CXC Chemokine Ligand 13 and CC Chemokine Ligand 19 Cooperatively Render Resistance to Apoptosis in Lineage Acute and Chronic Lymphocytic This information is current as CD23+CD5+ B Cells of September 28, 2021. Hu Chunsong, He Yuling, Wang Li, Xiong Jie, Zhou Gang, Zhang Qiuping, Gao Qingping, Zhang Kejian, Qiao Li, Alfred E. Chang, Jin Youxin and Tan Jinquan

J Immunol 2006; 177:6713-6722; ; Downloaded from doi: 10.4049/jimmunol.177.10.6713 http://www.jimmunol.org/content/177/10/6713 http://www.jimmunol.org/ References This article cites 54 articles, 31 of which you can access for free at: http://www.jimmunol.org/content/177/10/6713.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

by guest on September 28, 2021 • No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

CXC Chemokine Ligand 13 and CC Chemokine Ligand 19 Cooperatively Render Resistance to Apoptosis in B Cell Lineage Acute and Chronic Lymphocytic Leukemia CD23؉CD5؉ B Cells1

Hu Chunsong,2*† He Yuling,2† Wang Li,2† Xiong Jie,†‡ Zhou Gang,† Zhang Qiuping,† Gao Qingping,§ Zhang Kejian,§ Qiao Li,†¶ Alfred E. Chang,¶ Jin Youxin,‡ and Tan Jinquan3*†

CXCL13/CXCR5 and CCL19/CCR7 play a quite important role in normal physiological conditions, but the functions of both chemokine/ pairs in pathophysiological events are not well-investigated. We have investigated expression and functions of CXCL13/CXCR5 and CCL19/CCR7 in CD23؉CD5؉ and CD23؉CD5؊ B cells from cord blood (CB) and patients with B cell lineage acute or chronic lymphocytic leukemia (B-ALL or B-CLL). CXCR5 and CCR7 are selectively expressed on B-ALL, ؉ ؉ ؉ ؊ B-CLL, and CB CD23 CD5 B cells at high frequency, but not on CD23 CD5 B cells. Although no significant chemotactic Downloaded from responsiveness was observed, CXCL13 and CCL19 cooperatively induce significant resistance to TNF-␣-mediated apoptosis in B-ALL and B-CLL CD23؉CD5؉ B cells, but not in the cells from CB. B-ALL and B-CLL CD23؉CD5؉ B cells express elevated levels of paternally expressed gene 10 (PEG10). CXCL13 and CCL19 together significantly up-regulate PEG10 expression in the same cells. We have found that CXCL13 and CCL19 together by means of activation of CXCR5 and CCR7 up-regulate PEG10 expression and function, subsequently stabilize caspase-3 and caspase-8 in B-ALL and B-CLL CD23؉CD5؉ B cells, and further rescue the cells from TNF-␣-mediated apoptosis. Therefore, we suggest that normal , especially naive B and T cells, http://www.jimmunol.org/ use CXCL13/CXCR5 and CCL19/CCR7 for migration, homing, maturation, and cell homeostasis as well as secondary lymphoid tissues organogenesis. In addition, certain malignant cells take advantages of CXCL13/CXCR5 and CCL19/CCR7 for infiltration, resistance to apoptosis, and inappropriate proliferation. The Journal of Immunology, 2006, 177: 6713–6722.

lthough they decrease with age, CD5ϩ B cells are the and CD5ϩ B cells are indicated as a self-replenishing subpopula- major population in fetal life Most human IgMϩ cord tion, showing an increased propensity to malignant transformation A blood (CB)4 B cells express CD5; however, in the adult, (1). CD23ϩCD5ϩ B cells comprise the largest group of the ma- they represent 10–25% of B cells in blood and ϳ15–30% in tonsil, lignant cells in B cell chronic lymphocytic leukemia (B-CLL) and an important component in B cell lineage acute (B-ALL). by guest on September 28, 2021 CCL19 (EBV-induced gene-1 ligand chemokine (ELC)) and *Department of Immunology, School of Basic Medical Sciences, Anhui Medical CCL21 (secondary lymphoid tissue chemokine/6Ckine) are li- University, Hefei, China; †Department of Immunology, and Laboratory of and Clinical Immunology, Institute of Allergy and Immune-Related Diseases and gands for the chemokine receptor CCR7 (Burkitt re- Center for Medical Research, Wuhan University School of Medicine, Wuhan, China; ceptor-2), whereas CXCL13 (B cell-attracting chemokine 1 (BCA- ‡The State Key Laboratory of Molecular Biology, Institute of and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Science, 1)) is the only ligand for CXCR5 ( receptor-1) Shanghai, China; §Department of Hematology, The Renmin and Zongnan University (2). Homeostatic chemokines, such as CXCL13, CCL21, and ¶ Hospital, Wuhan University, Wuhan, China; and Department of Surgery, University CCL19, as well as their corresponding receptors, CXCR5 and of Michigan Medical Center, Ann Arbor, MI 48109 CCR7, have been shown to closely cooperate in the development Received for publication April 6, 2006. Accepted for publication August 22, 2006. of lymphoid organs and the maintenance of lymphoid tissue mi- The costs of publication of this article were defrayed in part by the payment of page croarchitecture (2). Expression of CXCR5 can be detected on ma- charges. This article must therefore be hereby marked advertisement in accordance ϩ with 18 U.S.C. Section 1734 solely to indicate this fact. ture recirculating B cells, small subsets of normal CD4 and ϩ 1 This work was supported by the National Key Basic Research Program of China CD8 T cells, and skin-derived migratory dendritic cells (3–6). from the Ministry of Science and Technology of People’s Republic of China (Nos. CXCR5 is essentially responsible for guiding B cells into the B 2001CB510004 and 2001CB510008), and by the National Natural Science Founda- cell zones of secondary lymphoid organs (7–9). However, the ex- tion of China (Nos. 39870674, 30572119, 30030130, and 30471509), Science Foun- dation of Anhui Province, China (No. 98436630), and Education and Research Foun- pression of CXCR5 on a subset of T cells strongly suggests an dation of Anhui Province, China (No. 98JL063) and Research Foundation from additional role for this receptor in migration (9, 10). CCR7 Health Department of Hubei Provincial Government, China (No. 301140344), and a special grant from the Personnel Department of Wuhan University, China. T.J. is a is highly expressed on naive T cells, but expressed at lower levels Chang Jiang Scholar supported by Chang Jiang Scholars Program from Ministry of on peripheral B cells. T cells and B cells show a transient increase Education, People’s Republic of China and Li Ka Shing Foundation, Hong Kong, in receptor expression following their activation (11), whereas T People’s Republic of China. cell differentiation toward effector cells is accompanied by a down- 2 H.C., H.Y., and W.L. contributed equally to this work. regulation of CCR7 on the cell surface (12–14). 3 Address correspondence and reprint requests to Dr. Tan Jinquan, Department of The synergy of CXCL13/CXCR5 and CCL19/CCR7 has been Immunology, School of Basic Medical Sciences, Anhui Medical University, 230032 Hefei, China or Department of Immunology, Wuhan University School of Medicine, shown in both physiological and pathological situations. CXCR5 Wuhan University, Dong Hu Road 115, 430071 Wuchang, Wuhan, People’s Republic cooperated with CCR7 control T cells and dendritic cell homing to of China. E-mail address: [email protected] 4 Abbreviations used in this paper: CB, cord blood; B-CLL, B cell chronic lympho- cytic leukemia; B-ALL, B cell lineage acute; ELC, EBV-induced gene-1 ligand che- apoptosis ; MCNC, migrating cells on negative control; sh, short hairpin; PKC, mokine; BCA, B cell-attracting chemokine; PEG10, paternally expressed gene 10; protein kinase C; CI, chemotactic index; SIAH1, human homolog of Drosophila HCC, human hepatocellular carcinoma; PI, propidium iodide; IAP, inhibitor of seven in absenia.

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00 6714 APOPTOTIC RESISTANCE BY CXCL13 AND CCL19 secondary lymphoid organs (12–14). The balanced expression of Real-time quantitative RT-PCR CCR7 and CXCR5 determines the positioning and proper function All real-time quantitative RT-PCR were performed as described elsewhere of follicular Th cells (11). CXCR5 and CCR7 double-deficient (28, 31). Briefly, the real-time quantitative PCR was performed in special mice lack lymphoid follicles due to an impaired migration of B optical tubes in a 96-well microtiter plate (Applied Biosystems) with an cells (15). Overexpression of CXCR5 and CCR7 on tumor lym- ABI PRISM 7700 Sequence Detector Systems (Applied Biosystems). By phocytes closely related to cells apoptosis as well as their migra- using the SYBR Green PCR Core Reagents , fluorescence signals were generated during each PCR cycle via the 5Ј to 3Ј endonuclease activity of tion and infiltration (16–19). The coexpression of CXCR5 and AmpliTaq Gold to provide real-time quantitative PCR information. The CCR7 were also found in B cell leukemia. However, the functional sequences of the specific primers are: CXCR5 sense, 5Ј-GGTCTTCATCT importance of CXCR5-CXCL13/BCA-1 and CCR7-CCL19/ELC TGCCCTTTG-3Ј; CXCR5 antisense, 5Ј-ATGCGTTTCTGCTTGGTTCT- receptor-ligand pairs in the pathophysiological events of malignant 3Ј; CCR7 sense, 5Ј-GCTCCAGGCACGCAACTTT-3Ј; CCR7 antisense, 5Ј-ACCACGACCACAGCGATGA-3Ј; PEG10 sense, 5Ј-ATGATGACAT B cells trafficking, homing, and survival is not fully understood. CGAGCTCCG-3Ј; PEG10 antisense, 5Ј-GCTGGGTAGTTGTGCATCA-3Ј. A novel paternally expressed imprinted gene, paternally ex- All unknown cDNAs were diluted to contain equal amounts of ␤-actin pressed gene 10 (PEG10), is identified as a paternally expressed cDNA. PCR retain conditions were 2 min at 50°C, 10 min at 95°C, 40 gene from a newly defined imprinted region at human chromo- cycles with 15 s at 95°C, 60 s at 60°C for amplifications. Potential PCR some 7q21 (20). PEG10 shows parent-of-origin-specific expres- product contamination was digested by uracil-N-glycosylase because dTTP is substituted by dUTP. sion in monochromosomal hybrids (21). PEG10 knockout mice show early embryonic lethality owing to defects in the placenta, Northern and Western blot assays indicating a critical role for mouse parthenogenetic development For mRNA detection (Northern blot), as previously described (27, 32), (21). An elevated level of expression has been found in the ma- each 5 ␮g of total RNA was electrophoresed under denaturing conditions, Downloaded from jority of the human hepatocellular carcinoma (HCC) cells (22, 23). followed by blotting onto Nytran membranes, and cross-linked by UV ␣ 32 Exogenous expression of PEG10 confers oncogenic activity and irradiation. CXCR5 and CCR7 cDNA probes, labeled by [ - P]dCTP, were obtained by PCR amplification of the sequence mentioned above transfection of hepatoma cells with PEG10 antisense suppressing its from total RNA from PBMC from normal adults (for CXCR5) or thymo- expression results in cancer cell growth inhibition (22). In addition, cytes from the specimen of thymusectomy (for CCR7), or human hepatoma PEG10 protein associates with human homolog of Drosophila seven cell line HepG2 (for PEG10). The membranes were hybridized overnight with 1 ϫ 106 cpm/ml 32P-labeled probe, followed by intensively washing in absenia (SIAH1), a mediator of apoptosis. Overexpression of http://www.jimmunol.org/ with 0.2ϫ SSC and 0.1% SDS before being autoradiographed. For protein PEG10 reduces the cell from death mediated by SIAH1 (22). Knock- detection (Western blot), the cells were lysed in lysis buffer. Cell lysis was down of PEG10 inhibits the proliferation of pancreatic carcinoma and performed for 30 min at 4°C with lysis buffer. Expression of inhibitor of HepG2 hepatocellular carcinoma cells (23). apoptosis protein (IAP) family (or other proteins indicated) was In this study, we have found that CXCR5 and CCR7 are selec- semiquantified after Western blot analysis (33). Lysates were centrifuged at tively expressed on B-ALL, B-CLL, and CB CD23ϩCD5ϩ B cells 10,000 rpm for 5 min at 4°C. Protein concentration was measured by Bio- Rad protein assay. Protein (around 40 ␮g) was loaded onto 16% SDS- at high frequency. CXCL13 and CCL19 cooperate selectively to ϩ ϩ PAGE, transferred onto polyvinylidene difluoride membranes after elec- induce resistance to apoptosis in B-ALL and B-CLL CD23 CD5 trophoresis, and incubated with the appropriate Abs at 0.5 ␮g/ml. Analyses B cells by means of activation of PEG10. were conducted using ECL detection (Amersham Biosciences). All Abs (Bcl-2, Bcl-x, c-FLIPL, c-IAP1, c-IAP2, X-linked mammalian, and survivin) by guest on September 28, 2021 were obtained from Santa Cruz Biotechnology, except anti-livin which was Materials and Methods obtained from Imgenex, anti-␤-actin which was obtained from Sigma-Aldrich, Patients and cell purification and CXCR5 and CCR7 mAbs which were obtained from R&D Systems. All patients with B-ALL and B-CLL fulfilled the French-American-British Chemotaxis assay (FAB) Cooperative Group criteria (24) and the guidelines of the National The chemotaxis assay was performed in a 48-well microchamber (Neuro Cancer Institute Working Group (25, 26). All patients gave informed con- ϩ Probe) technique (27, 29). Briefly, chemokines in RPMI 1640 with 0.5% sent according to institutional guidelines. CD19 (from normal periphery), ␮ ϩ ϩ ϩ BSA were placed in the lower wells (25 l). Twenty-five microliters of cell CD23 , or CD23 CD5 cells were purified from PBMCs from peripheral suspension (2 ϫ 106 cells/ml) was added to the upper well of the chamber, blood of normal subjects, CB of uncomplicated births (IgM undetectable), ␮ Plus which was separated from the lower well by a 5- m pore size, polycar- or patients with B-ALL or B-CLL using a FACStar sorting (27, 28). The bonate, polyvinylpyrolidone-free membrane (Nucleopore). The chamber viability of all cultured cells Ͼ95% was tested by trypan blue exclusion. ϩ ϩ was incubated for 60 min at 37°C and 5% CO2. The membrane was then The malignancy of purified B-ALL or B-CLL CD23 CD5 cells was con- carefully removed, fixed in 70% methanol, and stained for 5 min in 1% firmed by expression of CD20 and FMC-7. The cell line was Raji cell (B cell Coomassie brilliant blue. The migrating cells were counted using a Burkitt lymphoma cell line) obtained from the American Type Culture Col- microscopy. Approximately 6% of the cells will migrate spontaneously lection. The anti-CXCR5 and anti-CCR7 mAbs and chemokines (CXCL13, (known as migrating cells on negative control (MCNC)). The chemotactic CCL19, CCL25, and CXCL12) were purchased from R&D Systems. index (CI) ϭ migrating cell number at tested well of chemoattractant sam-

ple/MCNC. The results were expressed as CI with SD. Gene silencing assay For detection of CXCR5 and CCR7, the cells were triple stained PE-la- beled CD23, FITC-labeled CD5 (DakoCytomation) and PerCP-labeled Short hairpin (sh) RNAs were produced in vitro as described (34) using chemokine receptor Ab (R&D Systems), or matched isotype Ab (Dako- chemically synthesized DNA oligonucleotide templates (Sigma-Aldrich). Cytomation) at 5 ␮g/ml in PBS containing 2% BSA and 0.1% sodium Transcription templates were designed such that they contained U6 pro- Ј azide for 20 min, followed washing twice with staining buffer (28). The moter sequences at the 5 end. shRNA transcripts subjected to in vitro analyses were performed with flow cytometer (Coulter XL; Coulter). For Dicer processing were synthesized using a Riboprobe kit (Promega). detection of apoptosis, cells were stained in staining medium (RPMI 1640, dsDNA oligonucleotides encoding shRNAs with homology to the targeted 2% FBS, and 0.1% sodium azide) with 1 ␮g/ml propidium iodide (PI) for PEG10 gene were ligated into the EcoRV site to produce expression con- 30 min at 4°C, then stained with FITC-conjugated annexin V with binding structs. The PEG10 sense sequence inserted immediately downstream of buffer (BD Pharmingen) as previously described (29, 30). Coulter XL was the U6 promoters was as follows: GAGCTCTCTGAAGAGATCAACtt. used for analyses. For detection of intracellular active caspases, Cytofix/ Negative control was DNAPEG10 with the same sequence. Cells were Cytoperm buffer (BD Pharmingen) was used according to the manufactur- cultured in DMEM containing 10% heat-inactivated FBS, penicillin, and er’s instructions to permeabilize cells, and cells were subsequently stained streptomycin. Cells were harvested 2 days after the transfection. with anti-active-capsase-3 or anti-active-caspase-8 mAb (BD Pharmingen). Plasmids and cell transfection After washing, active caspase-3 or caspase-8 fluorescence intensity was measured by flow cytometry. Data were analyzed by means of the WinList Plasmids encoding PEG10 and CCR7 used in this study have been previ- program (The Scripps Research Institute). ously described (23, 35). The cells were transiently transfected with vectors The Journal of Immunology 6715 encoding target genes as described elsewhere (23, 35). Briefly, the cells CXCL13 and CCL19 induced very strong chemotaxis in CB were cultured with DMEM containing 10% FCS, penicillin, and strepto- CD23ϩCD5ϩ B cells (Fig. 2). mycin. Cells were grown to ϳ70% confluence in 60-mm dishes for 24 h before transfection. The DNA constructs of expression vectors (0.4 ␮g unless indicated) or vector only were mixed with 12 ␮l of LipofectAMINE CXCL13 and CCL19 together rescue B-ALL CD23ϩCD5ϩ (Invitrogen Life Technologies) in 2 ml of opti-DMEM serum-free medium B cells from apoptosis and added to cells, and incubated for 6 h. The cells were further cultured in 2.5 ml of DMEM containing 10% FCS in 5% CO2. We examined the protective effects of CXCL13 and CCL19 on different types of cells on TNF-␣-mediated apoptosis. Flow cy- Statistical analysis tometric analysis (Fig. 3) revealed that the number of apoptotic Statistical significance was assessed by the paired or unpaired Student t and necrotic cells was significantly decreased in cultured B- test. Values of p Ͻ 0.05 were considered statistically significant. ALL and B-CLL CD23ϩCD5ϩ B cells in presence of both CXCL13 and CCL19 (Fig. 3A, o and p), in comparison with Results those in the absence of CXCL13 and CCL19 (Fig. 3A, c and d). CXCR5 and CCR7 are selectively expressed on B-ALL ϩ ϩ Interestingly, CXCL13 or CCL19 alone had no such effect in CD23 CD5 B cells at high frequency B-ALL or B-CLL CD23ϩCD5ϩ B cells (Fig. 3A, g, h, k, and l). We screened some CXC and CC chemokine receptors on different CXCL13 and/or CCL19 did not render resistance to apoptosis in CD23ϩCD5ϩ and CD23ϩCD5Ϫ B cells (as well as normal pe- normal peripheral CD19ϩ B cells and CB CD23ϩCD5ϩ B cells ripheral CD19ϩCD5ϩ and CD19ϩCD5Ϫ cells). Flow cytometric (Fig. 3A, e, f, i, j, m, and n), in comparison with those in the

analysis revealed that the data of CXCR3, CXCR6, CCR4, and absence of CXCL13 and CCL19 (Fig. 3A, a and b). Neither Downloaded from CCR9 were either in agreement with previous reports or that there CCL13 nor CXCL19 had such an effect as to rescue different types were no differences among four types of cell sources in a total of of B cells (normal peripheral CD19ϩ B cells, and CB, B-ALL 41 cases of typical B-ALL and B-CLL patients (Table I). Inter- B-CLL CD23ϩCD5ϩ B cells) from TNF-␣-mediated apoptotic re- estingly, CXCR5 and CCR7 were selectively expressed on B-ALL sponse (data not shown). Abs against CXCR5 and CCR7 could and B-CLL CD23ϩCD5ϩ B cells at high frequency (86, 81, 42, completely block the combined protective effect of CXCL13 and and 51%, respectively) (Fig. 1, A and B), whereas, they were ex- CCL19, indicating that the rescuing effect was indeed induced by http://www.jimmunol.org/ pressed at significantly lower frequency (14 and 15%, 12 and 11%, means of interaction of CXCL13/CCL19 and CXCR5/CCR7 (data respectively) on B-ALL and B-CLL CD23ϩCD5Ϫ B cells. shown in Fig. 4). The total numbers of dead cells (including ap- CXCR5 and CCR7 were expressed at rather low level on CD19ϩ optotic and necrotic) in different types of the cells exhibited a B cells from normal peripheral blood. In comparison, they were similar pattern as shown in Fig. 3. Thus, CXCL13 and CCL19 expressed at similar levels on CD23ϩCD5ϩ and CD23ϩCD5Ϫ B cooperatively rescued B-ALL and B-CLL CD23ϩCD5ϩ B cells cells from CB. from TNF-␣-mediated apoptosis, but not either normal peripheral CXCR5 mRNA were detected in similar patterns as mentioned CD19ϩ B cells or CB CD23ϩCD5ϩ B cells, confirming antiapop- above in different types of CD23ϩCD5ϩ B cells (Fig. 1C, left), totic effects mediated by CXCR5 and CCR7 in B-ALL and B-CLL ϩ ϩ which was confirmed by Northern blot (Fig. 1C, right, upper pan- CD23 CD5 B cells. by guest on September 28, 2021 els), as well as in Western blot (Fig. 1C, right, lower panels). The Many signaling events of binding of certain chemokines to che- same results of CCR7 were shown in Fig. 1D. CXCR5 and CCR7 mokine receptors were included such as PI3K, MAPK, or protein expression patterns in CD23ϩCD5Ϫ B cells from distinct subjects kinase C (PKC), that appear to be involved in chemokine-mediated were also confirmed by Northern and Western blots (data not chemotaxis in certain cell types (36–39). To determine whether shown). these kinases were responsible for resistance to apoptosis induced Functionally, CXCL13/BCA-1 (CXCR5 ligand) and/or CCL19/ by costimulation of CXCL13 and CCL19, B-ALL and B-CLL ELC (CCR7 ligand) failed to induce significant chemotaxis in B- CD23ϩCD5ϩ B cells were pretreated with varying concentrations ALL and B-CLL CD23ϩCD5ϩ B cells (Fig. 2). Interestingly, both of pertussis toxin (1 ␮g/ml), an inhibitor of PKC; wortmannin (50

Table I. Some chemokine receptor expression on CD23ϩCD5ϩ, CD23ϩCD5Ϫ B cellsa

Cell Typeb CXCR3 CXCR5 CXCR6 CCR4 CCR7 CCR9

Periphery CD19ϩCD5ϩ Ndc Nd Nd Nd Nd Nd CD19ϩCD5Ϫ 5 Ϯ 1d 6 Ϯ 214Ϯ 811Ϯ 17Ϯ 4–e CB CD23ϩCD5ϩ 21 Ϯ 11 53 Ϯ 15 25 Ϯ 16 11 Ϯ 359Ϯ 17 21 Ϯ 13 CD23ϩCD5Ϫ 11 Ϯ 921Ϯ 12 15 Ϯ 6–23Ϯ 11 – B-ALL CD23ϩCD5ϩ 56 Ϯ 20 86 Ϯ 10 84 Ϯ 12 50 Ϯ 14 89 Ϯ 12 – CD23ϩCD5Ϫ 35 Ϯ 17 29 Ϯ 15 15 Ϯ 820Ϯ 11 27 Ϯ 12 – B-CLL CD23ϩCD5ϩ 51 Ϯ 15 47 Ϯ 14 24 Ϯ 13 21 Ϯ 12 46 Ϯ 19 3 Ϯ 1 CD23ϩCD5Ϫ 6 Ϯ 412Ϯ 45Ϯ 313Ϯ 811Ϯ 10 5 Ϯ 2

a The CD23ϩCD5ϩ, CD23ϩCD5Ϫ B cells were freshly isolated and stained in triple colors of CD23 (PE), CD5 (FITC), and indicated chemokine receptor Abs (PerCP) as described in Materials and Methods. b The CD23ϩ B cells were isolated from peripheral blood from normal subjects (CD19ϩ), cord blood (CB) of uncomplicated births, B-ALL, or B-CLL patients. c Nd, No determination since few CD23ϩCD5ϩ B cells in normal preipheral blood. d The listed numbers in the table were percentages of chemokine receptor-positive B cells. The data were mean values Ϯ SD at least of eight similar experiments conducted. For detection of CXCR5 or CCR7, showing data were mean values Ϯ SD of 25 B-ALL and 16 B-CLL patients. e Under detectable level. 6716 APOPTOTIC RESISTANCE BY CXCL13 AND CCL19 Downloaded from

FIGURE 2. The chemotaxis of CD23ϩCD5ϩ B cells. The chemotaxis of freshly isolated in CD23ϩCD5ϩ T cells from CB of uncomplicated births and B-ALL patients toward CXCL13/BCA-1 (A)(f), CCL19/ELC (B)(f), CCL3/MIP-1␣ (u) (100 ng/ml), or PBS control (Ⅺ). All results were expressed as CI or percentage of adhesive cells with SD (ϮSD), and based on triplicate determination of chemotaxis and adhesion on each con- http://www.jimmunol.org/ centration of chemokine indicated as nanograms per milliliter. The showing bars were mean values Ϯ SD of eight experiments conducted. Statistical significant differences as compared with controls are indi- .p Ͻ 0.001. Values of p Ͼ 0.05 are considered nonsignificant ,ء cated as ϭ CI at optimal concentration of vs CI at medium control. MCNCCB Ϯ ϭ Ϯ 13,437 1,785; MCNCB-ALL 15,239 2,021.

nM), a potent PI3K inhibitor; PD98059 (50 nM), a p44/42 MAPK inhibitor; or vehicle, DMSO, further costimulated with CXCL13 by guest on September 28, 2021 and CCL19. As seen in Fig. 4, only pertussis toxin significantly diminished the protective effect of CXCL13 and CCL19 costimu- lation in B-ALL CD23ϩCD5ϩ B cells from apoptosis. Neither PD98059 nor wortmannin inhibited a protective effect in B-ALL CD23ϩCD5ϩ B cells from apoptosis, suggesting that CXCL13 and CCL19 cooperated via the PKC pathway to protect B-ALL CD23ϩCD5ϩ B cells from apoptosis. As expected, Z-VAD-FMK (20 nM) rescued B-ALL CD23ϩCD5ϩ B cells from TNF-␣-me- diated apoptosis. We observed similar results in B-CLL CD23ϩCD5ϩ B cells (data not shown). As expected, mAbs against CXCR5 and CCR7 together significantly inhibited the protective effect of costimulation with CXCL13 and CCL19 in B-ALL CD23ϩCD5ϩ B cells from apoptosis, whereas isotypes had no

the same results of CCR7 (D). In C and D, the procedure for quantitative RT-PCR amplification was described in Materials and Methods. The show- ing bars were mean values Ϯ SD of eight similar experiments conducted. ,p Ͻ 0.001, normal peripheral B cells vs CD23ϩCD5ϩ B cells from CB ,ء FIGURE 1. CXCR5 and CCR7 selectively express on CD23ϩCD5ϩ B cells. Triple-color flow cytometric analysis of the distribution of CXCR5 B-ALL, or B-CLL. In C and D (upper panels), the detections of mRNA of and CCR7 on CD23ϩCD5ϩ and CD23ϩCD5Ϫ B cells from B-ALL (A) and CXCR5 and CCR7 were by Northern blot for freshly isolated B-CLL (B) patients. The CD23ϩ B cells were freshly isolated and stained CD23ϩCD5ϩ B cells from normal PBMC (only CD19ϩ), CB of uncom- in triple colors of CD23 (PE), CD5 (FITC), and CXCR5 or CCR7 (PerCP) plicated births, B-ALL, or B-CLL patients. Total RNA from different cells as described in Materials and Methods. The indicated numbers in the was treated as described in Materials and Methods. The hybridization sig- graphs were percentages of CD23ϩCD5ϩ and CD23ϩCD5Ϫ B cells. The nals for CXCR5 or CCR7 mRNA from different cells were shown in each indicated percentages in the graphs were numbers of CXCR5ϩ or CCR7ϩ upper picture. The 28S rRNAs in lower pictures confirmed that comparable B cells. The data were from a single experiment, which was representative amounts of total RNA were used. In C and D (lower panels), the CXCR5 of 23 (B-ALL) and 18 (B-CLL) similar experiments performed. Isotype Ab or CCR7 protein was examined using Western blot analyses. Actins in each controls were expressed as dished curves. CXCR5 expression on lower picture indicated the quantity of total cellular protein from the tested CD23ϩCD5ϩ B cells were examined by real-time quantitative RT-PCR (C; samples loaded in each lane. Arrows indicate markers used to verify equiv- left), Northern blot (C; upper panels), Western blot (C; lower panels), and alent molecular weights of appropriate proteins in each lane. The Journal of Immunology 6717 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 4. Analysis of the inhibition of TNF-␣-mediated apoptosis in CD23ϩCD5ϩ B cells. Flow cytometric analysis of apoptotic (A) and total dead cells (B), B-ALL CD23ϩCD5ϩ B cells were freshly isolated from B-ALL patients as described in Materials and Methods, then serum starved overnight and treated with anti-CXCR5 plus anti-CCR7 (each 5 ␮g/ml, Abs), isotype Abs (each 5 ␮g/ml, Isotypes), pertussis toxin (PT) (1 ␮g/ml), wortmannin (WT, 50 nM), PD98059 (50 nM), or vehicle DMSO for 1 h, followed culture in the presence of CXCL13 and CCL19/ELC (ࡗ, each 100 ng/ml) for 6 h before apoptotic assay. As a control, Z-VAD-FMK (20 nM), a caspase inhibitor, was used for blocking apoptosis. The cells were analyzed by flow cytometry as described in the legend for Fig. 3. The data were from a single experiment, which was representative of six experi- FIGURE 3. Analysis of apoptotic and total dead (necrotic and apo- ments performed. The data for total dead cells (PIϪ annexin Vϩ plus PIϩ ␣ ϩ ptotic) cells for inhibition of TNF- -mediated apoptosis of annexin V )(B) were mean values Ϯ SD of six experiments performed. CD23ϩCD5ϩ B cells. Flow cytometric analysis of apoptotic (A) and ϩ ϩ Statistically significant differences as compared with different treated cells Ͻ ء total dead cells (B), the CD23 CD5 B cells were freshly isolated from ϩ were indicated ( , p 0.001, Ab- or PT-treated vs PD98059-, WT- or normal PBMC (only CD19 ), CB of uncomplicated births, B-ALL or Z-VAD-FMK-treated CD23ϩCD5ϩ B cells from B-ALL). B-CLL patients and were pretreated at absence or presence of chemo- kine as indicated (all at 100 ng/ml) for 24 h at 37°C, following stim- ulation with TNF-␣ (100 ng/ml) for 24 h at 37°C. The cells were an- alyzed by flow cytometry for PI (y-axis) and FITC-conjugated annexin such effect (Fig. 4), documenting that the protective effect of V(x-axis) as described in Materials and Methods. The gating in the CXCL13 and CCL19 in the cells from apoptosis was indeed by forward scatter and side scatter histograms were adhered to the lym- means of CXCR5 and CCR7 pathways. As shown in Fig. 4, the phocyte region. The percentages of PIϪ annexin Vϩ cells and PIϩ an- ϩ data of total fractions of dead cells (including apoptotic and ne- nexin V cells were indicated in the figure. The data (A) were from a crotic) were similar to patterns in the results in Fig. 3A. single experiment, which was representative of six experiments per- Ϫ ϩ ϩ formed. The data for total dead cells (PI annexin V plus PI annexin PEG10 expression is selectively increased in B-ALL ϩ Ϯ V )(B) were mean values SD of six experiments performed. Sta- CD23ϩCD5ϩ B cells by CXCL13 and CCL19 costimulation tistically significant differences as compared with untreated cells were -p Ͻ 0.001, untreated vs CXCL13ϩCXCL19-treated Western blot showed that the protein levels of one group of anti ,ء) indicated ϩ ϩ CD23 CD5 B cells from B-ALL or B-CLL). apoptotic members (Bcl-2, Bcl-x, and c-FLIPL) in different types of freshly isolated CD23ϩCD5ϩ B cells (CB, B-CLL, and B-ALL) 6718 APOPTOTIC RESISTANCE BY CXCL13 AND CCL19

Table II. Analysis of total dead cells in B-CLL CD23ϩCD5ϩ B cells in different treatments

Treatmenta CXCL13 CCL19 CXCL13/CCL19b

Nontreated 65 Ϯ 14c 56 Ϯ 18 14 Ϯ 5 Vector (2 ␮g) 66 Ϯ 18 63 Ϯ 17 12 Ϯ 4 DNAPEG10 (2 ␮g) 62 Ϯ 14 61 Ϯ 18 15 Ϯ 5 shRNAPEG10 (0.02 ␮g) 59 Ϯ 20 68 Ϯ 10 21 Ϯ 11 shRNAPEG10 (2 ␮g) 63 Ϯ 867Ϯ 11 68 Ϯ 13d

a The purified CD23ϩCD5ϩ B cells from B-ALL patients were cultured for 6 days in the presence or absence of shRNAPEG10 at low concentration (0.02 ␮g) and high concentration (2 ␮g) as described in Materials and Methods. b They were then pretreated at presence of CXCL13 or/and CCL19 (all at 100 ng/ml) described in Materials and Methods, following stimulation in the presence of TNF-␣ (100 ng/ml) for 24 h at 37°C. c The cells were analyzed by flow cytometry for PI (y-axis) and FITC-conjugated annexin V (x-axis) as described in Materials and Methods. The data for total dead cells (PIϪ annexin Vϩ plus PIϩ annexin Vϩ) were mean values Ϯ SD of six exper- iments performed. d Value of p Ͻ 0.005 vs nontreated. Downloaded from were identical (data not shown). Interestingly, after stimulation with CXCL13 or/and CCL19 and further apoptotic induction with TNF-␣, their expression levels were still not significantly altered in different types of CD23ϩCD5ϩ B cells (data not shown). The ex- pression levels of another group of antiapoptotic proteins in the

IAP family (XIAP, c-IAP1, c-IAP2, survivin, and livin) in different http://www.jimmunol.org/ types of freshly isolated CD23ϩCD5ϩ B cells (CB, B-CLL and B-ALL) were identical as well (data not shown). After stimulation with CXCL13 or/and CCL19 and apoptotic induction with TNF-␣, their expression levels were still not significantly changed in dif- ferent types of B cell CD23ϩCD5ϩ B cells (data not shown). We further examined the expression levels of PEG10 in distinct CD23ϩCD5ϩ B cells during stimulation with CXCL13/CCL19 and TNF-␣. Data obtained from real-time quantitative RT-PCR and Northern blot analyses (Fig. 5) revealed that freshly isolated by guest on September 28, 2021 normal peripheral CD19ϩ B cells and CB CD23ϩCD5ϩ B cells expressed almost no PEG10 (Fig. 5A) or at very low levels (Fig. 5B). PEG10 expression levels in the cells were not significantly altered after 24 h culture with CXCL13 and/or CCL19. Costimu-

ϩ ϩ lation with TNF-␣ did not change PEG10 expression levels (data FIGURE 5. PEG10 mRNA expression in CD23 CD5 B cells. The not shown). In contrast, freshly isolated B-ALL and B-CLL real-time quantitative detection of RT-PCR (upper panels) and Northern ϩ ϩ blot (lower panels) for PEG10 mRNA in the CD23ϩCD5ϩ B cells. The CD23 CD5 B cells expressed elevated level of PEG10 (Fig. 5, cells were isolated from normal PBMC (only CD19ϩ)(A), CB of uncom- C and D). The PEG10 expression levels were significantly up- plicated births (B), B-ALL (C and E) or B-CLL (D) patients as described regulated in the cells cultured for 24 h with CXCL13 and CCL19 in Materials and Methods, which were pretreated at absence or presence of costimulation, CXCL13 or CCL19 alone did not show such an chemokine as indicated (all at 100 ng/ml) for 24 h at 37°C, following effect (Fig. 5, C and D). Costimulation with TNF-␣ did not change stimulation with TNF-␣ (100 ng/ml) for 24 h at 37°C. The procedure for PEG10 expression levels (data not shown). The time-course study quantitative RT-PCR amplification was described in Materials and Meth- (Fig. 5E) showed a significant up-regulation of PEG10 mRNA ods. A linear relationship between the cycle threshold (C ) and log starting T level by CXCL13 and CCL19 costimulation was already observed quantity of standard DNA template or target cDNA (PEG10) was detected within 8 h and peaked at 24 h. The data suggested that PEG10 (data not shown). The correlation coefficients are ϳ0.97–0.99. The show- ing bars were mean values Ϯ SD of six similar experiments conducted expression and functionality might contribute to might involved in -Ͻ ϩ ϩ ϩ the mechanism of resistance to TNF-␣-mediated apoptosis in B ء ( , p 0.001, untreated vs CXCL13 CXCL19-treated CD23 CD5 B ϩ ϩ cells from B-ALL or B-CLL). Total RNA from different cells as indicated ALL and B-CLL CD23 CD5 B cells as afforded by CXCL13 were isolated, electrophoresed and blotted as described in Materials and and CCL19 costimulation.

Methods. The hybridization signals for PEG10 mRNA in different cells ϩ ϩ were shown in upper parts of the panels. The 28S rRNAs in lower parts of PEG10 expression in B-ALL CD23 CD5 B cells is essential the panels confirmed the comparable amounts of loaded total RNA. The for resistance to apoptosis data were from a single experiment, which was representative of six ex- We applied shRNA of PEG10 (shRNAPEG10) to knockdown the periments performed. In E, the real-time quantitative detection of RT-PCR ϩ ϩ endogenous PEG10 expression to verify whether it contributes to for PEG10 mRNA in the CD23 CD5 B cells from B-ALL patients as ␣ described in Materials and Methods, which were pretreated at absence or CXCL13/CCL19 afforded resistance to TNF- -mediated apopto- presence of CXCL13 and CCL19 (each 100 ng/ml) for the time intervals as sis. Northern blot results showed that culture with shRNAPEG10 indicated at 37°C. The data were mean values Ϯ SD of six experiments at high concentration (2 ␮g/ml) completely abolished expression ϩ ϩ performed. Statistically significant differences as compared with untreated of PEG10 in B-ALL and B-CLL CD23 CD5 B cells at mRNA ,(p Ͻ 0.01, untreated vs CXCL13ϩCXCL19- levels, whereas, low concentration shRNAPEG10 (0.02 ␮g/ml ,ء) cells (0 h) were indicated treated B-ALL CD23ϩCD5ϩ B cells). DNAPEG10, and vector showed no such effects (data not shown). The Journal of Immunology 6719

The shRNAPEG10 at high concentration significantly blocked the effects of CXCL13 and CCL19 costimulation on induction of re- sistance to TNF-␣-mediated apoptosis in B-ALL CD23ϩCD5ϩ B cells, no low concentration shRNAPEG10, DNAPEG10 and vec- tor had such effect (Table II). All treatments did not alter the pat- terns of effects of CXCL13 or CCL19 alone in resistance to TNF- ␣-mediated apoptosis in B-ALL CD23ϩCD5ϩ B cells (Table II). Similar results were obtained in B-CLL CD23ϩCD5ϩ B cells (data not shown). Expression levels of activated caspase-3 (Fig. 6A) and caspase-8 (Fig. 6B) were measured by intracellular staining in B-ALL CD23ϩCD5ϩ B cells in cultures with the distinct treatments (40). The shRNAPEG10 at high concentration also significantly blocked the effects of CXCL13, and CCL19 costimulation promoted sta- bilization of caspase-3 and caspase-8 in B-ALL CD23ϩCD5ϩ B cells. In contrast, low concentration shRNAPEG10, DNAPEG10, and vector showed no such effect (data not shown). All treatments did not alter effects of CXCL13 or CCL19 alone on stabilization of ϩ ϩ caspase-3 and caspase-8 in B-ALL CD23 CD5 B cells (Fig. 6). Downloaded from As a control, we also measured caspase-3 and caspase-8 expres- sion in untreated B-ALL CD23ϩCD5ϩ B cells during TNF-␣- mediated apoptosis. The shRNAPEG10 treatment itself did not alter the patterns of caspase-3 and caspase-8 expression during TNF-␣-mediated apoptosis (Fig. 6). We observed similar results of

shRNAPEG10 to block stabilization of caspase-3 and caspase-8 http://www.jimmunol.org/ expression by CXCL13 and CCL19 costimulation during TNF-␣- mediated apoptosis in B-CLL CD23ϩCD5ϩ B cells (data not shown). The Raji cells were used for transfection studies. The cells were transfected with either PEG10 alone (Fig. 7A), or cotransfected with CCR7 and varying concentrations of PEG10 (Fig. 7, B–D). The transfection with PEG10 alone conferred neither significant protection against apoptosis nor stabilization of caspase-3 (Fig. 7A) and caspase-8 (data not shown) in Raji cells, indicating that PEG10 should be activated by two chemokines to fully function. by guest on September 28, 2021 Therefore, we cotransfected Raji cells with PEG10 and CCR7. The data showing in Fig. 7B verified that successful transfection of PEG10, CXCR5 (existed), and CCR7 took place by Western blot. Flow cytometric analysis (Fig. 7C) revealed that only the cells transfected with high doses of PEG10 (400 or 800 ng) could be rescued by CXCL13 and CCL19 costimulation of the cells from TNF-␣-induced apoptosis (Fig. 7C, d and e), whereas, neither low dose PEG10 (100 ng), vector, nor untransfected control showed any effect (Fig. 7C, a–c). The transfection itself did not cause apoptosis either (Fig. 7Cf ). Expression levels of activated caspase-3 were measured by intracellular staining in transfected cells. Only in the cells transfected with high doses of PEG10 (400 or 800 ng) could caspase-3 be stabilized by CXCL13 and CCL19 costimulation during TNF-␣-induced apoptosis (Fig. 7D, d and e), whereas, in the cells transfected with either low dose PEG10 (100 ng), vector, or untransfected cells, caspase-3 could not be stabi- lized (Fig. 7D, a–c). The transfection itself did not cause destabi- FIGURE 6. Activation of caspases in CD23ϩCD5ϩ B cells. Flow cyto- ϩ ϩ lization (Fig. 7Df ). We also obtained similar results of caspase-8 metric analysis of active caspase-3 (A) and caspase-8 (B) in CD23 CD5 B cells. The purified CD23ϩCD5ϩ B cells from B-ALL patients were cultured in the transfected cells (data not shown). The results strongly sug- ␣ for 6 days in the presence or absence of shRNAPEG10 at low concentration gested a critical role of PEG10 in rescuing cells from TNF- - (0.02 ␮g) and high concentration (2 ␮g) as described in Materials and Meth- mediated apoptosis by CXCL13 and CCL19 costimulation. ods. They were then pretreated at presence of CXCL13 or/and CCL19 (all at 100 ng/ml) described in Materials and Methods, following stimulation at pres- ence of TNF-␣ (100 ng/ml) for 24 h at 37°C. They were then permeabilized Discussion and fixed as indicated in Materials and Methods and subsequently stained for Chemokine receptors were reported to express on neoplastic cells of intracellular activated (cleaved) caspase-3 or caspase-8. Activated caspase-3- or caspase-8-specific fluorescence intensity was measured by flow cytometry. hemopoietic and nonhematopoietic origin, and overexpression of The indicating percentages of cells with activated caspase-3 or caspase-8 were some of these receptors was closely related to tumor progression and quantitated from relative-frequency histograms. Isotype Ab controls were ex- metastasis (41–43). As for B cell-derived lymphoproliferative disor- pressed as dished curves. The data were from a single experiment, which was ders, CXCR5 has been detected in neoplastic B cells from B-ALL representative of six experiments performed. (44) and B-CLL (45). CCR7 has been detected in B-CLL (46) and in 6720 APOPTOTIC RESISTANCE BY CXCL13 AND CCL19

tumor cells from classical Hodgkin’s disease with pre- dominance (47). In this study, we showed that CD23ϩCD5ϩ B cells comprised the largest group ϳ70% of the malignant cells in B-CLL, and 40% in B-ALL. CD5ϩ B cells have been reported to be the major population in fetal life, but decreased with age, and indicated as a self-replenishing subpopulation, showing an increased propensity to malignant transformation (1). In the present study, we compared four different types of cells (CD23ϩCD5ϩ and CD23ϩCD5Ϫ B cells from normal CB, B-ALL, and B-CLL patients) in terms of expression and functions of CXCR5 and CCR7. We have found that CXCR5 and CCR7 are selectively, frequently, and functionally expressed on B- ALL and B-CLL CD23ϩCD5ϩ B cells, whereas, their expression in B-ALL and B-CLL CD23ϩCD5Ϫ B cells was at a low level. Without impressive chemotactic responsiveness, CXCR5 and CCR7 costimu- lation (but not alone) displayed a novel function to induce resistance to TNF-␣-mediated apoptosis in B-ALL and B-CLL CD23ϩCD5ϩ B cells. Chemokine receptor signaling is reported to provide antiapop- totic activity to hemopoietic cells in a natural context (48). CCR9/ Downloaded from CCL25 interaction provides a cell survival signal to the receptor- expressing cells (29). CXCL1 and CXCL4 are able to support the survival of endothelial cells and , respectively (49, 50). However, there are some controversial, even contradictory, re- ports. For instance, CXCR4 induces programmed cell death of human peripheral CD4ϩ T cells, malignant T cells, and CD4/ http://www.jimmunol.org/ CXCR4 transfectants (51). The interaction between HIV R5 Env and CCR5 activates the Fas pathway and caspase-8 and triggers FasL production, ultimately causing CD4ϩ T cell death (52). We have also reported that CCR3 expression induced by IL-2 and IL-4 functions as a death receptor for B cells (30). The results in this study, together with other observations, suggest that normal B and T cells use CXCR5/CXCL13 and CCR7/CCL19 for migration,

homing, development, maturation, selection, and cell homeostasis by guest on September 28, 2021 as well as secondary lymphoid tissue organogenesis. Mean- while, some malignant cells, particularly B-ALL and B-CLL CD23ϩCD5ϩ B cells, take advantages of CXCR5/CXCL13 and CCR7/CCL19 for infiltration, resistance to apoptosis, and inappro- priate proliferation. To our knowledge, this study is the first report FIGURE 7. CXCL13 and CCL19 require PEG10 for protection against of differential functions of CXCR5/CXCL13 and CCR7/CCL19 in apoptosis. Raji cells were transfected with vectors encoding PEG10 (100 or distinct types of cells in terms of induction of apoptotic resistance, 800 ng alone) (A), or cotransfected with vectors encoding CCR7 (400 ng and is direct evidence of the pathophysiological activity of B-ALL each) in the absence or presence of increasing concentrations of PEG10 ϩ ϩ (100, 400, or 800 ng) (B–D). The amount of transfected cDNA was kept and B-CLL CD23 CD5 B cells induced by CXCL13 and CCL19 constant in each sample by adding control pcDNA3 vector. In A, PEG10- costimulation. transfected cells were treated with TNF-␣ (100 ng/ml) for 24 h at 37°C. PEG10 is identified on human 7q21 (21, 53). The cells were then analyzed by flow cytometry as described in Materials Mouse homolog PEG10 has recently been located in a large im- and Methods for apoptotic cells (right panels) and active caspase-3 (left printed gene cluster on mouse proximal chromosome 6 and has panels). In B, Western blotting were showing equal expression levels of been confirmed to be imprinted (54). Because the protein products CXCR5 and CCR7 as well as increasing expression levels of PEG10. In C from the predicted open reading frames 1 and 2 of PEG10 show D and , cotransfected cells were pretreated at absence or presence of homology to the gag and pol proteins of vertebrate retrotransposon CXCL13 and CCL19 (all 100 ng/ml) described in Materials and Methods, some of cells were followed stimulation with TNF-␣ (100 ng/ml) for Ty3/Gypsy, PEG10 is speculated to be a retrotransposon-derived 24 h at 37°C before other assay as indicated (ࡗ, no stimulation with gene. Distinct expression of PEG10 is found in the brain, kidney, TNF-␣). C, analysis of total dead (necrotic and apoptotic) cells. The lung, testis, and placenta but not in the liver and a number of other cells were analyzed by flow cytometry as described in the legend for tissues (20). In contrast to this, expression of PEG10 is only de- Fig. 3. The data were from a single experiment, which was represen- tected in the placenta among the 14 adult mouse tissues examined tative of six experiments performed. D, Flow cytometric analysis of (54). Some experimental data suggest a role for preferential ex- active caspase-3 in transfected cells as indicated. Cells were pretreated pression of the imprinted genes in regulating growth control of as mentioned above. They were then permeabilized and fixed as indi- liver and pancreatic carcinoma cells (23). Exogenous expression of cated in Materials and Methods and subsequently stained for intracel- PEG10 promotes growth of certain HCC cell lines that do not lular activated (cleaved) caspase-3. The indicating percentages of cells with activated caspase-3 were quantitated from relative-frequency his- manifest endogenous expression of this gene. The interaction of tograms. Isotype Ab controls were expressed as dished curves. The data PEG10 protein with SIAH proteins plays important roles in resis- were from a single experiment, which was representative of six exper- tance to apoptosis (22). Even PEG10 is suggested to serve as a iments performed. novel molecular target for treatment of HCCs (22). In the present study, we have found that freshly isolated B-ALL and B-CLL The Journal of Immunology 6721

ϩ ϩ CD23 CD5 B cells express an elevated level of PEG10, com- 16. Luther, S. A., T. Lopez, W. Bai, D. Hanahan, and J. G. Cyster. 2000. BLC pared with that in normal peripheral CD19ϩ B cells and CB expression in pancreatic islets causes B cell recruitment and lymphotoxin-depen- ϩ ϩ dent lymphoid neogenesis. Immunity 12: 471–481. CD23 CD5 B cells. After 24 h of culture with CXCL13 and 17. Finke, D., H. Acha-Orbea, A. Mattis, M. Lipp, and J. Kraehenbuhl. 2002. ϩ Ϫ ␣ ␤ CCL19 costimulation, PEG10 expression levels in the cells had CD4 CD3 cells induce Peyer’s patch development: role of 4 1 integrin ac- been significantly up-regulated (Fig. 5). By using shRNAs, we tivation by CXCR5. Immunity 17: 363–373. 18. Smith, J. R., R. M. Braziel, S. Paoletti, M. Lipp, M. Uguccioni, and have found that abrogation of PEG10 expression significantly J. T. Rosenbaum. 2003. Expression of B-cell-attracting chemokine 1 (CXCL13) blocks the protective effects of CXCL13 and CCL19 costimu- by malignant lymphocytes and vascular endothelium in primary central nervous lation on TNF-␣-mediated apoptosis in B-ALL and B-CLL system lymphoma. Blood 101: 815–821. ϩ ϩ 19. Mazzucchelli, L., A. Blaser, A. Kappeler, P. Scharli, J. A. Laissue, M. Baggiolini, CD23 CD5 B cells (Table II). We suggest that CXCL13 and and M. Uguccioni. 1999. BCA-1 is highly expressed in Helicobacter pylori- CCL19 cooperate via frequently expressed and activated CXCR5 induced mucosa-associated lymphoid tissue and gastric lymphoma. J. Clin. In- and CCR7, which up-regulates PEG10 expression and function, vest. 104: R49–R54. and subsequently stabilizes caspase-3 and caspase-8 in B-ALL and 20. Ono, R., S. Kobayashi, H. Wagatsuma, K. Aisaka, T. Kohda, T. Kaneko-Ishino, ϩ ϩ and F. Ishino. 2001. A retrotransposon-derived gene, PEG10, is a novel imprinted B-CLL CD23 CD5 B cells as an important mechanism to rescue gene located on human chromosome 7q21. Genomics 73: 232–237. the cells from TNF-␣-mediated apoptosis (Figs. 6 and 7). This is 21. Ono, R., K. Nakamura, K. Inoue, M. Naruse, T. Usami, N. Wakisaka-Saito, the first report of this imprinted gene expressed in both human T. Hino, R. Suzuki-Migishima, N. Ogonuki, H. Miki, et al. 2006. Deletion of ϩ ϩ Peg10, an imprinted gene acquired from a retrotransposon, causes early embry- B-ALL and B-CLL CD23 CD5 B cells. We have also proven a onic lethality. Nat. Genet. 38: 101–106. mechanism by which overexpression of CXCR5 and CCR7 on 22. Okabe, H., S. Satoh, Y. Furukawa, T. Kato, S. Hasegawa, Y. Nakajima, Y. B-ALL and B-CLL CD23ϩCD5ϩ B cells renders direct resistance Yamaoka, and Y. Nakamura. 2003. Involvement of PEG10 in human hepa- tocellular through interaction with SIAH1. Cancer Res. 63: to apoptosis in these malignant cells. Understanding the molecular 3043–3048. Downloaded from basis of abnormal imprinting of PEG10 in both human B-ALL and 23. Li, C. M., A. A. Margolin, M. Salas, L. Memeo, M. Mansukhani, H. Hibshoosh, B-CLL will shed new light on the process and mechanism that M. Szabolcs, A. Klinakis, and B. Tycko. 2006. PEG10 is a c- target gene in cancer cells. Cancer Res. 66: 665–672. leads to malignant lymphoproliferative disorders. 24. Bennett, J. M., D. Catovsky, M. T. Daniel, G. Flandrin, D. A. Galton, H. R. Gralnick, and C. Sultan. 1981. The morphological classification of acute lymphoblastic leukaemia: concordance among observers and clinical correla- Disclosures tions. Br. J. Haematol. 47: 553–561. The authors have no financial conflict of interest. 25. Cheson, B. D., J. M. Bennett, M. Grever, N. Kay, M. J. Keating, S. O’Brien, and http://www.jimmunol.org/ K. R. Rai. 1996. National Cancer Institute-sponsored Working Group guidelines for chronic lymphocytic leukemia: revised guidelines for diagnosis and treatment. References Blood 87: 4990–4997. 1. Dono, M., V. L. Burgio, C. Tacchetti, A. Favre, A. Augliera, S. Zupo, 26. Bennett, J. M., D. Catovsky, M. T. Daniel, G. Flandrin, D. A. Galton, G. Taborelli, N. Chiorazzi, C. E. Grossi, and M. Ferrarini. 1996. Subepithelial B H. R. Gralnick, and C. Sultan. 1989. Proposals for the classification of chronic cells in the human palatine tonsil. I. Morphologic, cytochemical and phenotypic (mature) B and T lymphoid leukaemias. French-American-British (FAB) Coop- characterization. Eur. J. Immunol. 26: 2035–2042. erative Group. J. Clin. Pathol. 42: 567–584. 2. Muller, G., U. E. Hopken, and M. Lipp. 2003. The impact of CCR7 and CXCR5 27. Qiuping, Z., L. Qun, H. Chunsong, Z. Xiaolian, H. Baojun, Y. Mingzhen, on lymphoid organ development and systemic immunity. Immunol. Rev. 195: L. Chengming, H. Jinshen, G. Qingping, Z. Kejian, et al. 2003. Selectively in- 117–135. creased expression and functions of chemokine receptor CCR9 on CD4ϩ T cells 3. Wu, M. T., and S. T. Hwang. 2002. CXCR5-transduced bone marrow-derived

from patients with T-cell lineage acute lymphocytic leukemia. Cancer Res. 63: by guest on September 28, 2021 dendritic cells traffic to B cell zones of lymph nodes and modify antigen-specific 6469–6477. immune responses. J. Immunol. 168: 5096–5102. 28. Jinquan, T., S. Quan, H. H. Jacobi, C. Jing, A. Millner, B. Jensen, H. O. Madsen, 4. Forster, R., T. Emrich, E. Kremmer, and M. Lipp. 1994. Expression of the G- L. P. Ryder, A. Svejgaard, H. J. Malling, et al. 2000. CXC chemokine receptor protein-coupled receptor BLR1 defines mature, recirculating B cells and a subset ϩ 3 expression on CD34 hematopoietic progenitors from human cord blood in- of T-helper memory cells. Blood 84: 830–840. duced by granulocyte- colony-stimulating factor: chemotaxis and ad- 5. Saeki, H., M. T. Wu, E. Olasz, and S. T. Hwang. 2000. A migratory population hesion induced by its ligands, interferon ␥-inducible protein 10 and monokine of skin-derived dendritic cells expresses CXCR5, responds to B lymphocyte che- induced by interferon ␥. Blood 96: 1230–1238. moattractant in vitro, and co-localizes to B cell zones in lymph nodes in vivo. Eur. J. Immunol. 30: 2808–2814. 29. Jinquan, T., H. H. Jacobi, C. Jing, A. Millner, E. Sten, L. Hviid, L. Anting, 6. Yu, P., Y. Wang, R. K. Chin, L. Martinez-Pomares, S. Gordon, M. H. Kosco-Vibois, L. P. Ryder, C. Glue, P. S. Skov, et al. 2003. CCR3 expression induced by IL-2 J. Cyster, and Y. X. Fu. 2002. B cells control the migration of a subset of den- and IL-4 functioning as a death receptor for B cells. J. Immunol. 171: 1722–1731. dritic cells into B cell follicles via CXC chemokine ligand 13 in a lymphotoxin- 30. Kruse, N., M. Pette, K. Toyka, and P. Rieckmann. 1997. Quantification of cy- dependent fashion. J. Immunol. 168: 5117–5123. tokine mRNA expression by RT PCR in samples of previously frozen blood. 7. Ansel, K. M., V. N. Ngo, P. L. Hyman, S. A. Luther, R. Forster, J. D. Sedgwick, J. Immunol. Methods 210: 195–203. J. L. Browning, M. Lipp, and J. G. Cyster. 2000. A chemokine-driven positive 31. Sica, A., A. Saccani, A. Borsatti, C. A. Power, T. N. Wells, W. Luini, feedback loop organizes lymphoid follicles. Nature 406: 309–314. N. Polentarutti, S. Sozzani, and A. Mantovani. 1997. Bacterial lipopolysaccharide 8. Gunn, M. D., V. N. Ngo, K. M. Ansel, E. H. Ekland, J. G. Cyster, and rapidly inhibits expression of C-C chemokine receptors in human monocytes. L. T. Williams. 1998. A B-cell-homing chemokine made in lymphoid follicles J. Exp. Med. 185: 969–974. activates Burkitt’s lymphoma receptor-1. Nature 391: 799–803. 32. Massari, P., Y. Ho, and L. M. Wetzler. 2000. Neisseria meningitidis porin PorB 9. Forster, R., A. E. Mattis, E. Kremmer, E. Wolf, G. Brem, and M. Lipp. 1996. A interacts with mitochondria and protects cells from apoptosis. Proc. Natl. Acad. putative chemokine receptor, BLR1, directs B cell migration to defined lymphoid Sci. USA 97: 9070–9075. organs and specific anatomic compartments of the spleen. Cell 87: 1037–1047. 33. Youn, B. S., Y. J. Kim, C. Mantel, K. Y. Yu, and H. E. Broxmeyer. 2001. 10. Okada, T., V. N. Ngo, E. H. Ekland, R. Forster, M. Lipp, D. R. Littman, and Blocking of c-FLIPL-independent cycloheximide-induced apoptosis or Fas-me- J. G. Cyster. 2002. Chemokine requirements for B cell entry to lymph nodes and diated apoptosis by the CC chemokine receptor 9/TECK interaction. Blood 98: Peyer’s patches. J. Exp. Med. 196: 65–75. 925–933. 11. Reif, K., E. H. Ekland, L. Ohl, H. Nakano, M. Lipp, R. Forster, and J. G. Cyster. 34. Kawasaki, H., and K. Taira. 2003. Short hairpin type of dsRNAs that are con- 2002. Balanced responsiveness to chemoattractants from adjacent zones deter- trolled by tRNA(Val) promoter significantly induce RNAi-mediated gene silenc- mines B-cell position. Nature 416: 94–99. ing in the cytoplasm of human cells. Nucleic Acids Res. 31: 700–707. 12. Sallusto, F., D. Lenig, R. Forster, M. Lipp, and A. Lanzavecchia. 1999. Two 35. Kanbe, K., N. Shimizu, Y. Soda, K. Takagishi, and H. Hoshino. 1999. A CXC subsets of memory T lymphocytes with distinct homing potentials and effector chemokine receptor, CXCR5/BLR1, is a novel and specific coreceptor for human functions. Nature 401: 708–712. immunodeficiency virus type 2. Virology 265: 264–273. 13. Yanagihara, S., E. Komura, J. Nagafune, H. Watarai, and Y. Yamaguchi. 1998. EBI1/CCR7 is a new member of dendritic cell chemokine receptor that is up- 36. Youn, B. S., C. Mantel, and H. E. Broxmeyer. 2000. Chemokines, chemokine regulated upon maturation. J. Immunol. 161: 3096–3102. receptors and hematopoiesis. Immunol. Rev. 177: 150–174. 14. Gunn, M. D., K. Tangemann, C. Tam, J. G. Cyster, S. D. Rosen, and 37. Kampen, G. T., S. Stafford, T. Adachi, T. Jinquan, S. Quan, J. A. Grant, L. T. Williams. 1998. A chemokine expressed in lymphoid high endothelial P. S. Skov, L. K. Poulsen, and R. Alam. 2000. Eotaxin induces degranulation and venules promotes the adhesion and chemotaxis of naive T lymphocytes. Proc. chemotaxis of through the activation of ERK2 and p38 mitogen- Natl. Acad. Sci. USA 95: 258–263. activated protein kinases. Blood 95: 1911–1917. 15. Ohl, L., G. Henning, S. Krautwald, M. Lipp, S. Hardtke, G. Bernhardt, O. Pabst, 38. Boehme, S. A., S. K. Sullivan, P. D. Crowe, M. Santos, P. J. Conlon, and R. Forster. 2003. Cooperating mechanisms of CXCR5 and CCR7 in de- P. Sriramarao, and K. B. Bacon. 1999. Activation of mitogen-activated pro- velopment and organization of secondary lymphoid organs. J. Exp. Med. 197: tein kinase regulates eotaxin-induced migration. J. Immunol. 163: 1199–1204. 1611–1618. 6722 APOPTOTIC RESISTANCE BY CXCL13 AND CCL19

39. Wang, J. F., I. W. Park, and J. E. Groopman. 2000. Stromal cell-derived factor-1␣ lymphocyte-predominant Hodgkin disease correlates with distinct dissemination stimulates of multiple focal adhesion proteins and in- of neoplastic cells in lymphoid organs. Blood 99: 1109–1116. duces migration of hematopoietic progenitor cells: roles of phosphoinositide-3 48. Youn, B. S., K. Y. Yu, J. Oh, J. Lee, T. H. Lee, and H. E. Broxmeyer. 2002. Role kinase and protein kinase C. Blood 95: 2505–2513. of the CC chemokine receptor 9/TECK interaction in apoptosis. Apoptosis 7: 40. Thornberry, N. A., and Y. Lazebnik. 1998. Caspases: enemies within. Science 271–276. 281: 1312–1316. 49. Han, Z. C., M. Lu, J. Li, M. Defard, B. Boval, N. Schlegel, and J. P. Caen. 1997. 41. Muller, A., B. Homey, H. Soto, N. Ge, D. Catron, M. E. Buchanan, factor 4 and other CXC chemokines support the survival of normal he- T. McClanahan, E. Murphy, W. Yuan, S. N. Wagner, et al. 2001. Involvement of matopoietic cells and reduce the chemosensitivity of cells to cytotoxic agents. chemokine receptors in breast cancer metastasis. Nature 410: 50–56. Blood 89: 2328–2335. 42. Geminder, H., O. Sagi-Assif, L. Goldberg, T. Meshel, G. Rechavi, I. P. Witz, and 50. Scheuerer, B., M. Ernst, I. Durrbaum-Landmann, J. Fleischer, E. Grage- Griebenow, A. Ben-Baruch. 2001. A possible role for CXCR4 and its ligand, the CXC che- E. Brandt, H. D. Flad, and F. Petersen. 2000. The CXC-chemokine platelet factor 4 mokine stromal cell-derived factor-1, in the development of bone marrow me- promotes survival and induces monocyte differentiation into . tastases in neuroblastoma. J. Immunol. 167: 4747–4757. Blood 95: 1158–1166. 51. Berndt, C., B. Mopps, S. Angermuller, P. Gierschik, and P. H. Krammer. 1998. 43. Scotton, C. J., J. L. Wilson, D. Milliken, G. Stamp, and F. R. Balkwill. 2001. ϩ CXCR4 and CD4 mediate a rapid CD95-independent cell death in CD4 T cells. Epithelial cancer cell migration: a role for chemokine receptors? Cancer Res. 61: Proc. Natl. Acad. Sci. USA 95: 12556–12561. 4961–4965. 52. Algeciras-Schimnich, A., S. R. Vlahakis, A. Villasis-Keever, T. Gomez, 44. Durig, J., U. Schmucker, and U. Duhrsen. 2001. Differential expression of che- C. J. Heppelmann, G. Bou, and C. V. Paya. 2002. CCR5 mediates Fas- and mokine receptors in B cell malignancies. Leukemia 15: 752–756. caspase-8 dependent apoptosis of both uninfected and HIV infected primary hu- 45. Burger, J. A., M. Burger, and T. J. Kipps. 1999. Chronic lymphocytic leukemia man CD4 T cells. AIDS 16: 1467–1478. B cells express functional CXCR4 chemokine receptors that mediate spontaneous 53. Nakabayashi, K., L. Bentley, M. P. Hitchins, K. Mitsuya, M. Meguro, migration beneath bone marrow stromal cells. Blood 94: 3658–3667. S. Minagawa, J. S. Bamforth, P. Stanier, M. Preece, R. Weksberg, et al. 2002. 46. Till, K. J., K. Lin, M. Zuzel, and J. C. Cawley. 2002. The chemokine receptor Identification and characterization of an imprinted antisense RNA (MESTIT1) in ␣ CCR7 and 4 integrin are important for migration of chronic lymphocytic leu- the human MEST locus on chromosome 7q32. Hum. Mol. Genet. 11: 1743–1756. kemia cells into lymph nodes. Blood 99: 2977–2984. 54. Ono, R., H. Shiura, H. Aburatani, T. Kohda, T. Kaneko-Ishino, and F. Ishino. 47. Hopken, U. E., H. D. Foss, D. Meyer, M. Hinz, K. Leder, H. Stein, and M. Lipp. 2003. Identification of a large novel imprinted gene cluster on mouse proximal Downloaded from 2002. Up-regulation of the chemokine receptor CCR7 in classical but not in chromosome 6. Genome Res. 13: 1696–1705. http://www.jimmunol.org/ by guest on September 28, 2021