Mutation Plt and EBI-1 Ligand Chemokine Genes in the Of

Home , Ligand (biochemistry)

Gene Duplications at the Chemokine Locus on Mouse Chromosome 4: Multiple Strain-Speciﬁc Haplotypes and the Deletion of Secondary Lymphoid-Organ Chemokine and EBI-1 Ligand Chemokine Genes in the plt Mutation1

Hideki Nakano and Michael D. Gunn2

The murine paucity of lymph node T cell (plt) mutation leads to abnormalities in leukocyte migration and immune response. The causative defect is thought to be a loss of secondary lymphoid-organ chemokine (SLC) expression in lymphoid tissues. We now ﬁnd that the plt defect is due to the loss of both SLC and EBI-1 ligand chemokine (ELC) expression in secondary lymphoid organs. In an examination of the plt locus, we ﬁnd that commonly used inbred mouse strains demonstrate at least three different haplotypes. Polymorphism at this locus is due to duplications of at least four genes, three of them encoding chemokines. At least two cutaneous T cell-attracting chemokine (CTACK), three SLC, and four ELC genes or pseudogenes are present in some haplotypes. All haplotypes share a duplication that includes two SLC genes, which demonstrate different expression patterns, a single functional ELC gene, and an ELC pseudogene. The plt mutation represents a deletion that includes the SLC gene expressed in secondary lymphoid organs and the single functional ELC gene, leaving only an SLC gene that is expressed in lymphatic endothelium and an ELC pseudogene. This lack of CCR7 ligands in the secondary lymphoid organs of plt mice provides a basis for their severe abnormalities in leukocyte migration and immune response. The Journal of Immunology, 2001, 166: 361–369.

t is now recognized that chemokines mediate the trafficking The plt mutation arose spontaneously in a colony of DDD/1 of leukocytes to and within lymphoid organs and thereby mice at the University of Tokyo (11). Because this mutation was I participate in the development of an immune response (1–3). not initially recognized, the true parental line was lost, but a con- Several chemokines are constitutively expressed in lymphoid or- genic strain, DDD/1-Mtv2, still exists (12). In a comparison of gans, and predictions have been made concerning their function DDD/1 and DDD/1-Mtv2 mice, it was found that DDD/1 mice based on their expression patterns and in vitro activities (4, 5). In display a marked paucity of T cells in peripheral LNs (13). Further a few cases these predictions have been confirmed through the use analysis revealed that this abnormality was due to the development of in vivo models (6–8). A prominent member of the constitutive of a recessive mutation (now designated plt) in the DDD/1 inbred chemokine family is secondary lymphoid-tissue chemokine line. DDD/1-plt mice demonstrate a 5- to 10-fold decrease in the (SLC),3 which is believed to mediate the migration of T cell and number of naive T cells present in peripheral LNs and a defect in dendritic cell (DC) subsets into lymphoid organs. Much of our naive T cell homing to secondary lymphoid organs (14, 15). The current understanding of SLC function originated from studies of plt locus was mapped to mouse chromosome 4 in a region of mice homozygous for the paucity of lymph node (LN) T cell ( plt) conserved synteny to human chromosome 9p13. Three human che- mutation (8–10). plt mice do not express SLC in secondary lym- mokine genes map to 9p13: SLC (CCL21), EBI-1 ligand chemo- phoid organs and demonstrate severe abnormalities in leukocyte kine (ELC; CCL19), and cutaneous T cell-attracting chemokine migration and immune response (8). However, significant ques- (CTACK; CCL27), although this was not known at the time plt tions remain concerning the validity of plt mice as a model of SLC mice were identified (16–18). dysfunction, because the molecular basis of the plt mutation has SLC was identified by several groups as a novel chemokine not been determined. present in the National Center for Biologic Information expressed sequence tag (EST) database (17, 19–21). Three characteristics of SLC suggested that it may be the chemokine responsible for me- Division of Cardiology, Department of Medicine, Duke University Medical Center, diating the entry of T cells into secondary lymphoid organs. First, Durham, NC 27710 it is expressed in the high endothelial venules (HEV) of LNs and Received for publication July 11, 2000. Accepted for publication October 6, 2000. Peyer’s patches and within T cell zones of LNs, spleen, and Pey- The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance er’s patches (22). SLC is also expressed in thymic medulla and in with 18 U.S.C. Section 1734 solely to indicate this fact. the lymphatic endothelium of multiple tissues (9, 19, 22). Second, 1 This work was supported by an Established Investigator Award (0040030N) from SLC is a highly efficacious chemoattractant for naive T cells (22, the American Heart Association and a training grant from the Cancer Research In- 23). Third, SLC stimulates the integrin-mediated adhesion of naive stitute (to H.N.). T cells to ICAM-1 and MadCAM-1 (24–26). The chemokine most 2 Address correspondence and reprint requests to Dr. Michael D. Gunn, Box 3547, Duke University Medical Center, Durham, NC 27710. E-mail address: michael. similar to SLC is ELC (16). SLC and ELC share the same receptor, [email protected] CCR7, and their genes are separated by Ͻ100 kb in humans (16, 3 Abbreviations used in this paper: SLC, secondary lymphoid-tissue chemokine; DC, 27, 28). ELC is expressed by DC and stromal cells within LNs and dendritic cell; BAC, bacterial artificial chromosome; CTACK, cutaneous T cell-at- spleen (29). Based on its expression pattern and activities, ELC is tracting chemokine; ELC, EBI-1 ligand chemokine; HEV, high endothelial venules; LN, lymph node; RFLP, restriction fragment length polymorphism; EST, expressed believed to act within lymphoid organs to mediate naive T cell-DC sequence tag; Cklc4, chemokine locus chromosome 4. interactions (1). The most recently identified chemokine on human

Copyright © 2001 by The American Association of Immunologists 0022-1767/01/$02.00 362 VARIATIONS IN CHEMOKINE GENE NUMBER ON MOUSE CHROMOSOME 4 chromosome 9 is CTACK, which is expressed predominately in Southern blot analyses ϩ skin and is chemotactic for CLA memory T cells (30). DNA was prepared from murine tissue by standard procedures or was Once the probable function of SLC was recognized, its potential obtained from The Jackson Laboratory. For Southern blot analysis, 10 ␮g contribution to the plt mutation was examined. It was found that of genomic DNA or a normalized amount of P1 or BAC plasmid was the plt phenotype and the SLC gene map to the same genetic locus digested with restriction enzymes according to manufacturer’s instructions (Roche, Indianapolis, IN), separated on 1% agarose gels at 1 V/cm for on mouse chromosome 4. SLC mRNA is not expressed in the ϩ 10–18 h, and transferred to nylon membranes (Hybond-N , Amersham, secondary lymphoid organs of plt mice despite the fact that an Arlington Heights, IL) by alkaline blotting (33). Blots were hybridized intact SLC gene is present in plt DNA (8). The expression of ELC with 32P-labeled probe random primed from a BglII-NsiI fragment of mRNA is reduced in plt mice, but is clearly present. Subsequent Scya21a (probe A), a PvuII-XbaI fragment of Scya19 (probe B), a CTACK studies have demonstrated that rolling naive T cells do not attach EST, or an IL-11R␣ EST in dextran sulfate hybridization mixture overnight at 68°C. Blots were washed in 0.1ϫ SSC/0.1% SDS at 68°C before to HEV in the LNs or Peyer’s patches of plt mice (9, 10). In LN autoradiography. this defect can be partially reversed by the s.c. injection of SLC (9). plt mice also demonstrate abnormalities in DC localization and SLC expression studies migration (8). The number of DCs in the LN and splenic white For in situ hybridizations, paraffin sections (5 ␮m) from BALB/c and pulp of plt mice is markedly reduced, as is the number of DCs that BALB/c-plt mice were deparaffinized, fixed in 4% paraformaldehyde, and migrate to these areas after inflammatory stimuli. Similar defects treated with proteinase K. After washing in 0.5ϫ SSC, the sections were in DC migration are seen in mice after treatment with anti-SLC covered with hybridization solution, prehybridized for 1–3 h at 55°C, and 35 Abs (31). These studies strongly suggest that SLC is required for hybridized overnight with sense or antisense S-labeled riboprobe transcribed from the mouse SLC cDNA. After hybridization, sections were the migration of naive T cells and activated DC into the thymus- washed at high stringency, dehydrated, dipped in photographic emulsion dependent areas of secondary lymphoid organs. Support for this NTB2 (Eastman Kodak, Rochester, NY), stored at 4°C for 4 wk, developed, view has come from studies of CCR7-deficient mice, which dis- and counterstained with hematoxylin and eosin. For RT-PCR-restriction play a constellation of leukocyte trafficking abnormalities that are fragment length polymorphism (RFLP) analysis, total RNA was prepared similar, but not identical, to those seen in plt mice (7). from mouse LN and spleen using TRIzol reagent (Life Technologies, Gaithersburg, MD), reverse transcribed using a First Strand Synthesis kit To determine the basis of the plt phenotype, we initiated studies (Roche), and amplified with ELC-specific primers (AGGAGGACATCT to examine the DNA abnormality in plt mice. These studies were GAGCGATTCC and TGGTGAACACAACAGCAGGCAC). A portion of complicated by the finding that marked genetic heterogeneity ex- the RT-PCR product was digested with NcoI, and digested and undigested ists at this locus in wild-type mice. At least two CTACK, three samples were resolved by agarose gel electrophoresis. For immunohistochemistry, frozen tissue sections were acetone fixed, SLC, and four ELC genes or pseudogenes are present in some blocked with PBS/5% normal donkey serum, and incubated with goat anti- inbred strains of mice. This locus includes the previously described murine SLC (R&D Systems, Minneapolis, MN), biotin-conjugated donkey duplication of Il11ra genes. At least three wild-type haplotypes of anti-goat IgG (Jackson ImmunoResearch Laboratories, West Grove, PA), this locus are found in commonly used inbred mice. While these HRP-avidin-biotin conjugate (Vector, Burlingame, CA), developed with studies were in progress, another group demonstrated that wild- Vector VIP substrate (Vector), and counterstained with methyl green. type mice express two forms of SLC and that one of these is deleted in plt mice (32). Our results confirm and extend those Results findings. The plt mutation represents a genomic deletion that leads Identification of multiple SLC and ELC genes in 129 mice to a unique combination of SLC and ELC genes in mice, elimi- To provide a basis for analyzing the plt mutation, the wild-type nating the SLC gene expressed in secondary lymphoid organs and SLC/ELC locus was examined. Five P1 genomic clones (desig- the sole functional ELC gene. nated A–E) and two BAC clones were identified in murine 129/Ola or 129/SvJ genomic libraries. All clones contained both SLC and Materials and Methods ELC by Southern blot analysis. Restriction mapping of the P1 clones was performed, but the results were not consistent with a Animals and reagents single location for either SLC or ELC. Some Southern blots also BALB/c-plt mice were produced by backcrossing plt mice 10 generations into suggested that multiple SLC and ELC genes are present in the a BALB/c genetic background. BALB/cJ control mice were obtained from murine genome (data not shown). To examine this possibility, The Jackson Laboratory (Bar Harbor, ME). All mice were maintained under specific pathogen-free conditions in accordance with institutional guidelines. SLC- and ELC-containing fragments from all P1 clones were sub- Frozen tissues from DDD-Mtv2 and DDD-plt were obtained from Kazutosi cloned and sequenced. Sequence analysis of these fragments dem- Sayama (Shizuoka University, Shizuoka, Japan) and used for DNA prepara- onstrated three distinct SLC genes (designated Scya21a–c; Fig. tion. P1 clones A–D (controls 15738–15741) were obtained from Incyte 1A). The sequences of these genes are highly conserved, although Genomics (St. Louis, MO) by PCR screening a 129/Ola library with SLC- specific primers (CCATATGAGTGATGGAGGGGGACAGG and CCTC- they each have scattered small deletions relative to the consensus GAGCTATCCTCTTGAGGGCTGTG). Bacterial artificial chromosomes sequence. The exon sequences of Scya21b and Scya21c are iden- (BACs) from the RPCI-23 library were identified by screening the BAC end- tical. They differ from the exon sequence of Scya21a by several sequence database (http://www.tigr.org/tdb/humgen/bac_end_search/bac_end- single-nucleotide changes. One of these changes (C to T at position _search.html) with SLC, ELC, CTACK, and I1llra sequences and were ob- 251 of SLC mRNA) leads to an amino acid change at position 65 tained from Children’s Hospital Oakland Research Institute (Oakland, CA). P1 clone E, BAC1, and BAC2 were gifts from Jason Cyster. They were originally of the SLC protein (serine in Scya21a to leucine in Scya21b and obtained from Incyte Genomics by screening with primers specific for the Scya21c). In general, the sequences of the Scya21a and Scya21b Scya19 gene. genes correspond to the SLC-Ser (C6kine-Ser) and SLC-Leu (C6kine-leu) genes described by Vassileva et al. (32). Subcloning and sequencing Sequence analysis demonstrated four distinct ELC genes (Fig. P1 clones were digested with HindIII or Asp718, and the resultant frag- 1B). The exon sequences of all four differ from each other by ments were randomly ligated into HindIII- or Asp718-digested pBluescript several single-nucleotide changes. ESTs corresponding to the pre- (Stratagene, La Jolla, CA). Colonies were screened by colony lifts onto dicted transcripts of all ELC genes are found in the National Cen- nylon filters. After alkaline lysis, filters were probed with randomly primed SLC or ELC cDNA. Clones containing SLC or ELC were picked from the ter for Biologic Information EST database, demonstrating that all original plates and prepared by standard procedures. Sequencing was per- four are transcribed. Importantly, the predicted transcript of only formed using dye terminator technology. one ELC gene has a methionine codon (ATG) in a position that The Journal of Immunology 363

FIGURE 1. Structure of three murine SLC and four murine ELC genes. A, Structure and selected restriction sites of the SLC gene consensus sequence in comparison with speciﬁc SLC genes. Ⅺ, Exons; f, coding sequence; u, di- or trinucleotide repeats. Vertical lines represent the presence of the indicated restriction site. Breaks in horizontal lines represent deletions relative to the consensus sequence. A single amino acid difference (Ser vs Leu) in the SLC coding sequence is indicated above the arrowhead. B, Structure and selected restriction sites of ELC gene consensus sequence in comparison with speciﬁc ELC genes. Diagram features are explained in A. C, Sequence surrounding consensus initiation codon (shown in bold) for each ELC gene. Indicated restriction enzymes are: Bam, BamHI; Bgl2, BglII; E, EcoRI, H, HindIII; K, KpnI; Nco, NcoI; Nhe, NheI; Nsi, NsiI; Pvu, PvuII; Sac, SacII; Sma, SmaI, Sw, SwaI; and Xba, XbaI. Gene maps are based on DNA sequences of the entire region shown. The full sequences have been deposited in GenBank (accession nos. AF307985—AF307991).

would lead to translation of ELC protein (Fig. 1C). This sequence Chemokine gene duplications in various inbred mouse strains thus corresponds to the true Scya19 gene. The three other ELC The presence of SLC and ELC genes in various inbred mouse genes appear to be pseudogenes, transcribed but not translated, and strains was determined using RFLPs that were identiﬁed by se- are therefore designated Scya19-ps1, Scya19-ps2, and Scya19-ps3. quence analysis. When P1 clones are examined by Southern blot- The predicted Scya19 transcript corresponds to the published ting, three fragments are found that correspond to the predicted mELC cDNA sequence with the exception of three single-base sizes of the three SLC genes (Fig. 2A, SLC probe). When mouse changes in the 3Ј untranslated region (29). The bases at these po- sitions in the published ELC cDNA sequence correspond to those genomic DNA is digested and hybridized similarly, three distinct found in all ELC pseudogenes, suggesting that the published se- patterns of hybridization are found (Fig. 2B, SLC probe). In the quence in this region is a composite of multiple transcripts that ﬁrst pattern fragments corresponding to all three SLC genes are were thought to arise from a single gene. present. In the second pattern only fragments corresponding to

FIGURE 2. Southern blot analysis of P1 genomic clones, murine genomic DNA, and BAC genomic clones. A, P1 129/Sv genomic clones were digested with HindIII and NsiI (for SLC); BamHI, PvuII, and XbaI (for ELC); or EcoRI (for CTACK); separated by gel; transferred to nylon membranes; and hybridized with probes speciﬁc for SLC (probe A in Fig. 1), ELC (probe B in Fig. 1), or CTACK (cDNA). The genes that correspond to the hybridizing fragments are indicated on the left. The sizes of fragments are indicated on the far right. P1 clone D demonstrates SLC and ELC hybridizing patterns identical with those of clone C and is not shown. B, Genomic DNA (10 ␮g) from the indicated strains of inbred mice was digested, transferred, and probed as described above. C, BAC 129/Sv genomic clones were digested and probed as described above. 364 VARIATIONS IN CHEMOKINE GENE NUMBER ON MOUSE CHROMOSOME 4

Scya21a and Scya21b are present. In the third pattern a fragment cus chromosome 4, or Cklc4. The above results demonstrate that at that corresponds to Scya21b is present along with a second frag- least three distinct haplotypes exist at the Cklc4 locus in commonly ment of larger size. The evidence presented below demonstrates used inbred mouse strains. We have designated these haplotypes that this larger fragment (identified in Southern blots as Scya21aЈ) Cklc4a, Cklc4b, and Cklc4c. The features of each haplotype are is a RFLP of Scya21a. summarized in Table I. When P1 clones are digested and hybridized with an ELC genomic probe, fragments of four sizes are found, corresponding Organization of the mouse Cklc4 to the four ELC genes (Fig. 2A, ELC probe). When mouse All P1 clones examined were derived from 129 mice and therefore genomic DNA is examined, two patterns are seen (Fig. 2B, ELC represent the Cklc4c haplotype. To determine the position of these probe). Scya19 and Scya19-ps1 are present in all strains examined, clones relative to each other, two 129/Sv BAC clones obtained in while Scya19-ps2 and Scya19-ps3 are found only in 129/SvJ, a screen for Scya19 were examined for the presence of other genes. C57BL/6, and PL/J mice. Thus, most strains of mice appear to By Southern blot analysis, BAC1 contains Scya21a and Scya19, possess two SLC genes, one functional ELC gene, and one ELC but no other SLC, ELC, CTACK, or I1llra genes (Fig. 2C and data pseudogene. 129/Sv, C57BL/6, and PL/J mice possess at least one not shown). BAC2 contains SLC genes Scya21a and Scya21c, additional SLC gene and two additional ELC pseudogenes. ELC genes Scya19 and Scya19-ps3, the Scya27b CTACK gene, It is known that several inbred strains of mice possess two cop- and the I1llra2 gene (Fig. 2C and data not shown). ies of the I1llra gene (I1llra1 and I1llra2) (34–36). Because the 3Ј To better define the arrangement of genes at the Cklc4c locus, end of the I1llra gene overlaps the 3Ј end of the CTACK gene (18), BAC clones from the C57BL/6 RPCI-23 library were examined. this finding suggested that the I1llra and SLC/ELC gene duplica- The database of RPCI-23 end sequences was searched for the pres- tions may be related events and that CTACK and I1llra genes may ence of Scya19, Scya21, Scya27, and I1llra sequences. Eight be present at the SLC/ELC locus. Southern blot analysis of P1 BACs were identified as positive for Cklc4 sequences and were clones revealed that a CTACK gene is present on P1 clone C (Fig. subjected to Southern blot analysis to determine the presence of 2A, CTACK probe). This same clone contains the Scya21c and specific Scya19, Scya21, and I1llra genes. All P1 and BAC clones Scya19-ps3 genes. This clone also contains the I1llra2 gene (data were also examined for the presence the D4 Mit237 STS marker. not shown), suggesting that both murine I1llra genes are associ- During the mapping of the plt mutation, it was determined that an ated with CTACK genes. inability to amplify D4 Mit237 from genomic DNA is closely To determine the distribution of CTACK genes in various linked to the plt phenotype. PCR analysis revealed that the D4 mouse strains, we examined EcoRI digests of mouse genomic Mit237 marker is present on four BACs but none of the P1 clones. DNA. An EcoRI polymorphism has been demonstrated immedi- The results of these analyses are summarized in Table II. The ately downstream of the I1llra1 and I1llra2 genes in the region localization of genes on P1 and BAC clones allows the construc- where the CTACK gene is found (34). Southern blot analysis re- tion of a contig of the Cklc4c locus. This contig includes the D4 veals that two CTACK-hybridizing EcoRI fragments are present in Mit 237 marker, the three SLC genes, the four ELC genes, the most strains of mice (Fig. 2B, CTACK probe). A single CTACK- I1llra2 gene, and the Scya27b gene. The I1llra1 and Scya27a hybridizing EcoRI fragment is seen in C57BR, C57L, SJL, and genes are known to be located near this region (35), but were not SM/J mice (Fig. 2B and data not shown). Not surprisingly, the present on any the clones examined. The contig presented in Table strains of mice found to have two CTACK genes correspond ex- II suggests the order of genes in the Cklc4c haplotype. Determining actly to those in which the I1llra2 duplication has been demon- the organization of haplotypes Cklc4a and Cklc4b will require the strated (34). We have designated the CTACK gene that is associ- examination of genomic DNA or clones derived from mice that ated with the I1llra1 gene (the 3.4-kb EcoRI fragment) Scya27a. harbor these haplotypes. The CTACK gene that is associated with the I1llra2 gene (the 4-kb EcoRI fragment) is designated Scya27b. The strains of mice that Characterization of the SLC/ELC locus in plt mice possess a single CTACK gene correspond to those that demon- The Scya21a gene is not found in genomic digests of BALB/c-plt strate the Scya21aЈ form of the SLC-Ser gene. DNA (32). This and the loss of the D4 Mit237 STS marker suggest The distributions of SLC, ELC, CTACK, and I1llra genes in the that the plt mutation involves a genomic deletion. To identify wild-type mouse strains we examined fall into three distinct pat- genes involved in this deletion, Southern blots of genomic DNA terns. Because we find variations in the numbers of multiple genes from DDD-Mtv2 mice and BALB/c-plt mice were compared. present at this site, we have designated this region chemokine lo- DDD-Mtv2 mice are believed to be representative of the DDD/1

Table I. Genes present in Cklc4 haplotypes

Genes Present Locus Representative Designation SLC ELC CTACK IL11Ra Mouse Strains

Cklc4a Scya21a Scya19 Scya27a Il11ra1 C57BR/cdJ Scya21b Scya19-ps1 C57L/J DBA/2J Cklc4b Scya21a Scya19 Scya27a Il11ra1 A/J Scya21b Scya19-ps1 Scya27b Il11ra2 BALB/cBy C3H/HeJ Cklc4c Scya21a Scya19 Scya27a Il11ra1 129/SvJ Scya21b Scya19-ps1 Scya27b Il11ra2 C57BL/6J Scya21c Scya19-ps2 PL/J Scya19-ps3 Cklc4plt Scya21b Scya19-ps1 Scya27a Il11ra1 DDD/1 The Journal of Immunology 365

Table II. Genes present on mouse genomic clones

P1 End Sequence D4Mit237 Scya21a Scya19 Il11ra2 Scya19-ps3 Scya21c Scya19-ps2 Scya19-ps1 Scya21b

E —XX———— —— C———XXX——— D————XX——— A—————XX—— B ——————— X X BAC

BAC1 ϩ XX———— —— BAC2 — X X X X X — — — 432P21 Scya21* ——————— —— 333H19 Il11ra ϩ X X XES — — — — — 346C23 Il11ra ϩ X X XES — — — — — 382I12 Scya21* ϩ XXX X X — — — 430D22 Scya21c —XXXXES— —— 355O11 Scya21b ————— X X X ES 322A19 Scya21c —————ESX X —

ϩ, Clone positive for marker by PCR; X, clone positive for gene by Southern blotting; ES, clone end sequence positive for gene; *, clone end sequences similar to but not identical with each other and to sequence 5Ј of the Scya21 genes. strain on which the plt mutation arose. BALB/c-plt mice have re- mice, suggesting that the plt deletion does not include the CTACK tained the genomic organization of Cklc4 that is seen in DDD-plt or I1llra genes. Overall, the pattern of intact and deleted genes in mice (data not shown). Southern blot analysis reveals that the plt mice suggests that the proximal end of the plt deletion is lo- Scya21aЈ and Scya21b genes are present in DDD-Mtv2 mice in a cated between the Scya19 gene and the next upstream SLC, ELC, pattern corresponding to the Cklc4a haplotype (Fig. 3A). In BALB/ CTACK, or I1llra gene. The identity of this next upstream gene is c-plt mice, the fragment corresponding to Scya21aЈ is absent, leav- not known in the Cklc4a haplotype. Thus, the plt deletion includes ing only the Scya21b gene. Strains of mice demonstrated to pos- the SLC-Ser gene expressed in secondary lymphoid organs, the sess the Scya21aЈ gene (DDD-Mtv2, DBA/2) express SLC in lone functional ELC gene, and the D4 mit237 marker. secondary lymphoid organs, while strains of mice that have deleted this gene (BALB/c-plt, DDD/1) do not (9) (data not shown). This suggests that the Scya21aЈ fragment is a variant of the Scya21a Expression pattern of Scya21b gene, and that this gene is expressed in secondary lymphoid By Northern blot analysis, SLC is expressed in nonlymphoid or- organs. gans of plt mice, predominately heart, lung, and gastrointestinal The finding that up to four transcribed ELC genes exist in wild- tract (32). To determine specifically where SLC is expressed in plt type mice calls into question the conclusion that a functional ELC mice, in situ hybridization was performed on multiple tissues from gene is expressed in plt mice. To determine the effect of the plt BALB/c and BALB/c-plt mice. Consistent with published data, mutation on ELC genes, Southern blots of BamHI/XbaI/PvuII-di- expression of SLC is not seen in the spleen, LN, or Peyer’s patches gested genomic DNA from 129/SvJ, DDD-Mtv2, and BALB/c-plt of plt mice (data not shown). SLC hybridization signal is observed mice were compared (Fig. 3B). Although both Scya19 and Scya19- on lymphatic endothelial cells in the intestine, heart, lung, liver, ps1 are found in DDD-Mtv2 mice, only the Scya19-ps1 ELC pseu- and kidney of plt mice (Fig. 4 and data not shown). However, dogene is present in plt mice. To confirm the absence of an Scya19 compared with wild-type mice, the hybridization signal seen in gene in plt mice, BALB/c-plt DNA digested with NcoI and XbaI tissues from plt mice is less intense and less uniformly distributed was examined. The NcoI site present at the ELC initiation codon within lymphatic vessels. is unique to the Scya19 gene (Fig. 1C). A comparison with 129/SvJ Similar results are seen when SLC protein is localized by im- DNA confirms that the Scya19 gene is absent in BALB/c-plt mice munohistochemistry. However, the distributions of SLC protein (Fig. 3C). and SLC mRNA differ somewhat. In general, SLC protein is dis- To determine whether an ELC transcript that contains the tributed more broadly than SLC mRNA, suggesting that SLC pro- Scya19 NcoI site is present in plt mice, cDNAs derived from tein diffuses to localize in regions surrounding SLC-producing BALB/c and BALB/c-plt LNs and spleen were subjected to PCR- cells. Non-SLC-producing areas that accumulate SLC protein in- RFLP analysis. When mRNA from BALB/c mice is reverse tran- clude the endothelium of splenic arterioles (Fig. 5C, inset), the scribed and amplified with primers specific for all ELC transcripts, portal triads of liver (Fig. 5E), and the peribronchial regions of the majority of PCR products (bases 63–390 of mELC cDNA) can lung (Fig. 5G). Consistent with previously reported in situ hybrid- be cleaved at an internal NcoI site (Fig. 3D). In contrast, none of ization results, SLC protein is not detected in the spleen, LNs, or the PCR products from plt-BAB/c mice is cleaved with NcoI. This Peyer’s patches of plt mice (Fig. 5, B and D, and data not shown). finding supports the conclusion that plt mice do not express an In the lungs and liver of plt mice, SLC protein can be detected on ELC transcript that contains an initiation codon in the proper lymphatic endothelial cells (Fig. 5, F and H, insets). SLC protein location. accumulation is faintly seen in the tissues surrounding lymphatics To determine whether the CTACK and I1llra genes are affected in plt mice, but is markedly reduced compared with that in wild- by the plt mutation, these genes were analyzed by Southern blot type mice. These findings suggest that the Scya21a and Scya21b analysis as described above. DDD-Mtv2 genomic DNA contains a genes are both transcribed in lymphatic endothelial cells and that single copy of the CTACK gene (Fig. 3E) and a single copy of the total SLC expression in these cells is reduced by deletion of the I1llra gene (Fig. 3F). The pattern of I1llra-hybridizing fragments Scya21a gene. Alternatively, it is possible that only the native corresponds to the I1llra1 gene (34). A similar pattern is seen in plt Scya21b gene is expressed in lymphatic endothelium, but that a 366 VARIATIONS IN CHEMOKINE GENE NUMBER ON MOUSE CHROMOSOME 4

FIGURE 4. Analysis of SLC mRNA expression in wild-type and plt mice by in situ hybridization. All photomicrographs demonstrate binding of 35S-labeled SLC antisense riboprobe to wild-type (A and C)orplt (B and D) tissues. The hybridization signal is shown as black dots. A, Small intestine of wild-type mice. B, Small intestine of plt mice. C, Epicardial surface of wild-type mice. D, Epicardial surface of plt mice. No signal was seen with SLC sense probe (not shown). Original magniﬁcation, ϫ80.

phatic endothelium, and it is deleted in the plt mutation, suggesting its role in lymphocyte trafficking. Based on its expression pattern in plt mice, Scya21b appears to be expressed only at low levels in lymphatic endothelium. However, this finding should be viewed with caution because the plt mutation may have disrupted DNA segments that regulate Scya21b expression. Currently there is no evidence that the Scya21b gene product, SLC-Leu, differs in activity from the Scya21a gene product, SLC-Ser. The third SLC FIGURE 3. Southern blot and PCR-RFLP analysis of genes present in gene, Scya21c, is present in few strains of mice. Because the pre- plt mice. A, Complement of SLC genes present in DDD-Mtv2 and BALB/ dicted transcript of Scya21c is identical with that of Scya21b,itis c-plt mice. The restriction enzymes were HindIII and NsiI. The probe was probe A. DBA/2, BALB/c, and 129/Sv genomic digests demonstrate, re- not clear whether this gene is functional. SLC mRNA levels are a b c spectively, the Cklc4a, Cklc4b, and Cklc4c fragment patterns. B, Comple- similar in the lymphoid organs of Cklc4 , Cklc4 , and Cklc4 mice ment of ELC genes present in DDD-Mtv2 and BALB/c-plt mice. The re- (data not shown). Determining the full expression pattern and striction enzymes were BamHI, PvuII, and XbaI. The probe was probe B. function of all SLC genes will require a detailed analysis of SLC C, Further demonstration of ELC genes in DDD-Mtv2 and BALB/c-plt transcripts and phenotypes in mice in which various combinations mice. The restriction enzymes were NcoI and XbaI. The probe was probe of SLC genes are present. B. The Scya19-ps1 and Scya19-ps2 genes are not resolved in this digest. D, One functional ELC gene and three ELC pseudogenes are PCR-RFLP of ELC cDNA from BALB/c and BALB/c-plt lymphoid or- present in the mouse genome (Figs. 1B and 2B). Scya19 and gans. cDNA was amplified with ELC-specific primers and digested with Scya19-ps1 are present in all mouse strains examined, while NcoI before agarose gel electrophoresis. E, CTACK genes in DDD-Mtv2 Scya19-ps2 and Scya19-ps3 are present only in those strains that and BALB/c-plt mice. The restriction enzyme was EcoRI. The probe was CTACK cDNA. F, I1llra genes in DDD-Mtv2 and BALB/c-plt mice. The have an Scya21c gene. By sequence analysis, Scya19 represents restriction enzyme was BamHI. The probe was I1llra cDNA. The I1llra the only functional ELC gene in 129/Sv mice. Because all ELC probe hybridizes to multiple I1llra1 and I1llra2 gene fragments. Identifi- genes are transcribed, hybridization-based methods of measuring cation of fragments is based on Ref. 34. Probes A and B are diagramed in ELC mRNA levels, such as Northern blotting and in situ hybrid- Fig. 1. ization, will not be reliable measures of ELC function. It was the presence of ELC pseudogene transcripts that led us to falsely con- clude the ELC function is preserved in plt mice (8). It has been previously demonstrated that two distinct I1llra cis-acting regulatory element required for optimum Scya21b ex- genes are present in the mouse genome (34–36). I1llra1 represents pression, such as a locus control region, is deleted in the plt the predominant form, is present in all mouse strains examined, mutation. and is expressed in all tissues. I1llra2 is present in about half the mouse strains examined and is expressed in testes, thymus, and Discussion LNs. Targeted deletion of the I1llra1 gene leads to female infer- We investigated the genetic abnormality that occurs in mice with tility due to defective uterine decidualization, but has no demon- the plt mutation. In the process we found that a series of gene strated effect on hemopoiesis (37–39). No specific function has duplications occurs at the plt locus in wild-type mice. At least four been determined for I1llra2. genes are affected. We found that at least three SLC genes are Our data suggest that at least two CTACK genes are present in present in the mouse genome (Figs. 1A and 2B). Our findings sug- the mouse genome. Although we have not sequenced these re- gest Scya21a should be viewed as the prototypical SLC gene. It is gions, the two CTACK-hybridizing EcoRI genomic fragments we expressed in lymphoid organs, it is probably expressed in lym- observed (Fig. 2B) correspond to fragments predicted to exist at The Journal of Immunology 367

Second, it is possible that more than three cluster types exist at the Cklc4 locus. We have observed Southern blot hybridization patterns in some mouse strains that do not correspond to any of the cluster types we have defined (data not shown). Whether this is due to incomplete digestions, RFLPs, or additional cluster types is not yet known. Third, the actual number of SLC, ELC, CTACK, or I1llra genes present at the Cklc4 locus may be greater than we have demonstrated here. Evidence suggests that up to six I1llra2 genes are present in some strains of mice (34). There is also evidence that a fourth SLC gene may exist in the Cklc4c haplotype. Some ESTs derived from C57BL/6 mice represent SLC transcripts that do not correspond to any of the SLC genes we have sequenced. Also, the end sequence of BAC 432P21 matches SLC sequence upstream of the transcription initiation site but does correspond exactly to Scya21a–c. Because this BAC continues further upstream, we could not determine whether this sequence represents a true SLC gene. The functional consequences of the gene duplications we de- scribe are unknown. At least one strain-specific trait, leukocyte infiltration into the uterus in response to estrogen, has been mapped to the vicinity of the Cklc4 locus (46). It is possible that other strain-specific variations in lymphoid organ anatomy or immune response are due in part to heterogeneity at this locus in mice. Such variability in gene number does not occur at this locus in humans. The human genomic sequence that corresponds to the Cklc4 locus (GenBank accession no. AC026658) demonstrates no duplication of SLC, ELC, CTACK, or I1llra genes. Most importantly for our purposes, characterizing the Cklc4 locus has allowed a greater understanding of the plt mutation. This FIGURE 5. Analysis of SLC protein expression in wild-type and plt mutation arose on a Cklc4a background that contains two SLC and mice by immunohistochemistry. All photomicrographs demonstrate immu- two ELC genes, one CTACK gene, and one I1llra gene. In plt nohistochemical staining of frozen tissue sections with anti-SLC Ab, mice, the SLC gene that is expressed in secondary lymphoid or- shown as purple (A–D) or brown (E–H). In LNs (A and B) and spleen (C gans and the lone functional ELC gene are deleted, leaving SLC- and D), SLC staining is seen in wild-type, but not plt, tissues. In wild-type, but not plt, spleen, strong SLC staining is seen on central arterioles (C and Leu as the only known CCR7 ligand in these animals. It remains D, insets). In wild-type mice diffuse SLC staining is seen in the portal triads possible that a gene other than SLC or ELC has been deleted in plt of liver (E) and the peribronchial regions of lung (G). In plt mice, such mice. However, preliminary studies suggest that all nonchemokine staining is limited mainly to the body of lymphatic endothelial cells (F and genes and ESTs in this region of the human genome are intact in G, insets). plt mice. In light of our findings, some reassessment of the mechanisms that lead to the plt phenotype is required. First, the lack of ELC in the 3Ј end of the I1llra1 and I1llra2 genes, the known locations of plt mice raises the possibility that this protein contributes to the the CTACK gene in humans and mice. We have not examined the extravasation of T cells across HEV. Although ELC is not ex- relative transcription levels of the Scya27 genes. pressed by high endothelial cells, it is possible that ELC protein is The Cklc4 haplotypes seen in mice appear to have arisen from transported to HEV in a manner similar to the accumulation of a series of gene duplications and deletions. Such duplication events SLC on splenic arterioles. No localization of ELC protein has been are not uncommon and are thought to be due to homologous re- described. In contrast, SLC mRNA is expressed in HEV, SLC combination between repeated DNA segments on misaligned al- protein localizes to the luminal aspect of HEV, and the lack of T leles (40). An example in humans is the variation in numbers of red cell adhesion to HEV in plt mice can be reversed by the injection and green photoreceptor genes that leads to color blindness (41). In of exogenous SLC (9, 22). Thus, the available evidence suggests mice, repeated and inverted DNA segments have led to multiple that SLC is the chemokine responsible for stimulating the firm gene duplications and deletions at the T locus (42). adhesion of T cells to HEV, but does not entirely rule out a con- Although we have provided an initial characterization of Cklc4 tributory role for ELC. haplotypes, the true complexity of these haplotypes is probably Our findings also suggest that SLC contributes to the migration greater that we have demonstrated here. First, it is likely that other of activated DC from peripheral tissues into afferent lymphatics. genes are involved in this series of duplications. As an example, We have demonstrated that the migration of DC into dermal lym- the galactose-1-phosphate uridylyltransferase gene lies within 4 kb phatics is intact in plt mice (8). The preserved, albeit reduced, of the I1llra gene in humans, maps to the Cklc4 locus in mice, and expression of SLC-Leu by lymphatic endothelial cells in plt mice displays strain-specific differences in the number of hybridizing would account for this migration. It is possible that a chemotactic fragments in Southern blots of genomic DNA (43–45). Therefore, gradient is established by the diffusion of SLC into the tissues it is likely that the galactose-1-phosphate uridylyltransferase gene surrounding afferent lymphatics and that this gradient serves to in mice has been duplicated in a manner similar to the I1llra gene. attract DC, which express CCR7 upon activation (47–49). Con- Our preliminary data suggest that the region duplicated in mice sistent with this hypothesis, the migration of DC to draining LN spans at least 100 kb and that other genes are present in this region. after contact sensitization is undetectable in mice lacking CCR7, 368 VARIATIONS IN CHEMOKINE GENE NUMBER ON MOUSE CHROMOSOME 4 whereas it is only reduced in plt mice (7, 8). plt mice also dem- 13. Nakano, H., T. Yoshimoto, T. Kakiuchi, and A. Matsuzawa. 1993. Nonspecific augmentation of lymph node T cells and I-E-independent selective deletion of onstrate abnormalities in the localization of those DC that reach ϩ V␤14 T cells by Mtv-2 in the DDD mouse. Eur. J. Immunol. 23:2434. draining LN. DC in plt mice accumulate in the subcapsular sinus 14. Nakano, H., T. Tamura, T. Yoshimoto, H. Yagita, M. Miyasaka, E. C. Butcher, and superficial cortex rather than reaching the LN paracortex (8). H. Nariuchi, T. Kakiuchi, and A. Matsuzawa. 1997. Genetic defect in T lym- This localization defect may be due to the lack of SLC-Ser or ELC phocyte-specific homing into peripheral lymph nodes. Eur. J. Immunol. 27:215. 15. Nakano, H., S. Mori, H. Yonekawa, H. Nariuchi, A. Matsuzawa, and T. Kakiuchi. expression in the lymphoid organs of plt mice. Transgenic expres- 1998. A novel mutant gene involved in T-lymphocyte-specific homing into pe- sion of SLC in pancreatic islets is sufficient to stimulate the proper ripheral lymphoid organs on mouse chromosome 4. Blood 91:2886. 16. Yoshida, R., T. Imai, K. Hieshima, J. Kusuda, M. Baba, M. Kitaura, localization of DC within neolymphoid structures, suggesting that M. Nishimura, M. Kakizaki, H. Nomiyama, and O. Yoshie. 1997. Molecular SLC is the major determinant of DC localization (50). However, cloning of a novel human CC chemokine EBI1-ligand chemokine that is a spe- the possible induction of ELC expression in SLC transgenic mice cific functional ligand for EBI1, CCR7. J. Biol. Chem. 272:13803. 17. Nagira, M., T. Imai, K. Hieshima, J. Kusuda, M. Ridanpaa, S. Takagi, has not been evaluated, and it remains possible that ELC plays a M. Nishimura, M. Kakizaki, H. Nomiyama, and O. Yoshie. 1997. Molecular contributory role in this process. Determining the relative roles of cloning of a novel human CC chemokine secondary lymphoid-tissue chemokine SLC and ELC in leukocyte migration and immune response will that is a potent chemoattractant for lymphocytes and mapped to chromosome 9p13. J. Biol. Chem. 272:19518. require studies of mice in which the function of these chemokines 18. Ishikawa-Mochizuki, I., M. Kitaura, M. Baba, T. Nakayama, D. Izawa, T. Imai, is inhibited on an individual basis. H. Yamada, K. Hieshima, R. Suzuki, H. Nomiyama, et al. 1999. Molecular clon- Finally, the localization of SLC protein on splenic arterioles ing of a novel CC chemokine, interleukin-11 receptor ␣-locus chemokine (ILC), which is located on chromosome 9p13 and a potential homologue of a CC che- may suggest a role for these vessels in the homing of T cells to mokine encoded by molluscum contagiosum virus. FEBS Lett. 460:544. splenic white pulp. Both plt mice and CCR7-deficient mice dem- 19. Tanabe, S., Z. Lu, Y. Luo, E. J. Quackenbush, M. A. Berman, onstrate defects in the migration of T cells into splenic T cell L. A. Collins-Racie, S. Mi, C. Reilly, D. Lo, K. A. Jacobs, et al. 1997. Identifi- cation of a new mouse ␤-chemokine, thymus-derived chemotactic agent 4, with zones. By analogy with events demonstrated to occur in LN, SLC activity on T lymphocytes and mesangial cells. J. Immunol. 159:5671. is likely to be expressed at the site of T cell extravasation into 20. Hedrick, J., and A. Zlotnik. 1997. Identification and characterization of a novel ␤ white pulp, where it would be predicted to stimulate the activation chemokine containing six conserved cysteines. J. Immunol. 159:1589. 21. Hromas, R., C. Kim, M. Klemsz, M. Krathwohl, K. Fife, S. Cooper, of lymphocyte integrins and the firm adhesion of these cells. The C. Schnizlein-Bick, and H. Broxmeyer. 1997. Isolation and characterization of localization of SLC on splenic arterioles raises the possibility that Exodus-2, a novel C-C chemokine with a unique 37 amino acid carboxyl-terminal these vessels are the site of lymphocyte integrin activation, and extension. J. Immunol. 159:2554. 22. Gunn, M. D., K. Tangemann, C. Tam, J. G. Cyster, S. D. Rosen, and perhaps extravasation. L. T. Williams. 1998. A chemokine expressed in lymphoid high endothelial venules promotes the adhesion and chemotaxis of naive T lymphocytes. Proc. Natl. Acad. Sci. USA 95:258. Acknowledgments 23. Chan, V. W., S. Kothakota, M. C. Rohan, L. Panganiban-Lustan, J. P. Gardner, We thank Carmen Tam, currently at the Dana-Farber Cancer Institute, for M. S. Wachowicz, J. A. Winter, and L. T. Williams. 1999. Secondary lymphoid- tissue chemokine (SLC) is chemotactic for mature dendritic cells. Blood 93:3610. the performance of in situ hybridizations, Dr. Lorraine Robb of the Walter 24. Campbell, J. J., J. Hedrick, A. Zlotnik, M. A. Siani, D. A. Thompson, and and Eliza Hall Institute of Medical Research for information concerning E. C. Butcher. 1998. Chemokines and the arrest of lymphocytes rolling under the murine I1llra genes, and Dr. Jason Cyster of the University of Cali- flow conditions. Science 279:381. fornia, San Francisco, for supplying us with mouse genomic clones. 25. Pachynski, R. K., S. W. Wu, M. D. Gunn, and D. J. Erle. 1998. Secondary ␣ ␤ Thanks also to Lois Maltais of The Jackson Laboratory for her extensive lymphoid-tissue chemokine (SLC) stimulates integrin 4 7-mediated adhesion of lymphocytes to mucosal addressin cell adhesion molecule-1 (MAdCAM-1) under assistance with mouse genomic nomenclature. flow. J. Immunol. 161:952. 26. Tangemann, K., M. D. Gunn, P. Giblin, and S. D. Rosen. 1998. A high endothelial cell-derived chemokine induces rapid, efficient, and subset-selective arrest References of rolling T lymphocytes on a reconstituted endothelial substrate. J. Immunol. 1. Cyster, J. G. 1999. Chemokines and cell migration in secondary lymphoid organs. 161:6330. Science 286:2098. 27. Yoshida, R., M. Nagira, M. Kitaura, N. Imagawa, T. Imai, and O. Yoshie. 1998. 2. Sallusto, F., C. R. Mackay, and A. Lanzavecchia. 2000. The role of chemokine Secondary lymphoid-tissue chemokine is a functional ligand for the CC chemo- receptors in primary, effector, and memory immune responses. Annu. Rev. Im- kine receptor CCR7. J. Biol. Chem. 273:7118. munol. 18:593. 28. Campbell, J. J., E. P. Bowman, K. Murphy, K. R. Youngman, M. A. Siani, 3. Melchers, F., A. G. Rolink, and C. Schaniel. 1999. The role of chemokines in D. A. Thompson, L. Wu, A. Zlotnik, and E. C. Butcher. 1998. 6-C-kine (SLC), regulating cell migration during humoral immune responses. Cell 99:351. a lymphocyte adhesion-triggering chemokine expressed by high endothelium, is 4. Baggiolini, M. 1998. Chemokines and leukocyte traffic. Nature 392:565. an agonist for the MIP-3␤ receptor CCR7. J. Cell Biol. 141:1053. 5. Yoshie, O., T. Imai, and H. Nomiyama. 1997. Novel lymphocyte-specific CC 29. Ngo, V. N., H. L. Tang, and J. G. Cyster. 1998. Epstein-Barr virus-induced chemokines and their receptors. J. Leukocyte Biol. 62:634. molecule 1 ligand chemokine is expressed by dendritic cells in lymphoid tissues 6. Forster, R., A. E. Mattis, E. Kremmer, E. Wolf, G. Brem, and M. Lipp. 1996. A and strongly attracts naive T cells and activated B cells. J. Exp. Med. 188:181. putative chemokine receptor, BLR1, directs B cell migration to defined lymphoid 30. Morales, J., B. Homey, A. P. Vicari, S. Hudak, E. Oldham, J. Hedrick, R. Orozco, organs and specific anatomic compartments of the spleen. Cell 87:1. N. G. Copeland, N. A. Jenkins, L. M. McEvoy, et al. 1999. CTACK, a skin- 7. Forster, R., A. Schubel, D. Breitfeld, E. Kremmer, I. Renner-Muller, E. Wolf, and associated chemokine that preferentially attracts skin- homing memory T cells. M. Lipp. 1999. CCR7 coordinates the primary immune response by establishing Proc. Natl. Acad. Sci. USA 96:14470. functional microenvironments in secondary lymphoid organs. Cell 99:23. 31. Saeki, H., A. M. Moore, M. J. Brown, and S. T. Hwang. 1999. Cutting edge: 8. Gunn, M. D., S. Kyuwa, C. Tam, T. Kakiuchi, A. Matsuzawa, L. T. Williams, and secondary lymphoid-tissue chemokine (SLC) and CC chemokine receptor 7 H. Nakano. 1999. Mice lacking expression of secondary lymphoid organ che- (CCR7) participate in the emigration pathway of mature dendritic cells from the mokine have defects in lymphocyte homing and dendritic cell localization. skin to regional lymph nodes. J. Immunol. 162:2472. J. Exp. Med. 189:451. 32. Vassileva, G., H. Soto, A. Zlotnik, H. Nakano, T. Kakiuchi, J. A. Hedrick, and 9. Stein, J. V., A. Rot, Y. Luo, M. Narasimhaswamy, H. Nakano, M. D. Gunn, S. A. Lira. 1999. The reduced expression of 6Ckine in the plt mouse results from A. Matsuzawa, E. J. Quackenbush, M. E. Dorf, and U. H. von Andrian. 2000. The the deletion of one of two 6Ckine genes. J. Exp. Med. 190:1183. CC chemokine thymus-derived chemotactic agent 4 (TCA-4, secondary lym- 33. Koetsier, P. A., J. Schorr, and W. Doerfler. 1993. A rapid optimized protocol for phoid tissue chemokine, 6Ckine, exodus-2) triggers lymphocyte function-associ- downward alkaline Southern blotting of DNA. BioTechniques 15:260. ated antigen 1-mediated arrest of rolling T lymphocytes in peripheral lymph node 34. Robb, L., D. J. Hilton, T. A. Willson, and C. G. Begley. 1996. Structural analysis high endothelial venules. J. Exp. Med. 191:61. of the gene encoding the murine interleukin-11 receptor ␣-chain and a related 10. Warnock, R. A., J. J. Campbell, M. E. Dorf, A. Matsuzawa, L.M. McEvoy, and locus. J. Biol. Chem. 271:13754. E. C. Butcher. 2000. The role of chemokines in the microenvironmental control 35. Robb, L., D. J. Hilton, P.T. Brook-Carter, and C. G. Begley. 1997. Identification of T versus B cell arrest in Peyer’s patch high endothelial venules. J. Exp. Med. of a second murine interleukin-11 receptor ␣-chain gene (Il11Ra2) with a re- 191:77. stricted pattern of expression. Genomics 40:387. 11. Tanaka, S., A. Matsuzawa, H. Kato, K. Esaki, K. Sudo, and K. Yamanouchi. 36. Bilinski, P., M. A. Hall, H. Neuhaus, C. Gissel, J. K. Heath, and A. Gossler. 1996. 1987. Inbred strains of mice maintained at the Institute of Medical Science, Uni- Two differentially expressed interleukin-11 receptor genes in the mouse genome. versity of Tokyo. Jpn. J. Exp. Med. 57:241. Biochem. J. 320:359. 12. Matsuzawa, A., K. Sayama, A. Tsubura, and A. Murakami. 1990. A congenic line 37. Nandurkar, H. H., L. Robb, D. Tarlinton, L. Barnett, F. Kontgen, and of the DDD mouse strain, DDD/1-Mtv-2/Mtv-2: establishment and mammary C. G. Begley. 1997. Adult mice with targeted mutation of the interleukin-11 tumorigenesis. Jpn. J. Cancer Res. 81:639. receptor (Il11Ra) display normal hematopoiesis. Blood 90:2148. The Journal of Immunology 369

38. Robb, L., R. Li, L. Hartley, H. H. Nandurkar, F. Koentgen, and C. G. Begley. 45. Pilz, A., J. Fountain, J. Peters, and C. Abbott. 1993. Linkage mapping of the 1998. Infertility in female mice lacking the receptor for interleukin 11 is due to Aldo-2, Pax-5, Ambp, and D4h9S3E loci on mouse chromosome 4 in the region a defective uterine response to implantation. Nat. Med. 4:303. of homology with human chromosome 9. Genomics 18:705. ␣ 39. Bilinski, P., D. Roopenian, and A. Gossler. 1998. Maternal IL-11r function is 46. Griffith, J. S., S. M. Jensen, J. K. Lunceford, M. W. Kahn, Y. Zheng, E. A. Falase, required for normal decidua and fetoplacental development in mice. Genes Dev C. R. Lyttle, and C. Teuscher. 1997. Evidence for the genetic control of estradiol- 12:2234. regulated responses: implications for variation in normal and pathological hor- 40. Ji, Y., E. E. Eichler, S. Schwartz, and R. D. Nicholls. 2000. Structure of chro- mone-dependent phenotypes. Am. J. Pathol. 150:2223. mosomal duplicons and their role in mediating human genomic disorders. Ge- nome Res. 10:597. 47. Sallusto, F., P. Schaerli, P. Loetscher, C. Schaniel, D. Lenig, C. R. Mackay, 41. Lupski, J. R. 1998. Genomic disorders: structural features of the genome can lead S. Qin, and A. Lanzavecchia. 1998. Rapid and coordinated switch in chemokine to DNA rearrangements and human disease traits. Trends Genet. 14:417. receptor expression during dendritic cell maturation. Eur. J. Immunol. 28:2760. 42. Herrmann, B. G., D. P. Barlow, and H. Lehrach. 1987. A large inverted dupli- 48. Sozzani, S., P. Allavena, G. D’Amico, W. Luini, G. Bianchi, M. Kataura, T. Imai, cation allows homologous recombination between chromosomes heterozygous O. Yoshie, R. Bonecchi, and A. Mantovani. 1998. Differential regulation of che- for the proximal t complex inversion. Cell 48:813. mokine receptors during dendritic cell maturation: a model for their trafficking 43. Magrangeas, F., G. Pitiot, S. Dubois, E. Bragado-Nilsson, M. Cherel, S. Jobert, properties. J. Immunol. 161:1083. B. Lebeau, O. Boisteau, B. Lethe, J. Mallet, et al. 1998. Cotranscription and 49. Dieu, M. C., B. Vanbervliet, A. Vicari, J. M. Bridon, E. Oldham, S. Ait-Yahia, intergenic splicing of human galactose-1-phosphate uridylyltransferase and interleukin-11 receptor ␣-chain genes generate a fusion mRNA in normal cells: F. Briere, A. Zlotnik, S. Lebecque, and C. Caux. 1998. Selective recruitment of implication for the production of multidomain proteins during evolution. J. Biol. immature and mature dendritic cells by distinct chemokines expressed in different Chem. 273:16005. anatomic sites. J. Exp. Med. 188:373. 44. Nadeau, J. H., and E. M. Eicher. 1982. Conserved linkage of soluble aconitase 50. Fan, L., C. R. Reilly, Y. Luo, M. E. Dorf, and D. Lo. 2000. Cutting edge: ectopic and galactose-1-phosphate uridyl transferase in mouse and man: assignment of expression of the chemokine TCA4/SLC is sufficient to trigger lymphoid neo- these genes to mouse chromosome 4. Cytogenet. Cell Genet. 34:271. genesis. J. Immunol. 164:3955.