A peptomimetic inhibitor of BCL6 with potent antilymphoma effects in vitro and in vivo

Abstract

The BCL6 transcriptional repressor is the most commonly involved oncogene in diffuse large B-cell lymphomas (DLBCLs). BCL6 lymphomagenic activity is dependent on its ability to recruit corepressor proteins to a unique binding site on its N-terminal BTB domain. A recombinant peptide fragment of the SMRT (silencing mediator for retinoid and thyroid hormone receptor) corepressor that blocks this site can inhibit BCL6 biologic functions. Shortening and conversion of this peptide to D-amino acid and retro configuration as well as the addition of a fusogenic motif yielded a far more potent and stable BCL6 inhibitor that still retained the specificity of the original SMRT fragment. Like the L-peptide, retroinverso BCL6 peptide inhibitor (RI-BPI) selectively killed BCR rather than OxPhos-type DLBCL cells. The RI-BPI could recapitulate the failure to form germinal centers seen in BCL6 null mice yet was nontoxic and nonimmunogenic even when administered for up to 52 weeks. RI-BPI showed superior duration of tissue penetration and could accordingly powerfully suppress the growth of human DLBCLs xenografts in a dose-dependent manner. Finally, RI-BPI could kill primary human DLBCL cells but had no effect on normal lymphoid tissue or other tumors.

Figure 1

Rational design of a BCL6 peptide inhibitor. The figure shows the structure, average GI₅₀, and dose frequency (ie, the number of doses given over a 48-hour period) for successive versions of BPI. GI₅₀ for the different peptides is an average of those obtained in the OCI-Ly1, OCI-Ly7, and OCI-Ly10 DLBCL cell lines. Each peptide and its respective control were tested at multiple doses and frequencies, with the goal of reaching the same potency as recombinant BPI with a single dose administration. The S3 and S6.2 (RI-BPI) peptides (shown in bold) were selected for further studies for the rest of the experiments shown in this article. GI₅₀ indicates growth inhibitory concentration 50%; DF, dose frequency during a 48-hour interval; R, recombinant; S, synthetic; NA, not applicable; light-gray background, L peptides; white background, L-D hybrid peptides; and dark-gray background, D peptides.

Figure 2

RI-BPI specifically inhibits the transcriptional and biologic function of BCL6. (A) BCL6 immunoprecipitation (IP) in OCI-Ly1 cell lysates after exposure to RI-BPI^biotin or CP^biotin (control) peptides. Anti-IgG was used as control for the IP. Detection of complexes was done using avidin-HRP conjugates. (B) Reporter assays performed in 293T cells transfected with BTB-fusion constructs as indicated. Cells were exposed to control peptide (□) or RI-BPI 5 μM (), 10 μM (), or 20 μM (■). Fold repression is expressed versus the effect of each dose on a GAL4-DBD vector control for each experiment, relative to a TK-Renilla internal control. (C) Chromatin IP from OCI-Ly1 cells treated with CP 20 μM (□) or RI-BPI 20 μM (■) using antibodies against SMRT, BCL6, and actin (as a negative control) and amplifying the promoter region surrounding the BCL6 binding site on the TP53 gene by quantitative PCR. Results are expressed as percentage relative to the input. (D) Real-time detection of mRNA of the endogenous BCL6 target genes ATR (□) and TP53 (), and the control gene CD20 (■), performed in the BCL6-dependent cell lines OCI-Ly7, OCI-Ly10, and OCI-Ly1 and in the BCL6-independent cell line OCI-Ly4, after treatment with RI-BPI 20 μM and CP 20 μM. Results are expressed as fold change in mRNA abundance mediated by RI-BPI over CP.

Figure 3

RI-BPI selectively kills BCR-type DLBCL cells. Dose-response curves for RI-BPI in a panel of 11 DLBCL cell lines. The x-axis shows the dose of BPI in micromoles. The y-axis shows the effect of RI-BPI compared with CP on cell viability. ● indicates BCL6-dependent cell lines (BCR-type); and , BCL6-independent (OxPhos-type) cell lines. The goodness-of-fit for the experimental data to the median-effect equation is represented by the linear correlation coefficient (r) obtained from the logarithmic form of this equation. There was a statistically significant difference between the average of the GI₅₀ values of BCL6-dependent versus BCL6-independent cell lines (P < .001, t test).

Figure 4

RI-BPI effectively distributes to lymphomas after parenteral administration. (A) The serum concentration of RI-BPI^biotin and L-BPI^biotin was determined after the intraperitoneal administration of 500 μg to mice carrying SU-DHL4 xenografts. Serum was taken at several time points (x-axis), and the concentration of biotinylated peptides was determined by chemical reaction with avidin-HRP (y-axis). (B) Histochemistry of the SU-DHL4 xenografts injected with RI-BPI^biotin and L-BPI^biotin performed at similar time points as in panel A. The presence of peptide was detected using Texas Red–avidin conjugates followed by fluorescence microscopy. Slides were mounted with permanent mounting medium (Vectashield Hard set; Vector Laboratories, Burlingame, CA) to prevent photobleaching. Slides were viewed with a fluorescent microscope (AxioSkop 2; Carl Zeiss, Jena, Germany) using a Plan-neofluar lens at a 10×/0.50 air objective and a 25×/0.80 oil objective. Images were acquired using a color camera (AxioCam; Carl Zeiss), and were processed using Axiovision software (Carl Zeiss).

Figure 5

RI-BPI has antilymphoma activity in vivo. (A) Tumor growth plots in SUDHL4 and SUDHL6 xenografted mice treated with control (●) or RI-BPI at 150 μg/day (□) or at 500 μg/day () for 10 consecutive days. The control peptide group includes mice treated with both the 150 and 500 μg/day doses, which were pooled to facilitate visualization. The y-axis represents the percentage of tumor volume (in mm³) compared to day 1 of treatment and x-axis represents treatment day. (B) Tumor burden (in milligrams) at day 10 in control (■), RI-BPI 150 μg/day (□), and RI-BPI 500 μg/day () treated SUDHL4 and SUDHL6 mice. (C) Serum levels of human β₂-microglobulin (measured in micrograms per milliliter and expressed as percentage to their respective controls) at day 10 in control (■), RI-BPI 150 μg/day (□), and RI-BPI 500 μg/day () treated SUDHL4 and SUDHL6 mice. (D) Kaplan-Meier survival curves for the pooled mice treated with control (dashed black line), 150 μg/day (gray line), and RI-BPI 500 μg/day (black line). A positive event was defined preanalysis as either death of the animal or tumor equal to 10 times the initial volume, whichever occurs first.

Figure 6

RI-BPI has no antilymphoma activity in the Toledo xenograft. (A) Tumor growth plot in Toledo (an OxPhos-type cell line) xenografted mice treated with control peptide (●) or RI-BPI 500 μg/day () for 10 consecutive days. The y-axis represents the percentage of tumor volume (in mm³) compared with day 1 of treatment, and the x-axis represents treatment day. (B) Tumor burden (in milligrams) at day 10 in control peptide (■) and RI-BPI 500 μg/day () treated Toledo mice. (C) Serum levels of human β2-microglobulin (in micrograms per milliliter) at day 10 in control peptide (■) and RI-BPI 500 μg/day () treated Toledo mice. (D) Representative images from Toledo mice tumors after being treated with control peptide (first column) or RI-BPI 500 μg/day (second column), and assayed for apoptosis by TUNEL. The plot on the far right represents the apoptotic index (apoptotic cells over total cells) with the percentage of apoptotic cells in the y-axis for control peptide (■) and RI-BPI 500 μg/day (). Slides were mounted with permanent mounting medium (Vectamount; Vector Laboratories). Slides were viewed with a light microscope (AxioSkop 2; Carl Zeiss) using a Plan-neofluar lens at a 10×/0.50 air objective, a 25×/0.80 oil objective, a 40×/0.90 oil objective, and a 100×/1.30 oil objective. Images were acquired using a color camera (AxioCam; Carl Zeiss), processed using Axiovision software (Carl Zeiss), and scored using ImageJ software (National Institutes of Health, Bethesda, MD).

Figure 7

RI-BPI inhibits BCL6 transcriptional repression and induces DLBCL apoptosis in vivo. (A) Representative images from SU-DHL4 and SU-DHL6 mice tumors after being treated with control peptide (first column), RI-BPI 150 μg/day (second column), or RI-BPI 500 μg/day (third column), and assayed for apoptosis by TUNEL. The plot on the far right represents the apoptotic index (apoptotic cells over total cells) with the percentage of apoptotic cells in the y-axis for pooled control peptide doses (■), RI-BPI 150 μg/day (□), and RI-BPI 500 μg/day () in the x-axis. (B) The same tumors as in panel A were assayed for proliferation by proliferating cell nuclear antigen immunostaining. Nuclei were classified as negative, low-intensity positive, and high-intensity positive as shown in the inset (digital zoom) of the SUDHL4 control tumor, by green, yellow, or red circles, respectively. The plot on the far right represents the proliferation index (using the same color coding for negative and positive cells) with the percentage of proliferating and nonproliferating cells in the y-axis for pooled control peptide samples (first stacking column), RI-BPI 150 μg/day (second stacking column), and RI-BPI 500 μg/day (third stacking column). (C) The same tumors were examined for the presence of mitotic cells. The plot on the far right represents the mitotic index (mitotic cells over total cells) with the percentage of mitotic cells in the y-axis in pooled control peptide samples (■), RI-BPI 150 μg/day (□), and RI-BPI 500 μg/day (). (A-C) Slides were mounted with permanent mounting medium (Vectamount; Vector Laboratories). Slides were viewed with a light microscope (AxioSkop 2; Carl Zeiss) using a Plan-neofluar lens at a 10×/0.50 air objective, a 25×/0.80 oil objective, a 40×/0.90 oil objective, and a 100×/1.30 oil objective. Images were acquired using a color camera (AxioCam; Carl Zeiss), processed using Axiovision software (Carl Zeiss), and scored using ImageJ software (National Institutes of Health, Bethesda, MD). (D) The mRNA abundance of TP53 and ATR was determined in the same tumors by quantitative RT-PCR. The y-axis represents the normalized amount of mRNA in arbitrary units as measured by the relative standard curve method.

Figure 8

RI-BPI inhibits the growth of primary human DLBCLs. Single-cell suspensions were obtained from lymph node biopsies of patients suspected of having DLBCL and were treated with either BPI 10 μM (bars) or CP 10 μM (line). indicates non-DLBCL samples; and ■, DLBCLs. The y-axis represents the percentage of viable cells compared with control peptide, which is represented by the line at 100%. Error bars represent the SEM for triplicates. BCL6 status as assessed by immunohistochemistry is shown on the bottom for each DLBCL case. HD indicates Hodgkin disease; TCL, T-cell lymphoma; and Misc, miscellaneous diagnosis.