Prolactin and estrogen enhance the activity of activating protein 1 in breast cancer cells: role of extracellularly regulated kinase 1/2-mediated signals to c-fos

Abstract

Despite the important roles of both prolactin (PRL) and 17beta-estradiol (E2) in normal mammary development as well as in breast cancer, and coexpression of the estrogen receptor (ER) and PRL receptor in many mammary tumors, the interactions between PRL and E2 in breast cancer have not been well studied. The activating protein 1 (AP-1) transcription factor, a known regulator of processes essential for normal growth and development as well as carcinogenesis, is a potential site for cross-talk between these hormones in breast cancer cells. Here we demonstrate that PRL and E2 cooperatively enhance the activity of AP-1 in MCF-7-derived cells. In addition to the acute PRL-induced ERK1/2 activation, PRL and E2 also individually elicited delayed, sustained rises in levels of phosphorylated p38 and especially ERK1/2. Together, these hormones increased the dynamic phosphorylation of ERK1/2 and c-Fos, and induced c-fos promoter activity. Synergistic activation of the transcription factor, Elk-1, reflected the PRL-E2 interaction at ERK1/2 and is a likely mechanism for activation of the c-fos promoter via the serum response element. The enhanced AP-1 activity resulting from the interaction of these hormones may increase expression of many target genes that are critical for oncogenesis and may contribute to neoplastic progression.

Fig. 1

PRL and E2 Interact to Enhance AP-1 Activity at 24 h A, Effect of time on E2 activation of AP-1 and interaction with PRL. Cells were cotransfected with 4XAP-1-luc, lPRLR, and β-galactosidase. After transfection, cells were washed and treated with or without 4 nM PRL and/or 1 nM E2 for 6 or 24 h. Cell lysates were harvested and assayed for luciferase and β-galactosidase activity as described in Materials and Methods. B, Continued presence of PRL is required for AP-1 activation. Cells were cotransfected with 4XAP-1-luc, lPRLR, and β-galactosidase. After transfection, cells were treated with or without 4 nM PRL and/or 1 nM E2. Incubation in hormone(s) is indicated by the solid arrow, followed by washout (time shown); continued incubation in serum-free conditions is indicated by the dashed arrow. All cell lysates were harvested 24 h after the initial treatment. Lysates were assayed as in panel A. In panels A and B, relative activity represents the mean of the corrected luciferase activity from at least three independent experiments, represented as mean fold change relative to the vehicle-treated control transfection ± SEM. For statistical analyses, hormone treatment groups harvested at the same times (panel A), or washed at the same times (panel B), were compared using one-way ANOVA, followed by Student-Newman-Keuls multiple comparison test. Different letters denote significant differences within these groups (P < 0.05). In panel A, the asterisk denotes a significant difference between the same treatment groups harvested at 6 and 24 h using Student’s t test (P < 0.005).

Fig. 2

Proximal Signaling Pathways for PRL and E2 Induction of AP-1 Activity at 24 h A, E2-BSA, which activates only membrane ER, does not activate AP-1 alone or enhance PRL induction. Left, Cells were cotransfected with 4XAP-1-luc, lPRLR, and β-galactosidase. After transfection, cells were treated with 4 nM PRL, 1 nM E2, 1 nM E2-BSA, or 100 nM ICI 182,780 as indicated. Lysates were harvested 24 h after treatment and assayed for luciferase and β-galactosidase activity. Relative activity represents the mean of the corrected luciferase activity from at least three independent experiments, represented as mean fold change relative to the vehicle-treated control transfection ± SEM. For statistical analysis, groups were compared using one-way ANOVA, followed by Student-Newman-Keuls multiple comparison test; different letters denote significant differences (P < 0.05). Right, To confirm that E2-BSA cannot activate nuclear ER, cells were cotransfected with oxytocin receptor (OT)-ERE-luc, lPRLR, and β-galactosidase. Cells were treated with vehicle, 1 nM E2, 1 nM E2-BSA, or 10 nM E2-BSA for 24 h and assayed as above. Representative experiment with each data point in triplicate ± SD and graphed as relative luciferase units (RLU). B, ICI 182,780 slightly reduces unstimulated AP-1 activity but does not alter the ability of PRL to activate AP-1. Cells were transfected and treated with or without PRL, with or without ICI 182,780 as above, and activation of the luciferase reporter was determined as above. Results are presented and analyzed as for panel A. C, Importance of Jak2 in PRL and E2 activation of AP-1. Cells were cotransfected with 4XAP-1-luc, lPRLR, and β-galactosidase, as well as dominant negative (DN) Jak2, where indicated. After transfection, cells were washed and treated with or without 4 nM PRL and/or 1 nM E2 in serum-free media. Lysates were harvested 24 h after treatment and assayed as in panel A. Relative activity represents the mean of the corrected luciferase activity from at least three independent experiments, represented as mean fold change relative to the vehicle-treated control transfection ± SEM. For statistical analysis, hormone treatment groups transfected with the same constructs were compared using one-way ANOVA, followed by Student-Newman-Keuls multiple comparison test; different letters denote significant differences within these groups (P < 0.05). The asterisk denotes a significant difference in fold activation between the control group and the DN Jak2 group receiving the same hormonal treatment using Student’s t test (P < 0.005).

Fig. 3

ERK1/2 Is Important in PRL and E2 Activation of AP-1 at 24 h A and B, Phosphorylation of ERK1/2 after PRL and/or E2 treatment. PRL-deficient cells were transfected with the lPRLR isoform as described in Materials and Methods. Cells were treated with 4 nM PRL and/or 1 nM E2 for the times indicated. Cell lysates were harvested, and equal amounts of protein were examined by Western analysis for phosphorylation of ERK1/2, as well as levels of total ERK1/2 (representative experiments). C, Effect of chemical inhibition of ERK1/2 on PRL and E2 activation of AP-1. Cells were cotransfected with 4XAP-1-luc, lPRLR, and β-galactosidase. After transfection, cells were washed and pretreated for 1 h with 10 μM U0126 and then treated with or without 4 nM PRL and/or 1 nM E2 in the presence of inhibitor for 24 h. Relative activity represents the mean of luciferase values from at least three independent experiments, represented as mean fold change relative to the vehicle-treated control transfection ± SEM. For statistical analysis, hormone treatment groups treated with the vehicle or inhibitor were compared separately using one-way ANOVA, followed by Student-Newman-Keuls multiple comparison test; different letters denote significant differences within these groups (P < 0.05). D, Effect of chemical inhibition of ERK1/2 on PRL- and E2-induced increases in cell number. Cells were cultured in serum-free media for 24 h before treatment with vehicle, 4 nM PRL, and/or 1 nM E2 with or without 10 μM U0126 for an additional 48 h, when total viable cells were counted using a hemocytometer. Results are expressed as percent of the number of cells present before treatment ± SD (representative experiment). For statistical analysis, the effect of hormone treatment was assessed as for panel C. The asterisk indicates a significant difference from the nonhormonally treated samples at time zero. DMSO, Dimethylsulfoxide.

Fig. 4

Role of Secreted Factors in PRL and E2 Activation of AP-1 A, CM do not contain additional factors that increase ERK1/2 phosphorylation. Cells transfected with lPRLR, as described in Materials and Methods, were treated for 24 h with vehicle, PRL, E2, or both hormones together. After 24 h, cell media were harvested and used to treat untransfected cells for 15 min. Cell lysates were harvested, and phosphorylated ERK1/2 and total ERK1/2 protein was examined by Western analysis (representative experiment). B, CM do not contain factors that further stimulate AP-1 activity. Left, Cells were transfected and treated as in Fig. 1A. Right, Media from cells treated for 24 h as in Fig. 4A were used to treat cells similarly transfected. AP-1 activity was determined 6 h after treatment. Relative activity represents the mean of the corrected luciferase activity from at least three independent experiments, represented as mean fold change relative to the vehicle-treated control transfection ± SEM. For statistical analyses, control and CM treatment groups were compared using one-way ANOVA, followed by Student-Newman-Keuls multiple comparison test. Different letters denote significant differences within these groups (P < 0.05).

Fig. 5

Role for p38 in PRL and E2 Activation of AP-1 A, Phosphorylation of p38 after PRL and/or E2 treatment. Cells were plated, transfected with the lPRLR isoform as described in Materials and Methods, and treated with 4 nM PRL and/or 1 nM E2 for the times indicated. Cell lysates were harvested, and equal amounts of protein were examined by Western analysis for phosphorylation of p38, as well as levels of total p38 (representative experiment). B, Effect of chemical inhibition of p38 on PRL and E2 activation of AP-1. Cells were cotransfected with 4XAP-1-luc, lPRLR, and β-galactosidase. After transfection, cells were washed and pretreated for 1 h with vehicle or 20 μM SB202190 and then treated with or without 4 nM PRL and/or 1 nM E2 in the continued presence of inhibitor for 24 h. Lysates were harvested and analyzed. Relative activity represents the mean of luciferase values from at least three independent experiments, represented as mean fold change relative to the vehicle-treated control transfection ± SEM. For statistical analysis, hormone treatment groups treated with the same inhibitor were compared using one-way ANOVA, followed by Student-Newman-Keuls multiple comparison test; different letters denote significant differences within these groups (P < 0.05). DMSO, Dimethylsulfoxide.

Fig. 6

Role for JNK1/2 and c-Jun in PRL and E2 Activation of AP-1 at 24 h A, Phosphorylation of JNK1/2 and c-Jun after PRL and/or E2 treatment. Cells were plated and transfected with the lPRLR isoform as described in Materials and Methods. Cells were treated with 4 nM PRL and/or 1 nM E2 for the times indicated. Cell lysates were harvested, and equal amounts of protein were examined by Western analysis for activation of JNK1/2, total JNK1/2, phosphorylation of c-Jun, and total c-Jun (representative experiments). B, Effect of JIP-1 and TAM-67 on PRL- and E2-induced AP-1 activity. Cells were cotransfected with 4XAP-1-luc, lPRLR, and β-galactosidase, as well as JIP-1 or TAM-67 where indicated. After transfection, cells were washed and treated with or without 4 nM PRL and/or 1 nM E2 for 24 h. Lysates were harvested and analyzed for luciferase and β-galactosidase activity. Relative activity represents the mean of the corrected luciferase activity from at least three independent experiments, represented as mean fold change relative to the vehicle-treated control transfection ± SEM.

Fig. 7

PRL and E2 Activation of c-Fos A, Effect of PRL and E2 treatment on c-Fos. Cells were plated and transfected with the lPRLR isoform as described in Materials and Methods. Cells were treated with 4 nM PRL and/or 1 nM E2 for the times indicated. Cell lysates were harvested, and equal amounts of protein were examined by Western analysis for c-Fos. Arrowhead indicates phosphorylated c-Fos (representative experiment). B, Effect of PRL and E2 on activation of the c-Fos promoter. Cells were cotransfected with c-Fos promoter-luciferase reporter construct (56), lPRLR, and β-galactosidase. After transfection, cells were washed and treated with or without 4 nM PRL and/or 1 nM E2 for 6 or 24 h. Lysates were harvested and analyzed for luciferase and β-galactosidase activity. C, Effect of inhibition of ERK1/2 on PRL and E2 activation of the c-Fos promoter. Cells were cotransfected with c-Fos promoter reporter construct, lPRLR, and β-galactosidase. After transfection, cells were washed and pretreated for 1 h with vehicle or 10 μM U0126, treated with or without 4 nM PRL and/or 1 nM E2 for 24 h, and then harvested and analyzed as in panel B. D, Effect of PRL and E2 on activation of Elk-1. Cells were cotransfected with the pFR-luciferase reporter construct containing five Gal4 binding sites, the pFA2-Elk-1 construct containing the Elk-1 transcriptional activator fused to the Gal4 DNA-binding domain, lPRLR, and β-galactosidase. For panels B, C, and D, relative activity represents the mean of the corrected luciferase activity from at least three independent experiments, represented as mean fold change relative to the vehicle-treated control transfection ± SEM. For statistical analysis in panels B and C, hormone treatment groups treated under the same conditions were compared using one-way ANOVA, followed by Student-Newman-Keuls multiple comparison test; different letters denote significant differences within these groups (P < 0.05). In panel D, asterisks denote significant differences between the vehicle-treated control and the indicated hormone treatment using Student’s t test, (*, P = 0.027; **, P = 0.0012; ***, P = 0.0002).