Dai and Pendergast, 1995; Shi et al., 1995; Ziemnicka-Kotula et al., 1998; Fan and Goff, 2000; Ikeguchi et al., 2001).

Leiomyomas are benign uterine tumours thought to be derived from a transformation of myometrial cells. Because leiomyomas occur during the reproductive years and are suppressed following menopause and in women who receive GnRH analogue (GnRHa) therapy, ovarian steroids play a central role in their growth. In addition to ovarian steroids, several growth factors such as EGF, PDGF and transforming growth factor (TGF)-β as well as the product of many genes involved in cell and tissue structure regulate the growth of leiomyoma (Tsibris et al., 2002; Chegini et al., 2003b; Wang et al., 2003; Chegini, 2005; Luo et al., 2005a,b). We have recently identified Abi-2 among differentially expressed genes in leiomyoma and myometrium and their smooth muscle cells (SMCs) in response to TGF-β1 (Luo et al., 2005a,b). Because of the importance of Abi in Abl tyrosine kinase signalling and Abl-mediated cellular transformation, growth and cell/tissue structure, this protein family may play a major role in the pathogenesis of leiomyoma. We therefore evaluated the expression of Abi-2 in leiomyomas and myometrium during the menstrual cycle and in patients who received GnRHa therapy and its regulation by TGF-β1 and GnRHa direct action in leiomyoma and myometrial smooth muscle cells (LSMC and MSMC). We selected TGF-β1 because the TGF-β system is expressed in leiomyomas and it functions as a key regulator of many genes whose products are critical in cell/tissue structural organization (Dou et al., 1996; Lee and Nowak, 2001; Chegini et al., 2003a; Ding et al., 2004; Luo et al., 2005a). In addition to the GnRHa therapy central action in regressing leiomyoma growth, GnRHa can act directly on leiomyoma and myometrial cells in vitro (Ding et al., 2004; Luo et al., 2005b). Using a specific synthetic inhibitor and SiRNA, we further examined whether TGF-β and GnRHa actions in regulating Abi-2 expression involve mitogen-activated protein kinase (MAPK) and Smad pathways, respectively.

Subjects and methods

Portions of paired leiomyoma and myometrium (n = 27) were collected from premenopausal women, ages ranging from 29 to 41 years, scheduled to undergo hysterectomy for indications related to symptomatic leiomyomas at the University of Florida-affiliated Shands Hospital, Gainesville, FL, USA. Of these patients, seven received GnRHa therapy for a period of 3 months before surgery. The untreated patients did not receive any medications during the previous 3 months before surgery, and on the basis of endometrial histology and the patients’ last menstrual period, they were identified as being from proliferative (n = 8) and secretory (n = 12) phases of the menstrual cycle. Randomly selected portions of leiomyomas of 2–3 cm in diameter and matched myometrium collected distal from the tumours were used in this study. Prior approval was obtained from the University of Florida Institutional Review Board for the experimental protocol of this study. Following collection, total RNA and protein were isolated from these tissues and subjected to real-time PCR and Western blotting or processed for immunohistochemistry and cell culturing as previously described (Xu et al., 2003; Ding et al., 2004).

All the materials for real-time PCR, Western blotting, immunohistochemistry and cell isolation and culturing were purchased from Applied Biosystem (Foster City, CA, USA), BioRad (Hercules, CA, USA), Vector Laboratories (Burlingame, CA, USA), Sigma Chemical (Saint Louis, MO, USA) and Fisher Scientific (Suwanee, GA, USA), respectively, as previously described in detail (Xu et al., 2003; Ding et al., 2004). Polyclonal antibody generated against synthetic Abi-2 and β-actin monoclonal antibody were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) and Sigma Chemical, respectively.

Real-time PCR

Total RNA was isolated from leiomyoma and myometrium using Trizol Reagent (Invitrogen, Carlsbad, CA, USA), and cDNA was generated from 2 µg of total RNA using Taqman reverse transcription reagent. The newly synthesized cDNA was used for PCR performed in 96-well optical reaction plates, with cDNA equivalent to 100 ng of RNA in a volume of 50 µl reaction containing 1× Taqman Universal Master Mix, optimized concentrations of 6-carboxyfluorescein (FAM)-labelled probe and specific forward and reverse primer for Abi-2 selected from Assay on Demand (Applied Biosystem). The results were analysed using a comparative method, with values normalized to 18S rRNA expression and converted into fold change on the basis of a doubling of PCR product in each PCR cycle, as previously described (Ding et al., 2004; Luo et al., 2005a).

Western blotting and immunohistochemistry

For Western blotting, tissue pieces were lysed in a lysis buffer and centrifuged and the supernatants were collected, and their total protein content was determined using a conventional method (Pierce, Rockford, IL, USA), as previously described (Ding et al., 2004). Equal amount of protein were subjected to polyacrylamide gel electrophoresis (PAGE) and transferred to polyvinyldiene difluoride membranes, and following further processing, the blots were incubated with Abi-2 antibody or monoclonal antibody to β-actin (control) for 1 h at room temperature. The blots were then washed with washing buffer and exposed to horse-radish peroxidase-conjugated immunoglobulin G (IgG), and immunostained proteins were visualized using enhanced chemiluminescence reagents (Amersham-Pharmacia Biotech, Piscataway, NJ, USA). The band densities were determined as previously described (Chegeni 2003a; Xu et al., 2003).

For immunohistochemistry, tissue sections were prepared from formalin-fixed and paraffin-embedded leiomyoma and myometrium and, following standard processing, were immunostained using Abi-2 antibody at 5 µg of IgG/ml for 2–3 h at room temperature. Following further standard processing, chromogenic reaction was detected with 3,3′-diaminobenzidine tetrahydrochloride solution (Chegini et al., 2003a). Omission of primary antibody or incubation of tissue sections with non-immune goat IgG instead of primary antibody at the same concentration served as controls.

The expression of Abi-2 in leiomyoma and myometrial smooth muscle cells and regulation by TGF-β and GnRHa

LSMC and MSMC were isolated, cultured and characterized as previously described (Chegini et al., 2002; Xu et al., 2003). The cells were cultured in 6-well plates at an approximate density of 10⁶ cells/well in Dulbecco’s modified Eagle’s medium-supplemented media containing 10% fetal bovine serum (FBS). After reaching visual confluence, the cells were washed in serum-free media and incubated for 24 h under serum-free, phenol red-free condition (Chegini et al., 2002).

The effect of TGF-β1 and GnRHa on Abi-2 expression in LSMC and MSMC was determined by culturing the cells as above and by treating with TGF-β1 (2.5 ng/ml) or GnRHa (0.1 µM) for 2, 6 and 12 h (Ding et al., 2004). To determine whether TGF-β1 and GnRHa mediate their actions through MAPK pathway, LSMC and MSMC were cultured as above and pretreated with a MEK-1/2 inhibitor, U0126 (20 µM), for 2 h and then were treated with TGF-β1 or GnRHa for an additional 2 h (Ding et al., 2004). To determine the involvement of Smad pathway in mediating TGF-β action on Abi-2 expression, LSMC and MSMC were cultured and, at approximately 80% confluence, were transfected with 200 pmol of Smad3 SiRNA using transfectamine 2000 reagent (10 µl) (Invitrogen) for 48 h, as previously described (Levens et al., 2005). The cells were then treated with TGF-β1 for 2 h. Untreated or cells treated with scrambled Smad3 SiRNA were used as a negative control. Total RNA was isolated from the treated and untreated control cells and subjected to real-time PCR.

Where appropriate, the results are expressed as mean ± SEM and are statistically analysed using unpaired Student’s t-test, analysis of variance (ANOVA) and Tukey test. A probability level of P < 0.05 was considered significant.

Results

Expression of Abi-2 in leiomyoma and myometrium

Real-time PCR indicated that leiomyomas and myometrium during the menstrual cycle express Abi-2 mRNA (Figure 1). The relative level of Abi-2 mRNA expression was higher in leiomyomas as compared with myometrium from proliferative (P < 0.05; Figure 1) but not the secretory phase of the menstrual cycle. This was because of an increase in the relative level of Abi-2 expression in myometrium during the secretory phase (Figure 1). The expression of Abi-2 was significantly reduced in leiomyomas and myometrium in patients who received GnRHa therapy, reaching levels observed in myometrium from the proliferative phase (P < 0.05; Figure 1).

Western blot analysis also indicated that leiomyoma and myometrial tissue extracts contain immunoreactive Abi-2 (Figure 2). There was a considerable variation in Abi-2 band density among the paired tissues from proliferative and secretory phases, as well as the GnRHa-treated groups. However, the bands’ mean density displayed a similar trend in these tissues as seen with Abi-2 mRNA expression but did not reach a statistical significance (Figure 2). Immunohistochemically, Abi-2 was localized in leiomyomas (Figure 3A–C) and myometrium (Figure 3D–F) from proliferative (Figure 3A and D) and secretory (Figure 3B and E) phases of the menstrual cycle, as well as in tissues in patients who received GnRHa therapy (Figure 3C and F). Immunostaining was associated with LSMC and MSMC and cellular components of connective tissue and vasculature, with reduced intensity in GnRHa-treated group. Deletion of the primary antibody or replacement with non-immune goat IgGs at the same concentration resulted in reduction in staining intensity associated with these cells (Figure 3G).

Expression of Abi-2 in leiomyoma and myometrial smooth muscle cells and regulation by TGF-β and GnRHa

LSMC and MSMC also expressed Abi-2 mRNA and were regulated by TGF-β1 (2.5 ng/ml) and GnRHa (0.1 µM) in a cell- and time-dependent manner (Figure 4A and B). TGF-β1 increased the expression of Abi-2 in LSMC and MSMC within 2 h of treatment, which reached control levels after 6–12 h (P < 0.05, Figure 4A). The expression of Abi-2 was also increased in MSMC after 2 h of treatment with GnRHa, but it was inhibited in MSMC and LSMC after 6 and 12 h of treatments (Figure 4B, P < 0.05).

Pretreatment of MSMC and LSMC with U0126, a MEK-1/2 inhibitor, or transfection with Smad3 SiRNA, indicated that TGF-β1 and GnRHa actions on Abi-2 expression are mediated, at least in part, through Smad and MAPK pathways. U0126 increased the basal expression of Abi-2 in LSMC but not in MSMC. However, pretreatment with U0126-inhibited TGF-β and reversed GnRHa actions on Abi-2 expression as compared with TGF-β- and GnRHa-treated cells and untreated cells, respectively (Figure 4C; P < 0.05). Transfection of LSMC and MSMC with Smad3 SiRNA also reduced Abi-2 expression induced by TGF-β, reaching levels expressed in controls (Figure 4D; P < 0.05).

Discussion

In the present study, we demonstrated that leiomyomas and myometrium express Abi-2 mRNA and protein. Comparatively, leiomyomas express an elevated level of Abi-2 as compared with myometrium from the proliferative phase but not the secretory phase of the menstrual cycle, due in part to an increased expression in myometrium. GnRHa therapy reduced the expression of Abi-2 in these tissues, reaching levels observed in proliferative phase myometrium. Similarity in the expression of Abi-2 in leiomyomas and myometrium during the secretory phase, and reduction under the hypo-estrogenic condition created by GnRHa therapy, suggests a regulatory function for the ovarian steroids on Abi-2 expression. Although the mechanism(s) involved in regulating the elevated expression of Abi-2 in leiomyoma during the proliferative phase is unclear from our study, it may be due in part to over expression of estrogen receptors. Leiomyomas also express elevated levels of progesterone receptor, and progesterone has been regarded as a key regulator of leiomyoma growth. As such, the ovarian steroids, in a tissue-specific manner, may influence the expression of Abi-2, but the biological significance of Abi-2 expression and the implication of its over expression in leiomyoma require detailed investigation.

The Abi proteins interact with Abl, a non-receptor tyrosine kinase, whose activation results in regulation of cytoskeleton function, cell growth and apoptosis, Abl-mediated transformation and signal transduction from Ras to Rac (Pendergast, 2002; Tani et al., 2003; Van Etten, 2003; Hernandez et al., 2004). At the cellular level, Abi proteins are localized to sites of actin polymerization, which regulates actin dynamics, thus affecting cell morphogenesis and migration (Courtney et al., 2000; Stradal et al., 2001). These activities are critical in leiomyoma growth and regression and are influenced by several growth factors such as EGF, PDGF and TGF-β (Chegini, 2005). We demonstrated that TGF-β, a multifunctional cytokine with a key regulatory action on the expression of genes whose products are involved in cell and tissue structural organization, also regulates the expression of Abi-2 in LSMC and MSMC. We found that TGF-β1 action in regulating the expression of Abi-2 in LSMC is more pronounced as compared with MSMC. Since the expression of TGF-β, TGF-β receptors and their receptor signalling is altered in leiomyomas and LSMC as compared with myometrium and MSMC; functionally, these changes could account for the difference observed in Abi-2 expression in these cells (Chegini et al., 2002, Chegini et al., 2003a; Xu et al., 2003). The selection of TGF-β1 in our study is based on the dominant expression of this isoform in leiomyomas, although other TGF-β isoforms, namely TGF-β3, which has also been shown to be expressed at increased levels in leiomyomas (Lee and Nowak, 2001), could equally regulate the expression of Abi-2. As with the expression of other genes (Ding et al., 2004; Levens et al., 2005; Luo et al., 2005a), the Abi-2 expression was also the target of GnRHa direct action in LSMC and MSMC. GnRHa inhibited the expression of Abi-2 in these cells, despite a moderate increase in expression in MSMC after 2 h, returning to untreated control levels after longer treatment. As such, the results further support a potential direct action for GnRHa at the uterine level, although GnRHa therapy’s major site action in suppressing leiomyoma growth is central.

Our results indicate that TGF-β and GnRHa action in regulating the expression of Abi-2 in LSMC and MSMC mediated, at least in part, through the activation of Smad and MAPK pathways, respectively. Smad and MAPK are major signalling pathways utilized by TGF-β receptors (Shi and Massague, 2003) and GnRH receptor (Cheng and Leung, 2005) in their target cells, including in LSMC and MSMC (Xu et al., 2003; Ding et al., 2004). However, blocking MEK-1/2 activity had a distinct influence on TGF-β and GnRHa actions on Abi-2 expression in LSMC and MSMC. The reasons for these differences are unclear; however, TGF-β and GnRH receptors activate other components of the MAPK pathway, whose crosstalks with MAPK/extracellular signal-regulated protein kinase (ERK) or with multiple other signalling pathways may influence Abi-2 expression in these cells.

Because Abi-2 and related proteins act as key regulators of cytoskeleton function, cell growth and apoptosis and cellular transformation, through these activities, Abi-2 may regulate leiomyomas cellular/tissue structural organization. As such, over expression of Abi-1 has been reported to suppress the transforming activity of v-Abl in NIH-3T3 fibroblasts by inhibiting v-Abl-stimulated ERK activation (Fan and Goff, 2000; Ikeguchi et al., 2001). In addition to Abl kinase, Abi also interacts with Abl-related gene product Arg, adopter protein EGF receptor substrate Eps8, spectrins, guanine nucleotide exchange factor Sos, Grb4 and Wave-1 (Wang et al., 1996; Biesova et al., 1997; Ziemnicka-Kotula et al., 1998; Scita et al., 1999; Cowan and Henkemeyer, 2001; Yamamoto et al., 2001; Eden et al., 2002; Soderling et al., 2002; Kunda et al., 2003; Echarri et al., 2004; Leng et al., 2005). Abi-1, through complex formation with Eps8 and Sos-1, has been reported to regulate Rac-specific guanine nucleotide exchange factor activities, a key process in actin re-organization (Taki et al., 1998; Fan and Goff, 2000; Garcia-Cuellar et al., 2000;     Dai et al., 2001; Pendergast, 2002; Fan et al., 2003). Activation of Rac signalling has been shown to result in degradation of Abi-2 which competes with Abi-1 for complex formation with Sos and Eps8 (Scita et al., 1999; Pendergast, 2002; Fan et al., 2003), whereas blocking Abi-1 activity resulted in abrogation of Rac-dependent membrane ruffling in fibroblasts in response to PDGF stimulation (Courtney et al., 2000). Because several components of Ras–Rac, Eps8, EGF and PDGF are expressed in leiomyomas and myometrium (Rossi et al., 1992; Chegini et al., 2003b; Luo et al., 2005a,b) and Rac1 acts as a signalling partner for TGF-β-mediated cellular transformation, their potential interactions with Abi proteins could influence leiomyomas growth and GnRHa-induced regression.

In summary, we have shown that Abi-2 is expressed in leiomyomas and myometrium in a menstrual cycle-dependent manner and is regulated by TGF-β1 and GnRHa in LSMC and MSMC, at least in part, through the activation of Smad and MAPK pathways. Because Abi-2 and related proteins are key regulators of cytoskeleton functions and cellular transformation, through these activities they may regulate leiomyomas cellular/tissue organization during growth and regression

Acknowledgements

This work was supported by NIH grant HD37432.

References