E3 ubiquitin ligase HOIP attenuates apoptotic cell death induced by cisplatin

Abstract

The genotoxin cisplatin is commonly used in chemotherapy to treat solid tumors, yet our understanding of the mechanism underlying the drug response is limited. In a focused siRNA screen, using an siRNA library targeting genes involved in ubiquitin and ubiquitin-like signaling, we identified the E3 ubiquitin ligase HOIP as a key regulator of cisplatin-induced genotoxicity. HOIP forms, with SHARPIN and HOIL-1L, the linear ubiquitin assembly complex (LUBAC). We show that cells deficient in the HOIP ligase complex exhibit hypersensitivity to cisplatin. This is due to a dramatic increase in caspase-8/caspase-3-mediated apoptosis that is strictly dependent on ATM-, but not ATR-mediated DNA damage checkpoint activation. Moreover, basal and cisplatin-induced activity of the stress response kinase JNK is enhanced in HOIP-depleted cells and, conversely, JNK inhibition can increase cellular resistance to cisplatin and reverse the apoptotic hyperactivation in HOIP-depleted cells. Furthermore, we show that HOIP depletion sensitizes cancer cells, derived from carcinomas of various origins, through an enhanced apoptotic cell death response. We also provide evidence that ovarian cancer cells classified as cisplatin-resistant can regain sensitivity following HOIP downregulation. Cumulatively, our study identifies a HOIP-regulated antiapoptotic signaling pathway, and we envisage HOIP as a potential target for the development of combinatorial chemotherapies to potentiate the efficacy of platinum-based anticancer drugs.

Figure 1

siRNA screen identifies HOIP as an enhancer of cisplatin-induced genotoxicity. A, schematic illustration of the primary siRNA screen, using the siRNA “ubiquitome” library in U2OS cells. B, maximum log₂SF is plotted for each gene that was classed as cisplatin-hypersensitive in the primary screen. C, paired wild-type HOIP^+/+ and HOIP^C879S/C879S knock-in mouse embryonic fibroblasts; cell viability was assayed 72 hrs after cisplatin treatment by MTS cell proliferation assay. D, HEK293 cells were transfected with the indicated siRNAs and cell lysates were assayed for efficient knockdown by immunoblotting with the indicated antibodies (top panel). Cell viability was assayed as in (C). Data in (C) and (D) are represented as mean ± SEM from three independent experiments.

Figure 2

HOIP is not required for DNA damage checkpoint activation. A, HEK293 cells were transfected with siCON or siHOIP and treated with 5 μM cisplatin for the indicated times. Cell lysates were analyzed by immunoblotting with the antibodies indicated. B, siCON or siHOIP transfected HEK293 cells were treated with 8 μM cisplatin for 2 hours before being allowed to recover for the times indicated. Cells were then fixed and stained with DAPI to visualize nuclei and γ-H2AX antibody to visualize γ-H2AX foci. C, quantitation of the data in (B): cells with more than 6 foci were classed as γ-H2AX positive. Data in (C) is presented as mean ± SEM from three independent experiments.

Figure 3

HOIP depletion sensitizes cells to cisplatin-induced apoptotic cell death. A, U2OS (left panel) or HEK293 (right panel) cells were transfected with mock siRNA (siCON) or siRNA targeting HOIP (siHOIP) and treated with 5 μM cisplatin for the indicated time. Cell lysates were analyzed by immunoblotting with the antibodies indicated. B, siCON or siHOIP transfected U2OS cells were treated with 5 μM cisplatin for 48 hours before caspase-3 activity was measured using a caspase-3 Glo assay; * indicates a p-value of 0.0028 C, HEK293 cells were treated as described in (B); * indicates a p-value of 0.0222. D, paired wild-type HOIP^+/+ and HOIP^C879S/C879S knock-in mouse embryonic fibroblasts were treated with 3 μM cisplatin for the indicated time. Cell lysates were analyzed by immunoblotting. E and F, as (A) except that cells were transfected with siSHARPIN or siHOIL-1L. G, siCON or siHOIP transfected HEK293 cells were treated with vehicle only (−) or 20 μM Z-IETD-FMK caspase-8 inhibitor (+) one hour before treatment with 5 μM cisplatin for the indicated times. Cell lysates were analyzed by immunoblotting. H, as (B) except that a caspase-8 Glo assay was used to measure caspase-8 activity; * indicates a p-value of 0.0076. I, HEK293 cells were treated as described in (H). * indicates a p-value of 0.014. Data in (B), (C), (H) and (I) are represented as mean ± SEM from three independent experiments and Student’s t-test was used to calculate significance.

Figure 4

Apoptotic sensitization of HOIP-depleted cells requires ATM but is independent of ATR activity. A, siRNA transfected HEK293 cells were treated with vehicle only (−) or 10 μM KU-55933 ATM inhibitor (+) one hour before treatment with 5 μM cisplatin for the indicated times. Cell lysates were analyzed by immunoblotting. B, quantitation of the data shown in (A). Image J software was used to quantify immunoblots; * indicates a p-value of 0.0078. C, as (A) except that 5 μM ETP-46464 ATR inhibitor was used. D, quantitation of the data shown in (C). E, As (A) except that U2OS cells were used. F, HA-IκBa(S32A,S36A)-expressing HEK293 cells were transfected either with siCON or siHOIP and treated with 5 μM cisplatin for the indicated times. Cell lysates were subjected to immunoblot analyses. Data in (B) and (D) are represented as mean ± SEM from three independent experiments and Student’s t-test was used to calculate significance.

Figure 5

JNK activity is elevated and essential for apoptotic cell death in HOIP-depleted cells. A, siRNA transfected HEK293 cells were treated with 25 μM cisplatin for the indicated times. Cell lysates were analyzed by immunoblotting. B, siCON or siHOIP transfected HEK293 cells were pre-treated with 3 μM JNK inhibitor or 10 μM ATM inhibitor before vehicle only (−) or 5 μM cisplatin treatment (+) as indicated. JNK kinases were immunoprecipitated from cell lysates, and kinase activity was measured using an in vitro kinase assay. Activity is presented as mU/mg cell lysate; * Student’s t-test was used to calculate significance: p=0.0396. C, siCON- or siHOIP-transfected HEK293 cells were pre-treated with vehicle only (−) or 3 μM JNK inhibitor (+) before treatment with 5 μM cisplatin for the indicated times. Cell lysates were analyzed by immunoblotting. D, Cell viability of siCON or siHOIP-transfected cells pre-treated with JNK inhibitor in response to cisplatin was measured by MTS cell proliferation assay. E, As (C) except that U2OS cells were used. Data in (B) and (D) are represented as mean ± SEM from three independent experiments.

Figure 6

HOIP-depletion sensitizes a range of cancer cell lines to cisplatin-induced apoptotic cell death. HCT116 (A), HeLa (B), PC3 (C), ZR.75.1 (D), MDA-MB-231 (E) or PEA1 (F) cells were transfected with mock siRNA (siCON) or siRNA targeting HOIP (siHOIP) and treated with 5 μM cisplatin for the indicated time before cell lysates were analyzed by immunoblotting with the antibodies indicated (upper panels). Cell viability of siCON- or siHOIP-transfected cells in response to the indicated concentration range of cisplatin was measured by MTS cell proliferation assays (lower panels). Data shown are represented as mean ± SEM from three independent experiments.

Figure 7

HOIP depletion sensitizes cisplatin-resistant ovarian cancer cell lines. A and B (left), A2780^CisS and A2780^CisR cells were transfected with siCON or siHOIP as indicated and cell viability assessed by MTS cell proliferation assays. A and B (right), A2780^CisS and A2780^CisR cells were transfected with siCON or siHOIP and treated with 5 μM cisplatin for the indicated time. Cell lysates were analyzed by immunoblotting. C–F, cell lines were transfected with siCON or siHOIP as indicated and cell viability assessed by MTS cell proliferation assays. Data shown are presented as mean ± SEM from three independent experiments.