The first line treatment of chronic myeloid leukemia patients is the TKIs, which are applied in clinical treatment with satisfactory efficacy and minor side effects[35]. Nonetheless, CML cells develop TKIs resistance because of the mutations of Bcr-Abl kinase region and pathogenicity of leukemia stem cells. Furthermore, TKIs have inconspicuous effect on some patients who transitioned from the chronic phase to the acute phase[36]. Therefore, identification of novel drug targets needed for CML treatment.
Many progresses have been made in identifying downstream signaling molecules activated by Bcr-Abl during CML pathogenesis[4, 6, 37]. Yet not much is known about the molecular mechanisms sustaining Bcr-Abl localization in the cytoplasm, especially those critical for the development of CML. The localization of Bcr-Abl is an important determinator the development of CML disease, and the CC domain of Bcr-Abl mainly affects its location in the cytoplasm[19, 20]. Protein interactions in leukemia cells can affect protein subcellular localization, hence we speculated that certain proteins in the cytoplasm may bind to the CC domain to retain Bcr-Abl in the cytoplasm[38]. In this study, we screened the target protein HSP90AB1 which is over-expressed in a large variety of cancer cells and belongs to highly conserved ATP-dependent molecular chaperone. HSP90AB1 interacts with transcription factors, cellular kinases, and various molecules to participate in lots of pathophysiological processes of cells[39]. However, the role of HSP90AB1 in the pathogenesis of leukemia is rarely reported. In this report, we have found that HSP90AB1 interacts with the CC domain of Bcr-Abl in the cytoplasm to prevent its degradation. Finding the specific binding site can help us target the Bcr/Abl-HSP90AB1 complex for dissociation. So, we modeled a three-dimensional structure diagram of the Bcr/Abl-HSP90AB1 complex. The direct interaction between the CC domain and the NTD of HSP90AB1 was determined by co-immunoprecipitation. At present, the development and application of HSP90 inhibitors have become a hotspot in tumor therapy, and the targets of inhibitors are also various[25]. The exploration of specific site can make the HSP90 inhibitor more accurate in the selection of therapeutic targets for CML and other Bcr-Abl-addicted disease.
The specific localization of Bcr-Abl in the cytoplasm can cause malignant transformation of blood cells, whereas Bcr-Abl induces the apoptosis of CML cells after transporting into the nucleus[15]. In this study, we found the decisive cause for the retention of Bcr-Abl in cytoplasm is the formation of Bcr/Abl-HSP90AB1 complex. For the previously identified binding site, we selected the 17AAG inhibitor that can promote HSP90AB1 dissociation with chaperone protein by antagonizing the NTD ATP function[40]. Interestingly, the Bcr-Abl is transported into the nucleus from the cytoplasm when dissociated with HSP90AB1 under the treatment of 17AAG. The nuclear Bcr-Abl down-regulates the cytoplasmic proliferation signaling activated by the tyrosine kinase of cytoplasmic Bcr-Abl. At the same time, the nuclear Bcr-Abl induces p73 and its downstream targets through c-Abl kinase (Fig. 7). The above experiment results illustrate that the Bcr-Abl can be directed into the nucleus, and Bcr-Abl located in nucleus can induce apoptosis and inhibit the proliferation of CML cells by DNA damage and inhibition of the tyrosine kinase activity. Our research in the early stage conducted a series of studies on the localization of Bcr-Abl, which confirmed its significance to CML. The exploration of the mechanism of specific localization of Bcr-Abl lays the foundation for further study on the pathogenic effect of Bcr-Abl in CML. Furthermore, these data describe a previously neglected strategies, promote the apoptosis of CML cells by inducing the transport of Bcr-Abl into the nucleus.
Targeting protein subcellular localization is considered challenging, because the protein may be transported out of the nucleus after being induced into the nucleus[41]. The Bcr-Abl and c-Abl have similar structural sites, both containing three nuclear localization signals and a nuclear output signal, so it can shuttle between the cytoplasm and nucleus, but is mainly located in the cytoplasm. Considering the nuclear shuttle function of Bcr-Abl, we used the protein nuclear export inhibitor Leptomycin B (LMB)[17]. Our immunofluorescence results displayed that using LMB alone does not affect the localization of Bcr-Abl, but can transport Bcr-Abl into the nucleus in combination with 17AAG. Based on the above results, we then clearly indicated that targeting chronic myeloid leukemia cells, the combination of the two inhibitors can enhance the killing ability. Malignant leukemia cells are particularly sensitive to HSP90 inhibition, leading to the steady development of clinical HSP90 inhibitors[42, 43]. Recently, there are more than thirty positive clinical trials involving the use of HSP90 inhibitors[44, 45]. Conceivably, HSP90 inhibitors will be used as potential alternative therapies to benefit CML patients with Imatinib resistance. However, most current studies on HSP90 inhibitors for the treatment of leukemia have focused on the functional structure of HSP90 and its effect on the phosphorylation or tyrosine kinase activity of the chaperone protein[25, 29, 46]. Following the previous results, our study focuses on the effects of HSP90AB1 on the localization and function of Bcr-Abl, and our findings provide an innovative strategy to develop new therapy of CML. The identification of interaction site in this study can help the development of new HSP90 inhibitors to find the effectively targets[46].