Introduction

Several studies have reported either the direct or indirect involvement of insulin-like growth factor-I (IGF-I) and IGF-II in the development of the normal breast epithelium and in the progression of breast carcinoma (Pollak, 2000; Sachdev and Yee, 2001). It is generally accepted that the mitogenic actions of IGFs are mediated through binding to and activation of the type I IGF receptor (IGF-IR) (Cullen et al., 1990). The IGF-IR is a transmembrane tyrosine kinase with the ability to regulate mitogenesis, transformation, and survival (Blakesley et al., 1996; O'Connor et al., 2000). The activation of IGF-IR leads to the subsequent phosphorylation and activation of various adaptor proteins involved in IGF signal transduction, such as insulin receptor substrate-1 (IRS-1), IRS-2, Shc, and Crk (Sun et al., 1991, 1995; Beitner-Johnson and LeRoith, 1995; Dey et al., 1996).

Phosphorylated IRS-1 serves as a docking molecule for a number of Src homology (SH) 2 domain containing proteins such as Grb2, Syp, and the regulatory subunit of PI 3-kinase (p85) (Lee and Pilch, 1994; White and Kahn, 1994). The interaction of IRS-1 with p85 activates PI 3-kinase activity and subsequent activation of the serine/threonine protein kinase AKT, which mediates antiapoptotic pathways (Strange et al., 2001). The association of IRS-1 with the Grb2–Sos complex results in the activation of Ras and subsequent activation of mitogen-activated protein (MAP) kinases, which are critical regulators of breast cancer cell growth. IRS-1 has been shown to be the main adaptor molecule phosphorylated by IGF-I in estrogen receptor-positive breast cancer cell lines, and this activation was associated with increased activation of the PI 3-kinase and MAP kinase pathways (Jackson et al., 1998). Although increasing evidence in breast cancer cell models suggests a major role of IRS-2 and SHC in regulating cell motility, the predominant role of IRS-1 appears to be in transmitting proliferative signals (Nolan et al., 1997; Mauro et al., 1999; Jackson et al., 2001). Recent studies have correlated high levels of IRS-1 in human breast tumors with increased recurrence of the disease (Rocha et al., 1997; Lee et al., 1999). The knowledge that IRS-1 can be overexpressed in breast cancer makes this component of the IGF signal transduction pathway a potential target for agents used in the treatment of breast cancer.

All-trans retinoic acid (RA) is a vitamin A derivative that regulates cell growth, differentiation, and apoptosis (DeLuca, 1991; Sporn et al., 1994). Various synthetic retinoids, including RA, have shown promise for the treatment and prevention of several cancers, including carcinoma of the breast (Veronesi et al., 1999; Dragnev et al., 2000; Guruswamy et al., 2001). Numerous groups, including ours, have found that retinoids potently inhibit the growth of breast cancer cell lines (Zhu et al., 1997; Rosenauer et al., 1998; Wang et al., 2000), and others have reported the capacity of retinoids to inhibit mammary carcinogenesis in animal models (Anzano et al., 1994; Moon and Constantinou, 1997). Many mechanisms have been proposed to explain the inhibition of breast cancer cell growth by retinoids including: cyclin D degradation, RAR induction, and inhibition of AP-1 activity (Liu et al., 1996; Zhou et al., 1997; Agadir et al., 1999). Another possible mechanism of RA-mediated regulation of breast cancer cell growth may be through interference with the IGF signal transduction pathway. A number of the known effects of retinoids on the IGF system in breast cancer cells include: abrogation of IGF-I-stimulated growth (Fontana et al., 1991), increased production of the growth inhibitory IGF binding protein-3 in breast cancer cell models (Shang et al., 1999), and downregulation of plasma IGF-I levels in patients with breast cancer (Torrisi et al., 2001). However, the effects of retinoids on key signaling molecules in the IGF-I pathway have not been well characterized (Li et al., 1994; Rubin et al., 1994). Thus, we examined the effect of RA on the IGF-IR and its main intracellular substrates, IRS-1, IRS-2, and SHC. Although we did not observe regulation of IGF-IR itself, we found that RA-mediated growth inhibition is associated with a selective reduction in IRS-1 protein and activity levels. We propose that the abrogation of IRS-1 protein signaling acts as a novel mechanism of RA-mediated growth inhibition of MCF-7 cells. We present evidence that decreasing IRS-1 levels may result in the selective downregulation of the PI 3-kinase/AKT pathway in MCF-7 cells treated with RA. The relevance of IRS-1 regulation to the growth inhibitory action of RA is supported by our finding that forced expression of IRS-1 abrogates RA's ability to significantly inhibit MCF-7 cell growth.

Results

RA regulates a specific IGF-IR downstream substrate in MCF-7 cells

Although previous studies of regulation of IGF-IR mRNA by RA have yielded conflicting results (Li et al., 1994; Rubin et al., 1994), in the ATCC clones of MCF-7 currently growing in our laboratory, we observe no change in the level of IGF-IR protein expression (Figure 1a) or mRNA expression (data not shown) in cells treated with 1 M RA for 24, 48, and 72 h. We also observed no change in the level of IGF-I-stimulated tyrosine phosphorylation of the IGF-IR in cells pretreated with 1 M RA for 24, 48, or 72 h (Figure 1b).

Figure 1.

RA selectively decreases IRS-1 levels in MCF-7 cells. (a) MCF-7 cells were treated with 1 M RA for 24, 48, and 72 h in SFM. Western blot was used to determine the expression level of IGF-IR. Lamin B was used as a loading control. (b) MCF-7 cells were pretreated with 1 M tRA for 24, 48, and 72 h in SFM. Prior to protein extraction, cells were stimulated with 5 nM IGF-I for 10 min. To detect phosphorylated IGF-IR, cell lysate was immunoprecipitated with an anti-IGF-IR antibody and immunoblotted using a phosphotyrosine antibody (PY100). Shown in the upper panel is the phosphorylation status of the -subunit of the IGF-IR. The lower panel shows the immunoblot in (b) stripped and immunoblotted with an anti-IGF-IR antibody. (c) MCF-7 cells were treated as in (b). Western blot was used to determine the expression levels of IRS-1, IRS-2, and SHC. Lamin B was used to show equal loading (fourth panel). These data represent three independent experiments

Full figure and legend (150K)

Since we did not observe any change in IGF-IR levels or phosphorylation, we proceeded to look at key IGF-IR downstream substrates: IRS-1, IRS-2, and SHC. The expression of IRS-1 mRNA was found to be unchanged in MCF-7 cells treated with 1 M RA (data not shown). However, we found the expression of IRS-1 protein decreased after exposure of MCF-7 cells to 1 M RA for 48 and 72 h (Figure 1c, panel 1). The protein expressions of IRS-2 (Figure 1c, panel 2) and SHC (isoforms p46, p52) (Figure 1c, panel 3) were unaffected by treatment of MCF-7 cells with 1 M RA, suggesting that IRS-1 is a novel and specific target for RA in the IGF-I signaling pathway.

RA modulates IGF-I-stimulated IRS-1 tyrosine phosphorylation in MCF-7 cells

IRS-1 is a key regulator of cell proliferation, and so we hypothesize that the lower IRS-1 protein levels in RA-treated MCF-7 cells may contribute to an inhibition of proliferation. Since IGF-I uses IRS-1 to transmit its mitogenic signal via a signal transduction pathway that begins with IRS-1 tyrosine phosphorylation, we examined the effect of RA on the level of IRS-1 tyrosine phosphorylation. MCF-7 cells were pretreated with 1 M RA for the times indicated, treated with IGF-I for 10 min, and then tyrosine-phosphorylated IRS-1 was assessed. MCF-7 cells have barely detectable phosphorylated IRS-1, which increases dramatically by treatment with IGF-I (Figure 2a, lane 2). This stimulation was decreased in MCF-7 cells pretreated with RA for 24 h and more significantly in cells pretreated with RA for 48 and 72 h (lanes 4 and 5, respectively).

Figure 2.

RA modulates IGF-I-stimulated IRS-1 tyrosine phosphorylation in MCF-7 cells. MCF-7 cells were treated as in Figure 1(b). (a) Phosphorylated IRS-1 was detected by immunoprecipitating cell lysates with an anti-IRS-1 pAb and immunoblotting using a phosphotyrosine antibody (PY100). (b) Phosphorylated SHC was detected by immunoprecipitating cell lysate with an anti-SHC pAb and immunoblotting using PY100. (c) To detect SHC/Grb2 association, cell lysate was immunoprecipitated with an anti-Grb2 mAb and immunoblotted with an anti-SHC pAb. All data are representative of three independent experiments

Full figure and legend (103K)

In addition to inducing phosphorylation of IRS-1, the activated IGF-IR can recruit and phosphorylate another downstream substrate, SHC. Although RA selectively decreases the levels and phosphorylation status of IRS-1 (Figure 2a), we found that RA does not have any effect on IGF-I-induced tyrosine phosphorylation of SHC (Figure 2b) or on the ability of SHC to associate with GRB2, as detected by immunoprecipitation studies (Figure 2c). We conclude that RA selectively regulates IRS-1 and that RA-mediated growth inhibition does not require the regulation of the IGF-I-stimulated SHC/Grb2 pathway.

RA inhibits IGF-I-stimulated downstream signaling in MCF-7 cells

The phosphorylated IRS-1 serves as a docking protein to recruit other molecules, including the p85 subunit of PI 3-kinase, tyrosine phosphatase Syp and the adaptor protein Grb2 (White, 1997). Grb2 interacts with Sos, the guanylnucleotide exchange factor for Ras, and the IRS-1/Grb2/Sos complex subsequently activates the MAPK pathway that has been shown to regulate breast cancer cell growth (Skolnik et al., 1993). In MCF-7 cells, the PI 3-kinase pathway has been reported to be the specific pathway transmitting IGF-I mitogenic signals (Dufourny et al., 1997). We examined the possibility that the decreased IRS-1 expression and phosphorylation after RA treatment might alter IRS-1/Grb-2 and IRS-1/p85 binding and thus regulate MAPK and PI 3-kinase activities, respectively. IGF-I stimulation increased the amounts of Grb2 coimmunoprecipitated with IRS-1 from control cells (Figure 3a, lane 2), but the levels of coimmunoprecipitated Grb2 were reduced in cells pretreated with 1 M RA for 48 and 72 h (lanes 4 and 5, respectively), and this was not because of an effect of RA on total levels of Grb2 (Figure 3b). Similarly, the levels of IRS-1 coimmunoprecipitated with p85 from MCF-7 cells pretreated with 1 M RA for 48 and 72 h (Figure 3c, lanes 4 and 5, respectively) were reduced compared with controls. This decrease in p85 associated with IRS-1 was not due to a reduction in the level of total p85 (Figure 3d). The finding that RA inhibits the association of IRS-1 with Grb2 and p85 confirms the observed downregulation of IRS-1 protein levels (Figure 1c, panel 1) and tyrosine phosphorylation (Figure 2a).

Figure 3.

RA suppresses IGF-I-stimulated IRS-1 /GRB2 and IRS-1/p85 complex formation. MCF-7 cells were treated as in Figure 1(b). (a) To detect IRS-1/Grb2 association, cell lysate was immunoprecipitated with an anti-GRB2 mAb and immunoblotted with an anti-IRS-1 pAb (upper panel). To confirm IRS-1/Grb2 binding, cell lysate was immunoprecipitated with an anti-IRS-1 pAb and immunoblotted with an anti-GRB2 mAb (lower panel). (b) Western blot showing equal levels of GRB2. (c) To detect IRS-1/p85 binding, cell lysate was immunoprecipitated with an anti-p85 pAb and immunoblotted with an anti-IRS-1 pAb. (d) Western blot showing equal levels of p85. Data depicted are representative of at least three independent experiments

Full figure and legend (113K)

RA impairs IGF-I-stimulated AKT activity but not ERK1/2 activation in MCF-7 cells

We sought to verify that the RA-mediated decrease in associations of IRS-1 with GRB2 and p85 correspond to reductions in the activation of downstream targets of these complexes. We looked at the effect of RA on phosphorylation of the p44 and p42 (erk1 and erk2) members of the MAPK pathway and phosphorylation of AKT, a serine/threonine kinase activated downstream of PI 3-kinase. Phosphorylated IRS-1 can activate erk1 and erk2 by binding Grb2 and activating the ras/MAPK pathway (Myers et al., 1994). In spite of decreased levels of phosphorylated IRS-1, we found that IGF-I-stimulated erk1 and erk2 phosphorylation (Figure 4a, lane 2) were not changed by pretreatment of MCF-7 cells with 1 M RA. In contrast, IGF-I-stimulated AKT activation (Figure 4b, lane 2), as assessed by phosphorylation of the AKT kinase in Ser473, was markedly decreased in MCF-7 cells that were pretreated with 1 M RA for 24 and 48 h (Figure 4b, lanes 3 and 4, respectively). In MCF-7 cells that had been pretreated for 72 h with RA, a slight recovery of AKT kinase activation was observed (Figure 4b, lane 5). The effects of RA on the serine phosphorylation status of AKT kinase were not due to a change in the level of total AKT (Figure 4c). To strengthen our hypothesis that RA-mediated growth inhibition involves the selective downregulation of the PI 3-kinase/AKT pathway, we modified the experimental conditions used in Figures 4a and b, using culture conditions more related to RA-mediated growth inhibition. To this end, in Figures 4d and e, we did not stimulate cells with IGF-I, we cultured MCF-7 cells in medium containing 5% FBS, which is known to contain IGF-I, to determine whether RA could still regulate AKT activity under these culture conditions. Indeed, RA decreases AKT kinase activation (Figure 4d) but not erk1/2 activity (Figure 4e) in medium containing 5% FBS, that is, using conditions similar to those employed in the proliferation assays showing RA-mediated growth arrest (see Figure 5a).

Figure 4.

RA regulates the phosphorylation of AKT but not of ERK1/2. MCF-7 cells were treated as in Figure 1(b). Western blot was used to determine the levels of: (a) phosphorylated p44/42 MAP kinase (on Thr-202/Tyr-204). Lamin B was used to show equal loading of lanes, (b) phosphorylated AKT (on Ser-473), and (c) AKT to show equal levels of total AKT. In (d) and (e), MCF-7 cells were cultured in SFM and stimulated with 5 nM IGF-I for 10 min (left panels) or cultured in the presence of 5% FBS (right panels). (d) Phosphorylated AKT (on Ser-473) (upper panel) and total AKT (lower panel). (e) Phosphorylated p44/42 MAP kinase (on Thr202/Tyr204) (upper panel) and lamin B (lower panel). These data represent three independent experiments

Full figure and legend (137K)

Figure 5.

RA inhibits the growth of MCF-7 cells over-expressing specific components of the IGF-I signal transduction pathway. (a) Parental (vector alone), IGF-IR-transfected (clone 17/IGF-IR), and IRS-1-transfected (clone 18/IRS-1) MCF-7 cells were plated at 1 10⁴ cells/well in 24-well plates in DMEM/F12 medium+5% FBS. All cell lines were incubated with 1 M RA for 5 days. Cells were counted using a hemacytometer on days 3 and 5. Each data point represents an average of triplicate wellsstandard deviation. (^* There was a significant difference between clone 18 and the MCF-7 parental P<0.005, while there was no significant difference between clone 17 and MCF-7). Insets: Western blots confirming an overexpression of IGF-IR in clone 17, and an overexpression of IRS-1 in clone 18. For (b)–(d): MCF-7 cells overexpressing IRS-1 (clone 18) were pretreated with 1 M RA for 24, 48, and 72 h in SFM. Prior to protein extraction, cells were stimulated with 5 nM IGF-I for 10 min. (b) Western blot was used to determine the levels of IRS-1 in clone 18. (c) Phosphorylated IRS-1 in clone 18 was detected by immunoprecipitating cell lysates with an anti-IRS-1 pAb and immunoblotting with PY100. (d) Western blot was used to determine the levels of phosphorylated AKT (on Ser-473) (upper panel) and total AKT (lower panel). Data depicted are representative of at least three independent experiments

Full figure and legend (132K)

Overexpression of specific components of the IGF-I signaling pathway decreases the growth inhibitory action of RA

Although both IGF-IR and IRS-1 have been implicated in the control of breast cancer cell growth (Ando et al., 1998), our data suggest that IRS-1 plays a more important role in the growth inhibitory response to RA. To confirm the importance of IRS-1 regulation by RA, we examined the response to RA in two previously characterized MCF-7 cell lines overexpressing IGF-IR (clone 17) (Guvakova and Surmacz, 1997) and IRS-1 (clone 18) (Surmacz and Burgaud, 1995). On days three and five of the growth curve, MCF-7 cells over-expressing IGF-IR were significantly growth inhibited by 1 M RA (Figure 5a). However, MCF-7 cells overexpressing IRS-1 (clone 18) were significantly less responsive to the growth inhibitory effects of 1 M RA in a representative 5-day growth curve than the parental MCF-7 cells. This supports our data that the effect of retinoids on the IGF signaling pathway is not mediated at the level of IGF-IR, but at the level of IRS-1. We next examined whether downregulation of IRS-1 signaling by RA is blocked by the overexpression of IRS-1 in clone 18. We found that RA does not decrease IRS-1 protein expression (Figure 5b), does not inhibit IGF-I-stimulated IRS-1 tyrosine phosphorylation (Figure 5c), and is unable to inhibit serine phosphorylation of AKT kinase in these cells (Figure 5d). This supports our hypothesis that the selective downregulation of IRS-1 signaling by RA mediates its growth inhibitory effect.

RA-mediated growth arrest involves IRS-1 regulation

We have previously reported, and include here (Figure 6a), that RA can inhibit the growth of the ER-negative breast cancer cell line, MDA-MB-231 stably transfected to express wild-type ER (S30), but not of MDA-MB-231 (Rosenauer et al., 1998). To provide additional evidence that RA-mediated growth arrest involves the regulation of IRS-1, we assessed the regulation of IRS-1 in the S30 retinoid sensitive cell versus its regulation in the RA-resistant MDA-MB-231 cell line. The Western blot shown in Figure 6b shows that RA decreases IRS-1 levels in the S30 cell line but not in the parental MDA-MB-231. The lack of IRS-1 regulation in MDA-MB-231 is consistent with the data obtained in the RA-resistant IRS-1 overexpressing MCF-7 cell line (clone 18) (see Figure 5a), and supports our hypothesis that RA-mediated growth arrest involves decreasing IRS-1 levels.

Figure 6.

RA-mediated growth inhibition involves IRS-1 regulation. (a) MCF-7 cells, MDA-MB-231, and S30 were plated at 1 10⁴ cells/well in 24-well plates in the presence of 5% FBS (MCF-7) or 5% charcoal stripped serum (MDA-MB-231 and S30). MCF-7 cells were incubated with 1 M RA for 6 days. MDA-MB-231 and S30 were incubated with 10 M RA for 6 days. Cells were counted in triplicate using a hemacytometer on days two, four, and six. There was a significant difference between S30 and MDA-MB-231 (P<0.005). (b) Western blot was used to determine the levels of IRS-1 in S30 and MDA-MB-231 cells treated with 10 M RA for 48 and 72 h. Lamin B was used to show equal loading of all lanes

Full figure and legend (84K)

Discussion

To our knowledge, this is the first study elucidating the effect of RA on signaling elements downstream of the IGF-IR, and from the results presented herein, we propose a novel mechanism by which RA inhibits the growth of MCF-7 cells. Although RA-mediated growth inhibition of MCF-7 cells is not directly mediated via alterations in IGF-IR, we show a significant reduction in IRS-1 protein levels when MCF-7 cells are treated with RA. This reduction in IRS-1 protein corresponds to a decrease in IGF-I-stimulated IRS-1 tyrosine phosphorylation at a concentration and time that is consistent with RA-mediated growth inhibition of these cells. The findings presented in this report are directly relevant to RA-mediated regulation of breast cancer cell growth, since previous studies downregulating IRS-1 expression using antisense strategies in several cellular systems implicate IRS-1 as a key mediator of cellular growth (Waters et al., 1993; Nolan et al., 1997; Taouis et al., 1998).

Our hypothesis that reducing IRS-1 protein levels and tyrosine phosphorylation may be a key step in RA-mediated growth arrest of MCF-7 cells is supported by our observations in MCF-7 cells that stably overexpress IRS-1 (clone 18). In these cells, RA does not reduce IRS-1 protein levels or IRS-1 tyrosine phosphorylation, nor does it lead to a significant inhibition of cell proliferation. The study of other anticancer agents has revealed similar findings. Specifically, the antiestrogen ICI 182 780 can decrease IRS-1 levels in MCF-7 cells; however, this agent is not effective in reducing IRS-1 levels or inhibiting the growth of MCF-7 cells overexpressing IRS-1 (Salerno et al., 1999). Although additional compounds such as 12-O-tetradecanoylphorbol-13-acetate (de Vente et al., 1996) and endothelin-1 (Ishibashi et al., 2001) have been reported to decrease IRS-1 levels, not all growth-regulatory agents are capable of doing so. For example, like RA, vitamin D₃ derivatives have potent differentiating and cell growth inhibitory activity and can effectively inhibit IGF-I-stimulated cell growth (Xie et al., 1999) and reduce IGF-IR levels (Xie et al., 1997), but they exert no changes on IRS-1 expression (Rozen and Pollak, 1999). To further support the observation that RA-mediated growth arrest of breast cancer cells involves IRS-1 downregulation, we show in the RA-resistant cell line MDA-MB-231, that IRS-1 levels remain unchanged, while in the RA-sensitive MDA-MB-231 subclone, S30, RA does decrease IRS-1 protein levels. These results suggest that resistance of breast cancer cells to the growth inhibitory effects of RA is accompanied by an inability of RA to decrease IRS-1 levels.

To extend our hypothesis that IRS-1 regulation may mediate RA-induced growth inhibition, we looked at the ability of IRS-1 to transmit the IGF-I-stimulated mitogenic signal downstream. IRS-1 possesses 20–22 potential tyrosine phosphorylation sites and interacts with many SH2 domain-containing proteins, including Grb2 and p85. Thus, tyrosine phosphorylation of IRS-1 represents a key step in activating distinct downstream IGF-I signaling pathways, including the MAP kinase and PI 3-kinase pathways. Although we observe less association of Grb2 with IRS-1 in RA-treated MCF-7 cells, MAP kinase activity was not abolished, as assessed by ERK1/2 phosphorylation. Both IRS-1/Grb2 and SHC/Grb2 lie upstream of ERK1/2, thus it is feasible that the SHC branch of the IGF-I signaling pathway can directly activate Erk1/2. Consistent with this possibility, decreasing IRS-1 expression using antisense IRS-1 has been reported to increase SHC levels and SHC tyrosine phosphorylation (Taouis et al., 1998). It has also been shown that blocking IRS-1 function, using IRS-1 deletion mutants, has no compensatory effect on the SHC/MAP kinase pathway (Goalstone et al., 2001). In this study, RA did not affect SHC protein levels, SHC tyrosine phosphorylation status, or the complex formation between SHC and Grb2. Thus, we show that the SHC/Grb2/MAP kinase pathway is unaffected by the RA-induced decrease in IRS-1 levels.

In MCF-7 cells, the mitogenic action of IGF-I is not via the MAP kinase pathway, but through PI 3-kinase activation (Dufourny et al., 1997). Furthermore, the serine/threonine kinase AKT is one of the major downstream effectors of PI 3-kinase reported to mediate the effects of IGF-I (Franke et al., 1995; Alessi et al., 1996; Kulik et al., 1997). AKT acts to regulate a number of cellular processes involved in transmitting key survival signals (Nicholson and Anderson, 2002) and can be found overexpressed in a variety of cancer cell lines, including MCF-7 cells (Jones et al., 1991; Downward, 1995). The observed abrogation of IRS-1/p85 association in cells that had been pretreated with RA for 48 h correlated well with a decrease in the phosphorylation status of AKT in MCF-7 cells. It remains unclear why we observe a partial recovery of AKT activation in MCF-7 cells that had been pretreated with RA for 72 h (Figure 4b, lane 5). We are currently using biological inhibitors of the PI 3-kinase and MAP kinase pathways in an attempt to elucidate if there is a feedback mechanism involved in activating AKT. It is also of interest that in the presence of normal cell culture conditions, that is in 5% FBS, RA is still able to decrease AKT activity, albeit at a later time point than in the absence of serum growth factors (Figure 4d). This is the first study reporting an effect of RA in downregulating AKT kinase activation in MCF-7 cells, and from recent studies, it appears that RA can now be added to a growing list of chemotherapeutic agents, such as Herceptin, Genistein, and CI-1033, that inhibit cancer cell growth via a mechanism involving the regulation of PI 3-kinase and AKT activity (Allen et al., 2002; Li and Sarkar, 2002; Yakes et al., 2002).

Our finding that RA-mediated growth inhibition involves the regulation of the PI 3-kinase/AKT pathway is in accordance with a recent paper showing that RA-mediated degradation of RAR gamma requires the downregulation of the PI 3-kinase/AKT pathway (Gianni et al., 2002). Furthermore, the observed abrogation of AKT activation may be a requirement for RA-mediated growth inhibition, as we do not observe AKT regulation in RA-resistant MCF-7 cells overexpressing IRS-1 (clone 18). Consistent with the data obtained in clone 18, another group has recently shown that the transfection of a dominant-negative AKT construct in RA-resistant, Her2/neu-overexpressing cells can overcome RA resistance (Tari et al., 2002). Additional work will be needed to elucidate the molecules involved in RA-mediated downregulation of the IRS-1/PI 3-kinase/AKT pathway. To this end, we are currently elucidating the mechanism of RA-mediated regulation of IRS-1. We speculate that RA regulates IRS-1 via a post-translational modification, due to our findings that RA has no effect on IRS-1 mRNA levels (data not shown), the RA-mediated decrease in IRS-1 protein expression was not altered by either actinomycin D (an inhibitor of transcription) or cycloheximide (an inhibitor of translation) (data not shown), and RA affects IRS-1 protein levels at later time points (Figure 1c). Based on previous reports showing that IRS-1 undergoes ubiquitin-dependent degradation (Lee et al., 1999; Sun et al., 1999; Zhang et al., 2000; Zhande et al., 2002) and unpublished preliminary data from our laboratory, we hypothesize that IRS-1 is degraded by the ubiquitin-proteasome pathway during RA treatment. There is indeed a growing list of proteins known to be regulated by RA via mechanisms involving the ubiquitin–proteasome pathway; cyclin D1 (Spinella et al., 1999), PML/RAR (Yoshida et al., 1996), RAR alpha, and RAR gamma (Tanaka et al., 2001), CDK-4 (Sueoka et al., 1999), Skp2 (Dow et al., 2001), and future experiments will elucidate whether IRS-1 can be added to this list.

In conclusion, we find that RA-mediated growth inhibition of MCF-7 cells involves alterations in specific downstream IGF signaling elements. We have shown that RA induces a significant reduction in the expression and tyrosine phosphorylation of IRS-1 protein in MCF-7 cells treated with RA. These data emphasize growth factor receptor adaptor molecules, such as insulin receptor substrate-1, rather than the cell surface IGF-IR as targets for antitumor therapeutic strategies. In fact, a recent study demonstrates that constitutive activation of IRS-1 is not restricted to breast tumors, thus presenting IRS-1 as a potential drug target in other human malignancies (Chang et al., 2002). In addition, our data suggest that a possible functional outcome of reduced IRS-1 protein levels is the selective downregulation of the PI 3-kinase/AKT pathway, which may be important for RA-mediated growth inhibition. These novel targets of RA could lead to the development of novel combination therapies in the treatment of breast cancer, making use of known biological inhibitors of the PI 3-kinase/AKT (Li and Zhu, 2002; Mitsiades et al., 2002) pathway in combination with retinoids.

Materials and methods

Reagents

All-trans (RA) was purchased from Sigma. Recombinant human IGF-I was purchased from PeproTech (Princeton, NJ, USA). Protein G-agarose, and Nonidet^® P40 were purchased from Sigma (Oakville, Canada). Enhanced chemiluminescence (ECL) detection system was purchased from Amersham Pharmacia Biotech. The following antibodies (Abs) were used for immunoprecipitations: for the IGF-IR: anti-IGF-IR mAb alpha-IR3 (Oncogene Science); for IRS-1: anti-C-terminal IRS-1 pAb (Upstate Biotechnology); for SHC: anti-SHC pAb (Transduction Laboratories); for GRB2: anti-GRB2 mAb (Transduction Laboratories); for the p85 subunit of PI3K: anti-PI3K-p85 pAb (Upstate Biotechnology). Tyrosine phosphorylation was detected with an antiphosphotyrosine mAb PY100 (Cell Signaling Technology). The following antibodies were used for Western blotting: for IRS-1 and IRS-2: anti-IRS-1 pAb and anti-IRS-2 pAb (Upstate Biotechnology); for IGF-IR: anti-IGF-IR pAb (Santa Cruz Biotechnology); for SHC and GRB2: anti-SHC mAb and anti-GRB2 mAb (Transduction Laboratories); for total AKT levels and active AKT: anti-AKT pAb and anti-phospho-Akt (on Ser-473) pAb (Cell Signaling Technology); and for active ERK1/2: anti-phospho-p44/42 MAP kinase (Thr202/Tyr204) pAb (Cell Signaling Technology).

Cell stimulation and cell lysate preparation

MCF-7 cells (ATCC) were plated in phenol red containing -MEM (Life Technologies, Inc.) supplemented with 5% fetal bovine serum (FBS). At 70% confluence cells were washed twice with phosphate-buffered saline (PBS) and changed to phenol red-free -MEM supplemented with BSA and holo-transferrin (serum free media – SFM) in the presence or absence of 1 M RA or DMSO vehicle control for the times indicated. MCF-7 cells were washed twice with cold PBS and lysed with RIPA buffer (50 mM Tris-HCl pH 8, 150 mM NaCl, 1% Nonidet^® P40, 0.05% sodium deoxycholate, and 0.1% SDS). The extracts were centrifuged at 13 000 g at 4°C for 30 min to remove insoluble material. After centrifugation, the protein content was measured by the Bradford assay using Bio-Rad reagents and BSA as standard. For IRS-1, IRS-2, and SHC Western blotting, cells were immediately lysed as described above. For experiments requiring IGF-I stimulation, prior to harvesting cells, they were incubated at 37°C for 10 min with 5 nM IGF-I.

Immunoprecipitation and Western blotting

For Western blotting, cell lysates were boiled for 3 min in 2 SDS sample buffer (250 mM Tris-HCl pH 6.8, 8% SDS, 8 mM EDTA, 35% glycerol, 2.5% -mercapto-ethanol, bromophenol Blue) and resolved in 8–10% SDS–polyacrylamide gels (SDS–PAGE). For immunoprecipitation, after IGF-I stimulation, 500 g–1 mg of precleared cell lysates were incubated with the indicated antibody overnight at 4°C, followed by the addition of protein G agarose overnight to collect immune complexes. The immunoprecipitates were washed in RIPA buffer, resuspended in SDS sample buffer, and boiled for 5 min. The solubilized proteins were resolved by SDS–PAGE. Proteins on the gel were transferred to nitrocellulose membrane (Bio-Rad) and detected by immunoblotting with the indicated antibody using ECL. Some membranes were stripped to prepare them for a second round of immunoblotting.

Cell proliferation assay

Exponentially growing cells were plated at subconfluent densities in 24-well plates. MCF-7 cells overexpressing IGF-IR (clones 12 and 17) or IRS-1 (clone 18) were incubated in phenol red containing DMEM-F12 (Life Technologies, Inc.) supplemented with 5% FBS for 5 days in the presence of 1 M (RA) or DMSO (vehicle control). MDA-MB-231 (ATCC) and the ER-positive subclone of MDA-MB-231, S30 (courtesy of Dr VC Jordan) were maintained in phenol red free -MEM (Life Technologies, Inc.) supplemented with 5% charcoal stripped serum. Cells in triplicate wells were counted by a hemacytometer. The Student's t-test was used to analyse the significance of the results obtained in cell growth curves.

References

1. Agadir A, Chen G, Bost F, Li Y, Mercola D and Zhang X. (1999). J. Biol. Chem., 274, 29779–29785. | Article | PubMed | ISI | ChemPort |
2. Alessi DR, Andjelkovic M, Caudwell B, Cron P, Morrice N, Cohen P and Hemmings B. (1996). EMBO J., 15, 6541–6551. | PubMed | ISI | ChemPort |
3. Allen LF, Lenehan PF, Eiseman IA, Elliott WL and Fry DW. (2002). Semin. Oncol., 29, 11–21. | Article | PubMed | ISI | ChemPort |
4. Ando S, Panno ML, Salerno M, Sisci D, Mauro L, Lanzino M and Surmacz E. (1998). Biochem. Biophys. Res. Commun., 253, 315–319. | Article | PubMed |
5. Anzano MA, Byers SW, Smith JM, Peer CW, Mullen LT, Brown CC, Roberts AB and Sporn MB. (1994). Cancer Res., 54, 4614–4617. | PubMed | ISI | ChemPort |
6. Beitner-Johnson D and LeRoith D. (1995). J. Biol. Chem., 270, 5187–5190. | Article | PubMed | ChemPort |
7. Blakesley VA, Kalebic T, Helman LJ, Stannard B, Faria TN, Roberts Jr CT and LeRoith D. (1996). Endocrinology, 137, 410–417. | PubMed |
8. Chang Q, Li Y, White MF, Fletcher JA and Xiao S. (2002). Cancer Res., 62, 6035–6038. | PubMed | ISI | ChemPort |
9. Cullen KJ, Yee D, Sly WS, Perdue J, Hampton B, Lippman ME and Rosen N. (1990). Cancer Res., 50, 48–53. | PubMed |
10. DeLuca ML. (1991). FASEB J., 5, 2924–2933. | PubMed | ISI | ChemPort |
11. de Vente JE, Carey JO, Byrant WO, Pettit GJ and Ways DK. (1996). J. Biol. Chem., 271, 32276–32280.
12. Dey BR, Frick K, Lopaczynski W, Nissley SP and Furlanetto RW. (1996). Mol. Endocrinol., 10, 631–641. | PubMed |
13. Dow R, Hendley J, Pirkmaier A, Musgrove EA and Germain D. (2001). J. Biol. Chem., 276, 45945–45951. | Article | PubMed | ISI | ChemPort |
14. Downward J. (1995). Nature, 376, 553–554. | Article | PubMed | ISI | ChemPort |
15. Dragnev KH, Rigas JR and Dmitrovsky E. (2000). Oncologist, 5, 361–368. | Article | PubMed | ChemPort |
16. Dufourny B, Alblas J, van Teeffelen HA, van Schaik FM, van der Burg B, Steenbergh PH and Sussenbach JS. (1997). J. Biol. Chem., 272, 31163–31171. | Article | PubMed | ISI | ChemPort |
17. Fontana JA, Burrows-Mezu A, Clemmons DR and LeRoith D. (1991). Endocrinology, 128, 1115–1152.
18. Franke T, Yang S, Chan TO, Datta K, Kazlauskas A, Morrison DK, Kaplan DR and Tsichilis PN. (1995). Cell, 81, 727–736. | Article | PubMed | ISI | ChemPort |
19. Gianni M, Kopf E, Bastien J, Oulad-Abdelghani M, Garattini E, Chambon P and Rochette-Egly C. (2002). J. Biol. Chem., 277, 24859–24862. | Article | PubMed | ISI | ChemPort |
20. Goalstone ML, Leitner JW, Berhanu P, Sharma PM, Olefsky JM and Draznin B. (2001). J. Biol. Chem., 276, 12805–12812.
21. Guruswamy S, Lightfoot S, Gold MA, Hassan R, Berlin KD, Ivey RT and Benbrook DM. (2001). J. Natl. Cancer Inst., 93, 516–525. | Article | PubMed | ChemPort |
22. Guvakova MA and Surmacz E. (1997). Exp. Cell Res., 231, 149–162. | Article | PubMed |
23. Ishibashi KI, Imamura T, Sharma PM, Huang J, Ugi S and Olefsky JM. (2001). J. Clin. Invest., 107, 1193–1202. | PubMed | ChemPort |
24. Jackson JG, White MF and Yee D. (1998). J. Biol. Chem., 273, 9994–10003. | Article | PubMed | ISI | ChemPort |
25. Jackson JG, Zhang X, Yoneda T and Yee D. (2001). Oncogene, 20, 7318–7325. | Article | PubMed | ISI | ChemPort |
26. Jones PF, Jakubowitz T, Pitossi FJ, Maurer F and Hemmings BA. (1991). Nature, 376, 599–602.
27. Kulik G, Klippel A and Weber MJ. (1997). Mol. Cell. Biol., 17, 1595–1606. | PubMed | ISI | ChemPort |
28. Lee AV, Jackson JG, Gooch JL, Hilsenbeck SG, Coronado-Heinsohn E, Osborne CK and Yee D. (1999). Mol. Endocrinol., 13, 787–796. | Article | PubMed | ISI | ChemPort |
29. Lee J and Pilch PF. (1994). Am. J. Physiol., 266, C319–C334.
30. Li Q and Zhu GD. (2002). Curr Top Med. Chem., 2, 939–971. | Article | PubMed | ChemPort |
31. Li XS, Chen JC, Sheikh MS, Shao ZM and Fontana JA. (1994). Exp. Cell Res., 211, 68–73.
32. Li Y and Sarkar FH. (2002). Clin. Cancer Res., 8, 2369–2377. | PubMed | ISI | ChemPort |
33. Liu Y, Lee MO, Wang HG, Li Y, Hashimoto Y, Klaus M, Reed JC and Zhang X. (1996). Mol. Cell. Biol., 16, 1138–1149. | PubMed | ISI | ChemPort |
34. Mauro L, Sisci D, Bartucci M, Salerno M, Kim J, Tam T, Guvakova MA, Ando S and Surmacz E. (1999). Exp. Cell Res., 252, 439–448. | Article | PubMed | ISI | ChemPort |
35. Mitsiades CS, Mitsiades N, Poulaki V, Schlossman R, Akiyama M, Chauhan D, Hideshima T, Treon SP, Munshi NC, Richardson PG and Anderson KC. (2002). Oncogene, 21, 5673–5683. | Article | PubMed | ISI | ChemPort |
36. Moon RC and Constantinou AI. (1997). Breast Cancer Res. Treatment, 46, 181–189.
37. Myers Jr MG, Wang LM, Sun XJ, Zhang Y, Yenush L, Schlessinger J, Pierce JH and White MF. (1994). Mol. Cell. Biol., 14, 3577–3587. | PubMed | ISI | ChemPort |
38. Nicholson KM and Anderson NG. (2002). Cell. Signal., 14, 381–395. | Article | PubMed | ISI | ChemPort |
39. Nolan MK, Jankowska L, Prisco M, Xu S, Guvakova MA and Surmacz E. (1997). Int. J. Cancer, 72, 828–834. | Article | PubMed | ISI | ChemPort |
40. O'Connor R, Fennelly C and Krause D. (2000). Biochem. Soc. Trans., 28, 47–51. | PubMed | ISI | ChemPort |
41. Pollak, M. (2000). Eur. J. Cancer, 36, 1224–1228. | Article | PubMed | ISI | ChemPort |
42. Rocha RL, Hilsenbeck SG, Jackson JG, VanDenBerg CL, Weng CN, Lee AV and Yee D. (1997). Clin. Cancer Res., 3, 103–109. | PubMed | ChemPort |
43. Rosenauer A, Nervi C, Davison K, Lamph WW, Mader S and Miller Jr WH. (1998). Cancer Res., 58, 5110–5116. | PubMed | ChemPort |
44. Rozen F and Pollak M. (1999). Int. J. Oncol., 15, 589–594.
45. Rubin M, Fenig E, Rosenauer A, Menendez-Botet C, Achkar C, Bentel JM, Yahalom J, Mendelsohn J and Miller Jr WH. (1994). Cancer Res., 54, 6549–6556. | PubMed | ISI | ChemPort |
46. Sachdev D and Yee D. (2001). Endocr. Relat. Cancer, 8, 197–209. | Article | PubMed | ISI | ChemPort |
47. Salerno M, Sisci D, Mauro L, Guvakova MA, Ando S and Surmacz E. (1999). Int. J. Cancer, 81, 299–304. | Article | PubMed |
48. Shang Y, Baumrucker CR and Green MH. (1999). J. Biol. Chem., 274, 18005–18010. | PubMed |
49. Skolnik EY, Lee CH, Batzer A, Vicentini LM, Zhou M, Daly R, Myers Jr MJ, Backer JM, Ullrich A and White MF. (1993). EMBO J., 12, 1929–1936. | PubMed | ISI | ChemPort |
50. Spinella MJ, Freemantle SJ, Sekula D, Chang JH, Christie AJ and Dmitrovsky E. (1999). J. Biol. Chem., 274, 22013–22018. | Article | PubMed | ISI | ChemPort |
51. Sporn MB, Roberts AB and Goodman DS. (1994). The Retinoids: Biology, Chemistry and Medicine. Raven Press: New York, pp. 573–595.
52. Strange R, Metcalfe T, Thackray L and Dang M. (2001). Microsc. Res. Tech., 52, 171–181. | Article | PubMed | ISI | ChemPort |
53. Sueoka N, Lee HY, Walsh GL, Hong WK and Kurie JM. (1999). Cancer Res., 59, 3838–3844. | PubMed | ChemPort |
54. Sun XJ, Goldberg JL, Qiao LY and Mitchell JJ. (1999). Diabetes, 48, 1359–1364.
55. Sun XJ, Rothenberg P, Kahn CR, Backer JM, Araki E, Wilden PA, Cahill DA, Goldstein BJ and White MF. (1991). Nature, 352, 73–77. | Article | PubMed | ISI | ChemPort |
56. Sun XJ, Wang LM, Zhang Y, Yenush L, Myers Jr MG, Glasheen E, Lane WS, Pierce JH and White MF. (1995). Nature, 377, 173–177. | Article | PubMed | ISI | ChemPort |
57. Surmacz E and Burgaud JL. (1995). Clin. Cancer Res., 1, 1429–1436. | PubMed |
58. Tanaka T, Rodriquez de la Concepcion ML and De Luca LM. (2001). Biochem. Pharmacol., 61, 1347–1355. | Article |
59. Taouis M, Dupont J, Gillet A, Derouet M and Simon J. (1998). Mol. Cell. Endocrinol., 137, 177–186.
60. Tari AM, Lim SJ, Hung MC, Esteva FJ and Lopez-Berestein G. (2002). Oncogene, 21, 5224–5232. | Article | PubMed | ISI | ChemPort |
61. Torrisi R, Decensi A, Formelli F, Camerini T and De Palo G. (2001). Drugs, 61, 909–918.
62. Veronesi U, De Palo G, Marubini E et al. (1999). J. Natl. Cancer Inst., 91, 1847–1856. | Article | PubMed | ChemPort |
63. Wang Q, Yang W, Uytingco MS, Christakos S and Wieder R. (2000). Cancer Res., 60, 2040–2048. | PubMed | ISI | ChemPort |
64. Waters SB, Yamauchi K and Pessin JE. (1993). J. Biol. Chem., 268, 22231–22234. | PubMed | ChemPort |
65. White MF. (1997). Diabetologia, 40, S2–S17. | Article | PubMed | ISI | ChemPort |
66. White MF and Kahn CR. (1994). J. Biol. Chem., 269, 1–4. | PubMed | ISI | ChemPort |
67. Xie SP, James SY and Colston KW. (1997). J. Endocrinol., 154, 495–504.
68. Xie SP, Pirianov G and Colston KW. (1999). Eur. J. Cancer, 35, 1717–1723.
69. Yakes FM, Chinratanalab W, Ritter CA, King W, Seelig S and Arteaga CL. (2002). Cancer Res., 62, 4132–4141. | PubMed | ISI | ChemPort |
70. Yoshida H, Kitamura K., Tanaka K, Omura S, Miyazaki T, Hachiya T, Ohno R and Naoe T. (1996). Cancer Res., 56, 2945–2948. | PubMed | ISI | ChemPort |
71. Zhande R, Mitchell JJ, Wu J and Sun XJ. (2002). Mol. Cell. Biol., 22, 1016–1026. | Article |
72. Zhang H, Hoff H and Sell C. (2000). J. Biol. Chem., 275, 22558–22562. | Article | PubMed |
73. Zhou Q, Stetler-Stevenson M and Steeg PS. (1997). Oncogene, 15, 107–115. | Article | PubMed | ISI | ChemPort |
74. Zhu WY, Jones CS, Kiss A, Matsukuma K, Amin S and De Luca LM. (1997). Exp. Cell Res., 234, 293–299. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank Dr Ewa Surmacz for helpful discussion and for providing the MCF-7 cells stably transfected with various IGF signaling components. This work was supported by a predoctoral fellowship award from the US Army Medical Research and Material Command Breast Cancer Research Program (award number DAMD1701-1-0320 to SV del Rincón) and a grant from the Canadian Breast Cancer Research Initiative. Wilson H Miller Jr is an Investigator of the Canadian Institutes of Health Research.