Hypoxia induces phenotypic plasticity and therapy resistance in melanoma via the tyrosine kinase receptors ROR1 and ROR2 - PubMed (original) (raw)

. 2013 Dec;3(12):1378-93.

doi: 10.1158/2159-8290.CD-13-0005. Epub 2013 Oct 8.

Katie Marchbank, Marie R Webster, Alexander A Valiga, Amanpreet Kaur, Adina Vultur, Ling Li, Meenhard Herlyn, Jessie Villanueva, Qin Liu, Xiangfan Yin, Sandy Widura, Janelle Nelson, Nivia Ruiz, Tura C Camilli, Fred E Indig, Keith T Flaherty, Jennifer A Wargo, Dennie T Frederick, Zachary A Cooper, Suresh Nair, Ravi K Amaravadi, Lynn M Schuchter, Giorgos C Karakousis, Wei Xu, Xiaowei Xu, Ashani T Weeraratna

Affiliations

Hypoxia induces phenotypic plasticity and therapy resistance in melanoma via the tyrosine kinase receptors ROR1 and ROR2

Michael P O'Connell et al. Cancer Discov. 2013 Dec.

Erratum in

Abstract

An emerging concept in melanoma biology is that of dynamic, adaptive phenotype switching, where cells switch from a highly proliferative, poorly invasive phenotype to a highly invasive, less proliferative one. This switch may hold significant implications not just for metastasis, but also for therapy resistance. We demonstrate that phenotype switching and subsequent resistance can be guided by changes in expression of receptors involved in the noncanonical Wnt5A signaling pathway, ROR1 and ROR2. ROR1 and ROR2 are inversely expressed in melanomas and negatively regulate each other. Furthermore, hypoxia initiates a shift of ROR1-positive melanomas to a more invasive, ROR2-positive phenotype. Notably, this receptor switch induces a 10-fold decrease in sensitivity to BRAF inhibitors. In patients with melanoma treated with the BRAF inhibitor vemurafenib, Wnt5A expression correlates with clinical response and therapy resistance. These data highlight the fact that mechanisms that guide metastatic progression may be linked to those that mediate therapy resistance.

Significance: These data show for the fi rst time that a single signaling pathway, the Wnt signaling pathway, can effectively guide the phenotypic plasticity of tumor cells, when primed to do so by a hypoxic microenvironment. Importantly, this increased Wnt5A signaling can give rise to a subpopulation of highly invasive cells that are intrinsically less sensitive to novel therapies for melanoma, and targeting the Wnt5A/ROR2 axis could improve the efficacy and duration of response for patients with melanoma on vemurafenib.

©2013 AACR.

PubMed Disclaimer

Conflict of interest statement

Conflict of Interest Statement: No known conflicts of interest exist.

Figures

Figure 1

Figure 1. ROR1 Expression is Inversely Correlated to Wnt5A/ROR2

(A) Gene expression of ROR1 in multiple microarray databases of melanoma cell lines divided into a more proliferative, less metastatic cohort (Cohort A) versus a highly metastatic cohort (Cohort C). (***p<0.001) (B) Quantitative real time PCR of ROR1 mRNA levels in a panel of poorly invasive compared to highly invasive cell lines (**p<0.01, ***p<0.001). (C) Quantitative real time PCR of ROR2 mRNA levels in a panel of poorly invasive compared to highly invasive cell lines (**p<0.01, ***p<0.001). (D) Western blot analysis of ROR1 and ROR2 expression in poorly invasive and highly invasive cell lines; densitometry is shown on the right. (E) Immunofluorescent analysis of ROR1 (green), Wnt5A (red) and ROR2 (green) protein expression in organotypic 3D reconstruct models of melanocytes and melanoma cells in radial growth phase (RGP-WM35), vertical growth phase (VGP-WM793) and metastatic (MET-1205Lu) stages. Hematoxylin and Eosin (H&E) staining of organotypic 3D reconstruct models of melanocytes and melanoma cells in RGP, VGP and MET stages is shown on the right.

Figure 2

Figure 2. Wnt5A Decreases ROR1 Expression Through Proteasomal Degradation

(A) Real-time PCR analysis of ROR1 mRNA levels in Wnt5A low UACC1273 and G361 cell lines +/− treatment with rWnt5A (200 ng/ml, 16 hours) (***p<0.001; error bars=STDEV). (B) Surface biotinylation and Western blot analyses of ROR1 protein expression in UACC1273 cells (+/− rWnt5A) as well as metastatic UACC903, M93-047 and UACC647 cell lines. (C) Time course analysis (5 minutes, 1 hour, 16 hours) of ROR1, Wnt5A and HSP90 in G361 and UACC1273 cells. (D) Expression of ROR1 (green) protein in G361 cells following pre-treatment with the proteasome inhibitor (MG132, 10 µM, 1 hour) in the presence or absence of rWnt5A (200 ng/ml, 10 minutes). (E) Western blot analysis of ROR1 protein in UACC1273 cells following pre-treatment with the proteasome inhibitor (MG132, 10 µM, 1 hour) in the presence or absence of rWnt5A (200 ng/ml, 16 hours). (F) Western blot analysis of ROR1 in poorly invasive UACC1273 cells following treatment with the PKC inhibitor GO6983 (1 µM, 17 hours). (G) Western blot analysis of ROR1 protein in UACC1273 cells following pre-treatment with the PKC inhibitor GO6983 (1 µM, 17 hours) in the presence or absence of rWnt5A (200 ng/ml, 16 hours).

Figure 3

Figure 3. Knockdown of ROR1 Increases the Invasive Potential of Melanoma Cells in vitro and in vivo

(A) ROR1 protein expression examined by Western blot analysis following treatment with CTRL or ROR1 siRNA (using multiple siRNAs) for 72 hours in WM35, WM983B and WM1799 cells. (B) Expression of ROR1 following ROR2 knockdown after 48 and 72 hours in cells. (C) Invasive potential of WM35, WM983B and WM1799 cells following treatment with ROR1 siRNA assessed by 3D spheroid assays. (D) Tumor growth assay (in vivo) in WM35 cells following treatment with CTRL or ROR1 siRNA (2 different ROR1 siRNAs). (E) Representative images of lung metastases following treatment with CTRL or ROR1 siRNA. (F) Graphical representation of counts of metastatic colonies seen in lungs of CTRL, ROR1_5 or ROR1_6 siRNA treated mice.

Figure 4

Figure 4. Hypoxia Induces a Switch in ROR Receptor Expression and Increases Wnt5A Expression and Motility

A) Wound healing assay of UACC903 cells seeded onto collagen coated plates and exposed to hypoxia (2% O2). Cells were scratched and imaged at 0 and 8 hours. (B) Protein expression of HIF1α in UACC1273 cells following exposure to hypoxia (2% O2) for 6 hours. (C) mRNA levels of Wnt5A and MITF following exposure to hypoxia (2%O2, 6 hours) analyzed by real-time PCR (*p<0.05, ***p<0.001; error bars=STDEV). (D) Western blot analysis of HIF1α protein expression in UACC903 cells following treatment with CTRL or HIF1α siRNA under normoxic and hypoxic conditions. (E) Western blot analysis of HIF1α and Wnt5A protein expression in M93-047 cells following treatment with CTRL or HIF1α siRNA under normoxic and hypoxic conditions. (F) Real-time PCR analysis of ROR1 mRNA levels in UACC1273 cells following exposure to hypoxia (2%O2, 6 hours and 48 hours) (***p<0.001; error bars=STDEV). Western blot analysis of ROR1 (G) and ROR2 (H) protein expression in UACC1273 cells following exposure to hypoxia. (I) Analysis of ROR2 and Wnt5A mRNA levels, by real-time PCR, following treatment with CTRL or HIF1α siRNA in M93-047 cells (*p<0.05; error bars=STDEV). (J) Wnt5A and ROR2 mRNA levels assessed by real time PCR following treatment with CTRL or Wnt5A siRNA in M93-047 cells under normoxic or hypoxic conditions (*p<0.05, **p<0.01, ***p<0.001; error bars=STDEV).

Figure 5

Figure 5. Wnt5A Degrades β-Catenin and Stabilizes HIF1α via Siah2

(A) Analysis of Siah2 mRNA levels in melanoma cells that are increasingly invasive (*p<0.05, **p<0.01; error bars=STDEV). (B) Analysis of Wnt5A and Siah2 mRNA levels by real-time PCR in M93-047 cells treated with either a CTRL or Wnt5A siRNA (*p<0.05, **p<0.01; error bars=STDEV). (C) Siah2 mRNA levels in poorly (UACC1273) and highly (M93-047) invasive melanoma cells following exposure to hypoxia (2% O2, 6 hours and 24 hours) (*p<0.05, **p<0.01; error bars=STDEV). (D) Siah2 mRNA levels following treatment with CTRL or Wnt5A siRNA in M93-047 cells under normoxic or hypoxic conditions (2%O2, 24 hours) (*p<0.05, **p<0.01). (E) HIF1α mRNA levels in M93-047 cells following treatment with CTRL or Wnt5A siRNA under normoxic or hypoxic conditions (2% O2, 24 hours) (**p<0.01; error bars=STDEV). (F) Expression of active β-catenin in cells following pre-treatment with the proteasome inhibitor (MG132, 10 µM, 1 hour) in the presence or absence of rWnt5A (200 ng/ml, 16 hours). (G) Expression of active β-catenin in cells after knockdown of SIAH2 (20nM, 48h) in the presence or absence of rWnt5A (200 ng/ml, 16 hours). (H) Schematic representation of hypoxic induction of Wnt5A, Siah2, HIF1α and ROR2 and subsequent inhibition of β -catenin.

Figure 6

Figure 6. Wnt5A and ROR2 Contribute to Intrinsic Resistance to BRAF Inhibitors

(A) Wnt5A and (C) ROR2 mRNA levels in sensitive and resistant melanoma cell lines analyzed by real-time PCR (*p< 0.05, and **p<0.005, respectively). (B) PLX-sensitive 451LU melanoma cells were treated with increasing doses of PLX4720 and proliferation was analyzed by an MTS assay. (D) PLX-resistant 1205LU cells were transfected with CNTR1 or ROR2 siRNA, and were then treated with increasing doses of PLX4720. (E) Tumor volume analysis in athymic nude mice injected with 1205LU cells transfected with CTRL or ROR2 siRNA. Once tumor formation had occurred (∼200 mm3), mice were fed either control AIN-76A chow or AIN-76A chow containing 417 mg/kg PLX4720 and tumors were tracked for 16 days. The percent change in tumor volume was recorded. (F) ROR1 mRNA levels in sensitive and resistant melanoma cell lines analyzed by real-time PCR (**p<0.005). (G) Tumor volume analysis in athymic nude mice injected with WM35 cells transfected with CTRL or ROR1 siRNA. Once tumor formation had occurred, mice were fed either control AIN-76A chow or AIN-76A chow containing 417 mg/kg PLX4720 and tumors were tracked for 21 days. The change in tumor volume was recorded.

Figure 7

Figure 7. Wnt5A and ROR2 Contribute to Acquired Resistance to BRAF Inhibitors

(A) Immunohistochemical staining of Wnt5A in 4 different patient samples pre-Vemurafenib treatment. (B) Wnt5A expression was scored and correlated to clinical response following treatment with Vemurafenib. Wnt5A expression correlates to poor patient response (**p<0.01). (C) Immunohistochemical staining of Wnt5A in patient samples pre- and post-treatment with Vemurafenib or Vemurafenib/Trametinib. Wnt5A expression increases in relapsing tumors (*p<0.05) (D) Western blot analysis of PO4-ERK and total-ERK expression in WM983B melanoma cells that are sensitive to BRAF inhibitors (parental (Par)) and in resistant (BR) subclones of the parental WM983B cells. Cells were untreated or treated with the BRAF inhibitor PLX4720 (1 µM, 48 hours). (E) ROR1, ROR2 and Wnt5A protein expression in parental and resistant WM983B cells analyzed by Western blot. (F) Western blot analysis of PO4-ERK and total-ERK expression in WM983B parental cells treated with rWnt5A (200 ng/ml) for 24 hours.

Similar articles

Cited by

References

    1. Ghislin S, Deshayes F, Middendorp S, Boggetto N, Alcaide-Loridan C. PHF19 and Akt control the switch between proliferative and invasive states in melanoma. Cell Cycle. 2012;11:1634–1645. - PubMed
    1. Hoek KS, Schlegel NC, Brafford P, Sucker A, Ugurel S, Kumar R, et al. Metastatic potential of melanomas defined by specific gene expression profiles with no BRAF signature. Pigment Cell Res. 2006;19:290–302. - PubMed
    1. Dissanayake SK, Olkhanud PB, O'Connell MP, Carter A, French AD, Camilli TC, et al. Wnt5A regulates expression of tumor-associated antigens in melanoma via changes in signal transducers and activators of transcription 3 phosphorylation. Cancer Res. 2008;68:10205–10214. - PMC - PubMed
    1. Cheli Y, Giuliano S, Botton T, Rocchi S, Hofman V, Hofman P, et al. Mitf is the key molecular switch between mouse or human melanoma initiating cells and their differentiated progeny. Oncogene. 2011;30:2307–2318. - PubMed
    1. Cheli Y, Giuliano S, Fenouille N, Allegra M, Hofman V, Hofman P, et al. Hypoxia and MITF control metastatic behaviour in mouse and human melanoma cells. Oncogene. 2012;31:2461–2470. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources