Acute myeloid leukemia (AML) is a hematopoietic malignancy characterized by strong invasiveness, poor prognosis, high recurrence rate and short overall survival, which is the most common type of adult leukemia and poses significant clinical challenges [1,2]. Mutations in the Fms-like tyrosine kinase 3 (FLT3) gene are recognized as the most common genetic aberrations in AML patients, with an incidence of about 30% [3]. Particularly, approximately 25% of AML patients harbor the internal tandem duplication (FLT3-ITD) mutations that frequently result in inferior prognosis and relapse; while 5% of AML patients have the tyrosine kinase domain (FLT3-TKD) mutations [4,5]. Numerous studies have shown that FLT3 mutations are closely related to the occurrence, development and prognosis of AML. Hence, targeting FLT3 mutations represents one of the current major therapeutic strategies for AML [[6], [7], [8]].

To date, plentiful small-molecule FLT3 inhibitors are disclosed, some of them have advanced to clinical stage and showed positive therapeutic efficacy for AML [8,9], such as midostaurin (1) [10], gilteritinib (2) [11], FN-1501 (3) [12], crenolanib (4) [13], sorafenib (5) [14] and quizartinib (6) [15] (Fig. 1). Additionally, multiple FLT3 inhibitors with novel structures are reported recently, such as compound 7 [16], compound 8 [17] and compound 9 [18] (Fig. 1), etc., which exhibit significant preclinical anti-AML activity. However, resistance mutations in FLT3 TKD region frequently occur in clinical application of these inhibitors, which lead to transient response period and limited clinical efficacy [[19], [20], [21]]. Moreover, the upregulation of FLT3 ligand induces hyperactivation of FLT3 in AML models, resulting in resistance to FLT3 inhibitors [22,23]. Developing alternative strategies to circumvent the resistance of AML to FLT3 inhibitors refers to a new direction for more potent anti-AML drugs.

Proteolysis targeting chimeras (PROTACs) have emerged as a novel therapeutic strategy in drug discovery, which induce the degradation of pathogenic target proteins through hijacking the ubiquitin-proteasome system in living cells [24,25]. PROTACs are heterobifunctional molecules consisting of three moieties: a ligand that binds a protein of interest, another ligand that recruits an E3 ubiquitin ligase and a linker that concatenates the two ligands [24]. Recently, a variety of PROTAC molecules have been developed and shown significant preclinical or clinical effects, particularly targeting cancer-related kinases [[26], [27], [28]], such as bruton's tyrosine kinase (BTK) [29], cyclin-dependent kinase (CDK) [30], etc. Given the apparently different mechanism of PROTACs from that of inhibitors, i.e. eliminating disease-relevant proteins, it is believed that PROTACs can avoid the occurrence of drug resistance induced by the original target proteins [31,32]. Since the first FLT3 PROTAC molecule was reported [33], there has been sustaining research interests in PROTACs targeting FLT3 [[34], [35], [36], [37]].

In previous studies, we have made great efforts to develop FLT3 inhibitors [12,[38], [39], [40], [41], [42], [43]]. In this work, we designed and synthesized a series of FLT3-targeting PROTACs based on a potent FLT3 inhibitor [42] previously reported by us, through a structure-based drug design strategy. A new FLT3-targeting PROTAC molecule (compound 35) with strong and selective antiproliferative activities against AML cells harboring FLT3 mutations was discovered. Its binding mode of forming the FLT3—PROTAC—cereblon (CRBN) ternary complex was simulated via the comparative molecular dynamics study, which would provide helpful insights into further rational discovery of FLT3-targeting PROTACs in the situation of lacking FLT3 PROTAC ternary complex crystal. To the best of our knowledge, this is the first time to report the dynamic interactions of FLT3 PROTAC molecules.