Atypical Protein Kinase C and Protein Kinase A in Drug-Resistant Basal Cell Carcinoma

Basal cell carcinoma (BCC) is driven by overactivation of the Hedgehog signaling pathway whose current FDA-approved chemical inhibitors can efficiently suppress protein receptor Smoothened (SMOi) but have presented serious resistance problems in clinical settings. In SMOi-resistant BCCs, we documented an important correlation between the increase in mRNA level of the constitutively active isoform of atypical protein kinase C (aPKC-K) and the surge in mutational load of ciliary proteins. We demonstrated, for the first time, that aPKC protein expression was upregulated upon cilia loss and that chemical inhibition of aPKC increased both percent ciliation and cilia shaft length, both in cells and BCC mice. We also found that the Hedgehog pathway activity (a) was significantly diminished when ciliogenesis was disrupted by transient knockdown of important ciliary genes and (b) was adequately rescued when aPKC-K was simultaneously overexpressed. At the same time, cells that overexpressed aPKC-K displayed resistance to Vismodegib, a widely prescribed SMOi. We also recorded higher aPKC protein expression and lower ciliogenesis in human superficial BCC subtype in comparison to nodular BCC subtype, suggesting a connection between the interfollicular epidermal origin of superficial BCCs and the SMOi-resistance tendency, presumably via aPKC overactivation coupled with cilia loss. In a second project, we detailed the functional effects of phosphorylation by protein kinase A (PKA) at individual and combinatorial PKA-specific phosphosites on GLI1. Artificial loss of PKA-phosphorylation promoted growth in murine BCC cells and allografts; whereas overexpression of clinical mutations found in resistant BCCs near the PKA-sites on GLI1 altered GLI1 activity, localization, and stability. We also described, for the first time, a truncated protein product of GLI1 when PKA-phosphorylation was mimicked. These findings challenged the currently accepted notion that GLI1 could only be completel