T Cells + Resides in a Subpopulation of CD43 Small Intestinal Intraepithelial Lymphocytes Maximum Immunobioactivity of Murine

Maximum Immunobioactivity of Murine Small Intestinal Intraepithelial Lymphocytes Resides in a Subpopulation of CD43+ T Cells

This information is current as Heuy-Ching Wang, Dina Montufar-Solis, Ba-Bie Teng and of September 29, 2021. John R. Klein J Immunol 2004; 173:6294-6302; ; doi: 10.4049/jimmunol.173.10.6294 http://www.jimmunol.org/content/173/10/6294 Downloaded from

References This article cites 49 articles, 23 of which you can access for free at: http://www.jimmunol.org/content/173/10/6294.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 29, 2021 *average

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

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Maximum Immunobioactivity of Murine Small Intestinal Intraepithelial Lymphocytes Resides in a Subpopulation of CD43؉ T Cells1

Heuy-Ching Wang,* Dina Montufar-Solis,* Ba-Bie Teng,† and John R. Klein2*

CD43 has been linked to many function-associated T cell activities. Using mAbs that recognize two different CD43 determinants, we show that, although mouse small intestinal intraepithelial lymphocytes (IELs) expressed the CD43 core molecule reactive with mAb R2/60, only about one-half of the total IELs—including some but not all of the TCR␣␤ and TCR␥␦ cells—expressed the -CD43 S7؊ reactive determinant. CD43 S7؉ IELs secreted more IL-2, IL-4, IL-10, IL-17, and IFN-␥ following anti-CD3 stimu .lation, and were >4-fold more cytotoxic in fresh isolates and >16-fold more cytotoxic after anti-CD3 stimulation, than S7؊ IELs -S7؉ but not S7؊ IELs from the ileum of IL-10؊/؊ mice spontaneously produced IFN-␥. In vivo BrdU uptake by IELs in non Downloaded from Ag-primed mice was greatest in the S7؉ population, indicating that significantly more S7؉ IELs than S7؊ IELs undergo cell expansion under normal homeostatic conditions. DNA microarray analyses showed that S7؉ IELs expressed higher levels of genes associated with activated T cells, whereas S7؊ IELs expressed genes used in the regulation of NK cells. These findings define two functionally distinct populations of IELs based on CD43 expression independent of TCR class, and they identify a subset of IELs that may serve as a target to better control intestinal inflammation. The Journal of Immunology, 2004, 173: 6294–6302. http://www.jimmunol.org/ urine small intestinal intraepithelial lymphocytes CD43 is a ubiquitous but enigmatic molecule on thymocytes, (IELs)3 exist in a novel state of activation. This can be peripheral T cells, and B cells, and is also found on early myeloid M seen, for example, by the fact that freshly isolated and lymphoid bone marrow hemopoietic precursors (10–13). IELs without deliberate in vitro activation have cytotoxic activity CD43 has been functionally linked to a variety of T cell activities, (1–5), yet they do not undergo proliferation or secrete T cell cy- including coactivation and enhanced cytotoxicity of CD8ϩ T cells tokines until after a TCR/CD3-mediated signal has been received (14–16). Moreover, CD43 is known to exert positive or negative (4–6). Within 12–24 h of CD3-mediated activation, however, regulatory effects depending upon when CD43 stimulation occurs IELs secrete IFN-␥ more quickly and to higher levels than CD3- during the immune response (17–19). In lymphocytic choriomen- Ϫ/Ϫ ϩ stimulated lymph node T cells (5). Consistent with that, murine ingitis virus-infected CD43 mice, for example, CD8 effector by guest on September 29, 2021 IELs express CD69 in vivo and ex vivo (5, 7), although they lack T cells were generated with the same efficiency as in CD43ϩ/ϩ other markers commonly associated with activated T cells, includ- mice, but were less capable of expansion of naive Ag-specific T ing Ly6C and CD25 (5). Further evidence for an unconventional cells, had a slight delay in effector cell trafficking into the CNS, activation state of IELs is supported by studies using DNA mi- and displayed a significantly retarded contraction phase of the im- croarray analyses of ␥␦ IELs in Yersinia pseudotuberculosis-in- mune response after infection (19). Additionally, high levels of fected and uninfected mice (8), and by serial analyses of gene CD43 expression on CD4ϩ T cells have been shown to curtail expression of ␣␤ and ␥␦ IELs (9). Of interest, in both studies, IELs activation-induced cell death (20). Interestingly, T cell prolifera- expressed genes associated with T cell activation while concomi- tion and activation in CD43-deficient mice is elevated rather than tantly expressing genes expressed during the deactivation of T depressed (17, 18), as would be expected from the costimulatory cells. However, IELs are phenotypically and functionally hetero- effects in normal mice. Taken together, these findings point to geneous. Thus, analyses of cells based on TCR␣␤ or TCR␥␦ dif- dynamic regulatory effects of CD43 depending upon the circum- ferences alone could easily reflect functionally different subsets, or stances of the immune response and the type of T cells involved. cells at different stages of differentiation, within those groups. Previous studies from our laboratory examined the expression of CD43 on murine IELs from normal mice and from mice with graft- vs-host disease (21). That study revealed the surprising finding that the pattern of CD43 expression differed for IELs compared with *Department of Diagnostic Sciences, Dental Branch, and †Research Center for Hu- peripheral T cells in that, although all murine IELs were reactive man Genetics, Institute of Molecular Medicine, University of Texas Health Science Center, Houston, TX 77030 with mAb R2/60, an Ab that reacts with the CD43 core molecule, Received for publication September 11, 2003. Accepted for publication September only about one-half of the IELs expressed the neuraminidase-sen- 3, 2004. sitive CD43 determinant recognized by mAb S7. By comparison, The costs of publication of this article were defrayed in part by the payment of page all peripheral T cells expressed the S7 determinant (21). In an charges. This article must therefore be hereby marked advertisement in accordance effort to understand that disparity for IELs, we have conducted with 18 U.S.C. Section 1734 solely to indicate this fact. studies using populations of IELs differentiated by S7 expression. 1 This work was supported in part by National Institutes of Health Grant DK35566. As reported here, there were major differences among IELs such 2 Address correspondence and reprint requests to Dr. John R. Klein, Department of that S7ϩ cells 1) synthesized significantly more IL-2, IL-4, IL-10, Diagnostic Sciences, Room 3.094F, Dental Branch, University of Texas Health Sci- ␥ ence Center at Houston, 6516 M. D. Anderson Boulevard, Houston, TX 77030. E- IL-17, and IFN- , 2) were more cytolytic in redirected cytotoxicity mail address: [email protected] assays, 3) had markedly greater in vivo proliferation, and 4) dis- 3 Abbreviation used in this paper: IEL, intraepithelial lymphocyte. played major differences in lymphocyte gene activation profiles

Isolation and purification of IELs Isolation of small intestine IELs was done as previously described (4, 5). Briefly, after tissue digestion with 2 mM DTT and 5 mM EDTA, IELs were separated from epithelial cells by centrifugation over a 40/70% Percoll gradient. For IEL fractionation into S7ϩ and S7Ϫ populations, IELs were stained by direct labeling at 4°C with PE-anti-CD43 (S7) and sorted using a Vantage SE Turbo high-speed cell sorter (BD Biosciences, Mountain View, CA). Cell sorting was done by first selecting the IEL population according to parameters of forward- and right-angle scatter, which we have previously determined to include Ͼ95% CD3ϩ cells, and second, by gating onto the S7ϩ and S7Ϫ subsets.

FIGURE 1. Nearly all murine small intestinal IELs express the CD43 Abs and flow cytometry core molecule recognized by mAb R2/60. Data are representative of four Abs used in this study were as follows: purified NA/LE anti-CD3 (145- independent experiments. 2C11); FITC-anti-TCR␤ (H57-597); FITC-anti-TCR␦ (GL3); CyChrome- anti-CD8␣ (53-6.7); biotynylated-anti-CD8␤ (53-5.8); purified, PE-, FITC-, and biotin-anti-CD43 (S7); FITC-CD44 (IM7); purified NA/LE anti- Ϫ hamster Ab (G94-56); streptavidin-allophycocyanin; strepavidin-Cy- than S7 IELs. These findings thus identify populations of IELs Chrome; streptavidin-PE; FITC-anti-BrdU; GolgiStop (monensin) (BD that can be applied to studies aimed at more precisely understand- Pharmingen, San Diego, CA); FITC-anti-IL-4 (BVD6-24G2); FITC-anti- ing IELs according to their state of activation, and they may have IL-10 (JES5-16E3); PE-anti-IFN-␥ (XMG1.2); FITC control (eBRG1); PE Downloaded from value for therapeutically targeting specific IEL subsets under con- control (eBRG1); PE-anti-CD94 (18d3); PE-anti-NK (Ly-49F/I/C/H) (eBioscience, San Diego, CA). Anti-CD43 mAb R2/60 was biotinylated in ditions of nonphysiological inflammation or autoimmunity. our laboratory. Alkaline phosphatase anti-rat IgG was purchased from Southern Biotechnology Associates (Birmingham, AL). Flow cytometric Materials and Methods analyses were done on a FACSCalibur flow cytometer using CellQuest Mice software (BD Biosciences). Intracellular staining was done using IELs cultured in RPMI 1640 (Sigma-Aldrich, St. Louis, MO) supplemented with

Adult female BALB/c and C57BL/6 mice were purchased from Harlan 10% FBS (Sigma-Aldrich) for 24 h with or without anti-CD3 stimulation http://www.jimmunol.org/ Sprague Dawley (Houston, TX). IL-10Ϫ/Ϫ (B6.129P2-Il10tm1Cgn/J) mice as previously described (4). Monensin (1 ␮l/ml) was added for the last 5 h were purchased from The Jackson Laboratory (Bar Harbor, ME). Animals of culture. Intracellular staining of IELs from 14-mo-old IL-10Ϫ/Ϫ mice were used in accord with the University of Texas Animal Welfare was done using cells recovered from the lower one-third of the small in- Guidelines. testine, followed by in vitro culture in RPMI 1640 containing 10% FBS by guest on September 29, 2021

FIGURE 2. The CD43 S7 Ag is differentially expressed on murine IELs. A–C, The CD43 S7 Ag is expressed on nearly all TCR␣␤ϩCD8␤ϩ IELs, and most but not all TCR␣␤ϩCD8␣ϩ IELs. D–F, TCR␥␦ϩ CD8␣ϩ IELs consist of both S7ϩ and S7Ϫ subsets. G–I, Three-color staining demon- strating that CD8␤ϩ IELs are S7ϩ cells, whereas the S7 determinant is differentially expressed on CD8␣ϩ IELs. Data are representative of ﬁve independent experiments. 6296 DIFFERENTIAL ROLE OF CD43 ON MURINE IELs

reacted for 2 h with 1/2000 dilution of alkaline phosphatase anti-rat IgG. Membranes were washed, developed using an ECL protocol (ICN Phar- maceuticals, Costa Mesa, CA), and exposed to Kodak (Rochester, NY) x-ray photographic film. In vivo BrdU labeling Mice were injected i.p. twice 4 h apart with 1 mg of BrdU (Sigma-Aldrich) suspended in PBS. Twenty-four hours after the last injection, IELs were recovered and stained for intracellular BrdU uptake using FITC-anti-BrdU Ab with PE-anti-CD43 Ab. DNA microarrays IELs, enriched for S7ϩ and S7Ϫ cells by high-speed cell sorting, were washed in PBS and lysed with buffer provided in the RNeasy mini-kit 50 (Qiagen, Valencia, CA). Three independent samples were used for S7Ϫ gene chip analyses, and two independent samples were used for S7ϩ gene chip analyses. Each sample consisted of cell-sorted IELs from three to four mice. RNAs were extracted and stored at Ϫ80°C until a sufficient amount FIGURE 3. CD43 S7Ϫ IELs are nonreactive with S7 mAb by Western was obtained to yield 8 ␮g for each sample. cDNAs were prepared from blotting. Cell lysates from equivalent numbers of whole IELs and IELs samples. The first-strand cDNA was primed with a T7-(dT)24 primer at depleted of S7ϩ cells by MACS sorting were electrophoresed and blotted 70°C and catalyzed by SuperScript II RT (Invitrogen Life Technologies, with mAb CD43 S7 as described in Materials and Methods. The lack of Carlsbad, CA) at 42°C. The second-strand cDNA was synthesized by nick Downloaded from reactivity of the S7 mAb against S7-depleted cell lysates confirms that the translational replacement in the presence of Escherichia coli DNA poly- merase, E. coli RNase H, and E. coli DNA ligase (Invitrogen Life Tech- S7 determinant is not expressed on those cells. Data are representative of nologies). To synthesize the biotin-labeled cRNA, the dsDNA was used as two independent experiments. template for in vitro transcription with Enzo Bioarray High Yield RNA Transcription Labeling kit (Affymetrix, Santa Clara, CA). Fragmentation of the labeled cRNA was done at 95°C before the hybridization mixture and 1 ␮l/ml monensin for 5 h to allow for intracellular accumulation of was generated; the hybridization mixture was made by adding 10 ␮gof http://www.jimmunol.org/ cytokines. Cells were collected and stained for CD43 S7, IFN-␥, and IL-4, biotin-labeled cRNA from each sample along with Eukaryotic Hybridiza- or IL-10. tion Controls and Control Oligonucleotide B2 (Affymetrix). The mixture was then hybridized to GeneChip Murine Genome U74Av2 (Affymetrix). Enzyme-linked immunoassays and redirected cell-mediated Chips were washed and stained with streptavidin-PE (Molecular Probes, cytotoxicity assay Eugene, OR). Signals were amplified by hybridizing with biotin-labeled anti-streptavidin-PE (Vector Laboratories, Burlingame, CA) and restained Commercial cytokine assay kits for IL-2, IL-4, IL-10, IL-17, and IFN-␥ ϩ Ϫ with streptavidin-PE (Molecular Probes). The arrays were scanned with were purchased from eBioscience. S7 and S7 IEL populations were GeneArray Scanner and visualized with Microarray Suite, version 5.0 (Af- obtained by flow cytometric cells sorting. In vitro anti-CD3 stimulation of Ϫ ␮ fymetrix). Gene expression levels were averaged for the three S7 samples IELs was done by coating plates overnight at 4°C with 10 g/ml anti- and for the two S7ϩ samples for determination of fold-change values and hamster Ig in PBS. Wells were washed, and 1 ␮g/ml anti-CD3 mAb was p values. Statistical analysis was done using Data Mining Tool, version 3 by guest on September 29, 2021 added in PBS for2hat37°C followed by 20-min blocking with 50% (Affymetrix); statistical significance was defined as Ն2-fold increase with newborn calf serum. Cell-free supernatants were collected after 24 h and a p value of Յ0.05, similar to that previously reported (22). Absolute call assayed for the presence of cytokines. Redirected cell-mediated assays ϩ Ϫ values for gene expression levels were determined for the five samples were done as previously reported (4, 5) using S7 and S7 IELs purified using Gene Expression MAS 5.0 software (Affymetrix), using default val- by flow cytometric cell sorting. For determining the effects of anti-CD3 ues provided by the manufacturer. stimulation on the cytotoxic activity of S7Ϫ and S7ϩ IELs, cell-sorted populations were cultured for 48 h in plates coated with anti-CD3 mAb (4). Viable cells were collected and assayed in the redirected cytotoxicity assay. Results CD8ϩ IELs express the CD43 core molecule but differentially Western blot express a murine CD43 S7 isoform ϩ A total of 107 whole unfractionated IELs, or IELs depleted of S7 cells by Studies from our laboratory demonstrated that the CD43 determinant MACS (autoMACS; Miltenyi Biotec, Auburn, CA), were solubilized in 50 recognized by the S7 mAb is expressed on 85–90% of all CD8ϩ ␮l of detergent buffer consisting of 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EGTA, 1 mM NaF, 1 ␮g/ml aprotinin, 1 ␮g/ml leupeptin, 1 peripheral T cells in mice (16). However, although all IELs expressed ␮g/ml pepstatin, and 1% Nonidet P-40 (Sigma-Aldrich). A volume of 12.5 CD43 as seen by reactivity with mAb R2/60 (Fig. 1), the S7 isoform ␮l of cell lysates was mixed with 2ϫ sample buffer (Bio-Rad, Hercules, was differentially expressed on small intestinal IELs. Thus, the S7 CA) containing 5% 2-ME (Sigma-Aldrich) and boiled for 5 min. A volume determinant was expressed on 99.2% of the TCR␣␤ϩCD8␤ϩ IELs of 15 ␮l of sample was added to precast 7.5% SDS-glycine cells (Bio-Rad). ϩ ϩ (Fig. 2, A and B), and on 80.7% of the TCR␣␤ CD8␣ IELs (A and After electrophoresis, proteins were transferred to PVDF membranes (Bio- ϩ ϩ Ϫ Rad). Membranes were blocked overnight with 5% nonfat dry milk in PBS, C). A total of 19.3% of the TCR␣␤ CD8␣ cells were S7 cells ϩ ϩ washed, and reacted for 2 h with 1 ␮g/ml purified S7 mAb, and washed and (Fig. 2, A and C). In contrast, 42.4% of the TCR␥␦ CD8a IELs

FIGURE 4. CD43 S7ϩ IELs produce significantly greater levels of IL-2, IL-4, IL-10, IL-17, and IFN-␥ than S7Ϫ IELs. IELs were enriched into S7ϩ and S7Ϫ populations by flow cytometric cells sorting. Purified cells were stimulated with immobilized anti-CD3 mAb for 24 h; cell-free supernatants were collected and assayed for cytokine activity using enzyme-linked immunoassays. Data are mean values of two independent experiments; bars indicate range of values for both experiments. The Journal of Immunology 6297

lysates from unfractionated IELs identiﬁed a 114- to 118-kDa component characteristic of that recognized by mAb S7 (21), thereby conﬁrming the lack of expression of the S7 determinant on CD43 S7Ϫ IELs.

Following anti-CD3 activation, S7ϩ IELs secrete higher levels of cytokines, and both S7ϩ and S7Ϫ populations contain Th1, Th2, and Th0 cells To determine how S7ϩ and S7Ϫ IELs differ functionally, purified S7ϩ and S7Ϫ IELs were assayed for their ability to secrete five IEL cytokines that are prominently involved in IEL regulation and/or activation (IL-2, IL-4, IL-10, IL-17, and IFN-␥). After isolation and purification by Percoll gradient centrifugation, S7ϩ and S7Ϫ IELs were enriched to Ͼ98% purity by high-speed flow cytometric Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 5. Anti-CD3 stimulation of IELs results in Th1, Th2, and Th0 profiles. A and B, IELs were cultured for 24 h with plate-bound anti-CD3 mAb and stained for three-color flow cytometric analysis for surface CD43 S7, intracellular IFN-␥, and intracellular IL-4 or IL-10. In both the S7ϩ and S7Ϫ populations, the majority of IELs were Th1 cells that synthesized IFN-␥. IL-4ϩ and IL-10ϩ cells were present among both the S7ϩ and S7Ϫ populations, although proportionally more S7ϩ cells were IL-4ϩ or IL-10ϩ cells. Note the presence of Th0 cells, particularly among the S7ϩ population. Cursor settings are positioned according to staining with FITC-Ig and PE-Ig control Abs. Data are representative of four independent experiments for each group. were S7ϩ cells (Fig. 2, D and F), and 57.6% were S7Ϫ cells (D and F). The overall distribution of S7 expression on CD8␣ and CD8␤ IELs is shown in Fig. 2, G–I. The fact that not all IELs express the S7 Ag by flow cytometry Ϫ/Ϫ did not formally rule out the possibility that the S7 determinant FIGURE 6. IELs from the ileum of IL-10 mice with rectal prolapse ␥ ϩ was expressed on S7Ϫ cells, because the lack of staining could be spontaneously produce IFN- predominantly within the S7 population. A, Ileal IELs from normal C57BL/6 mice cultured for 5 h without anti-CD3 due to a failure of mAb S7 to react with some IELs. To address stimulation fail to synthesize IFN-␥ or IL-4. B, Ileal IELs from a 14-mo-old this, Western blotting was done using the S7 mAb with cell lysates Ϫ/Ϫ ϩ IL-10 mice with rectal prolapse cultured for 5 h without anti-CD3 stim- from whole IELs (i.e., cells that included the S7 population), and ␥ ϩ ϩ ulation synthesized IFN- in the S7 population but only minimally in the IELs from which the S7 subset had been removed by negative S7Ϫ population, indicating that the S7ϩ subset is the likely source of proin- MACS sorting. As seen in Fig. 3, the S7 mAb was nonreactive by flammatory cytokines in IL-10Ϫ/Ϫ with active intestinal inflammation. Western blotting for cell lysates from S7-depleted IELs, whereas Data are representative of three IL-10Ϫ/Ϫ mice. 6298 DIFFERENTIAL ROLE OF CD43 ON MURINE IELs

FIGURE 7. CD43 S7ϩ IELs are more cytotoxic than S7Ϫ IELs. A, IELs from normal mice were puriﬁed into S7ϩ and S7Ϫ populations by ﬂow cytometric cell sorting, and assayed against P815 target cells in the presence of anti-CD3 mAb. S7ϩ IELs expressed ϳ4-fold greater cytotoxic activity than S7Ϫ IELs. B, After 48 h of anti-CD3 stimulation, there is ϳ2-fold increase in the lytic activity of S7Ϫ IELs compared with fresh S7Ϫ IELs (A); however, there is ϳ16-fold increase in the lytic activity of the S7ϩ cell population, indicating that S7ϩ IELs are more cytotoxic than S7Ϫ IELs both before and after CD3-mediated activation. C, A greater proportion of CD43 S7ϩ IELs undergo in vivo than CD43 S7Ϫ IELs in normal nonprimed mice. Twenty-four hours Downloaded from after in vivo BrdU injection, IELs were isolated and stained for intracellular BrdU with anti-CD43 S7 staining. Data are representative of two independent experiments. cell sorting. Equal numbers of each population were cultured for mice with rectal prolapse were isolated and cultured for 5 h with-

24 h in 24-well tissue culture plates coated with anti-hamster Ig out anti-CD3 stimulation in RPMI 1640 supplemented with 10% http://www.jimmunol.org/ plus hamster anti-CD3 mAb or control mAb. Cell-free superna- FBS plus monensin to permit the accumulation of intracellular tants were collected and screened for IL-2, IL-4, IL-10, IL-17, and cytokines generated in situ during the inflammatory process. Cells IFN-␥ activities by enzyme immunoassay. Findings from these were harvested and stained for CD43 S7 expression and intracel- experiments were notable in that S7ϩ cells produced significantly lular IFN-␥ and IL-4. As seen in Fig. 6A, IELs from the ileum of higher levels of all five cytokines compared with S7Ϫ IELs (Fig. normal C57BL/6 mice cultured without anti-CD3 stimulation pro- 4), indicating that there was a major difference in the levels of duced neither IFN-␥ nor IL-4 (B). In contrast, a significant pro- cytokine production by S7ϩ IELs following anti-CD3 stimulation. portion of S7ϩ ileal IELs from IL-10Ϫ/Ϫ mice without anti-CD3 Because S7ϩ and S7Ϫ IELs secreted both Th1 (IL-2 and IFN-␥) stimulation spontaneously produced IFN-␥ (Fig. 6B). Moreover, Ϫ ϩ and Th2 (IL-4 and IL-10) cytokines, albeit at quantitatively dif- IFN-␥-producing cells by S7 IELs was Ͻ10% of that of S7 by guest on September 29, 2021 ferent levels, it was of interest to determine the extent to which IELs (Fig. 6B). In neither the S7ϩ nor the S7Ϫ IEL subset was activated IELs were Th1 or Th2 cells, or whether they were Th0 there a significant number of IL-4-producing cells (Fig. 6B). These cells that produced both Th1 and Th2 cytokines (23, 24). IELs findings make an important point in that they indicate that the basic were cultured for 24 h with or without anti-CD3 stimulation. Cells relationship between S7ϩ IELs and activation observed for CD3- were collected and stained for three-color flow cytometric analysis stimulated IELs also held true in a model of intestinal inflamma- based on expression of S7, intracellular IFN-␥, and intracellular tion induced by cytokine dysregulation.

IL-4 or IL-10. After anti-CD3 stimulation, the majority of cyto- ϩ ϩ Ϫ CD43 S7 cells have higher cytolytic activity and in vivo kine-secreting cells in both the S7 and S7 groups were Th1 cells Ϫ that produced IFN-␥ but not IL-4, although proportionally more proliferation than S7 IELs ϩ Th1 cells were present in the S7 population (Fig. 5A). Addition- Small intestinal IELs are known to express cytolytic activity in ϩ Ϫ ally, a small proportion of the S7 and S7 cells were Th2 cells freshly isolated preparation without overt activation; this can be that produced IL-4 only following anti-CD3 stimulation (Fig. 5A). demonstrated using anti-CD3 or anti-TCR Abs to bridge effector ϩ Ϫ Interestingly, some cells in both the S7 and S7 populations cells to target cells (1–5). Freshly isolated IELs were enriched by were Th0 cells that produced IFN-␥ and IL-4; this was particularly high-speed cell sorting into S7ϩ and S7Ϫ populations and tested ϩ true for the S7 cells (Fig. 5A). The pattern for IL-10 staining for lytic activity in the redirected assay. As seen in Fig. 7A, there ϩ reﬂected that observed for IL-4 in that most IL-10-producing S7 were ϳ4-fold higher levels of cytolytic activity of S7ϩ cells com- cells also produced IFN-␥ (Fig. 5B). Cells cultured without anti- pared with an equivalent number of S7Ϫ cells. CD3 stimulation did not synthesize IFN-␥, IL-4, or IL-10 (data not To determine how CD3-mediated activation might affect the cy- shown). totoxic activity of S7Ϫ and S7ϩ IELs, S7Ϫ and S7ϩ cells were

ϩ Ϫ Ϫ/Ϫ purified by flow cytometric cell sorting and were cultured in plates CD43 S7 but not S7 IELs in IL-10 mice are the primary containing immobilized anti-CD3 mAb. After 48 h, cells were col- ␥ source of IFN- lected and assayed for lytic activity in the redirected cytotoxicity Studies involving anti-CD3 stimulation of IELs were relevant for assay. Although the lytic activity of both the S7Ϫ and the S7ϩ understanding the process of activation as it would occur in re- population increased over that of fresh IELs, there were quantita- sponse to Ag involving a TCR-mediated signal. It was of interest, tive differences in the increase in lytic activity of the S7Ϫ vs the however, to also examine the responses of S7ϩ and S7Ϫ IELs S7ϩ cells. Whereas the cytolytic activity of S7Ϫ IELs increased by during pathophysiological conditions of chronic intestinal inflam- ϳ2-fold, the activity of the S7ϩ IELs increased by ϳ16-fold over mation. Because most adult IL-10Ϫ/Ϫ mice Ͼ15 wk of age de- that of fresh IELs (Fig. 7, A and B). Additionally, the cytotoxic velop intestinal inflammation in the colon, ileum, and jejunum activity of the S7ϩ IELs was at least 16-fold greater than that of the (25), small intestinal IELs from the ileum of 14-mo-old IL-10Ϫ/Ϫ S7Ϫ IELs in CD3-stimulated cultures (Fig. 7B). The Journal of Immunology 6299

Murine IELs have been shown to undergo cell proliferation in levels than for S7ϩ cells (fold increase, and value of p Ͻ 0.05). vivo as demonstrated by BrdU incorporation (26). Based on the Although this likely points to a real difference between those pop- above findings of heightened cytolytic activity and cytokine secre- ulations at a functional level, it similarly reinforces the tendency of tion by S7ϩ IELs, we sought to determine whether S7 expression S7ϩ IELs to have more broad-ranging effector activities than S7Ϫ correlated with the in vivo proliferation capacity of IELs. Normal cells. mice were injected twice with BrdU as described in Materials and To confirm at the protein level the gene expression patterns Methods. Twenty-four hours after the last injection, animals were observed in Tables I and II, freshly isolated IELs were stained for sacrificed, IELs were recovered, and cells were stained for intra- expression of CD43 S7, CD44, Ly-49F/I/C/H, and CD94. Consis- cellular BrdU in conjunction with CD43 S7 expression. As seen in tent with the gene array findings, CD43 S7ϩ cells expressed CD44 Fig. 7C, significantly more CD43 S7ϩ than S7Ϫ IELs incorporated at high levels, whereas CD44 was differentially expressed on S7Ϫ BrdU in vivo, thus indicating that S7ϩ cells are an actively divid- IELs. Conversely, more than twice as many CD43 S7Ϫ IELs ex- ing cell population in vivo, either from a homeostatic proliferation pressed the NK markers, Ly-49F/I/C/H and CD94, than S7ϩ IELs process or possibly due to Ag-driven activation in vivo. (Fig. 8).

Gene array studies point to significant differences in the ϩ Ϫ functional roles of S7 and S7 IELs Discussion To more extensively examine the differences in IELs based on The distribution of IELs throughout the intestinal epithelium sug- CD43 S7 expression, DNA microarray analyses were done using gests that they play an important strategic role in the protection ϩ Ϫ cell-sorted S7 and S7 IELs. Findings from these studies rein- against the entry and dissemination of foreign Ag. Yet the exten- Downloaded from forced the observations that functional differences exist between sive phenotypic and functional complexities of the IELs have S7ϩ and S7Ϫ IELs. By three criteria (fold increase; value of p Ͻ made it difficult to fully appreciate their involvement in local im- 0.05; and absolute call for gene expression), S7ϩ IELs expressed munity. The findings described in this study offer new insight into significantly higher levels of genes for a number of functionally IEL biology by defining functionally distinct populations of cells relevant genes. The most notable of these were the CD6, CD44, that differ according to the expression of a CD43 glycoform. The and CD97 cell surface activation markers; CCR2, CXCR3, presence of the CD43 S7 Ag on IELs correlated with enhanced http://www.jimmunol.org/ and MIP-␤ gene expression; and the cathepsin L protease gene activity of three fundamental properties of activated CD8ϩ T cells, (Table I). S7Ϫ IELs also displayed several prominent features. namely, CD3-mediated cytotoxicity, cell proliferation, and CD3- Perhaps the most significant of these was the strong association induced cytokine secretion. Interestingly, although S7ϩ IELs pro- between S7Ϫ cells and the expression of genes for receptors linked duced significantly more cytokines than S7Ϫ cells for all of the to NK cell activation/inhibition (Ly49E-GE, Ly49G.2, CD94/ cytokines tested, the wide spectrum of the immunobiological ef- NKG2, Ly49H, and Ly49C) (Table II). It should be noted that, fects mediated by those cytokines suggests that additional func- although the majority of the genes with increased gene expression tional heterogeneity also may exist within the S7ϩ population. for the S7Ϫ also were expressed by S7ϩ cells (absolute call), gene Additional studies can now focus on defining those subpopulations expression for those nonetheless occurred at significantly higher in more detail. by guest on September 29, 2021

Table I. Results of gene array analyses of CD43 S7ϩ and S7Ϫ IEL: genes expressed higher in CD43 S7ϩ IELs

Absoluteb S7 cellsc Gene Title GenBank Folda p Value ϪϪϪ/ϩϩ

Cell surface CD6 U37543 45.00 0.037 AAA/PP CD44 X66084 3.75 0.003 AAA/PP CD97 AA754887 2.46 0.039 AMA/PP Chemokine CCR2 U56819 3.53 0.014 AAA/PP CXCR3 AF045146 2.39 0.004 AMA/PP MIP-1␤ X62502 2.04 0.024 AMA/PP Protease Cathepsin L X06086 18.45 0.027 AAA/PP Cell cycle Granulin; agranulin precursor (epithelium growth factor) D16195 3.64 0.014 AAA/PP GADD45␤; MyD118 AV138783 2.96 0.015 PPP/PP EMP-3; epithelium membrane protein 3 U87948 2.59 0.010 AAA/PP Cyclin-dependent kinase U19597 2.49 0.034 AAA/PP ORC6 AW045261 2.43 0.018 AAA/PP Signal transduction Nucleobindin 2; CalNuc (Ca binding) AJ222586 10.23 0.001 AAA/PP Ros1 proto-oncogene regulator of G-protein signaling 10 AI847399 3.34 0.007 AAA/PP Diacylglycerol kinase, ␣ AF085219 2.27 0.004 PPP/PP Other mGK-6kallikrein M13500 5.83 0.017 AAA/PP ADAM19; meltrin␤ AA726223 4.26 0.003 AAA/PP SH3 binding glutamic acid-rich; SH3BGR AW048272 3.66 0.014 AAA/PP Transcription factor 7, T cell speciﬁc AO019193 2.32 0.025 PPP/PP

a Fold increase. b Absolute call, relative gene expression value based on ␶ values from Affymetrix program deﬁned as absent (A), marginal (M), or present (P). c Indicates ﬁve samples consisting of three samples for S7Ϫ IELs and two samples for S7ϩ IELs. 6300 DIFFERENTIAL ROLE OF CD43 ON MURINE IELs

Table II. Results of gene array analyses of CD43 S7ϩ and S7Ϫ IEL: genes expressed higher in CD43 S7Ϫ IELs

Absoluteb S7 cellsc Gene Title GenBank Folda p Value ϪϪϪ/ϩϩ

Cell surface/TCR-associated CD72 J04170 2.28 0.025 PPP/PP CD24a M58661 2.12 0.029 PPP/PP Fc⑀RI␥ AV012229 4.77 0.044 PPP/PP Cytotoxic protein related Defensin-related cryptidin 16 U03066 4.63 0.022 PPP/PM Intestinal function and homeostasis Fibrinogen-like protein; T49 M16238 3.69 0.039 PPP/PP NK activating/inhibitory receptors Ly49E-GE U10091 3.00 0.026 PPP/PP Ly49G.2 U10094 2.39 0.021 PPP/PP CD94/NKG2 AF030311 2.25 0.017 PPP/PP Ly49H U12889 2.18 0.028 PPP/PP Ly49C U56404 2.06 0.049 PPP/PP Signal transduction Protein tyrosine phosphates, U28217 5.24 0.009 MPP/AA nonreceptor type 5 Downloaded from Septin 4 Z61452 3.31 0.009 PPP/AA EAT-2 AF020263 2.26 0.007 PPP/PP TYRO protein tyrosine kinase- AF024637 2.14 0.013 PPP/PP binding protein Rb2 X95403 2.08 0.021 PPP/PP

a Fold increase. b Absolute call, relative gene expression value based on ␶ values from Affymetrix program deﬁned as absent (A), marginal (M), or present (P). http://www.jimmunol.org/ c Indicates ﬁve samples consisting of three samples for S7Ϫ IELs and two samples for S7ϩ IELs.

A particularly important aspect of this study concerns the infor- the CD6 gene also has been shown to be down-regulated on acti- mation derived from the gene array analyses, which revealed major vated tumor-specific T cells (29). Moreover, a role for CD6 during differences in S7ϩ and S7Ϫ cells that, in some cases, could not intrathymic positive selection has recently been reported (30), thus have been easily predicted using other approaches. The presence of raising the possibility that CD6 expression on S7ϩ IELs may be high levels of CD44, CD97, and CD6 gene expression in S7ϩ IELs emblematic of lineage-associated differences between S7ϩ and Ϫ parallels the stronger effector responses of those cells as seen by S7 IELs. by guest on September 29, 2021 their in vitro cytotoxic and cytokine responses (Figs. 4 and 7). In The higher levels of gene expression of CCR2, CXCR3, and studies of whole unfractionated IELs, CD44 spans a range of ex- MIP-1␤ in CD43 S7ϩ IELs are significant for several reasons. pression from low to high, with CD44high IELs most likely con- CCR2 is expressed on activated T cells (31), and is increased in IL- stituting recently activated cells as predicted by the up-regulation 10Ϫ/Ϫ mice with colitis (32). Moreover, CCR2Ϫ/Ϫ mice have less of CD44 shortly after anti-CD3 stimulation (5). The gene array intestinal adhesion and ulceration than wild-type animals following findings regarding CD44 thus further underscore the notion that dextran sodium sulfate exposure (33). CXCR3 expression is elevated most S7ϩ IELs are highly activated cells. Similarly, greater gene in Rag-2Ϫ/Ϫ mice with colonic inflammation following transfer of expression of CD97 suggests that S7ϩ IELs are activated T cells, CD45RBhigh cells (32), and CXCR3 appears to play a role in lym- possibly with enhanced cell adhesion (27), that would augment the phocyte trafficking and localization during inflammation (34). effector activity of the S7ϩ population. Perhaps the most striking MIP-1␤ is elevated in patients with ulcerative colitis and Crohn’s aspect of these data, however, was the enormously higher level of disease (35). Taken together, those findings are in line with likelihood CD6 gene expression (45-fold) in S7ϩ vs S7Ϫ IELs. Although that S7ϩ cells may be prone to participate in the inflammatory re- CD6 has been linked to T cell activation (28), expression of sponse within the intestinal mucosa.

FIGURE 8. To conﬁrm the validity of gene expression patterns observed in Tables I and II, freshly isolated IELs were stained according to S7 expression and CD44, Ly-49F/I/C/H, and CD94. Note that nearly all CD43 S7ϩ cells were CD44ϩ cells, whereas CD44 was differentially expressed on S7Ϫ IELs. Conversely, a greater proportion of CD43 S7Ϫ IELs expressed Ly-49F/I/C/H and CD94 than CD43 S7ϩ IELs, thus conﬁrming the basic pattern observed from the gene array studies. Data are representative of two independent experiments. The Journal of Immunology 6301

The most significant feature of the gene array studies for the 3. Corazza, N., S. Muller, T. Brunner, D. Kagi, and C. Mueller. 2000. Differential S7Ϫ population was the expression of genes for receptors associ- contribution of Fas- and perforin-mediated mechanisms to the cell-mediated activity of naı¨ve and in vivo-primed intestinal intraepithelial lymphocytes. J. Im- ated with NK cell activation/inhibition. Although the majority of munol. 164:398. those genes (Ly49E-GE, Ly49G.2, CD94/NKG2, and Ly49C) 4. Wang, H. C., and J. R. Klein. 2001. Multiple levels of activation of murine CD8ϩ intraepithelial lymphocytes defined by OX40 (CD134) expression: effects on cell- have NK inhibitory activities (36, 37), the Ly49H receptor has mediated cytotoxicity, IFN-␥, and IL-10 regulation. J. Immunol. 167:6717. been shown to activate NK cells (38–42). On human IELs, the 5. Wang, H. C., Q. Zhou, J. Dragoo, and J. R. Klein. 2002. Most murine CD8ϩ NKG2D receptor for the MHC class I chain-related epithelium- intestinal intraepithelial lymphocytes are partially but not fully activated T cells. J. Immunol. 169:4717. expressed stress-induced molecules (MIC-A and MIC-B) is con- 6. Yamamoto, M., K. Fujihashi, K. Kawabata, J. R. McGhee, and H. Kiyono. 1998. stitutively expressed at low levels, and is up-regulated following A mucosal intranet: intestinal epithelial cells down-regulate intraepithelial, but anti-CD3 stimulation or upon culture with IL-15 (43), resulting in not peripheral, T lymphocytes. J. Immunol. 160:2188. ␥ 7. Kuo, S., A. E. Guindy, C. M. Panwala, P. M. Hagan, and V. Camerini. 2001. enhanced NKG2D-mediated cytotoxicity and heightened IFN- Differential appearance of T cell subsets in large and small intestine of neonatal production (43). Clearly, differences in the expression of NK re- mice. Ped. Res. 49:543. ceptor molecules could reflect discrete NK subsets present among 8. Fahrer, A. M., Y. Konigshofer, E. M. Kerr, G. Ghandour, D. H. Mack, Ϫ M. M. Davis, and Y.-H. Chien. 2001. Attributes of ␥␦ intraepithelial lymphocytes the S7 IEL population. However, regulation of NK activity could as suggested by their transcriptional profile. Proc. Natl. Acad. Sci. USA 98:10261. occur at the level of ligand expression, particularly if different NK 9. Shires, J., E. Theodoridis, and A. C. Hayday. 2001. Biological insights into ϩ ϩ receptors are coexpressed on the same cell; this would be true for TCR␥␦ and TCR␣␤ intraepithelial lymphocytes provided by serial analyses of gene expression (SAGE). Immunity 15:419. the Ly49H NK activating receptor, which binds to the m157 pro- 10. Baecher, C. M., A. J. Infante, K. L. Semcheski, and J. G. Frelinger. 1988. Iden- tein of the mouse CMV (44), for example. More important in the tification and characterization of a mouse cell surface antigen with alternative context of the present study, however, the association of NK re- molecular forms. Immunogenetics 28:295. Downloaded from Ϫ 11. Gulley, M. L., L. C. Ogata, J. A. Thorson, M. O. Dailey, and J. D. Kemp. 1988. ceptors on S7 IELs implies that those IELs differ significantly Identification of a murine pan-T cell antigen which is also expressed during ϩ from S7 IELs in the mechanisms by which they recognize Ag. terminal phases of B-cell differentiation. J. Immunol. 140:3751. The observation that the CD43 S7 marker is expressed to vary- 12. Moore, T., S. Huang, L. W. Terstappen, M. Bennett, and V. Kumar. 1994. Ex- pression of CD43 on murine and human pluripotent hematopoietic stem cells. J. ing degrees across both TCR␣␤ and TCR␥␦ cells, coupled with the Immunol. 153:4978. ϩ Ϫ differences in the functional responses of S7 and S7 cells, lends 13. Rosenstein, Y., A. Santana, and G. Pedraza-Alva. 1999. CD43, a molecule with multiple functions. Immunol. Res. 20:89. credence to the notion that many similar aspects of the IEL im- http://www.jimmunol.org/ 14. Sperling, A. I., J. M. Green, R. L. Mosley, P. Smith, J. R. Klein, J. A. Bluestone, mune response are distributed across subsets independent of TCR and C. B. Thompson. 1995. CD43 is a murine T cell costimulatory receptor that expression (45). Moreover, in addition to the functional differences functions independently of CD28. J. Exp. Med. 182:139. between S7ϩ and S7Ϫ IELs, the possibility that the CD43 S7 Ag 15. Bagriacik, E. U., M. Tang, H. C. Wang, and J. R. Klein. 2001. CD43 potentiates CD3-induced proliferation of murine intestinal intraepithelial lymphocytes. Im- may have significance as a lineage-related marker of IELs should munol. Cell Biol. 79:303. not be discounted. Note that the majority (though not all) of 16. Mosley, R. L., M. Hamad, M. Whetsell, and J. R. Klein. 1994. A novel marker TCR␥␦ IELs were CD43 S7Ϫ cells, whereas the majority (though of murine bone marrow hematopoietic stem cells that is expressed on peripheral ϩ T cells and is associated with a functionally important molecule on activated not all) of the TCR␣␤ IEL were S7 cells. This is in line with the cytotoxic T lymphocytes. Hybridoma 13:353. purported classification of IELs according to thymus-independent 17. Manjunath, N. M., M. Correa, M. Ardman, and B. Ardman. 1995. Negative ␣␤ ␣␣ ␥␦ ␣␣ regulation of T-cell adhesion and activation by CD43. Nature 377:535. by guest on September 29, 2021 (TCR CD8 and TCR CD8 ), and thymus-dependent 18. Woodman, R. C., M. J. Johnston, D. Hickey, D. Teoh, P. Reinhardt, B. Y. Poon, (TCR␣␤ CD8␣␤) lineages (46). Although extensive experimental and P. Kubes. 1998. The functional paradox of CD43 in leukocyte recruitment: work involving bone marrow reconstitution of athymic mice will a study using CD43-deficient mice. J. Exp. Med. 188:2181. 19. Onami, T. M., L. E. Harrington, M. A. Williams, M. Galvan, C. P. Larsen, be needed to define the relationship between S7 expression and T. C. Pearson, N. Manjunath, L. G. Baum, B. D. Pearce, and R. Ahmed. 2002. IEL lineage derivation, if true, analyses of IELs by S7 expression Dynamic regulation of T cell immunity by CD43. J. Immunol. 168:6022. would provide a valuable tool for penetrating into the abstruse 20. He, Y.-W., and M. J. Bevan. 1999. High level expression of CD43 inhibits T cell receptor/CD3-mediated apoptosis. J. Exp. Med. 190:1903. lineage/function relationship of IELs. 21. Bagriacik, E. U., M. D. Armstrong, M. Okabe, and J. R. Klein. 1999. Differential By all criteria (cytokine production following anti-CD3 stimu- expression of CD43 isoforms on murine T cells and their relationship to acute lation and in IL-10Ϫ/Ϫ mice, cytotoxicity in fresh IELs and fol- intestinal graft versus host disease: studies using enhanced-green fluorescent pro- ϩ tein transgenic mice. Int. Immunol. 11:1651. lowing anti-CD3 stimulation, and gene array studies), CD43 S7 22. Dutta, R., U. Singh, T.-B. Li, M. Fornage, and B.-B. Teng. 2003. Hepatic gene IELs have the potential to be the most potent effector cells, but also expression profiling reveals perturbed calcium signaling in a mouse model lack- to be the most destructive within the local intestinal environment. ing both LDL receptor and Apobec1 genes. Atherosclerosis 169:51. 23. Miner, K. T., and M. Croft. 1998. Generation, persistence, and modulation of Th0 Thus, CD43 may have considerable usefulness for further dissect- effector cells: role of autocrine IL-4 and IFN-␥. J. Immunol. 160:5280. ing regulatory and effector properties of IELs, and for therapeutic 24. Hu, C., and P. Salmage. 1999. Inability of interleukin-12 to modulate T-helper 0 approaches aimed at controlling hyperactivated IELs involved in effectors: a possible distinct subset of T cells. Immunology 97:84. 25. Kuhn, R., J. Lohler, D. Rennick, K. Rajewsky, and W. Muller. 1993. Interleukin- intestinal immunopathology. This could occur by selective elimi- 10-deficient mice develop chronic enterocolitis. Cell 75:263. nation or inactivation of IELs expressing the CD43 S7 glycoform 26. Penney, L., P. J. Kilshaw, and T. T. MacDonald. 1995. Regional variation in the ϩ ϩ through Ab immunotherapy. In cases of chronic intestinal inflam- proliferation rate and lifespan of ␣␤TCR and ␥␦TCR intraepithelial lymphocytes in the murine small intestine. Immunology 86:212. mation, this may be particularly important when coupled with ther- 27. Hamann, J., B. Vogel, G. M. W. Schijndel, and A. W. van Lier. 1996. The apeutic approaches targeted to specific inflammatory molecules seven-span transmembrane receptor CD97 has a cellular ligand (CD55, DAF). (47, 48). Similar combined therapies aimed at multiple entities J. Exp. Med. 184:1185. 28. Bott, C., J. Doshi, L. Li, S. Mcmurtry, J. Sanders, and D. Fox. 1994. Transcrip- have been shown to have beneficial synergistic effects in situations tional regulation of CD6 expression on human T lymphocytes by phorbol ester. such as with the transplantation of allograft islets in type 1 J. Immunol. 153:1. diabetes (49). 29. Zhang, X., Z. Chen, H. Huang, J. R. Gordon, and J. Xiang. 2002. DNA microarray analysis of the gene expression profiles of naive versus activated tumor- specific T cells. Life Sci. 71:3005. 30. Singer, N. G., D. A. Fox, T. M. Haqqi, L. Baretta, J. S. Endres, S. Prohaska, References J. R. Parnes, J. Bromberg, and R. M. Sramkoski. 2002. CD6: expression during 1. Viney, J., P. Kilshaw, and T. MacDonald. 1990. Cytotoxic ␣␤ϩ and ␥␦ϩ T cells development, apoptosis and selection of human and mouse thymocytes. Int. Im- in murine intestinal epithelium. Eur. J. Immunol. 20:1623. munol. 14:585. 2. Sydora, B. C., P. F. Mixter, H. R. Holcombe, P. Eghtesady, K. Williams, 31. Charo, I. F., S. J. Myers, A. Herman, C. Franci, A. J. Connolly, and M. C. Amara, A. Nel, and M. Kronenberg. 1993. Intestinal intraepithelial lym- S. R. Coughlin. 1994. Molecular cloning and functional expression of two mono- phocytes are activated and cytolytic but do not proliferate as well as other T cells cyte chemoattractant protein 1 receptors reveals alternative splicing of the car- in response to mitogenic signals. J. Immunol. 150:2179. boxyl-terminal tails. Proc. Natl. Acad. Sci. USA 91:2752. 6302 DIFFERENTIAL ROLE OF CD43 ON MURINE IELs

32. Scheetens, H., E. Hessel, R. de Waal-Malefyt, M. W. Leach, and D. Rennick. 41. George, T. C., L. H. Mason, J. R. Ortaldo, K. V. Kumar, and M. Bennett. 1999. 2001. Characterization of chemokines and chemokine receptors in two murine Positive recognition of MHC class I molecules by the Ly49D receptor of murine models of inflammatory bowel disease: IL-10Ϫ/Ϫ mice and Rag-2Ϫ/Ϫ mice re- NK cells. J. Immunol. 162:2035. constituted with CD4ϩCD45RBhigh T cells. Eur. J. Immunol. 31:1465. 42. Raziuddin, A., D. L. Longo, M. Bennett, R. Winkler-Pickett, J. R. Ortaldo, and 33. Andres, P. G., P. L. Beck, E. Mizoguchi, A. Mizoguchi, A. K. Bhan, T. Dawson, W. J. Murphy. 2002. Increased bone marrow allograft rejection by depletion of W. A. Kuziel, N. Maeda, R. P. MacDermott, D. K. Podolsky, and H.-C. Re- NK cells expressing inhibitory Ly49 NK receptors for donor class I antigens. inecker. 2000. Mice with a selective deletion of the CC chemokine receptors 5 or Transplantation 100:3026. 2 are protected from dextran sodium sulfate-mediated colitis: lack of CC che- 43. Roberts, A. I., L. Lee, E. Schwarz, V. Groh, T. Spies, E. C. Ebert, and B. Jabri. mokine receptor 5 expression results in a NK1.1ϩ lymphocyte-associated Th2- 2001. Cutting edge: NKG2D receptors induced by IL-15 costimulate CD28-neg- type immune response in the intestine. J. Immunol. 164:6303. ative effector CTL in the tissue microenvironment. J. Immunol. 167:5527. 44. Arase, H., E. S. Mocarski, A. E. Campbell, A. B. Hill, and L. L. Lanier. 2002. 34. Agace, W. W., A. I. Roberts, L. Wu, C. Greineder, E. C. Ebert, and C. M Parker. Direct recognition of cytomegalovirus by activating and inhibitory NK cell re- 2000. Human intestinal lamina propria and intraepithelial lymphocytes express ceptors. Science 296:1323. receptors specific for chemokines induced by inflammation. Eur. J. Immunol. 45. Hayday, A., E. Theodoridis, E. Ramsburg, and J. Shires. 2001. Intraepithelial 30:819. lymphocytes: exploring the third way in immunology. Nat. Immunol. 2:997. 35. Banks, C., A. Bateman, R. Payne, P. Johnson, and N. Sheron. 2003. Chemokine 46. Poussier, P., and M. Julius. 1994. Thymus independent T cell development and expression in IBD: mucosal chemokine expression is unselectively increased in selection in the intestinal epithelium. Annu. Rev. Immunol. 12:521. both ulcerative colitis and Crohn’s disease. J. Pathol. 199:28. 47. Targan, S. R., S. B. Hanauer, S. J. van Deventer, L. Mayer, D. H. Present, 36. Long, E. O. 1999. Regulation of immune responses through inhibitory receptors. T. Braakman, K. L. DeWoody, T. F. Schaible, and P. J. Rutgeerts. 1997. A Annu. Rev. Immunol. 17:875. short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor-␣ 37. Ravetch, J. V., and L. L. Lanier. 2000. Immune inhibitory receptors. Science for Crohn’s disease: Crohn’s Disease cA2 Study Group. N. Engl. J. Med. 290:84. 337:1029. 48. Marini, M., G. Bamias, J. Rivera-Nieves, C. A. Moskaluk, S. B. Hoang, 38. Colucci, F., M. A. Califiuri, and J. P. Di Santo. 2003. What does it take to make W. G. Ross, T. T. Pizarro, and F. Cominelli. 2003. TNF-␣ neutralization ame- a natural killer cell? Nat. Rev. Immunol. 3:413. liorates the severity of murine Crohn’s-like ileitis by abrogation of intestinal 39. Raulet, D. H., and W. Held. 1995. Natural killer cell receptors: the offs and ons epithelial cell apoptosis. Proc. Natl. Acad. Sci. USA 100:8366. Downloaded from of NK cell recognition. Cell 82:697. 49. Berney, T., A. Pileggi, R. D. Molano, R. Poggioli, E. Zahr, C. Ricordi, and 40. Smith, K. M., J. Wu, A. B. Bakker, J. H. Phillips, and L. L. Lanier. 1998. Ly-49D L. Inverardi. 2003. The effect of simultaneous CD154 and LFA-1 blockade on the and Ly-49 associate with mouse DAP12 and form activating receptors. J. Immu- survival of allogeneic islet grafts in nonobese diabetic mice. Transplantation nol. 161:7. 76:1669. http://www.jimmunol.org/ by guest on September 29, 2021