Differential Expression of Chemokine Receptors in B Cell Malignancies

JDu¨rig, U Schmu¨cker and U Du¨hrsen

Department of Haematology, University Hospital Essen, D-45122, Essen, Germany

Chemokines are a family of 8–10 kDa proteins with a wide range haemopoietic cells.8 The non-Hodgkin’s lymphomas (NHLs) of biological activities including the regulation of leukocyte represent the malignant counterparts of normal lymphocytes, trafficking, modulation of haemopoietic cell proliferation and adhesion to extracellular matrix molecules. Using a panel of arrested at specific stages of maturation. Using a panel of chemokine receptor-specific monoclonal antibodies (MoAb) in chemokine receptor-specific antibodies directed against a multicolour flow cytometry approach we analysed the CXCR4, CXCR5, CCR5 and CCR6 (Table 1) in a multicolour expression of the lymphocyte-associated chemokine receptors flow cytometry approach, we investigated the expression CXCR4, CXCR5, CCR5 and CCR6 in B cell acute lymphoblastic pattern of these receptors in B cell malignancies of varying leukaemia (precursor B-ALL; six cases), B cell chronic lympho- maturational stages. cytic leukaemia (B-CLL; 31 cases), multiple myeloma (10 cases), mantle cell lymphoma (MCL, four cases), follicular lymphoma (FL, three cases) and hairy cell leukaemia (HCL, five cases). We demonstrate that CXCR4, CXCR5 and CCR6 are dif- Materials and methods ferentially expressed in these B lymphoproliferative disorders depending on the maturational stage of the malignant B cell Patients and cells population investigated. In particular, we found that CXCR4 is strongly expressed on immature ALL blasts whereas no surface immunoreactivity for CXCR5, CCR5 and CCR6 was Normal bone marrow (NBM) and peripheral blood (PB) was observed. By contrast, non-Hodgkin’s lymphomas (NHLs) cor- obtained from normal blood donors and donations for allo- responding to more mature peripheral B cell subsets (ie B-CLL geneic transplantation and leftovers of routine diagnostic mar- and MCL) exhibited high expression levels of CXCR4 and row aspirations from haematologically normal patients. Diag- CXCR5. Analysis of terminally differentiated myeloma cells nosis was based on the Revised European American revealed a down-regulation of CXCR4, CXCR5 and CCR6. Lymphoma Classification.9 All samples were taken with CCR5, which is not expressed in normal B cells, was also absent from the majority of NHLs. However, CCR5 staining was appropriate informed consent from the patients according to seen in three of five cases of HCL, representing the first institutional guidelines. example of cross-lineage aberrant chemokine receptor expression in malignant haemopoietic cells. Leukemia (2001) 15, 752–756. Cell surface staining and flow cytometry Keywords: non-Hodgkin’s lymphoma; chemokine receptors; hairy cell leukaemia; CXCR4; CXCR5; CCR5; CCR6 Fresh heparinised PB and BM samples were prepared for flow cytometry by ammonium chloride erythrocyte lysis (Orthomune Lysing Reagent; Ortho Diagnostic Systems, Rari- Introduction tan, NJ, USA). As a first step in all cases the immunophenotype was characterized using antibodies directed against CD19, An efficient immune response is critically dependent on the CD10, CD79a, IgM, CD34, CD22, κ and λ light chains (all presence of lymphocyte clones with the right antigen speci- from DAKO, Glostrup, Denmark), CD38, CD56, CD4, CD8, ficity at the site of inflammation. Many of the cues needed CD3, CD11c (all from Becton Dickinson, Heidelberg, to direct lymphocytes into the right position are provided by Germany), CD103, CD138 (from DPC Biermann, Bad members of the chemokine family that are produced and Nauheim, Germany) CD5 (Immunotech, Marseille, France) 1 secreted in inflamed tissues. In addition to their role in employing a standard three-colour flow cytometry inflammation, chemokines have recently been shown to play approach.10 Chemokine receptor expression was subsequently an important role in the homing of circulating lymphocytes analysed in well-defined B cell populations, usually contain- to secondary lymphoid organs, eg lymph nodes and Peyer’s ing more than 95% malignant cells. For chemokine receptor patches, under physiological conditions.2 The specific effects of chemokines on their target cells are mediated by members of a family of 7-transmembrane-spanning G-protein-coupled Table 1 Details of the chemokine receptors investigated in this receptors, 18 of which have been cloned and characterized study in the past decade.1 It is now well established that unique patterns of chemokine receptor expression may help define Receptor Ligand (high affinity) Cellular distribution lymphocyte populations that differ in differentiation, acti- α β vation stages and biological properties.3,4 CXCR4 SDF-1 , SDF-1 Myeloid, T, B, Ep, En, DC Chemokine receptors previously implicated in B cell CXCR5 BCA-1 B migration and proliferation include CXCR4, CXCR5 and CCR5 RANTES, MIP-1α,T,M,DC CCR6.5–7 There is accumulating evidence that apart from their MIP-1β, MCP-2 physiologic functions, chemokine receptors may also be CCR6 MIP-3α T, B, DC involved in the migration and dissemination of malignant SDF, stroma-derived factor; BCA, B cell-attracting chemokine; RANTES, regulated upon activation, normal T cell expressed and secreted; MIP, macrophage inflammatory protein; MCP, monocyte Correspondence: J Du¨rig; Fax: 0049 201 723 2304 chemoattractant protein; T, T cell; B, B cell; En, endothelial cell; Ep, Received 5 September 2000; accepted 20 January 2001 epithelial cell; DC, dendritic cell; M, monocyte. Chemokine receptor expression in B cell malignancies JDu¨rig et al 753 analysis cells were incubated with antibodies directed against (Figure 1c, PB vs NBM P , 0.05), representing developmen- CCR5 (FITC-conjugated mouse IgG clone 2D7/CCR5), CXCR4 tally mature and more primitive B cell subsets, respectively.13 (PE-conjugated mouse IgG clone clone 12G5, both from CCR6 immunostaining was observed in the majority of both Pharmingen, Heidelberg, Germany), CXCR5 (PE-conjugated PB and NBM B cells with significantly higher expression levels mouse IgG clone 51505.111) and CCR6 (PE-conjugated in PB B cells. No CCR5 surface immunoreactivity was mouse IgG clone 53103.111, both from R&D Systems, Wies- observed in either PB or NBM-derived B cells. However, mon- baden, Germany). Negative isotype-matched controls (Becton ocytes and a subset of T cells, both of which have previously Dickinson) were used to define the threshold line separating been reported to express CCR5,1 stained brightly positive, chemokine receptor positive and negative cells such that less thereby serving as a positive internal control in each experi- than 1% of isotype positive cells were present to the right of ment (Figure 1c). the line. Results are expressed as the percentage of chemokine We have recently demonstrated that there is a significant receptor positive cells in the respective populations. A mem- up-regulation of CXCR4 as bone marrow progenitor cells brane marker was considered positive when more than 20% develop from CD34+ to more mature B lymphocytes.14 While of the gated population expressed it.11 In all experiments a 57.7 ± 3.5% of NBM CD34+ cells stained positive for CXCR4 minimum of 10 000 cells was analysed. The flow cytometer (n = 2314), no CXCR5, CCR6 and CCR5 expression (5.3 ± 0.8, (FACScan, Becton Dickinson) was calibrated with CAliBRITE- 8.5 ± 1.4 and 6.0 ± 3.1%, respectively, n = 3, data not shown) 3 beads (Becton Dickinson) and FACSComp Software (Becton was detected in these cells using flow cytometry. Dickinson). Data acquisition and analysis were performed using CellQuest software (Becton Dickinson). Chemokine receptor immunostaining in B-precursor ALL Statistical analysis CXCR4 was the only chemokine receptor detected in six cases Data are expressed as mean ± s.e.m. Student’s t-test was used of B-precursor ALL (Table 2). CXCR4 expression levels on to compare results between the indicated experimental CD34+ ALL blasts were significantly higher than those groups. reported for unfractionated NBM CD34+ cells (P , 0.05) and comparable to those of normal lymphoid progenitors (CD34/CD10 positive bone marrow cells, P . 0.0514). Inter- Results estingly, we have recently shown similar amounts of CXCR4 in bcr/abl positive CD34+/CD19+/CD10+ cells isolated from Using a flow cytometry assay we analysed the expression of CML patients with B lymphoid blastic crisis,14 suggesting that CXCR4, CXCR5, CCR5 and CCR6 in normal as compared to CXCR4 up-regulation may be a common phenomenon in malignant B cell populations from patients with precursor B- normal and malignant early B lymphopoiesis. ALL, CLL, multiple myeloma, mantle cell lymphoma, follicular lymphoma and hairy cell leukaemia (Table 2). Chemokine receptor immunostaining in B-CLL

Chemokine receptor immunostaining in normal PB All B-CLL cases exhibited a CXCR4 and CXCR5 positive and NBM-derived B cells phenotype with signiﬁcantly stronger expression of CXCR4 in comparison to normal B cells (P , 0.05). In contrast to the In line with previous reports12 we found that CXCR4 is homogenous expression pattern observed for CXCR4 and expressed at comparable levels in PB and NBM B cells CXCR5, CCR6 surface immunoreactivity on CLL cells was (Table 2, P . 0.05). All PB B cells expressed CXCR5, whereas highly variable, ranging from 2.1 to 100% of CD19+/CD5+ in NBM two distinct B cell populations, a large CXCR5 posi- cells. As a recent report has suggested that CLL can be subdiv- tive one and a smaller CXCR5 negative one, could be detected ided into two biologically different groups according to CD38

Table 2 Chemokine receptor expression in B cell malignancies

B cell population CXCR4 CXCR5 CCR5 CCR6 (CD19 positive cells)

NBM 93.8 ± 2.3 (11/11) 68.4 ± 8.1 (11/11) 3.3 ± 0.9 (0/11) 65.8 ± 15.4 (4/4) PB 94.4 ± 0.5 (3/3) 95.9 ± 1.1 (3/3)a 6.0 ± 2.1 (0/3) 97.3 ± 0.7 (3/3)a Precursor B-ALL 90.5 ± 4.2 (6/6) 1.3 ± 0.5 (0/6)b,c 1.6 ± 0.7 (0/6) 4.6 ± 1.2 (0/3)b,c CLL 99.1 ± 0.5 (23/23)b,c 97.6 ± 1.0 (23/23)b 0.7 ± 0.1 (0/23) 42.3 ± 7.3 (13/18)c Multiple myeloma 65.4 ± 11.2 (8/10)b 7.2 ± 3.2 (0/10)b,c ND 5.0 ± 2.4 (0/4)b,c Mantle cell lymphoma 95.5 ± 2.8 (4/4) 93.2 ± 5.2 (4/4) 2.0 ± 0.4 (0/4) 58.2 ± 17.2 (4/4) Follicular lymphoma 87.3 ± 4.8 (3/3)b 90.0 ± 7.2 (3/3) 1.4 ± 0.7 (0/3) ND Hairy cell leukaemia 78.8 ± 10.9 (5/5)b 54.8 ± 10.6 (5/5)b,c 34.0 ± 11.9 (3/5)b,c 85.3 ± 11.8 (5/5)

Flow cytometric detection using MoAbs described in Materials and methods with gating on CD19+ lymphocytes; data are percentages of CD19+ B lymphocytes co-expressing the indicated chemokine receptor, mean ± standard error; numbers in brackets denote positive cases of n tested. aP , 0.05 by Student’s t-test for unpaired samples, comparing chemokine receptor expression to that seen on NBM CD19+ B cells. bP , 0.05 by Student’s t-test for unpaired samples, comparing chemokine receptor expression to that seen on NBM CD19+ B cells. cP , 0.05 by Student’s t-test for unpaired samples, comparing chemokine receptor expression to that seen on normal PB CD19+ B cells.

Leukemia Chemokine receptor expression in B cell malignancies JDu¨rig et al 754

Figure 1 CCR5 and CXCR5 expression in hairy cell leukaemia (HCL) and normal B cells. CCR5 (a) and CXCR5 (b) expression in ﬁve cases of HCL and in normal BM and PB-derived B cell populations (c). Two-colour ﬂow cytometry was used to analyse chemokine receptor expression in lymphocyte gated CD19+ B cells as detailed in Materials and methods.

expression,15 we correlated CD38 expression with the per- showed high interindividual variability (ranging from 20 to centage of cells staining positive for CXCR4, CXCR5 and 100%). CCR6. No positive correlation was observed in either group. To determine whether CCR6 expression may serve as a diagnostic factor identifying clinical subsets of patients with Chemokine receptor immunostaining in hairy cell favourable vs poor prognoses, we compared the CCR6+ stain- leukaemia (HCL) ing of B-CLL cells in patients with different clinical stages of the disease. The percentages of CCR6+ B-CLL cells in the Binet CXCR4 and CCR6 expression was not significantly different A(n = 9), B (n = 3) and C (n = 6) patient cohorts were from that of normal B cells. By contrast, CXCR5 immunostain- 49.5 ± 8.9, 22.1 ± 10.0 and 41.2 ± 16.5%, respectively ing was reduced (P , 0.05) and dot plot analysis in all five (P . 0.05). cases tested revealed two distinct populations (Figure 1), representing residual normal peripheral B cells (staining brightly positive for CXCR5, Figure 2) and HCL cells (CXCR5 dim to Chemokine receptor immunostaining in multiple intermediate, Figure 2). In 3/5 cases, HCL cells were found to myeloma express CCR5 (patients FS, RG and JK, Figure 1a). Three- colour analysis (Figure 3) confirmed that CCR5 expression was Comparing CD138 positive malignant plasma cells to NBM B restricted to the CD19+/CD11c+ HCL cell population, whereas cells a significant down-regulation of CXCR4, CXCR5 and residual normal (CD11c negative) B cells in the same sample CCR6 was observed (P , 0.05) in the majority of patients did not exhibit CCR5 immunoreactivity (Figure 3). (Table 2). While the percentage of plasma cells co-expressing CXCR4 ranged from 5.9 to 100%, CXCR5 and CCR6 were not detectable in any of the samples analysed. Discussion

Employing a panel of chemokine receptor-speciﬁc antibodies Chemokine receptor immunostaining in mantle cell directed against CXCR4, CXCR5, CCR6 and CCR5 in a multi- lymphoma and follicular lymphoma colour ﬂow cytometry approach we show that malignant B cells corresponding to different stages of B cell development Both lymphoma entities expressed CXCR4 and CXCR5, exhibit differential chemokine receptor expression patterns. whereas CCR5 immunoreactivity was absent. In analogy to In accordance with previous studies12,14 our data demon- the B-CLL data CCR6 expression in mantle cell lymphoma strate that CXCR4 is expressed in immature haemopoietic pro-

Leukemia Chemokine receptor expression in B cell malignancies JDu¨rig et al 755 phomas with a nodular dissemination pattern (eg B-CLL, MCL, follicular lymphoma). These observations are consistent with a report by Fo¨rster et al13 showing that CXCR5 expression defines mature recirculating B cells but is absent from early bone marrow lymphoid precursor cells. Furthermore, our finding that CXCR5 expression is particularly high in non- Hodgkin’s lymphomas with lymph node involvement supports the concept that CXCR5 and its ligand B cell-attracting chemokine-1 (BCA-1) may be involved in the homing of circulating malignant B cells to lymphoid tissues. In this context it is worth mentioning that BCA-1 is specifically expressed in B cell follicles, presumably by follicular dendritic cells (FDC18). Disruption of CXCR5 in transgenic mice leads to disturbed development of primary follicles and Peyer’s patches,19 under- scoring the functional importance of CXCR5/BCA-1 in B lymphocyte homing under physiological conditions. Interestingly, we show that CXCR5 is down-regulated in HCL and com- pletely absent in multiple myeloma. This latter observation Figure 2 CXCR5 is differentially expressed on HCL cells and confirms earlier findings of Fo¨rster et al using Northern blot residual normal PB B cells. Three-colour flow cytometric analysis hybridisation in murine plasmacytoma cell lines20 and flow using a CXCR5-specific MoAb with gating on the lymphocyte popu- 13 lation. Residual normal PB B cells (CD19+/CD103−,R1+ R5) stain cytometry in one patient with multiple myeloma and brightly positive for CXCR5 whereas the majority of HCL cells extends them to a larger patient cohort. Our data showing (CD19+/CD103+,R1+ R4) exhibit a substantially lower CXCR5 weak or even absent staining for CXCR5 on HCL and ALL expression. cells, respectively (Table 1), contrast with those reported by Jones et al,17 who demonstrated strong CXCR5 immunoreactivity on 30 cases of B cell lymphoma/leukemia corresponding to all maturational stages of B cell development. However, this discrepancy may be explained by differences in the experimental systems, the most important probably being the use of biotinylated CXCR5-specific antibodies by Jones et al17 as compared to a directly PE-conjugated antibody in the present study. CCR6 expression on normal mature circulating PB B cells was significantly higher than in NBM B cells, suggesting that CCR6 may be up-regulated in the course of B cell development. This observation is in accordance with a recent report by Bowman et al4 demonstrating in a murine in vivo model that MIP-3α responsiveness and expression of its cognate receptor CCR6 are only acquired after emigration of B cells to the periphery. CCR6 was found to be absent on precursor B-ALL blasts and markedly expressed on all of the remaining lymphoma entities tested with the exception of multiple myeloma. Thus, its expression pattern in B cell tumours paralleled Figure 3 CCR5 is expressed on HCL cells but not on residual nor- that observed for CXCR5, however, interindividual differences mal PB B cells. Three-colour flow cytometric analysis using a CCR5- within the different lymphoma subgroups were more pro- specific MoAb with gating on the lymphocyte population. Note differ- nounced than those noted for CXCR5. ential expression of CCR5 in HCL cells (CD19+/CD11c lymphocytes, + + − Unlike CXCR4, CXCR5 and CCR6, CCR5 has not been R1 R3) and residual normal B cells (CD19 /CD11c lymphocytes, recognized as a B cell associated chemokine receptor1,5,21 and R1 + R2). its expression in normal haemopoietic cells appears to be restricted to T cells, monocytes and dendritic cells.1 Conse- genitor cells and up-regulated as CD34+ cells develop into quently, our finding that CCR5 is not expressed in the majority mature B lymphocytes. CXCR4 surface immunoreactivity of B cell non-Hodgkin’s lymphomas was not unexpected. observed in malignant CD34+/CD19+/CD10+ ALL cells was Notably, CCR5 was expressed in three out of five HCL cases comparable to that of CD34+/CD10+ NBM immature lymph- (Figure 1) with no apparent association with the clinical stage oid progenitor cells,14 suggesting that CXCR4 up-regulation of the disease. To our knowledge, this is the first example of may be a common phenomenon in normal and malignant a cross-lineage aberrant chemokine receptor expression early B lymphopoiesis. CXCR4 expression levels in B-CLL described in a human lymphoproliferative disorder. Interest- were higher than in the respective PB B cell controls con- ingly, one of these HCL patients in whom CCR5 expression firming a recent study by Mo¨hle et al16 showing CXCR4 over- was observed, received IFN-α on a regular basis. In light of expression in B-CLL. Furthermore, we found that CXCR4 is our own recent results22 demonstrating up-regulation of MIP- down-regulated in multiple myeloma complementing results 1α receptors in CD34+ haemopoietic progenitor cells in recently reported for normal plasma cells.17 response to IFN-γ, it appears possible that IFN-α treatment in By contrast, CXCR5 was not detected on immature normal these patients may have contributed to the up-regulation of and leukemic B lymphoid progenitor cells but was strongly CCR5 in the malignant HCL clone. To further investigate this expressed by essentially all PB B cells and leukemic B lym- possibility HCL cells isolated from an untreated patient with

Leukemia Chemokine receptor expression in B cell malignancies JDu¨rig et al 756 a CCR5 negative phenotype (HG, Figure 1) were cultured in receptor for a novel lymphocyte-directed CC chemokine LARC. J the presence of varying concentrations of IFN-α (103 to 105 Biol Chem 1997; 272: 14893–14898. U/ml) for 24 h and subsequently analysed for CCR5 7 Legler DF, Loetscher M, Roos RS, Clark-Lewis I, Baggiolini M, α Moser B. B cell-attracting chemokine 1, a human CXC chemokine expression using flow cytometry. However, IFN- -treated expressed in lymphoid tissues, selectively attracts B lymphocytes cells did not exhibit increased CCR5 surface expression com- via BLR1/CXCR5. J Exp Med 1998; 187: 655–660. pared with that of normal controls (data not shown). 8 Drillenburg P, Pals ST. Cell adhesion receptors in lymphoma dis- Taken together, our findings raise several points. First, the semination. Blood 1900; 95: 1900–1910. pattern of chemokine receptor expression in normal B cells 9 Harris NL, Jaffe ES, Stein H, Banks PM, Chan JK, Cleary ML, Delsol appears to be paralleled by that seen in B cell non-Hodgkin’s G, De Wolf-Peeters C, Falini B, Gatter KC. A revised European– American classification of lymphoid neoplasms: a proposal from lymphoma depending on the maturational stage of the cells the International Lymphoma Study Group. Blood 1994; 84: investigated. In particular, while development of immature B- 1361–1392. lymphoid progenitors into more differentiated cells is associa- 10 Ho¨ffkes H, Schmidtke G, Uppenkamp M, Schmucker U. Multi- ted with a step-like up-regulation of CXCR4, CXCR5 and parametric immunophenotyping of B cells in peripheral blood of CCR6, terminal differentiation of mature B cells results in healthy adults by flow cytometry. Clin Diagn Lab Immunol 1996; down-regulation of these receptors. Second, CXCR5 which is 3: 30–36. 11 Bene M, Castoldi G, Knapp W, Ludwig WD, Matutes E, Orfao A, known to play a pivotal role in normal B lymphocyte homing van’t Veer MB (EGIL). Proposal for the immunological classi- may also be involved in the dissemination of malignant B cells fication of acute leukemias. Leukemia 1995; 9: 1783. to lymphoid tissues. Thus, MoAbs aimed at this receptor may 12 Aiuti A, Tavian M, Cipponi A, Ficara F, Zappone E, Hoxie J, Peault be used as a potential adjunct in the treatment of patients with B, Bordignon C. Expression of CXCR4, the receptor for stromal CXCR5 positive non-Hodgkin’s lymphomas. Third and finally, cell-derived factor-1 on fetal and adult human lympho- given the differential expression of chemokine receptors in dif- hematopoietic progenitors. Eur J Immunol 1999; 29: 1823–1831. 17 13 Fo¨rster R, Emrich T, Kremmer E, Lipp M. Expression of G-protein- ferent lymphoma entities, chemokine receptor-specific anti- coupled receptor BLR1 defines mature, recirculating B cells and bodies may serve as diagnostic tools in the flow cytometric a subset of T-helper memory cells. Blood 1994; 84: 830–840. evaluation of non-Hodgkin’s lymphomas. 14 Du¨rig J, Rosenthal C, Elmaagacli A, Heyworth C, Halfmeyer K, Kasper C, Novotny J, Du¨hrsen U. Biological effects of stroma In general, recent data have shown that receptor expression + alone is not an adequate parameter to predict a cell’s chemo- derived factor-1 on normal and CML CD34 haemopoietic pro- tactic potential.4,23 For instance, mature B cells express higher genitor cells. Leukemia 2000; 14: 1652–1660. 15 Damle R, Wasil T, Fais F, Ghiotto F, Valetto A, Allen SL, Buch- levels of CXCR4 receptors than uncommitted progenitors but binder A, Budman D, Dittmar K, Kolitz J, Lichtman SM, Schulman exhibit a reduced chemotactic response to SDF-1α suggesting P, Vinciguerra VP, Rai KR, Ferrarini M, Chiorazzi N. Ig V gene that signal transduction events downstream from CXCR4 may mutation status and CD38 expression as novel prognostic indi- also regulate responsiveness of B cells to SDF-1α.23 Further- cators in chronic lymphocytic leukemia. Blood 1999; 94: 1840– more, our own previous work14,24 and that of others25 has 1847. demonstrated that even in the presence of normal receptor 16 Mo¨hle R, Failenschmid C, Bautz F, Kanz L. Overexpression of the chemokine receptor CXCR4 in B cell chronic lymphocytic leuke- levels aberrant responses to chemokines can be observed in mia is associated with increased functional response to stromal malignant haemopoietic cells that may also be due to defects cell-derived factor-1 (SDF-1). Leukemia 1999; 13: 1954–1959. in intracellular signalling. Thus, the flow cytometric analysis 17 Jones D, Benjamin RJ, Shahsafaei A, Dorfman DM. The chemo- of chemokine receptor expression described here needs to be kine receptor CXCR3 is expressed in a subset of B-cell lymphomas complemented by functional studies investigating the biologi- and is a marker of B-cell chronic lymphocytic leukemia. Blood cal response of lymphoma cells to the respective chemokines. 2000; 95: 627–632. 18 Gunn M, Ngo V, Ansel K, Ekland E, Cyster J, Williams L. A B-cell- homimg chemokine made in lymphoid follicles activates Burkitt’s lymphoma receptor-1. Nature 1998; 391: 799–803. Acknowledgements 19 Fo¨rster R, Mattis A, Kremmer E, Wolf E, Brem G, Lipp M. A puta- tive chemokine receptor, BLR-1, directs B cell migration to defined This work was supported by a grant from the Deutsche For- lymphoid organs and specific anatomic compartments of the schungsgemeinschaft (SFB 545/B5). spleen. Cell 1996; 87: 1037–1047. 20 Kaiser E, Fo¨rster R, Wolf I, Ebensperger C, Kuehl W, Lipp M. The G-protein-coupled receptor BLR1 is involved in murine B-cell differentiation and is also expressed in neuronal tissues. Eur J Immu- References nol 1993; 23: 2532–2539. 21 Rollins BJ. Chemokines. Blood 1997; 90: 909–928. 1 Murdoch C, Finn A. Chemokine receptors and their role in 22 Du¨rig J, de Wynter EA, Kasper C, Cross MA, Chang J, Testa NG, inflammation and infectious diseases. Blood 2000; 95: 3032– Heyworth CM. Expression of macrophage inflammatory protein- 3043. 1 receptors in human CD34(+) hematopoietic cells and their 2 Cyster JG. Chemokines and cell migration in secondary lymphoid modulation by tumor necrosis factor-alpha and interferon-gamma. organs. Science 1999; 286: 2098–2102. Blood 1998; 92: 3073–3081. 3 Bleul C, Wu L, Hoxie J, Springer T, Mackay C. The HIV corecep- 23 Fedyk E, Ryyan D, Ritterman I, Springer T. Maturation decreases tors CXCR4 and CCR5 are differentially expressed and regulated responsiveness of human bone marrow B lineage cells to stromal- on human T lymphocytes. Proc Natl Acad Sci USA 1997; 94: derived factor 1 (SDF-1). J Leukoc Biol 1999; 66: 667–673. 1925–1930. 24 Du¨rig J, Testa NG, Heyworth CM. Distinct biological effects of 4 Bowman EP, Campbell JJ, Soler D, Dong Z, Manlongat N, Picarella macrophage inflammatory protein-1alpha and stroma-derived fac- D, Hardy R, Butcher EC. Developmental switches in chemokine tor-1alpha on CD34+ hemopoietic cells. Stem Cells 1999; 17: response profiles during B cell differentiation and maturation. J 62–71. Exp Med 2000; 191: 1303–1318. 25 Salgia R, Quackenbush E, Lin J, Souchkova N, Sattler M, Ewaniuk 5 Baggiolini M. Chemokines and leukocyte traffic. Nature 1998; DS, Klucher KM, Daley GQ, Kraeft SK, Sackstein R, Alyea III EP, 392: 565–568. von Andrian UH, Chen LB, Gutierrez-Ramos J-C, Pendergast AM, 6 Baba M, Imai T, Nishimura M, Kakizaki M, Takagi S, Hieshima Griffin JD.The bcr/abl oncogene alters the chemotactic response K, Nomiyama H, Yoshie O. Identification of CCR6, the specific to stromal-derived factor-1alpha. Blood 1999; 94: 4233–4246.

Leukemia