Pan-HDAC Inhibitors Restore PRDM1 Response to IL21 in CREBBP

Published OnlineFirst October 22, 2018; DOI: 10.1158/1078-0432.CCR-18-1153

Cancer Therapy: Preclinical Clinical Cancer Research Pan-HDAC Inhibitors Restore PRDM1 Response to IL21 in CREBBP-Mutated Follicular Lymphoma Fabienne Desmots1,2, Mikael€ Roussel1,2,Celine Pangault1,2, Francisco Llamas-Gutierrez3,Cedric Pastoret1,2, Eric Guiheneuf2,Jer ome^ Le Priol2, Valerie Camara-Clayette4, Gersende Caron1,2, Catherine Henry5, Marc-Antoine Belaud-Rotureau5, Pascal Godmer6, Thierry Lamy1,7, Fabrice Jardin8, Karin Tarte1,9, Vincent Ribrag4, and Thierry Fest1,2

Abstract

Purpose: Follicular lymphoma arises from a germinal cen- is enriched in the potent inducer of PRDM1 and IL21, highly ter B-cell proliferation supported by a bidirectional crosstalk produced by Tfhs. In follicular lymphoma carrying CREBBP with tumor microenvironment, in particular with follicular loss-of-function mutations, we found a lack of IL21-mediated helper T cells (Tfh). We explored the relation that exists PRDM1 response associated with an abnormal increased between the differentiation arrest of follicular lymphoma cells enrichment of the BCL6 protein repressor in PRDM1 gene. and loss-of-function of CREBBP acetyltransferase. Moreover, in these follicular lymphoma cells, pan-HDAC Experimental Design: The study used human primary cells inhibitor, vorinostat, restored their PRDM1 response to IL21 obtained from either follicular lymphoma tumors character- by lowering BCL6 bound to PRDM1. This finding was rein- ized for somatic mutations, or inflamed tonsils for normal forced by our exploration of patients with follicular lympho- germinal center B cells. Transcriptome and functional analyses ma treated with another pan-HDAC inhibitor. Patients were done to decipher the B- and T-cell crosstalk. Responses showed an increase of plasma cell identity genes, mainly were assessed by flow cytometry and molecular biology PRDM1 and XBP1, which underline the progression of follic- including ChIP-qPCR approaches. ular lymphoma B cells in the differentiation process. Results: Conversely to normal B cells, follicular lymphoma Conclusions: Our data uncover a new mechanism by which cells are unable to upregulate the transcription repressor, pan-HDAC inhibitors may act positively to treat patients with PRDM1, required for plasma cell differentiation. This defect follicular lymphoma through the induction of the expression occurs although the follicular lymphoma microenvironment of plasma cell genes.

the microenvironment including IL21. This latter is a potent Introduction þ inducer of PCs and is produced by GC CD4 helper T cells called Adaptive immune responses involve the formation of germinal follicular helper T cells (Tfh; ref. 1). Akin to most cytokines, IL21 centers (GC), which are specialized structures allowing the dif- activates mainly JAK3/STAT3 signaling pathways (2), which are ferentiation of high-affinity B cells into long-lived memory B cells known to play a highly selective role in PC differentiation (3) and and plasma cells (PC). PCs are generated through specific tran- apoptosis, depending on the context (4). In humans, blockade of scriptional programs influenced by numerous signals delivered by IL21 inhibits PC generation. Tfh interacting with IL21 receptor expressing GC B cells could thus impact B-cell destiny via the 1Universite de Rennes, Etablissement Francais¸ du Sang (EFS) de Bretagne, Inserm, MICMAC–UMR_S1236, Rennes, France. 2Laboratoire d'hematologie, delivery of IL21 signal (3, 5). Human PCs emerging through Centre Hospitalier Universitaire de Rennes, Rennes, France. 3Laboratoire IL21/pSTAT3 signaling enhancement has concomitantly upregu- d'Anatomie Pathologique, Centre Hospitalier de Rennes, Rennes, France. lated PRDM1/BLIMP1 and downregulated BCL6 gene expression 4Gustave Roussy Cancer Campus, DITEP, Villejuif, France. 5Laboratoire de (6, 7). PRDM1/BLIMP1 is considered as the master PC factor that Cytogen etique, Centre Hospitalier de Rennes, Rennes, France. 6Service antagonizes BCL6 which sustains the B-cell identity (8, 9). There- 7 d'Hematologie Clinique, CH Bretagne Atlantique, Vannes, France. Service fore, PRDM1/BLIMP1 and BCL6 mutually inhibit each other and d'Hematologie Clinique, Centre Hospitalier de Rennes, Rennes, France. 8Inserm 9 BCL6-related inhibition of PC generation is predominantly relat- U918, Centre Henri Becquerel, Rouen, France. Laboratoire d'Immunologie, PRDM1 Centre Hospitalier de Rennes, Rennes, France. ed to BCL6 binding at intron 3 of where BCL6 recruits MTA3, which acts as a corepressor (10). Moreover, this PRDM1 Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/). intronic region contains an enhancer bound by CREBBP (11). CREBBP acts as a transcriptional coactivator of many different Current address for E. Guiheneuf: Laboratoire d'Hematologie, CHU Amiens- transcription factors through its intrinsic acetylation function on Picardie, Amiens, France. histone but also on nonhistone proteins including BCL6. Indeed, Corresponding Author: Thierry Fest, Rennes 1 University, 2 avenue du Pr Leon CREBBP binds and acetylates BCL6, leading to the inactivation of Bernard, Rennes 35043, France. Phone: 332-9928-4127; Fax: 332-9928-4152; its transcriptional repressor function (12, 13). E-mail: [email protected] In follicular lymphoma, we demonstrated earlier that the doi: 10.1158/1078-0432.CCR-18-1153 tumor microenvironment is enriched for Tfhs that sustain neo- 2018 American Association for Cancer Research. plastic cells (14, 15). It is generally accepted in the field that

www.aacrjournals.org OF1

Desmots et al.

2.0 Array and strategy for raw data normalization and ﬁltering Translational Relevance is detailed in Supplementary Materials and Methods. In a number of phase I/II clinical trials, HDACi therapies have shown clinical responses, including complete remissions Gene expression analysis in previously multitreated patients with follicular lymphoma. The relative quantiﬁcation of gene expression was determined DDC Even if multiple biological effects have been described for using the 2 t method, then normalized to at least an internal these drugs, no effect on the master plasma cell regulator gene, control gene (ABL1, GAPDH, and/or HPRT1) and relative to a PRDM1, has been so far described in the context of the calibrator control sample corresponding to a mix of cDNA of follicular lymphoma. Indeed, our study revealed that follicular peripheral blood mononuclear cells (PBMC) from several healthy lymphomas with a CREBBP loss-of-function (concerns more donors. Statistical analysis using GraphPad Prism Software used than 50% of follicular lymphomas) were unable to upregulate Mann–Whitney test. PRDM1 expression despite the presence of IL21 in the tumor microenvironment, a potent inductor of PRDM1. In this Flow cytometry analysis, tissue immunostaining, Western context, we found that pan-HDACi can restore PRDM1 expres- blotting, and FISH analysis sion as well as other plasma cell genes, indicating a possible All antibodies and technical details concerning protein expres- reinitiation of follicular lymphoma B-cell differentiation. Our sion and FISH analyses are in Supplementary Materials and results highlight one effect of pan-HDACis to overcome fol- Methods, Supplementary Table S4, and Supplementary Figs. licular lymphomas' differentiation blockade. S3, S4, and S8.

Culture conditions and molecular analyses Cells were cultured for 2 hours before receiving indicated follicular lymphomas are tumors reflective of centrocytes that fail treatments for 24 hours, and then CD19/CD20 viable B cells to differentiate beyond the GC exit point. Purified follicular were collected and used for subsequent DNA and RNA extraction. lymphoma B cells compared with normal GC B cells does not Detailed procedures are given in Supplementary Materials and identify a radically opposed signature, but rather substantial Methods. modification of normal GC expression driving most likely by numerous genetic and epigenetic somatic modifications Somatic mutations assessment (14, 16–18). Both histone acetyl transferases, CREBBP and EP300, We performed the SureSelect targeted-capture strategy from are commonly mutated in follicular lymphoma and CREBBP loss- Agilent Technologies using a panel of 34 genes described previ- of-function affects preferentially H3K27 acetylation whose deple- ously (22) with subsequent paired-end sequencing on an Illumina tion leads to downregulation of genes involved in GC output HiSeq 1500 Platform (Illumina). For variants detection, SureCall (19). Interestingly, the expression of these genes can be restored software from Agilent technologies was used (see Supplementary after inhibition of HDAC3 (20). Materials and Methods, Supplementary Tables S2B and S3B). In this study, we assessed PRDM1 gene expression and regulation to achieve new insights on the differentiation blockade that Statistical analyses characterizes the follicular lymphoma. We showed that despite a Statistical analyses were performed with the GraphPad Prism functional capacity to activate the IL21/pSTAT3 signaling, non- software V5 (GraphPad Software) as indicated in Supplementary functional CREBBP follicular lymphoma cells were unable to Materials and Methods section. increase PRDM1 expression. In these follicular lymphomas, induction of PRDM1 in response to IL21 could be restored Results through the use of pan-HDAC inhibitors. Follicular lymphoma B cells are functionally unable to regulate genes involved in plasma cell differentiation Materials and Methods B-cell differentiation process is supported by the microenvi- Samples ronment through the delivery of soluble and membrane signals Subjects were recruited under Institutional Review Board (23, 24). Among them, CD40L and IL21 are major contributors to approval and informed consent process (French Minister autho- the transcriptional emergence of PC identity genes in committed B rization DC-2016-2565). An informed written consent was cells (3, 7, 25). To decipher specific CD40L- and IL21-targeted obtained from each subject or subject's guardian. The study was genes, freshly purified centroblasts (CB) were cultured for 3 hours conducted in accordance of the Declaration of Helsinki. Normal in presence or not of CD40L and IL21 before transcriptome GC-derived B and T cells were isolated from human tonsils and analysis were performed. We found a total of 4,932 genes differ- reactive lymph nodes. Follicular lymphoma tumors were entially expressed with 2,154 up- and 2,778 downregulated genes obtained from patients that underwent a surgical biopsy during between unstimulated and stimulated conditions (FDR < 0.05; a diagnosis procedure and from patients with refractory/relapsed Fig. 1A). In particular, PRDM1 expression is increased 3.5 times follicular lymphoma recruited during the phase I/II study based upon CD40L and IL21 stimulation (P ¼ 0.001; Fig. 1B). In on an oral pan-HDACi drug (21). parallel, using the same approach, we analyzed highly purified human centroblastic and centrocytic GC-derived B cells from Transcriptomic data tonsils and highly purified follicular lymphoma B cells. At first, For transcriptomic analysis, we used highly purified B- a total of 4,654 genes (FDR < 0.05) with 2,347 up- and 2,307 lymphocyte fractions that were sorted using combinations of downregulated genes were found differentially expressed between mAbs and FACSARIA (BD Biosciences) system. Extracted RNAs centroblasts (CB) and centrocytes (CC; Fig. 1C). We display list were hybridized on an Affymetrix Human Genome U133 Plus comparisons using Venn diagrams to find which subsets of genes

OF2 Clin Cancer Res; 2018 Clinical Cancer Research

HDACi and PRDM1 Expression in Follicular Lymphoma

Figure 1. Gene expression profiling revealed a functional impairment of the PRDM1 response in follicular lymphoma (FL) compared with normal centrocytes (CC). A, Freshly sorted normal CBs were cultured for 3 hours with or without CD40L þ IL21. Extracted RNAs were subsequently hybridized on Affymetrix U133þ2.0 microarrays, and expression of 10,000 probesets (PS) with the highest intensity was clustered using unsupervised hierarchical clustering (HCA). Differentially expressed genes were obtained for an FDR < 0.05 comparing stimulated versus unstimulated conditions (n ¼ 6). B, A new hierarchical clustering was performed only on a subset of genes known to be upregulated by either IL21 or CD40L pathways or during the plasma cell differentiation process. P values (FDR) were calculated by moderated paired t tests. C, CBs and CCs were sorted from tonsils based on the following markers: CD19posIgDnegCD38hiCD10pos plus CXCR4-positive for CBs only. Extracted RNAs were subsequently hybridized on Affymetrix U133þ2.0 microarrays, and the same clustering strategy as in A was performed. List of genes corresponding to genes differentially expressed first in CBs upon CD40L þ IL21 treatment and second, between CBs and CCs were compared using Venn diagrams, thus identifying 804 up- and 965 downregulated genes during CB/CC transition that could be linked to the CD40L þ IL21 stimulation. D, Follicular lymphoma B cells (n ¼ 10) were sorted on the basis of CD20hiCD44loCD38posIgDnegCD138neg definition. HCA was performed comparing CBs, CCs, and follicular lymphoma B cells for the 1,769 genes of the IL 21/CD40L signature (corresponding to 804 up- and 965 downregulated genes previously identified). Two PSs corresponding to the PRDM1 gene are highlighted, and on right of the figure are indicated genes described further in F. E, Table shows proportions of genes highly or weakly expressed in CB, CC, and follicular lymphoma for each box. F, Ingenuity pathway enrichment on seven different items in box1 compared with box2. G, PRDM1 gene expression (for Affymetrix PS 228964_at) on a 23-independent cohort of follicular lymphomas (follicular lymphoma B) compared with reactive lymph nodes (RLN-B). In this study (14), B cells were obtained by immunomagnetic sorting based on the CD19 expression for both follicular lymphomas and normal tissues. www.aacrjournals.org Clin Cancer Res; 2018 OF3

Desmots et al.

are involved in both CB/CC transition and CB response to IL21/ CD40L signaling. A total of 804 up- and 965 downregulated genes (total of 1,769 genes) were common to both lists (Fig. 1C) and major hallmark pathways associated with these genes were related to immune responses and cell proliferation, respectively (Sup- plementary Table S1). We next applied this 1,769-gene signature to compare CB, CC, and follicular lymphoma B-cell transcrip- tomes. Unsupervised hierarchical analysis clustered follicular lymphoma B cells separately from other cells with a particular and an opposite pattern of expression compared with CBs (Fig. 1D). We delimited two boxes of probe sets based on the clustering and found probe sets that positively expressed in Box1 were vastly (>95%) connected to CBs, whereas the Box2-positive probe sets were associated to follicular lymphoma B cells (>82%). In addition, Box2 contained 95.8% of the 804 genes upregulated in the 1,769-gene signature. Interestingly, in this analysis, the CCs presented somewhere an intermediate position, suggesting a polarized axis where follicular lymphoma cells occupy the most advanced position in the B-cell differentiation (Fig. 1E). To complete our characterization, we use Ingenuity Pathways Anal- ysis (IPA, Ingenuity Systems, www.ingenuity.com) and focused on previously described molecular pathways enriched in B cells and PCs (26). Globally, we confirmed that Box1 genes are related to CB signatures (cell cycle and FOXM1 transcription factor network), whereas Box2 presented plasmablast and early PC features (Fig. 1F). In this context, we noticed with interest the strong repression of PRDM1 in follicular lymphomas (highlighted in Fig. 1D). This finding was confirmed using an independent cohort of 23 follicular lymphomas previously explored and where þ total CD19 B cells were compared with B cells issued from reactive lymph nodes (Fig. 1G; ref. 14). IHC for BLIMP1 protein showed a totally negative staining in follicular lymphoma, whereas normal GCs showed very weak expression and tumor plasmo- cytoma, a strong positivity (Supplementary Fig. S1C). Our transcriptome findings in follicular lymphoma are in contradiction with the transcriptional network of normal B cells (27) and also with our previous data (28) on the hierarchical clustering, show- ing a transcriptional switch between CB and CC subtypes. Alto- gether, our data suggest the existence of a deregulation of the transcriptional BCL6/PRDM1 balance in follicular lymphomas. Indeed, follicular lymphomas maintain the expression of B-cell identity genes including BACH2 and BCL6 (PAX5 is missing in our 1,769-gene signature) besides low PRDM1 expression, whereas, on the other hand, these cells express IRF4 and XBP1, both tightly connected to the PC identity. By IPA Ingenuity Systems, we found that the Box2 was significantly enriched for the CD40L/CD40 pathway (Supplementary Fig. S1A). Gene set enrichment assays Figure 2. (GSEA) applied on the 1,769-gene signature identified a signif- PRDM1 is not induced by IL21 in follicular lymphoma (FL) B cells. PRDM1 (A)and icant enrichment in follicular lymphomas compared with CB for BCL6 (B) expression in control L3055 (five independent experiments) and in CD40L, IL21, and plasma cell upregulated genes signatures (Sup- fi n ¼ primary culture of puri ed follicular lymphoma B cells ( 18) stimulated with plementary Fig. S1B). Collectively, our results indicate that fol- IL21 for 24 hours and compared with the unstimulated condition (Unstim). Relative quantities of PRDM1 and BCL6 genes were calculated after licular lymphoma B cells are most likely blocked at a terminal normalization to internal control gene expressions and relative to an external B-cell differentiation step characterized by a low expression of calibrator used as control sample. Upon IL21, PRDM1 expression was significantly PRDM1, despite the presence of substantial CD40L and IL21 induced in control L3055, whereas no modification was observed in follicular microenvironment signaling. lymphoma. For the BCL6 expression, no changes were observed in control – L3055 or in follicular lymphoma. Statistical analysis using Mann Whitney test Impaired PRDM1 response after IL21 stimulation in follicular comparing globally all follicular lymphomas upon IL21 stimulation versus unstimulated conditions (see Supplementary Fig. S3C and S3D) were not lymphoma significant. C, PRDM1 and BCL6 expressions upon IL21 stimulation are correlated During the differentiation of CCs into PCs, the upregulation of and the majority of follicular lymphomas, mutated for CREBBP, clustered in the PRDM1 through the IL21 signaling was previously demonstrated lower values of expression. in normal and lymphoma B cells (4, 29). On the basis of these

OF4 Clin Cancer Res; 2018 Clinical Cancer Research

HDACi and PRDM1 Expression in Follicular Lymphoma

Figure 3. Investigations of the IL21/p-STAT3 signaling in follicular lymphomas. A, Percentage of Tfh from 15 follicular lymphomas (FL), 9 reactive lymph nodes (rLN), and 7 tonsils. B, Percentage of IL21-expressing Tfh from 5 follicular lymphomas and 6 tonsils among the total Tfh population; C, RMFI comparison for pSTAT3 in Tfh from 15 follicular lymphomas, 9 rLNs, and 7 tonsils. D, RMFI comparison for pSTAT3 in 20 follicular lymphoma B cells (BCL2-positive for tumor B cells and BCL2-negative for nonmalignant B cells) and GC B-cell counterparts from 9 rLNs and 9 tonsils. E, pSTAT3 RMFI correlation between BCL2-positive tumor B cells and BCL2-negative nonmalignant B cells. F, The relative expression of the pSTAT3 target gene BATF was increased after IL21 stimulation in all follicular lymphomas that we tested (n ¼ 14). G, Representative single or double IHC staining for PD1 and pSTAT3 onto parafﬁn-embedded tissues from one rLN (left) and a representative case of follicular lymphoma (case FL_6103; right). In the rLN, PD1-positive cells are polarized at the light zone of the GC and are associated with few pSTAT3-positive cells. In follicular lymphomas, PD1-positive cells are more frequent, show a diffuse distribution, and are associated with clusters of pSTAT3-positive cells. Unlike rLNs, in follicular lymphomas we visualized cells stained for both PD1 and pSTAT3 (stars) corresponding to Tfhs (, P < 0.05; , P < 0.01; , P < 0.001). www.aacrjournals.org Clin Cancer Res; 2018 OF5

Desmots et al.

data and for reasons of sample saving to complete our study on Altogether, our data showed that follicular lymphomas present an primary follicular lymphoma cells, we decided to focus our enhanced and functional IL21/pSTAT3 signaling. investigations solely on IL21 signaling. Highly purified follicular lymphoma B cells of 18 patients (clinical details are in Sup- Nonfunctional CREBBP follicular lymphomas increased BCL6 plementary Table S2A) were cultured for 24 hours in the binding to PRDM1 gene in response to IL21 presence or not of IL21 and cell viability was systematically Genetic alterations of CREBBP may alter its acetyltransferase monitored (Supplementary Fig. S3). As control, we used L3055 activity, which may abolish BCL6 acetylation and maintain its centroblastic cell line (named hereafter, control L3055), an repressor activity on PRDM1 (10, 12, 13). We next sought for EBV-negative Burkitt lymphoma cell line, phenotypically and somatic mutations in CREBBP-coding region included in our 34- functionally similar to the normal CBs (30) carrying besides the gene lymphopanel beside other chromatin-modifier genes rearranged MYC locus no additional alterations affecting BCL6 (EP300, EZH2, KMT2D, and MEF2B; ref. 34) used for capture- and PRDM1 loci, as shown by CGH array (Supplementary Fig. targeted deep-sequencing strategy. Fourteen follicular lympho- S2). Control L3055 significantly increased the PRDM1 expres- mas of 20 were mutated for CREBBP at least once (70%), includ- sion upon IL21 without effect on BCL6 (Fig.2AandB).Protein ing the 10 nonsynonymous variants (SNV) located in the HAT BLIMP1/PRDM1 was detectable by flow cytometry for control catalytic domain (histone acetyltransferase domain), 2 SNVs tonsil–derived GC B cells and by Western blot analysis for outside the HAT, and 2 frameshifts (fs) in the N-terminal region. L3055 cells after 24 hours of IL21 stimulation (Supplementary For 12 of these variants (HAT-located and fs), a nonfunctional Fig. S2). In contrast, the 18 follicular lymphomas did not CREBBP protein is encoded (Fig. 4A; Supplementary Table S2C; significantly modify their expression of PRDM1 and BCL6 (Fig. ref. 12). EP300 (another gene with a HAT domain) was mutated in 2A and B; Supplementary Fig. S3C and S3D). Spearman anal- two cases but outside HAT domain (Supplementary Table S2C). ysis to find a correlation between the expression of PRDM1 and Altogether, 12 of 20 follicular lymphomas presented a functional BCL6, showed that a large majority of follicular lymphomas did loss of CREBBP owing to genetic alterations. not respond for both genes after IL21 stimulation (Fig. 2C). To determine the binding capacity of BCL6 protein to intron These results are in line with our transcriptome analysis find- 3(INT3)ofPRDM1, we used a chromatin immunoprecipita- ings and confirm that most follicular lymphoma B cells exhibit tion (ChIP) approach followed by qPCR (see procedure details a defect in PRDM1 response. in Supplementary Fig. S5). Highly purified follicular lymphoma B cells and control L3055 were cultured for 24 hours in the Follicular lymphoma Tfhs and follicular lymphoma B cells presence or absence of IL21. BCL6 binding to PRDM1 was show a functional and increased IL21/pSTAT3 response decreased in L3055 upon IL21, and correlates with its increased We used total cell suspensions from follicular lymphoma expression due to a probable releasing BCL6-mediated tran- tumors to evaluate the functional capacities of B cells and Tfhs. scriptional repression (Fig. 4B). We then explored 10 follicular Control cell counterparts came from nonmalignant tonsils or lymphoma of 18 previously explored for PRDM1 expression reactive lymph nodes (rLN). The gating strategy of the flow (Fig. 2A), for which we had sufficient viable B cells for ChIP cytometry analysis defined a specific CD3/CD4/CxCR5/PD-1– assays. For two (FL_6108 and FL_5511) out of four wild-type positive cell population corresponding to Tfhs (Supplementary follicular lymphomas for CREBBP, the IL21-induced PRDM1 Fig. S4). As described previously, Tfhs increased in number in expression was associated with a decrease of BCL6 enrichment follicular lymphomas compared with rLNs (Fig. 3A; refs. 14, 31). at INT3 of PRDM1, thus behaving like control L3055. Six other Functional experiments on Tfhs showed an enhanced production follicular lymphomas, characterized by a positive BCL6 protein of IL21 after stimulation by PMA–ionomycin (Fig. 3B; ref. 32) as IHC staining and a loss-of-function variant of CREBBP, we well as a significant increase of the pSTAT3 response after 10 observed a strong enrichment of the BCL6 binding, except for minutes of IL21 (Fig. 3C). Follicular lymphoma B cells from 20 FL_5008. This finding correlated with the absence of IL21 patients showed after IL21 stimulation a significant increase in induction of PRDM1 gene expression and data indicate that pSTAT3 compared with controls. Interestingly, this tonic response in response to IL21 stimulation, the tumor cells operate an was detected in malignant and nonmalignant B cells, both dis- active silencing of the PRDM1 gene, possibly facilitating local criminated with BCL2 intracellular staining (Fig. 3D; Supplemen- recruitment of the BCL6 repressor (Fig. 4B). Globally, our data tary Fig. S4). In addition, both B-cell populations for the same suggest that a link exists between BCL6 binding, PRDM1 follicular lymphoma proportionally increased their pSTAT3 response to IL21, and the functional activity of CREBBP. There- expression (Fig. 3E). Moreover, 24 hours of IL21 stimulation led fore, we decided to explore further the binding of BCL6 protein to the expression of BATF, a specific pSTAT3-target (ref. 33; Fig. to INT3 of PRDM1 in response to IL21 with the use of a histone 3F). We completed our analysis by evaluating the expression of acetyltransferase inhibitor (HDACi). pSTAT3 by IHC in 7 follicular lymphomas and compared the results with 2 rLNs. In normal GC, B cells presented pSTAT3 Pan-HDAC inhibitor vorinostat restores PRDM1 response to staining mainly in medium and large B cells, localized within the IL21 in nonfunctional CREBBP follicular lymphomas light zone of the GC in the vicinity of PD1-positive T cells (Fig. 3G, We postulated that CREBBP-mutated follicular lymphomas left). In follicular lymphoma, this pattern of expression was lost might present a diminished acetylated form of BCL6 protein and pSTAT3-positive B cells were more numerous and formed conferring a potential oncogenic activity (12, 13). Thus, we clusters (Fig. 3G, right). Overall, we found a mean number of 40 hypothesized that by restoring acetylation state of histones and pSTAT3-positive cells per GC in follicular lymphoma (ranged proteins in follicular lymphoma B cells using an HDAC inhibitor, from 29 to 51) compared with a mean of 19 in rLNs. Unlike rLNs, we would be able to restore PRDM1 response to IL21. We used some follicular lymphoma Tfhs were positive for pSTAT3 in vorinostat, also known as suberanilohydroxamic acid (SAHA), a agreement with our above flow cytometry data (Fig. 3G, right). potent pan-HDACi, on 6 CREBBP-mutated follicular lymphomas

OF6 Clin Cancer Res; 2018 Clinical Cancer Research

HDACi and PRDM1 Expression in Follicular Lymphoma

Figure 4. CREBBP somatic mutations in follicular lymphoma (FL) and functional impact on PRDM1 response under IL21 stimulation. A, Schematic representation (based on Human Protein Database) of CREBBP and its key functional domains delimited with amino acid positions with distribution of frameshift (green diamonds) and missense mutations (brown diamonds) in 14 CREBBP-mutated follicular lymphomas. Conserved domains are ZF, Zinc Finger-TAZ type; KIX, coactivator CBP (CREB-binding protein and MYB interaction domain); BD, bromodomain; HAT, histone acetyl transferase domain; ZZ, zinc ﬁnger, ZZ type; NR, nuclear receptor. B, Besides control L3055, 10 follicular lymphomas were analyzed and classiﬁed in the table according to their HAT-CREBBP–mutated status (4 wild-type cases, plus control L3055, labeled WT, and 6 HAT-CREBBP–mutated cases). After 24 hours of culture with or without IL21 (100 ng/mL), cells were split for subsequent

RNA extraction or chromatin immunoprecipitation. Log2 values of IL21/Unstim fold change ratio (FC) of PRDM1 expression from Fig. 2C are indicated in this table with paralleled paired follicular lymphoma levels of BCL6 enrichment IL21/Unstim ratio (Log2) at the INT3 site of PRDM1 relative to control IgG enrichment used as control for each condition. This table also indicates the presence of (þ), absence of (), or undetermined (nd) expression of BCL6 by IHC (BCL6 prot. IHC column) in parallel with the BCL6 translocation with t(3q27) determined by FISH (3q27 pos. column). and on control L3055. Purified cells were analyzed after 24 hours cases (Fig. 5C and D). In conclusion, our observations could be of culture using three different conditions: IL21 alone, IL21 plus summarized schematically using a scale where SAHA addition vorinostat, and vorinostat alone. None of these conditions trig- to IL21 will reverse the abnormal BCL6/PRDM1 transcriptional gered cell death (Supplementary Fig. S7). In control L3055, gene expression equilibrium in follicular lymphomas by addition of SAHA to IL21 induced a significant increase of increasing PRDM1 while decreasing BCL6 expression (Fig. 5E). BLIMP1/PRDM1 mRNA and protein expression without modification of BCL6 binding to PRDM1 gene (Fig. 5; Supplementary Increased expression of PC-related genes in patients with Fig. S6). For 4 of 6 follicular lymphomas, PRDM1 expression follicular lymphoma treated with a new pan-HDACi increased under IL21 plus vorinostat compared with IL21 alone To determine whether HDACi therapy in follicular lymphoma and was associated with a BCL6 expression decrease (Fig. 5A may affect the expression of PC master genes, we analyzed 4 of 7 and B) leading overall to a significant positive effect on the patients with follicular lymphoma included in a multicenter PRDM1/BCL6 expression balance as shown for the control phase I/II study testing a new pan-HDACi in refractory/relapsed L3055 (P ¼ 0.041; Fig. 5D). In addition, we detected for all B-cell lymphoproliferative disease (21). Somatic mutation 6 follicular lymphomas a clear BCL6 occupancy decrease to the screening was performed on tumor DNA at diagnosis and when INT3 of PRDM1 under IL21 plus vorinostat (Fig. 5C). The patients reprogressed (Supplementary Table S3). FL_CAN patient vorinostat-only condition compared with IL21 alone did not still in complete remission 5 years after the inclusion (the drug significantly increase the PRDM1/BCL6 ratio (P ¼ 0.93; Fig. discontinued after 4 years) was CREBBP wild-type at diagnosis, 5D), whereas a decrease of BCL6 binding occurred in five of six but mutated in the HAT domain of EP300, which encoded a

www.aacrjournals.org Clin Cancer Res; 2018 OF7

Desmots et al.

Figure 5. Assessment of BCL6 protein occupancy at the INT3 of PRDM1 in follicular lymphoma (FL) B cells using the histone acetyl transferase inhibitor, SAHA, and impact on the PRDM1/BCL6 expression balance. Sorted follicular lymphoma B cells from 6 HAT/CREBBP–mutated patients and control L3055 were cultured in presence of IL21 with or without addition of SAHA, or with SAHA alone. For each follicular lymphoma, fold change ratio (FC) of the tested condition compared with unstimulated condition (Unstim) was calculated for PRDM1 (A)andBCL6 (B) expression, and for BCL6 enrichment (C). D, The PRDM1/BCL6 FC ratio was calculated comparing IL21 alone (blue) to either SAHA þ IL21 (red) or SAHA alone (green). A statistically signiﬁcant increase of PRDM1 relative to BCL6 was observed for the former condition but not for the latter one. L3055 data are shown, yellow triangle, for comparison. E, Globally the addition of SAHA to IL21 compared with IL21 alone induced an increase of PRDM1 expression concomitant with a decrease of BCL6 expression and a decrease of BCL6 recruitment at INT3 of PRDM1 modifying the BCL6-PRDM1 balance.

protein that, like CREBBP, sustains acetylation-mediated inacti- response to the drug (FL_CON), presented a clonal evolution in vation of BCL6 repressive activity (ref. 12; Fig. 6A). For patients their rebiopsy (Fig. 6A), probably leading to the therapeutic FL_CON and FL_DAD, we detected nonsynonymous variants resistance. FL_CAS patient who had an objective 28-month within, respectively, the HAT domain of EP300 or CREBBP before response to the drug before relapse was not mutated for CREBBP and after treatment (Fig. 6A). These two patients, one with 7 or EP300. However, he showed also a clonal evolution upon months of stable disease (FL_DAD) and the other without treatment and notably the loss of the CD79B mutation, which

OF8 Clin Cancer Res; 2018 Clinical Cancer Research

HDACi and PRDM1 Expression in Follicular Lymphoma

Figure 6. Mutation proﬁling and B-cell differentiation gene expression analysis of 4 patients with follicular lymphoma (FL) treated with a new HDACi. Patients with a refractory or multiple relapsing follicular lymphoma were enrolled in a phase I/II study (21) testing a new pan-HDACi inhibitor. Various responses to the drug were observed for the 4 patients analyzed and are indicated as follows: NR, nonresponder; SD, stable disease; CR, complete response; and OR, objective response. Samples were collected at diagnosis (Diag.) or after drug administration (Drug.T.), except for the patient who reached a CR. Molecular analyses were performed before and after treatment. A, R-graphic representation using "ggpubr" package, of clonal evolution following pan-HDACi administration compared with diagnosis respective to variant allelic frequency (VAF; VAF Diag. vs. VAF Drug T.) of major variants found. B and C, Results from RQ-PCR analysis of basal expression at diagnosis and after treatment for PRDM1, XBP1,andIRF4 (B, in green), for B-cell identity genes BCL6, BACH2,andPAX5 (B, in orange) and for XBP1s and ERN1 genes (C) relative to an external calibrator and normalized with control gene expressions. D, The expression ratio of each gene was calculated, comparing the levels of expression after drug treatment with those at diagnosis. Plasma cell identity gene ratios are positive compared with B-cell identity genes, except for BCL6. www.aacrjournals.org Clin Cancer Res; 2018 OF9

Desmots et al.

plays a role in antigenic stimulation in chronic active BCR sig- stopped in a step of differentiation characterized by their inability naling (Fig. 6A). to positively regulate the expression of PRDM1 in absence of any Our investigations on tumor RNA extracts showed globally an genetic alteration of this gene (44). In chronic lymphocytic increase in the expression of the 3 tested PC transcription factors, leukemia, Duckworth and colleagues described a transcriptional PRDM1, IRF4, and XBP1 in rebiopsies compared with the initial repression of PRDM1 upon IL21 owing to the loss of chromatin tissues, whereas B-cell identity factors remain stable except for active marks (45). In some aggressive lymphomas, silencing of BCL6 and BACH2 in patient FL_CON only (Fig. 6B). The expres- PRDM1 was related to DNA hypermethylation in regulatory sion of PRDM1 in FL_CON and FL_CAS increased during treat- regions of the gene (46). In our study, we investigated the ment, whereas FL_DAD showed stable expression but had a proximal promoter and intron 3 of the gene and found very low higher level of PRDM1 expression at the time of inclusion com- level of methylation, ruling out this mechanism (data not shown). pared with other patients. In addition, FL_DAD was the only one The IL21 signaling goes mostly through a potent induction which showed a marked increase in the expression of IRF4. The 3 of the STAT3 pathway involving the binding to a pSTAT3-IRF4 patients for whom the treatment failed showed a clear increase of consensus site of PRDM1 leading to the gene upregulation (2, 29). XBP1 expression. Interestingly, for patients with FL_DAD and Herein, we confirmed previous descriptions in follicular lympho- FL_CON, we observed an elevation of the expression of the active ma with the presence of a Tfh-enriched microenvironment (14), form of XBP1 and XBP1s, associated with a clear upregulation of an enhanced capacity of Tfhs to secrete IL21 (15), and the the unfolding protein response (UPR) sensor ERN1, a gene increased proportion of both follicular lymphoma B cells and encoding the endoplasmic reticulum kinase involved in XBP1 Tfhs to produce pSTAT3 in response to IL21 compared with their maturation (Fig. 6C; ref. 35). These mechanisms are essential normal counterparts. These findings may reflect the existence of a for PC differentiation (36). Overall, increased expressions of chronically active stimulation in follicular lymphoma microen- key PC-identity genes under pan-HDACi treatment suggest that vironment leading to a prompt pSTAT3 functional response. follicular lymphomas underwent some differentiation, but that Indeed, we found an enhanced expression of BATF, a target gene tumor cells still escape the drug (Fig. 5D). of pSTAT3, after IL21 stimulation and the presence of pSTAT3- positive cell clusters in tumor tissues. Both, BCL6 and BLIMP1 are known to negatively cross-regulate Discussion each other. Therefore, during B-cell differentiation, crossing the In 2006, two different groups (37, 38) described the presence of restriction point, thanks to a reverse transcriptional balance (i.e., PRDM1/BLIMP1 inactivation in 50% of non-GC DLBCL owing to PRDM1 goes up while BCL6 decreases), will be crucial to com- alterations on both alleles by either deletion or mutations. This plete the final cell commitment (6, 47). The role played by IL21 in loss of function is critical for the lymphomagenesis process giving this control has been little explored mainly because of the absence to BLIMP1 the role of a bona fide tumor suppressor. This inacti- of reliable B-cell lines. Our cell culture experiments on control vation is mutually exclusive with BCL6 alterations (39). However, L3055 showed that the IL21-mediated PRDM1 upregulation was they also identified few non-GC DLBCLs without PRDM1 genetic associated with a decrease of BCL6 occupancy on PRDM1. For alterations but with a lack of BLIMP1 protein, suggesting that primary cultured follicular lymphoma B cells, only two cases other mechanisms could be responsible for PRDM1/BLIMP1 presented similar results whereas the other eight showed a inactivation. In our study, we identified in follicular lymphoma decrease in their PRDM1 expression by IL21 and, in parallel, they B cells compared with normal counterparts a downregulation of increased the binding of BCL6 protein to PRDM1. Among them, 6 PRDM1 expression, whereas our screening for somatic mutations follicular lymphomas had a functional loss of CREBBP, which has rules out the existence of PRDM1 genetic alterations. Our tran- an impact on the BCL6 acetylation status. Indeed, in normal cells, scriptome analysis of normal GC B cells subjected to IL21/CD40L CREBBP binds and acetylates BCL6, which disrupts its ability to stimuli identified a specific signature representative of the Tfh- recruit histone deacetylases, thereby enhancing its capacity to delivery signal promoting terminal B-cell differentiation, that is, repress transcription of target genes, such as PRDM1 (12, 13). upregulation of PC identity genes including PRDM1 associated to We, therefore, speculate that HDACi treatments could lead to downregulation of B-cell identity genes (6, 7). In comparison, the accumulation of inactive acetylated BCL6, cell-cycle arrest, and IL21/CD40L signature in follicular lymphomas transcriptome is apoptosis in B-cell lymphoma cells (13). Our experiments with reversed with upregulation of some PC identity genes (e.g., IRF4 vorinostat on nonfunctional CREBBP follicular lymphoma cells and XBP1) and downregulation of PRDM1, whose gene expres- showed globally an increase in PRDM1 expression after IL21 sion is mandatory for PC maturation and maintenance (40, 41). exposure. This gain of PRDM1 response was associated to a Therefore, our observations that follicular lymphoma cells are low decrease of BCL6 protein occupancy on INT3 of PRDM1 and a PRDM1 expressers suggest a blockade of follicular lymphoma decrease in BCL6 gene expression leading indirectly to an cells maturation that might be due to the impossibility of tumor enhanced PRDM1 functional response through the PRDM1/BCL6 cells to express PRDM1/BLIMP1. balance, which drives the B-cell terminal differentiation (26). It is Studies of human and mouse models revealed that IL21 of interest to notice that wild-type follicular lymphoma cases for plays a crucial role in the development of B-cell immunoglobulin CREBBP maintained a positive PRDM1 response to IL21, whereas responses through the induction of PRDM1 expression (42, 43). 3 of 4 presented a deregulated BCL6 gene (3q27-positive) with Our functional experiments in control L3055 found a positive limited modifications of BCL6 binding to PRDM1 (Fig. 4). The IL21-mediated PRDM1 response while follicular lymphoma cells fourth case wild-type for CREBBP and without 3q27 abnormality showed low baseline expression of PRDM1 and a lack of induc- responded to IL21 like the control L3055, that is, a positive tion with IL21. This finding suggested that follicular lymphoma PRDM1 response to IL21 accompanied with a decrease of BCL6 cells take advantage of the GC's "fertile soil" to develop and enrichment at INT3 of PRDM1. Taken altogether these findings, initiate the terminal differentiation; however, the process is despite the lack of protein verification on follicular lymphoma

OF10 Clin Cancer Res; 2018 Clinical Cancer Research

HDACi and PRDM1 Expression in Follicular Lymphoma

cells due to material scarcity, we speculate that the decrease of gene and protein expression for PRDM1/Blimp1 and other key BCL6 binding to PRDM1 in response to IL21, in nonfunctional players in PC differentiation, thus providing the ultimate proof of CREBBP follicular lymphomas treated by vorinostat, is likely not the effect of these drugs in follicular lymphoma. Collectively, our due to the decrease of BCL6 gene expression. We suspect that data uncover a new mechanism by which pan-HDAC inhibitors additional modifications on the BCL6 protein complex could may act positively to treat patients with follicular lymphoma and have been induced by the vorinostat allowing the decrease of in particular, those with nonfunctional CREBBP. BCL6 enrichment at INT3 and thereby suppressing the repression on PRDM1. Disclosure of Potential Conflicts of Interest The loss of PRDM1 contributes to the lymphomagenesis by V. Ribrag reports receiving commercial research grants from ArgenX and blocking PC differentiation (39), knowing, however, that in Epizyme and speakers bureau honoraria from Servier, and is a consultant/ follicular lymphomas, the loss of CREBBP is not sufficient to advisory board member for Bristol-Myers Squibb, Epizyme, Gilead, Infinity, fl drive lymphomagenesis, but need the cooccurrence of BCL2 MSD, Nanostring, Pharmamar, and Roche. No potential con icts of interest were disclosed by the other authors. translocation (11). Our analysis of 4 massively pretreated patients with follicular lymphoma with a new pan-HDACi showed that the drug might increase PC identity genes expression within the Authors' Contributions tumor, including PRDM1, in agreement with our in vitro data. Conception and design: F. Desmots, T. Fest Development of methodology: F. Desmots, E. Guiheneuf, T. Fest The restoration of PC identity genes with a pan-HDACi is in line Acquisition of data (provided animals, acquired and managed patients, with the recent report of Jiang and colleagues who described broad provided facilities, etc.): M. Roussel, C. Pangault, C. Pastoret, E. Guiheneuf, effects of CREBBP on the transcriptional regulation of B cells (19) J. Le Priol, G. Caron, C. Henry, M.-A. Belaud-Rotureau, P. Godmer and here, in our study, specifically on a master gene of the B-cell Analysis and interpretation of data (e.g., statistical analysis, biostatistics, differentiation, that is, PRDM1. Pan-HDAC inhibitors may allow a computational analysis): F. Desmots, M. Roussel, C. Pangault, C. Pastoret, new step in follicular lymphoma cell differentiation with varying E. Guiheneuf, M.-A. Belaud-Rotureau, V. Ribrag, T. Fest fi Writing, review, and/or revision of the manuscript: F. Desmots, C. Pastoret, ef cacy depending on the presence of somatic abnormalities and E. Guiheneuf, T. Lamy, F. Jardin, K. Tarte, V. Ribrag, T. Fest clonal tumor diversity. Rebiopsied tissues presented an increase of Administrative, technical, or material support (i.e., reporting or organizing the BCL6 expression, which could be due to the reprogression data, constructing databases): F. Desmots, J. Le Priol, V. Camara-Clayette, status of the disease connected with the proliferation properties of T. Lamy, K. Tarte BCL6 rather than its induction by the drug. Despite this effect, the Study supervision: F. Desmots, T. Fest drug allowed a positive increase in the PRDM1/BCL6 balance. Others (histologic interpretation): F. Llamas-Gutierrez Others (assisted with experiments): V. Camara-Clayette Among the identified upregulated PC identity genes, we found XBP1 gene, which occupies a downstream position in the tran- Acknowledgments scriptional cascade that governs B-cell differentiation (26). PRDM1 This work was supported by an internal grant from the Hematology Labo- Indeed, /BLIMP-1 is known to regulate UPR components ratory, CHU de Rennes, France, and the Ligue Regionale contre le Cancer de ATF6 ERN1 like and that are required for full-length XBP1 expres- l'Ouest. The NGS experiments are part of the RELYSE project supported by an sion and the production of subsequent active spliced-form XBP1s NCI translational grant. Sequencing was performed in the Biogenouest Geno- (48, 49). Two of 3 patients had increased XBP1s and ERN1 mics/Human and Environmental Genomics core facility of Rennes (Biosit/ expressions with the pan-HDACi drug supporting the idea of a OSUR). Human samples were obtained from the processing of biological successful BLIMP1/PRDM1 protein restoration in these cells. samples through the Centre de Ressources Biologiques (CRB)-Sante of Rennes ERN1 (BB-0033-00056, http://www.crbsante-rennes.com). Cell sorting was per- Recently, Bujisic and colleagues described that impaired formed by flow cytometry facility, Biosit, University of Rennes 1 (Rennes, expression and XBP1 activation contribute to tumor growth in France). Part of this work was supported by the Carte d'identite des Tumeurs GC-type diffuse large B-cell lymphoma (50). (CIT) program (http://cit.ligue-cancer.net/index.php/en) from the Ligue Natio- HDACi drugs seem to be promising medications in largely nale Contre le Cancer. The research protocol was conducted under French legal pretreated follicular lymphomas as confirmed recently by Even guidelines and fulfilled the requirements of the local institutional ethics and colleagues, which showed a significant clinical activity in this committee. disease, including highly durable responses (51). In this context, The costs of publication of this article were defrayed in part by the payment of our study shows that monitoring effects of such drug by analyzing advertisement fi PRDM1, IRF4, XBP1 page charges. This article must therefore be hereby marked in speci c gene expressions, for example, , and accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ERN1, thanks to iterative fine-needle punctures to easily accessible tumors, might be of interest. Furthermore, additional studies will Received April 14, 2018; revised July 26, 2018; accepted October 17, 2018; also be needed to confirm that pan-HDAC inhibitors restore both published first October 22, 2018.

References 1. Chtanova T, Tangye SG, Newton R, Frank N, Hodge MR, Rolph MS, et al. 5. Crotty S. Follicular helper CD4 T cells (TFH). Annu Rev Immunol T follicular helper cells express a distinctive transcriptional proﬁle, reﬂect- 2011;29:621–63. ing their role as non-Th1/Th2 effector cells that provide help for B cells. 6. Diehl SA, Schmidlin H, Nagasawa M, van Haren SD, Kwakkenbos MJ, J Immunol 2004;173:68–78. Yasuda E, et al. STAT3-mediated up-regulation of BLIMP1 Is coordinated 2. Zeng R, Spolski R, Casas E, Zhu W, Levy DE, Leonard WJ. The molecular with BCL6 down-regulation to control human plasma cell differentiation. basis of IL-21-mediated proliferation. Blood 2007;109:4135–42. J Immunol 2008;180:4805–15. 3. Moens L, Tangye SG. Cytokine-mediated regulation of plasma cell gener- 7. Ding BB, Bi E, Chen H, Yu JJ, Ye BH. IL-21 and CD40L synergistically ation: IL-21 takes center stage. Front Immunol 2014;5:65. promote plasma cell differentiation through upregulation of Blimp-1 in 4. Ozaki K, Spolski R, Ettinger R, Kim HP, Wang G, Qi CF, et al. Regulation of B human B cells. J Immunol 2013;190:1827–36. cell differentiation and plasma cell generation by IL-21, a novel inducer of 8. Martins G, Calame K. Regulation and functions of Blimp-1 in T and B Blimp-1 and Bcl-6. J Immunol 2004;173:5361–71. lymphocytes. Annu Rev Immunol 2008;26:133–69.

www.aacrjournals.org Clin Cancer Res; 2018 OF11

Desmots et al.

9. Cubedo E, Maurin M, Jiang X, Lossos IS, Wright KL. PRDM1/Blimp1 31. Myklebust JH, Irish JM, Brody J, Czerwinski DK, Houot R, Kohrt HE, et al. downregulates expression of germinal center genes LMO2 and HGAL. High PD-1 expression and suppressed cytokine signaling distinguish T cells FEBS J 2011;278:3065–75. infiltrating follicular lymphoma tumors from peripheral T cells. Blood 10. Parekh S, Polo JM, Shaknovich R, Juszczynski P, Lev P, Ranuncolo SM, et al. 2013;121:1367–76. BCL6 programs lymphoma cells for survival and differentiation through 32. Ame-Thomas P, Hoeller S, Artchounin C, Misiak J, Braza MS, Jean R, et al. distinct biochemical mechanisms. Blood 2007;110:2067–74. CD10 delineates a subset of human IL-4 producing follicular helper T cells 11. Zhang J, Vlasevska S, Wells VA, Nataraj S, Holmes AB, Duval R, et al. The involved in the survival of follicular lymphoma B cells. Blood 2015; crebbp acetyltransferase is a haploinsufficient tumor suppressor in B cell 125:2381–5. lymphoma. Cancer Discov 2017;7:322–37. 33. Liao J, Humphrey SE, Poston S, Taparowsky EJ. Batf promotes growth arrest 12. Pasqualucci L, Dominguez-Sola D, Chiarenza A, Fabbri G, Grunn A, and terminal differentiation of mouse myeloid leukemia cells. Mol Cancer Trifonov V, et al. Inactivating mutations of acetyltransferase genes in B-cell Res 2011;9:350–63. lymphoma. Nature 2011;471:189–95. 34. Dubois S, Viailly PJ, Mareschal S, Bohers E, Bertrand P, Ruminy P, et al. Next 13. Bereshchenko OR, Gu W, Dalla-Favera R. Acetylation inactivates the generation sequencing in diffuse large B cell lymphoma highlights molec- transcriptional repressor BCL6. Nat Genet 2002;32:606–13. ular divergence and therapeutic opportunities: a LYSA Study. Clin Cancer 14. Pangault C, Ame-Thomas P, Ruminy P, Rossille D, Caron G, Baia M, et al. Res 2016;22:2919–28. Follicular lymphoma cell niche: identification of a preeminent IL-4- 35. Meier JA, Zhang F, Sanjana NE. GUIDES: sgRNA design for loss-of-function dependent T(FH)-B cell axis. Leukemia 2010;24:2080–9. screens. Nat Methods 2017;14:831–2. 15. Ame-Thomas P, Le Priol J, Yssel H, Caron G, Pangault C, Jean R, et al. 36. Iwakoshi NN, Lee AH, Vallabhajosyula P, Otipoby KL, Rajewsky K, Characterization of intratumoral follicular helper T cells in follicular Glimcher LH. Plasma cell differentiation and the unfolded protein lymphoma: role in the survival of malignant B cells. Leukemia 2012;26: response intersect at the transcription factor XBP-1. Nat Immunol 1053–63. 2003;4:321–9. 16. Kuppers R. Mechanisms of B-cell lymphoma pathogenesis. Nat Rev Cancer 37. Tam W, Gomez M, Chadburn A, Lee JW, Chan WC, Knowles DM. Muta- 2005;5:251–62. tional analysis of PRDM1 indicates a tumor-suppressor role in diffuse large 17. Kretzmer H, Bernhart SH, Wang W, Haake A, Weniger MA, Bergmann AK, B-cell lymphomas. Blood 2006;107:4090–100. et al. DNA methylome analysis in Burkitt and follicular lymphomas 38. Pasqualucci L, Compagno M, Houldsworth J, Monti S, Grunn A, Nandula identifies differentially methylated regions linked to somatic mutation SV, et al. Inactivation of the PRDM1/BLIMP1 gene in diffuse large B cell and transcriptional control. Nat Genet 2015;47:1316–25. lymphoma. J Exp Med 2006;203:311–7. 18. Koues OI, Kowalewski RA, Chang LW, Pyfrom SC, Schmidt JA, Luo H, et al. 39. Mandelbaum J, Bhagat G, Tang H, Mo T, Brahmachary M, Shen Q, et al. Enhancer sequence variants and transcription-factor deregulation syner- BLIMP1 is a tumor suppressor gene frequently disrupted in activated B cell- gize to construct pathogenic regulatory circuits in B-cell lymphoma. like diffuse large B cell lymphoma. Cancer Cell 2010;18:568–79. Immunity 2015;42:186–98. 40. Kallies A, Hasbold J, Fairfax K, Pridans C, Emslie D, McKenzie BS, et al. 19. Jiang Y, Ortega-Molina A, Geng H, Ying HY, Hatzi K, Parsa S, et al. CREBBP Initiation of plasma-cell differentiation is independent of the transcription inactivation promotes the development of HDAC3-dependent lympho- factor Blimp-1. Immunity 2007;26:555–66. mas. Cancer Discov 2017;7:38–53. 41. Jourdan M, Caraux A, Caron G, Robert N, Fiol G, Reme T, et al. 20. Green MR, Kihira S, Liu CL, Nair RV, Salari R, Gentles AJ, et al. Mutations in Characterization of a transitional preplasmablast population in the early follicular lymphoma progenitors are associated with suppressed process of human B cell to plasma cell differentiation. J Immunol antigen presentation. Proc Natl Acad Sci U S A 2015;112:E1116–25. 2011;187:3931–41. 21. Morschhauser F, Terriou L, Coiffier B, Bachy E, Varga A, Kloos I, et al. Phase 42. Jourdan M, Caraux A, De Vos J, Fiol G, Larroque M, Cognot C, et al. 1 study of the oral histone deacetylase inhibitor abexinostat in patients An in vitro model of differentiation of memory B cells into plasmablasts with Hodgkin lymphoma, non-Hodgkin lymphoma, or chronic lympho- and plasma cells including detailed phenotypic and molecular character- cytic leukaemia. Invest New Drugs 2015;33:423–31. ization. Blood 2009;114:5173–81. 22. Bohers E, Viailly PJ, Dubois S, Bertrand P, Maingonnat C, Mareschal S, et al. 43. Marasco E, Farroni C, Cascioli S, Marcellini V, Scarsella M, Giorda E, et al. Somatic mutations of cell-free circulating DNA detected by next-generation B-cell activation with CD40L or CpG measures the function of B-cell sequencing reflect the genetic changes in both germinal center B-cell-like subsets and identifies specific defects in immunodeficient patients. Eur J and activated B-cell-like diffuse large B-cell lymphomas at the time of Immunol 2017;47:131–43. diagnosis. Haematologica 2015;100:e280–4. 44. de Jong D, Fest T. The microenvironment in follicular lymphoma. Best 23. Zhang Y, Garcia-Ibanez L, Toellner KM. Regulation of germinal center B-cell Pract Res Clin Haematol 2011;24:135–46. differentiation. Immunol Rev 2016;270:8–19. 45. Duckworth A, Glenn M, Slupsky JR, Packham G, Kalakonda N. Variable 24. Shulman Z, Gitlin AD, Weinstein JS, Lainez B, Esplugues E, Flavell RA, et al. induction of PRDM1 and differentiation in chronic lymphocytic leukemia Dynamic signaling by T follicular helper cells during germinal center B cell is associated with anergy. Blood 2014;123:3277–85. selection. Science 2014;345:1058–62. 46. Zhang YW, Zhang J, Li J, Zhu JF, Yang YL, Zhou LL, et al. Methylation 25. Le Gallou S, Caron G, Delaloy C, Rossille D, Tarte K, Fest T. IL-2 require- contributes to imbalance of PRDM1alpha/PRDM1bbeta expression in ment for human plasma cell generation: coupling differentiation and diffuse large B-cell lymphoma. Leuk Lymphoma 2015;56:2429–38. proliferation by enhancing MAPK-ERK signaling. J Immunol 2012; 47. Calame KL, Lin KI, Tunyaplin C. Regulatory mechanisms that determine 189:161–73. the development and function of plasma cells. Annu Rev Immunol 2003; 26. Kassambara A, Reme T, Jourdan M, Fest T, Hose D, Tarte K, et al. Geno- 21:205–30. micScape: an easy-to-use web tool for gene expression data analysis. 48. Tellier J, Shi W, Minnich M, Liao Y, Crawford S, Smyth GK, et al. Blimp-1 Application to investigate the molecular events in the differentiation of controls plasma cell function through the regulation of immunoglobulin B cells into plasma cells. PLoS Comput Biol 2015;11:e1004077. secretion and the unfolded protein response. Nat Immunol 2016;17: 27. Nutt SL, Hodgkin PD, Tarlinton DM, Corcoran LM. The generation of 323–30. antibody-secreting plasma cells. Nat Rev Immunol 2015;15:160–71. 49. Yoshida H, Matsui T, Yamamoto A, Okada T, Mori K. XBP1 mRNA is 28. Caron G, Le Gallou S, Lamy T, Tarte K, Fest T. CXCR4 expression func- induced by ATF6 and spliced by IRE1 in response to ER stress to produce a tionally discriminates centroblasts versus centrocytes within human ger- highly active transcription factor. Cell 2001;107:881–91. minal center B cells. J Immunol 2009;182:7595–602. 50. Bujisic B, De Gassart A, Tallant R, Demaria O, Zaffalon L, Chelbi S, et al. 29. Kwon H, Thierry-Mieg D, Thierry-Mieg J, Kim HP, Oh J, Tunyaplin C, et al. Impairment of both IRE1 expression andXBP1activationisahallmark Analysis of interleukin-21-induced Prdm1 gene regulation reveals func- of GCB DLBCL and contributes to tumor growth. Blood 2017;129: tional cooperation of STAT3 and IRF4 transcription factors. Immunity 2420–8. 2009;31:941–52. 51. Evens AM, Balasubramanian S, Vose JM, Harb W, Gordon LI, Langdon R, 30. Choe J, Li L, Zhang X, Gregory CD, Choi YS. Distinct role of follicular et al. A phase I/II multicenter, open-label study of the oral histone dendritic cells and T cells in the proliferation, differentiation, and apo- deacetylase inhibitor abexinostat in relapsed/refractory lymphoma. Clin ptosis of a centroblast cell line, L3055. J Immunol 2000;164:56–63. Cancer Res 2016;22:1059–66.

OF12 Clin Cancer Res; 2018 Clinical Cancer Research

Pan-HDAC Inhibitors Restore PRDM1 Response to IL21 in CREBBP-Mutated Follicular Lymphoma

Fabienne Desmots, Mikaël Roussel, Céline Pangault, et al.

Clin Cancer Res Published OnlineFirst October 22, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-18-1153

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2018/10/20/1078-0432.CCR-18-1153.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2018/12/07/1078-0432.CCR-18-1153. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.