<<

REPRODUCTIONRESEARCH

Metformin decreases IGF1-induced cell proliferation and protein synthesis through AMP-activated protein in cultured bovine granulosa cells

Lucie Tosca, Christelle Rame´, Christine Chabrolle, Sophie Tesseraud1 and Joe¨lle Dupont Unite´ de Physiologie de la Reproduction et des Comportements, Institut National de la Recherche Agronomique (INRA), UMR85, F-37 380 Nouzilly, France and 1Unite´ de Recherches Avicoles, INRA, URA83, F-37 380 Nouzilly, France Correspondence should be addressed to J Dupont; Email: [email protected]

Abstract

Although its mechanism of action is still unclear, metformin is an anti-diabetic drug effective to restore cyclicity and spontaneous ovulation in women with polycystic ovary syndrome. It may also reduce the risk of cancer. We have recently shown that metformin treatment decreases steroidogenesis through AMP-activated kinase (AMPK) in granulosa cells of various species. Here, we investigated the effects and the molecular mechanisms of metformin in IGF1-induced proliferation and protein synthesis in cultured bovine granulosa cells. Treatment with metformin (10 mM) for 24 h reduced cell proliferation and the levels of cyclin D2 and E, and increased the K associations cyclin D2/p21 and cyclin D2/p27 without affecting cell viability in response to IGF1 (10 8 M). It also decreased IGF1-induced protein synthesis and phosphorylation of P70S6 kinase and ribosomal S6 protein. Interestingly, metformin treatment for 1 h decreased MAPK3/1 (ERK1/2) and P90RSK phosphorylation without affecting AKT phosphorylation in response to IGF1. Adenovirus- mediated expression of dominant-negative AMPK totally abolished the effects of metformin on cell proliferation and phosphorylation of P70S6K in response to IGF1. It also eliminated the inhibitory effects of metformin on MAPK3/1 and P90RSK phosphorylation. Taken together, our results strongly suggest that metformin reduces cell growth, protein synthesis, MAPK3/1, and P90RSK phosphorylation in response to IGF1 through an AMPK-dependent mechanism in cultured bovine granulosa cells. Reproduction (2010) 139 409–418

Introduction insulin action, relatively little is known about the mechanism of action of this compound on the ovarian Metformin, a biguanide, is the most widely used drug for tissue. Several studies have shown in vitro that the treatment of type 2 diabetes but also for the treatment metformin decreases steroids production in response to of the polycystic ovary syndrome (PCOS). This syndrome FSH but also to insulin-like growth factor 1 (IGF1) in affects 5–10% of women of childbearing age (Franks 1995). It is characterized by the presence of two or more granulosa cells of different species including rodent of the following features: chronic oligo-ovulation or (Tosca et al. 2006), human (Mansfield et al. 2003), and anovulation, androgen excess, and polycystic ovaries bovine (Tosca et al. 2007a). Several studies have shown (Franks 1995). In women with PCOS, a long-term that there is an increased granulosa cell proliferation in metformin treatment increases ovulation rates, improves some follicles in women with PCOS (Stubbs et al. 2007, menstrual cyclicity, and reduces serum androgen levels Das et al. 2008). Thus, it is important to determine the (Velazquez et al. 1997, Vandermolen et al. 2001). These effects and the molecular mechanisms of metformin, beneficial effects have been attributed to a reduction in frequently used for the treatment of PCOS, on prolifer- hyperinsulinemia partly due to an increase in peripheral ation and protein synthesis in granulosa cells. In the glucose utilization and a decrease in endogenous present work, we chose to investigate these actions of glucose production by the liver (Bailey 1992, Scheen metformin on cultured bovine granulosa cells, where we 1997). Some evidence indicates that metformin has also have already studied the effects of this drug on significant effects on tumorigenesis and cancer cell steroidogenesis. growth (Evans et al. 2005, Bowker et al. 2006, Ben Here, we studied the effect of metformin on the Sahra et al. 2008). Although the molecular mechanisms proliferation of granulosa cells in response to IGF1. IGF1 of metformin have been studied in tissues such as liver, is a potent mitogenic factor, and it is critical for ovarian muscle, and fat, in relation to glucose homeostasis and follicle development. In transgenic mice lacking IGF1,

q 2010 Society for Reproduction and Fertility DOI: 10.1530/REP-09-0351 ISSN 1470–1626 (paper) 1741–7899 (online) Online version via www.reproduction-online.org Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access 410 L Tosca and others follicle development is arrested at the small antral stage, in IGF1-induced cell proliferation and protein synthesis and mature Graafian follicles are not produced (Baker in response to metformin was AMPK dependent by using et al. 1996, Zhou et al. 1997). IGF1 activates two major an adenovirus-mediated expression of dominant- signal transduction pathways: the phosphoinositide negative AMPK in bovine granulosa cells. 30-OH kinase (PI3K) pathway and the MAP kinase pathway (MAPK pathway, also known as the ERK pathway; Dupont et al. 2003). Some of the downstream Results effectors of PI3K are the serine/threonine protein AKT/PKB and P70/S6 kinase. P70/S6 kinase (P70S6K) is a Effects of the metformin treatment on basal and key involved in the control of protein synthesis IGF1-stimulated cell proliferation in cultured bovine (Dufner & Thomas 1999). Activation of the MAPK granulosa cells pathway by IGF1 requires activation of Ras and is We investigated whether the metformin treatment mediated by a cascade of successive protein phosphory- affected the number of granulosa cells in response to lation reactions involving Raf, MAPK kinase (MEK-1 IGF1 by inducing mitosis or altering cell viability. and -2), and finally MAPK3 and -1 (ERK1 and -2; Pearson [3H]-thymidine incorporation into granulosa cells K et al. 2001). After activation, MAPK3/1 phosphorylates treated with 10 mM metforminGIGF1 (10 8 M) was several downstream elements, including Elk-1 or determined after 24 h of culture. As expected, IGF1 P90RSK. In granulosa cells, activation of the PI3K or treatment significantly increased [3H]-thymidine incor- the MAPK pathway by IGF1 has been reported to poration (P!0.001; Fig. 1A).However,metformin stimulate proliferation, differentiation, and/or cell survi- treatment decreased IGF1-stimulated [3H]-thymidine val (Hu et al. 2004, Tosca et al. 2005, Goto et al. 2009). incorporation (Fig. 1A) by about 20% (P!0.001) without Metformin treatment decreases cell proliferation of affecting the level of [3H]-thymidine incorporation in various cells including breast cancer cells and endome- basal state (in absence of IGF1). These effects of triotic stromal cells (Takemura et al. 2007, Alimova metformin on IGF1-induced cell proliferation were et al. 2009). However, the mechanisms of action by confirmed by evaluating cyclin D2 and cyclin E protein which metformin mediates cell cycle arrest are not levels in response to IGF1 by western blotting (Fig. 1B completely understood. and C). Indeed, metformin treatment reduced IGF1- Several studies have shown both in vivo and in vitro induced cyclin D2 protein levels by 30% (P!0.05, that metformin can act through activation of AMP- Fig. 1B) and IGF1-stimulated cyclin E protein levels by activated protein kinase (AMPK; Zhou et al. 2001, Musi 35% (P!0.05, Fig. 1C). We have also shown that et al. 2002). AMPK is a heterotrimeric enzyme, metformin treatment increased significantly the associ- consisting of one catalytic subunit, a, and two regulatory ation levels between the cyclin-dependent kinase subunits, b and g (Cheung et al. 2000). AMPK activity inhibitors (CDKIs) p21 and p27 with CDK2 in the results partly from phosphorylation of the Thr172 residue presence of IGF1 (P!0.05, Fig. 1D and E). These data of the a-subunit by two known upstream kinases: were observed without any variation of the protein LKB1 and calcium–calmodulin-dependent kinase amount of p21 and p27 whatever the treatment used kinase (CaMKK; Hardie 2004, Birnbaum 2005). AMPK (data not shown). Trypan blue staining demonstrated that is activated by a change in AMP:ATP ratio in response metformin treatment had no effect on cell viability in the to several stimuli, including exercise (Hayashi et al. presence or absence of IGF1 (data not shown). 1998), hypoxia (Mu et al. 2001), hormones (Minokoshi et al. 2002, Yamauchi et al. 2002), and pharmacological drugs such as metformin (Zhou et al.2001)or Effects of the metformin treatment on basal and 5-aminoimidazole-4-carboxamide-riboside-5-phos- IGF1-stimulated protein synthesis in cultured bovine granulosa cells phate (Corton et al. 1995). AMPK is a multisubstrate enzyme well characterized in many tissues, including Cell growth is very often accompanied with an increase the liver, muscle, lung, heart, kidney, brain (Stapleton in protein synthesis. We next determined the effect of et al. 1996), and ovary (Downs et al. 2002, Bilodeau- metformin treatment on IGF1-induced protein synthesis. Goeseels et al. 2007, Tosca et al. 2007a, 2007b). We [14C]-phenylalanine incorporation into granulosa cells K recently characterized AMPK in bovine ovary and treated with 10 mM metforminGIGF1 (10 8 M) was demonstrated that metformin-induced AMPK activation determined after 4 h of culture. As expected, IGF1 decreased MAPK3/1 phosphorylation and steroid treatment significantly increased [14C]-phenylalanine secretion by granulosa cells (Tosca et al. 2007a). incorporation (P!0.001, Fig. 2A). However, metformin The aims of this study were to investigate the effects treatment halved IGF1-stimulated [14C]-phenylalanine of metformin on cultured bovine primary granulosa incorporation (P!0.001, Fig. 2A). To elucidate the cell proliferation and protein synthesis in response molecular mechanism associated with this latter to IGF1 and the underlying molecular mechanisms. result, we determined the effect of metformin on the More precisely, we tested whether the decrease observed IGF1-induced phosphorylation of P70S6 kinase and

Reproduction (2010) 139 409–418 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access Role of metformin in bovine granulosa cells 411

Figure 1 Effect of metformin treatment on IGF1-induced cell proliferation and the expression of several cell cycle components in bovine granulose cells. (A) Effect of metformin on IGF1-induced proliferation. Thymidine incorporation was determined in granulosa cells cultured for 24 h in serum- K free medium in the presence or absence of IGF1 (10 8 M) and 10 mM metformin as described in Materials and Methods. Data are expressed as the percentage of the result obtained from unstimulated cells. Results are expressed as meansGS.E.M. and are representative of at least three independent experiments. Different letters indicate significant differences at P!0.05. (B and C) Effect of metformin on IGF1-induced cyclins D2 and E expression. K Bovine granulosa cells were serum-starved for 16 h and then stimulated with IGF1 (10 8 M) for 6 h (B, for cyclin D2 measurement) or 24 h (C, for cyclin E measurement) in the absence or presence of metformin (10 mM). Cells were then harvested and cyclins D2 and E protein levels were measured by western blotting. Samples contained equal levels of protein, as confirmed by reprobing each membrane with an anti-vinculin . In each panel, immunoreactivity was quantified by scanning densitometry, and the ratio cyclin/vinculin was represented. These results are representative of three independent experiments. Different letters indicate significant differences at P!0.05. (D and E) Effect of metformin on the K association level of CDKIs, p21, and p27 with CDK2. Bovine granulosa cells were serum-starved for 16 h and then stimulated with IGF1 (10 8 M) for 24 h in the absence or presence of metformin (10 mM). CDK2 association was determined with p21 (D) or with p27 (E). CDK2 was immunoprecipitated from whole cell lysates. Samples were then subjected to western blotting with against either p21 or p27. Membranes were then reprobed with aCDK2 to evaluate CDK2 levels in each lane. These results are representative of three similar experiments. Different letters indicate significant differences at P!0.05. ribosomal S6 protein (two key components involved in 2006, 2007a). We next investigated whether the protein synthesis). Bovine granulosa cells were serum- metformin-induced decrease in cell proliferation and deprived for 16 h, and then pre-incubated in the protein synthesis was mediated by AMPK. We used presence or in the absence of metformin (10 mM) for an adenoviral vector (Ad.aDN) to overexpress a various times and then stimulated for 5 min with IGF1 dominant-negative AMPKa1 in bovine granulosa cells K (10 8 M). As expected, IGF1 treatment significantly and treated these cells with IGF1 as indicated in the increased phosphorylation of P70S6K on Thr389 and on Materials and Methods. Granulosa cells were infected Ser424 Thr421 residues, and phosphorylation of S6 with Ad.aDN and Ad.GFP (Ad.GFP as a control) (Fig. 2B). However, this effect was totally abolished when adenovirus (100 pfu) and were analyzed by western cells were pre-incubated with metformin (10 mM) for at blotting for production of mutant and endogenous least 1 h (Fig. 2B). Thus, metformin treatment decreased AMPKa subunits. Ad.aDN proteins significantly attenu- IGF1-induced protein synthesis, and this was associated ated basal and metformin-induced Thr172 phosphoryl- with a strong reduction in P70S6K and S6 ation (Fig. 3A) without affecting cell viability (data not phosphorylation. shown). Cell infection with control (Ad.GFP, 100 pfu) had no effect on AMPKa expression or AMPKa Thr172 phosphorylation (data not shown). Expression of Effects of the overexpression of dominant-negative Ad.aDN in bovine granulosa cells abolished the a AMPK 1 on IGF1-induced cell proliferation and metformin-induced decrease in the cell proliferation in P70S6K phosphorylation in response to metformin the presence of IGF1 (Fig. 3B). It also abolished the treatment in cultured bovine granulosa cells metformin-induced decrease of P70S6K phosphoryl- We have demonstrated that metformin treatment ation in the presence of IGF1 (Fig. 3C). The expression activates AMPK in bovine granulosa cells (Tosca et al. of Ad.GFP in the same conditions did not restore these www.reproduction-online.org Reproduction (2010) 139 409–418

Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access 412 L Tosca and others

Figure 2 Effect of metformin treatment on IGF1-induced protein synthesis in bovine granulosa cells. (A) Effects of metformin on IGF1-induced K phenylalanine incorporation. Bovine granulosa cells were cultured for 24 h in serum-free medium and then stimulated or not with IGF1 (10 8 M) in the absence or presence of metformin (10 mM) for 4 h. Protein synthesis was determined as described in Materials and Methods. Different letters indicate significant differences at P!0.05. (B) Bovine granulosa cells were serum-deprived for 16 h, and then pre-incubated in the presence or in the K absence of metformin (10 mM) for various times (5, 30, 60, and 120 min) and then stimulated for 5 min with IGF1 (10 8 M). Lysates (50 mg) were resolved by SDS-PAGE, transferred to nitrocellulose membrane, and probed with anti-phospho-P70S6K Thr389 and anti-phospho-P70S6K Ser424/Thr421 antibodies. Membranes were then reprobed with P70S6K antibodies to evaluate the ratio phospho-P70S6K Thr389 or Ser424/Thr421/P70S6K in each lane. Membranes were also probed with anti-phospho-S6 Ser235/236 antibodies. In this case, the membranes were reprobed with vinculin antibodies to confirm that samples contained equal levels of protein in each lane. We used vinculin antibodies because we did not find good antibodies against the S6 protein that cross-react with bovine species. Representative blots from three independent experiments are shown. Blots were quantified, and the phosphorylated protein/total protein or pS6/vinculin ratio is shown. The results are represented as means GS.E.M. Different letters indicate significant differences at P!0.05. inhibitory effects of metformin (Fig. 3B and C). Thus, restore these inhibitory effects of metformin (Fig. 4Cand metformin reduces proliferation and P70S6 kinase D). Thus, metformin reduces IGF1-induced MAPK3/1 phosphorylation through the activation of AMPKa in and P90RSK phosphorylation through the activation of bovine granulosa cells. AMPKa in bovine granulosa cells. We did not observe effects of metformin on IGF1-induced AKT phosphoryl- ation (data not shown). Effects of the metformin treatment on IGF1-stimulated MAPK3/1 and p90RSK: potential involvement of AMPK Discussion We also determined the effects of metformin on IGF1- We have recently shown that metformin decreases stimulated MAPK3/1 and P90RSK and tested a potential steroidogenesis through AMPK in the presence or K K implication of AMPK by using Ad.aDN. Bovine absence of FSH (10 8 M) and IGF1 (10 8 M) in cultured granulosa cells were serum-deprived for 16 h, and then bovine granulosa cells (Tosca et al. 2007a). However, the pre-incubated in the presence or in the absence of effects and the mechanisms through which metformin metformin (10 mM) for various times and then stimulated affects cell growth and protein synthesis have never been K8 for 5 min with IGF1 (10 M). IGF1 treatment alone defined in these granulosa cells. In the present paper, we significantly increased the phosphorylation of MAPK3/1 have shown that metformin inhibits proliferation and and P90RSK (Fig. 4A and B). This effect was totally protein synthesis in response to IGF1 through an AMPK- abolished when cells were pre-incubated with metfor- dependent mechanism in bovine granulosa cells (Fig. 5). min (10 mM) for at least 1 h (Fig. 4A and B). Moreover, At the molecular level, the inhibitory effect of metformin expression of Ad.aDN totally abolished the metformin- on IGF1-induced cell proliferation is associated with a induced decrease of MAPK3/1 and P90RSK phosphoryl- decrease in the cyclins D2 and E protein amount, and an ation in the presence of IGF1 (Fig. 4C and D). The increase in the association level between CDK2 and the expression of Ad.GFP in the same conditions did not CDKIs, p21 and p27. Moreover, the metformin-induced

Reproduction (2010) 139 409–418 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access Role of metformin in bovine granulosa cells 413

Figure 3 Metformin-reduced proliferation and P70S6K phosphorylation in the presence of IGF1 in bovine granulosa cells infected with adenovirus expressing AMPKa1 mutant constructs. (A) The expression and phosphorylation levels of AMPKa in bovine granulosa cells infected with adenovirus constructs expressing mutant AMPKa1 (Ad.aDN) or the GFP protein (Ad.GFP). Granulosa cells were infected with 100 Pfu/cell of the viruses and K were stimulated 16 h later in the presence or not of metformin 10 mMGIGF1 10 8 M for 2 h. Cells were then prepared, and the lysates were resolved by SDS-PAGE, transferred to nitrocellulose membrane, probed with anti-phospho-AMPKa Thr172, and then reprobed with anti-AMPKa1/2 antibodies. Blots were quantified, and the phosphorylated protein/total protein ratio is shown. The immunoblots shown are representative of three independent experiments. (B and C) Thymidine incorporation (B) and P70S6K Thr389 phosphorylation (C) in granulosa cells expressing a1-DN or K GFP in the presence or absence of metformin 10 mMGIGF1 10 8 M. (B) Granulosa cells were infected or not with 100 Pfu/cell of Ad.aDN or GFP K virus for 16 h, and stimulated or not with metformin 10 mMGIGF1 10 8 M for 24 h. Thymidine incorporation was then determined as described in the Materials and Methods. The data shown represent meansGS.E.M. from three independent experiments. Data are expressed as the percentage of the results obtained from unstimulated cells. Different letters indicate significant differences at P!0.05. (C) Granulosa cells were infected or not with 100 Pfu/cell of Ad.aDN or GFP virus for 16 h, and then pre-incubated in the presence or in the absence of metformin (10 mM) for 60 min and then K stimulated for 5 min with IGF1 (10 8 M). Lysates (50 mg) were resolved by SDS-PAGE, transferred to nitrocellulose membrane, and probed with anti- phospho-P70S6K Thr389 and then reprobed with anti-P70S6K total antibodies. Representative blots from three independent experiments are shown. Blots were quantified, and the phosphorylated protein/total protein ratio is shown. The results are represented as meansGS.E.M. Different letters indicate significant differences at P!0.05. decrease in the protein synthesis in response to IGF1 is shown that metformin inhibits IGF1-induced cyclins D2 associated with a reduction in P70S6K phosphorylation. and E protein expression, suggesting that it inhibits cell Metformin through AMPK activation has also inhibited cycle progression from G0/G1 into S phase in bovine IGF1-induced phosphorylation of MAPK3/1 and granulosa cells. Cyclin D2 is an important regulator of P90RSK, two kinases involved in the cell growth and the cell cycle in granulosa cells. Indeed, cyclin D2- protein synthesis (Fig. 5). deficient females are sterile, ovarian granulosa cells In our study, we used bovine granulosa cells cultured being unable to proliferate (Sicinski et al. 1996). for 48 h in the presence of serum (10%). These culture Downregulation of cyclin D1 in response to metformin conditions are luteinizing conditions for this cell type. has also been demonstrated in LNCaP prostate cancer Thus, our cultured cells are more representative cells cells and in human breast cancer cell lines (Ben Sahra from the corpus luteum than from the growing follicles. et al. 2008, Zhuang & Miskimins 2008). In mammalian The fact that metformin treatment reduces cell prolifer- cells, the rate of transit through G1 phase is regulated by ation induced by IGF1 in cultured bovine granulosa cells coordinated action of CDKs in association with specific is in good agreement with some results observed both regulatory cyclin proteins. The primary known regulators in vivo and in vitro in other cell types. Indeed, metformin of G1 progression are the D-type cyclins (D1, D2, and has inhibitory effects on tumor cell growth in vitro and D3), cyclin E, and their catalytic partners CDKs-2, -4, in vivo (Evans et al. 2005, Bowker et al. 2006, Buzzai and -6 (Dulic et al. 1992, Sherr 1993, 1994). The activity et al. 2007, Phoenix et al. 2009). In our study, we have of these CDKs is regulated by changes in cyclin www.reproduction-online.org Reproduction (2010) 139 409–418

Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access 414 L Tosca and others

Figure 4 Effects of metformin on IGF1-induced MAPK3/1 and P90RSK phosphorylation in bovine granulosa cells infected or not with adenovirus expressing AMPKa1 mutant con- structs. (A and B) Bovine granulosa cells were serum-deprived for 16 h, and then pre-incu- bated in the presence or in the absence of metformin (10 mM) for various times (5, 30, 60 and 120 min) and then stimulated for K 5 min with IGF1 (10 8 M). Lysates (50 mg) were resolved by SDS-PAGE, transferred to nitrocellulose membrane, and probed with anti-phospho-MAPK3/1 and then reprobed with MAPK1 antibodies (A) or with anti-phospho-P90RSK and then reprobed with P90RSK total (B) antibodies. Representative blots from three independent experiments are shown. Blots were quantified, and the phosphorylated protein/total protein ratio is shown. The results are represented as means GS.E.M. Different letters indicate significant differences at P!0.05. (C and D) Granulosa cells were infected or not with 100 Pfu/cell of Ad.aDN or GFP virus for 16 h, and then pre-incubated in the presence or in the absence of metformin (10 mM) for 60 min and K then stimulated for 5 min with IGF1 (10 8 M). Lysates were treated as described above. The results are represented as meansGS.E.M. Different letters indicate significant differences at P!0.05. levels, interaction with CDKIs, and by regulatory growth inhibition of MCF7 cells in response to phosphorylation (Sherr & Roberts 1999). Two distinct metformin treatment. Moreover, the compound C, a classes of cellular CDKIs are known, the Cip/Kip and specific inhibitor of AMPK, blocked metformin-induced Ink4 families. The Cip/Kip family includes p21 and p27, downregulation of cyclin D1 in MCF7 cells. Thus, which inhibit the formation of cyclin D-CDK4/6 and metformin-induced AMPK activation could inhibit the cyclin E-CDK2 complexes and consequently the pro- expression of cyclins D2 and E expression and gression of the cell cycle. p27 and p21 play crucial roles consequently cell proliferation in response to IGF1 in in mammalian ovarian development and more precisely bovine granulosa cells. AKT and MAPK3/1 are the main in the proliferation of granulosa cells (Jirawatnotai et al. signaling pathways that control the growth of 2003, Rajareddy et al. 2007). In the present paper, we the granulosa cells in response to IGF1 (Hu et al. 2004, have observed that metformin increased the association Goto et al. 2009). In our study, metformin did not affect CDK2/p21 or CDK2/p27 in the presence of IGF1 without IGF1-induced AKT phosphorylation (data not shown). In altering the protein level of p21 and p27. Active CDK– contrast, we have observed that metformin decreased cyclin complexes are essential for the activation of IGF1-induced MAPK3/1 and P90RSK phosphorylation, required for progression through the S phase (Weinberg and this was totally abolished when cells were infected 1995). Thus, metformin could inhibit IGF1-induced cell with a dominant-negative AMPK adenovirus. P90RSK proliferation by blocking cyclins D2 and E expression is one of the downstream elements phosphorylated by and inhibiting the activity of the cyclin/CDK complexes. MAPK3/1. We have previously shown that metformin- In bovine granulosa cells, we have shown that the induced AMPK activation reduced steroidogenesis in metformin-induced decrease in proliferation in response bovine granulosa, and this was associated with an to IGF1 was AMPK dependent by using a dominant- inhibition of the MAPK3/1 phosphorylation (Tosca negative AMPK adenovirus. Zakikhani et al. (2006) et al. 2007a). Thus, in these cells, metformin through have also shown that siRNAs targeting AMPK prevent AMPK activation and inhibition of MAPK3/1 could

Reproduction (2010) 139 409–418 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access Role of metformin in bovine granulosa cells 415

differentiation. In the present study, we have shown that metformin decreased protein synthesis in response to IGF1, and this was associated with an inhibition of P70S6K. Moreover, this inhibition was AMPK depen- dent. In response to growth factors, the serine/threonine P70S6K phosphorylates several proteins, including eukaryotic initiation factor 4B (EIF4B) and ribosomal protein S6. It also induces dephosphorylation of the eukaryotic elongation factor 2 (EEF2), therefore affecting both the initiation and elongation stages of mRNA translation (Dufner & Thomas 1999, Proud 2006). P70S6K possesses different sites of phosphorylation. One of them, T389 in the linker domain, is considered as critical for kinase function (Dennis et al. 1998). Phosphorylation of T421 and S424 located in the autoinhibitory domain of P70S6K is also supposed to be required for opening the kinase into a fully active conformation, which allows the sequential phosphoryl- ation of other phosphorylation sites, such as T389. Phosphorylation of all these sites of P70S6K was Figure 5 Schematic representation of the different signaling pathways inhibited by metformin in bovine granulosa cells in the involved in the metformin effects on IGF1-induced proliferation and presence of IGF1. These results are in good agreement protein synthesis. Metformin treatment (10 mM) activates AMPK and with results observed in breast cancer cells. Indeed, consequently inhibits MAPK3/1, P70S6K, and P90RSK phosphorylation metformin through AMPK activation inhibited protein induced by IGF1. The inhibition of these different pathways leads to a synthesis in MCF-7 cells, and this was associated with an reduction in proliferation and protein synthesis in bovine granulosa cells. The arrow from metformin to AMPK represents a positive effect, inhibition of the P70S6 kinase phosphorylation (Dowling and effects downstream of AMPK are negative effects on IGF-mediated et al. 2007). Metformin has also been reported to cause pathways. an inhibition of protein synthesis in cardiac myocytes stimulated with phenylephrine (Chan et al. 2004). Some decrease steroid production and cell proliferation in studies demonstrated the importance of MAPK3/1 response to IGF1. signaling in the stimulation of P70S6K (Iijima et al. Some studies indicate that there are significant 2002) and protein synthesis by growth factors (Toku- differences in the rate of cell death and proliferation in dome et al. 2004). Thus, metformin could inhibit protein granulosa cell populations in PCOS patients (Das et al. synthesis in bovine granulosa cells through inhibition of 2008). The ovarian phenotype in PCOS is characterized MAPK3/1 and P90RSK (Fig. 5). by an increased number of small antral follicles (Stubbs In conclusion, we showed that activation of AMPK by et al. 2007). The follicles frequently contain an increased metformin in bovine granulosa cells leads to a reduction number of steroidogenic cells in the theca interna, in IGF1-induced proliferation and protein synthesis although they are not present in all follicles. There is probably due to an inhibition in the cyclins D2 and E also an increased granulosa cell proliferation. Therefore, protein expression and phosphorylation levels of it is clear that there are abnormalities of proliferation and P70S6K, MAPK3/1, and P90RSK. Our findings have differentiation in both the theca and granulosa compart- significant implications for the understanding of the ments of the polycystic ovary. Metformin treatment is mechanism of action of metformin on ovarian cells. now widely prescribed in women with PCOS. Several However, further investigations are needed in human studies have demonstrated that metformin is able to exert ovarian cells to determine whether metformin through direct effects on the ovary, inhibiting progesterone AMPK activation is important in the treatment of PCOS and estradiol production by human granulosa cells and/or cancer. (Mansfield et al. 2003, Rice et al. 2009). However, the effects and the molecular mechanism of metformin on the proliferation of granulosa cell are unknown. In our study, Materials and Methods we have shown that metformin inhibits proliferation and protein synthesis in response to IGF1 through an AMPK- Hormones and reagents dependent mechanism in bovine granulosa cells. Thus, Recombinant human IGF1 and metformin were from Sigma metformin treatment could reduce these abnormalities of Chemical Co. McCoy 5A modified culture medium, penicillin, proliferation in granulosa cells in women with PCOS. streptomycin were purchased from Invitrogen. Thymidine Protein synthesis is required for many cellular methyl-H3 and phenylalanine C14 were purchased from Perkin processes, including growth, proliferation, and Elmer Life and Technological Sciences (Boston, MA, USA). www.reproduction-online.org Reproduction (2010) 139 409–418

Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access 416 L Tosca and others

Antibodies (4 h), cells were washed once in ice-cold PBS, and then in ice-cold 10% TCA. TCA-insoluble material was washed thrice Rabbit polyclonal antibodies to phospho-AMPKa (Thr172), with 10% TCA and solubilized in 0.5 M NaOH at 37 8C for phospho-MAPK3/1 (Thr202/Tyr204), AKT, phospho-P70S6 determination of protein (BCA, Pierce, Fisher Scientific, kinase (Thr389), phospho-P70S6 kinase (Ser424/Thr421), Strasbourg, France) and radioactivity incorporated into granu- phospho-S6 (Ser235/236), phospho-P90RSK (Thr359/Ser363), losa cell protein. Protein synthesis was expressed as nmol and P90RSK were purchased from New England Biolabs Inc. phenylalanine incorporated per mg protein/h (Dardevet et al. (Beverly, MA, USA). Rabbit polyclonal antibodies to MAPK1 1996) and then as the percentage of the result obtained from (C14), cyclin D2 (C17), cyclin E (C19), p21 p27, CDK2, and unstimulated cells. phospho-AKT (Ser 473) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit polyclonal antibodies to AMPKa1/2 were obtained from Upstate Bio- Adenoviruses and infection of bovine granulosa cells technology Inc. (Lake, Placid, NY, USA). Mouse monoclonal antibody to vinculin was obtained from Sigma. All antibodies Dominant-negative AMPK adenovirus was constructed from a were used at 1/1000 dilution in western blotting. AMPK 1 carrying the Asp-157 to Ala (D157A) mutation as previously described (Woods et al. 2000). Recombinant adenovirus was propagated in HEK293 cells, purified by Animals, isolation, and culture of bovine granulosa cells cesium chloride density centrifugation, and stored as pre- viously described (Woods et al. 2000). Bovine granulosa cells All procedures were approved by the Agricultural Agency and were pre-infected with 100 pfu/cell adenovirus in serum- the Scientific Research Agency, and conducted in accordance starved McCoy’s 5A during 16 h, and then cultured for 24 h with the guidelines for Care and Use of Agricultural Animals in with 100 pfu/cell adenovirus and [3H]-thymidine in the K Agricultural Research and Teaching. presence or absence of IGF1 (10 8 M) and/or metformin Adult cow ovaries were collected at the slaughterhouse, (10 mM) for cell proliferation measurement or for different time removed aseptically, and transferred to culture medium. as indicated in the legends of figures. Preliminary studies ! Granulosa cells from small follicles ( 6 mm) were harvested revealed that within 16 h of infection (100 pfu/cell) with a by puncturing the follicles, allowing expulsion of the cells. green fluorescent protein (GFP)-expressing virus, 70% of Cells were recovered by centrifugation, washed with fresh granulosa cells expressed GFP. medium, and counted in a hemocytometer. The culture medium used was McCoy’s 5A supplemented with 20 mmol/l Hepes, penicillin (100 U/ml), streptomycin (100 mg/l), Immunoprecipitation and immunoblot L-glutamine (3 mmol/l), 0.1% BSA, 0.1 mmol/l androstene- Total proteins from granulosa cells were extracted on ice in dione, 5 mg/l transferrin, 20 mg/l selenium, and 10% FBS. The lysis buffer A (10 mM Tris (pH 7.4), 150 mM NaCl, 1 mM cells were initially cultured for 48 h with no other treatment EDTA, 1 mM EGTA, 0.5% Igepal) containing various protease and then incubated in fresh culture medium with or without inhibitors (2 mM PMSF, 10 mg/ml leupeptin, 10 mg/ml test reagents for the appropriate time. Even if the cells collected aprotinin) and phosphatase inhibitors (100 mM sodium were granulosa cells, after 48 h of culture, the cells have fluoride, 10 mM sodium pyrophosphate, 2 mM sodium ortho- probably become luteinized granulosa cells. All cultures were vanadate (Sigma)). Lysates were centrifuged at 13 000 g performed under a water-saturated atmosphere of 95% for 20 min at 4 8C, and the protein concentration in the air/5% CO at 37 8C. 2 supernatants was determined using a colorimetric assay (kit BC Assay, Uptima Interchim, Montluc¸on, France). Thymidine incorporation into granulosa cells After normalization for protein concentration (500 mg), CDK2 proteins were immunoprecipitated from the super- ! 5 Granulosa cells (2 10 viable cells/500 ml) were cultured in natants using 5 mg of the appropriate antibody at 4 8C 24-well dishes in McCoy’s 5A medium and 10% FBS during overnight. The immunocomplexes were precipitated with 48 h, and were then serum-starved for 24 h followed by the 40 ml protein A-agarose for 1 h at 4 8C. After two sequential 3 addition of 1 mCi/mlof[H]-thymidine in the presence or washes using buffer A with a 1/2 dilution, the resulting pellets K8 absence of metformin (10 mM) and/or IGF1 (10 M). Cultures were boiled for 4 min in reducing Laemmli buffer containing were maintained at 37 8C under 5% CO2 in air. After 24 h of 80 mM dithiothreitol. Proteins were resolved by SDS-PAGE and culture, excess of thymidine was removed by washing twice transferred to nitrocellulose membranes. with PBS, fixed with cold trichloroacetic acid (TCA) 50% for Cell extracts (50 mg) or immunoprecipitated proteins were 15 min, and lysed by 0.5 mol/l NaOH. The radioactivity was subjected to electrophoresis on 12% (w:v) SDS-polyacrylamide determined by scintillation fluid (Packard Bioscience, Perkin gel under reducing conditions. The proteins were then Elmer, Courtaboeuf, France) counting in a b-photomultiplier. electrotransferred onto nitrocellulose membranes (Schleicher and Schuell, Ecquevilly, France) for 2 h 30 min at 80 V. Membranes were incubated for 1 h at room temperature with Phenylalanine C14 incorporation into granulosa cells Tris-buffered saline (TBS, 2 mM Tris–HCl, pH 8.0, 15 mM In order to measure protein synthesis in granulosa cells, NaCl, pH 7.6), containing 5% nonfat dry milk powder 14 L-[U- C]phenylalanine (0.25 mCi/ml) was added to the (NFDMP) and 0.1% Tween-20 to saturate nonspecific sites. medium during the stimulation period. At the end of incubation Then, membranes were incubated overnight at 4 8C with

Reproduction (2010) 139 409–418 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access Role of metformin in bovine granulosa cells 417 appropriate primary antibodies (final dilution 1:1000), in TBS Chan AY, Soltys CL, Young ME, Proud CG & Dyck JR 2004 Activation of containing 0.1% Tween-20 and 5% NFDMP. After washing in AMP-activated protein kinase inhibits protein synthesis associated with TBS–0.1% Tween-20, the membranes were incubated for 2 h at hypertrophy in the cardiac myocyte. Journal of Biological Chemistry 279 32771–32779. room temperature with a HRP-conjugated anti-rabbit or anti- Cheung PC, Salt IP, Davies SP, Hardie DG & Carling D 2000 mouse IgG (final dilution 1:10 000; Diagnostic Pasteur, Characterization of AMP-activated protein kinase gamma-subunit Marnes-la-Coquette, France) in TBS–0.1% Tween-20. After isoforms and their role in AMP binding. Biochemical Journal 346 washing in TBS–0.1% Tween-20, the signal was detected by 659–669. ECL (ECL, Amersham Pharmacia Biotech). The films were Corton JM, Gillespie JG, Hawley SA & Hardie DG 1995 5-Aminoimida- zole-4-carboxamide ribonucleoside. A specific method for activating analyzed and signals quantified with the software Scion Image AMP-activated protein kinase in intact cells? European Journal of (4.0.3.2 version; Scion Corporation, Frederick, MD, USA). Biochemistry 229 558–565. Dardevet D, Sornet C, Vary T & Grizard J 1996 Phosphatidylinositol 3-kinase and p70 s6 kinase participate in the regulation of protein Statistical analysis turnover in skeletal muscle by insulin and insulin-like growth factor I. Endocrinology 137 4087–4094. All experimental data are presented as meansGS.E.M. One-way Das M, Djahanbakhch O, Hacihanefioglu B, Saridogan E, Ikram M, ANOVA was used to test differences. If ANOVA revealed Ghali L, Raveendran M & Storey A 2008 Granulosa cell survival and significant effects, the means were compared by Newman’s proliferation are altered in polycystic ovary syndrome. Journal of Clinical test, with P!0.05 considered significant. In the various graphs, Endocrinology and Metabolism 93 881–887. Dennis PB, Pullen N, Pearson RB, Kozma SC & Thomas G 1998 different letters indicate significant differences. Phosphorylation sites in the autoinhibitory domain participate in p70(s6k) activation loop phosphorylation. Journal of Biological Chemistry 273 14845–14852. Declaration of interest Dowling RJ, Zakikhani M, Fantus IG, Pollak M & Sonenberg N 2007 Metformin inhibits mammalian target of rapamycin-dependent trans- The authors declare that there is no conflict of interest that lation initiation in breast cancer cells. Cancer Research 67 could be perceived as prejudicing the impartiality of the 10804–10812. research reported. Downs SM, Hudson ER & Hardie DG 2002 A potential role for AMP- activated protein kinase in meiotic induction in mouse oocytes. Developmental Biology 245 200–212. Funding Dufner A & Thomas G 1999 signaling and the control of translation. Experimental Cell Research 253 100–109. Lucie Tosca is a PhD student supported by the Re´gion Centre. Dulic V, Lees E & Reed SI 1992 Association of human cyclin E with a This work was supported by the GIS-AGENAE Programme, periodic G1-S phase protein kinase. Science 257 1958–1961. ANR and APIS-. Dupont J, Dunn SE, Barrett JC & LeRoith D 2003 Microarray analysis and identification of novel molecules involved in insulin-like growth factor-1 receptor signaling and . Recent Progress in Hormone Acknowledgements Research 58 325–342. Evans JM, Donnelly LA, Emslie-Smith AM, Alessi DR & Morris AD 2005 We thank Pascal Ferre´ and Fabienne Foufelle for the AMPKa Metformin and reduced risk of cancer in diabetic patients. BMJ 330 dominant-negative adenovirus. 1304–1305. Franks S 1995 Polycystic ovary syndrome. New England Journal of Medicine 333 853–861. Goto M, Iwase A, Harata T, Takigawa S, Suzuki K, Manabe S & Kikkawa F References 2009 IGF1-induced AKT phosphorylation and cell proliferation are suppressed with the increase in PTEN during luteinization in human Alimova IN, Liu B, Fan Z, Edgerton SM, Dillon T, Lind SE & Thor AD 2009 granulosa cells. Reproduction 137 835–842. Metformin inhibits breast cancer cell growth, colony formation and Hardie DG 2004 The AMP-activated protein kinase pathway – new players induces cell cycle arrest in vitro. Cell Cycle 8 909–915. upstream and downstream. Journal of Cell Science 117 5479–5487. Bailey CJ 1992 Biguanides and NIDDM. Diabetes Care 15 755–772. Hayashi T, Hirshman MF, Kurth EJ, Winder WW & Goodyear LJ 1998 Baker J, Hardy MP, Zhou J, Bondy C, Lupu F, Bellve AR & Efstratiadis A 0 Evidence for 5 AMP-activated protein kinase mediation of the effect of 1996 Effects of an Igf1 gene null mutation on mouse reproduction. muscle contraction on glucose transport. Diabetes 47 1369–1373. Molecular Endocrinology 10 903–918. Hu CL, Cowan RG, Harman RM & Quirk SM 2004 Cell cycle progression Ben Sahra I, Laurent K, Loubat A, Giorgetti-Peraldi S, Colosetti P, and activation of Akt kinase are required for insulin-like growth factor Auberger P, Tanti JF, Le Marchand-Brustel Y & Bost F 2008 The I-mediated suppression of apoptosis in granulosa cells. Molecular antidiabetic drug metformin exerts an antitumoral effect in vitro and in vivo through a decrease of cyclin D1 level. Oncogene 27 3576–3586. Endocrinology 18 326–338. Bilodeau-Goeseels S, Sasseville M, Guillemette C & Richard FJ 2007 Effects Iijima Y, Laser M, Shiraishi H, Willey CD, Sundaravadivel B, Xu L, of adenosine monophosphate-activated kinase activators on bovine McDermott PJ & Kuppuswamy D 2002 c-Raf/MEK/ERK pathway controls oocyte nuclear maturation in vitro. Molecular Reproduction and -mediated p70S6K activation in adult cardiac muscle Development 74 1021–1034. cells. Journal of Biological Chemistry 277 23065–23075. Birnbaum MJ 2005 Activating AMP-activated protein kinase without AMP. Jirawatnotai S, Moons DS, Stocco CO, Franks R, Hales DB, Gibori G & Molecular Cell 19 289–290. Kiyokawa H 2003 The cyclin-dependent kinase inhibitors p27Kip1 and Bowker SL, Majumdar SR, Veugelers P & Johnson JA 2006 Increased p21Cip1 cooperate to restrict proliferative life span in differentiating cancer-related mortality for patients with type 2 diabetes who use ovarian cells. Journal of Biological Chemistry 278 17021–17027. sulfonylureas or insulin: response to Farooki and Schneider. Diabetes Mansfield R, Galea R, Brincat M, Hole D & Mason H 2003 Metformin has Care 29 1990–1991. direct effects on human ovarian steroidogenesis. Fertility and Sterility 79 Buzzai M, Jones RG, Amaravadi RK, Lum JJ, DeBerardinis RJ, Zhao F, 956–962. Viollet B & Thompson CB 2007 Systemic treatment with the antidiabetic Minokoshi Y, Kim YB, Peroni OD, Fryer LG, Muller C, Carling D & Kahn BB drug metformin selectively impairs p53-deficient tumor cell growth. 2002 Leptin stimulates fatty-acid oxidation by activating AMP-activated Cancer Research 67 6745–6752. protein kinase. Nature 415 339–343. www.reproduction-online.org Reproduction (2010) 139 409–418

Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access 418 L Tosca and others

Mu J, Brozinick JT Jr, Valladares O, Bucan M & Birnbaum MJ 2001 A role Tosca L, Froment P, Solnais P, Ferre P, Foufelle F & Dupont J 2005 Adenosine for AMP-activated protein kinase in contraction- and hypoxia-regulated 50-monophosphate-activated protein kinase regulates progesterone glucose transport in skeletal muscle. Molecular Cell 7 1085–1094. secretion in rat granulosa cells. Endocrinology 146 4500–4513. Musi N, Hirshman MF, Nygren J, Svanfeldt M, Bavenholm P, Rooyackers O, Tosca L, Solnais P, Ferre P, Foufelle F & Dupont J 2006 Metformin-induced 0 Zhou G, Williamson JM, Ljunqvist O, Efendic S et al. 2002 Metformin stimulation of adenosine 5 monophosphate-activated protein kinase increases AMP-activated protein kinase activity in skeletal muscle of (PRKA) impairs progesterone secretion in rat granulosa cells. Biology of subjects with type 2 diabetes. Diabetes 51 2074–2081. Reproduction 75 342–351. Pearson G, Robinson F, Beers Gibson T, Xu BE, Karandikar M, Berman K & Tosca L, Chabrolle C, Uzbekova S & Dupont J 2007a Effects of metformin Cobb MH 2001 -activated protein (MAP) kinase pathways: on bovine granulosa cells steroidogenesis: possible involvement of 0 regulation and physiological functions. Endocrine Reviews 22 153–183. adenosine 5 monophosphate-activated protein kinase (AMPK). Biology Phoenix KN, Vumbaca F & Claffey KP 2009 Therapeutic metformin/AMPK of Reproduction 76 368–378. 0 activation promotes the angiogenic phenotype in the ERalpha negative Tosca L, Uzbekova S, Chabrolle C & Dupont J 2007b Possible role of 5 MDA-MB-435 breast cancer model. Breast Cancer Research and AMP-activated protein kinase in the metformin-mediated arrest of bovine Treatment 113 101–111. oocytes at the germinal vesicle stage during in vitro maturation. Biology Proud CG 2006 Regulation of protein synthesis by insulin. Biochemical and of Reproduction 77 452–465. Biophysical Research Communications 34 213–216. Vandermolen DT, Ratts VS, Evans WS, Stovall DW, Kauma SW & Nestler JE Rajareddy S, Reddy P, Du C, Liu L, Jagarlamudi K, Tang W, Shen Y, 2001 Metformin increases the ovulatory rate and pregnancy rate from Berthet C, Peng SL, Kaldis P et al. 2007 p27kip1 (cyclin-dependent clomiphene citrate in patients with polycystic ovary syndrome who are kinase inhibitor 1B) controls ovarian development by suppressing follicle resistant to clomiphene citrate alone. Fertility and Sterility 75 310–315. endowment and activation and promoting follicle atresia in mice. Velazquez E, Acosta A & Mendoza SG 1997 Menstrual cyclicity after Molecular Endocrinology 21 2189–2202. metformin therapy in polycystic ovary syndrome. Obstetrics and Rice S, Pellatt L, Ramanathan K, Whitehead SA & Mason HD 2009 Gynecology 90 392–395. Metformin inhibits aromatase via an extracellular signal-regulated Weinberg RA 1995 The retinoblastoma protein and cell cycle control. Cell 81 323–330. kinase-mediated pathway. Endocrinology 150 4794–4801. Woods A, Azzout-Marniche D, Foretz M, Stein SC, Lemarchand P, Ferre P, Scheen AJ 1997 Drug treatment of non-insulin-dependent diabetes mellitus Foufelle F & Carling D 2000 Characterization of the role of AMP- in the 1990s. Achievements and future developments. Drugs 54 activated protein kinase in the regulation of glucose-activated gene 355–368. expression using constitutively active and dominant negative forms of the Sherr CJ 1993 Mammalian G cyclins. Cell 73 1059–1065. 1 kinase. Molecular and Cellular Biology 20 6704–6711. Sherr CJ 1994 G phase progression: cycling on cue. Cell 79 551–555. 1 Yamauchi T, Kamon J, Minokoshi Y, Ito Y, Waki H, Uchida S, Yamashita S, Sherr CJ & Roberts JM 1999 CDK inhibitors: positive and negative Noda M, Kita S, Ueki K et al. 2002 Adiponectin stimulates glucose regulators of G -phase progression. Genes and Development 13 1 utilization and fatty-acid oxidation by activating AMP-activated protein 1501–1512. kinase. Nature Medicine 8 1288–1295. Sicinski P, Donaher JL, Geng Y, Parker SB, Gardner H, Park MY, Robker RL, Zakikhani M, Dowling R, Fantus IG, Sonenberg N & Pollak M 2006 Richards JS, McGinnis LK, Biggers JD et al. 1996 Cyclin D2 is an FSH- Metformin is an AMP kinase-dependent growth inhibitor for breast responsive gene involved in gonadal cell proliferation and oncogenesis. cancer cells. Cancer Research 66 10269–10273. Nature 384 470–474. Zhou J, Kumar TR, Matzuk MM & Bondy C 1997 Insulin-like growth factor I Stapleton D, Mitchelhill KI, Gao G, Widmer J, Michell BJ, Teh T, regulates gonadotropin responsiveness in the murine ovary. Molecular House CM, Fernandez CS, Cox T, Witters LA et al. 1996 Mammalian Endocrinology 11 1924–1933. AMP-activated protein kinase subfamily. Journal of Biological Chemistry Zhou G, Myers R, Li Y, Chen Y, Shen X, Fenyk-Melody J, Wu M, Ventre J, 271 611–614. Doebber T, Fujii N et al. 2001 Role of AMP-activated protein kinase in Stubbs SA, Stark J, Dilworth SM, Franks S & Hardy K 2007 Abnormal mechanism of metformin action. Journal of Clinical Investigation 108 preantral folliculogenesis in polycystic ovaries is associated with 1167–1174. increased granulosa cell division. Journal of Clinical Endocrinology Zhuang Y & Miskimins WK 2008 Cell cycle arrest in metformin treated and Metabolism 92 4418–4426. breast cancer cells involves activation of AMPK, downregulation of Takemura Y, Osuga Y, Yoshino O, Hasegawa A, Hirata T, Hirota Y, Nose E, cyclin D1, and requires p27Kip1 or p21Cip1. Journal of Molecular Morimoto C, Harada M, Koga K et al. 2007 Metformin suppresses Signaling 3 18. interleukin (IL)-1beta-induced IL-8 production, aromatase activation, and proliferation of endometriotic stromal cells. Journal of Clinical Endocrinology and Metabolism 92 3213–3218. Received 14 August 2009 Tokudome T, Horio T, Yoshihara F, Suga S, Kawano Y, Kohno M & First decision 22 September 2009 Kangawa K 2004 Direct effects of high glucose and insulin on protein synthesis in cultured cardiac myocytes and DNA and collagen synthesis Revised manuscript received 8 November 2009 in cardiac fibroblasts. Metabolism 53 710–715. Accepted 11 November 2009

Reproduction (2010) 139 409–418 www.reproduction-online.org

Downloaded from Bioscientifica.com at 10/01/2021 05:02:53PM via free access