TAp63–miRNA–AURKA Axis as a Therapeutic Target for Cutaneous Squamous Cell Carcinoma

Cutaneous squamous cell carcinoma (cuSCC) is the second most common form of skin cancer in the United States, and the incidence of cuSCC is increasing worldwide with about 1 million cases per year (1). Although the vast majority of cuSCC cases have good prognosis, patients with invasive cuSCC show poor outcomes, with a recurrence rate of 6%–10% and cancer-associated mortality rate of approximately 9% (1). This is mainly due to lack of defined molecular markers associated with this disease. In addition, the main treatments for invasive cuSCC include surgical resection, adjuvant radiation, and immune checkpoint inhibitors, thus, targeted therapies need to be developed. Clinical and in vivo studies determining the molecular mechanisms that contribute to cuSCC progression are required to develop novel treatment strategies and to improve the prognosis of cuSCC. Previous studies by Chitsazzadeh and colleagues (2) identified common genetic drivers of SCC from different tissues of origin through transcriptome analyses of human SCC and UV-induced mouse SCC. These include pathways associated with RB1, TP53, and TP63, however, functional validation of these candidate markers has not yet been performed.

Accumulating literature has suggested involvement of TP63 in development of cuSCC and other SCCs of different origins (lung, head and neck, esophagus, bladder), although TP63 primarily plays crucial roles in epithelial biology including development of stratified epithelial tissues. Studies using genetically engineered mouse models and correlative studies using human cancer tissues have provided evidence for the role of TP63 in SCC and other types of cancer (2–4). TP63 belongs to the TP53 family transcription factors (4). Like other TP53 family members (TP53, TP73), TP63 binds to TP53-responsive elements and transactivates numerous downstream target genes involved in cell-cycle progression and cell death. TP63 has multiple isoforms, which can be roughly categorized into two major groups, TAp63 with the N-terminal transactivation domain and ΔNp63 without the N-terminal domain. TAp63 appears to have suppressive activities on tumor development and metastasis like TP53, while ΔNp63 has a dominant-negative activity on TAp63 and other cancer-promoting functions (3). It should be noted that several studies do not discriminate between TAp63 and ΔNp63, mainly due to the presence of various TAp63/ΔNp63 isoforms in human cancers/tissues, lack of specificity of many available antibodies, and insufficient consideration for the presence of isoforms. Hence, animal models that specifically deplete or overexpress a specific isoform of TP63 are very useful to understand the role of each isoform in tissue development, biology, and cancer progression.

In this issue, Davis and colleagues (5) have defined the crucial roles of TAp63 and its downstream target miRNAs in the suppression of ultraviolet radiation (UVR)-associated cuSCC through analyses of cuSCC mouse models and human cuSCC. They treated wild-type and TAp63^−/− mice on a pure C57BL/6 genetic background with UVR and showed that TAp63^−/− mice had increased incidence of cuSCC, as compared with wild-type mice. Genetic deletion of TAp63 in mice resulted in changes of gene expression for 1,993 mRNAs and 90 miRNAs in cuSCC. Functional pair analyses of miRNA signatures from TAp63^−/− cuSCC and human cuSCC tissues identified 10 conserved differentially expressed miRNAs, including miR-30c-2* and miR-497. Reduced miR-30c-2* and miR-497 expression by TAp63 knockdown in human epithelial keratinocytes was dependent on Dicer, a downstream target of TAp63. Biologically, expression of miR-30c-2* or miR-497 mimics induced cell-cycle arrest and/or apoptosis in cuSCC cell lines. In addition, proteogenomic analysis and its comparison with mRNAs altered in mouse TAp63^−/− cuSCC and human cuSCC identified potential targets of these miRNAs, including FAT2 (miR-30c-2*), ORC1 (miR-30c-2*), KIF18B (miR-30c-2* and miR-497), PKMYT1 (miR-30c-2* and miR-497), AURKA (miR-497), and CDK6 (miR-497). Indeed, knockdown of these miRNA targets in cuSCC cell lines led to either decrease in cell proliferation or increase in cell death, suggesting tumor-suppressive functions of these miRNAs and tumor-promoting activities of the target proteins. Of these potential targets, knockdown of AURKA, a target of miR-497, consistently inhibited cell growth and induced apoptosis in multiple cuSCC cell lines. Furthermore, there was an inverse relationship between miR-497 and AURKA expression in human cuSCC tissues. In addition, low miR-497 and high AURKA levels were associated with poor survival in patients with cuSCC. These observations strongly suggest that miR-497 suppresses cuSCC progression primarily through AURKA inhibition and that AURKA could be a viable therapeutic target for cuSCC. Indeed, treatment of cuSCC cells with alisertib, an AURKA-selective inhibitor, suppressed cell proliferation and induced apoptosis of cuSCC cells in culture, while it had a moderate inhibitory effect on tumor growth in immunocompromised mice. Intriguingly, transfection of miR-30c-2* or miR-497 mimics in COLO16 cuSCC cells and the subsequent xenograft of transfected cells into mice almost completely abolished its tumor-forming potential. These results may suggest that use of miR-30c-2* or miR-497 miRNA mimics is more effective for the treatment of advanced cuSCC than alisertib alone. It should be noted that TAp63^−/− mice used in this study express normal levels of ΔNp63 (5). Thus, phenotypes observed in this study, including downregulation of miR-30c-2* or miR-497 by TAp63 knockout, are unlikely due to altered expression of ΔNp63 and its transcriptional target DGCR8, a protein that plays a crucial role in Dicer-independent miRNA biogenesis (5, 6).

Although this study provides evidence for TAp63′s suppressive activity in cuSCC, some questions remain. These include: (i) the roles of miR-30c-2* and miR-497 in the progression of other types of SCC, (ii) whether other targets of miR-30c-2* and miR-497 (FAT2, ITGA6, KIF18B, ORC1, CDK6, PKMYT1) also promote cuSCC progression, (iii) feasibility of using miR-30c-2* and miR-497 mimics as therapeutic strategies for cuSCC, and (iv) whether AURKA inhibitors could be used to treat cuSCCs. It would also be important to examine how efficiently exogenous AURKA rescues cellular phenotypes induced by miR-497 in cuSCC cells. In addition, because KIF18B and PKMYT1 are the common targets of miR-30c-2* and miR-497, these proteins may play crucial roles in cuSCC survival, proliferation, and progression. Thus, the functional roles of TAp63-regulated miRNAs and their targets in SCC progression need to be further investigated.

This study shows only a modest effect of alisertib on cuSCC suppression in mouse models (5). However, this does not mean that AURKA cannot serve as an important marker and/or therapeutic target for cuSCC and other types of SCC. Indeed, AURKA is frequently (∼70%) overexpressed in SCC and is implicated in the progression of SCC and other types of cancer including breast, liver, pancreatic, bladder, and gastrointestinal (7). Moreover, knockdown of AURKA inhibits proliferation and migration of head and neck SCC cells (8). Because alisertib is in clinical trials of multiple cancers including malignant rhabdoid tumors, lung cancer, and triple-negative breast cancer, experiments examining whether alisertib has any cooperative effects with other clinically available drugs and identifying new alisertib analogs that have increased potency of suppressing cuSCC should be performed in the future. These approaches could be used to overcome the observed modest antitumor effect of alisertib. In addition, this study does not examine whether AURKA activity is inhibited in cuSCC tumors formed in mice following alisertib treatment (5). AURKA inhibitors that can efficiently reach cuSCC tumor sites and inhibit AURKA activity in tumors should be identified.

This study also proposes the therapeutic delivery of miR-30c-2*/miR-497 as a potential novel therapeutic strategy. One major concern is how to specifically deliver miRNAs into tumors, which is a common concern in gene therapy, because delivery of these miRNAs to nontumor tissues could cause adverse effects. Thus, in addition to determining possible adverse effects caused by these miRNAs, a tumor-specific miRNA delivery system should be discovered. If cuSCC-specific delivery of these miRNAs is achieved, this could be used to treat recurrent and metastatic cuSCC. Hence, identification of tumor-specific cell surface markers of cuSCC would be one of the critical future studies to achieve this goal.

TAp63 acts not only as a tumor suppressor, but also as a metastasis suppressor. Indeed, this study shows that UVR-treated TAp63^−/− mice have increased lung metastases as compare with wild-type mice (5). This is consistent with previous studies by this group showing the metastasis-suppressive role of TAp63 and its downstream target, Dicer, in SCC (3). In addition, AURKA is shown to promote SCC metastasis (8). However, it remains unclear whether miR-30c-2* and miR-497 also play crucial roles in suppressing cuSCC metastasis, although these miRNAs are implicated in metastasis suppression of other types of cancer (9, 10). Further detailed studies are required to define the roles of these miRNAs and their downstream targets in cuSCC metastasis. In summary, this study, together with previous findings, may accelerate development of novel therapeutic strategies for advanced cuSCC by targeting the TAp63-miR-30c-2*/miR-497-AURKA axis.

No potential conflicts of interest were disclosed.

This work was supported by grants R01 CA174735 and R01 CA214916 (to T. Iwakuma) from the NIH. The authors thank Elizabeth Thoenen for her helpful comments.