Angiogenesis inhibitors (AIs) and photosensitizers (Ps) are two types of drugs used in tumor treatment [[1], [2], [3], [4]]. AIs target angiogenesis related proteins to block signaling pathways achieving the goal of “starving tumors” by inhibiting angiogenesis. Currently, several AIs, such as sorafenib, have been approved for tumor treatment. However, clinical observations have revealed that most AIs develop resistance following short-term benefits [5]. This has emerged as the primary bottleneck in the clinical application of AIs [6]. Photodynamic therapy (PDT), characterized by minimal invasiveness, low toxicity, high effectiveness, and absence of resistance with repeated applications, presents a promising approach to cancer treatment. In PDT, Ps convert oxygen (O2) into cytotoxic radicals and reactive oxygen species (ROS), capable of irreversibly oxidizing DNA, proteins, or fatty acids. However, conventional Ps suffer from an “Achilles' heel”, including strong oxygen dependence, limited targeting ability, and shallow tissue penetration depth [[7], [8], [9], [10], [11], [12]]. One of the current hot areas in PDT research focuses on overcoming these limitations to enhance its clinical utility.

Motivated by the aforementioned concerns, this study, we employed a hybridization strategy to design and synthesize novel angiogenesis inhibitors (NAIs). By leveraging the respective strengths of AIs and Ps, we aim to address their clinical application limitations. The NAIs comprise three components: the target recognition moiety (AIs), the photoactive compound (Ps) and the linker (Scheme 1). The target recognition moiety is designed to selectively bind to vascular endothelial growth factor receptor 2 (VEGFR-2), which is abundantly expressed on the surface of tumor cells and vascular endothelial cells. In the presence of light, Ps generate superoxide anion radicals (O2−•), which are among the primary and most toxic ROS [[13], [14], [15]]. Importantly, intracellular superoxide dismutase (SOD) can catalyze disproportionation reactions, converting O2−• into O2 and highly toxic hydroxyl radical (OH) [[16], [17], [18]]. O2 is recyclable in these cascade bioreactions, thus enhancing anti-hypoxia performance (Scheme 1) [16]. Furthermore, the excitation wavelength of the Ps must fall within the “therapeutic window (600–900 nm)” to ensure deeper tissue penetration. The linker serves to minimize the steric hindrance between AIs and Ps, ensuring that the presence of Ps does not interfere with the binding of AIs to target proteins.

Sorafenib (a commercially available angiogenesis inhibitor) and BD7 (an angiogenesis inhibitor previously discovered in our laboratory) were selected as target recognition molecules due to their excellent targeting ability for VEGFR-2 [19,20]. The selection of Ps is crucial for the efficacy of NAIs. Benzophenothiazine photosensitizers (BPs) have been selected for their capability to generate O2−• via the Type I mechanism, with excitation wavelengths falling within the “therapeutic window” [[21], [22], [23]]. The NAIs demonstrate targeting of both vascular endothelial cells and tumor cells, while their capacity to generate O2−• inducing apoptosis under both hypoxic and normoxic conditions. In conclusion, these findings indicate that linking AIs to Ps via linkers can effectively address the clinical limitations they encounter, offering novel avenues for the discovery of angiogenesis inhibitors. Remarkable, the BPs's ability to generate O2−• was significantly enhanced upon connection to sorafenib, albeit with a significant reduction in fluorescence intensity. To our knowledge, this is the first instance that BPs have shown an augmented ability to generate ROS following linkage with sorafenib. Through the introduction of various lengths and types of linkers, it was observed that sorafenib could serve as an enhancer, amplifying the ROS generation capacity of BPs.