Calreticulin as a novel B cell receptor antigen in chronic lymphocytic leukemia

Transcription

1 Published Ahead of Print on July 27, 2017, as doi: /haematol Copyright 2017 Ferrata Storti Foundation. Calreticulin as a novel B cell receptor antigen in chronic lymphocytic leukemia by Elisa ten Hacken, Maria Gounari, Jaap Willem Back, Ekaterina Shimanovskaya, Lydia Scarfo, Ekaterina Kim, Jared Burks, Maurilio Ponzoni, Giuseppe Alvise Ramirez, William G. Wierda, Zeev Estrov, Michael J. Keating, Alessandra Ferrajoli, Kostas Stamatopoulos, Paolo Ghia, and Jan A. Burger Haematologica 2017 [Epub ahead of print] Citation: ten Hacken E, Gounari M, Back JW, Shimanovskaya E, Scarfo L, Kim E, Burks J, Ponzoni M, Ramirez GA, Wierda WG, Estrov Z, Keating MJ, Ferrajoli A, Stamatopoulos K, Ghia P, and Burger JA. Calreticulin as a novel B cell receptor antigen in chronic lymphocytic leukemia. Haematologica. 2017; 102:xxx doi: /haematol Publisher's Disclaimer. E-publishing ahead of print is increasingly important for the rapid dissemination of science. Haematologica is, therefore, E-publishing PDF files of an early version of manuscripts that have completed a regular peer review and have been accepted for publication. E-publishing of this PDF file has been approved by the authors. After having E-published Ahead of Print, manuscripts will then undergo technical and English editing, typesetting, proof correction and be presented for the authors' final approval; the final version of the manuscript will then appear in print on a regular issue of the journal. All legal disclaimers that apply to the journal also pertain to this production process.

2 Calreticulin as a novel B cell receptor antigen in chronic lymphocytic leukemia Elisa ten Hacken 1, Maria Gounari 2,3, Jaap Willem Back 4, Ekaterina Shimanovskaya 4, Lydia Scarfo 2, Ekaterina Kim 1, Jared Burks, 1 Maurilio Ponzoni 2, Giuseppe Alvise Ramirez 2, William G. Wierda 1, Zeev Estrov 1, Michael J. Keating 1, Alessandra Ferrajoli 1, Kostas Stamatopoulos 3, Paolo Ghia 2 and Jan A. Burger 1 1 Department of Leukemia, The University of Texas M.D. Anderson Cancer Center, Houston, TX 2 IRCCS Ospedale San Raffaele and Università Vita-Salute San Raffaele, Milan, Italy 3 Institute of Applied Biosciences, Centre for Research and Technology Hellas, Thessaloniki, Greece 4 Pepscan, Lelystad, The Netherlands E.t.H. and M.G. equally contributed to the work; current address for E.t.H: Department of Medical Oncology, Dana Farber Cancer Institute, Boston, Massachusetts; current address for M.G: Institute of Applied Biosciences, Centre for Research and Technology Hellas, Thessaloniki, Greece Running title: Calreticulin as a CLL antigen The work was supported by a Leukemia & Lymphoma Society Scholar Award in Clinical Research (J.A.B.), and the MD Anderson s Moon Shot Program in CLL. This research is also supported in part by the MD Anderson Cancer Center Support Grant CA016672, by Associazione Italiana per la Ricerca sul Cancro AIRC Investigator grants #15189 (P.G.), and Special Program Molecular Clinical Oncology AIRC 5 per mille #9965 (P.G.). Correspondence: Jan A. Burger, MD, PhD, Department of Leukemia, Unit 428, The University of Texas MD Anderson Cancer Center, PO Box , Houston, TX ; jaburger@mdanderson.org; Elisa ten Hacken, PhD, Department of Medical Oncology, Dana Farber Cancer Institute, Boston, MA 02215; Elisa_TenHacken@dfci.harvard.edu. d

3 Chronic lymphocytic leukemia (CLL) is characterized by the accumulation of mature monoclonal B lymphocytes with a distinct immune-phenotype (CD19 + CD5 + CD23 + ), and a markedly heterogeneous clinical course, ranging from indolent disease to more aggressive presentations. B cell receptor (BCR) signaling plays a central role in disease pathogenesis and is primarily activated in secondary lymphatic tissues.(1) The mutational status of the immunoglobulin heavy chain variable region (IGHV) genes of the BCR, which distinguishes patients with mutated IGHV from those with unmutated IGHV (M-CLL or U-CLL, with 2% deviation or <2% deviation from the germline sequence, respectively) is associated with BCR signaling capacity and major differences in disease progression.(2) Patients with high BCR signaling responsiveness typically belong to the U-CLL subset, and present with more aggressive disease, resulting in an inferior prognosis, while patients with M-CLL generally have more indolent disease and better prognosis.(3, 4) Approximately 30% of patients with CLL, in both the M-CLL and U-CLL subgroups, express quasi-identical surface BCRs with stereotyped CDR3 regions, which are commonly classified into subsets (reviewed in (5)). Interestingly, similarities between cases belonging to distinct stereotyped subsets link BCR immunoglobulin sequences with shared genetic and biological characteristics and clinical behavior. For example, stereotyped subsets #1, #2, #6 and #8 often present as more aggressive CLL.(6) Furthermore, independent studies demonstrated frequent associations between specific genetic aberrations in CLL patients and stereotyped subsets.(7-9) For example, subset #1 cases frequently harbor NOTCH1 and NFKBIE mutations,(10) subset #2 patients often carry del(11q) and SF3B1 mutations, and subset #8 patients present with trisomy 12, NOTCH1 mutations and Richter s transformation.(11) In contrast, subset #4 patients are characterized by relatively young age at diagnosis, an indolent disease course, and a rare need for therapy.(6) The mechanisms that trigger BCR activation in d

4 CLL have not been fully elucidated, although BCR activation in the lymphoid tissues by autoantigens and microbial antigens is the most plausible mechanism, together with homotypic interaction of BCR-binding epitopes within the heavy and light chain of selected stereotyped BCRs (12). Relevant antigens for CLL have been characterized, particularly for BCRs from patients with U-CLL, including self-antigens, such as non-muscle myosin heavy chain IIA, vimentin, dsdna, oxidized lipoproteins (13-16) and fungal antigens. (17) BCR signaling is primarily activated in secondary lymphoid organs, presumably by interactions between CLL cells and the microenvironment, resulting in activation of key survival pathways for CLL cells, including c-myc and NF-κB proteins (18). Importantly, when CLL cells are co-cultured in the presence of monocyte-derived nurselike cells (NLCs), an in vitro model system for the lymphoid tissue microenvironment, gene signatures associated to BCR signaling are recapitulated (19), and associated with activation of BCR-associated kinases and IgM internalization, suggesting engagement of the BCRs by antigenic determinants. (20) Based on these notions, we hypothesized that NLC may carry surface proteins, which can be recognized by the CLL-BCRs, thereby initiating BCR signaling activation in the CLL lymph node microenvironment. We focused our work on the protein Calreticulin (CRT), a known autoantigen for ulcerative colitis (UC), systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA),(21) which was recently described to be expressed on the surface of tumor-associated macrophages (TAMs), facilitating cancer cell phagocytosis.(22) Under physiologic conditions, CRT is an endoplasmic-reticulum (ER) resident chaperone, involved in protein folding, antigen presentation and calcium homeostasis.(23) CRT is over-represented in the plasma of CLL patients with aggressive disease,(24) and is upregulated on the surface of apoptotic CLL d

5 cells,(25) although its functional role in CLL has not been defined. Provided the importance of autoantigen stimulation in CLL biology, and given the fact that NLC represent CLL- associated TAMs, (26) we hypothesized that CRT could be a putative CLL-BCR antigen, expressed in the context of CLL-NLC interactions. We analyzed CRT expression by immunofluorescence staining of 4 CLL-NLC preparations after a 14-day co-culture with CLL cells, and noted diffuse CRT expression, predominantly on CD68 + NLC (Figure 1A and Supplemental Table 1). We also detected CRT expression on CD68 + macrophages (Figure 1B) of a CLL lymph node biopsy. In this context, both CLL cells (red staining) and macrophage-like cells (double signal in red for calreticulin and granular brown for CD68) displayed strong immunoreactivity for CRT. We further asked whether CRT could be exposed on the surface of NLC in the context of CLL-NLC interactions, and tested its expression by surface immunofluorescence staining (sif), using IgM as a marker for CLL cells (Figure 1C and Supplemental Table 1). We were able to detect CRT expression on the surface of NLC on CLL-NLC co-cultures of 2 U-CLL and 2 M-CLL cases. Surface expression of CRT was further confirmed by Western Blot analysis of 3 NLC membrane preparations (Supplemental Figure 1A and Table 1), and flow cytometry analysis on 4 cases (Supplemental Figure 1B and Table 1). We then explored CRT reactivity of 14 different recombinant monoclonal antibodies (mabs) derived from CLL patients carrying BCRs from various different stereotyped subsets, representative of the most common IGHV genes used by CLL-BCRs (Figure 1D, Supplemental Table 2), and sera from 5 patients with systemic lupus erythematosus (SLE), whose antibodies have known reactivity against CRT as positive control. Of the 14 CLL antibodies, one mab belonging to subset #1, two belonging to subset #8, and one to subset #10 (a minor subset d

6 enriched for trisomy 12 with no definitive correlation with clinical outcome) (27) showed robust CRT binding (Figure 1D). Lower, but detectable binding, was also observed when testing M- CLL subset #4 antibodies (Figure 1D, mabs #13 and #14). We further tested the four strongest binders for dose-dependent reactivity to CRT and confirmed their protein binding properties (Figure 1E). Next, we characterized CLL-BCR specific binding epitopes within CRT. We assessed binding specificity of three CLL-mAbs [1 (subset #1), 8 (subset #8) and 12 (subset #10) from Figure 1D and Supplemental Table 2] to linear epitopes of the full CRT protein, and identified high binding of the subset #1 CLL-mAb to eighteen 15-mers of a region comprising aminoacids of the C-terminal domain of CRT protein (Figure 1F). Interestingly, the predicted linear epitopes included a common aminoacidic core EDK(D/E)(D/E)DE(D/E) (Figure 1G), which has also been described as a putative target region for SLE antibodies.(21) Discrete binding of the subset #8 and subset #10 CLL mabs to linear epitopes of CRT was not identified (Supplemental Figure 2), indicating that such mabs may preferentially recognize conformational epitopes or post-translational modifications of CRT protein. Taken all together, these results support the concept that CRT may function as a putative CLL-BCR antigen for selected U-CLL patients belonging to clinically more aggressive subsets. We also cannot exclude CRT binding to M-CLL cases, in particular those belonging to stereotyped subset #4. Further functional analysis in selected CLL primary cell subsets is required to fully understand the contribution of these interactions to CLL survival. Since CRT is overexpressed in the plasma of CLL patients,(24) upregulated in CLL cells undergoing apoptosis (25) and exposed on the surface of NLCs, conceivably CRT-BCR interactions could occur not only in the lymph node microenvironment, but also to some extent in the peripheral blood, particularly in relation to treatment-induced apoptosis mainly associated to cytotoxic therapies. d

7 This evidence poses the basis for development of novel combinatorial therapies, whereby monoclonal antibodies directed against CRT (or its specific BCR-binding epitopes) could be employed in combinatorial treatments together with BCR signaling inhibitors. Overall, our findings provide insight into subset-specific recognition of CLL-BCR antigens, with implications for differential BCR activation profiles of patients CLL cells in vivo and optimization of novel BCR-subset specific therapeutic strategies. Author contribution. E.t.H and MG conceived the research, performed experiments and wrote the manuscript; JWB, ES, EK, JB and MP performed experiments; LS, GAR, WGW, ZE, MJK, AF provided patients samples; KS and PG wrote the manuscript; JAB supervised the research and wrote the manuscript. All authors have reviewed and approved the final version of the paper. Conflict of interest. None of the authors declare competing financial interests for the presented research. e

8 References 1. Burger JA, Chiorazzi N. B cell receptor signaling in chronic lymphocytic leukemia. Trends Immunol. 2013;34(12): Lanham S, Hamblin T, Oscier D, Ibbotson R, Stevenson F, Packham G. Differential signaling via surface IgM is associated with VH gene mutational status and CD38 expression in chronic lymphocytic leukemia. Blood. 2003;101(3): Damle RN, Wasil T, Fais F, et al. Ig V gene mutation status and CD38 expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood. 1999;94(6): Hamblin TJ, Davis Z, Gardiner A, Oscier DG, Stevenson FK. Unmutated Ig V(H) genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood. 1999;94(6): Stamatopoulos K, Agathangelidis A, Rosenquist R, Ghia P. Antigen receptor stereotypy in chronic lymphocytic leukemia. Leukemia. 2017;31(2): Baliakas P, Hadzidimitriou A, Sutton LA, et al. Clinical effect of stereotyped B-cell receptor immunoglobulins in chronic lymphocytic leukaemia: a retrospective multicentre study. Lancet Haematol. 2014;1(2):e Rossi D, Spina V, Bomben R, et al. Association between molecular lesions and specific B-cell receptor subsets in chronic lymphocytic leukemia. Blood. 2013;121(24): Strefford JC, Sutton LA, Baliakas P, et al. Distinct patterns of novel gene mutations in poor-prognostic stereotyped subsets of chronic lymphocytic leukemia: the case of SF3B1 and subset #2. Leukemia. 2013;27(11): Sutton LA, Young E, Baliakas P, et al. Different spectra of recurrent gene mutations in subsets of chronic lymphocytic leukemia harboring stereotyped B-cell receptors. Haematologica. 2016;101(8): Mansouri L, Sutton LA, Ljungstrom V, et al. Functional loss of IkappaBepsilon leads to NF-kappaB deregulation in aggressive chronic lymphocytic leukemia. J Exp Med. 2015;212(6): Rossi D, Spina V, Cerri M, et al. Stereotyped B-cell receptor is an independent risk factor of chronic lymphocytic leukemia transformation to Richter syndrome. Clin Cancer Res. 2009;15(13): Minici C, Gounari M, Ubelhart R, et al. Distinct homotypic B-cell receptor interactions shape the outcome of chronic lymphocytic leukaemia. Nat Commun. 2017;8: Lanemo Myhrinder A, Hellqvist E, Sidorova E, et al. A new perspective: molecular motifs on oxidized LDL, apoptotic cells, and bacteria are targets for chronic lymphocytic leukemia antibodies. Blood. 2008;111(7): Chu CC, Catera R, Hatzi K, et al. Chronic lymphocytic leukemia antibodies with a common stereotypic rearrangement recognize nonmuscle myosin heavy chain IIA. Blood. 2008;112(13): Catera R, Silverman GJ, Hatzi K, et al. Chronic lymphocytic leukemia cells recognize conserved epitopes associated with apoptosis and oxidation. Mol Med. 2008;14(11-12): Zwick C, Fadle N, Regitz E, et al. Autoantigenic targets of B-cell receptors derived from chronic lymphocytic leukemias bind to and induce proliferation of leukemic cells. Blood. 2013;121(23): e

9 17. Hoogeboom R, van Kessel KP, Hochstenbach F, et al. A mutated B cell chronic lymphocytic leukemia subset that recognizes and responds to fungi. J Exp Med. 2013;210(1): Herishanu Y, Perez-Galan P, Liu D, et al. The lymph node microenvironment promotes B-cell receptor signaling, NF-kappaB activation, and tumor proliferation in chronic lymphocytic leukemia. Blood. 2011;117(2): Burger JA, Quiroga MP, Hartmann E, et al. High-level expression of the T-cell chemokines CCL3 and CCL4 by chronic lymphocytic leukemia B cells in nurselike cell cocultures and after BCR stimulation. Blood. 2009;113(13): Ten Hacken E, Sivina M, Kim E, et al. Functional Differences between IgM and IgD Signaling in Chronic Lymphocytic Leukemia. J Immunol. 2016;197(6): Eggleton P, Ward FJ, Johnson S, et al. Fine specificity of autoantibodies to calreticulin: epitope mapping and characterization. Clin Exp Immunol. 2000;120(2): Feng M, Chen JY, Weissman-Tsukamoto R, et al. Macrophages eat cancer cells using their own calreticulin as a guide: roles of TLR and Btk. Proc Natl Acad Sci U S A. 2015;112(7): de Bruyn M, Wiersma VR, Helfrich W, Eggleton P, Bremer E. The ever-expanding immunomodulatory role of calreticulin in cancer immunity. Front Oncol. 2015;5: Molica S, Digiesi G, D'Arena G, et al. Serum levels of soluble calreticulin predict for time to first treatment in early chronic lymphocytic leukaemia. Br J Haematol. 2016;175(5): Martinez-Torres AC, Quiney C, Attout T, et al. CD47 agonist peptides induce programmed cell death in refractory chronic lymphocytic leukemia B cells via PLCgamma1 activation: evidence from mice and humans. PLoS Med. 2015;12(3):e Galletti G, Caligaris-Cappio F, Bertilaccio MT. B cells and macrophages pursue a common path toward the development and progression of chronic lymphocytic leukemia. Leukemia. 2016;30(12): Maura F, Cutrona G, Fabris S, et al. Relevance of stereotyped B-cell receptors in the context of the molecular, cytogenetic and clinical features of chronic lymphocytic leukemia. PLoS One. 2011;6(8):e e

10 Figure 1. Calreticulin is a putative CLL-BCR antigen. (A) Representative confocal imaging analysis of CLL-NLC co-cultures stained with DAPI (blue) for nuclear staining, anti-crt (green), anti-cd68 (red). The merge of the three colors and the scale bar of 40µm is indicated in the image. (B) Representative immunohistochemistry staining of a CLL lymph node section stained with anti-cd68 antibodies (brown) and with anti-crt antibodies (red). Black arrows indicate CD68 + macrophages. Scale bar, 20µm. (C) Surface immunofluorescence staining of CLL-NLC preparations from 2 U-CLL (upper panels) and 2 M-CLL cases (lower panels) with anti-crt antibodies (green) and anti-igm antibodies (red). Scale bar, 20µm. (D) Calreticulin reactivity, expressed as OD ratio of negative control wells of 14 stereotyped CLL-mAbs, derived from CLL patients belonging to distinct subsets (i.e. U-CLL subsets #1, #2, #6, #8, #9, #10 and M-CLL subset #4), and tested by ELISA at a concentration of 20 µg/ml. Sera from systemic lupus erythematosus (SLE) were tested in a 1:100 dilution, as positive control. CLL-mAbs with positive reactivity to Calreticulin (OD sample at least 2,5 fold of OD blank) were color-coded: Subset #1 (red), Subset #8 (blue and green), Subset #10 (yellow). (E) Titration of Calreticulin reactivity of the 4 CLL-BCRs for which we detected high binding activity, when tested at increasing concentrations (µg/ml) in ELISA assays. The color code is consistent to the one in Figure D. (F) Epitope binding intensity of one subset #1 CLL-mAb to 15-mer linear epitopes spanning the full Calreticulin protein. ELISA-based signal intensity (in mau) is displayed. A schematic representation of full-length Calreticulin is depicted underneath. (G) Curated alignment of the 18 peptides bound by the subset #1 CLL-mAb. Fully conserved residues are highlighted in yellow, conservative mutations are highlighted in cyan. Peptides 8 and 14 are present twice in this alignment (second incidence indicated by B). e

11

12 Supplemental Materials and Methods Patients samples and reagents Peripheral blood (PB) samples were obtained from previously untreated patients fulfilling diagnostic and immunophenotypic criteria for CLL at the Leukemia Department at MD Anderson Cancer Center after informed consent and Institutional Review Board approval (Supplemental Table 1). Sera from patients with active systemic lupus erythematosus (SLE) were obtained from patients fulfilling diagnostic criteria for SLE at the Ospedale Vita-Salute San Raffaele (Milano, Italy) after informed consent and approval by the Institutional Ethical Committee. CLL- PBMCs were isolated by density gradient centrifugation over Ficoll-Plaque (GE Healthcare), and cultured at 1.5 x 10 7 cells/ml in RPMI 1640 medium supplemented with 10% Fetal Bovine Serum (FBS; Gibco), and penicillin-streptomycin-l-glutamine solution (Corning Inc.) (complete RPMI). Immunofluorescence and confocal microscopy CLL-PBMCs were cultured at 1.5 x 10 7 cells/ ml in complete RPMI for 14 days onto 8-well microscope-compatible chamber slides (ibidi). Media was then removed, and NLC outgrowth (1) confirmed by phase contrast microscopy. Cells were then washed twice in PBS, then fixed with 4% paraformaldehyde (PAF) for 10 minutes at room temperature. PBS washes were repeated prior to permeabilization with 0.5% Triton X-100 in PBS for 10 minutes at room temperature. PBS washes were again repeated, then cells were blocked for one hour in blocking solution (3% BSA + 1% horse serum) at room temperature. Rabbit monoclonal anti-calreticulin primary antibody (Cell Signaling Technology, 1:400) and mouse monoclonal anti-cd68 (Abcam, 1:200) were added and incubated overnight at 4ºC. Cells were then washed twice with PBS and secondary anti-rabbit AlexaFluor488 and anti-mouse AlexaFluor594 were added at 1:1000 concentration, and incubated 1

13 for 1 hour at room temperature. Cells were again washed, then stained with DAPI (1:10.000) for 3 minutes at room temperature, prior to addition of 300µL mounting media (Dako) to the chamber, and imaged through an Olympus FV-1000 laser confocal microscope (60X objective). For surface immunofluorescence staining, anti-calreticulin antibody (1:400) and mouse monoclonal anti-igm (Abcam, 1:200) were added prior to fixation and permeabilization of the CLL-NLC preparations. Immunohistochemistry For double immunostain, section of a lymph node effaced by CLL underwent two-step procedure. During the first step, anti-cd68 (clone KP1, Leica Biosystem, dilution 1:400) mouse monoclonal IgG1 followed antigen retrieval with Tris-EDTA at ph9 and was revealed with DAB. The second step utilized anti-calreticulin rabbit monoclonal IgG at 1:200 dilution following antigen retrieval for 5 minutes of microwave treatment with citrate buffer at ph6 and the reaction was revealed with FastRed. Western Blot and flow cytometry Immunoblot (2) and flow cytometry (3) were performed as previously described. After 14 days of CLL-NLC co-culture, CLL cells were washed off of the NLC layer by repeated washes with complete RPMI medium. To detach NLC, 5mM EDTA in PBS was added for 2 minutes, prior to scraping. Membrane and cytosolic fractions were isolated with the MEM-PER extraction kit from Thermo Fisher Scientific, according to the manufacturer s instructions. Anti-Calreticulin, anti- HSP90 and anti-gapdh primary antibodies, all of them used at a 1:1000 concentration, were purchased from Cell Signaling Technology. For flow cytometry, cells were incubated for 20 minutes with PE-conjugated anti-calreticulin (1:100) from Enzo Life Sciences at 4 C, or isotype 2

14 control. Samples were acquired on a FACSCalibur (Beckton Dickinson) and data analyzed using FlowJo Software version (TreeStar). Expression of recombinant CLL monoclonal antibodies and ELISA assays Soluble CLL monoclonal antibodies (mabs) were prepared as recombinant human IgG1 paired with kappa or lambda light chain depending on the isotype expressed by the leukemic clone, using standard methodology (4). In brief, heavy and light chain variable regions were amplified by PCR using gene-specific primers and cloned into a human IgG1, IgK, or IgL expression vector containing the human Igγ1, Igκ or Igλ constant regions and a murine Ig gene signal peptide sequence (GenBank accession no. DQ407610). Recombinant mabs were expressed in human embryonic kidney (HEK) 293 cells by co-transfection with immunoglobulin (IG) heavy chain and corresponding IG light chain encoding plasmids using Jet Pei reagent (Polyplus transfections). Twelve hours after transfection, cell culture medium was replaced with serum-free DMEM supplemented with 1% Nutridoma SP (Roche Diagnostics). Supernatants were collected after 7 days of culture and cell debris was removed by centrifugation at 800 x g for 10 minutes. Recombinant mabs were purified using Protein G beads (GE Healthcare), eluted in 0.1M glycine (ph 3.0) and neutralized with 1M Tris-HCl (ph 8.0). Their integrity and purity was confirmed by SDS-PAGE analysis and Coomassie blue staining and their concentration were determined by quantitative ELISA (Human IgG ELISA Quantitation Kits, Bethyl laboratories Inc.). Reactivity against Calreticulin was assessed by ELISA. Briefly, 96-well plates (Corning) were coated with 100µl of 5 µg/ml of purified Calreticulin (Abcam) overnight at 4 C, and then blocked with 150 µl 3% BSA in PBS for 3 hours at room temperature. CLL mabs were used as primary antibodies at a concentration of 20 µg/ml and incubated overnight at 4 C. In parallel, sera from systemic lupus 3

15 erythematosus (SLE) patients were used as positive control in a 1:100 dilution. Next, 100µl of horseradish peroxidase (HRP) conjugated goat anti-human IgG antibody (Bethyl laboratories) at 1: dilution was added for 1 hour at room temperature and assays were developed using 50µl of TMB Peroxidase Substrate (Bethyl laboratories). The reaction was stopped after 15 minutes by adding 50µl 0,18 M H 2 SO 4 (Bethyl laboratories) and the absorbance was measured at 450 nm. All samples were tested in triplicates. The analysis was performed on a BioTeK ELx800 ELISA reader (BioTek), and data analyzed by GraphPad Prism 5 software. Anti-Calreticulin binding of CLL mabs was plotted as the signal-to-background ratio. CLL mabs with a ratio >2.5 were considered as Calreticulin binders. The titration of anti-calreticulin reactivity of the positive mabs was performed by ELISA as described above, and CLL mabs were used at various concentrations (2,5-80 µg/ml). Epitope mapping To reconstruct epitopes of the target molecule a library of peptides was synthesized. An amino functionalized polypropylene support was obtained by grafting with a proprietary hydrophilic polymer formulation, followed by reaction with t-butyloxycarbonyl-hexamethylenediamine (BocHMDA) using dicyclohexylcarbodiimide (DCC) with Nhydroxybenzotriazole (HOBt) and subsequent cleavage of the Boc-groups using trifluoroacetic acid (TFA). Standard Fmoc-peptide synthesis was used to synthesize peptides on the amino-functionalized solid support by custom modified JANUS liquid handling stations (Perkin Elmer). Synthesis of structural mimics was done using Pepscan s proprietary Chemically Linked Peptides on Scaffolds (CLIPS) technology. CLIPS technology allows to structure peptides into single loops, double loops, triple loops, sheet-like folds, helix-like folds and combinations thereof. CLIPS templates are coupled to cysteine residues. 4

16 The side-chains of multiple cysteines in the peptides are coupled to one or two CLIPS templates. For example, a 0.5 mm solution of the P2 CLIPS (2,6-bis(bromomethyl)pyridine) is dissolved in ammonium bicarbonate (20 mm, ph 7.8)/acetonitrile (1:3(v/v)). This solution is added onto the peptide arrays. The CLIPS template will bind to side-chains of two cysteines as present in the solid-phase bound peptides of the peptide-arrays (455 wells plate with 3 µl wells). The peptide arrays are gently shaken in the solution for 30 to 60 minutes while completely covered in solution. Finally, the peptide arrays are washed extensively with excess of H 2 O and sonicated in disruptbuffer containing 1% SDS/0.1 %beta-mercaptoethanol in PBS (ph 7.2) at 70 C for 30 minutes, followed by sonication in H 2 O for another 45 minutes. The T3 CLIPS carrying peptides were made in a similar way but now with three cysteines. The binding of antibody to each of the synthesized peptides was tested in a PEPSCAN-based ELISA. The peptide arrays were incubated with primary antibody solution (overnight at 4 C). After washing, the peptide arrays were incubated with a 1:1000 dilution of an appropriate antibody peroxidase conjugate (SBA) for one hour at 25 C. After washing, the peroxidase substrate 2,2 azino-di-3-ethylbenzthiazoline sulfonate (ABTS) and 20 µl/ml of 3% H 2 O 2 were added. After one hour, color development was measured and quantified with a charge coupled device (CCD) - camera and an image processing system. Values obtained from the CCD camera range from 0 to 3000 mau, similar to a standard 96-well plate ELISA reader. The results were quantified and stored into the Peplab database. References 1. Burger JA, Tsukada N, Burger M, Zvaifler NJ, Dell'Aquila M, Kipps TJ. Blood-derived nurse-like cells protect chronic lymphocytic leukemia B cells from spontaneous apoptosis through stromal cell-derived factor-1. Blood. 2000;96(8): ten Hacken E, Scielzo C, Bertilaccio MT, Scarfo L, Apollonio B, Barbaglio F, et al. Targeting the LYN/HS1 signaling axis in chronic lymphocytic leukemia. Blood Mar 21;121(12):

17 3. Ten Hacken E, Sivina M, Kim E, O'Brien S, Wierda WG, Ferrajoli A, et al. Functional Differences between IgM and IgD Signaling in Chronic Lymphocytic Leukemia. J Immunol Aug Gounari M, Ntoufa S, Apollonio B, Papakonstantinou N, Ponzoni M, Chu CC, et al. Excessive antigen reactivity may underlie the clinical aggressiveness of chronic lymphocytic leukemia stereotyped subset 8. Blood Apr 21. 6

18 Supplemental Table 1 Clinical and biological features of the analyzed CLL patient samples. Rai stage Age Sex IGHV status CD38 ZAP70 Cytogenetics CRT expression analysis I 67 M M Negative Negative del (13q) sif, FC,WB 0 55 M M Negative Negative Negative IF, sif, FC 0 72 M nd Positive Positive tri12 IF,WB 0 42 M U Negative nd del (13q) IF, WB IV 73 M U Negative nd Negative IF, sif, FC II 62 M U Positive Negative del (13q) sif, FC Rai stage of disease, age at diagnosis, sex, IGHV gene mutational status (M, M-CLL; U, U-CLL), CD38 expression (pos, positive; neg, negative), ZAP70 expression (pos, positive; neg, negative), cytogenetic abnormalities (del, deletion; tri, trisomy), and type of analysis conducted to assess CRT expression. IGHV gene mutational analysis was performed by PCR, followed by direct sequencing, and 98% cutoff was used for mutational status assessment. CD38 expression was determined by flow cytometry, and 30% positivity was used as cutoff. ZAP70 expression was evaluated by immunohistochemical staining of bone marrow biopsies. Cytogenetic abnormalities were determined by fluorescence in situ hybridization analysis. IF: immunofluorescence; sif: surface immunofluorescence; FC: flow cytometry; WB: Western Blot. 7

19 Supplemental Table 2. Molecular characteristics and stereotyped subset assignment of recombinant CLL monoclonal antibodies tested for anti-calreticulin reactivity CLL mab # IGHV IGΗV % IGHD IGHJ HCDR3 amino acid sequence IGLV IGLV % IGLJ LCDR3 amino acid sequence 1 #1 IGHV1-03*13 100,00 IGHD6-19*01 IGHJ4*02 CAREQWLVLNYFDYW IGKV1-39*01 100,00 IGKJ2*01 CQQSYSTPPYTF 2 #1 IGHV1-2*02 100,00 IGHD6-19*01 IGHJ4*02 CARGQWLVQLNFDYW IGKV1-39*01 100,00 IGKJ2*02 CQQSYSTPPYTF 3 #1 IGHV5-a*03 100,00 IGHD6-19*01 IGHJ4*02 CAREQWLVLEHFDYW IGKV1-39*01 100,00 IGKJ3*01 CQQSYSTPRFTF 4 #2 IGHV3-21*01 97,22 IGHD1-1*01 IGHJ6*02 CASDKNGMDVW IGLV3-21*01 99,57 IGLJ3*02 CQVWDSSSDHPWVF 5 #2 IGHV3-21*01 98,61 ND IGHJ6*02 CARDQNAMDVW IGLV3-21*01 97,85 IGLJ3*02 CQVWDSGSDHPWVF 6 #6 IGHV1-69*06 100,00 IGHD3-16*02 IGHJ3*02 CARGGNYDYIWGSYRSNDAFDIW IGKV3-20*01 100,00 IGKJ4*01 CQQYGSSPPLTF 7 #6 IGHV1-69*06 100,00 IGHD3-16*02 IGHJ3*02 CARGGDYDYVWGSYRSNDAFDIW IGKV3-20*01 100,00 IGKJ4*01 CQQYGSSPTF 8 #8 IGHV4-39*01 100,00 IGHD6-13*01 IGHJ5*02 CARHNLGYSSSWYSRNNWFDPW IGKV1-39*01 100,00 IGKJ1*01 CQQSYSTPRTF 9 #8 IGHV4-39*07 100,00 IGHD6-13*01 IGHJ5*02 CARRF_GYSSSWYGLD_WFDPW IGKV1-39*01 99,64 IGKJ1*01 CQQSYSTPRTF 10 #8 IGHV4-39*01 99,31 IGHD6-13*01 IGHJ5*02 CASKT_GYSSSWYGRD_WFDPW IGKV1-39*01 99,64 IGKJ1*01 CQQSYSTPRTF 11 #9 IGHV3-21*01 100,00 IGHD3-3*01 IGHJ6*02 CARGVLNYDFWSVYYYYGMDVW IGKV4-1*01 100,00 IGKJ4*01 CQQYYSTPLTF 12 #10 IGHV4-39*01 100,00 IGHD2-2*01 IGHJ6*02 CARHRLGYCSSTSCYYYYYGMDVW IGLV1-40*01 98,96 IGLJ2*01 CQSYDSSLSVVF 13 #4 IGHV4-34*01 96,84 IGHD3-10*01 IGHJ6*04 CARGYADTPVFRRYYYYGMDVW IGKV2-30*02 97,62 IGKJ2*01 CMQGTHWPPYTF 14 #4 IGHV4-34*01 96,84 IGHD5-18*01 IGHJ6*02 CARGYGDTPTIRRYYYYGMDVW IGKV2-30*02 98,62 IGKJ2*01 CMQGTHWPPYTF Stereotyped CLL-mAb heavy (IGHV, IGHD, IGHJ) and light chain (IGLV, IGLJ) characteristics, percentage (%) identity to germline IGHV and IGLV genes, and CDR3 aminoacidic sequences of heavy (HCDR3) and light (LCDR3) chains. 8

20 Supplemental Figure 1. Calreticulin is expressed on the surface of NLC. (A) Western blot analysis of Calreticulin (CRT) expression after membrane to cytosolic fractionation of 3 NLC preparations. htert stroma (S) is used as positive control. HSP90 expression is analyzed to exclude cytosolic contamination in membrane fractions. GAPDH is tested, as loading control. (B) Flow cytometry staining of Calreticulin expression on the surface of NLC, gated based on FSC- SSC parameters. Isotype histogram is shown, as control. 9

21 Subset #8 CLL-mAb Epitope Binding Intensity Subset #10 CLL-mAb Epitope position (aa) Supplemental Figure 2. Intensity profiles recorded for one subset #8 (CLL mab #8 as in Supplemental Table 2) and one subset #10 (CLL mab #12 as in Supplemental Table 2) antibody under high (red traces) and moderate stringency conditions (black traces) with linear 15-mer peptides. 10