The focus of this study was to measure the impact of MAL3-101 on in vitro and in vivo growth of MM cells and to predict the clinical effectiveness of this compound against MM. Our results provide evidence that MAL3-101 is cytotoxic to MM cells, and that synergistic effects are observed when this compound is used in combination with Hsp90 or proteasome inhibitors.
In MM, the protein-folding machinery is overloaded due to increased IG synthesis and secretion, and a significant population of the IGs may misfold; in fact, some MM cells produce unassembled IG single chains. The concentration of these proteins in the ER may be decreased by being retrotranslocated back to the cytoplasm, where they are degraded by the proteasome, a process referred to as ER-associated degradation (ERAD) [48
]. Since IGs fold in the ER and traffic through the secretory pathway, proteasome inhibition results in an increased load of misfolded proteins in the ER, thus triggering apoptosis. Because Hsp70 homologs in the ER and in the cytoplasm play critical roles in ERAD [28
], it may not be surprising that we observed synergistic cytotoxic effects on MM cell survival when the proteasome and Hsp70 were simultaneously inhibited. We also observed synergistic cytotoxicity in MM cells by a combined inhibition of Hsp70 and Hsp90, further supporting the notion that MM cells are susceptible to treatments that compromise protein quality control.
The most MAL3-101-responsive cell line, NCI-H929, is a high secretor of monoclonal IG [50
] and is consistent with other published data that IG load correlates with sensitivity to inhibitors of the quality control machinery [17
]. Even though the lack of UPR induction in MM cells has been previously reported [12
], we envisioned that this response might be triggered when both Hsp70 and the proteasome were inhibited since ERAD substrates should accumulate. The lack of UPR induction by MAL3-101, Hsp90, and proteasome inhibitors at the low concentrations that induced synergistic apoptosis may be multifactorial: While NCI-H929 cells synthesize and secrete large amounts of IGs, they may be unable to induce a sustained UPR. Alternatively, apoptosis may occur by pathways independent of UPR activation, including autophagy [45
] or aggresome disposal [14
]; therefore, it will be important to better define at the molecular level the mechanisms that regulate cell stress responses [51
]. Further work is necessary to establish whether MAL3-101 has effects on pathways that are upregulated in MM cells [14
]. Nevertheless, Davenport and colleagues [15
] showed that proteasome inhibition in MM cell lines by bortezomib does not necessarily result in the production of XBP1s, but ER stress and activation of the intrinsic apoptotic pathway are evident. Moreover, 2
h after exposure to 17-AAG, there was a rapid initiation of the UPR, which leads to ER stress and activation of intrinsic apoptotic pathways. More recently, the same group discovered that the 17AAG-induced UPR was synergistically increased by Hsp72 inhibition in MM cell lines [21
]. By analogy, we found that the level of XBP1s increased after exposure to MAL3101 after 48
h. However, at effective antimyeloma concentrations, we did not observe XBP1s when MAL3-101 and 17-AAG were coadministered, unless much higher concentrations and a prolonged incubation were used. We suggest that the different effects observed may result from the nature of the compounds that were employed to inhibit Hsp70, and/or that the activities of unique Hsp70 family members may be differentially affected by these chemicals.
A critical result of our studies is the demonstration of the synergistic effects of Hsp70 and proteasome inhibition on the MM microenvironment. We and others have observed IG gene rearrangement in MM EPCs [40
]; therefore, protein misfolding and the ensuing ER stress response may determine the susceptibility of tumor endothelial cells, in at least a population of patients, to Hsp70 and proteasome inhibition. This possibility is supported by a lack of sensitivity to MAL3-101 in normal bone marrow, lymphocytes, and endothelial cells, most of which do not produce IG. As a result, the clinical application of MAL3-101 in MM may not only help overcome drug resistance by potentiating the effects of other protein quality control inhibitors, but it may also limit MM growth via inhibition of tumor angiogenesis.
The antimyeloma effects of MAL3-101 were apparent at concentrations similar to those used with 15-deoxyspergualin, a Hsp70 modulator with a KD for Hsp70 of ~5μ
M. 15-Deoxyspergualin has proven successful in clinical trials [53
]; however, this compound is a nonspecific chaperone inhibitor and may exhibit other off-pathway effects [54
]. Our results are also in agreement with recent data showing strong expression of Hsp70 expression in MM cell lines at baseline and significant upregulation after extracellular matrix (ECM) adhesion [26
]. In the same report, it was also demonstrated that inhibition of Hsp70 gene expression induced a reduction in tumor cell adhesion to ECM followed by an increase in apoptosis in MM cells.
In summary, we suggest that MAL3-101, an Hsp70 modulator that specifically binds to Hsp70 and affects its interaction with cochaperones [1
], may provide a viable means to enhance the antimyeloma effects of both proteasome and Hsp90 inhibitors. Coadministration strategies may reduce drug resistance in MM treatments and allow those patients who currently cannot tolerate bortezomib [13
] to benefit from the drug because lower concentrations can be used in conjunction with an MAL3-101-like molecule. In addition to direct antitumor properties, MAL3-101 may affect the microenvironment by targeting the microvasculature. Our experiments using a xenograft plasmacytoma model show that tumor cell growth in vivo
is also delayed and reduced by MAL3-101, and that this inhibitory effect is synergistic for proteasome inhibition which was observed within the first week after combination treatment was initiated. In this model, because we started treatment with compounds within 24
h of tumor initiation, the findings may possibly be translated clinically to a restricted effect at earlier stages of the disease prior to extensive skeletal involvement. It is also possible that combinations containing higher doses of MAL3-101 with proteasome inhibition may be required to control larger tumors and MM at its later stages. The need to elucidate the pharmacokinetic properties of MAL3-101 is underscored by these results for optimization of dose and frequency schedule of MAL3-101 exposure. Particularly important will be the measurement of plasma levels as well as rate and extent of clearance of the compound when determining effects versus toxicity. In the latter regard, we do know that concentrations up to 160
mg/kg i.p. have been tolerated without toxicity (data not shown). The results presented here strongly suggest that more detailed studies of MAL3-101 pharmacokinetics are warranted.