HPSCC is a less prevalent yet aggressive malignancy in the head and neck region. Absence of symptoms in its early phase and a high rate of regional and systemic metastases make HPSCC the most lethal of head and neck cancers[2, 4]. A large proportion of HPSCC patients lack specific early symptoms or pathologic signs in part because of the unique anatomy of the hypopharynx, and, in addition, laryngoscopy is not routine in most countries. Consequently, early detection of HPSCC remains challenging, and it was reported that more than 75% of patients were already in stage III or IV when initially diagnosed. HPSCC patients at advanced stages have limited treatment options and their clinical outcomes are far from satisfactory, although the overall 5-year survival rate has increased from 37.5–41.3%. Elucidation of the molecular events that contribute to cancer cell proliferation and migration is thus of critical importance in achieving early diagnosis and finding appropriate therapeutic targets.
As a member of the homologous AMIGO family, AMIGO2 was first found by ordered differential display in neuronal tissues treated with neurite-promoting protein, and it was regarded as a novel cell-adhesion molecule that contributes to neuronal formation. Also known as Alivin 1, AMIGO2 was discovered by another research group that reported its promotion of the neuronal activity-dependent survival of cerebellar granule neurons. An etiologic role for AMIGO2 (also known as DEGA) in gastric adenocarcinoma was subsequently reported. Furthermore, emerging data suggested a detrimental action of AMIGO2 in multiple cancer types, including melanoma, liver, gastric, colorectal, and ovarian cancer. Based on these previous reports, we hypothesized an oncogenic role for AMIGO2 in HPSCC.
In the current study, we found AMIGO2 to be up-regulated in HNSC tumors when compared to adjacent normal tissues based on TCGA data, and high expression levels of AMIGO2 were correlated with a poor prognosis in HNSC. We further confirmed that AMIGO2 exhibited a higher mRNA abundance in the HPSCC cell line FaDu relative to the laryngeal cancer cell line HEp-2. AMIGO2 silencing by shRNAs markedly suppressed cellular proliferation and migration while promoting the apoptosis of FaDu cells in vitro, and knockdown of AMIGO2 in vivo also significantly reduced xenograft tumor growth, as expected. Similar to our results, AMIGO2 was reported to be differentially expressed in nearly half of the gastric adenocarcinoma patients, and its knockdown led to nearly complete abrogation of tumorigenicity. Additionally, AMIGO2 was identified as the most highly differentially expressed gene in the process of in-vivo selection of a metastatic ovarian cancer cell line. Collectively, our results indicated that AMIGO2 promoted tumor proliferation and migration in HPSCC.
Based on LC-MS/MS and bioinformatics analysis results, we uncovered five proteins that potentially interact with AMIGO2. Co-IP experiments confirmed that in the FaDu cellular context, AMIGO2 elicited protein-protein interactions with CAPN2. AMIGO2-silenced inhibition of cellular proliferation might thus be partially attenuated by the overexpression of CAPN2. Furthermore, CAPN2 overexpression notably counteracted the inhibition on migration by AMIGO2 silencing. These results suggested that CAPN2 is a key molecule responsible for the biologic effects of AMIGO2 in HPSCC. CAPN2 encodes a distinct catalytic subunit of calpain-2, a calcium-dependent intracellular thiol protease that is ubiquitously expressed in tissues. Investigators had previously demonstrated an oncogenic role for calpain-2 in numerous types of cancers, including colorectal, mammary, and hepatocellular cancers—as well as acute myelogenous leukemia. For example, in breast cancer calpain-2 promotes tumor cell proliferation and migration both in vitro and in vivo via regulation by Akt. Moreover, calpain-2 serves as a diagnostic marker in non-small cell lung cancer, and as a potential treatment target in colorectal cancer as well as in gefitinib‑resistant lung adenocarcinoma[25–27].
Upon investigation of the synergistic effects of AMIGO2 and CAPN2 (calpain-2) in HPSCC, we found that multiple signaling pathways were significantly enriched in KEGG analysis, including HPV infection and PI3K-Akt signaling pathway. HPV infection has been known as an aetiological risk factor for HNSCC for decades. HPV-positive HNSC shows a unique tumorigenic biology and exhibits distinct differences from HPV-negative HNSC in many aspects, including gene expression and immune profiles. While HPV positivity were relevantly less common in HPSCC, it was associated with an improved survival, indicating a potientially different clinical management strategy[30, 31]. Although HPV infection pathway was markedly enriched upon regulation of AMIGO2 and CAPN2, it is still unknown whether these genes (proteins) could facilitate or hinder HPV infection in hypopharynx, and more experiments are needed in future to answer this question. For another, PI3K-Akt signaling is a well-documented pathway and regarded as one of the most frequently dysregulated pathways in human cancers[32–35]. In brief, phosphoinositide 3-kinase (PI3K) activated by multiple stimuli phosphorylates phosphatidylinositol-4,5-bisphosphate (PIP2) to generate phosphatidylinositol-3,4,5-trisphosphate (PIP3). PIP3 then recruits phosphoinositide-dependent protein kinase-1 (PDK1) to the plasma membrane, resulting in the activation of Akt by its phosphorylation by PDK1[36, 37]. Notably, the C-terminal region of AMIGO2 can directly bind to PDK1, regulate the membrane localization of PDK1, and subsequently facilitate Akt activation. Furthermore, CAPN2 has been reported to mediate the degradation of phosphatase and tensin homolog (PTEN), a protein that dephosphorylates PIP3 to generate PIP2. And degradation of PTEN can then cause accumulation of PIP3, resulting in prolonged activation of the PI3K-Akt pathway[36, 37]. However, in the context of HPSCC, the participation of AMIGO2 or CAPN2 in PI3K-Akt signaling pathway still remains unknown and awaits further experiments to validate.
Taken together, our results showed that AMIGO2 interacts with CAPN2 and promotes tumor proliferation and migration in HPSCC. Our work thus provided a potential novel molecular mechanism underlying the oncogenesis of HPSCC and supports AMIGO2 as a potential marker for the diagnosis and therapeutic intervention in HPSCC.