Protein kinase C-theta regulates KIT expression and proliferation in gastrointestinal stromal tumors

Abstract

Oncogenic KIT or PDGFRA receptor tyrosine kinase mutations are compelling therapeutic targets in gastrointestinal stromal tumors (GISTs), and the KIT/PDGFRA kinase inhibitor, imatinib, is standard of care for patients with metastatic GIST. However, most of these patients eventually develop clinical resistance to imatinib and other KIT/PDGFRA kinase inhibitors and there is an urgent need to identify novel therapeutic strategies. We reported previously that protein kinase C-theta (PKCtheta) is activated in GIST, irrespective of KIT or PDGFRA mutational status, and is expressed at levels unprecedented in other mesenchymal tumors, therefore serving as a diagnostic marker of GIST. Herein, we characterize biological functions of PKCtheta in imatinib-sensitive and imatinib-resistant GISTs, showing that lentivirus-mediated PKCtheta knockdown is accompanied by inhibition of KIT expression in three KIT+/PKCtheta+ GIST cell lines, but not in a comparator KIT+/PKCtheta- Ewing's sarcoma cell line. PKCtheta knockdown in the KIT+ GISTs was associated with inhibition of the phosphatidylinositol-3-kinase/AKT signaling pathway, upregulation of the cyclin-dependent kinase inhibitors p21 and p27, antiproliferative effects due to G(1) arrest and induction of apoptosis, comparable to the effects seen after direct knockdown of KIT expression by KIT short-hairpin RNA. These novel findings highlight that PKCtheta warrants clinical evaluation as a potential therapeutic target in GISTs, including those cases containing mutations that confer resistance to KIT/PDGFRA kinase inhibitors.

Figure 1

Immunoblotting evaluations of GIST cell lines show strong PKCθ activation and expression in KIT-positive lines (GIST882, GIST430 and GIST48) but not in KIT-negative lines (GIST62 and GIST522). The Jurkat T-cell acute lymphoblastic leukemia provides a positive control for PKCθ activation and expression.

Figure 2

A) Immunoblotting evaluations of KIT and PKCθ expression and activation, in KIT-positive GIST cell lines (GIST882, GIST430 and GIST48) at 96 hours after infection by lentiviral KIT or PKCθ shRNA constructs. KIT shRNA infection inhibited KIT expression and activation in each cell line. PKCθ shRNA inhibition inhibited PKCθ expression and activation in each cell line, and also inhibited KIT expression and activation in the imatinib-resistant GIST48 and GIST430 cell lines. Control lanes, for each cell line, include uninfected cells (untreated lane) and cells infected with empty lentiviral vector. B) Comparison immunoblotting evaluations of KIT and PKCθ expression, after infection of KIT-positive Ewing's sarcoma (EWS502) cells with lentiviral KIT or PKCθ shRNA constructs. These studies show that PKCθ shRNA treatment does not inhibit KIT expression, in the absence of PKCθ expression, indicating that PKCθ shRNA-mediated KIT inhibition in GIST (2A) does not result from nonspecific interactions between the PKCθ shRNA and KIT mRNA. The GIST882 cells provide a positive control for KIT and PKCθ expression. EWS502 no-treatment controls include uninfected cells (untreated lane) and EWS502 infected with empty lentiviral vector. C) Immunoblotting evaluations of KIT-positive GIST cell lines (GIST882, GIST430 and GIST48) at 96 hours after infection by lentiviral KIT or PKCθ shRNA constructs. The immunoblotting assays evaluated affects of KIT and PKCθ knockdown on signaling intermediates (AKT, MAPK p42/44, S6, STAT1), proliferation markers (Cyclin A and PCNA), and cell cycle checkpoint proteins (p27, p21 and p53). The empty vector lane is a parallel control. The PI3-K and Actin immunostains show equivalence of lane loading. Control lanes, for each cell line, include uninfected cells (untreated lane) and cells infected with empty lentiviral vector.

Figure 2

A) Immunoblotting evaluations of KIT and PKCθ expression and activation, in KIT-positive GIST cell lines (GIST882, GIST430 and GIST48) at 96 hours after infection by lentiviral KIT or PKCθ shRNA constructs. KIT shRNA infection inhibited KIT expression and activation in each cell line. PKCθ shRNA inhibition inhibited PKCθ expression and activation in each cell line, and also inhibited KIT expression and activation in the imatinib-resistant GIST48 and GIST430 cell lines. Control lanes, for each cell line, include uninfected cells (untreated lane) and cells infected with empty lentiviral vector. B) Comparison immunoblotting evaluations of KIT and PKCθ expression, after infection of KIT-positive Ewing's sarcoma (EWS502) cells with lentiviral KIT or PKCθ shRNA constructs. These studies show that PKCθ shRNA treatment does not inhibit KIT expression, in the absence of PKCθ expression, indicating that PKCθ shRNA-mediated KIT inhibition in GIST (2A) does not result from nonspecific interactions between the PKCθ shRNA and KIT mRNA. The GIST882 cells provide a positive control for KIT and PKCθ expression. EWS502 no-treatment controls include uninfected cells (untreated lane) and EWS502 infected with empty lentiviral vector. C) Immunoblotting evaluations of KIT-positive GIST cell lines (GIST882, GIST430 and GIST48) at 96 hours after infection by lentiviral KIT or PKCθ shRNA constructs. The immunoblotting assays evaluated affects of KIT and PKCθ knockdown on signaling intermediates (AKT, MAPK p42/44, S6, STAT1), proliferation markers (Cyclin A and PCNA), and cell cycle checkpoint proteins (p27, p21 and p53). The empty vector lane is a parallel control. The PI3-K and Actin immunostains show equivalence of lane loading. Control lanes, for each cell line, include uninfected cells (untreated lane) and cells infected with empty lentiviral vector.

Figure 2

A) Immunoblotting evaluations of KIT and PKCθ expression and activation, in KIT-positive GIST cell lines (GIST882, GIST430 and GIST48) at 96 hours after infection by lentiviral KIT or PKCθ shRNA constructs. KIT shRNA infection inhibited KIT expression and activation in each cell line. PKCθ shRNA inhibition inhibited PKCθ expression and activation in each cell line, and also inhibited KIT expression and activation in the imatinib-resistant GIST48 and GIST430 cell lines. Control lanes, for each cell line, include uninfected cells (untreated lane) and cells infected with empty lentiviral vector. B) Comparison immunoblotting evaluations of KIT and PKCθ expression, after infection of KIT-positive Ewing's sarcoma (EWS502) cells with lentiviral KIT or PKCθ shRNA constructs. These studies show that PKCθ shRNA treatment does not inhibit KIT expression, in the absence of PKCθ expression, indicating that PKCθ shRNA-mediated KIT inhibition in GIST (2A) does not result from nonspecific interactions between the PKCθ shRNA and KIT mRNA. The GIST882 cells provide a positive control for KIT and PKCθ expression. EWS502 no-treatment controls include uninfected cells (untreated lane) and EWS502 infected with empty lentiviral vector. C) Immunoblotting evaluations of KIT-positive GIST cell lines (GIST882, GIST430 and GIST48) at 96 hours after infection by lentiviral KIT or PKCθ shRNA constructs. The immunoblotting assays evaluated affects of KIT and PKCθ knockdown on signaling intermediates (AKT, MAPK p42/44, S6, STAT1), proliferation markers (Cyclin A and PCNA), and cell cycle checkpoint proteins (p27, p21 and p53). The empty vector lane is a parallel control. The PI3-K and Actin immunostains show equivalence of lane loading. Control lanes, for each cell line, include uninfected cells (untreated lane) and cells infected with empty lentiviral vector.

Figure 3

A) Immunoblotting evaluations of GIST882 cells, at 10 and 20 days after infection by lentiviral KIT or PKCθ shRNA constructs. KIT shRNA resulted in decreased expression of KIT, phospho-AKT, phospho-S6, and the proliferation markers Cyclin A and PCNA. PKCθ shRNA resulted in decreased expression of PKCθ, KIT, phospho-AKT, and Cyclin A. The PI3-K and β-actin immunostains show equivalence of lane loading. B) GIST882 cell culture appearance, when evaluated at 10 and 20 days after infection by lentiviral KIT or PKCθ shRNA constructs, showing growth inhibition compared to control GIST882 cultures infected with lentiviral empty constructs.

Figure 3

A) Immunoblotting evaluations of GIST882 cells, at 10 and 20 days after infection by lentiviral KIT or PKCθ shRNA constructs. KIT shRNA resulted in decreased expression of KIT, phospho-AKT, phospho-S6, and the proliferation markers Cyclin A and PCNA. PKCθ shRNA resulted in decreased expression of PKCθ, KIT, phospho-AKT, and Cyclin A. The PI3-K and β-actin immunostains show equivalence of lane loading. B) GIST882 cell culture appearance, when evaluated at 10 and 20 days after infection by lentiviral KIT or PKCθ shRNA constructs, showing growth inhibition compared to control GIST882 cultures infected with lentiviral empty constructs.

Figure 4

A) Cell viability was evaluated in KIT-positive GIST882 (black bars), GIST48 (gray bars), and GIST430 (white bars) cell lines, at day 8 and day 12 after infection with lentiviral KIT and PKCθ shRNAs. Viability was analyzed using Cell-titer Glo® ATP-based luminescence assay. The data were normalized to the empty lentivirus control infections, and represent the mean values (± s.d.) from quadruplicate cultures. B) Apoptosis was evaluated in GIST882 (black bars) and GIST48 (gray bars) cell lines, at day 4 after infection with lentiviral KIT and PKCθ shRNAs. Caspase 3/7 activity was measured using a Caspase-Glo® luminescence assay. The data were normalized to the empty lentivirus control infections, and represent the mean values (± s.d.) from quadruplicate cultures. C) Cell cycle analyses were performed in GIST882 and GIST48 cells at 8 days and 4 days after infection by lentiviral KIT or PKCθ shRNA constructs, respectively. Apoptotic response was greater, in both cell lines, after treatment with PKCθ shRNA than after treatment with KIT shRNA.

Figure 4

A) Cell viability was evaluated in KIT-positive GIST882 (black bars), GIST48 (gray bars), and GIST430 (white bars) cell lines, at day 8 and day 12 after infection with lentiviral KIT and PKCθ shRNAs. Viability was analyzed using Cell-titer Glo® ATP-based luminescence assay. The data were normalized to the empty lentivirus control infections, and represent the mean values (± s.d.) from quadruplicate cultures. B) Apoptosis was evaluated in GIST882 (black bars) and GIST48 (gray bars) cell lines, at day 4 after infection with lentiviral KIT and PKCθ shRNAs. Caspase 3/7 activity was measured using a Caspase-Glo® luminescence assay. The data were normalized to the empty lentivirus control infections, and represent the mean values (± s.d.) from quadruplicate cultures. C) Cell cycle analyses were performed in GIST882 and GIST48 cells at 8 days and 4 days after infection by lentiviral KIT or PKCθ shRNA constructs, respectively. Apoptotic response was greater, in both cell lines, after treatment with PKCθ shRNA than after treatment with KIT shRNA.

Figure 4

A) Cell viability was evaluated in KIT-positive GIST882 (black bars), GIST48 (gray bars), and GIST430 (white bars) cell lines, at day 8 and day 12 after infection with lentiviral KIT and PKCθ shRNAs. Viability was analyzed using Cell-titer Glo® ATP-based luminescence assay. The data were normalized to the empty lentivirus control infections, and represent the mean values (± s.d.) from quadruplicate cultures. B) Apoptosis was evaluated in GIST882 (black bars) and GIST48 (gray bars) cell lines, at day 4 after infection with lentiviral KIT and PKCθ shRNAs. Caspase 3/7 activity was measured using a Caspase-Glo® luminescence assay. The data were normalized to the empty lentivirus control infections, and represent the mean values (± s.d.) from quadruplicate cultures. C) Cell cycle analyses were performed in GIST882 and GIST48 cells at 8 days and 4 days after infection by lentiviral KIT or PKCθ shRNA constructs, respectively. Apoptotic response was greater, in both cell lines, after treatment with PKCθ shRNA than after treatment with KIT shRNA.