Targeted therapy of BCL6- dependent diffuse large B-cell lymphomas by heat-shock protein 90 inhibition
Abstract
Heat-shock protein (HSP)90 is a molecular chaperone involved in the proper folding and cellular transportation of many signaling proteins that are deregulated in lymphoma. HSP90 inhibition results in proteasomal degradation of these proteins, leading to antitumoral activity. Recent studies have focused on the use of HSP90 inhibitors as potential therapies for non-Hodgkin lymphoma. BCL6 plays a critical role in the pathogenesis of most diffuse large B-cell lymphoma (DLBCL), the most frequent non-Hodgkin lymphoma. The current study demonstrates that HSP90 forms a complex with BCL6, and inhibition of HSP90 with the drug PU-H71 selectively kills BCL6-positive DLBCL in animal models. These data support the use of HSP90 inhibitor PU-H71 for treating patients with BCL6-positive DLBCL.
KEYWORDS: BCL6 • heat-shock proteins • non-Hodgkin lymphoma • PU-H71
Introduction
Heat-shock proteins (HSPs) are ubiquitous, highly conserved proteins that function as molecular chaperones [1]. Members of the HSP family include HSP90, HSP70, HSP40, p23 and calreticulin, among others. Some members of the family are constitutively expressed, while oth- ers are induced by several forms of cellular stress to protect the cells and enhance survival. HSPs regulate the normal folding, intracellular trans- portation and proteolytic turnover of many pro- teins, including those with critical implications in cell growth, signal transduction and apoptosis. HSP90 plays a critical role in cell proliferation and survival [2]. HSP90 interacts with a num- ber of proteins facilitating their proper folding, assembly and intracellular function. Proteins regulated by HSP90 include receptor tyrosine kinases, telomerase, apoptosis-regulating pro- teins and proteins involved in angiogenesis and metastasis. Overexpression of HSP, particularly HSP90, has been documented in several can- cers, including non-Hodgkin lymphoma, and contributes to the oncogenic process. HSP90, thus, has been considered a preferential target for cancer therapy. Inhibition of HSP90 leads to the degradation of many protein substrates by the proteasome complex and, thus, reduces their oncogenic potential [3].
BCL6 is a proto-oncogene encoding a nuclear transcription factor that is specifically expressed by the normal germinal center [4]. BCL6 expres- sion is required for germinal center development and immunoglobulin affinity maturation on B cells. Deregulation of BCL6 is the most fre- quent genetic abnormality in diffuse large B-cell lymphoma (DLBCL). Constitutive expression of BCL6 in DLBCL is associated with translocations involving 3q37 or BCL6 promoter mutations in the majority of cases. However, other mechanisms might account for its expression in DLBCL, since a substantial amount of DLBCL cases expressing BCL6 lack such genetic lesions. BCL6 protein is required to promote proliferation and survival of DLBCL cells. Specifically, BCL6 suppresses apop- tosis and cell cycle arrest by repressing genes, such as p27, and upregulating c-myc [5]. Further proof of the implication of BCL6 in the pathogenesis of DLBCL comes from the fact that constitu- tive expression of BCL6 in B cells recapitulates this type of lymphoma in mice [6]. For all of this, BCL6 has been considered an attractive target for DLBCL therapy. Targeting transcription factors such as BCL6 with small drugs is often limited by the complexity of action of these proteins. This has been partially overcome by using a peptide inhibitor that spe- cifically targets BCL6, blocking its cellular function. In vitro and in vivo studies with this blocking peptide have demonstrated anti- lymphoma activity, providing further evidence that targeting BCL6 is a promising therapeutic approach for patients with DLBCL. In their key paper, Cerchietti et al. demonstrated that BCL6 protein is regulated by HSP90, and inhibition of HSP90 specifically kills BCL6-positive DLBCL [7].
Methods
In this study, the antilymphoma effect of a new HSP90 inhibitor, PU-H71, was analyzed. Details on the mechanism of action of this drug were analyzed, in particular the interactions of HSP90 with BCL6 protein.
Investigators first tested the in vitro effect of PU-H71 on sev- eral DLBCL cell lines, correlating the antitumor effect with the expression of BCL6. Then, by using immunoprecipitation and protein affinity chromatography analysis they studied the interac- tion of HSP90 with BCL6 and how the HSP90 inhibition affected the kinetics of BCL6 mRNA and protein.
Studies on the expression of HSP90 and BCL6 proteins in pri- mary DLBCL tumors were conducted using immunohistochem- istry on tissue microarrays. The antitumoral effect of PU-H71 on primary DLBCL cells was also tested. First, in vitro studies with the tumors exposed to the drug were performed and apoptosis was analyzed. In addition, in vivo studies in xenograft animal models were done to asses the antitumor efficacy of the drug; severe com- bined immunodeficient mice were injected with different DLBCL cell lines and further treated with several doses of PU-H7, and tumor volumes and overall survival was assessed. Additionally, the effect of PU-H71 in DLBCL cells was further character- ized by analyzing BCL6 protein expression and specific gene- expression changes in treated tumors. Pharmacokinetic studies were done in this model, and the concentration of the drug was assessed in the growing tumors and in normal tissues. Specific toxicity studies were performed in severe combined immuno- deficient mice treated with PU-H71 and also in a different model, the immunocompetent Balb/c mouse.
Results
PU-H71-induced apoptosis in DLBCL cell lines, in a dose-depen- dent manner, mediated through capase-3 and capase-7. This effect was found in cells expressing BCL6, while no significant apoptotic effect was detected in those DLBCL cell lines lacking BCL6.
HSP90 colocalizes with BCL6 in the nucleus of the lymphoma cell, and forms a complex with this protein contributing to the BCL6 action. Treatment of BCL6-positive DLBCL cells with PU-H71 led to a complete depletion of BCL6 protein, mainly mediated through the proteasome, and this effect was indispensable for the cell death to occur. BCL6 mRNA was also affected by the treatment with PU-H71 because treated cells demonstrated decreased BCL6 mRNA levels compared with untreated tumor cells. Additionally, expression of target genes repressed by BCL6 was upregulated after treatment with PU-H71 confirming that BCL6 was the most important target protein after treatment with PU-H71.
Next, a series of studies were performed in primary DLBCL tumors. First, expression of HSP90 and BCL6 proteins was dem- onstrated by immunohistochemistry in DLBCL cells. More than 80% of the tumor specimens were positive for both proteins, with BCL6 expressed in the nucleus and HSP90 in both the cyto- plasm and nucleus of the tumor cells. Second, investigators dem- onstrated that PU-H71 was able to kill primary BCL6-positive DLBCL cells in vitro in 19 out of 21 cases tested.
The antitumor effect of PU-H71 was also tested in vivo in xenografted DLBCL models. PU-H71 was preferentially retained in tumor cells compared with normal tissues in all DLBCL mod- els tested. PU-H71 induced apoptosis and BCL6 protein deple- tion in the tumor cells of treated mice. Treatment of mice with established DLBCL tumors led to a significant reduction in the tumor volume and prolonged survival of mice. Again, this effect was detected only in BCL6-positive DLBCL xenografts, while no antitumor effect was detected in mice xenografted with BCL6- negative tumors. Importantly, no toxic effects were documented in mice even after long-term treatment with PU-H71.
Discussion
This key paper evaluates the antitumoral effect of a new HSP90 inhibitor, PU-H71. More importantly, it demonstrates, for the first time, that the HSP90 complex interacts with BCL6 protein and plays an important role in BCL6 functionality. HSP90 interacts with several proteins that function as key regulators of cellular proliferation, differentiation and apoptotic pathways, such as Akt, cdk4, bcl-2 and bcl-X. HSP90 client proteins implicated in the pathogenesis of lym- phoma include Akt, the NF-B complex and nucleophosmin- ALK. For all of this, HSP90 is an interesting target for developing therapies against lymphoma [2,3].
The benzoquinone ansamycin, 17-AAG, was the first HSP90 inhibitor to enter clinical trials. However, its hepatic toxicity and poor water solubility severely compromise its wide use for the treatment of cancer patients. New HSP90 inhibitors with improved pharmacokinetics have been developed, and PU-H71 is the most potent in its class.
The study by Cerchietti et al. demonstrated that PU-H71 induces in vitro apoptosis of DLBCL cell lines and, more importantly, of DLBCL primary tumors. Moreover, PU-H71 has significant anti- tumor activity in DLBCL animal models with no significant toxic- ity. An interesting data with clinical implications is the finding that PU-H71 accumulates preferentially in the tumor cells compared with normal cells. The explanation for this selectivity comes from data revealing that HSP90 is present in the tumors in its high- affinity activated form, whereas in normal tissues HSP90 is in an inactive form, which has a considerably lower drug affinity [8]. The most striking data of this work is that the antitumor effect of PU-H71 in DLBCL is dependant on the expression of BCL6. BCL6 is the most commonly involved oncoprotein in DLBCL. Translocations affecting 3q37, promoter mutations, and other less well-defined genetic lesions account for the con- stitutive expression of BCL6 in DLBCL. For this reason, BCL6 has been considered an attractive therapeutic target for DLBCL. Previous work by Melnick’s group focused on the inhibition of BCL6 to kill DLBCL tumors [9]. The fact that BCL6 mediates its actions indirectly, via protein–protein interactions, makes the use of small BCL6-inhibitor drugs of limited efficacy for treating DLBCL. For this reason, a BCL6-specific peptide inhibitor fused to a nuclear localization signal was designed and proven to be therapeutically effective in DLBCL animal models. However, this approach requires frequent administration of the peptide to exert its therapeutic action, which limits its translation to the patient. Evidence is provided in the paper that HSP90 interacts with both BCL6 mRNA and protein, and contributes to the maintainance of sufficient BCL6 levels for exerting the repressing action over their target genes, which are eventually implicated in lymphoma- genesis. Moreover, among the client proteins associated to HSP90 in DLBCL, BCL6 represents one of the most important. Hence, inhibition of HSP90 by PU-H71 in DLBCL leads to degradation of BCL6 without need to use other complex strategies, such as specific peptide inhibitors.
There is evidence from previous work that BCL6 is constitutively expressed in those DLBCL with a B-cell-receptor gene signature [10]. In line with this, PU-H71 preferentially targets the BCR DLBCL subtype, suggesting that HSP90 inhibitors may cooperate with the new drugs inhibiting the BCR signaling cascade (i.e., syk inhibitors) to enhance their antitumor effect in DLBCL [11].
Key issues
• Heat-shock protein 90 interacts with BCL6 in diffuse large B-cell lymphoma (DLBCL) and regulates BCL6 function.
• BCL6 is involved in the pathogenesis of most cases of DLBCL.
• PU-H71 is the most potent heat-shock protein 90 inhibitor.
• PU-H71 selectively induces in vitro and in vivo apoptosis of BCL6-positive DLBCL.
References
Papers of special note have been highlighted as:
• of interest
•• of considerable interest
1 Takayama S, Reed J, Homma S. Heat-shock proteins as regulators of apoptosis. Oncogene 22, 9041–9047 (2004).
2 Whitesell L, Lindquist S. HSP90 and the chaperoning of cancer. Nat. Rev. Cancer 5, 761–772 (2005).
3 Kamal A, Boehm M, Burrows F. Therapeutic and diagnostic implications of Hsp90 activation. Trends Mol. Med. 10, 283–290 (2004).
4 Klein U, Dalla-Favera R. Germinal centres: role in B-cell physiology and malignancy. Nat. Rev. Immunol. 8, 22–33 (2008).
5 Shaffer AL, Lin K, Kuo T et al. BCL-6 represses genes that function in lymphocyte differentiation, inflammation, and cell cycle control. Immunity 13, 199–212 (2000).
• In-depth description of the function of BCL6 in B cells and its implications for lymphomagenesis.
6 Cattoretti G, Pasqualucci L, Ballon G et al. Deregulated BCL6 expression recapitulates the pathogenesis of human diffuse large B cell lymphomas in mice. Cancer Cell 7, 445–455 (2005).
•• Demonstrates the role of BCL6 in the pathogenesis of diffuse large B-cell lymphoma.
7 Cerchietti L, Lopes E, Yang S et al. A purine scaffold Hsp90 inhibitor destabilizes BCL-6 and has specific antitumor activity in BCL-6-dependent B cell lymphomas. Nat. Med. 15, 1369–1377 (2009).
8 Kamal A, Thao L, Sensintaffar J et al. A high-affinity conformation of Hsp90 confers tumour selectivity on Hsp90 inhibitors. Nature 425, 407–410 (2003).
9 Polo JM, Dell’Oso T, Ranuncolo SM et al. Specific peptide interference reveals BCL6 transcriptional and oncogenic mechanisms in B-cell lymphoma cells. Nat. Med. 10, 1329–1335 (2004).
•• Demonstrates for the first time how inhibition of BCL6 constitutes a target for the therapy of diffuse large B-cell lymphoma.
10 Polo JM, Juszczynski P, Monti S et al. Transcriptional signature with differential expression of BCL6 target genes accurately identifies BCL6-dependent diffuse large B cell lymphomas. Proc. Natl Acad. Sci. USA 104, 3207–3212 (2007).
11 Juszczynski P, Chen L, O’Donnell E et al. BCL6 modulates tonic BCR signaling in diffuse large B-cell lymphomas by repressing the SYK phosphatase, PTPROt. Blood 114, 5315–5321 (2009).