Targeting SOX9 for degradation to inhibit chemoresistance, metastatic spread, and recurrence

ABSTRACT

Cancer cells with stem-like properties are believed to contribute to treatment resistance, dissemination, and recurrence. SOX9 controls stem cell plasticity and its deregulation may provide a basis for tumor progression. Here, we summarize our findings of targeted SOX9 destruction by SCF^FBW7 (Skp1/Cul1/F-box) in medulloblastoma and its potential for therapeutic intervention.

SOX9 (sex-determining region Y [SRY] - containing box 9), a member of the high-mobility group (HMG) family of transcription factors, is known to be required for chondrocyte specification and was first identified as causal for campomelic dysplasia syndrome. Given the context of its discovery, many of the early studies focused on functions of SOX9 in the developmental process of the endochondral skeleton, where it directs chondrogenic programming in the early mesenchymal cell population to differentiate toward hypertrophic chondrocytes.¹ However, in recent years the oncogenic potential of SOX9, in combination with other cancer drivers, during the initiation and progression of various malignancies has been increasingly recognized. This has been exemplified in several studies demonstrating that SOX9 functions as a critical transcription factor in MYCN-driven sonic hedgehog–dependent (SHH) medulloblastoma² and Hedgehog/WNT-driven basal cell carcinoma,³ in addition to co-operating with SNAI2 in the conversion of differentiated cells to stem cells and endowing tumor-initiating and metastatic abilities in breast cancer cells.⁴

Mechanistic studies have linked SOX9 oncogenicity to the regulation of several cancer-specific gene networks, including self-renewal, differentiation, and extracellular matrix/cytoskeleton remodeling.³ These broad SOX9 regulatory capacities have been proposed to depend on differential SOX9 interaction with the global genome, either through indirect binding of SOX9 proximal to the transcriptional start site via the basal transcriptional complex or through direct binding of SOX9 to multiple enhancer elements via low-affinity SOX9 dimeric motifs.¹ It remains to be investigated how changes in SOX9 protein levels affect its preferential binding distribution in the genome and/or its transcriptional activity.

Dysregulation of specific signaling pathways including, but not limited to, Hedgehog, WNT, EGFR, and NOTCH1, have been shown to transcriptionally induce SOX9. However, little is known about how different post-translational modifications affect SOX9 protein stability in cancer cells. Our discovery that SOX9 is a substrate of the SCF^FBW7 (Skp1/Cul1/F-box) ubiquitin ligase initially arose from a previous proteomic screen aimed to identify novel targets of F-box and WD repeat domain-containing 7 (FBW7).⁵ Our recent findings demonstrate that FBW7 suppresses SOX9 protein via the ubiquitin-proteasome system (UPS), and that dysregulation of this process contributes to metastatic potential and resistance to cisplatin treatment in medulloblastoma (Fig. 1;⁶). FBW7 binds SOX9 through an evolutionary conserved GSK3-phosphorylated motif, which spans Thr236–Thr240 of the SOX9 protein. The direct physical interaction also requires intact FBW7 substrate recognition residues (Arg465, 479, and 505) to trigger SOX9 poly-ubiquitylation and proteasomal degradation. Indeed, we identified frequent FBW7 mutations in primary medulloblastoma specimens that disrupt FBW7-SOX9 binding. Importantly, an independent study by Hong et al. further extends our observation to several other cancer types, including colorectal adenocarcinoma, osteosarcoma, and lung adenocarcinoma, in response to specific DNA-damaging drugs.⁷

Figure 1.

The SCF^FBW7 ubiquitin ligase targets SOX9 for proteasomal destruction. The PI3K/AKT/mTOR signaling axis regulates GSK3/FBW7-mediated SOX9 ubiquitylation and proteasomal degradation. Loss of FBW7 or oncogenic activation of PI3K/AKT/mTOR signaling stabilizes SOX9 protein, which contributes to increased medulloblastoma malignancy. Inhibitors of the PI3K/AKT/mTOR pathway may be used to overcome chemoresistance in tumors with functional FBW7.

Additionally, our clinical observations indicated that FBW7 has an important function in restraining medulloblastoma malignancy. Inactivating mutations and/or downregulation of FBW7 expression in all medulloblastoma subtypes were linked to elevated SOX9 levels, increased brain stem and spinal cord metastasis, and poor patient outcome. Using tetracycline-inducible FBW7 and/or SOX9 expression systems we were able to model SOX9-induced medulloblastoma cell migration in vitro and metastasis in vivo, and attenuated such phenotypes through FBW7 induction. Transcriptional profiling of medulloblastoma cells expressing either wild-type SOX9 or FBW7-resistant SOX9 mutants cultured in stem cell-like conditions further linked aggressive metastatic behavior to an epithelial-to-mesenchymal (EMT)–like reprograming involving upregulation of SNAI2, among other proteins. As SNAI2 not only functionally cooperates as transcription factor,⁴ but also regulates SOX9 protein stability via the UPS,⁸ it will be important to explore the function of SNAI2 in SOX9-driven medulloblastoma and other malignancies in future studies.

These results should also be interpreted in light of resistance of medulloblastoma to treatment and patient relapse. SOX9 expression in medulloblastoma promoted resistance to cisplatin, a clinically relevant drug that is commonly used for treatment of this disease. Intriguingly, and in agreement with Hong et al., we observed that cisplatin treatment promoted SOX9 degradation through a FBW7/GSK3-mediated process. While the exact molecular mechanisms underlying SOX9-associated cisplatin resistance remain to be fully elucidated, we presume that the combination of EMT-like reprogramming and SOX9-regulated expression of several genes linked to cisplatin sensitivity, including the copper transporter ATP7A, and the modulation of intracellular ERK1/2 signaling contributes to the overall reduced efficacy of cisplatin treatment.

Global genome sequencing and expression profiling of SHH medulloblastoma has documented frequent mutations of PIK3CA and PTEN together with hyperactivity of the PI3K/AKT/mTOR pathway.⁹ As activation of the PI3K/AKT/mTOR pathway attenuates GSK3 activity, we hypothesized that stimulating GSK3 by means of PI3K/AKT/mTOR pathway inhibition would increase SOX9 degradation by FBW7 and consequently lead to re-sensitization to cisplatin treatment. In support of this hypothesis, we demonstrated enhanced SOX9 protein turnover by FBW7 upon treatment of medulloblastoma cells with various PI3K, AKT, and/or mTOR inhibitors, and synthetically lethality of their combination with cisplatin (Fig. 1;⁶). Thus, our findings suggest that targeting the PI3K/AKT/mTOR pathway could be used for clinical intervention, not only in patients with recurrent, chemoresistant medulloblastoma with functional FBW7 but potentially also for other SOX9-driven cancers.

Last but not least, a recent study from Massague's laboratory revealed that SOX9 is enriched and important for maintenance of latency-competent cancer cells.¹⁰ This observation suggests the potential importance of targeting SOX9 not only as a strategy to enhance treatment efficacy but also possibly for circumventing relapse through targeting the slow-cycling and quiescent cancer stem-like subpopulations. To this end, we believe that development of specific SOX9 inhibitors or compounds that attenuate SOX9 expression, dependent or independent of FBW7, may be of significant importance for the development of future personalized anticancer therapies.

Disclosure of potential conflict of interest

OS receives a research grant from Astra Zeneca for a separate project.

Funding

Work from our study cited in this review was funded by the Swedish Childhood Cancer Foundation (FJS, OS, ASR), the Swedish Cancer Society (FJS, OS), the Swedish Research Council (FJS, OS), the European Research Council (FJS), the Ragnar Söderberg Foundation (FJS), the Swedish Society of Medicine (FJS), the Åke Wiberg Foundation (FJS), Science for Life Laboratories (FJS, OS), Åke Olssons Stiftelse (OS), Radiumhemmets Forskningsfonder (OS, ASR), Karolinska Institute Foundations (OS, ASR), and Worldwide Cancer Research (FJS).