ABSTRACT
The activated B-cell (ABC) subtype of diffuse large B-cell lymphoma (DLBCL) is characterized by the chronic activation of signaling initiated by immunoglobulin-μ (IgM). By analyzing DNA copy profiles of 1,000 DLBCLs, we identified gains of 18q21.2 as the most frequent genetic alteration in ABC-like DLBCL. We show that these alterations target the TCF4 (E2-2) transcription factor, and that over-expression of TCF4 leads to its occupancy on immunoglobulin gene enhancers and increased expression of IgM at the transcript and protein level. The TCF4 gene is one of the top BRD4-regulated genes in DLBCL. Using a BET proteolysis-targeting chimera (PROTAC) we show that TCF4 and IgM expression can be extinguished, and ABC-like DLBCL cells can be killed in vitro and in vivo. This highlights a novel genetic mechanism for promoting immunoglobulin signaling in ABC-like DLBCL and provides a functional rationale for the use of BET inhibitors in this disease.
INTRODUCTION
Diffuse large B-cell lymphoma (DLBCL) is the most common form of lymphoma and is curable in ∼60% of patients using a combination chemo-immunotherapy regimen, R-CHOP1,2. However, those that are refractory to, or relapse following, first-line therapy have a dismal outcome3. Chimeric antigen receptor (CAR)-T cells are likely to change the landscape of outcomes in relapsed/refractory patients, but a large number of patients are not eligible for CAR-T therapy and ∼50% of those that received CAR-T progress within 12 months4. Novel rationally-targeted therapeutic strategies are therefore needed for DLBCL.
The clinical heterogeneity of DLBCL is underpinned by molecular heterogeneity, with the major distinction being between the germinal center B-cell (GCB)-like and activated B-cell (ABC)-like ‘cell of origin’ (COO) subtypes that were identified by gene expression profiling5. The GCB-like subtype shows transcriptional similarities to normal germinal center B-cells, whereas the ABC-like subtype shows transcriptional similarities to CD40-activated B-cells or plasmablasts. Patients with ABC-like DLBCL have significantly worse overall survival compared to patients with GCB-like DLBCL, when treated with the standard-of-care combination chemotherapy (CHOP) plus rituximab (R-CHOP) regimen6. The ABC-like DLBCL subtype expresses immunoglobulin μ (IgM)7 in >90% of cases, which forms the B-cell receptor (BCR) signaling complex in association with CD79A and CD79B and drives chronically active BCR signaling. Several genetic alterations have been shown to promote this signaling, including mutations of the CD79A, CD79B, CARD11, and MYD88 genes8-11. However, these mutations only account for approximately two thirds of ABC-like DLBCL cases12, suggesting that other significant genetic drivers remain to be defined.
A common mechanism for tumorigenesis is the gain or loss of DNA encoding oncogenes or tumor suppressor genes, respectively. These copy number alterations (CNAs) perturb a higher fraction of the cancer genome than somatic nucleotide variants (SNVs) and small insertion/deletions (InDels) and are critically important to cancer etiology13. Here, we have integrated multiple datasets, including DNA copy number profiles of 1,000 DLBCLs, and identified DNA copy number gain of the E2 transcription factor TCF4 as the most frequent genetic alteration in ABC-like DLBCL. We show that TCF4 is capable of driving IgM expression and is amenable to therapeutic targeting through BET inhibition. These data therefore highlight a novel genetic basis for ABC-like DLBCL with potential implications for future clinical studies.
RESULTS
DNA copy number gains of chromosome 18 are the most frequent genetic alteration in DLBCL
In order to identify significant CNAs in DLBCL, we interrogated the genomic profiles of 1,000 DLBCLs using the GISTIC2 algorithm14. These included high-resolution SNP microarrays from 860 previously published cases, in addition to next generation sequencing (NGS)-derived profiles from our own cohort of 140 cases (Table S1-2). Our analysis revealed 20 significant DNA copy number gains and 21 significant DNA copy number losses (false discovery rate [FDR] <0.1; Fig. 1A and Table S3). Using a subset of 448 cases for which COO subtype data was available, we identified 9 DNA copy number alterations that were significantly more frequent in ABC-like DLBCL and 11 that were significantly more frequent in GCB-DLBCL (Fisher Q-value<0.1; Fig. 1B and S1; Table S4). The most frequent genetic alteration in ABC-like DLBCL was gain of 18q21.2, which was observed in 44% of tumors. In line with the association with ABC-like DLBCL, 18q21.2 gains were associated with significantly reduced overall survival in both CHOP and R-CHOP treated patients (Fig 1C-D). Using 199 tumors with matched COO subtype, DNA copy number data and mutation status for 40 genes, we observed that the frequency of 18q21 gains (23.1% of all tumors; 40.7% of ABC-like tumors) was higher than other ABC-like DLBCL-associated somatic mutations including MYD88 mutation (16.6% of all tumors; 33.3% of ABC-like tumors), CD79B mutation (7.5% of all tumors; 18.5% of ABC-like tumors) and other ABC-associated genes (Fig 1E, S2 and Table S5). Because multiple genetic alterations are associated with ABC-like DLBCL, we employed the REVEALER algorithm15 to identify the set of genetic alterations that best explained the ABC-like DLBCL signature. Using a set of 87 DNA copy number alterations and recurrently mutated genes as the feature set and MYD88 mutations as the seed feature, REVEALER identified an additional 4 genetic alterations including 18q21.2 gain as those best associating with the ABC-like signature (Fig. 1F). Gains of 18q21 are therefore the most frequent genetic feature of ABC-like DLBCL and are predicted to contribute to this molecular phenotype.
The TCF4 (E2-2) transcription factor is the target of 18q21 gains in ABC-like DLBCL
Gains of 18q have been previously attributed to the BCL2 oncogene16,17. However, our analysis of this large cohort provided the resolution to identify two significant peaks of DNA copy gain on chromosome 18; 18q21.2 (16 genes, Q=4.8×10−14) and 18q22.1 (70 genes, Q=1.1×10−7; Table S3). We further integrated GEP data from 249 tumors to identify the likely targets of these lesions by testing for the increase in expression of genes within the most significant peaks of DNA copy gain. This highlighted TCF4 and BCL2 as likely targets of the 18q21.2 and 18q22.1 gains, respectively (Fig. 2A; Table S6). Notably, most 18q copy number alterations incorporated both of these genes (Fig 2A-C). Only 7.3% or 1.0% of ABC-like DLBCLs have copy number alterations targeting TCF4 or BCL2 alone, respectively (Fig. 2B-C). In addition, we observed that TCF4 was more highly expressed in ABC-like DLBCL compared to GCB-like DLBCL generally, but was further increased by DNA copy gain (Fig 2D). In line with this, ABC-like DLBCL cell lines expressed TCF4 protein irrespective of DNA copy number, but these levels were significantly increased by DNA copy gain (Fig 2E).
The TCF4 gene encodes an E2 family transcription factor, E2-2. Mutations of another E2 transcription factor, TCF3, and its negative regulators ID2 and ID3 are frequent in Burkitt’s lymphoma (BL) and promote immunoglobulin signaling18,19. We therefore interrogated the mutation status of these genes and TCF4 copy gains in our cohort of 140 DLBCLs and a prior cohort of 108 BLs that were sequenced and analyzed with the same approach20. We did not observe recurrent mutations of TCF4 or ID2 in this BL cohort, and mutations of TCF3 and ID3 were infrequent in DLBCL (data not shown). However, in BL, gains of TCF4 were present at the same frequency as TCF3 mutations (18%). Furthermore, TCF4 gains were significantly mutually-exclusive from TCF3 and ID3 mutations (Fisher P=0.019; Fig. 2F), suggesting that TCF4 gains may serve a similar function as TCF3/ID3 mutations in promoting immunoglobulin signaling. These data therefore show that the TCF4 gene is highly expressed in ABC-like DLBCL, with expression further promoted by frequent 18q21.2 DNA copy gains, and implicates TCF4 in immunoglobulin signaling.
TCF4 regulates IgM and MYC expression in ABC-like DLBCL
To identify potential target genes of TCF4, we performed differential gene expression analysis of primary DLBCL tumors with TCF4 DNA copy gain (n=51) compared to those without (n=59). This analysis was limited to ABC-like tumors so as to eliminate the confounding effect of genes that differ in expression between COO subtypes. A total of 355 genes (472 probe-sets) and 87 genes (107 probe-sets) were found to be expressed at significantly higher or lower levels in tumors with TCF4 gain, respectively (Q<0.1, fold-change≥1.2; Fig 3A; Table S7). We performed ChIP-seq of ABC-like DLBCL cell lines, SUDHL2 and TMD8, with tetracycline-inducible Myc-DDK-tagged TCF4 in order to define whether these genes were direct transcriptional targets of TCF4 (Fig 3A). Importantly, TCF4 was expressed at a level comparable to that in the U2932 cell line with TCF4 copy gain (Fig. S3). Using the intersection of significant peaks from both cell lines, we identified TCF4 binding proximal to 180/355 genes with increased expression and 46/87 genes with decreased expression in tumors with TCF4 copy gain (Fig. 3A-B; Table S8). These peaks showed a highly significant enrichment of motifs containing E-box consensus sequences (CANNTG; Fig. S4), and many of the same regions are also bound by TCF4 in plasmacytoid dendritic cell neoplasms21 (Fig. S4), providing strong evidence that we detected on-target binding. Among the most significant ChIP-seq peaks were those within the immunoglobulin heavy chain locus (Fig. 3B), in line with the significantly higher expression of IGHM in ABC-like DLBCL tumors with TCF4 copy gain (Fig. 3C). This included peaks immediately upstream and downstream of the IGHM and IGHD genes, respectively, in regions with corresponding H3K27Ac in normal CD20+ B-cells that indicates they are bona fide enhancers (Fig. 3D). Tetracycline-inducible expression of TCF4 in ABC-like DLBCL cell lines led to a marked increase in IGHM at the transcript (Fig. 3E) and protein level (Fig. 3F). In comparison, BCL2 expression was not induced by TCF4 over-expression and MYC induction was restricted to the two cell lines that lacked MYC translocation (SUDHL2 and TMD8; Fig. S5). These data show that IgM is a direct target of TCF4 and can be induced by its over-expression in ABC-like DLBCL.
TCF4 can be targeted by the BET proteolysis-targeting chimera (PROTAC), ARV771
The TCF4 gene is one of the most highly BRD4-loaded genes in DLBCL, including in ABC-like DLBCL cell lines with TCF4 copy gain (Fig. S6). We therefore evaluated small molecule BET inhibitors and a BET protein degrader, ARV771, as a potential avenue for reducing TCF4 expression in ABC-like DLBCL cell lines with high-copy number of TCF4. The small molecule BET inhibitors, JQ1 and OTX015, resulted in an up-regulation of BRD4 that was not observed with ARV771 due to its role as a sub-stoichiometric BRD4 degrader (Fig 4A and S7). This was associated with a greater efficacy of ARV771 in reducing the BRD4 target genes, MYC and TCF4 (Fig 4A), and the ability of ARV771 to induce apoptosis of these cell lines (Fig 4B). However, as MYC is also a target of TCF4, the down-regulation of MYC is likely partially mediated through TCF4.
Reductions of TCF4 by ARV771 treatment were accompanied by reduced expression of the TCF4 target genes. This included significant reductions of IgM at the transcript and protein level (Fig. 4C-D and Table S9). ARV771 treatment led to significant down-regulation of the set of genes that were identified as being increased in association with TCF4 DNA copy number gain in primary tumors (Fig. 4E). The promising in vitro activity of ARV771 led us to test whether this compound would be efficacious in vivo. In xenografts of the U2932 (Fig. 4F-I) and RIVA (Fig. 4J-M) cell lines that express high levels of TCF4, we observed that ARV771 was able to significantly reduce tumor growth. At the end of treatment, tumors were significantly smaller in ARV771-treated mice and this led to a significant prolongation of survival in these mice (Log Rank P-value < 0.05). Together these data demonstrate a clear functional rationale for BET inhibition in ABC-like DLBCL, and show that ARV771 is effective at eliminating TCF4 and its target genes and treating ABC-like DLBCL in vivo.
DISCUSSION
The ABC-like subtype is one of two major molecular subtypes of DLBCL that is recognized by the WHO classification22. These tumors are driven by chronic active B-cell receptor signaling that emanates from autoreactive IgM that is localized to the cell surface and intracellular lysosomes8,23-25. Mutations in CD79B and MYD88 deregulate this signaling through the reduction of LYN-mediated negative feedback and by activation of IRAK signaling, respectively8,10. However, recent murine studies have shown that MYD88 mutation alone drove a phenotype that was reminiscent of peripheral tolerance, and this was only relieved by the combination of MYD88 and CD79B mutations together, or by increased expression of surface IgM26. The ABC-like phenotype is therefore the result of cumulative epistatic genetic alterations, rather than a single dominant driver mutation. In further support of this notion, recent genomic studies have defined co-associated sets of genetic alterations that co-segregate with unique genetic subsets of ABC-like DLBCL27. The “cluster 5” subset of ABC-like DLBCL included frequent MYD88 and CD79B mutations, but the most frequent genetic alteration in this subtype was DNA copy number gain of 18q27.
We identified the TCF4 (aka E2-2) gene as the most significant target of 18q DNA copy number gains in DLBCL. The TCF4 gene (aka E2-2) is closely related to TCF3 (aka E2A), with both encoding helix-loop-helix transcription factors that form dimers and recognize E-box consensus sequences (CANNTG)28. Murine conditional knock-out studies showed that TCF3 and TCF4 are critical regulators of germinal center B-cell and plasma cell development, in part due to their role in activating immunoglobulin heavy- and light-chain enhancer elements29,30. The ID2 and ID3 proteins bind to and inhibit the activity of TCF3 and TCF4 by preventing their dimerization and DNA binding28. The TCF3 and ID3 genes are recurrently mutated in another form of B-cell lymphoma, Burkitt Lymphoma, with the mutations residing in the interface between TCF3 and ID3 and preventing their interaction18,19. Mutations in ID3 are approximately twice as frequent as mutations of TCF3, and presumably also reduce the interaction between ID3 and TCF4 considering the high degree of homology between these two proteins. We observed that TCF4 DNA copy number gains are also frequent in Burkitt lymphoma, and that they mutually exclude TCF3 and ID3 mutations, providing further evidence for the importance of TCF3/TCF4 deregulation in this disease. In contrast to Burkitt lymphoma, TCF3 and ID3 mutations are rare in ABC-like DLBCL, but TCF4 DNA copy number gains are present at more than twice the frequency. In line with the murine studies, we observed a marked up-regulation of IgM transcript expression in primary tumors with TCF4 DNA copy number gain. We also identified binding sites for TCF4 in the immunoglobulin heavy chain locus and showed that induced expression of TCF4 was sufficient for increasing the expression of IgM at the transcript and protein level. Together, this provides strong evidence for a functional role of TCF4 in promoting IgM expression in ABC-like DLBCL. This is particularly important in this disease, because >90% of ABC-like DLBCL cases express IgM and the disease etiology centers on pathogenic signaling downstream of this receptor7,8. Notably, TCF4 was more highly expressed in ABC-like DLBCL compared to GCB-like DLBCL generally, even in cases without DNA copy number gain of the locus. This suggests that this axis may be active in all ABC-like DLBCLs, and further enhanced in the ∼40% that harbor 18q DNA copy number gains. This is akin to the role of EZH2 in GCB-like DLBCL, which promotes the survival and proliferation of all germinal center B-cells but has enhanced activity in the context of hypomorphic somatic mutations31,32. We therefore hypothesize that TCF4 may participate in a critical functional axis of immunoglobulin regulation in all ABC-like DLBCL.
Proteins in the BET family, including BRD4, are attractive therapeutic targets in cancer due to their role in the transcriptional activation of oncogenes such as MYC33,34. In DLBCL, BRD4 targets include key transcription factors such as BCL6, PAX5 and IRF435. We have highlighted TCF4 as another prominent target of BRD4 in DLBCL, as has been previously described in plasmacytoid dendritic cell neoplasms21. Due to the difficulty in directly drugging transcription factors, BET inhibition therefore represents a logical avenue for reducing TCF4 expression in ABC-like DLBCL. Notably, cell line studies have shown that the small molecule BET inhibitor OTX015 induces apoptosis in ABC-like DLBCL cell lines, as compared to a predominantly cytostatic effect in GCB-like DLBCL cell lines36. However, small molecule inhibitors have also been shown to result in the up-regulation of BRD4 expression37. We therefore evaluated a novel BET protein PROTAC, ARV-771, which combines a BET-targeting warhead from OTX015 with a moiety that recruits the VHL ubiquitin ligase37. Because the PROTAC is not degraded, this results in sub-stoichiometric proteolysis of BET proteins, including BRD4. We found that ARV-771 was able to inhibit the expression of TCF4 at 10-fold lower concentrations than small molecule BET inhibitors. The inhibition of TCF4 expression in ABC-like DLBCL cell lines with high TCF4 copy number was accompanied by the coordinate down-regulation of genes that were highly expressed in primary tumors with TCF4 DNA copy gain, suggesting that a subset of the broad transcriptional changes associated with BRD4 degradation were the consequence of reduced TCF4. This was also associated with the induction of apoptosis at low nanomolar doses of ARV-771, significant reduction of tumor growth in vivo, and a significant prolongation in the life of tumor-bearing animals. Together, this highlights TCF4 DNA copy gains as a functional rationale for BET inhibition in ABC-like DLBCL and shows that the BET PROTAC ARV-771 has significant activity in this context. The over-expression of BCL2 has been described as a resistance mechanism for BET inhibitors38. We observed that the majority of 18q DNA copy number gains in DLBCL encompass both the TCF4 and the BCL2 gene, and we therefore posit that the promising activity of BET inhibitors in ABC-like DLBCL may be further enhanced by combination with a BCL2 inhibitor such as Venetoclax. In support of this, BET inhibitors have been shown to act synergistically with Venetoclax in myeloid leukemia38 and in another form of B-cell lymphoma, mantle cell lymphoma39. Combination of BET and BCL2 inhibition therefore represents an attractive therapeutic avenue for future investigation in ABC-like DLBCL.
In conclusions, we have identified DNA copy number gains of TCF4 as the most frequent genetic alteration in ABC-like DLBCL. Increased expression of TCF4 leads to its occupancy on IgM enhancer elements and increased expression of IgM at the transcript and protein level. We have shown that BET-targeting PROTACs efficiently reduce the expression of TCF4 and its target genes, induce apoptosis in ABC-like DLBCL cells, and prolong the life of mice bearing ABC-like DLBCL tumors. This study therefore highlights the BRD4-regulated TCF4 and IgM axis as a functional rational for the use of BET inhibitors in ABC-like DLBCL.
MATERIALS AND METHODS
DNA copy number data acquisition and processing
Publicly available data for single nucleotide polymorphism microarrays and array comparative genome hybridization platforms with >200,000 markers were downloaded from the gene expression omnibus16,17,40-46 (Table S1; www.ncbi.nlm.nih.gov/geo/). These included Affymetrix 250K and SNP 6.0 platforms, and the Agilent 244K platform. For Affymetrix microarrays, raw CEL files were extracted and copy number predicted using the Affymetrix Copy Number Analysis for Genechip (CNAG) tool, with reference to data from 100 Caucasian HapMap samples. Agilent data was analyzed using BioConductor, as previously described17. Data for all arrays were represented as Log2 copy number change and segmented using the circular binary segmentation (CBS) tool in GenePattern47. Peaks of significant DNA copy number loss and gain were identified using GISTIC2.014. The thresholds utilized for DNA copy number gain and loss were 0.2 copies over a region encompassing 100 markers.
Targeted next generation sequencing (NGS) and variant calling
Genomic DNA for 140 fresh/frozen DLBCL tumors were obtained from the University of Nebraska Medical Center (UNMC) lymphoma tissue bank (IRB 161-95-EP) and interrogated by targeted sequencing of a panel of 380 genes, as previously described20. In brief, 500-1,000ng of genomic DNA was sheared using a Covaris S2 instrument and libraries prepared using KAPA Hyper Prep kits (Roche) and Illumina TruSeq Adapters (Bioo Scientific) according to the manufacturer’s protocol. A maximum of 6 cycles of PCR was used for library preparation. Samples were 12-plexed, subjected to hybrid capture with a 5.3Mbp Nimblegen SeqCap custom reagent (Roche), and amplified by 8 cycles of PCR. Each pool was sequenced on a single lane of an Illumina HiSeq 2500 instrument in high-output mode using 100bp paired-end reads at the Hudson Alpha Institute for Biotechnology Genome Sequencing Laboratory. Raw sequencing reads were aligned to the human genome (hg19) using BWA-Mem48, realigned around InDels using GATK49, sorted and deduplicated using Picard tools, and variants were called according to a consensus between VarScan250 and GATK Unified Genotyper49. This approach has been validated to have a specificity of 92.9% and a sensitivity of 86.7%51. Average on-target rate for this dataset was 88% and average depth of coverage 623X (min = 122X, max = 1396X). Raw FASTQ files for the targeted NGS of previously published DLBCLs (n=119; European Nucleotide Archive Accession ERP021212)52 and Burkitt’s lymphomas20 were also analyzed with the same pipeline, and the results integrated. The DNA copy number of UNMC DLBCL and Burkitt lymphoma cohorts was determined using CopyWriteR53 with 200kB windows.
DLBCL Cell Lines
The SU-DHL-2 cell line was obtained from ATCC. The RIVA (aka RI-1), HBL1, TMD8, U2932 and OCI-Ly10 cell lines were obtained from DSMZ. The DNA copy number profile of DLBCL cell lines were derived from previously reported SNP6.0 data54 or targeted next generation sequencing, as described above. Cell of origin subtype was determined according to previous descriptions8. U2932, RIVA, TMD8, HBL1, and SUDHL2 were maintained in RPMI-1640 media with 10% FBS and 1% penicillin/streptomycin. OCI-LY1, OCI-LY7 and OCI-LY10 were maintained in IMDM supplemented with 20% human serum and 1% penicillin/streptomycin. Cell lines were regularly tested for mycoplasma, and identity confirmation by Short Tandem Repeat at core facility of MD Anderson Cancer Center. Tetracyline-inducible expression of TCF4 was performed in the TMD8, HBL1, and SUDHL2 cell lines. Detailed methodology can be found in the supplementary methods.
ChIP-sequencing of TCF4
For inducible TCF4 expression, TMD8-TCF4, or SU-DHL-2-TCF4 cell lines were treated with doxycycline (60ng/ml) for 24 hours. For chromatin immuno-precipitation, five million cells were fixed with 1% formaldehyde for 10 min, quenched by addition of 125 mM Glycine for 5 min at RT, washed with ice-cold PBS then resuspended and incubated in ice-cold ChIP buffer (10mM Tris-HCl pH 8.0, 6.0 mM EDTA, 0.5% SDS and protease inhibitor) for 1hour. In the same time, 5μg of antibodies (TCF421) or control rabbit IgG (Cell Signaling; 2729) were allowed to bind to dynabeads Protein-G (Invitrogen; 10003D) in binding buffer (0.2% BSA, 0.1% Tween-20 in PBS) for 2 hours. Chromatin was sheared using Covaris M220. Sonicated lysates were diluted in dilution buffer (10mM Tris-HCl pH8, 140 mM NaCl, 1mM EDTA pH 8, 0.5 mM EGTA, 1% Triton X-100, and 0.1% Sodium Deoxycholate) and added to antibody bound Protein-G beads for immunoprecipitation overnight at 4ºC. Note; for ChIP normalization, spike in chromatin/antibody was added to sonicated lysates (53083/61686; Active Motif). Next day, bead-bound complexes were washed 5 times with RIPA buffer (1% NP40, 0.1% SDS, and 0.5% Sodium Deoxycholate in PBS), 2 times with LiCl buffer (10 mM Tris-HCl pH8, 250 mM LiCl, 0.5% NP40, 0.5% Sodium Deoxycholate, 1 mM EDTA), once with TE buffer pH 8.0 and finally resuspended in 50μl of TE buffer containing 20μg of proteinase K and RNase A (0.2 μg/μl). TE buffer, RNase A and proteinase K mixture was also added in total chromatin samples in parallel as input reference. Reverse cross-linking was performed at thermal cycler (4 hours 37°C, 4 hours 50°C, and overnight 65°C). DNA purification was performed with SPRIselect beads (Beckman Coulter; B23317) and further processed for library generation with KAPA HyperPrep kit (KK8502) according to the kit protocol.
Sequencing reads were aligned to the human genome (hg19) using BWA-Mem48, realigned around InDels using GATK49, sorted and deduplicated using Picard tools. Peaks were called in TCF4 ChIP samples compared to their input control using EaSeq with global thresholding. Peaks were annotated according to the transcription start site of the nearest RefSeq gene and filtered based upon FDR (<0.1), log2ratio of TCF4 ChIP vs. isotype control (≥2.0), peaks that overlapped between TMD8 and SUDHL2, and peaks corresponding to genes with differential expression between ABC-like DLBCL tumors with or without TCF4 DNA copy number gain. Peaks within 2kbp of the transcription start site were defined as ‘promoter’ peaks, those outside of the promoter region but within the coding region of the gene were defined as ‘intragenic’ peaks, and those outside of these regions but within 50kbp of the transcription start site were defined as distant ‘enhancer’ peaks. For visualization, files were converted to wiggle format and viewed using the Integrative Genomics Viewer55. The wiggle file for H3K27Ac ChIP-seq for CD20+ B-cells was downloaded directly from UCSC Genome Browser (https://genome.ucsc.edu/ENCODE/downloads.html). Significantly over-represented DNA sequence motifs (FDR<0.05) were identified in TCF4 ChIP-seq peaks compared to the reference genome (hg19) using CisFinder56 with the default settings. Motifs with 75% homology were collapsed to motif clusters.
For ChIP-PCR, chromatin immuno-precipitation was performed with BRD4 antibody (Bethyl, Cat No. A301-985A) following the protocol as described above for TCF4. Chromatin DNA was also purified from the input samples. The purified DNA was used to perform quantitative PCR using SYBR Green/ROX qPCR Master Mix (Applied Biosystem; 4309155). Percentage of input was quantified from the adjusted input Ct values and further used to determine ΔCt values for BRD4 or IgG ChIP. Primers used for ChIP-PCR have been listed in Table S10.
BET Inhibitors and Treatments
BRD4 inhibitors JQ1 and OTX015 were obtained from Selleck Chemicals. BRD4-PROTAC (ARV-771) was provided by Arvinas, Inc. (New Haven, CT). U2932 and RIVA cell lines were treated with indicated concentrations of BET-inhibitors (JQ1, OTX015) or BRD4-PROTAC (ARV771) for 24 hours before immunoblotting. For apoptosis analysis, U2932 and RIVA cell lines were seeded at 2.5 × 105 cells/ml and treated with ARV771 at indicated concentration for 48 hours. Cells were stained with Annexin V (Thermo Fisher; A35122)/To-PRO-3 and analyzed using flow cytometry (BD LSRFortessa) and FlowJo software. For gene expression analysis, cell lines (U2932 and RIVA) were un-treated or treated with ARV771 (50ng/ml) for 24 hours. Total RNA was extracted using All prep DNA/RNA kit (Qiagen; 80204) and RNA integrity was assessed using an Agilent-4200 TapeStation system. Libraries were generated using KAPA RNA HyperPrep kit with RiboErase (KK8560) according to the manufacturer’s instructions. Libraries were pooled and run on a single land of a HiSeq 4000 instrument at the MD Anderson Sequencing and Microarray Core Facility. Fastq files were first aligned to the GRCh37 assembly with GENCODE37lift37 annotations using STAR 2.6.0c, using a two-pass protocol with alignment parameters from the ENCODE long RNA-seq pipeline. The transcript-coordinate output files were then pre-processed with RSEM version 1.2.31’s convert-sam-for-rsem tool before quantifying with rsem-calculate-expression, assuming the data is from an unstranded paired-end library. Tximport version 1.6.0 was then used under R version 3.4.3 to read individual RSEM output files and aggregate to gene-level expressions based on the gene-transcript relationships in GENCODE27lift37’s Gene symbol metadata. DESeq2 version 1.18.1 was used to identify differentially expressed genes using a two-variable (Cell line and Treatment) analysis with default settings. Gene set enrichment analysis57 was performed using GenePattern and a list of all genes from RNA-seq ranked by the fold-change in expression following ARV771 treatment. The gene set consisted of all genes that showed significantly higher expression in ABC-like DLBCL tumors with TCF4 DNA copy number gain compared to those without, as shown in Figure 3.
Murine Xenograft Experiments
Reagents and antibodies
ARV-771 was kindly provided by Arvinas, Inc. (New Haven, CT) D-Luciferin (potassium salt) was obtained from Gold Biotechnology, Inc. (St Louis, MO). BD Matrigel Matrix High Concentration was obtained from BD Biosciences (Franklin Lakes, NJ) (Catalog number 354248).
Cell lines
Luciferase-expressing RIVA and U2932 cells were created by transducing cells with Luc-ZSGreen. pHIV-Luc-ZsGreen was a gift from Bryan Welm (Addgene plasmid # 39196). High GFP-expressing cells were isolated by flow sorting for GFP expression in the M. D. Anderson Flow Cytometry and Cellular Imaging Core Facility (FCCICF), a shared resource partially funded by NCI Cancer Center Support Grant P30CA16672.
In vivo studies
All animal studies were performed under a protocol approved by the IACUC at M.D. Anderson Cancer Center, an AAALAC-accredited institution. Five million RIVA or U2932 cells (mixed with Matrigel at a volume ratio of 1:1) were subcutaneously injected in the left flank of male athymic nude mice (nu/nu) (n = 8 per group). Tumor volume was calculated by the 1/2(length x width2) method. Treatment was initiated when the mean tumor volumes reached ∼150 mm3. Mice were treated with vehicle (10% [1:1 solutol: ethanol] and 90% D5-water, s.c. daily x 5 days per week) or ARV-771 (30 mg/kg, s.c., daily x 5 per week). The RIVA mouse model was treated for two weeks. Due to slower tumor growth, the U2932 mouse model was treated for three weeks. For bioluminescent imaging, mice were IP-injected with 100 μL of 75 mg/kg D-Luciferin potassium salt (reconstituted in 1X PBS and sterile-filtered through a 0.2 um filter) incubated for 5 minutes, anesthetized with isoflurane and imaged once per week utilizing a Xenogen IVIS-200 imaging system (PerkinElmer) to monitor disease status and treatment efficacy. One mouse from each cohort was euthanized after three weeks of treatment for biomarker analysis. Mice bearing tumors greater than 1500 mm3 were removed from study and humanely euthanized (carbon dioxide inhalation and cervical dislocation) according to the IACUC-approved protocol. Veterinarians and veterinary staff assisting in determining when euthanasia was required were blinded to the experimental conditions of the study. Tumor size was compared among cohorts by unpaired t-test. The survival of the mice is represented by a Kaplan Meier plot. Differences in survival were calculated by a Mantel-Cox log-rank test. P values less than 0.05 were considered significant.
COMPETING INTERESTS
The authors have no competing interests to declare.
AVAILABILITY OF DATA
The data produced in this study are available in the gene-expression omnibus (www.ncbi.nlm.nih.gov/geo/), accession number GSE119241. The SNP and gene expression microarray accessions for the previously published data are listed in Table S1. Raw next generation sequencing data will be provided upon reasonable request to the corresponding author and the completion of confidentiality non-disclosure and material transfer agreements.
AUTHOR CONTRIBUTIONS
NJ, KH, ST, WF, DK and OH performed experiments. NJ, KH, ST, KB and MRG analyzed data and wrote the manuscript. MJM, AB, TH, QD, DM, CP, AG, SR, JI, FG, SSN, JW, RED and KB analyzed or interpreted data. AA and CLL provided computational resources. EH, RK, KES, GJ, RR, RDG, AR, JV, ML, and TG provided samples or data. MRG conceived and supervised the study. All authors read and approved the manuscript.
ACKNOWLEDGEMENTS
This research was supported by the Nebraska Department of Health and Human Services (LB506 2016-16, M.R.G.), the Schweitzer Family Fund (J.W.), RO1 CA210250 (K.B.) and the MD Anderson Cancer Center NCI CORE Grant (P30 CA016672). Arvinas, Inc. kindly provided ARV-771 for the studies.