Slamf1, the NKT Cell Control Gene Nkt11

of February 11, 2016.This information is current as

Nkt1, the NKT Cell Control Gene Slamf1

Alan G. BaxterMargaret A. Jordan, Julie M. Fletcher, Daniel Pellicci and

http://www.jimmunol.org/content/178/3/1618doi: 10.4049/jimmunol.178.3.1618

2007; 178:1618-1627; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/178/3/1618.full#ref-list-1

, 23 of which you can access for free at: cites 43 articlesThis article

Subscriptionshttp://jimmunol.org/subscriptions

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/ji/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/cgi/alerts/etocReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2007 by The American Association of9650 Rockville Pike, Bethesda, MD 20814-3994.The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on February 11, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

Slamf1, the NKT Cell Control Gene Nkt11

Margaret A. Jordan,2* Julie M. Fletcher,2* Daniel Pellicci,† and Alan G. Baxter3*

Invariant NKT cells play a critical role in controlling the strength and character of adaptive immune responses. We havepreviously reported deficiencies in the numbers and function of NKT cells in the NOD mouse strain, which is a well-validatedmodel of type 1 diabetes and systemic lupus erythematosus. Genetic control of thymic NKT cell numbers was mapped to twolinkage regions: Nkt1 on distal chromosome 1 and Nkt2 on chromosome 2. In this study, we report the production and charac-terization of a NOD.Nkrp1b.Nkt1b congenic mouse strain, apply microarray expression analyses to limit candidate genes within the95% confidence region, identify Slamf1 (encoding signaling lymphocyte activation molecule) and Slamf6 (encoding Ly108) aspotential candidates, and demonstrate retarded signaling lymphocyte activation molecule expression during T cell development ofNOD mice, resulting in reduced expression at the CD4�CD8� stage, which is consistent with decreased NKT cell production andderanged tolerance induction in NOD mice. The Journal of Immunology, 2007, 178: 1618–1627.

I nvariant NKT (iNKT)4 cells are an immunoregulatory pop-ulation of lymphocytes that plays a critical role in controllingthe adaptive immune system and contributes to the regulation

of autoimmune responses (1–3). We have previously reported de-ficiencies in the numbers and function of NKT cells in the NODmouse strain (4, 5), which is a well-validated model of type 1diabetes (6) and systemic lupus erythematosus (7, 8), and mappedgenetic control of thymic NKT cell numbers in a first backcross(BC1) from C57BL/6 to NOD.Nkrp1b mice (9). The numbers ofthymic NKT cells of 320 BC1 mice were determined by fluores-cence-activated cell analysis using CD1d/�-galactosylceramide(CD1d/�-GalCer) tetramer (10). Tail DNA of 138 female BC1mice was analyzed for PCR product length polymorphisms at 181simple sequence repeats, providing �90% coverage of the auto-somal genome with an average marker separation of 8 cM. Twoloci exhibiting significant linkage to NKT cell numbers were iden-tified; the most significant (Nkt1; log-likelihood ratio 6.82) wasmapped near D1mit15 on distal chromosome 1 (9) in the sameregion as the NOD mouse lupus susceptibility gene Babs2/Bana3(11). The second locus (Nkt2; log-likelihood ratio 4.90) wasmapped between D2mit490 and D2mit280 on chromosome 2 (9) inthe same region as Idd13, a NOD-derived diabetes susceptibilitygene (12). In an attempt to identify the genetic sequences on chro-

mosome 1 that control NKT cell numbers, we produced and char-acterized a NOD mouse line congenic for the C57BL/6 allele at theNkt1 locus.

Materials and MethodsMice

NOD.Nkrp1b, C57BL/6J, and congenic mice were maintained at the Im-munogenetics Research Facility at the James Cook University in specificpathogen-free conditions. The NOD.Nkrp1b strain carries B6-derived al-leles at the NK complex on chromosome 6 (from D6mit105 to D6mit135),permitting the use of the NK1.1 marker (13, 14). NOD.Nkrp1b.Nkt1b micewere produced by intercrossing NOD.Nkrp1b and C57BL/6J mice and per-forming serial backcrosses to NOD.Nkrp1b to N10, before intercrossingand selection of homozygous congenic founders. These studies have beenreviewed and approved by the James Cook University Institutional AnimalCare and Ethics Committee.

DNA preparation

Extraction of genomic DNA from NOD.Nkrp1b, NOD.Nkrp1b.Nkt1b,and C57BL/6 mouse strains was conducted using the CAS-1810 X-TractorGene (Corbett Robotics) and the XTR2 X-tractor gene solid samplereagent pack (Sigma-Aldrich), which is based on a method developed inthis laboratory. Briefly, DNA was extracted by digesting an 11-mm tailtipin 400 �l of digest buffer (100 mM Tris-HCl (pH 8), 10 mM EDTA, 100mM NaCl, 0.5%SDS, 50 mM DTT, and 100 mM proteinase K), O/N, 56°C,40 rpm in a VORTEMP 56EVC (Labnet). Samples were lysed by additionof 700 �l of 5.25 M guanidine thiocyanate lysis buffer (5.25 M guanidinethiocyanate, 10 mM Tris-HCl (pH 6.5), 20 mM EDTA, 4% Triton X-100,and 64.8 mM DTT), loaded on a glass filter (GF/B) polypropylene micro-plate (Whatman International), and washed twice in propanol wash bufferand once in 100% ethanol. Samples were eluted in 150 �l of elution buffer.The DNA yield was quantified spectrophotometrically.

Genotyping

Identification of the congenic segment boundaries and the backgroundscreen were conducted by genotyping the extracted tail DNA using simplesequence repeats chosen from the Whitehead Institute simple sequencelength polymorphism library, as well as markers designed in-house on thebasis of PCR product length polymorphisms between C57BL/6 andNOD/Lt strains, as described previously (9).

RNA preparation and microarray expression analyses

To minimize activation of the apoptosis cascade, thymi were removed from4-wk-old female mice and placed in RNA-later (Qiagen) within 120 s ofthe mouse being placed in CO2 for asphyxiation. In our hands, this pro-cedure substantially improved the signal to noise ratio of expression anal-ysis, greatly reducing the numbers of differentially expressed genesidentified.

The thymi were individually homogenized in the RLT buffer of anRNeasy kit (Qiagen), with contamination minimized by extensive washing

*Comparative Genomics Center, James Cook University, Townsville, Queensland,Australia; and †University of Melbourne, Department of Microbiology and Immu-nology, Parkville, Victoria, Australia

Received for publication January 19, 2006. Accepted for publication November21, 2006.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 A.G.B. is supported by an Australian National Health and Medical Research Coun-cil Senior Research Fellowship, J.M.F. is the recipient of an Australian PostgraduateAward, and M.A.J. is the recipient of a James Cook University intramural scholarship.This project was funded by the Australian National Health and Medical ResearchCouncil.2 M.A.J. and J.M.F. contributed equally to this manuscript.3 Address correspondence and reprint requests to Dr. Alan G. Baxter, ComparativeGenomics Center, Molecular Sciences Building 21, James Cook University, Towns-ville, Queensland 4811, Australia. E-mail address: Alan.Baxter@jcu.edu.au4 Abbreviations used in this paper: iNKT, invariant NKT; CD1d/�-GalCer, CD1/�-galactosylceramide; DP, double positive; SAP, SLAM-associated protein; SH, Srchomology; SLAM, signaling lymphocyte activation molecule; SP, single positive.

Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00

The Journal of Immunology

www.jimmunol.org

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

with RNase-off and RNase-free-DNase-free-water between samples. Ho-mogenates were passed through Qiashedder columns (Qiagen) and ex-tracted (RNeasy; Qiagen). The RNA yield was quantified spectrophoto-metrically and aliquots electrophoresed for determination of sampleconcentration and purity.

Expression microarray hybridizations were performed by the AustralianGenome Research Facility using the one-cycle cDNA synthesis kit (Af-fymetrix) and Affymetrix 430 2.0 mouse gene microarray, which contains�45,000 probe sets, representing �34,000 well-substantiated mousegenes.

The probed arrays were scanned using the GeneChip Scanner 3000, andthe images (.dat files) were processed using GeneChip Operating System(GCOS, Affymetrix) and imported into Avadis Prophetic3.3 (StrandGenomics) for further analysis. The statistical significance threshold wasset by permutative analysis (10,000 permutations) and a Kruskal-Wallistest. A conservative significance threshold of p � 0.001 was set; this valuecoincided with a lack of overlap in signal values between the two groups(n � 7/group).

First-strand cDNA synthesis

First-strand cDNA was synthesized from 5 �g of total RNA using oli-go(dT) primers and Superscript II reverse transcriptase following manu-facturer’s instructions (Invitrogen Life Technologies).

Real-time quantitative PCR

Primers were designed to verify microarray data on independent samples ofRNA from NOD.Nkrp1b and NOD.Nkrp1b.Nkt1b mice. All PCR were con-ducted on the Rotorgene 3000 (Corbett) and PCR mixes set up using aCAS1200 liquid handling platform (Corbett Robotics). Each 25-�l reactioncontained 12.5 �l of Platinum Sybr Green qPCR Supermix UDG (Invitro-gen Life Technologies), 0.5 �l each primer (5 �M), 1 �l of dNTP (10mM), and 5 �l of cDNA. Slamf1 and Slamf6 expression values were nor-malized against Gapdh, as microarray expression analyses had shown that thisgene was not differentially expressed between NOD.Nkrp1b andNOD.Nkrp1b.Nkt1b mice. The primers used for quantitation were as follows:Slamf1 exons 3–5 (microarray probe 1425570_at), F primer, 5�-TAATCTTCATCCTGGTTTTCACGGC-3�, and R primer, 5�-TTGGGCATAAATAGTAAGGC-3�; Slamf1 exon7 (microarray probe 1425569_a_at), F primer, 5�-AGATGAAGAGGGAACAAAGC-3�, and R primer, 5�-TTGTTTGAAGCATAAGAGGC-3�; Slamf6 S-isoform (microarray probe 1457773_at), F primer,5�-CCTATTCCTGCTATCACG-3�, and R primer, 5�-AACTTAGAGGAAAATGGGTGC-3�; Slamf6 L-isoform (microarray probe 1420659_at), Fprimer, 5�-TGTTTGACCTCTGTGACCTTT-3�, and R primer, 5�-TACAGGAGGAACCCAACAGGC-3�; and Gapdh, F primer, 5�-TGCCGCCTGGAGAAACCTGCCAAGTATG-3�, and R primer, 5�-TGGAAGAGTGGGAGTTGCTGTTGAAGT-3�.

Analyses of unknown samples were conducted by comparison to a stan-dard curve for both the gene of interest and the housekeeper. Templatestandards were prepared by PCR amplification of cDNA from C57BL/6thymi using primers flanking those used for quantitation: Slamf1 exons 3–5(microarray probe 1425570_at), F primer, 5�-ACCACAGTCCATGCCATCAC-3�, and R primer, 5�-TCCACCACCCTGTTGCTGTA-3�; Slamf1exon7 (microarray probe 1425569_a_at), F primer, 5�-CTGGACTTTATTCTGGAAGC-3�, and R primer, 5�-TTGAGGTTCCAGAGTTTTGC-3�;Slamf6 S-isoform (microarray probe 1457773_at), F primer, 5�-TGTGTGGTATTACTCCAAGGA-3�, and R primer, 5�-AGTAACTCCATCCCCATAGC-3�; Slamf6 L-isoform (microarray probe 1420659_at), F primer,

5�-ATTTTGCTCTTGTCTCTGC-3�, and R primer, 5�-GGAATCCCTCTTTAGGTAGACTGC-3�; and Gapdh, F primer, 5�-ACCACAGTCCATGCCATCACT-3�, and R primer, 5�-TCCACCACCCTGTTGCTGTA-3�.

Titrated template standards were processed in parallel with unknowncontrols.

Primer design and sequencing

Primers for sequencing were designed using BioTechnix 3d 1.1.0, based onsequence obtained from UCSC Genome Bioinformatics (http://genome.ucsc.edu) such that overlapping sequences would be amplified across thepromoter, coding, and noncoding mRNA sequences. PCR were performedfor all mouse strains for all regions on Omn-E thermal cyclers (Hybaid).Each 100-�l reaction included 10 �l of 10� PCR buffer with 3 mM MgCl2(Roche), 0.4 mM each of dATP, dCTP, dGTP, and dTTP (Astral), 1.6 U ofTaq Polymerase (Roche), and 2 �l of DNA (cDNA). Approximately 20 �lof mineral oil overlaid the reaction mix. PCR protocol included denatur-ation 95°C, 3 min, then 32 (40) cycles (95°C, 1 min; 50–62°C (primerdependant annealing) 1 min, 72°C, 1 min), followed by an extension stepof 72°C, 7 min. Reactions were verified by 1% agarose gel electrophoresis.Reactions were then purified using the Qiagen PCR purification kit fol-lowing manufacturers directions. Twenty to 100 ng of PCR product be-tween 200 and 500 bp/100–160 ng of PCR product between 500 and 1000bp were prepared with 6.4 pmol primer and sent to the Australian GenomeResearch Facility for sequencing (both forward and reverse reactions foreach). The raw data were retrieved by FTP and analyzed using Sequencher3.1.1. (Gene Codes).

Cell suspension preparation

Thymocyte cell suspensions were prepared by gently grinding the thymusbetween two frosted microscope slides in MACS buffer (PBS containing 2

FIGURE 1. Characterization of the NOD.Nkrp1b.Nkt1b congenic mouse line. The boundaries of the Nkt1 congenic segment on distal chromosome 1 areindicated (A). Proportions (B) and absolute numbers (C) of thymic NKT cell numbers in 5-wk-old mice from the congenic line and the NOD.Nkrp1b parentalline as determined by CD1d/�-GalCer tetramer binding are shown. Values for NOD.Nkrp1b mice are indicated by � whereas those for NOD.Nkrp1b.Nkt1b

mice by �. Proportion means and SEM are shown, whereas for numbers, individual values and statistical analysis (Mann-Whitney U test) are given.

Table I. List of genetic markers tested to confirm genetic homogeneityof NOD.Nkrp1b.Nkt1b mouse line

D1mit58, D1mit72, D1mit279, D1mit124, D1mit180, D1mit438, D1mit306,D1mit494, D1mit348, D1mit445, D1mit103, D1mit199, D1mit199,D1mit102, D1mit449, D1mit288, D1mit 369, D1mit396*, D1mit33,*D1Bax208,* D1Bax15,* D1mit406,* D1mit209,* D1mit155,* D2mit1,D2mit362, D2mit458, D2mit92, D2mit256, D2mit490, D2mit283,D2mit412, D2mit528, D2mit265, D3mit60, D3mit203, D3mit25, D3mit187,D3mit13, D3mit84, D3mit147, D3mit19, D4mit264, D4mit139, D4mit301,D4mit9, D4mit11, D4mit204, D4mit226, D4mit59, D5mit48, D5mit387,D5mit81, D5mit113, D5mit239, D5mit406, D5mit245, D5mit169, D6mit86,D6mit224, D6mit209, D6mit178, D6mit343, D6mit259, D6mit15, D7mit76,D7mit225, D7mit84, D7mit301, D7mit101, D7mit334, D8mit155,D8mit281, D8mit191, D8mit54, D8mit81, D8mit211, D8mit166, D9mit90,D9mit285, D9mit26, D9mit335, D9mit165, D9mit269, D9mit136, D9mit17,D9mit52, D10mit104, D10Bax30, D10mit15, D10mit198, D10mit42,D10mit95, D11mit77, D11mit131, D11mit5, D11mit357, D11mit198,D11mit61, D12mit190, D12mit156, D12mit259, D12mit141, D13mit158,D13mit221, D13mit7, D13mit202, D13mit230, D13mit78, D14mit207,D14mit62, D14mit63, D14mit239, D14mit197, D14mit178, D15mit174,D15mit179, D15mit255, D15mit121, D15mit71, D15mit159, D15mit193,D15mit35, D16mit131, D16mit98, D16mit189, D17mit133, D17mit176,D17mit68, D17mit70, D17mit93, D17mit130, D18mit60, D18mit123,D18mit206, D18mit4, D19mit78, D19mit79, D19mit73, D19mit119,D19mit91, D19mit35

� Presence of C57BL/6-derived alleles.

1619The Journal of Immunology

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

mM EDTA (Amresco) and 0.5% (w/v) BSA (ICN Biomedicals). Spleenswere disrupted using a 26-gauge needle and forceps and the resulting cellsuspension treated with RBC lysing buffer (Sigma-Aldrich).

Flow cytometric analysis

For flow cytometric analyses cells were labeled with anti-�TCR-FITC (clone H57-597), anti-CD3-FITC (clone 145-2C11), anti-CD3-allophycocyanin (clone 145-2C11), anti-CD3-allophycocyanin-Cy7 (clone145-2C11), anti-CD4-allophycocyanin (clone GK1.5), anti-CD4-PerCP-Cy5.5 (clone RM 4-5), anti-NK1.1-PE-Cy7 (clone PK136), anti-CD8-FITC (clone 53-6.7), anti-CD45R/B220-allophycocyanin (cloneRA3-6B2), anti-CD44-FITC (clone IM7), all from BD Pharmingen, andanti-CD150(SLAM)-PE (clone TC15-12F12.2; Biolegend). Mouse CD1d

tetramer, conjugated to either PE or PE-Cy7 and loaded with �-GalCer,was produced in house as previously described (10) using recombinantbaculovirus encoding his-tagged mouse CD1d and mouse �2-microglobu-lin, provided by Prof. M. Kronenberg’s laboratory (La Jolly Institute forAllergy and Immunology, San Diego, CA).

For surface staining, Abs were diluted in MACS buffer. Cells werepreincubated for 15 min with CD16/32 (clone 93; eBioscience), followedby an additional 20-min incubation with 10% mouse serum to prevent FcRbinding, before addition of surface staining Ab mixtures. Viable lympho-cytes were identified by the forward and side scatter profile and in somecases by propidium iodide exclusion. A forward scatter-area against for-ward scatter-height gate was used to exclude doublets from analysis.Where possible, an empty fluorescent channel was used to exclude

FIGURE 2. Thymic and splenic sub-sets of NKT cells in the NOD.Nkrp1b.Nkt1b congenic mouse line. Flow cyto-metric analyses of thymic CD1d/�-GalCer tetramer-binding NKT cellsubset proportions of mice aged 2 and 4wk (top panels) and 6 wk (middle leftpanel) of age from the NOD.Nkrp1b.Nkt1b congenic and the NOD.Nkrp1b pa-rental line as defined by the CD44, CD4,and NK1.1 markers are illustrated. Datafor splenic NKT cell subset proportionsof mice aged 6 wk are also shown forboth inbred lines, as well as for (NOD.Nkrp1b � NOD.Nkrp1b.Nkt1b)F1 mice(middle right panel). Histograms illus-trating thymic (bottom left panel) andsplenic (bottom right panel) NKT cellnumbers for 6-wk-old mice are shown inwhich values from NOD.Nkrp1b miceare indicated by �, those for NOD.Nkrp1b.Nkt1b mice are indicated by f,and those for (NOD.Nkrp1b � NOD.Nkrp1b.Nkt1b)F1 mice indicated in u

(mean � SEM; n � 4–7).

Table II. Thymic and splenic NKT cell numbers at 6 wk

Organ Strain nTotal Cell

Number (�10�6) % NKT CellsNumber NKTCells (�10�5)

Thymus NOD.Nkrp1 7 113 � 7 0.09 � 0.01 1.0 � 0.2F1 7 143 � 9 0.16 � 0.02 2.3 � 0.3NOD.Nkrp1.Nkt1 4 184 � 7 0.17 � 0.01 3.1 � 0.2

Spleen NOD.Nkrp1 5 75 � 3 0.30 � 0.02 2.3 � 0.3F1 4 65 � 1 0.44 � 0.03 2.9 � 0.3NOD.Nkrp1.Nkt1 5 62 � 3 0.55 � 0.02 3.4 � 0.2

1620 GENETIC CONTROL OF NKT CELL NUMBERS

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

autofluorescent cells. In general, flow cytometry was performed on aFACSVantage SE with FACSDiVa option (BD Biosciences) and dataanalyzed using either CellQuest Pro or FACSDiVa software (BD Bio-sciences). The data in Fig. 8E were acquired on a CyAn ADP flowcytometer (DakoCytomation) and analyzed with Summit 4.3 software(DakoCytomation).

Proliferation assays

Single-cell suspensions were cultured in triplicate at 37°C, 5% CO2 for 3–5days in RPMI 1640 medium with L-glutamine (Invitrogen Life Technolo-gies) supplemented with 100 U/ml penicillin, 100 �g/ml streptomycin sul-fate, and 50 �M 2-ME. Stimulation was achieved by the addition of Dyna-bead Mouse CD3/CD28 T cell expander beads (Dynal Biotech) in varyingproportions. Some cultures were established in the presence of 100 �g/mlof the blocking signaling lymphocyte activation molecule (SLAM) peptide132–146 (FCKQLKLYEQVSPPE; Auspep), 100 �g/ml of the nonblockingSLAM peptide 83–97 (DLSKGSYPDHLEDGY; Auspep) or inhibiting con-centrations (6.25 �g/ml) of the TC15 anti-SLAM mAb (Biolegend). Prolif-eration was assayed by the addition of 0.25 �Ci of 6-3H-labeled thymidineper 200-�l well (GE Healthcare) 8–16 h before harvesting. At termination,plates were spun to pellet cells, 100 �l of supernatant was removed forcytokine assays, and the cells were harvested with a Tomtec Harvester 96Mach IIIM, the emission scintillated with MeltiLex A melt-on scintillatorsheets (Wallac) and detected with a Wallac 1450 Microbeta Jet liquid scin-tillation counter.

Cytokine measurement

Cytokine levels in cell culture supernatants were determined using MouseTh1/Th2 Cytokine Cytometric Bead Array (BD Biosciences). Capturebeads (30 �l, specific for IL-2, IL-4, IL-5, IFN-�, and TNF) together with30 �l of culture supernatant samples and 30 �l of PE detection reagent,were incubated for 2 h in 96-well plates. Beads were washed twice with200 �l of wash buffer, resuspended, and data were acquired using aFACSCalibur (BD Biosciences). Serial dilutions of the provided cyto-kine standards were prepared and assayed as described above. Standardcurves were generated and samples quantified using the BD CBA soft-ware (BD Biosciences).

ResultsNOD.Nkrp1b.Nkt1b congenic mice

A NOD.Nkrp1b.Nkt1b congenic mouse line carrying a C57BL/6-derived chromosomal segment spanning the 95% confidence in-terval of Nkt1 was produced by serial backcrossing to theNOD.Nkrp1b strain to N10, followed by intercrossing and selec-tion for Nkt1b homozygotes. The proximal boundary of the con-genic segment lies between D1mit369 and D1mit396 and the distalboundary is distal to the most telomeric marker available,D1mit155 (Fig. 1A). A background screen of 136 polymorphic locidistributed throughout the rest of the autosomal genome failed to

FIGURE 3. Averaged log signal intensity of Af-fymetrix Mouse 430 series 2 expression microarray pro-filing of thymi from NOD.Nkrp1b and NOD.Nkrp1b.Nkt1b

mice (n � 7/group; A). Results of genes for which a p �0.05 was obtained are illustrated. F, Genes that are highlydifferentially expressed (experiment-wise permutativeanalysis threshold of p � 0.001; Kruskal-Wallis test). Di-agonal lines indicate 2-fold differential expression. Num-bers indicate gene identities as listed in Fig. 4. The linkagedata from Ref. 9 are presented transformed to physicaldistances (B), the location of the Nkt1 congenic interval(indicated by the f) presented on the same scale (C), andthe locations of the highly differentially expressed genesdisplayed as a histogram (D). The probability of 21 of the28 locatable highly differentially expressed genes mappingto the Nkt1 congenic region by chance is p � 10�200 (�2

one sample test).

1621The Journal of Immunology

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

detect any residual C57BL/6-derived genomic contamination (Ta-ble I). Flow cytometric analyses of thymic NKT cell numbers andproportions, as determined by CD1d/�-GalCer tetramer binding,confirmed that thymi from the NOD.Nkrp1b.Nkt1b congenic linehave larger proportions (Figs. 1B and 2) and numbers (Fig. 1C,Table II) of iNKT cells than those from the NOD.Nkrp1b parentalstrain controls. The increase in thymic NKT cell numbers in thecongenic line is a product of both a higher proportion of NKT cellsin the thymus and an increase in total thymic cellularity (Table II).The proportions of thymic NKT cell numbers in (NOD.Nkrp1b �NOD.Nkrp1b.Nkt1b)F1 mice is intermediate between the congenicand parental strains.

Splenic NKT cell proportions and numbers are also increased in thecongenic line compared with the parental line (Fig. 2, Table II; p �0.05; Mann-Whitney U test). Again, the proportions and numbers ofNKT cells in the spleens of (NOD.Nkrp1b � NOD.Nkrp1b.Nkt1b)F1

mice are intermediate between the congenic and parental strains.

NKT cell subsets in NOD.Nkrp1b.Nkt1b congenic mice

Thymic NKT cell subsets, which are related to each other by adevelopmental pathway, can be defined by the cell surface markersCD4, CD44, and NK1.1 (15, 16). Flow cytometric analyses ofthymic and splenic NKT cells indicate that the majority of theadditional NKT cells found in NOD.Nkrp1b.Nkt1b mice belong tothe CD4�CD44highNK1.1� population (Fig. 2), which is consid-ered to be relatively developmentally immature. Similarly, in thespleen, the majority of the additional NKT cells found inNOD.Nkrp1b.Nkt1b mice are CD4�NK1.1�.

Microarray gene expression analysis

To identify a subset of candidate genes within the Nkt1 linkage95% confidence interval, microarray gene expression analysis wasperformed on thymi of 4-wk-old NOD.Nkrp1b and NOD.Nkrp1b.Nkt1b mice (n � 7/group; Fig. 3A), following procedures to min-imize activation of the apoptosis cascade. Thymic RNA was ex-tracted, hybridization and scanning of Affymetrix Mouse 430series 2 expression microarrays performed by the Australian Ge-nome Research Facility, and data imported into Avadis Propheticusing an RMA summarization algorithm. The statistical signifi-cance threshold was set by permutative analysis (10,000 permuta-tions) and a Kruskal-Wallis test applied. A total of only 28 geneswere identified as being highly differentially expressed (i.e., thosewith a p � 0.001), of which 21 mapped to the Nkt1 congenicregion (1.6% of genome; �2 � 986; df � 1; p � 10�200; �2 onesample test; Fig. 3, B–D). This result is indicative of an extremelygood signal-to-noise ratio.

Only 15 of the 21 highly differentially expressed genes mappingto the Nkt1 congenic region lie within the 95% confidence limitsobtained in the original linkage analysis (Ref. 9; Fig. 4). Theirphysical positions and expression fold change are shown in Fig.4D, and their identities given in the figure legend. Of these genes,

FIGURE 4. Physical locations of highly differentially expressed geneson chromosome 1 in relationship to linkage data from Ref. 9 (A) and lo-cation of the Nkt1 congenic segment (indicated by f; B). The 95% linkageconfidence interval is shown (fine lines). C and D, Genes are indicated byhistogram. The width of each bar indicates the physical length of the geneand the height represents the fold change of differential expression betweenthe NOD.Nkrp1b and NOD.Nkrp1b.Nkt1b mice, with a positive displace-ment indicating higher expression in the NOD.Nkrp1b.Nkt1b congenic line.D illustrates the 95% confidence interval at higher resolution, and the num-bers indicate the identities of individual genes, as indicated: Nr locus de-scription—1) 2010005O13rik, RIKEN cDNA 2010005O13 gene; 2) Rgs5,regulator of G-protein signaling 5; 3) Hsd17b7, hydroxysteroid (17-�) de-hydrogenase 7; 4) Nr1i3, nuclear receptor subfamily 1, group I, member 3;5) Ppox, protoporphyrinogen oxidase; 6) 6030405P05rik, RIKEN cDNA6030405P05 gene; 7) Slamf1, SLAM; 8) Slamf6, SLAM family member 6;9) Ltap, loop tail-associated protein; 10) Pex19, peroximal biogenesis fac-tor 19; 11) Loc623121, similar to IFN-activated gene 203; 12) A1447904,similar to IFN-activated gene 203; 13) Ifi203, IFN-activated gene 203; 14)Ifi202b, IFN-activated gene 202B; and 15) Ifi205, IFN-activated gene 205.The same key is applied to gene identities in Fig. 3.

FIGURE 5. Microarray profiling of expression levels of SLAM familymembers of thymi from NOD.Nkrp1b and NOD.Nkrp1b.Nkt1b mice. Af-fymetrix 430 2.0 mouse gene microarray probes and locations were as follows:Slamf1, probe 1 (1425569_a_at) 707–962 bp, probe 2 (1425570_at) 1528–2038 bp, and probe 3 (1425571_at) 2312–2738 bp; Slamf2 (1427301_at) 522–705 bp; Slamf3 (1449156_at) 1928–2296 bp; Slamf4, probe 1 (1426120_a_at)748–978 bp, and probe 2 (1449991_at) 1495–1862 bp; Slamf5, probe 1(1422875_at) 942–1444 bp, and probe 2 (1446505_at) 68–486 bp; Slamf6,probe 1 (1420659_at) 1877–2323 bp, probe 2 (1425086_a_at) 979–1134bp, and probe 3 (1457773_at) 210–667 bp; Slamf7 (1453472_a_at) 1337–1758 bp; Slamf8 (1425294_at) 997–1541 bp; and Slamf9 (1419315_at)550–1064 bp. Mann-Whitney U test applied with significance thresholdcorrected for multiple hypothesis testing.

1622 GENETIC CONTROL OF NKT CELL NUMBERS

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

the most prominent candidates for control of NKT cell numbersare Slamf1 and Slamf6 because signaling through SLAM-associ-ated protein (SAP) appears to be essential for thymic positive se-lection of NKT cells (reviewed in Ref. 17; see Discussion).

The expression data from the thymi of NOD.Nkrp1b andNOD.Nkrp1b.Nkt1b mice for Slam family members 1–9 are shownin Fig. 5. Only Slamf1 and Slamf6 were significantly differentiallyexpressed in the congenic line compared with the parental line( p � 0.001; Mann-Whitney U test).

Characterization of SLAM expression in thymus and spleen

Validation of Slamf1 and Slamf6 microarray data was obtained byquantitative RT-PCR of the sequences probed by the array on anindependent sample set (Fig. 6, n � 6–9; Fig. 7, n � 5–9). Val-idation of SLAM expression on thymic and splenic lymphocyteswas also performed by flow cytometry (Fig. 8). Consistent withmicroarray and RT-PCR quantitation of thymic SLAM expression,thymocytes from NOD.Nkrp1b.Nkt1b congenic mice expressedsignificantly more SLAM on their surfaces than those of the pa-rental strain (Fig. 8A). The cell surface markers CD3, CD4, andCD8 can be used to define the developmental pathway of T cellsfrom the least mature CD4�CD8� (double-negative) CD3�,through a double-positive (DP) intermediate stage, to the mostmature CD4 or CD8 single-positive (SP) subsets immediately be-fore thymic export (Fig. 8B). Flow cytometric analysis of SLAMexpression on thymocytes from each developmental stage revealedmajor differences in the developmental program of thymic SLAMexpression between the NOD.Nkrp1b and NOD.Nkrp1b.Nkt1b

strains. While the NOD.Nkrp1b.Nkt1b mice and C57BL/6 mice(data not shown) express high levels of SLAM on DP thymocytes,with a relatively lower level expressed on mature SP cells, expres-sion of SLAM on the developing T cells of NOD.Nkrp1b mice isretarded, reaching its peak of expression only at the mature SPstage (Fig. 8C, D). At each developmental stage, the levels ofexpression of SLAM on the thymocytes of (NOD.Nkrp1b �NOD.Nkrp1b.Nkt1b)F1 mice were intermediate between the twoparental strains (data not shown). Levels of expression of SLAMon mature thymic NKT cells are similar (Fig. 8E).

Consistent with the levels of SLAM expression on mature SPthymocytes, splenic expression was relatively similar on both Tand B cells of both strains and the (NOD.Nkrp1b � NOD.Nkrp1b.Nkt1b)F1 mice (Fig. 8, F–I; data for F1 mice not shown).

Functional consequences of differences in SLAM expression

To determine whether the difference in SLAM expression on DPthymocytes between the NOD.Nkrp1b and NOD.Nkrp1b.Nkt1b

strains was sufficient to have functional effects, an assay of SLAMfunction was established. As SLAM acts as a costimulator throughhomotypic interactions, and the difference in levels of expressionwas largely restricted to the DP population of thymocytes, SLAMfunction was assessed by measuring TCR-stimulated proliferationof whole thymocytes. Whole thymocytes or purified CD4� spleno-cytes were stimulated in vitro with anti-CD3/anti-CD28 coatedbeads and the proliferative response detected by thymidine incor-poration five or three days later, respectively. The validity of thissystem as a surrogate measure of SLAM function was confirmed

FIGURE 6. Validation of expression microarray pro-filing of Slamf1 expression in thymi from NOD.Nkrp1b

and NOD.Nkrp1b.Nkt1b mice. Gene structure and exonstructure of the three SLAM isoforms are shown to-gether with the relative locations of microarray probetargets and RT-PCR primer sites (A). The microarrayresults (B; n � 7) and RT-PCR validation (C; n � 6–9)performed on an independent sample are shown.

1623The Journal of Immunology

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

by the inhibition of proliferation by the addition of 100 �g/ml ofthe blocking SLAM peptide 132–146 or inhibiting concentrations(6.25 �g/ml) of the TC15 anti-SLAM mAb (Fig. 9A).

Consistent with varying expression levels of SLAM on DP thy-mocytes modulating the signal threshold of the responding matureSP thymocytes, greater proliferation was seen in cultures of thy-mocytes from NOD.Nkrp1b.Nkt1b mice than from those ofNOD.Nkrp1b mice (Fig. 9B; n � 6; p � 0.05; Mann-Whitney Utest). The addition of exogenous IL-2 eliminated this differencebetween the strains, which is consistent with SLAM’s role as acostimulator. No significant difference in TCR-stimulated prolif-eration of splenic CD4� T cells was observed, which is consistentwith the absence of a difference in the levels of SLAM expressionon this population.

The supernatants from the cultures were then assayed for cyto-kines. Thymocytes from NOD.Nkrp1b mice produced significantlyless IL-4 and IL-5, and slightly more IFN-�, in a manner analo-gous to the cytokine phenotypes of Slamf1�/� and Sap�/� targetedmutant mice (18–20). A similar deviation in IL-4 production wasseen in cultures of CD4� splenocytes (Fig. 9, C–E).

DiscussionThe production and characterization of the NOD.Nkrp1b.Nkt1b

mouse strain described here formally confirmed the location of amajor NKT cell control gene on distal chromosome 1, as the con-genic mice had a 2-fold increase in proportions, and a three-foldincrease in absolute numbers, of thymic NKT cells at six weeks ofage. While conventional NOD lines lack the NK1.1 developmental

marker, the NOD.Nkrp1b parental line used in these studies is con-genic for Nkrp1b, the allele encoding NK1.1, and was specificallydeveloped by us to allow analysis of the major NKT cell subsets(13, 14). The presence of C57BL/6-derived alleles at the Nkrp1locus on chromosome 6 does not affect either the numbers of NKTcells, nor the strain’s susceptibility to autoimmune disease (14).Flow cytometric analyses of thymic and splenic NKT cell subsetsdefined by the cell surface markers CD44, CD4, and NK1.1 (15,16, 21) indicated that the majority of the additional NKT cellsfound in NOD.Nkrp1b.Nkt1b mice belonged to the CD4� NK1.1�

population, which is considered to be developmentally relativelyimmature (15, 16). This finding suggests that the addition of theC57BL/6-derived allele is sufficient to increase the number of cellsentering the NKT cell developmental pathway, but insufficient topush those cells through to maturity. The functional characteristicsof these cells, as well as the additional effects of Nkt2, the otherNKT cell control locus, are issues of great interest.

Two strategies were applied to reduce the number of Nkt1 can-didate genes under consideration. The first was the use of geneexpression microarrays. As a generalization, this has not been aparticularly helpful strategy in the past, and reports of hundreds orthousands of differentially expressed genes in congenic mice havebeen published (e.g., Ref. 22). In our experience, a dramatic im-provement in signal-to-noise ratio could be attained by avoidingengagement of the activation and apoptosis cascades. In this spe-cific case, thymi were removed from mice within 120 s of theinduction of anoxia and placed immediately in RNAlater. The sec-ond strategy applied was the use of a stringent statistical threshold,

FIGURE 7. Validation of expression microarrayprofiling of Slamf6 expression in thymi from NOD.Nkrp1b and NOD.Nkrp1b.Nkt1b mice. Gene structureand exon structure of the two Slamf6 isoforms areshown together with the relative locations of microarrayprobe targets and RT-PCR primer sites (A). The mi-croarray results (B; n � 7) and RT-PCR validation (C;n � 5–9) performed on an independent sample areshown.

1624 GENETIC CONTROL OF NKT CELL NUMBERS

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

rather than ad hoc fold difference thresholds, which have no ob-vious biological validity. As a consequence of these procedures, 21of the 28 locatable highly differentially expressed genes mapped tothe Nkt1 congenic region and only fifteen of these genes lay withinthe Nkt1 95% confidence interval. To our knowledge, no microar-ray expression analysis of congenic mice has produced a bettersignal to noise ratio. Of the 15 highly differentially expressedgenes lying within the Nkt1 95% confidence interval, the mostprominent candidates for control of NKT cell numbers are Slamf1and Slamf6, as signaling through SAP appears to be essential forthymic positive selection of NKT cells (reviewed in Ref. 17).

Slamf1 encodes the Ig-like receptor termed SLAM or CD150,which forms homotypic interactions and modulates immune re-sponses (17, 23). It associates with, and signals through, the Srchomology (SH)2-domain containing adaptor protein SAP, which ismutated in the human inherited immunodeficiency X-linked lym-phoproliferative disease (24–26), and FynT, a Src-related proteintyrosine kinase, which is recruited to SAP through a unique inter-action involving the SH2 domain of SAP and the SH3 domain ofFynT (27). Ligation of SLAM with mAbs enhanced TCR-stimu-lated proliferation and cytokine production by human and mouse Tcells (28–30), which is consistent with a role as a costimulator(31). T cells from Slamf1�/� targeted mutant mice deficient inSLAM expression have a severe defect in TCR-activated produc-tion of IL-4 in vitro but produce slightly more IFN-�, consistentwith an important role in modulating the character of immuneresponses (18). Consistent with its role in SLAM signal transduc-

tion, Sap�/� targeted mutant mice showed a similar defect in Tcell-mediated IL-4 production and slightly increased IFN-� pro-duction, compared with SAP-sufficient wild-type CD4� T cells(19, 20, 32, 33). Significantly, SAP-deficient X-linked lymphopro-liferative patients as well as mice bearing targeted deletions ofSAP (34–36) or FynT (37, 38) lack NKT cells, indicating a criticalrole for the SAP/FynT signaling pathway, presumably activatedfollowing recruitment to one or more members of the SLAM familyof cell surface receptors. As NKT cells are positively selected on DPthymocytes (39, 40), selection is dependent on the SAP/FynT sig-naling pathway (34–38), and SLAM is known to be expressed onthe surfaces of DP thymocytes (41), SLAM-SLAM interactionsmay be responsible (42). Slamf1 lies within a haplotype blockcontaining genes encoding nine SLAM family members, many ofwhich contain multiple polymorphism between the minority hap-lotype, expressed in C57BL/6, C57L, C57BR, C57BL/10, and RF(haplotype 1) and the majority haplotype, which is expressed in129/SvJ, A/J, AKR/J, BALB/cJ, C3H/HeJ, CBA/J, DBA/2J, MRL/MpJ, NOD/Lt, NZB/B1WJ, NZW, SJL/J, and others (haplotype 2;Ref. 43). The lupus susceptibility gene Sle1b has been localized tothis region by congenic mapping and is expressed in haplotype 2(43).

RT-PCR and flow cytometry confirmed a major difference inSLAM expression on the thymocytes of NOD.Nkrp1b.Nkt1b andNOD.Nkrp1b mice. Comparison of SLAM levels on thymic sub-sets revealed variation in the developmentally regulated pattern ofexpression between the strains. While the NOD.Nkrp1b.Nkt1b mice

FIGURE 8. SLAM expression on thymocytes (A–E)and splenocytes (F–I) of NOD.Nkrp1b and NOD.Nkrp1b.Nkt1b mice (n � 5/group). Profiles from NOD.Nkrp1b mice are indicated with the fine lines, whereasthose from NOD.Nkrp1b.Nkt1b are indicated with theheavy lines. The T cell developmental pathway is indi-cated (B) and example profiles of SLAM expression onthymocyte subsets gated for CD3, CD4, and CD8 ex-pression presented (C). Means and SEM of mean fluo-rescence intensities (MFI) are shown for each stage (n �5; D). Values for NOD.Nkrp1b mice are indicated by �,whereas those for NOD.Nkrp1b.Nkt1b mice by �. SLAMlevels of thymic CD1d/�-GalCer tetramer-binding NKTcell CD4� and double-negative subsets were determinedby flow cytometry (n � 3; E). Splenocytes were gated by�TCR (T cells) and B220 (B cells; G). Example profilesare shown (H) as well as individual values (n � 5; I).

1625The Journal of Immunology

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

express increasing levels of SLAM through T cell development topeak on DP thymocytes and then decline to relatively lower levelson mature SP cells, expression of SLAM on developing T cells ofNOD.Nkrp1b is retarded, reaching its peak of expression only atthe mature SP stage. Consistent with the levels of SLAM expres-sion on mature SP thymocytes, splenic expression was relativelysimilar between the strains on both T and B cells. This differencein SLAM expression was of functional importance, as it affectedboth TCR-stimulated proliferation as well as cytokine production.Significantly, thymocytes and CD4� splenocytes from NOD.Nkrp1b mice produced less IL-4, and slightly more IFN-�, in amanner analogous to the cytokine phenotypes of Slamf1�/� andSap�/� targeted mutant mice (18–20).

The retardation of developmentally programmed SLAM expres-sion in NOD mice has three significant implications. First, as DPthymocytes account for �80% of the thymus, it explains the find-

ing of differential gene expression between the strains in wholethymic RNA preparations. Second, because NKT cells are positivelyselected on DP thymocytes via a mechanism dependent on the SAP/FynT signaling pathway, decreased SLAM expression at this de-velopmental stage may provide an explanation for the reducednumbers of NKT cells in NOD mice. Third, as SLAM also acts asa costimulator for conventional T cells, it is possible that the rel-atively lower levels of SLAM expression at the stage of negativeselection (late DP stage) compared with those at maturity (SP thy-mocytes and in the periphery) result in a lowering of the signalingthreshold of conventional T cells in the periphery. If true, this mayresult in an increased proportion of peripheral T cells capable ofresponding to self-Ags.

In conclusion, the data presented make a strong case for thehypothesis that the control of NKT cell numbers attributed to theNkt1 gene is mediated by differential expression of Slamf1 and are

FIGURE 9. The costimulation effect of SLAM ex-pression by DP thymocytes on anti-CD3/CD28-inducedproliferation of whole thymocytes from NOD.Nkrp1b

and NOD.Nkrp1b.Nkt1b mice is illustrated in A in theabsence or presence of blocking SLAM peptide 132–146 or inhibiting concentrations of the TC15 anti-SLAM mAb. Proliferation of control cultures containingno added inhibitors or the nonblocking SLAM peptide83–97 is also shown. Functional effects of allelic SLAMexpression on TCR signaling in vitro of thymocytes andsplenocytes from NOD.Nkrp1b (�) and NOD.Nkrp1b.Nkt1b mice (f; n � 6/group; B–E). Cultures of wholethymocytes (3 days) and purified CD4� splenocytes (5days) were stimulated with anti-CD3/anti-CD28-coatedbeads in the presence or absence of IL-2, and prolifer-ation assayed by thymidine incorporation (B). Levels ofIL-4 (C), IL-5 (D), and IFN-� (E) in culture superna-tants were measured by cytokine bead array.

1626 GENETIC CONTROL OF NKT CELL NUMBERS

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from

consistent with an additional contribution by Slamf6. In addition, itis possible that the retarded programmed expression of SLAM ondeveloping conventional T cells of NOD mice may contribute tolowering their TCR signaling threshold in the periphery therebycontributing to autoimmune disease in this strain.

AcknowledgmentsWe acknowledge the invaluable advice and suggestions of Dale Godfreyand Stuart Tangye, technical assistance by Tatiana Tsoutsman, Tim Butler,and Rhiana Magee, and the help and support of our animal attendantsNicole Fraser and Rohan Henderson.

References1. Godfrey, D. I., and M. Kronenberg. 2004. Going both ways: immune regulation

via CD1d-dependent NKT cells. J. Clin. Invest. 114: 1379–1388.2. Van Kaer, L. 2005. �-Galactosylceramide therapy for autoimmune diseases:

prospects and obstacles. Nat. Rev. Immunol. 5: 31–42.3. Brigl, M., and M. B. Brenner. 2004. CD1: antigen presentation and T cell func-

tion. Ann. Rev. Immunol. 22: 817–890.4. Baxter, A. G., S. J. Kinder, K. J. Hammond, R. Scollay, and D. I. Godfrey. 1997.

Association between ��TCR�CD4�CD8� T cell deficiency and IDDM inNOD/Lt mice. Diabetes 46: 572–582.

5. Hammond, K. J., L. D. Poulton, L. J. Palmisano, P. A. Silveira, D. I. Godfrey,and A. G. Baxter. 1998. �/�-T cell receptor (TCR)�CD4�CD8� (NKT) thy-mocytes prevent insulin-dependent diabetes mellitus in nonobese diabetic(NOD)/Lt mice by the influence of interleukin (IL)-4 and/or IL-10. J. Exp.Med. 187: 1047–1056.

6. Makino, S., K. Kunimoto, Y. Muraoka, Y. Mizushima, K. Katagiri, and Y.Tochino. 1980. Breeding of a non-obese, diabetic strain of mice. Jikken Dobutsu29: 1–13.

7. Baxter, A. G., A. C. Horsfall, D. Healey, P. Ozegbe, S. Day, D. G. Williams, andA. Cooke. 1994. Mycobacteria precipitate an SLE-like syndrome in diabetes-prone NOD mice. Immunology 83: 227–231.

8. Silveira, P. A., and A. G. Baxter. 2001. The NOD mouse as a model of SLE.Autoimmunity 34: 53–64.

9. Esteban, L. M., T. Tsoutsman, M. A. Jordan, D. Roach, L. D. Poulton, A. Brooks,O. V. Naidenko, S. Sidobre, D. I. Godfrey, and A. G. Baxter. 2003. Geneticcontrol of NKT cell numbers maps to major diabetes and lupus loci. J. Immunol.171: 2873–2878.

10. Matsuda, J. L., O. V. Naidenko, L. Gapin, T. Nakayama, M. Taniguchi, C. R.Wang, Y. Koezuka, and M. Kronenberg. 2000. Tracking the response of naturalkiller T cells to a glycolipid antigen using CD1d tetramers. J. Exp. Med. 192:741–754.

11. Jordan, M. A., P. A. Silveira, D. P. Shepherd, C. Chu, S. J. Kinder, J. Chen,L. J. Palmisano, L. D. Poulton, and A. G. Baxter. 2000. Linkage analysis ofsystemic lupus erythematosus induced in diabetes-prone nonobese diabetic miceby Mycobacterium bovis. J. Immunol. 165: 1673–1684.

12. Hamilton-Williams, E. E., D. V. Serreze, B. Charlton, E. A. Johnson, M. P.Marron, A. Mullbacher, and R. M. Slattery. 2001. Transgenic rescue implicates�2-microglobulin as a diabetes susceptibility gene in nonobese diabetic (NOD)mice. Proc. Natl. Acad. Sci. USA 98: 11533–11538.

13. Hammond, K. J., D. G. Pellicci, L. D. Poulton, O. V. Naidenko, A. A. Scalzo,A. G. Baxter, and D. I. Godfrey. 2001. CD1d-restricted NKT cells: an interstraincomparison. J. Immunol. 167: 1164–1173.

14. Poulton, L. D., M. J. Smyth, C. G. Hawke, P. Silveira, D. Shepherd, O. V.Naidenko, D. I. Godfrey, and A. G. Baxter. 2001. Cytometric and functionalanalyses of NK and NKT cell deficiencies in NOD mice. Int. Immunol. 13:887–896.

15. Pellicci, D. G., K. L. Hammond, A. P. Uldrich, A. G. Baxter, M. J. Smyth, andD. I. Godfrey. 2002. A natural killer T (NKT) cell developmental pathway in-volving a thymus-dependent NK1.1�CD4� CD1d-dependent precursor stage.J. Exp. Med. 195: 835–844.

16. Benlagha, K., T. Kyin, A. Beavis, L. Teyton, and A. Bendelac. 2002. A thymicprecursor to the NK T cell lineage. Science 296: 553–555.

17. Nichols, K. E., C. S. Ma, J. L. Cannons, P. L. Schwartzberg, and S. G. Tangye.2005. Molecular and cellular pathogenesis of X-linked lymphoproliferative dis-ease. Immunol. Rev. 203: 180–199.

18. Wang, N., A. Satoskar, W. Faubion, D. Howie, S. Okamoto, S. Feske, C. Gullo,K. Clarke, M. R. Sosa, A. H. Sharpe, and C. Terhorst. 2004. The cell surfacereceptor SLAM controls T cell and macrophage functions. J. Exp. Med. 199:1255–1264.

19. Wu, C., K. B. Nguyen, G. C. Pien, N. Wang, C. Gullo, D. Howie, M. R. Sosa,M. J. Edwards, P. Borrow, A. R. Satoskar, et al. 2001. SAP controls T cellresponses to virus and terminal differentiation of TH2 cells. Nat. Immunol. 2:410–414.

20. Czar, M. J., E. N. Kersh, L. A. Mijares, G. Lanier, J. Lewis, G. Yap, A. Chen,A. Sher, C. S. Duckett, R. Ahmed, and P. L. Schwartzberg. 2001. Altered lym-phocyte responses and cytokine production in mice deficient in the X-linkedlymphoproliferative disease gene SH2D1A/DSHP/SAP. Proc. Natl. Acad. Sci.USA 98: 7449–7454.

21. Gadue, P., and P. L. Stein. 2002. NK T cell precursors exhibit differential cyto-kine regulation and require Itk for efficient maturation. J. Immunol. 169:2397–2406.

22. Eaves, I. A., L. S. Wicker, G. Ghandour, P. A. Lyons, L. B. Peterson, J. A. Todd,and R. J. Glynne. 2002. Combining mouse congenic strains and microarray geneexpression analyses to study a complex trait: the NOD model of type 1 diabetes.Genome Res. 12: 232–243.

23. Veillette, A., and S. Latour. 2003. The SLAM family of immune-cell receptors.Curr. Opin. Immunol. 15: 277–285.

24. Sayos, J., C. Wu, M. Morra, N. Wang, X. Zhang, D. Allen, S. van Schaik,L. Notarangelo, R. Geha, M. G. Roncarolo, et al. 1998. The X-linked lympho-proliferative-disease gene product SAP regulates signals induced through theco-receptor SLAM. Nature 395: 462–469.

25. Nichols, K. E., D. P. Harkin, S. Levitz, M. Krainer, K. A. Kolquist, C. Genovese,A. Bernard, M. Ferguson, L. Zuo, E. Snyder, et al. 1998. Inactivating mutationsin an SH2 domain-encoding gene in X-linked lymphoproliferative syndrome.Proc. Natl. Acad. Sci. USA 95: 13765–11370.

26. Coffey, A. J., R. A. Brooksbank, O. Brandau, T. Oohashi, G. R. Howell, J. M.Bye, A. P. Cahn, J. Durham, P. Heath, P. Wray, et al. 1998. Host response to EBVinfection in X-linked lymphoproliferative disease results from mutations in anSH2-domain encoding gene. Nat. Genet. 20: 129–135.

27. Latour, S., R. Roncagalli, R. Chen, M. Bakinowski, X. Shi, P. L. Schwartzberg,D. Davidson, and A. Veillette. 2003. Binding of SAP SH2 domain to FynT SH3domain reveals a novel mechanism of receptor signalling in immune regulation.Nat. Cell Biol. 5: 149–154.

28. Cocks, B. G., C. C. Chang, J. M. Carballido, H. Yssel, J. E. de Vries, and G.Aversa. 1995. A novel receptor involved in T cell activation. Nature 376:260–263.

29. Aversa, G., C. C. Chang, J. M. Carballido, B. G. Cocks, and J. E. de Vries. 1997.Engagement of the signaling lymphocytic activation molecule (SLAM) on acti-vated T cells results in IL-2-independent, cyclosporin A-sensitive T cell prolif-eration and IFN-� production. J. Immunol. 158: 4036–4044.

30. Castro, A. G., T. M. Hauser, B. G. Cocks, J. Abrams, S. Zurawski, T. Churakova,F. Zonin, D. Robinson, S. G. Tangye, G. Aversa, et al. 1999. Molecular andfunctional characterization of mouse signaling lymphocytic activation molecule(SLAM): differential expression and responsiveness in Th1 and Th2 cells. J. Im-munol. 163: 5860–5870.

31. Howie, D., S. Okamoto, S. Rietdijk, K. Clarke, N. Wang, C. Gullo, J. P.Bruggeman, S. Manning, A. J. Coyle, E. Greenfield, et al. 2002. The role of SAPin murine CD150 (SLAM)-mediated T-cell proliferation and interferon � pro-duction. Blood 100: 2899–2907.

32. Cannons, J. L., L. J. Yu, B. Hill, L. A. Mijares, D. Dombroski, K. E. Nichols,A. Antonellis, G. A. Koretzky, K. Gardner, and P. L. Schwartzberg. 2004. SAPregulates Th2 differentiation and PKC-�-mediated activation of NF-�B1. Immu-nity 21: 693–706.

33. Davidson, D., X. Shi, S. Zhang, H. Wang, M. Nemer, N. Ono, S. Ohno, Y.Yanagi, and A. Veillette. 2004. Genetic evidence linking SAP, the X-linked lym-phoproliferative gene product, to Src-related kinase FynT in Th2 cytokine reg-ulation. Immunity 21: 707–717.

34. Nichols, K. E., J. Hom, S. Y. Gong, A. Ganguly, C. S. Ma, J. L. Cannons,S. G. Tangye, P. L. Schwartzberg, G. A. Koretzky, and P. L. Stein. 2005. Reg-ulation of NKT cell development by SAP, the protein defective in XLP. Nat.Med. 11: 340–345.

35. Pasquier, B., L. Yin, M. C. Fondaneche, F. Relouzat, C. Bloch-Queyrat, N.Lambert, A. Fischer, G. de Saint-Basile, and S. Latour. 2005. Defective NKT celldevelopment in mice and humans lacking the adapter SAP, the X-linked lym-phoproliferative syndrome gene product. J. Exp. Med. 201: 695–701.

36. Chung, B., A. Aoukaty, J. Dutz, C. Terhorst, and R. Tan. 2005. Signaling lym-phocytic activation molecule-associated protein controls NKT cell functions.J. Immunol. 174: 3153–3157.

37. Gadue, P., N. Morton, and P. L. Stein. 1999. The Src family tyrosine kinase Fynregulates natural killer T cell development. J. Exp. Med. 190: 1189–1196.

38. Eberl, G., B. Lowin-Kropf, and H. R. MacDonald. 1999. Cutting edge: NKT celldevelopment is selectively impaired in Fyn-deficient mice. J. Immunol. 163:4091–4094.

39. Gapin, L., J. L. Matsuda, C. D. Surh, and M. Kronenberg. 2001. NKT cells derivefrom double-positive thymocytes that are positively selected by CD1d. Nat. Im-munol. 2: 971–978.

40. Egawa, T., G. Eberl, I. Taniuchi, K. Benlagha, F. Geissmann, L. Hennighausen,A. Bendelac, and D. R. Littman. 2005. Genetic evidence supporting selection ofthe V�14i NKT cell lineage from double-positive thymocyte precursors. Immu-nity 22: 705–716.

41. Romero, X., D. Benitez, S. March, R. Vilella, M. Miralpeix, and P. Engel. 2004.Differential expression of SAP and EAT-2-binding leukocyte cell surface mole-cules CD84, CD150 (SLAM), CD229 (Ly9) and CD244 (2B4). Tissue Antigens64: 132–144.

42. Borowski, C., and A. Bendelac. 2005. Signaling for NKT cell development: theSAP-FynT connection. J. Exp. Med. 201: 833–836.

43. Wandstrat, A. E., C. Nguyen, N. Limaye, A. Y. Chan, S. Subramanian, X. H.Tian, Y. S. Yim, A. Pertsemlidis, H. R. Garner, Jr., L. Morel, and E. K.Wakeland. 2004. Association of extensive polymorphisms in the SLAM/CD2gene cluster with murine lupus. Immunity 21: 769–780.

1627The Journal of Immunology

by guest on February 11, 2016http://w

ww

.jimm

unol.org/D

ownloaded from