Letters to the Editor
1804 2 Grand FH, Hidalgo-Curtis CE, Ernst T, Zoi K, Zoi C, McGuire C et al. Frequent CBL mutations associated with 11q acquired uniparental disomy in myeloproliferative neoplasms. Blood 2009; 113: 6182–6192. 3 Makishima H, Cazzolli H, Szpurka H, Dunbar A, Tiu R, Huh J et al. Mutations of e3 ubiquitin ligase cbl family members constitute a novel common pathogenic lesion in myeloid malignancies. J Clin Oncol 2009; 27: 6109–6116. 4 Delhommeau F, Dupont S, Della Valle V, James C, Trannoy S, Masse A et al. Mutation in TET2 in myeloid cancers. N Engl J Med 2009; 360: 2289–2301. 5 Szpurka H, Gondek LP, Mohan SR, Hsi ED, Theil KS, Maciejewski JP. UPD1p indicates the presence of MPL W515L mutation in
RARS-T, a mechanism analogous to UPD9p and JAK2 V617F mutation. Leukemia 2009; 23: 610–614. 6 Jasek M, Gondek LP, Bejanyan N, Tiu R, Huh J, Theil KS et al. TP53 mutations in myeloid malignancies are either homozygous or hemizygous due to copy number-neutral loss of heterozygosity or deletion of 17p. Leukemia 2010; 24: 216–219. 7 O’Keefe C, McDevitt MA, Maciejewski JP. Copy neutral loss of heterozygosity: a novel chromosomal lesion in myeloid malignancies. Blood 2010; 115: 2731–2739. 8 Morin RD, Johnson NA, Severson TM, Mungall AJ, An J, Goya R et al. Somatic mutations altering EZH2 (Tyr641) in follicular and diffuse large B-cell lymphomas of germinal-center origin. Nat Genet 2010; 42: 181–185.
Targeting heat shock protein 72 enhances Hsp90 inhibitor-induced apoptosis in myeloma
Leukemia (2010) 24, 1804–1807; doi:10.1038/leu.2010.168; Published online 12 August 2010
Recent years have seen the development of a number of new small molecule therapies for the treatment of myeloma and as a result of their introduction there have been significant improvements in patient survival. However, patients continue to relapse, and the current challenge is to develop therapies able to effectively treat these refractory cases. An emerging feature of the use of single, highly specific molecules is that they can be clinically ineffective because of the redundancy in the pathway targeted and compensatory changes induced by drug exposure. Consequently therapeutic efficacy is reliant on the development of effective combinations targeting synergistic pathways. A simplified view of the function and biology of plasma cells is that they exist to produce immunoglobulin and this can provide a differential marker, which can be targeted. The requirement for immunoglobulin production means that the myeloma plasma cell is reliant upon the unfolded protein response (UPR) and molecular chaperones for survival.1 Heat shock protein (Hsp) 90 is a molecular chaperone that has an important role in the pathophysiology of myeloma and represents a relevant therapeutic target.2,3 It fulfils two essential roles in plasma cells: firstly it is a major pro-survival chaperone, which associates with and stabilizes an extensive range of signal transducers important to plasma cell survival and proliferation (insulin-like growth factor1, vascular endothelial growth factor, RAF and p42/44 mitogenactivated protein kinase), cell cycle regulation (AKT and nuclear
factor-kB), migration (phosphoinositide 3-kinase) and antiapoptotic (nuclear factor-kB) pathways. Secondly, it has a central role in the UPR, which is responsible for ensuring the correct folding of nascent immunoglobulin, as well as providing survival signals for clonal expansion. Although targeting these pathways using inhibitors of Hsp90 has proved successful in myeloma preclinical models, preliminary results in patients suggest that combinations of agents will be required for clinical use. To provide a rationale for the development of effective mechanism-based clinical combinations, we have tested the hypothesis that the induction of the Hsp70 family member, Hsp72, after treatment with an Hsp90 inhibitor, protects myeloma cells from apoptosis by enabling cells to increase their ability to fold damaged proteins. Hence inhibition of the function of Hsp72 may enhance the apoptosis induced by Hsp90 inhibition. We have therefore investigated the role of Hsp72 in myeloma using two chemical compounds, which have been shown to inhibit the protective effects of Hsp70, triptolide and KNK437. To confirm the impact of Hsp72 inhibition, we have verified these findings using small hairpin RNA (shRNA) specifically directed against Hsp72. Our results show that after treatment with the Hsp90 inhibitor 17-AAG for 24 h, myeloma cells increase the protein level of Hsp72. Combination treatment with KNK437 or triptolide blocks this increase (Figure 1a) and results in a decrease in proliferation compared with treatment with 17-AAG alone. Using a series of 16 concentrations of these agents applied in combination for a period of 4 days, we demonstrate that the concomitant application of 17-AAG with either of these
Figure 1 (a) Myeloma cells were treated with either: 17-AAG (5 mM), triptolide (T, 100 nM) and KNK437 (K, 100 nM) alone or in combination for 24 h, and Hsp72 expression analyzed by western blot. (b) Myeloma cells were treated with a 16-point concentration range of each agent. Agents were applied for 4 days, in three regimens: concomitantly on day 1, agent 1 applied 24 h before agent 2 and the inverse, as indicated. The combination index (CI) methods of Chou and Talalay were used to assess for synergistic effects. Graphs show the CI plotted against the survival fraction. Values o1 indicate synergism, values between 1 and 1.2 indicate additive effects, values 41.2 indicate antagonism. (c) Myeloma cells were left untreated (C), treated with triptolide alone (T) or in combination with 17-AAG (17) for 24 h and the % of cell death analyzed: Upper panel: trypan blue, middle and lower panels: annexin V/PI results for H929 and U266 cell lines. (d) Left panel: reverse transcriptase-PCR (RT-PCR) analysis of CHOP expression after treatment with 17-AAG or triptolide or a combination of both agents. Right panel: cleavage of XBP1 was analyzed by RT-PCR using LUX primers. The ratio of spliced to unspliced is shown. (e) Upper panel: western blot analysis of cell death pathways after treatment with drugs alone or in combination for 24 h. Lower panel: analysis of the dependence of drug-induced cell death on caspasedependent cell death pathways. Cells were pretreated with Z-VAD-FMK (Z) (50 mM) followed by the drugs alone or in combination 24 h. Samples were analyzed by annexin V/PI staining and the % of apoptotic cells displayed. Calculated P-values showed that there was a significant difference between T and T þ Z (P ¼ 0.019) and between 17 þ T and 17 þ T þ Z (P ¼ 0.034). n indicates statistically significant P-values o0.05. (f) Upper panel: myeloma cells transfected with P, C1 or C4 shRNA were treated with 17-AAG and cell proliferation measured by WST-1 assay over a 4-day period. Lower panel: western blot analysis of Hsp72 expression after transduction with Hsp72 shRNAs: P ¼ control, scrambled shRNA plasmid transfected cells. C1 and C4 ¼ transfected cells expressing two different Hsp72 shRNAs. All experiments were performed in triplicate and, unless otherwise stated representative data using H929 cells is shown throughout. Leukemia
Letters to the Editor
1805
a
d
Hsp72
C
17
17-AAG
K 17 +K
2
0
T 17 +T
Combination Index (CI)
6
0 -1 -2
24
C
2
4
6
24
C
2
4
6
24
C
2
4
6
2
0
2
4
6
1 0 -1 -2
24
2.5 Antagonistic
1.5 1 0.5
17-AAG + Triptolide
Additive 0
0.2
0.4
0.6
0.8
1
12 10 8 6 4 2 0
2
Ratio
2
0
1 0 -1 -2
0
Synergistic
2
4
6
24
-0.5 17-AAG
e
2.5
0
2
4
6
Triptolide 24
0
2
4
17-AAG + Triptolide
6 24
0
2
4
6
24
2 1.5
Additive
1 0.5
0
0.2
0.4
0.6
0.8
Caspase 8
1
Synergistic
0 -0.5
Caspase 3
17-AAG and Triptolide concomitantly 17-AAG 24 h then Triptolide
Actin
Triptolide 24 h then 17-AAG
Time (h)
H929 U266
6
24
100 90 80 70 60 50 40 30 20 10 0
*
*
Time (h) Treatment
100%
Early apoptosis
20
Early apoptosis
80%
Viable
40%
5
10
3
1
17-AAG (uM)
Late apoptosis
60%
0.7
0 0.75
100%
0.65
Time (h)
40
0.3
24
60
0.45
6
80
0
0
H929 P H929 C1 H929 C4
100
0.1
0%
120
0.25
20%
f
0.08
Viable
0.05
Late apoptosis
60% 40%
0.05
80%
% Cell Proliferation
% Cell Death
24
100 90 80 70 60 50 40 30 20 10 0
0 H929 % Total cells
6
% Cell Death
0
c
Time (h) Caspase 9
Antagonistic
Survival Fraction
U266 % Total cells
24
Time (h)
Time (h)
Survival Fraction
U266
12 10 8 6 4 2 0
4
1
15
H929
17
Fold Change
b
Combination Inddex (CI)
C
Triptolide alone
2
Ratio
Actin
Fold Change
Hsp72 U266
XBP1 Splicing
12 10 8 6 4 2 0
Ratio
Actin
Fold Change
CHOP
H929
P
C1
C4 Hsp72
20% 0%
Actin
Treatment
48 72
48 72 96 48
72 96 Time (h)
Leukemia
Letters to the Editor
1806 compounds induces either additive or synergistic, antiproliferative effects (Figure 1b). These effects are not seen in normal, human mononuclear cells (data not shown). We go on to show that both KNK437 and triptolide act synergistically with 17-AAG to increase the induction of apoptosis, as assessed by trypan blue exclusion and annexin V/PI staining (Figure 1c). Cell cycle analysis demonstrated that single-agent treatment with 17-AAG induced a G2 cell cycle arrest. However, the combination of triptolide with 17-AAG overcomes this arrest creating an increase in the sub-G1 cell population, indicative of increased cell death (data not shown). To determine the mechanism underlying the increase in apoptosis and decrease in cell proliferation induced by these inhibitors, we examined the cellular stress/UPR pathway and the caspase-dependent apoptotic pathways. Previously we have shown that Hsp90 inhibition in myeloma cells results in the induction of endoplasmic reticulum stress and UPR activation, leading to cell death via activation of the ER resident transmembrane protein kinase (PERK) branch of the UPR and the delivery of a proapoptotic signal via CCAAT/enhancer binding protein homologues protein (CHOP). In this study, we show that treatment with 17-AAG or triptolide alone induced upregulation of CHOP mRNA levels over a 24-h period (Figure 1d). Concomitant treatment with both agents resulted in a sustained increase in CHOP mRNA expression, above those seen with either agent alone, suggesting that this is one mechanism by which the combination enhances apoptosis. Activation of the Inositol requiring enzyme 1 (IRE1) branch of the UPR results in the splicing of X-box binding protein 1 (XBP1) to its active form, spliced XBP1 (XBP1s), which delivers a prosurvival signal.4 We show that 17-AAG induced rapid, transient cleavage of XBP1,whereas triptolide induced XBP1 cleavage that was sustained for up to 24 h (Figure 1d). Cleavage of XBP1 induced by the drug combination was readily detected, sustained and stronger than that induced by single agent therapy. However, by 24 h XBP1s was undetectable and a decrease in the total level of XBP1 was noted. These results suggest that by 24 h, the combination of agents has tipped the survival/apoptosis balance of UPR signaling towards cell death with an increase in the levels of pro-apoptotic CHOP and a decrease in the level of prosurvival XBP1s. 17-AAG activates both the intrinsic (caspases 9 and 3) and extrinsic caspase (caspases 8 and 3) pathways within 4 h of treatment (Figure 1e). Triptolide alone induces cleavage of caspases 9, 8 and 3 but only after 24 h of treatment. As expected, the combination of agents induced cleavage of both sets of caspases at 4 h which is sustained. Importantly, the combination of 17-AAG and triptolide induced a more significant cleavage of caspases 9 and 3 than of caspase 8, suggesting that the combination potentiates cell death via sustained and enhanced activation of predominantly the intrinsic caspase-dependent cell death pathway. The validity of this observation was confirmed using the caspase inhibitor Z-VAD-FMK, which showed that only a minority of cell death occurs via a caspase-dependent mechanism in response to 17-AAG treatment alone, whereas the majority of cell death in response to triptolide alone or the combination occurs via a caspase-dependent mechanism (Figure 1e). As triptolide is primarily an inhibitor of Heats Shock Factor 1 (HSF1)-mediated heat shock response and not a specific inhibitor of Hsp72, we determined the specific effects of Hsp72 knockdown using shRNA. Using a transient expression system a number of shRNAs were found to successfully knockdown expression of Hsp72 in H929 cells (Figure 1f). To verify whether knockdown of Hsp72 increased the efficacy of Leukemia
Hsp90 inhibitors in myeloma, we tested the effect of 17-AAG on the proliferation of cells transfected with Hsp72 shRNAs compared with those transfected with a scrambled shRNA. We demonstrate that Hsp72 knockdown induced a considerable increase, of up to 40%, in the susceptibility of myeloma cells to 17-AAG-induced apoptosis (Figure 1f). In this work, we demonstrate that the combination of Hsp90 inhibitors with inhibitors of Hsp72 induction, act synergistically to kill myeloma cells. This observation is compatible with the current understanding of the function of Hsp72, which interacts with members of the intrinsic apoptotic pathway, such as Bax, to prevent release of pro-apoptotic molecules from the mitochondria and with Apaf-1, blocking the recruitment of procaspase 9 to the apoptosome. In addition, it also associates with and inhibits precursor forms of caspases 3 and 7.5 Thus inhibition of Hsp72 results in an increase in caspase-dependent cell death. In addition, we show that combining inhibitors of Hsp90 with inhibitors of Hsp72 induction enhances cell death through modulation of the endoplasmic reticulum stress pathway and the UPR. We demonstrate increased expression of CHOP mRNA, suggesting that the combination efficiently induces cell death via the PERK branch of the UPR, which is initiated through the effects of Bax on mitochondria and the subsequent initiation of the mitochondrial-dependent cell death signaling cascades. When this pathway is activated by 17-AAG alone, it is normally interrupted downstream of CHOP by the Hsp70-DnaJ chaperone complex which inhibits Bax translocation. However, this inhibitory effect is overcome by the combined use of Hsp inhibitors and contributes to their synergistic effects. Further evidence of cell death signaling via the UPR comes from the demonstration that the drug combination induces earlier and stronger XBP1 cleavage than either drug alone. However, this early pro-survival effect is not maintained and subsequent downregulation of XBP1 promotes increased cell death. We confirm that the effects on cell death seen with the chemical inhibitors of Hsp72 induction are a direct effect of Hsp72 silencing using shRNA. shRNA gene silencing of Hsp72 increased the susceptibility of myeloma cells to 17-AAGinduced cell death in a manner similar to that seen with the chemical inhibitors. This result is in line with work in other cell types including colon and ovarian carcinoma cells, confirming the potential of Hsp72 as a molecular target.6 A number of high throughput screens have identified more specific Hsp72 small molecule inhibitors which are suitable for clinical use.7–9 Using these drugs either in combination with Hsp90 inhibitors or other drugs which induce cellular stress and upregulate Hsp72, offers a way to increase clinical efficacy. Our work specifically demonstrates that combinations of specific small molecule inhibitors of Hsp’s 90 and 72 may be particularly effective against immunoglobulin producing, myeloma cells, as these cells intrinsically rely heavily upon heat shock proteins to deal with their protein load and redirect and degrade misfolded proteins.
Conflict of interest The authors are employees of The Institute of Cancer Research, which has a commercial interest in Hsp90 inhibitors. The authors have been involved in funded research collaborations on Hsp90 inhibitors with Vernalis Ltd. and on chaperone and stress pathways with AstraZeneca. Intellectual property arising from the Hsp90 programme has been licensed to Vernalis and Novartis. PW has been a consultant to Novartis.
Letters to the Editor
1807 Acknowledgements We acknowledge NHS funding to the NIHR Biomedical Research Centre. This work was supported by the Kay Kendall Leukaemia Fund, Myeloma UK, Luck-Hille Foundation, Cancer Research UK Program Grant C309/A8274 and PW is a Cancer Research UK Life Fellow.
2 3 4
1
1
1
1
EL Davenport , A Zeisig , LI Aronson , HE Moore , S Hockley1, D Gonzalez1, EM Smith1, MV Powers2, SY Sharp2, P Workman2, GJ Morgan1 and FE Davies1 1 Section of Haemato-Oncology, The Institute of Cancer Research, Sutton, Surrey, UK and 2 Cancer Research UK Centre for Cancer Therapeutics, The Institute of Cancer Research, Sutton, Surrey, UK E-mail:
[email protected]
5
6 7 8
References 9 1 Davenport EL, Moore HE, Dunlop AS, Sharp SY, Workman P, Morgan GJ et al. Heat shock protein inhibition is associated with
activation of the unfolded protein response pathway in myeloma plasma cells. Blood 2007; 110: 2641–2649. Powers MV, Workman P. Inhibitors of the heat shock response: biology and pharmacology. FEBS Lett 2007; 581: 3758–3769. Mitsiades CS, Mitsiades NS, McMullan CJ, Poulaki V, Kung AL, Davies FE et al. Antimyeloma activity of heat shock protein-90 inhibition. Blood 2006; 107: 1092–1100. Davenport EL, Morgan GJ, Davies FE. Untangling the unfolded protein response. Cell Cycle 2008; 7: 865–869. Komarova EY, Afanasyeva EA, Bulatova MM, Cheetham ME, Margulis BA, Guzhova IV. Downstream caspases are novel targets for the antiapoptotic activity of the molecular chaperone hsp70. Cell Stress Chaperones 2004; 9: 265–275. Powers MV, Clarke PA, Workman P. Dual targeting of HSC70 and HSP72 inhibits HSP90 function and induces tumor-specific apoptosis. Cancer Cell 2008; 14: 250–262. Evans CG, Chang L, Gestwicki JE. Heat shock protein 70 (Hsp70) as an emerging drug target. J Med Chem 2010; 53: 4585–4602. Leu JI, Pimkina J, Frank A, Murphy ME, George DL. A small molecule inhibitor of inducible heat shock protein 70. Mol Cell 2009; 36: 15–27. Powers MV, Jones K, Barillari C, Westwood I, van Montfort RL, Workman P. Targeting HSP70: the second potentially druggable heat shock protein and molecular chaperone? Cell Cycle 2010; 9: 1542–1550.
HH-GV-678, a novel selective inhibitor of Bcr-Abl, outperforms imatinib and effectively overrides imatinib resistance a
Leukemia (2010) 24, 1807–1809; doi:10.1038/leu.2010.169; Published online 12 August 2010
Imatinib mesylate (STI-571, Gleevec, Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA) is now used worldwide for the treatment of chronic myelogenous leukemia (CML) with great success; however, CML often progresses or relapses because of the development of imatinib resistance after long-term treatment. The main cause for imatinib resistance is the acquisition of point mutations in the genetic sequences encoding the Abl kinase domain.1 Therefore, the discovery and development of a new generation of Bcr-Abl inhibitors capable of overriding imatinib resistance is of utmost importance. On the basis of the crystallographic structure of the imatinib–Abl complex, we identified a novel inhibitor of Bcr-Abl called HH-GV-678 (Figure 1a). The kinase activities of c-Abl, PDGFRb (platelet derived growth factor receptor-b) and c-Kit were inhibited by HH-GV-678 with IC50 values of 1.2±0.7, 307.6±259.0 and 665.5±306.8 nM; however, this drug had no effect on the kinase activities of vascular endothelial growth factor receptor 2/kinase insert domain receptor (VEGFR2/KDR), Flt3, Ret, VEGFR3, c-Src, epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor 2 (HER2). Imatinib had IC50 values of 100.9±91.8, 201.8±23.6 and 361.8±40.0 nM for c-Abl, PDGFRb and c-Kit, respectively. Consistent with the in vitro results of a kinase assay, HH-GV-678 completely blocked cellular Bcr-Abl autophosphorylation and Stat5 and Erk1/2 phosphorylation in K562 leukemia cells, with much more potent activity than did imatinib (30 vs 300 nM) (Figure 1b). However, HH-GV-678 inhibited the phosphorylations of c-Kit and PDGFRb with less potency (Supplementary Figure 1) and had no effect on the phosphorylation of EGFR, VEGFR, c-Src or HER2 (Supplementary Figure 2). These data indicate that HH-GV-678 is a selective inhibitor of the c-Abl kinase and has much stronger activity than imatinib.
b
0
HH-GV-678
1
3
10
30 100 300 1000
(nM) p-Bcr-Abl p-Stat5 p-Erk1/2 eIF4E
Imatinib
p-Bcr-Abl Bcr-Abl
c
p-Bcr-Abl
HH-GV-678 Bcr-Abl p-Bcr-Abl Imatinib Bcr-Abl
d
p-Bcr-Abl
HH-GV-678 Bcr-Abl p-Bcr-Abl Imatinib Bcr-Abl
Figure 1 Effects of HH-GV-678 and imatinib on cellular tyrosine phosphorylation of Bcr-Abl in cells expressing nonmutated or mutated Bcr-Abl. (a) Chemical structure of HH-GV-678. (b) K562 cells expressing nonmutated Bcr-Abl, (c) 32D cells expressing H396P Bcr-Abl mutant, and (d) 32D cells expressing T315I Bcr-Abl mutant were incubated with HH-GV-678 or imatinib for 3 h. Total cell lysates were analyzed by western blotting, and the levels of phospho- and un-phosphoproteins were determined using specific antibodies. Leukemia
