Prostate cancer (PCa) is a commonly diagnosed disease with one of the highest heritability estimates among cancers. Among germline variations that account for PCa heritability, a recurrent mutation (G84E) in HOXB13 is associated with a 3 to 6 fold increase in PCa risk. HOXB13 is a prostate-specific transcription factor that plays a role in prostate development. Other mutations in HOXB13 in PCa patients have further implicated HOXB13 in PCa biology. The goal of this study is to generate HOXB13 cell line models of normal prostate to examine the role of HOXB13 in driving cellular differentiation and/or transformation. To this end, we utilized 957E/hTERT, a cell line derived from a radical prostatectomy specimen, as a model for normal prostate epithelial cells. HOXB13 WT or G84E was stably expressed in 957E/hTERT with a lentivirus system in the presence and absence of exogenous AR. HOXB13 caused a morphology change whereby an epithelial morphology transitioned to a fibroblastic-like structure. Cells adopted elongated bodies with a frequent appearance of lamellipodia. The morphologic conversion was transient in AR- cells but was maintained in AR+ cells. Analyses of epithelial cell markers using qPCR revealed a change in the epithelial cell marker profile (AR-: no change p63, CD49f↓, K5↓, K14↑, K18↓; AR+: no change p63, CD49f, K5, K14↑, K18↓). K14 upregulation was striking with a >10 fold increase. AR alone could upregulate K14 to a level comparable to HOXB13-induced K14 in AR- cells, which was surprising since AR mainly functions in luminal cells where K14 expression is turned off. Coexpression of HOXB13 and AR had a synergistic effect in K14 upregulation. Despite the morphology change suggestive of EMT, vimentin was downregulated by HOXB13 in AR- and AR+ cells. In vitro growth assays showed that HOXB13 reduced the proliferation rate of AR+ cells but not AR- cells. Gap closure assays revealed that HOXB13 does not induce a qualitative change in migration. Similar results were obtained with G84E models. Our study demonstrates that HOXB13 may have a potential function in dictating a cell fate in the epithelial lineage as evidenced by a change in the epithelial cell marker profile. Conclusively, HOXB13 and AR can drive K14 by themselves and in collaboration. This finding is exciting in light of the recent discovery (Cheung KJ et al. 2016) that K14 is not merely a basal cell marker but a major mediator of collective invasion in breast cancer. During breast development, K14 is expressed in the cap cells at the growing tips of mammary ducts invading into the fat pad. K14 driven by HOXB13 in collaboration with AR may regulate prostate epithelial budding into the surrounding mesenchyme during development. Hypothetical dysregulation of this embryonic programming by HOXB13 mutation could predispose a carrier to PCa. This study warrants the need for HOXB13 mouse models to test these fundamental questions.

Citation Format: Dorhyun Johng, Charles M. Ewing, William B. Isaacs. HOXB13 collaborates with AR to alter the cellular phenotype of prostate epithelial cells through cytokeratin 14 upregulation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3524. doi:10.1158/1538-7445.AM2017-3524