Nitrogen mustards (NMs) are an important class of chemotherapeutic drugs and have been widely employed for the treatment of various cancers. However, due to the high reactivity of nitrogen mustard, most NMs react with proteins and phospholipids within the cell membrane. Therefore, only a very small fraction of NMs can reach the reach nucleus, alkylating and cross-linking DNA. To efficiently penetrate the cell membrane barrier, the hybridization of NMs with a membranolytic agent may be an effective strategy. Herein, the chlorambucil (CLB, a kind of NM) hybrids were first designed by conjugation with membranolytic peptide LTX-315. However, although LTX-315 could help large amounts of CLB penetrate the cytomembrane and enter the cytoplasm, CLB still did not readily reach the nucleus. Our previous work demonstrated that the hybrid peptide NTP-385 obtained by covalent conjugation of rhodamine B with LTX-315 could accumulate in the nucleus. Hence, the NTP-385-CLB conjugate, named FXY-3, was then designed and systematically evaluated both in vitro and in vivo. FXY-3 displayed prominent localization in the cancer cell nucleus and induced severe DNA double-strand breaks (DSBs) to trigger cell apoptosis. Especially, compared with CLB and LTX-315, FXY-3 exhibited significantly increased in vitro cytotoxicity against a panel of cancer cell lines. Moreover, FXY-3 showed superior in vivo anticancer efficiency in the mouse cancer model. Collectively, this study established an effective strategy to increase the anticancer activity and the nuclear accumulation of NMs, which will provide a valuable reference for future nucleus-targeting modification of nitrogen mustards.