In chronic lymphocytic leukemia (CLL), NOTCH1 mutations have been associated with clinical resistance to the anti-CD20 rituximab, although the mechanisms behind this peculiar behavior remain to be clarified. In a wide CLL series (n=692), we demonstrated that CLL cells from NOTCH1-mutated cases (87/692) were characterized by lower CD20 expression and lower relative lysis induced by anti-CD20 exposure in vitro. Consistently, CD20 expression by CLL cells was upregulated in vitro by γ-secretase inhibitors or NOTCH1-specific small interfering RNA and the stable transfection of a mutated (c.7541-7542delCT) NOTCH1 intracellular domain (NICD-mut) into CLL-like cells resulted in a strong downregulation of both CD20 protein and transcript. By using these NICD-mut transfectants, we investigated protein interactions of RBPJ, a transcription factor acting either as activator or repressor of NOTCH1 pathway when respectively bound to NICD or histone deacetylases (HDACs). Compared with controls, NICD-mut transfectants had RBPJ preferentially complexed to NICD and showed higher levels of HDACs interacting with the promoter of the CD20 gene. Finally, treatment with the HDAC inhibitor valproic acid upregulated CD20 in both NICD-mut transfectants and primary CLL cells. In conclusion, NOTCH1 mutations are associated with low CD20 levels in CLL and are responsible for a dysregulation of HDAC-mediated epigenetic repression of CD20 expression.
Chronic lymphocytic leukemia (CLL) is a heterogeneous disease with highly variable clinical courses and survivals ranging from months to decades. In particular, a subset of CLL patients is known to experience a progressive symptomatic disease poorly responsive to the common immuno-chemotherapeutic regimens.1, 2 A fraction of these high-risk CLL, overall accounting for 5%–10% of cases, can be identified by screening for TP53 mutation/deletion,1, 2 whereas an additional fraction of cases has been recently shown to bear mutations involving the NOTCH1, SF3B1 and BIRC3 genes. Overall, alterations of these genes occur in ~20% of CLL patients at diagnosis and have significant correlations with survival in consecutive series from independent institutions.3, 4, 5, 6, 7
Mutations of NOTCH1 are found in ~10% of CLL cases at diagnosis, with frequency increasing in advanced disease phases, in chemorefractory patients and during transformation to Richter syndrome.3, 4, 5, 7, 8 Moreover, NOTCH1 mutations are enriched in CLL patient subgroups defined by trisomy 12 and an unmutated IGHV gene status.9, 10 NOTCH1 encodes for a transmembrane receptor acting as a ligand-activated transcription factor.11, 12 In particular, NOTCH1 signaling initiates when the ligand, from either the JAGGED or DELTA families, binds to the receptor and induces successive proteolytic cleavages, resulting in the release and nuclear translocation of the NOTCH1 intracellular domain (NICD). In the nucleus, the NICD becomes part of an activation complex along with the transcription factor RBPJ, which leads to the de-repression/activation of specific target genes, including genes of the HES family.13, 14, 15, 16, 17, 18, 19, 20 At variance with normal B cells, CLL cells constitutively express the NOTCH1 receptor as well as its ligands JAGGED1 and JAGGED2, suggesting autocrine/paracrine loops for NOTCH1 signaling activation.21 In CLL, virtually all NOTCH1 mutations are frameshift or nonsense events clustering within exon 34, including a highly recurrent c.7541-7542delCT frameshift deletion, represented in 80% of cases.3, 4, 10 These mutations result in the truncation of the C-PEST regulatory domain of the protein and the subsequent impaired degradation of the NICD,3, 4, 22, 23, 24 which in turn determines to an intense and sustained activation of the NOTCH1 pathway.25
Recently, the presence of NOTCH1 mutations has been associated with a relative resistance to anti-CD20 immunotherapy in a prospective clinical study comparing the effectiveness of the fludarabine plus cyclophosphamide regimen vs the fludarabine plus cyclophosphamide plus rituximab regimen,26 although the biological mechanisms underlying the differential activity of rituximab in relation to NOTCH1 mutational status is still to be elucidated.
Materials and methods
Primary cells from CLL patients and healthy donors
The study was approved by the Internal Review Board of the Aviano Centro di Riferimento Oncologico (Approval n. IRB-05-2010) and included peripheral blood samples from 692 patients with CLL.27 Informed consent was obtained in accordance with the Declaration of Helsinki. CLL cases were characterized for IGHV mutational status, the main cytogenetic abnormalities, CD38, CD49d and ZAP70 expression, as described.28
Primary CLL cells and normal B cells from healthy donors (n=3) were obtained from peripheral blood samples by Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) density gradient centrifugation and used either directly or cryopreserved until use. All studies were performed on highly purified cells (>95% pure) as results of negative selection by immunomagnetic beads when required.29 In-vitro studies were performed in CLL cells from NOTCH1-mutated (NOTCH1-mut) cases with relevant NOTCH1 mutational burden, that is, >25% of total DNA, or in NOTCH1 wild-type (NOTCH1-wt) cases, as control.
CD20 expression was evaluated by flow cytometry at the Clinical and Experimental Onco-Hematology Unit (CRO, Aviano, Italy) in the 692 CLL cases entering this study, as part of the routine diagnostic procedures for CLL assessment. In particular, 495 cases were evaluated by a fluorescein isothiocyanate-conjugated anti-CD20 antibody, whereas in the remaining 197 cases a phycoerythrin-cyanine7-conjugated antibody was employed (clone L27, in both cases, BD Biosciences, Milan, Italy), owing to a modification of the flow cytometry diagnostic panel. For CD20 expression analyses, these two cohorts were kept separated. All experiments were performed on FACSCanto II (BD Biosciences).28, 29
NOTCH1 mutational status
The presence of c.7541-7542delCT NOTCH1 mutation was investigated by amplification refractory mutation system PCR, as described.3, 8, 10 The load of c.7541-7542delCT NOTCH1 mutation was evaluated by next-generation sequencing (NGS) using a MiSeq sequencer (Illumina, San Diego, CA, USA), with a ~1000 × coverage fold.
The presence of NOTCH1 mutations other than the c.7541-7542delCT was investigated by Sanger sequencing in the entire NOTCH1 PEST domain, as reported.30 The mutational load was roughly determined (~50%, 25%–50%, ~25% and <25% of mutated DNA) by visual inspection of sequence electropherograms, as reported.31
CLL cells from selected NOTCH1-mutated cases were sorted according to CD20 expression by using the PE-conjugated anti-CD20 antibody (BD Biosciences). The CD20low or CD20high fractions were selected below the 25th percentile or above the 75th percentile of CD20 expression, respectively. After complement-dependent cytotoxicity (CDC) assay, CLL cells from selected NOTCH1-mutated cases were sorted according to 7-aminoactinomycin (BD Biosciences) expression. Viable cell fraction was identified as 7-aminoactinomycin negative. Sorting was performed using a FACSAriaIII cell sorter (BD Biosciences), as described.29
NICD plasmids and transfection
NICD plasmids were engineered cloning the NICD coding sequence in a pcDNA3.1-NT-GFP-TOPO vector (Life Technologies, Monza, Italy). The c.7541-7542delCT mutation (NICD-mut) or c.5304G>A (NICD-null) mutation were inserted with the Quikchange II XL Mutagenesis kit (Agilent, Milan, Italy). MEC-1 cells were transfected with the Amaxa Nucleofector (Lonza, Basel, Switzerland).
Primary CLL cells were transfected with small interfering RNA (siRNA) for NOTCH1 (TriFECTa, RNAi kit, IDT, Leuven, Belgium) using the Amaxa Nucleofector, as reported.32 NOTCH1 protein expression was evaluated by flow cytometry using the phycoerythrin-conjugated anti-NOTCH1 antibody (clone MHN1-519, BD Biosciences).
Nuclear extracts were obtained as reported.33 Co-immunoprecipitation was performed using anti-RBPJ (clone ab25949, Abcam, Cambridge, UK) and isotype (Millipore, Milan, Italy) antibodies. Western blotting was performed using anti-NOTCH1 (D1E11, CST-Cell Signaling Technology, Leiden, The Netherlands), anti-histone deacetylase 1 (HDAC1) (10E2, Abcam), anti-HDAC2 (HDAC2-62, Abcam) and anti-RBPJ (D10A4, CST-Cell Signaling Technology) antibodies. Anti-ERK 1/2 (BD Biosciences) and anti-BRG1 (Santa Cruz Biotechnology, Heidelberg, Germany) were used as loading controls for cytoplasmic and nuclear lysates.
Chromatin immunoprecipitation assay
Chromatin immunoprecipitation (ChIP) assays were performed with SimpleChIP enzymatic Chromatin IP kit (CST-Cell Signaling Technology), according to standard manufacturer’s protocol, using anti-HDAC1 (10E2, Abcam), anti-HDAC2 (HDAC2-62, Abcam), anti-Hystone H3 (kit provided) or control isotype (kit provided) antibodies. Qualitative PCR amplification of MS4A1 promoter was performed as reported.34 Quantification of MS4A1 promoter DNA was determined by quantitative real time-PCR.
Further details regarding the methods and the statistical approaches are provided as Supplementary Information.
NOTCH1 mutational status and NOTCH1 protein expression in CLL
The presence of the c.7541-7542delCT NOTCH1 mutation was investigated by amplification refractory mutation system PCR in 692 CLL cases. With this approach, the c.7541-7542delCT was detected in 81 cases (Supplementary Table S1). Additional six cases with a NOTCH1 mutation other than the c.7541-7542delCT were detected by Sanger sequencing (Supplementary Table S1). Overall considered, NOTCH1-mut cases represented ~12% (that is, 87/692 cases) of the cohort, in keeping with previous studies.3, 4, 5 A quantitative detection of the c.7541-7542delCT was performed by NGS. As shown in Supplementary Table S2, the NOTCH1 mutational load ranged from 1 to 50% of total DNA, in agreement with the heterozygous nature of NOTCH1 mutations and with its subclonal representation in some instances.3, 4, 5
NOTCH1 protein expression was evaluated by western blotting in NOTCH1-mut cases, chosen among those with high mutational load (that is, >25% of NOTCH1-mutated DNA) and, for comparison, in NOTCH1-wt CLL. In keeping with the presence of the c.7541-7542delCT that generates truncated protein with impaired degradation,35 NOTCH1-mut cases showed high transmembrane NOTCH1 and NICD levels, both with molecular weights consistent with the truncation of the NOTCH1-mutated protein (Supplementary Figure S1).4, 21, 25 Conversely, NOTCH1-wt CLL, although expressing discrete amount of transmembrane NOTCH1 in some instances, usually expressed less NICD protein than NOTCH1-mut cases (Supplementary Figure S1).4, 21, 25
Correlation between CD20 expression and NOTCH1 mutational status in CLL
CD20 expression was investigated by flow cytometry using either a fluorescein isothiocyanate- or a phycoerythrin-cyanine7 antibody (Supplementary Table S1) and separately analyzed (Supplementary Figures S2a and S3a). In the cohort of 495 cases (60 NOTCH1-mut) in which CD20 expression was evaluated by the fluorescein isothiocyanate-conjugated antibody, CD20 levels were generally lower in the CLL component than in the normal non-neoplastic residual B-cell counterpart (Supplementary Figure S2a), as reported.27 Moreover, when CLL cases were stratified according to the classification of the main cytogenetic aberrations,36 variable CD20 levels were found, the highest levels being detected in trisomy 12 CLL (Supplementary Figure S2a).37 When the CD20 expression was evaluated with respect to NOTCH1 mutational status, NOTCH1-mut CLL expressed lower mean fluorescence intensity (MFI) values than NOTCH1-wt cases in both trisomy 12 CLL (mean MFI in 20 NOTCH1-mut cases=1893±196; mean MFI in 69 NOTCH1-wt cases=7051±819; P<0.0001) and non-trisomy 12 CLL (mean MFI in 40 NOTCH1-mut cases=1858±203; mean MFI in 366 NOTCH1-wt cases=2 426±112; P=0.017; Figure 1a and Supplementary Figure S2b).
Superimposable results were obtained in the remaining 197 CLL (27 NOTCH1-mut and 170 NOTCH1-wt cases), in which the CD20 expression was evaluated with a phycoerythrin-cyanine7 antibody (Supplementary Table S1), both in trisomy 12 CLL (mean MFI in 6 NOTCH1-mut cases=12 926±3676; mean MFI in 17 NOTCH1-wt cases=28 216±5228; P=0.027) and non-trisomy 12 CLL (mean MFI in 21 NOTCH1-mut cases=10 207±1 310; mean MFI in 153 NOTCH1-wt cases=15 208±1578; P=0.017, Supplementary Figures S3a and b).
In keeping with flow cytometry results, transcript levels of MS4A1, the gene encoding for CD20,38 as evaluated in 275 cases (46 NOTCH1-mut), were lower in NOTCH1-mut than in NOTCH1-wt cases both in the trisomy 12 (P=0.006) and in the non-trisomy 12 (P=0.019) CLL categories (Figure 1b).
To corroborate the correlation between CD20 expression and NOTCH1 mutations, we performed cell-sorting experiments to isolate the extreme CD20low and CD20high subpopulations in five CLL cases with different NOTCH1 mutational load (Supplementary Figure S4a), as determined by NGS, that is, 3% (CLL#406), 8% (CLL#34), 27% (CLL#171), 35% (CLL#243) and 41% (CLL#266) of total DNA. As shown by NGS re-sequencing of the separated subpopulations, CD20low sorted cells always had a relative enrichment in the NOTCH1 mutational burden when compared with the CD20high counterpart, that is, 9% vs 1% (CLL#406), 14% vs 3% (CLL#34), 32% vs 15% (CLL#171), 38% vs 32% (CLL#243) and 48% vs 39% (CLL#266). Consistently, the amount of MS4A1 transcripts was always significantly lower in the CD20low than in the CD20high subpopulation (Supplementary Figure S4b).
NOTCH1 mutational status and susceptibility to anti-CD20 in CLL
Then we investigated whether NOTCH1 mutational status could effectively influence susceptibility to anti-CD20 immunotherapy. To evaluate the capability of rituximab to kill in-vitro CLL cells bearing or not NOTCH1 mutations, CDC assay was performed using purified CLL cells from nine NOTCH1-mut and nine NOTCH1-wt cases. NOTCH1-mut CLL cells showed significantly lower relative lysis induced by rituximab than NOTCH1-wt CLL cells (mean % of relative lysis=2.5±0.8 vs 26.3±8.9, P=0.021) and the killing capacity of rituximab directly correlated with CD20 levels (Figure 1c).
We further investigated the correlation between NOTCH1 mutational status and susceptibility to rituximab by evaluating in three NOTCH1-mut cases the enrichment of NOTCH1 mutational burden after CDC assay on rituximab and subsequent cell sorting of the residual viable cell population. The NOTCH1 mutational burden, as detected by NGS, resulted higher in the post CDC-sorted viable cells than in the pre-CDC-unsorted counterpart in all the three tested cases (Supplementary Figure S4c). Consistently, the amount of MS4A1 transcripts, as detected by quantitative real time-PCR, were lower in the viable cell populations than in the pre-CDC-unsorted counterparts (Supplementary Figure S4c).
We also evaluated the capability of the alternative anti-CD20 antibody ofatumumab to kill in-vitro CLL cells from nine NOTCH1-mut and nine NOTCH1-wt cases. Although the killing capacity of ofatumumab resulted generally higher than that of rituximab, NOTCH1-mut CLL cells showed significantly lower relative lysis than NOTCH1-wt CLL cells (mean % of relative lysis=30.6±8.5 vs 60.6±5.8, P=0.011), again consistently with CD20 expression levels (Figure 1d).
NOTCH1 signaling and CD20 expression in CLL
To evaluate whether NOTCH1 signaling could influence CD20 expression in primary CLL cases, CLL cells from five NOTCH1-mut and six NOTCH1-wt cases were treated at different time points with the γ-secretase inhibitor (GSI) L-685,458, able to block the proteolytic generation of NICD.21 On GSI treatment, NOTCH1 signaling was consistently impaired, as defined by a reduction of HES1 expression (at 6 h) in both NOTCH1-wt and NOTCH1-mut CLL, although decreases were lower in the NOTCH1-mut category (P=0.005), according to the presence of higher levels of NICD in the latter cases (Supplementary Figures S5a and S1). More important, both MS4A1 transcripts (at 6 h) and CD20 expression levels (at 24 h) were significantly upregulated by GSI in NOTCH1-wt and, to a lesser extent, in NOTCH1-mut cases (Supplementary Figure S5b). No effect on CD20 expression was observed in purified normal B cells from healthy donors exposed in vitro to GSI, in keeping with the notion of a lack of NOTCH1 expression in these cells (not shown).21
To further confirm the association between NOTCH1 signaling and CD20 expression, CLL cells from six NOTCH1-mut and five NOTCH1-wt cases were transiently transfected with siRNA for NOTCH1. In both NOTCH1-mut and NOTCH1-wt cases, siRNA transfection effectively reduced NOTCH1 transcript at 6 h (P=0.001, not shown) and protein at 24 h (NOTCH1-mut cases, mean MFI=538±119 vs 184±32, P=0.011; NOTCH1-wt cases, mean MFI=524±64 vs 204±17, P=0.003). Consistently, CD20 expression result augmented both at transcript level (at 6 h, NOTCH1-mut, P=0.034; NOTCH1-wt, P=0.012, not shown) and protein level (at 24 h, NOTCH1-mut cases, mean MFI=2685±887 vs 3035±916, P=0.001; NOTCH1-wt cases, mean MFI=1707±434 vs 1923±434, P=0.003, Supplementary Figure S5c).
Establishment of an in-vitro model of mutated NICD-transfected CLL-like cells
To investigate the mechanism(s) through which NOTCH1 mutations may affect CD20 expression in CLL, we established an in-vitro model of NICD-transfected cells by taking advantage of the CLL-like MEC-1 cell line. MEC-1 cells, constitutively expressing a wt NOTCH1 form, were stably transfected with vectors encoding for the following: (i) a modified NICD with the c.7541-7542delCT (NICD-mut) and (ii) a modified NICD with a nonsense mutation inserted after the beginning of the coding sequence, as a null control (NICD-null). NICD-mut cells showed higher constitutive NOTCH1 protein levels than NICD-null cells (Figure 2a). Consistently, HES1 and HES5 transcript levels were higher in NICD-mut than in NICD-null cells (Supplementary Figure S6a).
When CD20 expression was tested, NICD-mut cells showed constitutive lower CD20 expression at both protein and transcript level than NICD-null cells (Figures 2a and b and Supplementary Figure S6b) and, consistently, lower relative lysis induced by rituximab and ofatumumab by CDC assay (P=0.043 and P=0.025, respectively, Figure 2c). Moreover, on GSI treatment, CD20 protein and transcript expression was significantly upregulated in both NICD-null cells and NICD-mut cells (Figure 2d and not shown).
According to these validations, we assumed the NICD-mut cells as in-vitro model of NOTCH1-mut CLL, in which the increased NICD accumulation, owing to a decreased degradation of truncated form,15 is mimicked by the enforced expression of an exogenously transfected mutated NICD.
Immunoprecipitation of the RBPJ transcription factor in NICD transfectants
When released by proteolytic cleavages and translocated into the nucleus on activation of the NOTCH1 pathway, NICD interacts with the RBPJ transcription factor and converts its function from repressor to activator of gene transcription.13, 15, 35 In fact, NICD is able to displace RBPJ from a HDAC-containing repression complex, thus forming, with RBPJ itself and other co-activators, the major gene transcriptional activation complex of the NOTCH1 pathway.13, 15, 35
To evaluate whether NICD accumulation, as it occurs on NOTCH1 mutations, could alter the balancing of the two functions of RBPJ, that is, transcriptional activator (complexed with NICD) or transcriptional repressor (complexed with HDACs),13, 15 we performed co-immunoprecipitation experiments aimed at investigating the alternative presence of NICD or HDACs (namely HDAC1 and HDAC2) bound to RBPJ in NICD transfectants. As shown in Figure 3a, co-immunoprecipitation experiments revealed that NICD-mut cells had higher levels of NICD bound with RBPJ than NICD-null cells. On the contrary, NICD-mut cells showed lower levels of HDAC1 or HDAC2 co-immunoprecipitated with RBPJ than NICD-null cells (Figure 3a). Notably, no difference was found by comparing NICD transfectants regarding the levels of immunoprecipitated RBPJ, and the nuclear and cytoplasmic levels of RBPJ, HDAC1 and HDAC2, as evidenced by control western blotting experiments (Supplementary Figures S7a–c). Consistently, comparable constitutive HDAC1/HDAC2 expression levels were found in NOTCH1-mut vs NOTCH1-wt primary CLL (Supplementary Figure S8).
The unbalancing of the transcriptional activation/repression equilibrium of RBPJ turned in favor of the activation of NOTCH1 signaling detected in NICD-mut cells was also in keeping with the higher HES1 and HES5 transcript levels detected in these cells (Supplementary Figure S6a).
HDAC-mediated ChIP in NICD transfectants
Previous studies identified epigenetic silencing of CD20 expression via HDACs as a mechanism conferring resistance to rituximab in lymphomas.34, 39, 40 To evaluate whether the preferential interaction of RBPJ with NICD could result in higher levels of HDAC1/HDAC2 available for the transcriptional repression of MS4A1,13, 15 ChIP assays were performed on nuclear lysates from NICD transfectants. As shown in Figure 3b, higher levels of DNA corresponding to the MS4A1 promoter were found in HDAC1 and HDAC2 chromatin immunoprecipitates from NICD-mut compared with NICD-null cells. Of note, a higher involvement of HDAC2 with respect to HDAC1 was evidenced ChIP experiments, in keeping with the higher levels of HDAC2 expressed by NICD transfectants (Supplementary Figure S7c). On the other hand, lower levels of DNA corresponding to the HES1 promoter were found by ChIP of NICD-mut cells compared with NICD-null cells (not shown).
These results suggest that higher NICD levels, as occurring in NICD-mut cells, may cause a NICD-dependent dislodgement of RBPJ from the HDAC-containing repression complexes. This phenomenon is associated with an increased availability of HDACs to repress transcription of the MS4A1 gene.
HDAC inhibition and CD20 expression
To further evaluate whether the higher levels of HDACs bound to the MS4A1 promoter could effectively affect CD20 expression, NICD-transfected cells were treated with the HDAC inhibitor valproic acid (VPA) for 48 h. In both NICD-mut and NICD-null cells, VPA treatment was able to significantly increase MS4A1 transcript levels (NICD-mut, mean fold increase=1.7, P=0.001; NICD-null, mean fold increase=1.5, P=0.003, Supplementary Figure S9a) and CD20 protein expression (NICD-mut, mean fold increase=1.3, P=0.041; NICD-null, mean fold increase=1.4, P=0.029; Figure 4a and Supplementary Figure S9b).
Similar results were obtained by treating with VPA primary CLL cells of seven NOTCH1-mut and six NOTCH1-wt cases. In both categories, VPA treatment was able to significantly increase MS4A1 transcripts (NOTCH1-mut, mean fold increase=1.5, P=0.05; NOTCH1-wt, mean fold increase=1.8, P=0.02; Supplementary Figure S9c) and CD20 protein (NOTCH1-mut, mean fold increase=1.3, P=0.05; NOTCH1-wt, mean fold increase=1.3, P=0.005; Figure 4b and Supplementary Figure S9d). These increments were not associated with significant increases of relative lysis by in-vitro CDC assays (not shown).
The fludarabine plus cyclophosphamide plus rituximab immuno-chemotherapy combination still represents the frontline regimen for treatment of patients in good physical conditions.1, 2 In particular, the addition of rituximab to the fludarabine plus cyclophosphamide combination has been definitely proved to improve the clinical outcome of CLL patients, despite the relative low levels of CD20 usually expressed on the surface of CLL cells.26, 27 Recently, however, it has been clearly demonstrated that such a benefit does not include patients affected by CLL-bearing NOTCH1 mutations,26, 41 although the reason for this different clinical behavior remains to be elucidated.
In the present study, we demonstrated that NOTCH1 mutations identify a CLL subset characterized by particularly low levels of CD20, both in non-trisomy 12 CLL and in the trisomy 12 CLL category, which usually has relatively higher CD20 levels and a higher frequency of NOTCH1 mutations.9, 10, 37 Conversely, Stilgenbauer et al.26 did not find any difference in CD20 expression between NOTCH1-mut and NOTCH1-wt CLL, although in this study CD20 levels were checked exclusively by flow cytometry in a minority of cases. Here, the lower CD20 expression by NOTCH1-mut cases was corroborated by the parallel finding of lower MS4A1 transcript levels. Moreover, in cell-sorting experiments of CLL cases with different NOTCH1 mutation levels, higher percentages of NOTCH1-mutated DNA were found in the sorted CD20low component compared with the CD20high counterpart. Finally, the dramatic downregulation of CD20 expression levels obtained by stably transfecting the CLL-like MEC-1 cells with a mutated NICD definitely confirmed this inverse correlation.
The low CD20 expression by NOTCH1-mut CLL cells is consistent with their lower sensitivity to rituximab and ofatumumab exposure in vitro, as shown here, in agreement with previous reports.42 Results of the present study also indicate that the residual CLL cells surviving on CDC assay with rituximab usually expressed lower CD20 levels and a greater NOTCH1 mutational load. In keeping, NOTCH1 mutations have been demonstrated to impact on rituximab sensitivity of CLL patients also when present at subclonal level.26, 41, 43
These data may also suggest that in CLL, the constitutive expression of NOTCH1, in its mutated configuration but also in the wt form,21 could be related with the generally lower CD20 levels observed in neoplastic vs normal B cells, in which NOTCH1 is not expressed at all.21 In keeping, we demonstrated here that GSI treatment in vitro was able to substantially augment CD20 expression both in NOTCH1-wt and NOTCH1-mut CLL cells, although in the latter the accumulation of NICD due to truncating mutations makes these cells relatively less susceptible to NOTCH1 signaling perturbation. As theoretically GSI may have off-target genes,44 NOTCH1 was also inhibited by specific siRNA. Again, transfection with siRNA increased CD20 expression both in NOTCH1-wt and NOTCH1-mut CLL cells.35
In humans, the balance of histone acetylation/deacetylation, respectively, induced by histone acetyl transferases and HDACs, represents one of the main epigenetic mechanisms of modification of chromatin conformation and regulation of gene expression.45, 46 In particular, the transcriptional activity owing to the triggering of the NOTCH1 pathway is known to be greatly sensitive to chromatin modifications and histone rearrangements.35 In this context, the main effector of the NOTCH1 pathway at nuclear level is a DNA-binding protein named RBPJ.13, 15, 35 This protein, in association with NICD and other co-activators, forms an activation complex that is essential for NICD-dependent transcription and target gene expression. Such an activation complex is degraded via NICD phosphorylation and its subsequent ubiquitination, with these molecular reactions requiring an intact C-terminal PEST region of the NICD protein.13, 15, 35 The specific degradation of NICD results in the dissociation among RBPJ and the other co-activators. In the absence of NICD, RBPJ is free to associate with specific co-repressors, which in turn recruit HDAC1 and HDAC2; this newly obtained repression complex represses NOTCH1 signaling.13, 15, 35. A simplified scheme of these multi-protein interactions is reported in Figure 5a.
Results of this study suggest that NOTCH1 with C-terminal truncations, as those determined by the c.7541-7542delCT, may influence the epigenetic downregulation of CD20 by HDACs allegedly via an impaired ubiquitination and degradation of the truncated NICD. In fact, as defined by co-immunoprecipitation experiments, in the condition of NICD accumulation due to the c.7541-7542delCT, RBPJ showed a preferential binding to NICD, in the context of the activation complex, rather than to HDACs, in the context of the repression complex. In NICD-mut cells, in turn, HDACs were mainly associated to the MS4A1 promoter, as defined by ChIP experiments. A necessary prerequisite is the persistence of the activation complex due to the lack of degradation of the truncated NICD (Figure 5b).13, 15, 35 In keeping, the rare NOTCH1-mut CLL carrying truncating mutations other than the c.7541-7542delCT (six cases in our cohort) were all characterized by low CD20 levels, comparable with those of NOTCH1-mut CLL carrying the c.7541-7542delCT. Conversely, three cases (not included in this cohort) carrying NOTCH1 missense mutations (for example, p.G2292R, p.V2214M and p.T2484M) expressed CD20 levels comparable with those of NOTCH1-wt CLL (FP, personal communication).
To restore epigenetic regulation, a wide range of compounds inhibiting HDAC functionality have been identified, some of them employed in anticancer therapies.45, 46 In addition, HDAC inhibitors are known to augment the cytotoxic activity of rituximab by increasing CD20 expression in lymphoma cells.34, 39, 40 In this study, treatment with the HDAC inhibitor VPA was capable to upregulate both MS4A1 transcript and CD20 protein either in NICD-transfected cells or in primary CLL cells from NOTCH1-mut and NOTCH1-wt cases.
In conclusion, we provided evidence that truncating NOTCH1 mutations in CLL are associated with low CD20 expression, and with a relative resistance to anti-CD20 immunotherapy in vitro. The low CD20 expression in NOTCH1-mut CLL can be ascribed to a NOTCH1 mutation-driven epigenetic dysregulation of a transcriptional repression mechanism involving HDACs. Clinically, drugs interfering with the NOTCH1 pathway and/or inhibiting HDACs might have a role to increase CD20 expression in vivo, thus overcoming the relative resistance of NOTCH1-mut CLL to rituximab-containing therapies.
Hallek M, Cheson BD, Catovsky D, Caligaris-Cappio F, Dighiero G, Dohner H et al. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working Group 1996 guidelines. Blood 2008; 111: 5446–5456.
Hallek M . Chronic lymphocytic leukemia: 2013 update on diagnosis, risk stratification and treatment. Am J Hematol 2013; 88: 803–816.
Fabbri G, Rasi S, Rossi D, Trifonov V, Khiabanian H, Ma J et al. Analysis of the chronic lymphocytic leukemia coding genome: role of NOTCH1 mutational activation. J Exp Med 2011; 208: 1389–1401.
Puente XS, Pinyol M, Quesada V, Conde L, Ordonez GR, Villamor N et al. Whole-genome sequencing identifies recurrent mutations in chronic lymphocytic leukaemia. Nature 2011; 475: 101–105.
Rossi D, Rasi S, Spina V, Bruscaggin A, Monti S, Ciardullo C et al. Integrated mutational and cytogenetic analysis identifies new prognostic subgroups in chronic lymphocytic leukemia. Blood 2013; 121: 1403–1412.
Quesada V, Conde L, Villamor N, Ordonez GR, Jares P, Bassaganyas L et al. Exome sequencing identifies recurrent mutations of the splicing factor SF3B1 gene in chronic lymphocytic leukemia. Nat Genet 2012; 44: 47–52.
Wang L, Lawrence MS, Wan Y, Stojanov P, Sougnez C, Stevenson K et al. SF3B1 and other novel cancer genes in chronic lymphocytic leukemia. N Engl J Med 2011; 365: 2497–2506.
Del Poeta G, Dal Bo M, Del Principe MI, Pozzo F, Rossi FM, Zucchetto A et al. Clinical significance of c.7544-7545 delCT NOTCH1 mutation in chronic lymphocytic leukaemia. Br J Haematol 2013; 160: 415–418.
Del Giudice I, Rossi D, Chiaretti S, Marinelli M, Tavolaro S, Gabrielli S et al. NOTCH1 mutations in +12 chronic lymphocytic leukemia (CLL) confer an unfavorable prognosis, induce a distinctive transcriptional profiling and refine the intermediate prognosis of +12 CLL. Haematologica 2012; 97: 437–441.
Rossi D, Rasi S, Fabbri G, Spina V, Fangazio M, Forconi F et al. Mutations of NOTCH1 are an independent predictor of survival in chronic lymphocytic leukemia. Blood 2012; 119: 521–529.
Lobry C, Oh P, Aifantis I . Oncogenic and tumor suppressor functions of Notch in cancer: it’s NOTCH what you think. J Exp Med 2011; 208: 1931–1935.
Yuan JS, Kousis PC, Suliman S, Visan I, Guidos CJ . Functions of notch signaling in the immune system: consensus and controversies. Annu Rev Immunol 2010; 28: 343–365.
Castel D, Mourikis P, Bartels SJ, Brinkman AB, Tajbakhsh S, Stunnenberg HG . Dynamic binding of RBPJ is determined by Notch signaling status. Genes Dev 2013; 27: 1059–1071.
Davis RL, Turner DL . Vertebrate hairy and enhancer of split related proteins: transcriptional repressors regulating cellular differentiation and embryonic patterning. Oncogene 2001; 20: 8342–8357.
Hsieh JJ, Hayward SD . Masking of the CBF1/RBPJ kappa transcriptional repression domain by Epstein-Barr virus EBNA2. Science 1995; 268: 560–563.
Iso T, Kedes L, Hamamori Y . HES and HERP families: multiple effectors of the Notch signaling pathway. J Cell Physiol 2003; 194: 237–255.
Jarriault S, Brou C, Logeat F, Schroeter EH, Kopan R, Israel A . Signalling downstream of activated mammalian Notch. Nature 1995; 377: 355–358.
Kato H, Taniguchi Y, Kurooka H, Minoguchi S, Sakai T, Nomura-Okazaki S et al. Involvement of RBP-J in biological functions of mouse Notch1 and its derivatives. Development 1997; 124: 4133–4141.
Klinakis A, Szabolcs M, Politi K, Kiaris H, Artavanis-Tsakonas S, Efstratiadis A . Myc is a Notch1 transcriptional target and a requisite for Notch1-induced mammary tumorigenesis in mice. Proc Natl Acad Sci USA 2006; 103: 9262–9267.
Weng AP, Millholland JM, Yashiro-Ohtani Y, Arcangeli ML, Lau A, Wai C et al. c-Myc is an important direct target of Notch1 in T-cell acute lymphoblastic leukemia/lymphoma. Genes Dev 2006; 20: 2096–2109.
Rosati E, Sabatini R, Rampino G, Tabilio A, Di IM, Fettucciari K et al. Constitutively activated Notch signaling is involved in survival and apoptosis resistance of B-CLL cells. Blood 2009; 113: 856–865.
Paganin M, Ferrando A . Molecular pathogenesis and targeted therapies for NOTCH1-induced T-cell acute lymphoblastic leukemia. Blood Rev 2011; 25: 83–90.
Weng AP, Ferrando AA, Lee W, Morris JP, Silverman LB, Sanchez-Irizarry C et al. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science 2004; 306: 269–271.
Sportoletti P, Baldoni S, Cavalli L, Del PB, Bonifacio E, Ciurnelli R et al. NOTCH1 PEST domain mutation is an adverse prognostic factor in B-CLL. Br J Haematol 2010; 151: 404–406.
Arruga F, Gizdic B, Serra S, Vaisitti T, Ciardullo C, Coscia M et al. Functional impact of NOTCH1 mutations in chronic lymphocytic leukemia. Leukemia 2014; 28: 1060–1070.
Stilgenbauer S, Schnaiter A, Paschka P, Zenz T, Rossi M, Dohner K et al. Gene mutations and treatment outcome in chronic lymphocytic leukemia: results from the CLL8 trial. Blood 2014; 123: 3247–3254.
Matutes E, Owusu-Ankomah K, Morilla R, Garcia MJ, Houlihan A, Que TH et al. The immunological profile of B-cell disorders and proposal of a scoring system for the diagnosis of CLL. Leukemia 1994; 8: 1640–1645.
Gattei V, Bulian P, Del Principe MI, Zucchetto A, Maurillo L, Buccisano F et al. Relevance of CD49d protein expression as overall survival and progressive disease prognosticator in chronic lymphocytic leukemia. Blood 2008; 111: 865–873.
Zucchetto A, Vaisitti T, Benedetti D, Tissino E, Bertagnolo V, Rossi D et al. The CD49d/CD29 complex is physically and functionally associated with CD38 in B-cell chronic lymphocytic leukemia cells. Leukemia 2012; 26: 1301–1312.
Rossi D, Spina V, Bomben R, Rasi S, Dal-Bo M, Bruscaggin A et al. Association between molecular lesions and specific B-cell receptor subsets in chronic lymphocytic leukemia. Blood 2013; 121: 4902–4905.
Balatti V, Bottoni A, Palamarchuk A, Alder H, Rassenti LZ, Kipps TJ et al. NOTCH1 mutations in CLL associated with trisomy 12. Blood 2012; 119: 329–331.
Bomben R, Gobessi S, Dal BM, Volinia S, Marconi D, Tissino E et al. The miR-17 approximately 92 family regulates the response to Toll-like receptor 9 triggering of CLL cells with unmutated IGHV genes. Leukemia 2012; 26: 1584–1593.
Saborit-Villarroya I, Vaisitti T, Rossi D, D'Arena G, Gaidano G, Malavasi F et al. E2A is a transcriptional regulator of CD38 expression in chronic lymphocytic leukemia. Leukemia 2011; 25: 479–488.
Sugimoto T, Tomita A, Hiraga J, Shimada K, Kiyoi H, Kinoshita T et al. Escape mechanisms from antibody therapy to lymphoma cells: downregulation of CD20 mRNA by recruitment of the HDAC complex and not by DNA methylation. Biochem Biophys Res Commun 2009; 390: 48–53.
Bray SJ . Notch signalling: a simple pathway becomes complex. Nat Rev Mol Cell Biol 2006; 7: 678–689.
Dohner H, Stilgenbauer S, Benner A, Leupolt E, Krober A, Bullinger L et al. Genomic aberrations and survival in chronic lymphocytic leukemia. N Engl J Med 2000; 343: 1910–1916.
Tam CS, Otero-Palacios J, Abruzzo LV, Jorgensen JL, Ferrajoli A, Wierda WG et al. Chronic lymphocytic leukaemia CD20 expression is dependent on the genetic subtype: a study of quantitative flow cytometry and fluorescent in-situ hybridization in 510 patients. Br J Haematol 2008; 141: 36–40.
Tedder TF, Streuli M, Schlossman SF, Saito H . Isolation and structure of a cDNA encoding the B1 (CD20) cell-surface antigen of human B lymphocytes. Proc Natl Acad Sci USA 1988; 85: 208–212.
Hiraga J, Tomita A, Sugimoto T, Shimada K, Ito M, Nakamura S et al. Down-regulation of CD20 expression in B-cell lymphoma cells after treatment with rituximab-containing combination chemotherapies: its prevalence and clinical significance. Blood 2009; 113: 4885–4893.
Shimizu R, Kikuchi J, Wada T, Ozawa K, Kano Y, Furukawa Y . HDAC inhibitors augment cytotoxic activity of rituximab by upregulating CD20 expression on lymphoma cells. Leukemia 2010; 24: 1760–1768.
Bo MD, Del Principe MI, Pozzo F, Ragusa D, Bulian P, Rossi D et al. NOTCH1 mutations identify a chronic lymphocytic leukemia patient subset with worse prognosis in the setting of a rituximab-based induction and consolidation treatment. Ann Hematol 2014; 93: 1765–1774.
Golay J, Lazzari M, Facchinetti V, Bernasconi S, Borleri G, Barbui T et al. CD20 levels determine the in vitro susceptibility to rituximab and complement of B-cell chronic lymphocytic leukemia: further regulation by CD55 and CD59. Blood 2001; 98: 3383–3389.
Sportoletti P, Baldoni S, Del PB, Aureli P, Dorillo E, Ruggeri L et al. A revised NOTCH1 mutation frequency still impacts survival while the allele burden predicts early progression in chronic lymphocytic leukemia. Leukemia 2014; 28: 436–439.
Andersson ER, Lendahl U . Therapeutic modulation of Notch signalling—are we there yet? Nat Rev Drug Discov 2014; 13: 357–378.
Minucci S, Pelicci PG . Histone deacetylase inhibitors and the promise of epigenetic (and more) treatments for cancer. Nat Rev Cancer 2006; 6: 38–51.
Ropero S, Esteller M . The role of histone deacetylases (HDACs) in human cancer. Mol Oncol 2007; 1: 19–25.
This study was supported in part by the Associazione Italiana Ricerca Cancro (AIRC), Investigator Grant IG-13227; Progetto Ricerca Finalizzata I.R.C.C.S. n. RF-2009-1469205, n. RF-2010-2307262, Progetto Giovani Ricercatori n. GR-2009-1475467, n. GR-2010-2317594, n. GR-2011-02347441, n. GR-2011-02346826, Ministero della Salute, Rome, Italy; Fondazione Cariplo (grant 2012-0689); Associazione Italiana contro le Leucemie, linfomi e mielomi (AIL), Venezia Section, Pramaggiore Group, Italy; Fondazione per la Vita di Pordenone, Italy; Ricerca Scientifica Applicata, Regione Friuli Venezia Giulia (‘Linfonet’ Project), Trieste, Italy; and ‘5x1000 Intramural Program’, Centro di Riferimento Oncologico, Aviano, Italy. FA is supported by a Beat-Leukemia fellowship.
FP contributed to write the manuscript, analyzed the data and performed the research. TB performed the research. FA, PB, PM, ET, BG, FMR, RB, AZ, DB and MD contributed to perform the research. GDA, AC, FZ, GP, DR, GG., GDP and SD provided well-characterized biological samples and contributed to write the manuscript. VG and MDB designed the study, interpreted data and wrote the manuscript.
The authors declare no conflict of interest.
Supplementary Information accompanies this paper on the Leukemia website
About this article
Cite this article
Pozzo, F., Bittolo, T., Arruga, F. et al. NOTCH1 mutations associate with low CD20 level in chronic lymphocytic leukemia: evidence for a NOTCH1 mutation-driven epigenetic dysregulation. Leukemia 30, 182–189 (2016). https://doi.org/10.1038/leu.2015.182
Cold Spring Harbor Perspectives in Medicine (2021)
Cancer Management and Research (2021)
An Updated Perspective on Current Prognostic and Predictive Biomarkers in Chronic Lymphocytic Leukemia in the Context of Chemoimmunotherapy and Novel Targeted Therapy
Expert Opinion on Orphan Drugs (2020)