Aptamer/doxorubicin-conjugated nanoparticles target membranous CEMIP2 in colorectal cancer

The objectives were to identify the functional domains of a potential oncoprotein, cell migration inducing hyaluronidase 2 (CEMIP2), evaluate its expression levels and roles in colorectal cancer (CRC), and develop an aptamer-based nanoparticle for targeted therapy. Data mining on TCGA identified that CEMIP2 might play oncogenic roles in CRC. In a local cohort, CEMIP2 mRNA levels significantly stepwise increase in CRC patients with higher stages, and high CEMIP2 confers worse disease-free survival. In addition, CEMIP2 mRNA levels significantly correlated to hyaluronan levels in sera from CRC patients. Deletion mapping identified that CEMIP2 containing G8 and PANDER-like domains preserved hyaluronidase activity and oncogenic roles, including cell proliferation, anchorage-independent cell growth, cell migration and invasion, and human umbilical vein endothelial cell (HUVEC) tube formation in CRC-derived cells. A customized monoclonal mouse anti-human CEMIP2 antibody probing the PANDER-like domain (anti-289307) counteracted CEMIP2-mediated carcinogenesis in vitro. Cell-SELEX pinpointed an aptamer, aptCEMIP2(101), specifically interacted with the full-length CEMIP2, potentially involving its 3D structure. Treatments with aptCEMIP2(101) significantly reduced CEMIP2-mediated tumorigenesis in vitro. Mesoporous silica nanoparticles (MSN) carrying atpCEMIP2(101) and Dox were fabricated. Dox@MSN, MSN-aptCEMIP2(101), and Dox@MSN-aptCEMIP2(101) significantly suppressed tumorigenesis in vitro compared to the Mock, while Dox@MSN-aptCEMIP2(101) showed substantially higher effects compared to Dox@MSN and MSN-aptCEMIP2(101) in CRC-derived cells. Our study identified a novel oncogene and developed an effective aptamer-based targeted therapeutic strategy.

Human TMEM2 is not a catalytic hyaluronidase, but a regulator of hyaluronan metabolism via HYBID (KIAA1199/CEMIP) and HAS2 expression

Cutaneous hyaluronan (HA) is depolymerized to intermediate sizes in the extracellular matrix, and further fragmented in the regional lymph nodes. Previously, we showed that the HA-binding protein involved in HA depolymerization (HYBID), also known as KIAA1199/CEMIP, is responsible for the first step of HA depolymerization. Recently, mouse transmembrane 2 (mTMEM2) with high structural similarity to HYBID was proposed to be a membrane-bound hyaluronidase. However, we showed that knockdown of human TMEM2 (hTMEM2) conversely promoted HA depolymerization in normal human dermal fibroblasts (NHDFs). Therefore, we examined the HA-degrading activity and function of hTMEM2 using HEK293T cells. We found that human HYBID and mTMEM2, but not hTMEM2, degraded extracellular HA, indicating that hTMEM2 does not function as a catalytic hyaluronidase. Analysis of the HA-degrading activity of chimeric TMEM2 in HEK293T cells suggested the importance of the mouse GG domain. Therefore, we focused on the amino acid residues that are conserved in active mouse and human HYBID and mTMEM2, but are substituted in hTMEM2. The HA-degrading activity of mTMEM2 was abolished when its His248 and Ala303 were simultaneously replaced by the corresponding residues of inactive hTMEM2 (Asn248 and Phe303). In NHDFs, enhancement of hTMEM2 expression by proinflammatory cytokines decreased HYBID expression and increased hyaluronan synthase 2 (HAS2)-dependent HA production. The effects of proinflammatory cytokines were abrogated by hTMEM2 knockdown. Moreover, a decreased HYBID expression by interleukin-1β and transforming growth factor-β was canceled by hTMEM2 knockdown. In conclusion, these results indicate that hTMEM2 is not a catalytic hyaluronidase, but a regulator of HA metabolism.