Open Full Page

Bcl-2 Overexpression Leads to Increases in Suppressor of Cytokine Signaling-3 Expression in B Cells and De novo Follicular Lymphoma

Gary J. Vanasse,1 Robert K. Winn,4 Sofya Rodov,1 Arthur W. Zieske,2 John T. Li,4 Joan C. Tupper,3 Jingjing Tang,5 Elaine W. Raines,5 Mette A. Peters,6 Ka Yee Yeung,6 and John M. Harlan3

Departments of 1Internal Medicine and 2Laboratory Medicine, Yale University School of Medicine, New Haven, Connecticut and Departments of 3Internal Medicine, 4Surgery, and 5Pathology, School of Medicine, and 6Center for Expression Arrays, University of Washington, Seattle, Washington

Abstract hyperplastic tonsil tissue. These data suggest that Bcl-2 The t(14;18)(q32;q21), resulting in deregulated overexpression leads to the induction of activated signal expression of B-cell-leukemia/lymphoma-2 (Bcl-2), transducer and activator of transcription 3 (STAT3) represents the genetic hallmark in human follicular and to the induction of SOCS3, which may contribute lymphomas. Substantial evidence supports the to the pathogenesis of follicular lymphoma. hypothesis that the t(14;18) and Bcl-2 overexpression (Mol Cancer Res 2004;2(11):620–31) are necessary but not solely responsible for neoplastic transformation and require cooperating genetic derangements for neoplastic transformation to occur. To Introduction investigate genes that cooperate with Bcl-2 to influence Follicular lymphomas comprise approximately one third of cellular signaling pathways important for neoplastic all cases of non–Hodgkin’s lymphoma in humans. Follicular transformation, we used oligonucleotide microarrays to lymphomas are initially clinically indolent and chemosensitive determine differential gene expression patterns in but have a natural history marked by multiple relapses, CD19+ B cells isolated from EM-Bcl-2 transgenic mice becoming progressively chemoresistant and ultimately remain- and wild-type littermate control mice. Fifty-seven genes ing incurable. Twenty-five percent to 60% of follicular were induced and 94 genes were repressed by >_2-fold in lymphomas also transform into more aggressive subtypes of EM-Bcl-2 transgenic mice (P < 0.05). The suppressor of non–Hodgkin’s lymphoma (1-3). Eighty-five percent of cytokine signaling-3 (SOCS3) gene was found to be follicular lymphomas harbor t(14;18)(q32;q21), resulting in overexpressed 5-fold in B cells from EM-Bcl-2 transgenic juxtaposition of the B-cell-leukemia/lymphoma-2 (Bcl-2) mice. Overexpression of Bcl-2 in both mouse embryo proto-oncogene with the immunoglobulin heavy chain (IgH) fibroblast-1 and hematopoietic cell lines resulted locus, typically upstream of one of the JH segments (4-8). in induction of SOCS3 protein, suggesting a Deregulated expression of Bcl-2 prolongs survival of B and T Bcl-2-associated mechanism underlying SOCS3 lymphocytes via abrogation of the majority of apoptotic induction. Immunohistochemistry with SOCS3 antisera pathways (8-10). Substantial evidence supports the hypothesis on tissue from a cohort of patients with de novo that t(14;18) and Bcl-2 overexpression are necessary but not follicular lymphoma revealed marked overexpression solely responsible for the genesis of follicular lymphomas. EA- of SOCS3 protein that, within the follicular center Bcl-2 transgenic mice uniformly develop polyclonal B-cell cell region, was limited to neoplastic follicular hyperplasia, but only 5% to 15% eventually progress to lymphoma cells and colocalized with Bcl-2 expression aggressive monoclonal B-cell lymphomas following a pro- in 9 of 12 de novo follicular lymphoma cases examined. tracted latency period and often in conjunction with cooperating In contrast, SOCS3 protein expression was not cytogenetic lesions (10-13). Further evidence supporting the detected in the follicular center cell region of benign notion that t(14;18) is not causative for the development of follicular lymphoma is that B cells harboring t(14;18) have been detected by PCR screening of peripheral blood and hyperplastic lymphoid tissue from healthy individuals (14-16), Received 7/6/04; revised 9/20/04; accepted 10/6/04. Grant support: NIH grants CA78254 (G.J. Vanasse) and 5U24DK058813-02 and this phenomenon seems to increase with age (17). Finally, (K.Y. Yeung), American Society of Hematology fellow scholar grant (G.J. the rare hematologic disorder, persistent polyclonal B-cell Vanasse), and NIH research grant CA-16359 from the National Cancer Institute. lymphocytosis, is characterized by chronic stable polyclonal The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in lymphocytosis that, despite the presence of Bcl-2 rearrange- accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ments into the IgH locus, fails to overexpress Bcl-2 (18). These Notes: G.J. Vanasse is a past American Society of Hematology fellow scholar and data suggest that, although t(14;18) is sufficient to initiate an member of the Yale Cancer Center. Requests for reprints: Gary J. Vanasse, Section of Hematology, Department of oncogenic pathway, Bcl-2 alone is a relatively weak oncogene Internal Medicine, Yale University School of Medicine, 333 Cedar Street, WWW- and requires additional cooperating genetic lesions for neoplas- 403, Box 208021, New Haven, CT 06520. Phone: 203-737-2340; Fax: 203-785- 7232. E-mail: [email protected] tic transformation to occur. Although molecular analysis of Copyright D 2004 American Association for Cancer Research. human follicular lymphoma has revealed numerous cytogenetic

620 Mol Cancer Res 2004;2(11). November 2004 Downloaded from mcr.aacrjournals.org on September 25, 2021. © 2004 American Association for Cancer Research. Bcl-2-Associated SOCS3 Induction in Follicular Lymphoma 621

alterations potentially important for propagation of a neoplastic overexpressing polyclonal B cells as a template to identify clone (19), the significance of these secondary chromosomal genes also deregulated in de novo follicular lymphoma. We abnormalities in influencing clinical course and pathogenesis of purified primary B cells from EA-Bcl-2 transgenic mice and follicular lymphoma remains to be determined. wild-type LMC mice by negative selection and did oligonu- The application of oligonucleotide and cDNA microarray cleotide microarray analyses to formulate a differential gene technology to the study of non–Hodgkin’s lymphoma has expression profile. Phenotypes of the mice were similar to that provided insights into gene expression patterns that differentiate reported previously, with EA-Bcl-2 transgenic mice exhibiting malignant B cells from their normal counterparts, has defined B-cell hyperplasia as described (28). At the time of analysis, all prognostic subgroups, and has identified potential therapeutic mice were healthy and without evidence of tumor formation. targets (20-22). Gene profiling studies on follicular lymphoma Single cell suspensions of splenocytes were prepared from B cells have revealed a genetic signature similar to germinal spleens harvested from six 24-week-old EA-Bcl-2 transgenic center B cells; have identified differentially expressed genes and five age-matched wild-type LMC mice (C57BL/6 strain). involved in cellular pathways important for cell cycle reg- Immunomagnetic bead depletion was used to isolate naive B ulation, cell adhesion, cellular signaling, and B-cell develop- cells of primarily B2 subtype and devoid of B1 subtype, T cells, ment; and have shown that transformation of follicular monocytes, and natural killer cells. Negative selection of B cells lymphoma into diffuse large B-cell lymphomas requires distinct was done to avoid B-cell activation and its resultant gene genetic alterations (20, 23-27). However, much of the gene profile, which may confuse interpretation of the microarray expression analyses have been generated on follicular lympho- analysis. B cells were phenotyped and analyzed by flow ma B cells obtained from patients heavily treated for relapsed cytometry, revealing a >97% CD19+ pure population without disease, on t(14;18)+ cell lines rather than on primary cells, or detectable CD4+, CD8+, or CD56+ (data not shown). on RNA isolated from whole tissue biopsies rather than from purified follicular lymphoma cells. Therefore, these studies may be compromised in their ability to distinguish early, primary Oligonucleotide Microarray Analysis genetic events important for the genesis of follicular lymphoma To identify Bcl-2-mediated differential gene expression, we from the multitude of secondary genetic changes associated compared the CD19+ B-cell gene expression between trans- with disease progression, therapeutic intervention, or the cel- gene-positive and LMC mice by oligonucleotide microarray lular microenvironment. analysis. Target transcripts (15 Ag) from individual mice from The use of microarray technology to analyze gene each cohort were hybridized to Affymetrix murine U74v2 A, B, expression profiles in animal models of proto-oncogene and C chipsets (Affymetrix, Santa Clara, CA) consisting of deregulation may facilitate the identification of those primary >36,000 genes and expressed sequence tags. Individual array genetic events important for tumorigenesis in humans. We results obtained from the six EA-Bcl-2 transgenic and the five hypothesized that gene expression profiling of primary, LMC mice were summarized as one experimental array and one polyclonal B cells overexpressing Bcl-2 could serve as a tem- control array, respectively, and ratios built and analyzed using plate for the identification of candidate genes, the deregulation Rosetta Resolver version 3.1 software (Rosetta Biosoftware, of which affects pathways important for the biology of Bcl-2- Seattle, WA) were used to identify genes exhibiting z2-fold associated lymphomas. Using oligonucleotide microarrays to differential expression (P < 0.05). Comparison of the two analyze nonmalignant B cells from EA-Bcl-2 transgenic mice, composite arrays revealed that a total of 151 genes were we aimed to identify differentially expressed genes that also differentially expressed according to our parameters, with exhibited correlative deregulated expression in human follicular 57 genes induced and 94 genes repressed (Fig. 1; Table 1). Of lymphoma. In the present study, we show that CD19+ B cells this group, 103 represented known genes, whereas 48 were isolated from EA-Bcl-2 transgenic mice overexpress the cDNA without known function or homology based on Genbank suppressor of cytokine signaling-3 (SOCS3) gene when com- database references. Analysis revealed differential expression of pared with littermate control (LMC) mice. SOCS3 induction is both known and novel genes associated with cellular pathways mediated by overexpression of Bcl-2 in a manner independent important in apoptosis, B-cell growth and differentiation, cell of the site of transgene insertion. We also provide evidence cycle, intracellular signaling, inflammatory response mediators, showing overexpression of SOCS3 protein in a cohort of immunity, and DNA damage repair. Genes associated with patients diagnosed with de novo follicular lymphoma. Taken antiapoptotic pathways (HSP1a and HSP1b) as well as me- together, these studies suggest that the Bcl-2-associated diators of intracellular signaling (SOCS3 and MAP3K11) were induction of SOCS3 represents an early genetic event induced. Conversely, reduced expression was noted in the influencing cellular pathways important for the pathogenesis proapoptotic gene Bid; the c-myc and c-myb proto-oncogenes; of follicular lymphomas in humans. cyclin D2, an important regulator of progression through G1 phase of the cell cycle; and several genes associated with in- nate immunity (Pgrp, CR2, and TRAF1). To validate the array Results results, we measured mRNA levels by real-time quantitative Purification of B Cells from El-Bcl-2 Transgenic and Wild- reverse transcription-PCR (RT-PCR) using SYBR Green I. type Mice Differential gene expression was confirmed in 18 of 20 genes To characterize Bcl-2-mediated cellular signaling pathways tested (Table 2). The fold change for several genes tested by important for the development of follicular lymphoma in real-time quantitative RT-PCR was proven greater than that humans, we used differential gene expression in Bcl-2- reported on the microarray, suggesting that microarray analysis

FIGURE 1. Oligonucleotide microarray analysis of murine CD19+ B cells overexpressing Bcl-2. Affymetrix murine U74v2 A, B, and C chipsets were used to study CD19+ B cells obtained from EA-Bcl-2 transgenic and transgene-negative LMC mice. Normalized intensity data from individual arrays obtained from six EA-Bcl-2 transgenic and five LMC mice were summarized as one combined transgenic intensity experiment and one combined control intensity experiment, respectively, and analyzed using Rosetta Resolver version 3.1 software to identify genes exhibiting z2-fold differential expression (P < 0.05). The combined LMC intensity experiment was used as the baseline. Red crosses, induced expression; green crosses, reduced expression.

may underestimate the amplitude of differential gene expres- B-cell hyperplasia as well as a similar incidence of lymphoma sion. Interestingly, the SOCS3 gene was found to be induced development (29). Probing with SOCS3 antisera revealed a 2-fold on the microarray, and this was highly statistically marked increase in SOCS3 protein in both strains of EA-Bcl-2 significant (P = 0.003). To confirm SOCS3 induction, real-time transgenic mice relative to their respective LMC mice (Fig. 2). quantitative RT-PCR and RNase protection assays were done These results indicate that the induction of SOCS3 occurs on RNA samples from B cells from array mice. Both methods independently of the site of transgene insertion, thereby revealed 5-fold induction of SOCS3 mRNA in B cells from decreasing the likelihood that SOCS3 induction is the result EA-Bcl-2 transgenic mice, and this was consistent across of insertional mutagenesis. In contrast, probing with antisera samples (Table 2). An accumulating body of evidence indicates recognizing the SOCS family members CIS, SOCS1, and SOCS3 to be an important negative regulator of inflammatory SOCS5 failed to reveal detectable protein expression of these and immune responses. However, Bcl-2-associated transcrip- other SOCS family members in B cells from either strain of tional deregulation of SOCS3 has not been reported previously. EA-Bcl-2 transgenic mice (data not shown). Therefore, SOCS3 seemed an interesting candidate gene warranting further investigation. The complete differential gene Overexpression of Bcl-2 Leads to Increased SOCS3 expression analysis has been deposited in the Gene Expression Expression Omnibus at http://www.ncbi.nlm.nih.gov/geo/. We then wanted to determine whether SOCS3 induction was due to overexpression of Bcl-2 or merely a response to antigen- SOCS3 Protein Is Overexpressed in Distinct Strains of driven polyclonal B-cell hyperplasia common in EA-Bcl-2 El-Bcl-2 Transgenic Mice transgenic mice. A retroviral construct containing a human Bcl- To determine whether SOCS3 protein levels were induced in 2 cDNA and an IRES-enhanced green fluorescent protein EA-Bcl-2 transgenic mice relative to LMC mice, we did (EGFP) was then overexpressed in both mouse embryo fibroblast Western blot analysis on whole cell lysates prepared from four (MEF-1) and monocyte/macrophage hematopoietic (JAWSII) matched pairs of transgenic and control mice. To control for the cell lines. Western analysis for Bcl-2 revealed undetectable possibility that SOCS3 induction was due to the site of endogenous expression of Bcl-2 in MEF-1 and JAWSII cells transgene insertion and not mediated by Bcl-2, we also (Fig. 3A). Whole protein lysates from each cell line were prepared whole cell lysates from CD19+ B cells isolated as prepared as described and measured for SOCS3 protein levels. described from a distinct strain of EA-Bcl-2 transgenic mice When probed with SOCS3 antisera, MEF-1 cells overexpressing (29), which as a result of transgene insertion leads to Bcl-2 the Bcl-2:EGFP construct exhibited marked overexpression of overexpression in both B and T cells. Otherwise, both of the SOCS3 protein compared with MEF-1 cells expressing EGFP EA-Bcl-2 transgenic strains exhibit phenotypically similar alone, where SOCS3 levels were proven undetectable (Fig. 3B).

Mol Cancer Res 2004;2(11). November 2004 Downloaded from mcr.aacrjournals.org on September 25, 2021. © 2004 American Association for Cancer Research. Bcl-2-Associated SOCS3 Induction in Follicular Lymphoma 623

Table 1. Differential Gene Expression Analysis of Murine CD19+ B Cells Overexpressing Bcl-2

Accession (Unigene) Gene/Protein Name Fold Change P

Oncogene and tumor suppressor proteins AA839840 Putative RNA polymerase II elongation factor À2.00 0.02 M12848 c-myb proto-oncogene (c-myb) À2.00 0.04 L00039 c-myc proto-oncogene (c-myc) À2.04 0.01 Z31359 Neoplastic progression 2 gene (Npn2) À2.02 0.03

Intracellular signaling mediators and stress response proteins M12571 Heat shock protein 70.3 (HSP1a) 11.12 <0.001 AF109906 Heat shock protein 70.1 (HSP1b) 7.30 0.007 U24703 Reelin 6.03 0.003 AI842663 Osmotic stress protein 94 (OSP94) 3.95 0.02 X78667 Ryanodine receptor 2 (Ryr2) 3.08 0.007 AI846606 Hypothetical protein similar to MAP3K11 2.44 0.03 AV084051 IL-1 receptor antagonist 2.26 0.03 AV374868 SOCS3 2.05 0.003 M90388 Protein tyrosine phosphatase 70zpep À2.00 0.02 AI153935 Phosphatidylinositol 3-kinase, regulatory subunit, p150 À2.00 0.01 AA175606 Putative InB~ protein À2.00 0.04 Y09632 Rabkinesin-6 À2.14 0.03 AW122494 Ras-GRF2 À2.15 0.008 AI594690 Choline/ethanolaminephosphotransferase 1 À2.34 0.04 AV087622 Annexin 4 À3.01 0.03 U96635 Nedd4 À3.25 0.01 AW050293 Putative elongation factor Tu À3.97 <0.001 L35302 Tumor necrosis factor receptor – associated factor 1 (TRAF1) À4.13 0.03

Apoptosis and cell cycle regulation proteins AB021861 Apoptosis signal-regulating kinase 2 2.76 <0.001 L31532 Bcl-2a exon 2 2.00 0.03 U75506 BID BH3-only domain protein À2.00 0.01 L31532 Bcl-2h À2.02 0.01 AI447296 Ectodysplasin A receptor – associated death domain (EDAR) À2.07 0.01 AI605650 DNase g precursor À2.37 0.02 AA119627 Protein similar to M-phase phosphoprotein 9 À2.37 0.04 AI152882 Transglutaminase 2 À5.92 <0.001 M83749 Cyclin D2 À5.92 0.04

Transcription factors and DNA binding proteins AV349362 Myelin transcription factor 1 (Myt 1) 2.59 0.003 AI841913 Sclerostin-like protein 2.59 0.03 AI527205 Coup transcription factor 2 2.41 0.03 AW050036 Brain abundant membrane signal protein (Basp1) homologue 2.06 0.02 U08185 B-lymphocyte – induced maturation protein 1 (BLIMP1) 2.00 0.01 AA162644 Putative transcription regulator NT fin 12 À2.20 0.02 AF077861 Id2 gene À2.20 0.03 AI415206 IFN-induced Mx protein À2.24 0.009 AI957146 Putative MASL1 gene À2.28 0.01 AI594455 Trichorhinophalangeal 1 (TRPS1) À2.50 0.01 AA960657 Putative INF-g-induced protein IFI16 À2.97 0.04 AI019193 T-cell transcription factor 7 (Tcf 7) À4.71 <0.001

Receptors and cell surface proteins AI647643 Signal recognition particle 54-kDa protein (Srp54) 3.29 0.004 AF010254 C1 inhibitor 3.20 0.02 AF107847 Golgi protein 55 isoform 2.77 0.02 AI608001 Src H3 domain bp 4 2.67 0.009 AA510989 Protein similar to IL-6 receptor a 2.53 0.04 AI426271 Paired Ig-like type 2 receptor a 2.52 0.04 AJ132336 Chemomokine receptor 9 (CCR9) 2.37 0.03 AW124738 Lanthionine synthetase C – like protein (Lancl-1) 2.23 0.02 AI849185 Muscleblind-like protein 2.19 0.01 M29281 Complement receptor 2 (CR2) À2.00 <0.001 M63695 Cd1d1 À2.00 0.003 AV340322 IFN-g-induced Mg11 protein homologue À2.00 0.008 M18194 Fibronectin À2.05 0.004 AI747561 Mucolipin 3 À2.15 0.01 AI851899 Transmembrane protein 25 (Tmem25) À2.20 0.03 AF076482 Peptidoglycan recognition protein precursor (Pgrp) À2.23 0.04 U29678 Chemokine receptor 1 (CCR1) À2.47 0.04 U05265 Glycoprotein 49B (gp49B) À2.98 0.04 M65027 Glycoprotein 49A (gp49A) À3.37 0.01 L08115 CD9 À3.84 0.004 AI853884 Chemokine binding protein 2 À5.17 0.007

(Continued on following page)

Table 1. Differential Gene Expression Analysis of Murine CD19+ B Cells Overexpressing Bcl-2 (Cont’d)

Accession (Unigene) Gene/Protein Name Fold Change P

AF000236 RDC1 orphan chemokine receptor À5.89 0.008 AA822679 Hematopoietic cell signal transducer À5.97 0.01

Miscellaneous proteins and cDNA AV253089 cDNA 10.57 0.01 V00793 IgG1-C region 10.16 0.002 X67210 Rearranged IgG2b H-chain 4.11 0.01 AI451032 IgG1 H chain 4 3.90 0.04 AA416072 cDNA 3.73 0.03 D14625 IgG3 H chain 8 3.71 0.02 D78344 IgG 3.41 0.03 AW047643 cDNA 3.21 0.009 J00475 IgH DFL16.1 3.11 0.002 Ai850363 Muscle glycogen phosphorylase (Pygm) 2.92 0.04 AV038316 cDNA 2.79 0.03 AF002719 Secretory leukoprotease inhibitor (SLPi) 2.58 0.007 AW061234 cDNA 2.54 0.03 AI786089 Kininogen precursor 2.46 0.04 AA517032 cDNA 2.45 0.02 AA789553 Alstrom syndrome 1 (Alms1) protein homologue 2.39 0.02 M90766 Ig J chain 2.35 0.006 AI853664 cDNA 2.31 0.006 AV080003 IgH J558 family 2.25 0.03 AV258047 cDNA 2.23 0.03 J03482 Histone H1 gene 2.21 0.02 AV281523 cDNA 2.20 0.01 AV210037 cDNA 2.17 0.03 AF000913 WS2a43 mutated IgH 2.12 0.01 AI314284 hypothetical protein 2.00 0.02 AV259552 cDNA 2.00 0.03 AU045276 cDNA 2.00 0.04 AV320218 cDNA 2.00 0.03 M34597 Ig germ line E chain Vx-J2-C2 2.00 0.003 AV217136 cDNA 2.00 0.01 AV297816 cDNA 2.00 0.009 AV174430 cDNA 2.00 0.04 AV207625 Gene similar to protein phosphatase-2 inhibitor 2.00 0.02 AV225591 Protein similar to mouse glutathione peroxidase À2.00 0.03 AV310830 cDNA À2.00 0.03 AI448839 cDNA À2.00 0.01 M16819 Mouse tumor necrosis factor-h À2.00 0.02 AI462391 Hypothetical protein À2.07 0.03 AV235558 cDNA À2.09 0.008 AV012076 cDNA À2.10 0.02 AI152709 cDNA À2.10 0.04 AW045191 cDNA À2.17 0.006 AV101344 DNA ligase-3 À2.17 0.02 AI662280 cDNA À2.17 0.03 M60474 Myristoylated alanine-rich protein kinase C substrate À2.18 0.009 AV128327 cDNA À2.20 0.008 M19436 Myosin light chain À2.20 0.03 AF072697 Shyc À2.23 0.04 AI842144 cDNA À2.25 0.009 X51941 Methylmalonyl CoA mutase À2.25 0.04 AI627038 cDNA À2.32 0.03 AI481498 Procollagen, type V, a1 À2.34 0.009 AI182009 cDNA À2.35 <0.001 AV212587 cDNA À2.35 0.03 AV368065 Hypothetical protein À2.39 <0.001 AV229080 cDNA À2.41 0.03 AV332560 cDNA À2.48 0.04 AA671194 cDNA À2.51 0.01 X12905 Properdin factor À2.53 0.03 AI853854 ATP binding cassette, subfamily C protein À2.55 0.006 AI530075 cDNA À2.55 0.03 AI841689 Chemokine-like factor superfamily-3 À2.56 0.03 AW229312 cDNA À2.67 0.02 AW124025 Putative helicase-like protein non – Hodgkin’s lymphoma À2.72 0.03 AW214234 cDNA À2.74 0.01 AV365271 Nedd4 À2.77 <0.001 AV271750 cDNA À2.81 0.02 AA712022 cDNA À2.82 0.002 AI835567 Tubulin, g2 chain À2.86 <0.001

Table 1. Differential Gene Expression Analysis of Murine CD19+ B Cells Overexpressing Bcl-2 (Cont’d)

Accession (Unigene) Gene/Protein Name Fold Change P

AV092014 cDNA À2.87 0.04 AW228014 Hypothetical protein À2.93 0.001 M20878 TCR h chain, VDJ region À3.06 0.04 M21050 Lysozyme M À3.16 0.001 AI429433 cDNA À3.18 <0.001 X70057 Serine protease gene À3.67 0.04 AI159157 cDNA À4.20 0.02 AA289585 cDNA À4.60 0.008 AI450988 cDNA À4.64 0.02 U34277 Platelet-activating factor acetylhydrolase À4.87 0.03 AV260742 cDNA À4.91 0.04 X51547 Lysozyme P precursor À5.81 0.004 AI844675 cDNA À5.92 0.01 AI450144 cDNA À6.95 0.002 AV312050 cDNA À8.70 0.02 X15313 Myeloperoxidase À26.73 <0.001

NOTE: Comparison of composite arrays generated on CD19+ B cells obtained from both EA-Bcl-2 transgenic and transgene-negative LMC mice revealed 151 genes exhibiting z2-fold differential expression (P < 0.05), with 57 genes induced and 94 genes repressed. Genes are denoted according to their Genbank accession nos. Genes are grouped according to their function as reported.

JAWSII cells overexpressing the Bcl-2: EGFP construct within the follicular center cell region, was limited to neoplastic also revealed overexpression of SOCS3 protein compared with follicular lymphoma cells and colocalized with Bcl-2 in 9 of 12 cells expressing EGFP alone (Fig. 3B). In addition, to assess de novo follicular lymphoma cases examined (Fig. 4; Table 3). whether Bcl-2-associated induction of SOCS3 is linked to The antibodies stained mainly the nucleus of positive cells, with activation of signal transducer and activator of transcription slight cytoplasmic staining also noted in some cases. In con- (STAT) 3, we measured phospho-STAT3 levels relative to trast, SOCS3 protein was not detected by immunostaining in STAT3 in Bcl-2-overexpressing cells. When probed with germinal center follicular B cells from benign hyperplastic phospho-STAT3 antisera, both MEF-1 and JAWSII cells over- tonsil tissue (Fig. 4). SOCS3 staining was also noted in normal expressing the Bcl-2:EGFP construct exhibited overexpression as well as neoplastic follicular lymphoma cells outside the fol- of phospho-STAT3 protein compared with their respective EGFP licular center cell region. control cells, which failed to express detectable levels of phospho-STAT3 protein (Fig. 3C). Western blots were then Discussion stripped and the same blots were reprobed with STAT3 antisera, Although t(14;18) represents an early initiating genetic event revealing no difference in the levels of STAT3 protein between required for the development of follicular lymphoma, it is clear Bcl-2-overexpressing and control cells (Fig. 3C). Probing with that cooperating genetic errors are required to deregulate antisera recognizing the SOCS family members CIS, SOCS1, cellular pathways for neoplastic transformation to occur. In our and SOCS5 failed to reveal detectable protein expression of these search for genes that cooperate with Bcl-2 to mediate neoplastic specific family members in either MEF-1 or JAWSII cells overexpressing Bcl-2:EGFP (data not shown). These data Table 2. Selected Differential Gene Expression in EA-Bcl-2 indicate that the induction of SOCS3 is associated with Transgenic Mice as Determined by Microarray Analysis, overexpression of Bcl-2 and not simply a physiologic response RT-PCR, and RNA Protection Assay to increased cell turnover. Furthermore, overexpression of Bcl-2 mediates SOCS3 induction via cellular pathways linked to Gene Microarray P RT-PCR RNA Protection Assay activation of STAT3. SOCS3 2.05 0.003 5 5 HSP1a 11.12 <0.001 ND ND SOCS3 Is Expressed at High Levels in a Cohort of IgG1-C 10.16 0.002 10 ND IgG2b H chain 4.11 0.01 10 ND Patients with De novo Follicular Lymphoma CCR9 2.37 0.03 7 ND To determine whether Bcl-2-associated induction of SOCS3 Bcl-2 a exon 2 2.00 0.03 2 ND may occur in human follicular lymphoma, we measured SOCS3 Blimp 1 2.00 0.01 5 ND c-myb À2.00 0.04 À4 ND protein levels by immunohistochemistry in paraffin-embedded CR2 À2.00 <0.001 À4 ND biopsies from 12 patients diagnosed with de novo follicular BID À2.00 0.01 À2 ND c-myc À2.04 0.01 À3 ND lymphoma prior to the initiation of therapy. Follicular lym- NEDD4 À3.25 0.01 À4 ND phoma tissue specimens were diagnosed as either histologic Cyclin D2 À5.92 0.04 À8 ND grades I or II, and all harbored t(14;18) with concomitant marked overexpression of Bcl-2 in the follicular center cell NOTE: Microarray, RT-PCR, and RNA protection assay data reflect the fold change for the average of EA-Bcl-2 transgenic animals examined calculated region (Fig. 4). Immunostaining with two distinct antibodies to relative to the signal observed for the average of LMC control samples. ND, not SOCS3 revealed marked overexpression of SOCS3 protein that, determined.

FIGURE 2. SOCS3 protein is induced in distinct strains of EA-Bcl-2 transgenic versus LMC mice. Western blot for SOCS3 on whole protein extracts (30 Ag/lane) from CD19+ B cells isolated from individual mice from two distinct strains of EA-Bcl-2 transgenic mice and their respective transgene-negative LMC mice. Top, Western blot for SOCS3 in EA-Bcl-2 transgenic strain 1 (lanes a, c, d, and f), LMC strain 1 (lanes b and e), EA-Bcl-2 transgenic strain 2 (lanes g and i), and LMC strain 2 (lanes h and j). Bottom, Western blot for actin (lanes a-j) to confirm equivalent protein loading.

transformation in follicular lymphoma, our analysis revealed comparisons. Second, differences exist between studies that overexpression of Bcl-2 was associated with induction of concerning the cell type examined. Our arrays were done on the SOCS3 gene in both murine CD19+ B cells and human primary murine CD19+ B cells overexpressing Bcl-2. Bohen follicular lymphoma tissue specimens. To our knowledge, this et al. (25) noted SOCS3 overexpression in follicular lymphoma is the first description of an association between Bcl-2 and tissue from a group of patients noted to be nonresponders to SOCS3 as well as the first report of the induction of SOCS3 rituximab. However, in this study, interpretation of differential RNA and protein in purified populations of murine B cells and gene expression may be complicated by the fact that analysis human de novo follicular lymphoma cells. Our data suggest was done on a mixed population of cells rather than purely that SOCS3 transcript levels in normal B cells are barely isolated neoplastic B cells. In comparison with studies detectable, with transcriptional induction noted only in con- examining human follicular lymphoma cells (23, 26, 27) or junction with overexpression of Bcl-2. The finding that SOCS3 cultured t(14;18)+ cell lines (24), which have not revealed expression colocalized with that of Bcl-2 and was noted pri- SOCS3 induction, SOCS3 gene induction in murine B cells marily throughout neoplastic germinal center follicles indicates examined in our study may reflect differences between murine that overexpression of SOCS3 originated from malignant follic- and human B-cell biology or differences inherent to the ular lymphoma B cells and not from associated normal cells. developmental stage of the B cell examined. Finally, selection Previous studies examining gene expression in follicular biases resulting from heterogeneity in patient populations and lymphoma have reported discordant results regarding SOCS3 tumor biology as well as of Bcl-2 expression levels may induction, a finding potentially due to multiple factors. First, contribute to variable levels of SOCS3 expression. studies have used different microarray platforms as well as The SOCS3 gene is a member of a family of cytokine varied statistical methodologies for determining differential suppressors that inhibit cytokine-mediated signaling via classic gene expression, thus resulting in difficulty with database negative feedback loop inhibition (30-34). Transcripts

FIGURE 3. Overexpression of Bcl-2 induces SOCS3 protein levels in MEF-1 and JAWSII cells via activation of STAT3. Both MEF-1 and JAWSII cells were transduced with a retroviral construct harboring either a fusion Bcl-2:EGFP or EGFP alone, and whole protein lysates were isolated at 48 hours. Western blot for Bcl-2, SOCS3, or phospho-STAT3 was then done on whole proteinlysates (30 Ag/lane). A. Top, Western blot for Bcl-2 in MEF-1 cells harboring Bcl-2:EGFP (lane a) and EGFP control (lane b) and JAWSII cells harboring Bcl-2:EGFP (lane c) and EGFP control (lane d); bottom, Western blot for actin (lanes a-d) to confirm equivalent protein loading. B. Top, Western blot for SOCS3 in MEF-1 cells containing Bcl-2:EGFP (lane a) and EGFP control (lane b) and JAWSII cells containing EGFP control (lane c) and Bcl- 2:EGFP (lane d); bottom, Western blot for actin (lanes a-d) to confirm equivalent protein loading. C. Top, Western blot for phospho-STAT3 in both MEF-1 cells (lane a) and JAWSII cells (lane c) containing Bcl-2:EGFP and in respective EGFP control cells (lanes b and d); bottom, Western blot for STAT3 (lanes a-d).

FIGURE 4. Immunostaining reveals overexpression of SOCS3 in de novo follicular lymphoma. All staining was done on paraffin biopsies by Vectastain ABC detection. Representative case of de novo follicular lymphoma (no. 3 in Table 3) showing (A and B) Bcl-2-positive staining of neoplastic follicular lymphoma cells within the follicular center cell region. Concomitant SOCS3-positive staining is seen within the follicular center cell region limited to neoplastic follicular lymphoma cells using (C and D) anti-SOCS3 antibody 1 (Zymed) and (E) anti-SOCS3 antibody 2 (Santa Cruz Biotechnology). F and G. Representative case of benign hyperplastic tonsil germinal center B cells showing negative staining for SOCS3. Original magnification, Â10 (A, C, and F) and Â50 (B, D, E, and G).

encoding various SOCS family members are typically present at influence over cellular responses to IL-6 (33). Specificity of very low or undetectable levels but are rapidly up-regulated in SOCS3 activity may thus be nonredundant and dependent on response to a wide range of cytokines and hormones (33). The specific cytokine receptor interactions, thus possibly revealing expression of SOCS3 is tightly controlled via transcriptional a central role for SOCS3 in directing gp130-related cytokines regulation primarily mediated through STAT family proteins toward either proinflammatory or anti-inflammatory cellular STAT3 and STAT5 (35-38). SOCS3-mediated feedback responses. inhibition is primarily regulated via signaling through Janus- Although SOCS3 would seem to negatively regulate activated kinase and STAT family proteins and has been inflammatory responses (33), its role in tumorigenesis and the extensively reviewed elsewhere (33, 39-43). SOCS3 expres- underlying mechanisms that regulate its expression in B cells sion has been primarily noted in murine T cells and has been remain to be defined. Several studies have examined SOCS3 shown to regulate both T-cell development and activation via expression in a diverse group of tumors of hematopoietic cell numerous mechanisms (44-47). The recent generation of origin. Chronic myelogenous leukemia cell lines as well as mice lacking functional SOCS3 in hepatocytes, macrophages, leukemic cells from patients with chronic myelogenous and neutrophils reveals SOCS3 to be an essential regulator of leukemia blast crisis were noted to constitutively express interleukin-6 (IL-6) signaling via mediation of gp130-related SOCS3, resulting in attenuation of IFN-a signaling and re- cellular signaling pathways (48-50) as well as a negative sistance to its antiproliferative effects (52). Similarly, SOCS3 regulator of granulocyte colony-stimulating factor signaling overexpression in cancer cells derived from patients with (51). In addition to its role as a suppressor of IL-6-mediated cutaneous T-cell lymphoma was found dependent on aberrant signaling, SOCS3 may also have qualitative and quantitative expression of STAT3, representing a pathway that also results

Table 3. SOCS3 Immunostaining Patterns of Neoplastic B pathways important for tumorigenesis and does not serve as Cells in Paraffin-Embedded Biopsies from Cases of a tumor suppressor gene. On the other hand, similar to its Follicular Lymphoma emerging role in immunity as a negative regulator of cellular signaling that mediate inflammatory responses, SOCS3 over- Follicular Lymphoma Case Anti-SOCS3 Antibody 1 Anti-SOCS3 Antibody 2 expression in de novo follicular lymphoma may serve a regulatory role to suppress the proliferative capacity of the 1++ 2++ neoplastic clone and select for a more indolent lymphoma 3++ phenotype. In this scenario, the loss of SOCS3 induction may 4 ÀÀ 5++ then predispose to a more aggressive phenotype such as seen 6+/À +/À in transformed follicular lymphoma. Determining whether 7++ SOCS3 induction is present in intermediate and high-grade 8 ÀÀ 9 ÀÀ non–Hodgkin’s lymphoma subtypes and whether it carries 10 + + prognostic significance will help discern whether SOCS3 11 +/À +/À overexpression influences neoplastic transformation or acts 12 + + as a tumor suppressor. NOTE: Immunostaining using two distinct SOCS3 antisera: anti-SOCS3 antibody Taken together, our study suggests that the induction of 1 (Zymed) and anti-SOCS3 antibody 2 (Santa Cruz Biotechnology). À, all tumor SOCS3 in B cells is an early genetic event mediated by cells negative; +/À, staining in >50% of the tumor cells; +, staining in >90% of overexpression of Bcl-2 and that SOCS3 may cooperate with the tumor cells. Bcl-2 in deregulating cellular pathways important for the pathogenesis of de novo follicular lymphoma in humans. in decreased responsiveness of cutaneous T-cell lymphoma Examination of the cellular pathways by which Bcl-2 over- cells to IFN-a (53). In addition, acute myeloid leukemia cells expression leads to the induction of SOCS3 and other bearing IL-6-mediated constitutive STAT3 phosphorylation downstream effectors affected by deregulated expression of were found to also constitutively overexpress SOCS1 and SOCS3 should provide important insight into the genesis of SOCS3 (54). In contrast to our study, oligonucleotide micro- follicular lymphoma in humans as well as identify potential array analysis of IL-6-dependent multiple myeloma cells novel signaling intermediaries that lend to the development of revealed STAT3-mediated induction of SOCS3 via Bcl-2- novel targeted therapies. independent cellular pathways (55). It is possible that Bcl-2- associated induction of SOCS3 is restricted to an earlier stage of B-cell development rather than in late-stage plasma cells or Materials and Methods memory B cells. Collectively, these studies indicate that IL-6- B-Cell Isolation dependent, STAT3-mediated pathways serve as important regu- Single cell suspensions were prepared from individual lators of SOCS3 expression levels in diverse hematopoietic spleens from two distinct strains of male and female 24-week- A tumors. old E -Bcl-2 transgenic mice (28, 29) and 24-week-old The cellular pathways required for Bcl-2-mediated induction transgene-negative LMC mice (C57BL/6 strain). Naive un- of SOCS3, as well as the degree to which SOCS3 induction touched B cells were isolated from murine spleen cells by influences the biology of de novo follicular lymphoma in negative selection using an immunomagnetic bead B-Cell humans, remain to be elucidated. Given that Bcl-2 is not a Isolation Kit (Miltenyi Biotec, Inc., Auburn, CA) according to transcription factor, Bcl-2 overexpression likely induces the manufacturer’s instructions. Following immunomagnetic SOCS3 indirectly by modulating pathways that deregulate bead isolation, a small aliquot of cells was phenotyped by flow factors necessary for transcriptional up-regulation of SOCS3. It cytometry to assess B-cell purity. remains to be determined whether Bcl-2 and SOCS3 function in cooperation to cause an oncogenic hit important for neoplas- Flow Cytometry and Antibodies tic transformation of B cells or whether SOCS3 may be Immunophenotyping was done on cell suspensions using a suppressing propagation of malignant B cells harboring Bcl-2 FITC-conjugated monoclonal antibody directed against murine overexpression. It is well known that forced expression of CD19 (BD Biosciences/PharMingen, San Diego, CA). An oncogenes in concert with Bcl-2 overexpression cooperate to irrelevant isotype-matched control antibody was used in all deregulate pathways that speed the pace of tumorigenesis as experiments. Analysis was done within 1 hour using a dual- well as influence morphology of the neoplastic clone. laser FACSCalibur instrument (Becton-Dickinson, Franklin Transgenic animals with combined overexpression of c-myc Lakes, NJ). and Bcl-2 develop lymphomas of a more primitive histology and with markedly decreased latency compared with transgenic RNA Isolation animals solely overexpressing Bcl-2 (13), most likely reflecting Total RNA was prepared from B cells isolated as above combined deregulation of apoptotic and cell cycle pathways. It using the RNeasy Midi Kit (Qiagen, Valencia, CA) according to remains to be determined whether SOCS3 might also cooperate the manufacturer’s instructions. RNA quality was examined with Bcl-2 in affecting neoplastic transformation of B cells. The by the RNA 6000 LabChip Kit on the 2100 bioanalyzer finding that SOCS3 is overexpressed in our cohort of follicular (Agilent Technologies, Palo Alto, CA). Quantity and absor- lymphoma and is also overexpressed in other hematopoietic bance at 260/280 nm of total and cRNA were assessed by tumors suggests that SOCS3 deregulation activates cellular UV spectrophotometer.

Gene Expression Analysis by DNA Oligonucleotide RiboQuant kit (BD Biosciences/PharMingen) according to the Arrays manufacturer’s instructions. Following electrophoresis on a 5% Double-stranded cDNA was synthesized from total RNA, polyacrylamide gel containing urea (8 mol/L), radiolabeled amplified as cRNA, labeled with biotin, and hybridized to bands from experimental and control sample lanes were Affymetrix murine U74v2 A, B, and C Array chipsets, which quantitated using PhosphorImager and normalized to the values were washed and scanned at the University of Washington’s for glyceraldehyde-3-phosphate dehydrogenase and L32 in the Center for Expression Arrays according to procedures devel- same samples. oped by the manufacturer. Image processing was done using Affymetrix MAS-5 software. The quality of hybridization and Cell Culture overall chip performance was determined by visual inspection MEF-1 cell lines (American Type Culture Collection, of the raw scanned data and the MAS-5 generated report file. Manassas, VA) were cultured in DMEM with glucose (4.5 g/L) The raw data were loaded into the Rosetta Resolver Gene enriched with 10% heat-inactivated fetal bovine serum (Hyclone Expression Data Analysis System (56, 57) via standard Laboratories, Logan, UT), L-glutamine (2 mmol/L), nonessential methods. Using the Resolver system, the normalized intensity amino acids mixture (100Â), and sodium pyruvate (1 mmol/L) data from all control experiments and from all transgenic in the presence of penicillin (100 units/mL) and streptomycin experiments were summarized as one combined control (100 Ag/mL), all purchased from BioWhittaker, Inc. (Walkers- intensity experiment and one combined transgenic experiment, ville, MD). Mouse bone marrow cells (JAWSII; American Type respectively. These combined experiments take the spread and Culture Collection) were cultured in a-MEM with ribonucleo- distribution of the individual experiments into consideration, sides, deoxyribonucleosides enriched with 20% heat-inactivated hence facilitating our analysis without losing information. fetal bovine serum (Hyclone Laboratories), L-glutamine Resolver uses an error-weighed approach to compute expres- (4 mmol/L), and sodium pyruvate (1 mmol/L) in the presence sion log ratios for each probe set based on the spread of the of granulocyte macrophage colony-stimulating factor (5 ng/mL; replicate measurements. The software then computes a con- R&D Systems, Inc., Minneapolis, MN). fidence level, called P, for each probe set based on this error estimate. Background correction in the Resolver system is done on individual perfect match and mismatch probes. Resolver Construction of Expression Vectors adopts an error model (56) and a background correction A cDNA encoding the intron-less open reading frame of the strategy in estimating the probe set intensity levels. Their error 717-bp human Bcl-2a (pORF-hBcl-2; InvivoGen, San Diego, model is derived from extensive like-versus-like experiments, CA) was cloned into shuttle plasmid SL1180 (Amersham and their background correction approach uses local back- Pharmacia Biotech, Piscataway, NJ) using the NcoI and NheI ground estimates for probe sets in different regions of the chip. (New England Biolabs, Inc., Beverly, MA) restriction enzyme In effect, the error model minimizes false-positives, particularly sites. The pORF-hBcl-2 was subsequently cloned into the at low expression values. EcoRI and XhoI (New England Biolabs) sites on the multiple cloning region of the bicistronic retroviral expression plasmid, pBMN-IRES-EGFP (kindly provided by Dr. Garry Nolan, Real-time Quantitative PCR Analysis Stanford University, Palo Alto, CA). High-titer, second- Real-time quantitative RT-PCR analysis was done using a generation helper free retrovirus was produced by calcium LightCycler (Roche Diagnostics, Basel, Switzerland). Reverse phosphate–mediated transfection of the Phoenix ecotropic transcription was done by using SuperScript II (Invitrogen, packaging cell line (American Type Culture Collection) with Carlsbad, CA). PCR primers were designed with MacVector either 24 Ag of the hBcl-2 expression plasmid or pBMN-IRES- software (Accelrys, San Diego, CA). The nucleotide sequences EGFP control plasmid. Recombinant retroviral supernatant of the primer pairs are available on request. PCR reactions were was collected 48 hours after transfection and filtered through optimized using the FastStart DNA Master SYBR Green I Kit a Millex-HV 0.45 Am filter (Millipore Corp., Bedford, MA). (Roche Diagnostics) after verifying that no amplification was For transduction, cell culture medium from f70% confluent noted in the no-template controls. To ensure that any DNA MEF-1 cells or JAWSII cells in six-well plates (Corning Inc., contamination was removed by DNase I treatment of total Corning, NY) were replaced with 2.5 mL of retrovirus RNA, real-time quantitative RT-PCR was done on non-reverse- supernatant and centrifuged for 2 hours (1,430 Â g at 32jC) transcribed RNA. No amplification was observed in these and then incubated for 10 hours (5% CO2,37jC). On com- conditions for differentially expressed genes examined. The pletion of the incubation period, retroviral supernatant was size of the PCR product for each gene was verified by gel replaced by appropriate normal growth medium for each cell electrophoresis. Signals for genes from each RNA sample were type. Cells were sorted for stable retrovirus transfection based normalized to that sample’s signal for glyceraldehyde-3- on EGFP expression using a FACSVantage SE cell sorter phosphate dehydrogenase. The fold change for experimental (Becton-Dickinson). samples was calculated relative to the signal observed for control samples. Western Blot Analysis Cell lysis and preparation of whole protein lysates were RNase Protection Assay done as described (58). Proteins were separated by SDS-PAGE 32P-labeled riboprobes were incubated with total RNA and blotted onto nitrocellulose membrane. Filters were probed (10 Ag) and then subjected to RNase digestion using a with primary antibodies recognizing either SOCS3 or SOCS1

(Zymed, South San Francisco, CA), CIS (Novus Biologicals, 8. Yang E, Korsmeyer SJ. Molecular thanatopsis: a discourse on the BCL2 family and cell death. Blood 1996;88:386 – 401. Littleton, CO), SOCS5 (Imgenex, San Diego, CA), STAT3 or 9. Vaux DL, Cory S, Adams JM. Bcl-2 gene promotes haemopoietic cell survival phospho-STAT3 (Upstate, Charlottesville, VA), or Bcl-2 (BD and cooperates with c-myc to immortalize pre-B cells. Nature 1988;335:440 – 2. Biosciences/PharMingen) followed by horseradish peroxidase– 10. McDonnell TJ, Deane N, Platt FM, et al. Bcl-2-immunoglobulin transgenic conjugated anti-rabbit IgG (BD Biosciences/PharMingen) and mice demonstrate extended B cell survival and follicular lymphoproliferation. detected using an enhanced chemiluminescence kit (Amersham Cell 1989;57:79 – 88. Pharmacia Biotech). 11. McDonnell TJ, Korsmeyer SJ. Progression from lymphoid hyperplasia to high-grade malignant lymphoma in mice transgenic for the t(14:18). Nature 1991;349:254 – 6. Tissue Specimens 12. Strasser A, Harris AW, Cory S. EA-bcl-2 transgene facilitates spontaneous transformation of early pre-B and immunoglobulin secreting cells but not T cells. Paraffin-embedded biopsies of newly diagnosed, de novo Oncogene 1993;8:1 – 9. follicular lymphoma and benign hyperplastic tonsil were 13. Strasser A, Harris AW, Bath ML, Cory S. Novel primitive lymphoid tumours obtained from the Critical Technologies Shared Resource of induced in transgenic mice by cooperation between myc and Bcl-2. Nature 1990; the Yale Cancer Center according to approved Human 348:331 – 3. Investigation Committee protocols. In each case, the diagnosis 14. Limpens J, Stad R, Vos C, et al. Lymphoma-associated translocation t(14;18) in blood B cells of normal individuals. Blood 1995;85:2528 – 36. had been made based on conventional histologic and 15. Paltiel O, Zelenets A, Sverdlin I, Gordon L, Ben-Yehuda D. Translocation immunohistologic examination according to the criteria of the t(14;18) in healthy individuals: preliminary study of its association with family WHO classification (59). history and agricultural exposure. Ann Oncol 2000;11:78 – 80. 16. Aster J, Kobayashi Y, Shiota M, Mori S, Sklar J. Detection of the t(14;18) at similar frequencies in hyperplastic lymphoid tissues from American and Japanese Immunohistochemistry patients. Am J Pathol 1992;141:291 – 9. Slides containing 4-Am tissue sections were subjected to a 17. Liu Y, Hernandez A, Shibata D, Cortopassi G. Bcl2 translocation frequency conventional antigen retrieval protocol for 2 minutes using a rises with age in humans. Proc Natl Acad Sci U S A 1994;91:8910 – 4. pressure cooker and prepared as described (60). Slides were 18. Lancry L, Roulland S, Roue G, et al. No BCL-2 protein over expression but j BCL-2/IgH rearrangements in B cells of patients with persistent polyclonal B-cell then incubated overnight at 4 C with one of two distinct lymphocytosis. Hematol J 2001;2:228 – 33. antibodies to SOCS3 [a rabbit polyclonal antibody to SOCS3 19. Lestou VS, Gascoyne RD, Sehn L, et al. Muticolour fluorescence in situ (Zymed) and a goat polyclonal antibody to SOCS3 (Santa Cruz hybridization analysis of t(14;18)-positive follicular lymphoma and correla- Biotechnology, Santa Cruz, CA)] or Bcl-2 (BD Biosciences/ tion with gene expression data and clinical outcome. Br J Haematol 2003;122: 745 – 59. PharMingen) followed by detection the next day using a 20. Alizadeh AA, Eisen MB, Davis ER, et al. Distinct types of diffuse large Vectastain ABC detection kit (Vector Laboratories, Burlingame, B-cell lymphoma identified by gene expression profiling. Nature 2003;403: CA) according to the manufacturer’s instructions. Sections were 503 – 11. stained in parallel without primary antibody to provide a 21. Shipp MA, Ross KN, Tamayo P, et al. Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine negative control for each reaction. Two authors of this study learning. Nat Med 2002;8:68 – 74. (G.J.V. and A.W.Z.) independently evaluated the immunostain- 22. Huang JZ, Sanger WG, Greiner TC, et al. The t(14;18) defines a unique ing results. subset of diffuse large B-cell lymphoma with a germinal center B-cell gene expression profile. Blood 2002;1:2285 – 90. 23. Husson H, Carideo EG, Neuberg D, et al. Gene expression profiling of Acknowledgments follicular lymphoma and normal germinal center B cells using cDNA arrays. We thank Noel Blake for assistance with flow cytometry; Vicki Morgan- Blood 2002;99:282 – 9. Stephensen, Annie Minard, and Kristine Eiting for technical assistance; Drs. David Rimm and Robert Camp (Yale Department of Pathology) for providing 24. Robetorye RS, Bohling SD, Morgan JW, Fillmore GC, Lim MS, Elenitoba- tissue specimens and helpful advice; and Dr. Nancy Berliner for critical review of Johnson KS. Microarray analysis of B-cell lymphoma cell lines with the t(14;18). the article. J Mol Diagn 2002;4:123 – 36. 25. Bohen SP, Troyanskaya OG, Alter O, et al. Variation in gene expression patterns in follicular lymphoma and the response to rituximab. Proc Natl Acad Sci References 2003;100:1926 – 30. 1. Harris N, Jaffe E, Stein H, et al. A revised European-American classification 26. de Vos S, Hofmann WK, Grogan TM, et al. Gene expression profile of serial of lymphoid neoplasms: a proposal from the International Lymphoma Study samples of transformed B-cell lymphomas. Lab Invest 2003;83:271 – 85. Group. Blood 1994;84:1361 – 92. 27. Lossos IS, Alizadeh AA, Diehn M, et al. Transformation of follicular 2. Horning SJ. Natural history of and therapy for the indolent non-Hodgkin’s lymphoma to diffuse large-cell lymphoma: alternative patterns with increased or lymphomas. Semin Oncol 1993;20:75 – 88. decreased expression of c-myc and its regulated genes. Proc Natl Acad Sci U S A 3. Ersboll J, Schultz HB, Pedersen-Bjergaard J, Nissen NI. Follicular low-grade 2002;99:8886 – 91. non-Hodgkin’s lymphoma: long term outcome with or without tumor progression. 28. Strasser A, Whittingham S, Vaux DL, et al. Enforced BCL2 expression in Eur J Haematol 1989;42:155 – 63. B-lymphoid cells prolongs antibody responses and elicits autoimmune disease. 4. Finger LR, Harvey RC, Moore RCA, Showe LC, Croce CM. A common Immunology 1991;88:8661 – 5. mechanism of chromosomal translocation in T- and B-cell neoplasia. Science 29. Strasser A, Harris AW, Corcoran LM, Cory S. Bcl-2 expression promotes 1986;234:982 – 5. B- but not T-lymphoid development in scid mice. Nature 1994;368:457 – 60. 5. Bakhshi A, Jensen JP, Goldman P, et al. Cloning the chromosomal breakpoint 30. Starr R, Willson TA, Viney EM, et al. A family of cytokine-inducible of t(14;18) human lymphomas: clustering around JH on chromosome 14 and near inhibitors of signalling. Nature 1997;387:917 – 21. a transcriptional unit on 18. Cell 1985;41:899 – 906. 31. Naka T, Narazaki M, Hirata M, et al. Structure and function of a new STAT- 6. Cleary ML, Sklar J. Nucleotide sequence of a t(14;18) chromosomal induced STAT inhibitor. Nature 1997;387:924 – 9. breakpoint in follicular lymphoma and demonstration of a breakpoint cluster region near a transcriptionally active locus on chromosome 18. Proc Natl Acad 32. Endo TA, Masuhara M, Yokouchi M, et al. A new protein containing an SH2 Sci U S A 1985;82:7439 – 43. domain that inhibits JAK kinases. Nature 1997;387:921 – 4. 7. Tsujimoto Y, Gorham J, Cossman J, Jaffe E, Croce CM. The t(14;18) 33. Alexander WS, Hilton DJ. The role of cytokine signaling (SOCS) proteins in chromosome translocations involved in B-cell neoplasms result from mistakes in regulation of the immune response. Annu Rev Immunol 2004;22:503 – 29. VDJ joining. Science 1985;22:1390 – 3. 34. Hilton DJ, Richardson RT, Alexander WS, et al. Twenty proteins containing a

C-terminal SOCS box form five structural classes. Proc Natl Acad Sci U S A 47. Seki Y, Inoue H, Nagata N, et al. SOCS-3 regulates onset and maintenance of 1998;95:114 – 9. TH2-mediated allergic responses. Nat Med 2003;9:1047 – 54. 35. Davey HW, McLachlan MJ, Wilkins RJ, Hilton DJ, Adams TE. STAT5b 48. Lang R, Pauleau A-L, Parganas E, et al. SOCS3 regulates the plasticity of mediates the GH-induced expression of SOCS-2 and SOCS-3 mRNA in the liver. gp130 signaling. Nat Immunol 2003;4:546 – 50. Mol Cell Endocrinol 1999;158:111 – 6. 49. Croker BA, Krebs DL, Zhang J-G, et al. SOCS3 negatively regulates IL-6 36. Auernhammer CJ, Bousquet C, Melmed S. Autoregulation of pituitary signaling in vivo. Nat Immunol 2003;4:540 – 5. corticotroph SOCS-3 expression: characterization of the murine SOCS-3 50. Yasukawa H, Ohishi M, Mori H, et al. IL-6 induces an anti-inflammatory promoter. Proc Natl Acad Sci U S A 1999;96:6964 – 9. response in the absence of SOCS3 in macrophages. Nat Immunol 2003;4:551 – 6. 37. Emanuelli B, Peraldi P, Filloux C, et al. SOCS-3 is an insulin-induced 51. Croker BA, Metcalf D, Robb L, et al. SOCS3 is a critical physiological negative regulator of insulin signaling. J Biol Chem 2000;275:15985 – 91. negative regulator of G-CSF signaling and emergency granulopoiesis. Immunity 38. He B, You L, Uematsu K, et al. Cloning and characterization of a functional 2004;20:153 – 65. promoter of the human SOCS-3 gene. Biochem Biophys Res Commun 2003; 52. Sakai I, Takeuchi K, Yamauchi H, Narumi H, Fujita S. Constitutive 301:386 – 91. expression of SOCS3 confers resistance to IFN-a in chronic myelogenous 39. O’Shea JJ, Gadina M, Schreiber RD. Cytokine signaling in 2002: new leukemia cells. Blood 2002;100:2926 – 31. surprises in the JAK/STAT pathway. Cell 2002;109:5121 – 31. 53. Brender C, Nielsen M, Kaltoft K, et al. STAT3-mediated constitutive 40. Yasukawa H, Misawa H, Sakamoto H, et al. The JAK-binding protein JAB expression of SOCS-3 in cutaneous T-cell lymphoma. Blood 2001;97:1056 – 62. inhibits JAK tyrosine kinase activity through binding in the activation loop. 54. Schuringa JJ, Wierenga AT, Kruijer W, Vellenga E. Constitutive stat3, tyr705, EMBO J 1999;18:1309 – 20. and ser727 phosphorylation in acute myeloid leukemia cells caused by the 41. Sasaki A, Yasukawa H, Shouda T, Kitamura T, Dikic I, Yoshimura A. CIS3/ autocrine secretion of interleukin-6. Blood 2000;95:3765 – 70. SOCS-3 suppresses erythropoietin (EPO) signaling by binding the EPO receptor 55. Brocke-Heidrich K, Kretzschmar AK, Pfeifer G, et al. Interleukin-6- and JAK2. J Biol Chem 2000;275:29338 – 47. dependent gene expression profiles in multiple myeloma INA-6 cells reveal a 42. Nicholson SE, De Souza D, Fabri LJ, et al. Suppressor of cytokine signaling- Bcl-2 family-independent survival pathway closely associated with Stat3 ac- 3 preferentially binds to the SHP-2 binding site on the shared cytokine receptor tivation. Blood 2004;103:242 – 51. subunit gp130. Proc Natl Acad Sci U S A 2000;97:6493 – 8. 56. Hughes TR, Roberts CJ, Dai H, et al. Widespread aneuploidy revealed by 43. Lehmann U, Schmitz J, Weissenbach M, et al. SHP2 and SOCS3 contribute DNA microarray expression profiling. Nat Genet 2000;25:333 – 7. to tyr-759-dependent attenuation of interleukin-6 signaling through gp130. J Biol 57. Schadt EE, Monks SA, Drake TA, et al. Genetics of gene expression Chem 2003;278:661 – 71. surveyed in maize, mouse and man. Nature 2003;422:297 – 302. 44. Egwuagu CE, Yu C-R, Zhang M, Mahdi RM, Kim SJ, Gery I. Suppressors of 58. Drachman JG, Griffin JD, Kaushansky K. The c-mpl ligand (thrombopoietin) cytokine signaling proteins are differentially expressed in TH1 and TH2 cells: stimulates tyrosine phosphorylation of JAK2, Shc, and c-mpl. J Biol Chem 1995; implications for TH cell lineage commitment and maintenance. J Immunol 2002; 270:4979 – 82. 168:3181 – 7. 59. Jaffe ES, Harris NL, Stein H, Wardiman JW. WHO classification of tumours. 45. Cohney SJ, Sanden D, Cacalano NA, et al. SOCS-3 is tyrosine pathology and genetics: tumours of haematopoietic and lymphoid tissues. Lyon phosphorylated in response to interleukin-2 and suppresses STAT5 phosphory- (France): IARC Press; 2001. lation and lymphocyte proliferation. Mol Cell Biol 1999;19:4980 – 8. 60. Kluger HM, Dolled-Filhart M, Rodov S, Kacinski BM, Camp RL, Rimm DL. 46. Banerjee A, Banks AS, Nawijn MC, Chen XP, Rothman PB. Cutting edge: Macrophage colony-stimulating factor-1 receptor expression is associated with suppressor of cytokine signaling 3 inhibits activation of NFATp. J Immunol 2002; poor outcome in breast cancer by large cohort tissue microarray analysis. Clin 168:4277 – 81. Cancer Res 2004;10:173 – 7.

Mol Cancer Res 2004;2(11). November 2004 Downloaded from mcr.aacrjournals.org on September 25, 2021. © 2004 American Association for Cancer Research. Bcl-2 Overexpression Leads to Increases in Suppressor of Cytokine Signaling-3 Expression in B Cells and De novo Follicular Lymphoma 11NIH grants CA78254 (G.J. Vanasse) and 5U24DK058813-02 (K.Y. Yeung), American Society of Hematology fellow scholar grant (G.J. Vanasse), and NIH research grant CA-16359 from the National Cancer Institute. Notes: G.J. Vanasse is a past American Society of Hematology fellow scholar and member of the Yale Cancer Center.

Gary J. Vanasse, Robert K. Winn, Sofya Rodov, et al.

Mol Cancer Res 2004;2:620-631.

Updated version Access the most recent version of this article at: http://mcr.aacrjournals.org/content/2/11/620

Cited articles This article cites 57 articles, 25 of which you can access for free at: http://mcr.aacrjournals.org/content/2/11/620.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/2/11/620.full#related-urls

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://mcr.aacrjournals.org/content/2/11/620. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.