Chemoresistance is a daunting obstacle to the effective treatment of breast cancer patients receiving chemotherapy. Although the mechanism of chemotherapy drug resistance has been explored broadly, the precise mechanism at the proteome level remains unclear. Especially, comparative studies between widely used anticancer drugs in breast cancer are very limited. In this study, we employed proteomics and bioinformatics approaches on chemoresistant breast cancer cell lines to understand the underlying resistance mechanisms that resulted from doxorubicin (DR), paclitaxel (PR), and tamoxifen (TAR). In total, 10,385 proteins were identified and quantified from three TMT 6-plex and one TMT 10-plex experiments. Bioinformatics analysis showed that Notch signaling, immune response, and protein re-localization processes were uniquely associated with DR, PR, and TAR resistance, respectively. In addition, proteomic signatures related to drug resistance were identified as potential targets of many FDA-approved drugs. Furthermore, we identified potential prognostic proteins with significant effects on overall survival. Representatively, PLXNB2 expression was associated with a highly significant increase in risk, and downregulation of ACOX3 was correlated with a worse overall survival rate. Consequently, our study provides new insights into the proteomic aspects of the distinct mechanisms underlying chemoresistance in breast cancer.

Osteoclasts are derived from hematopoietic stem cells. Monocyte preosteoclasts obtain resorbing activity via cell–cell fusion to generate multinucleated cells. However, the mechanisms and molecules involved in the fusion process are poorly understood. In this study, we performed RNA sequencing with single nucleated cells (SNCs) and multinucleated cells (MNCs) to identify the fusion-specific genes. The SNCs and MNCs were isolated under the same conditions during osteoclastogenesis with the receptor activator of nuclear factor-κB ligand (RANKL) administration. Based on this analysis, the expression of seven genes was found to be significantly increased in MNCs but decreased in SNCs, compared to that in bone marrow-derived macrophages (BMMs). We then generated knockout macrophage cell lines using a CRISPR-Cas9 genome-editing tool to examine their function during osteoclastogenesis. Calcrl-, Marco-, or Ube3a-deficient cells could not develop multinucleated giant osteoclasts upon RANKL stimulation. However, Tmem26-deficient cells fused more efficiently than control cells. Our findings demonstrate that Calcrl, Marco, and Ube3a are novel determinants of osteoclastogenesis, especially with respect to cell fusion, and highlight potential targets for osteoporosis therapy.