New Variations of Human CC-Chemokine Receptors CCR3 and CCR4

Home , CCR3 (gene), CCR4, CCR5, CC chemokine receptors

H Kato1, N Tsuchiya1, S Izumi2, M Miyamasu2, T Nakajima3, H Kawasaki4, K Hirai3 and K Tokunaga1 1Department of Human Genetics, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan 113- 0033; 2Department of Allergy and Rheumatology, Graduate School of Medicine, University of Tokyo; 3Department of Bioregulatory Function, Graduate School of Medicine, University of Tokyo; 4Department of Clinical Immunology and AIDS Research Center, Institute of Medical Sciences, University of Tokyo, Japan

CCR3 and CCR4 are the members of CC chemokine receptor family expressed on Th2 type CD4+ T cells. In this study, variation screening of the entire coding regions of CCR3 and CCR4 was performed, and possible association with several autoimmune diseases was tested, using the genomic DNA from 304 Japanese healthy individuals and 272 Japanese patients with rheumatic diseases. One non-synonymous substitution was identified in CCR3 gene, whereas in CCR4 gene, two non-synonymous and two synonymous substitutions were detected. Among the synonymous substitutions, CCR4 1014(C→T) was observed in 7.2% of the healthy individuals and 6.6% of the patients, and was considered as a single nucleotide polymorphism. All other variations were observed in only one or two individuals. No significant association was observed between any of the variations and any of the rheumatic diseases. Among these variations, CCR3-C218S substitution coded by 652(T→A) substitution was localized in the region conserved among the G protein coupled receptor family. Reactivity of eosinophils to the monoclonal antibody against CCR3 and the chemotaxis to eotaxin were slightly reduced in this patient as compared with healthy controls or a patient with Behçet disease homozygous for the common allele, while CCR3 mRNA level was not different. These findings suggest that CCR3-C218S substitution may lead to the reduced function of CCR3 at the protein level. Further study will be of interest to test whether CCR3-C218S variation or any of the CCR4 variations has a significant role in rendering susceptibility to immunological diseases or resistance to HIV infection.

Keywords: CCR3; CCR4; SNP; Behc¸et disease; eotaxin

Introduction expressed on eosinophils but also on basophils,3–5 and on a subset of T lymphocytes with a certain type of Th2 The chemokines are a family of chemotactic cytokines helper properties that colocalize with eosinophils on that mediate a range of pro-inflammatory effects on leu- allergic sites,6–8 and is believed to be the principal chemo- kocytes, including chemotaxis, degranulation, and inte- kine receptor for the recruitment of these cells.1,9,10 CCR4 grin activation. Chemokines are divided into four groups is expressed on CD4+ T lymphocytes producing Th2 type based on the primary sequence of the first two cysteines: cytokines in thymus and peripheral blood, and is con- ␣ ␤ ␥ ␦ + the CXC ( -), CC ( -), C ( -), and CX3C ( -) chemokines. sidered to play a role in CD4 T cell homing.8 Both CCR3 Chemokines act via seven-transmembrane-domain (7TM) and CCR4 are recognized as specific markers of Th2 sub- receptors, which form a distinct group of structurally set among the CD4+ T cells.7,8,10 related proteins within the superfamily of receptors that CCR4 binds two chemokines, thymus and activation- signal through heterotrimeric GTP-binding proteins (G regulated chemokine (TARC)11 and macrophage-derived proteins). A number of conserved structural motifs have chemokine (MDC).12 They are lymphocyte specific been found mainly within the transmembrane chemokines and are believed to act specifically on chron- 1,2 domains. ically activated Th2 lymphocytes via CCR4.13–15 CCR3 and CCR4 are members of the CC chemokine It is remarkable that the pathophysiologically relevant receptor family. CCR3 was shown to be prominently leukocytes share CCR3 expression and can be recruited concomitantly to sites of allergic inflammation by the same set of chemokines. T lymphocytes expressing CCR3 Correspondence: Dr N Tsuchiya, Department of Human Genetics, Gradu- are found to colocalize with eosinophils in atopic derma- ate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, titis, nasal polyps and ulcerative colitis, whereas the lym- Tokyo, 113-0033, Japan. E-mail: [email protected] phocytes in non-allergic infiltrates are generally CCR3- This study was supported by a Grant-in-Aid for Scientific Research negative.16 CCR3 binds several chemokines. Eotaxin and (B109557215) from the Ministry of Education, Science, Sports and eotaxin-2 bind only to CCR3 whereas RANTES, mono- Culture, Uehara Memorial Foundation, Wakunaga Pharmaceutical Co Ltd, and a donation from Dr Hidehiko Masatsuka (Masatsuka cyte chemotactic protein (MCP)-2, MCP-3, and MCP-4 17,18 Clinic, Saitama, Japan). also bind to other CC-chemokine receptors. In mice Received 4 May 1999; accepted 18 May 1999 with a deletion of the eotaxin gene, allergen-induced Variations of human CCR3 and CCR4 H Kato et al 98 eosinophil infiltration into the lungs was retarded, but variation was identified in CCR3 (Figure 1, Table 1). A T- the defect was compensated at later stages by other to-A single nucleotide substitution was detected at pos- chemokines.17 Thus, a more sustained effect could result ition 652 (counting from the ATG start codon), which from disruption of the gene encoding CCR3. resulted in the non-conservative amino acid change of Another important role of chemokine receptor is to cysteine at position 218 to serine (CCR3-C218S) within serve as a coreceptor for human immunodeficiency virus the fifth transmembrane region (TM5) of the CCR3. Simi- (HIV)-1 infection. CCR3 is a coreceptor for some types larly, four novel single nucleotide substitutions were of pathogenic macrophage (M)-tropic strains, especially detected in CCR4 (Figure 1, Table 1), two of which neurotropic HIV-1 isolates that are likely to infect the resulted in the amino acid substitutions. In addition, we central nervous system and cause the dementia associa- incidentally detected four additional mismatches with ted with acquired immunodeficiency syndrome the published sequence28 at nucleotide positions (AIDS).19–23 In addition, HIV-1-specific lytic activity of 414(G→A), 417(A→T), 1068(T→C) and 1169(C→T). cytotoxic T lymphocytes (CTLs) derived from the patients Because all of the 12 sequenced alleles commonly con- was reported to be enhanced by RANTES, which acts on tained all four substitutions, it is likely that these the CTLs through CCR3. The mechanism involved in sequences define the common allele of CCR4 at least in such a phenomenon is different from the role of RANTES Japanese. as a competitor for HIV-1 coreceptor usage.24 Besides, MDC was reported to have a suppressive activity on HIV-1, which is not dependent on strain tropism,25 sug- Analysis of association between CCR3 or CCR4 gesting a role for CCR4 in HIV-1 entry. variations and rheumatic diseases Based on such evidence, we hypothesized that genetic We next examined whether any of the variations of CCR3 variations of CCR3 and CCR4 might be associated with or CCR4 is associated with the susceptibility to any of the the susceptibility to a range of immunological diseases, rheumatic diseases, using case-control association analy- especially autoimmune rheumatic diseases in which the sis. The results are summarized in Table 1. imbalance of Th1 and Th2 cells has been implicated.26,27 CCR3-C218S, CCR4-L130V[388(C→G)], CCR4- Thus far no attempts have been reported on the screening V166V[498(G→C)] were detected in one sample from a of gene variations of these receptors. In the present study, patient with Behçet disease, from a patient with rheuma- we performed a systematic variation screening of human toid arthritis (RA), and from a healthy individual, CCR3 and CCR4, and made an attempt to examine the respectively. CCR4-C178S[533(G→C)] was detected in possibility that any of the genetic variation may be asso- one patient with systemic lupus erythematosus (SLE) and ciated with the susceptibility to rheumatic diseases. one healthy individual. All of these individuals were heterozygous for the common allele and the variant alleles. Results Thus, these variations could be regarded as relatively rare variants. Identification of new variants of CCR3 and CCR4 On the other hand, CCR4-Y338Y [1014(C→T)] was Systematic screening of the entire coding regions of CCR3 detected in 22 of 304 healthy individuals (7.2%) and in and CCR4 for gene variations was performed on genomic 18 of 272 patients (6.6%). One patient with RA was homo- DNA samples from healthy Japanese individuals and zygous for CCR4-Y338Y. This variation could be con- patients, using a polymerase chain reaction-single-strand sidered as a single nucleotide polymorphism (SNP), conformation polymorphism (PCR-SSCP) method. Rep- although no association with any of the tested diseases resentative SSCP pattern is shown in Figure 1. The nucle- seemed to be present. Only one patient with RA pos- otide sequences of the detected variations were deter- sessed two of the variations (CCR4-L130V, Y338Y): all mined by direct sequencing. other individuals with variations had a base substitution Using this approach, one novel nucleotide sequence only at a single position.

Figure 1 SSCP patterns of ampliﬁed fragments of CCR3 (a) and CCR4 (b,c). (a) Lanes 1, 2, 3, 5, homozygotes of CCR3 common allele. Lane 4, a heterozygote of a variant allele CCR3-C218S and the common allele. (b) Lanes 4, 5, homozygotes of CCR4 common allele. Lanes 1, 2, 3, a heterozygote of variant alleles and the common allele. Subsequent direct sequencing revealed that the variants encoded CCR4-L130V (Lane 1), CCR4-V166V (Lane 2), CCR4-C178S (Lane 3) substitutions. (c) Lanes 1, 4, 5, homozygotes of CCR4 common allele. Lane 2, a homozy- gote of a variant allele CCR4-Y338Y, and Lane 3, a heterozygote of CCR4-Y338Y allele and the common allele. Variations of human CCR3 and CCR4 H Kato et al 99 Table 1 Detected nucleotide substitutions and positivity among the patients with rheumatic diseases and healthy individulals

Substitutions in Positivity (%)

Nucleotidea AAa positionb RAd SLEd myositis scleroderma Behc¸et Sjo¨gren MCTDd ASd Control (n = 145) (n = 66) (n = 13) (n = 12) (n = 12) (n = 12) (n = 7) (n = 5) (n = 304)

CCR3 652(T→A) C218S 5TM 0 (0) 0 (0) 0 (0) 0 (0) 1 (8.3) 0 (0) 0 (0) 0 (0) 0 (0)

CCR4c 388(C→G) L130V 3TM 1(0.7) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 498(G→C) V166V 4TM 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (0.3) 533(G→C) C178S 2nd EC loop 0 (0) 1 (1.5) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (0.3) 1014(C→T) Y338Y C tail 12 (8.3) 3 (4.5) 1 (7.7) 0 (0) 0 (0) 1 (8.3) 1 (14.3) 0 (0) 22 (7.2) aNumbering starts at the ATG translation initiation. AA, amino acid. bTM, transmembrane domain; EC, extracellular domain; C tail, C terminal tail. cIn addition, four mismatches with the published sequence, 414(G→A), 417(A→T), 1068(T→C), 1169(C→T), were detected in all sequenced alleles. 1169(C→T) was located within 3′-untranslated region, and all other three were synonymous substitutions. dRA, rheumatoid arthritis; SLE, systemic lupus erythematosus; MCTD, mixed connective tissue disease; AS, ankylosing spondylitis.

Flow-cytometric analysis of cell-surface expression of C218S variant of CCR3 Since CCR3-C218S substitution is located within the TM5 and changes the important conserved motif M(X2)CY, it was considered possible that this variation may have a significant effect on the level of expression or on the conformation of the expressed protein. To test whether the variant has significant alteration in the level of cell-surface expression, flow cytometric analysis using a monoclonal antibody (mAb) to CCR3 was carried out. The reactivity of mAb to peripheral blood eosinophils from one patient with Behçet disease carrying CCR3-C218S substitution was compared with that to eosinophils from healthy individuals and one patient with Behçet disease who were homozygous for the common allele. CD16 was used as the surface marker to exclude neutrophils. The results are shown in Figure 2. Relative mean fluorescence intensity (relative MFI) of CCR3 was calculated by the mean fluorescence intensity (MFI) of anti-CCR3 staining of CD16-negative polymorphonuclear cells divided by that of control mAb staining. Slight reduction in the fluorescence intensity was observed in the patient with CCR3-C218S substitution (relative MFI = 2.6) compared with the patient without substitution (relative MFI = 4.0) and two healthy individuals without substitution (relative MFI = 4.0).

Chemotaxis analysis of C218S variant of CCR3 We next examined whether CCR3-C218S substitution has a significant effect on the interaction of CCR3 with its specific ligand, eotaxin, using chemotaxis analysis. The Figure 2 Flow cytometric analysis of CCR3 expression on eosino- results of chemotactic activity of the eosinophils are phils from a patient with Behçet disease heterozygous for 218C and shown in Table 2. Chemotactic activity was slightly 218S alleles, a patient with Behçet disease and two healthy individ- reduced in the patient with Behçet disease carrying uals homozygous for 218C allele. Cells were stained with FITC- C218S as compared with a patient without C218S anti-CCR3 mAb (clone No. 444) and PE-anti-CD16 Ab. (a) a patient ± ± = with Behçet disease with 218 C/S genotype, (b) a patient with (30.5 3.7 vs 52.2 8.9 %, P 0.088) and a healthy indi- Behçet disease with 218 C/C genotype, (c, d) healthy individuals vidual not carrying C218S (30.5 ± 3.7 vs 54.0 ± 8.1%, with 218 C/C genotype. Total granulocytes were gated based on P = 0.059) at the eotaxin concentration of 200 ng/ml, forward and side scatters and analyzed. The majority of CD16- although the difference was of borderline significance. negative granulocytes was considered to be eosinophils, and the Both patients with Behçet disease were taking low dose remainder to be basophils. Inset of each profile indicates the mean colchicine (0.5 g/day); therefore, the difference could not fluorescence intensity of anti-CCR3 staining of CD16-negative polymorphonuclear cells divided by that of control mAb staining be explained by colchicine. These results suggested that (relative MFI). Relative MFI of the patient with heterozygous substi- the substitution may lead to a functional abnormality tution (a) was lower than that of the patient without substitution of CCR3. (b) and controls without substitution (c, d). Variations of human CCR3 and CCR4 H Kato et al 100 Table 2 Chemotaxis of eosinophils from a patient with Behçet dis- CCR3-C218S allele was detected in only one individual ease carrying CCR3–218C/S genotype to eotaxin with Behçet disease among 304 healthy Japanese individuals and 272 patients, indicating that this is probably a a Eotaxin conc. Migrated cells (%) rare variant, at least in Japanese. Since only one of 12 (ng/ml) patients with Behçet disease was positive for this allele, Behçet disease Behçet disease Control it seems unlikely that this mutation is associated with the 218 C/S 218 C/C 218 C/C susceptibility to Behçet disease. However, the accumu- lation of the T cells producing Th1 cytokine in the b ± 1.6 — 1.6 0.1 — inflamed tissues of the patients with Behcet disease32 sug- 8 — 3.8 ± 1.2 — ¸ 40 6.2 ± 0.8 10.9 ± 3.3 7.3 ± 1.2 gests the imbalance of Th1/Th2 cells, which might be 200 30.5 ± 3.7*† 52.2 ± 8.9 54.0 ± 8.1 related to the possible hypofunction of Th2 cells caused by the C218S mutation. Further analysis on a larger num- *P = 0.059 (Student’s unpaired t-test, Behçet disease-218C/S vs Con- ber of patients with Behçet disease will be necessary to trol-218C/C ). †P = 0.088 (Student’s unpaired t-test, Behçet disease- draw a definite conclusion. 218C/S vs Behçet disease-218C/C). aMean ± s.e.m. of the triplicate Human CCR4 gene is located at chromosome 3p24.29 In b experiments. no migration. CCR4 gene, we identified four novel nucleotide sequence variations, two of which result in amino acid substitution. RT-PCR analysis of mRNA expression of C218S Y338Y [1014(C→T)] substitution was observed in 7.2% of variant of CCR3 healthy individuals and 6.6% of patients, including one Reduced reactivity with the monoclonal antibody and patient with RA who was homozygous for this substi- lower response to eotaxin of the eosinophils from the het- tution. This variation can be considered as a SNP. erozygous patient may result from the conformational Although significant difference in the allele positivity change of the CCR3 or lower mRNA expression of the was not observed between patients and controls in this CCR3-C218S allele. To test whether the variant has sig- study, this SNP may potentially be useful for the analysis nificant alteration in the level of mRNA expression, quan- of susceptibility gene to other diseases, such as atopy titative RT-PCR analysis using competitor DNA was car- or asthma. ried out. The quantity of CCR3 mRNA from eosinophils In two non-synonymous substitutions, L130V substi- from the patient with Behçet disease carrying CCR3- tution is located in TM3 region close to the DRY motif, C218S substitution was estimated to be 1.8 pg/106 cells, and changes the IFF(X2)L(X2)I motif which is conserved and that from the patient without that substitution was in the CCR family.1 Because the DRY motif is thought to 1.6 pg/106 cells. These results indicated that the substi- play an important role in coupling G protein ␣ subunit,30 tution does not lead to reduction in the CCR3 mRNA it is also possible that L130V substitution has a substan- level. tial influence on subsequent signal transduction. L130V substitution was observed in one RA patient, who was Discussion also heterozygous for Y338Y substitution. Although it is unlikely that this combination is associated with the sus- In the present study, we carried out a variation screening ceptibility to RA in general, it cannot be excluded that of the entire coding regions of human CCR3 and CCR4, L130V substitution resulted in an abnormality of signal and detected a number of new variations. Although evi- transduction via CCR4 and hypofunction of Th2 cells, dence of association with rheumatic diseases was not leading to the imbalance of Th1/Th2 cells related to RA, obtained, some of the variations seem to be potentially in this particular patient. important in terms of the function of the receptors. Another non-synonymous substitution, C178S, located Because of the relatively small number of the patients in the second extracellular loop, was observed in one with each disease, the association study described in this healthy individual and one patient with SLE, both of paper should be considered as a preliminary one. Large them were heterozygous for C178S and the common scale case-control studies with age- and sex-matched con- allele. Because it is possible that C178S substitution trols and family-based association studies will be of parti- results in the conformational change, it will be interesting cular interest in the future. to test the reaction with monoclonal antibodies to CCR4, Human CCR3 gene is located at chromosome 3p21.3.29 which are as yet not available. Its product is expressed mainly on eosinophil surface.4 Another potential significance of CCR3 and CCR4 vari- The newly identified variation, C218S substitution, is ants relates to the HIV-1 infection. The important regions located in TM5 region, and changes the conserved motif of M-tropic HIV-1 gp120 binding to chemokine receptor in the G protein coupled receptor (GPCR) family, are considered to be N-terminal and the first extracellular M(X2)CY into M(X2)SY. The M(X2)CY motif is con- loop in CCR5,33 but other regions are also suggested to sidered to play an important role in the construction of be involved.9,34,35 It is theoretically possible that CCR3- the ligand binding pocket, and the change of that motif C218S allele might have a protective effect in the infection is likely to influence on ligand affinity.30,31 Slightly dimin- with HIV-1 that uses CCR3 as a coreceptor, especially ished reactivity to monoclonal antibody and the response neurotropic isolates. Although HIV-1 strain that uses to eotaxin in spite of normal level of mRNA in the eosino- CCR4 as a coreceptor has not been reported, CCR4 spe- phils from the heterozygous individual seem to be com- cific ligand, MDC, was reported to have an anti-HIV-1 patible with the possibility that C218S substitution may effect.25 Because MDC is considered to perform its anti- result in the conformational change of the ligand binding HIV-1 activity via CCR4, the newly identified variants in pocket of the CCR3. It will be of interest to test the effect this study might have a role in the infection with HIV-1 of this substitution on the response to other CCR3 ligand, or in the progression of AIDS. such as RANTES, in the future study. In conclusion, new variants of CCR3 and CCR4 genes Variations of human CCR3 and CCR4 H Kato et al 101 were detected in the Japanese population. Such infor- phoresis was carried out in 0.5 × TBE (45 mm Tris-borate mation will be important in future studies on the possible [pH 8.0], 1 mm EDTA) under constant electric current , involvement of CCR3 and CCR4 in immunological dis- using a minigel electrophoresis apparatus with a constant eases including atopy, and in HIV-1 infection. temperature control system (AE-6410 and AE-6370; ATTO, Tokyo, Japan). The optimal temperature and the duration of electrophoresis were determined by prelimi- Materials and methods nary experiments (Table 3). Single-strand DNA fragments in the gel were visualized by a staining dye (SYBR Patients and healthy individuals Gold, Molecular Probes, Inc, Eugene, OR, USA). A total of 304 healthy individuals and 272 patients with rheumatic diseases were examined. The patient group Direct sequencing consisted of 145 patients with RA, 66 patients with SLE, PCR products were amplified and both sense and anti- 13 patients with dermatomyositis/polymyositis, 12 sense strands were directly sequenced using the same patients with scleroderma, 12 patients with Behçet dis- primers as those for the SSCP analysis. Fluorescence- ease, 12 patients with Sjo¨gren syndrome, seven patients based automated cycle sequencing of PCR products was with mixed connective tissue disease (MCTD), and five performed on an automated sequencer (ABI PRISM 377 patients with ankylosing spondylitis (AS). These diseases DNA Sequencing System, Perkin-Elmer Applied Biosys- were diagnosed according to the generally accepted cri- tems, Foster City, CA, USA), using dye-terminator 36–38 teria. All healthy individuals and patients were unre- method according to the manufacturer’s instructions lated Japanese living in the Tokyo area. The central part (ABI PRISM dRhodamine Terminator Cycle Sequen- of Japan has been shown to be homogeneous with respect cing-Ready Reaction Kit). to the genetic background,39 permitting the case-control approach employed in this study. Healthy individuals Flow-cytometric analysis of CCR3 expression on consisted of the researchers, laboratory workers and stu- eosinophils dents at the institutions where this study was carried out. Fluorescein isothiocyanate (FITC)-mouse IgG1, phycoery- thin (PE)-mouse IgG1 and PE-anti-CD16 Ab were pur- Genomic DNA chased from Beckman Coulter, Inc (Fullerton, CA, USA). Genomic DNA from patients and healthy individuals The monoclonal antibody against human CCR3 (clone was purified from peripheral blood leukocytes using a No. 444) was generated from CCR3(+) transfectant immu- commercial kit (QIAamp blood kit, QIAGEN, Hilden, nized mouse as previously described.43 FITC labeled anti- Germany). CCR3 Ab (clone No. 444) was prepared by conjugation of purified Ab with FITC (Sigma Chemical Co, St Louis, PCR-single-strand conformation polymorphism (PCR- MO, USA). SSCP) Flow-cytometric analysis of CCR3 expression on The primers used for PCR and the annealing temperature eosinophils was performed with whole-blood immuno- are shown in Table 3. These primers were designed staining as described previously.44 Briefly, 50 ␮l of whole- according to the genomic DNA sequence of CCR3,4 and blood were diluted with the same volume of PBS. The the cDNA sequence of CCR4.28 CCR3 is a single exon leukocytes were stained with FITC-anti-CCR3 Ab and gene with the coding region of 1068 bp, and CCR4 is PE-anti-CD16 Ab at 4°C for 30 min. Erythrocytes were thought to be intronless at least within the coding region lysed by treatment with 10 mm potassium bicarbonate of 1083 bp [see reference 40, unpublished observations].40 containing 0.155 mm ammonium chloride and 0.1 mm The CCR3 coding region was divided into seven overlap- EDTA. To analyze the eosinophils within the whole- ping segments and the CCR4 coding region into six seg- blood, total granulocytes were gated based on forward ments of 250–300 bp, because the size of the fragment and side scatters on FACScan (Becton Dickinson, Moun- optimal for SSCP was determined to be 200–400 bp in the tain View, CA, USA). The CCR3 expression was analyzed previous studies41 (Table 3). PCR was carried out using in the CD16-negative population of the gated cells. The commercially available reagents (GeneAmp and Ampli- majority of the CD16-negative population of the gated Taq Gold, Perkin-Elmer, Norwalk, CT, USA). To activate cells was considered to be eosinophils, and the remainder DNA polymerase, preheating was performed for 10 min to be basophils. at 96°C. The amplification conditions consisted of 35 cycles of denaturation for 30 s at 96°C, annealing for 30 s Chemotaxis analysis of CCR3 chemoattractant to at the temperature shown in Table 3, and extension for eosinophils 1 min at 72°C, in a thermal cycler (Thermal cycler MP, Eosinophil chemotaxis assay was performed as described Takara, Kyoto, Japan). previously.5 In brief, eosinophils were partially purified The amplified DNA was analyzed using SSCP. The using dextran sedimentation, density centrifugation with SSCP analysis was performed essentially according to the Percoll (1.088 g/ml, Pharmacia, Uppsala, Sweden), and method described previously.42 One microliter of sol- hypotonic lysis of contaminating erythrocytes, as pre- ution containing the PCR product was mixed with 7 ␮l viously described.45 The migration assay was performed of denaturing solution (95% formamide, 20 mm EDTA, using an assembly consisting of a 96-well chamber 0.05% bromphenol blue, 0.05% xylene cyanol FF). The (Neuro Probe) and a polycarbonate filter membrane with mixtures were denatured at 96°C for 5 min and immedi- a pore diameter of 5 ␮m (Nucleopore, Pleasanton, CA, ately cooled on ice. Five microliters of the mixtures were USA). Samples to be tested were transferred to each well applied to 10% polyacrylamide gel (acrylamide: of a flat-bottom 96-well plate, and a cell suspension bis = 49:1). Five percent glycerol were included in the gel which contained 1–2 × 104 Percoll-separated eosinophils for the analyses of several fragments (Table 3). Electro- was introduced into the top compartment. The chamber Variations of human CCR3 and CCR4 H Kato et al 102 Table 3 Primers used in this study and SSCP conditions for each fragment

Name Primer sequence Positiona Fragment size Electrophoresis conditionsb

CCR3-1 CCR3A1 5′-ACTGGTGTGTTTTACGGAGG-3′ 5′ UTR 252 bp 15°C CCR3B1 5′-GATCATCACCACCACCACAT-3′ TM1 CCR3-2 CCR3A2 5′-ACTCCCTGGTGTTCACTGTG-3′ TM1 237 bp 10°C(+) CCR3B2 5′-CGCTGTACAAGCCTGTGTGA-3′ TM3 CCR3-3 CCR3A3 5′CTTCCTCGTCACCCTTCCAT-3′ TM2 253 bp 15°C CCR3B3 5′-AAGAGCTGCTAGCACTGCCA-3′ TM4 CCR3-4 CCR3A4 5′-GGTACCTGGCCATTGTCCAT-3′ CY 2nd loop 251 bp 22°C(+) CCR3B4 5′-AACGAGCAGAGGGAGAACGA-3′ TM5 CCR3-5 CCR3A5 5′-AGCTGGAGGCATTTCCACAC-3′ EC 2nd loop 249 bp 10°C(+) CCR3B5 5′-CGTCCGCTCACAGTCATTTC-3′ EC 3rd loop CCR3-6 CCR3A6 5′-GGACACCCTACAATGTGGCT-3′ TM6 255 bp 10°C CCR3B6 5′-CCAGCTTCTCACTAGGAAGG-3′ C terminal CCR3-7 CCR3A7 5′-ACAGGCACTTGCTCATGCAC-3′ C terminal 236 bp 15°C CCR3B7 5′-GAGCTTCAGTGTTGCACTGG-3′ 3′ UTR

CCR4-1 CCR4A1 5′-AGCAAGCTGCTTCTGGTTGG-3′ 5′ UTR 283 bp 18°C CCR4B1 5′-CATCAGTCATGGACCTGAGC-3′ CY 1st loop CCR4-2 CCR4A2 5′-AGCCTTGCACCAAAGAAGGC-3′ N terminal 280 bp 12°C CCR4B2 5′-CCCACCAAGTACATCCAGGA-3′ TM3 CCR4-3 CCR4A3 5′-ACTATGCAGCAGACCAGTGG-3′ TM2 302 bp 6°C CCR4B3 5′-CGTCGTGGAGTTGAGAGAGT-3′ EC 2nd loop CCR4-4 CCR4A4 5′-CCTGGCTTTCTGTTCAGCAC-3′ TM4 293 bp 8°C CCR4B4 5′-ACCAGGGTCTCTAGGAAGAG-3′ TM6 CCR4-5 CCR4A5 5′-GTGGTCCTCTTCCTTGGGTT-3′ TM6 299 bp 12°C CCR4B5 5′-GGACTGCGTGTAAGATGAGC-3′ C terminal CCR4-6 CCR4A6 5′-GTGCTCTGCCAATACTGTGG-3′ C terminal 291 bp 12°C(+) CCR6B6 5′-CCAGGCTGGACTTGTCTAGT-3′ 3′ UTR

PCR annealing temperature: Each fragment of CCR3 gene : 57°C; each fragment of CCR4 gene : 60°C. aPrimer position: UTR, untranslated region; EC, extracellular domain; TM, transmembrane domain; CY, cytoplasmic domain. bSSCP conditions: Except for CCR4–3, 90 min duration under 20 mA/gel; for CCR4–3 fragment, 150 min duration under 10 mA/gel; (+) : 5% glycerol was added in 10% polyacrylamide gel.

° was incubated at 37 C, 5% CO2 for 90 min. The plate was Amplification of cDNA was performed as previously centrifuged to detach the cells from the lower surface of described46 except that AmpliTaq Gold DNA polymerase the filter. To determine the number of migrated eosino- (Perkin-Elmer, Branchburg, NJ, USA) was used. Direct phils, the cell-associated eosinophil peroxidase content in and reverse oligo primers for CCR3 (5′-ATCCTACTAT the lower wells was determined by using o-phenylene GATGACGTGGGC-3′; position TM1 for sense and 5′- diamine as the substrate. The numbers of migrated GGCCAGGTACCTGTCGATTG-3′; position TM3 for eosinophils were calculated based on a calibration curve antisense) were constructed based on the published established with varying known numbers of eosinophils. sequences of mRNA. To activate DNA polymerase, preheating (9 min at 95°C) was performed. Then amplifi- RT-PCR analysis of CCR3 mRNA expression on cation was performed for 30 cycles of denaturation (30 s eosinophils at 94°C), annealing (30 s at 62°C), extension (30 s at 72°C) Total RNA was extracted with a SNAP Total RNA Iso- and final extension (10 min at 72°C). PCR products were lation Kit (Invitrogen, NV Leek, The Netherlands) accord- electrophoresed on a 2% agarose gel and visualized with ing to the manufacturer’s instructions except that deoxy- ethidium bromide. To identify the PCR products as ribonuclease purchased from Nippon Gene (Osaka, human CCR3, cDNA from L1.2 cells which was trans- Japan) was used. After precipitation with ethanol, the fected with full-length human CCR3 DNA was ampli- first-strand cDNA was reverse transcribed as described fied simultaneously. previously.46 The second-strand DNA synthesis and hot- The procedures for competitive PCR of CCR3 mRNA start amplification were performed using a thermal cycler were basically the same as those described previously.46 (Thermal cycler MP, Takara, Kyoto, Japan). Briefly, cDNA and varying amounts of competitor cDNA Variations of human CCR3 and CCR4 H Kato et al 103 were used as templates, and quantification of the ampli- ness of type 1 T helper cells (Th1s) and Th2s. J Exp Med 1998; fied PCR products was performed by ELISA.47 The ampli- 187: 129–133. fied products were immobilized on carboxylated-surface 8 Sallusto F, Lenig D, Mackay CR, Lanzavecchia A. Flexible pro- plates. The covalently-bound single-strand DNAs were grams of chemokine receptor expression on human polarized T hybridized with digoxigenin-labeled oligonucleotide helper 1 and 2 lymphocytes. J Exp Med 1998; 187: 875–883. 9 Ward SG, Bacon K, Westwick J. Chemokines and T lympho- probes (5′-GTCACCCTTCCATTCTGGATCC-3′ for CCR3 ′ ′ cytes: more than attraction. Immunity 1998; 9: 1–11. and 5 -AAGTACCGTCGACGTCGGA-3 for competitor). 10 Sallusto F, Mackay CR, Lanzavecchia A. Selective expression of Hybridized probes were detected with peroxidase-conju- the eotaxin receptor CCR3 by human T helper cells. Science 1997; gated anti-digoxigenin antibodies, and signals were vis- 277: 2005–2007. ualized with tetramethyl benzidine. Because the ratio of 11 Imai T, Baba M, Nishimura M et al. The T cell-directed CC the target and competitor templates remained constant chemokine TARC is a highly specific biological ligand for CC during amplification, the quantity of competitor DNA in chemokine receptor 4. J Biol Chem 1997; 272: 15036–15042. the PCR templates which yielded an equal amount of the 12 Imai T, Chantry D, Raport CJ et al. Macrophage-derived chemo- two PCR products was considered to indicate the initial kine is a functional ligand for the CC chemokine receptor 4. J amount of the target gene. Biol Chem 1998; 273: 1764–1768. 13 Imai T, Yoshida T, Baba M et al. Molecular cloning of a novel T cell-directed CC chemokine expressed in thymus by signal Statistical analysis sequence trap using Epstein-Barr virus vector. J Biol Chem 1996; Fisher’s exact test was used to analyze the significance 271: 21514–21521. of the difference in the frequencies of CCR4 genotypes 14 Chang M, McNinch J, Elias C III et al. Molecular cloning and between patients and healthy individuals. Results from functional characterization of a novel CC chemokine, stimulated the chemotaxis assay were compared using Student’s T cell chemotactic protein (STCP-1) that specifically acts on acti- unpaired t-test. vated T lymphocytes. J Biol Chem 1997; 272: 25229–25237. 15 Andrew DP, Chang M, McNinch J et al. STCP-1 (MDC) CC chemokine acts specifically on chronically activated Th2 lym- Acknowledgements phocytes and is produced by monocytes on stimulation with Th2 cytokines IL-4 and IL-13. J Immunol 1998; 161: 5027–5038. The authors are indebted to Dr Takeshi Suzuki 16 Gerber BO, Zanni MP, Uguccioni M et al. Functional expression (Department of Allergy and Rheumatology, University of of the eotaxin receptor CCR3 in T lymphocytes co-localizing Tokyo) for allowing us to study the patient’s sample. with eosinophils. Curr Biol 1997; 7: 836–843. Thanks to Yoko Hatta and Michiko Shiota (Department 17 Rothenberg ME, MacLean JA, Pearlman E et al. Targeted disrup- of Human Genetics, University of Tokyo) for helpful dis- tion of the chemokine eotaxin partially reduces antigen-induced cussions, assistance and encouragement. tissue eosinophilia. J Exp Med 1997; 185: 785–790. Dr M Miyamasu is a Research Fellow of the Japan 18 Forssmann U, Uguccioni M, Loetscher P et al. Eotaxin-2, a novel Society for the Promotion of Science. CC chemokine that is selective for the chemokine receptor Sequence accession number: The nucleotide sequence CCR3, and acts like eotaxin on human eosinophil and basophil leukocytes. J Exp Med 1997; 185: 2171–2176. data reported in this paper have been submitted to the 19 Choe H, Farzan M, Sun Y et al. The ␤-chemokine receptors CCR3 DDBJ database and have been assigned the accession and CCR5 facilitate infection by primary HIV-1 isolates. Cell numbers AB023887-AB023892. 1996; 85: 1135–1148. 20 Doranz BJ, Rucker J, Yi Y et al. A dual-tropic primary HIV-1 Note added in proof isolate that uses fusin and the ␤-chemokine receptors CKR-5, While this paper was in press, the authors became aware CKR3, and CKR-2b as fusin cofactors. Cell 1996; 85: 1149–1158. of a recent paper which described variations of CCR3 21 He J, Chen Y, Farzan M et al. CCR3 and CCR5 are co-receptors gene among Caucasian and African American popu- HIV-1 infection of microglia. Nature 1997; 385: 645–649. lations.48 Four single nucleotide substitutions including 22 Littman DR. Chemokine receptors: keys to AIDS pathogenesis? Cell 1998; 93: 677–680. two nonsynonymous amino acid substitutions, R275Q 23 Cairns JS, D’Souza MP. Chemokines and HIV-1 second recep- and L351P, are described. None of the four variations tors: the therapeutic connection. Nature Med 1998; 4: 563–568. were detected in our subjects. 24 Hadida F, Vieillard V, Autran B et al. HIV-specific T cell cytotox- icity mediated by RANTES via the chemokine receptor CCR3. J Exp Med 1998; 188: 609–614. References 25 Pal R, Garzino-Demo A, Markham PD et al. Inhibition of HIV- ␤ 1 Baggiolini M, Dewald B, Moser B. Human chemokines: an 1 infection by the -chemokine MDC. Science 1997; 278: 695–698. update. Annu Rev Immunol 1997; 15: 675–705. 26 Charlton B, Lafferty KJ. The Th1/Th2 balance in autoimmunity. 2 Baggiolini M. Chemokines and leukocyte traffic. Nature 1998; Curr Opin Immunol 1995; 7: 793–798. 392: 565–568. 27 Kroemer G, Hirsch F, Gonzalez-Garcia A, Martinez C. Differen- 3 Combadiere C, Ahuja SK, Murphy PM. Cloning and functional tial involvement of Th1 and Th2 cytokines in autoimmune dis- expression of a human eosinophil CC chemokine receptor. J Biol eases. Autoimmunity 1996; 24: 25–33. Chem 1995; 270: 16491–16494. 28 Power CA, Meyer A, Nemeth K et al. Molecular cloning and 4 Ponath PD, Qin S, Post TW et al. Molecular cloning and charac- functional expression of a novel CC chemokine receptor cDNA terization of a human eotaxin receptor expressed selectively on from a human basophilic cell line. J Biol Chem 1995; 270: eosinophils. J Exp Med 1996; 183: 2437–2448. 19495–19500. 5 Yamada H, Hirai K, Miyamasu M et al. Eotaxin is a potent chem- 29 Samson M, Soularue P, Vassart G, Parmentier M. The genes otaxin for human basophils. Biochem Biophys Res Commun 1997; encoding the human CC-chemokine receptors CC-CKR1 to CC- 231: 365–368. CKR5 (CMKBR1-CMKBR5) are clustered in the p21.3-p24 region 6 Loetscher P, Uguccioni M, Bordoli L et al. CCR5 is characteristic of chromosome 3. Genomics 1996; 36: 522–526. of Th1 lymphocytes. Nature 1998; 391: 344–345. 30 Herz JM, Thomsen WJ, Yabrough GG. Molecular approaches to 7 Bonecchi R, Bianchi G, Bordignon PP et al. Differential receptors as targets for drug discovery. J Recept Signal Transduc- expression of chemokine receptors and chemotactic responsive- tion Res 1997; 17: 671–776. Variations of human CCR3 and CCR4 H Kato et al 104 31 Sheer A, Cotecchia S. Constitutively active G protein-coupled with a homology to the human C-C chemokine receptor, CCR- receptors: potential mechanisms of receptor activation. J Recept 4. Blood 1997; 89: 4448–4460. Signal Transduction Res 1997; 17: 57–73. 41 Bannai M, Tokunaga K, Lin L et al. Discrimination of human 32 Raziuddin S, al-Dalaan A, Bahabri S et al. Divergent cytokine HLA-DRB1 alleles by PCR-SSCP (single-strand conformation production profile in Behçet disease. Altered Th1/Th2 cell cyto- polymorphism) method. Eur J Immunogenet 1994; 21: 1–9. kine pattern. J Rheumatol 1998; 25: 329–333. 42 Hatta Y, Tsuchiya N, Matsushita M et al. Identification of the ␤ 33 Rucker J, Samson M, Doranz BJ et al. Regions in -chemokine gene variations in human CD22. Immunogenetics 1999; 49: 280– receptors CCR5 and CCR2b that determine HIV-1 cofactor speci- 286. ficity. Cell 1996; 87: 437–446. 43 Sato K, Kawasaki H, Nagayama H et al. CC chemokine recep- 34 Samson M, LaRosa G, Libert F et al. The second extracellular tors, CCR-1 and CCR-3, are potentially involved in APC func- loop of CCR5 is the major determinant of lignad specificity. J tion of human peripheral blood monocyte-derived dendritic Biol Chem 1997; 40: 24934–24941. cells. Blood 1999; 93: 34–42. 35 Atchison RE, Gosling J, Monteclaro FS et al. Multiple extracellu- 44 Mawhorter SD, Stephany DA, Ottesen EA, Nutman TB. Identi- lar elements of CCR5 and HIV-1 entry: dissociation from fication of surface molecules associated with physiologic acti- response to chemokines. Science 1996; 274: 1924–1926. vation of eosinophils. Application of whole-blood flow cytome- 36 Arnett FC, Edworthy S, Bloch DA et al. The American Rheuma- try to eosinophils. J Immunol 1996; 156: 4851–4858. tism Association 1987 revised criteria for the classification of 45 Miyamasu M, Hirai K, Takahashi Y et al. Chemotactic agonists rheumatoid arthritis. Arthritis Rheum 1988; 31: 315–324. induce cytokine generation in eosinophils. J Immunol 1995; 154: 37 Tan EM, Cohen AS, Fries JF et al. The 1982 revised criteria for 1339–1349. the classification of systemic lupus erythematosus. Arthritis Rheum 1982; 25: 1271–1277. 46 Miyamasu M, Misaki Y, Izumi S et al. Glucocorticoids inhibit 38 International Study Group for Behçet’s Disease. Criteria for chemokine generation by human eosinophils. J Allergy Clin diagnosis of Behçet’s disease. Lancet 1990; 335: 1078–1080. Immunol 1998; 101: 75–83. 39 Tokunaga K, Imanishi T, Takahashi K, Juji T. On the origin and 47 Taniguchi A, Kohsaka H, Carson DA. Competitive RT-PCR dispersal of East Asian populations as viewed from HLA haplo- ELISA: a rapid sensitive and non-radioactive method to quantit- types. In: Akazawa T, Szathmary EJ (eds). Prehistoric Mongoloid ate cytokine mRNA. J Immunol Methods 1994; 169: 101–109. Dispersals. Oxford University Press: Oxford, 1996, pp 187–197. 48 Zimmermann N, Bernstein JA, Rothenberg ME. Polymorphisms 40 Youn BS, Kim SH, Lyu MS et al. Molecular cloning and charac- in the human CC chemokine receptor-3 gene. Biochim Biophys terization of a cDNA, CHEMR1, encoding a chemokine receptor Acta 1998; 1442: 170–176.