This information is current as of March 16, 2015.

Genetic Investigation of MHC-Independent Missing-Self Recognition by Mouse NK Cells Using an In Vivo Bone Marrow Transplantation Model Peter Chen, Oscar A. Aguilar, Mir Munir A. Rahim, David S. J. Allan, Jason H. Fine, Christina L. Kirkham, Jaehun Ma, Miho Tanaka, Megan M. Tu, Andrew Wight, Vicky Kartsogiannis, Matthew T. Gillespie, Andrew P. Makrigiannis and James R. Carlyle

Supplementary Material References Subscriptions Permissions Email Alerts

http://www.jimmunol.org/content/suppl/2015/02/13/jimmunol.140152 3.DCSupplemental.html This article cites 55 articles, 22 of which you can access for free at: http://www.jimmunol.org/content/194/6/2909.full#ref-list-1 Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscriptions Submit copyright permission requests at: http://www.aai.org/ji/copyright.html Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/cgi/alerts/etoc

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 9650 Rockville Pike, Bethesda, MD 20814-3994. Copyright © 2015 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606.

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

J Immunol 2015; 194:2909-2918; Prepublished online 13 February 2015; doi: 10.4049/jimmunol.1401523 http://www.jimmunol.org/content/194/6/2909

The Journal of Immunology

Genetic Investigation of MHC-Independent Missing-Self Recognition by Mouse NK Cells Using an In Vivo Bone Marrow Transplantation Model

MHC-I–specific receptors play a vital role in NK cell–mediated “missing-self” recognition, which contributes to NK cell activation. In contrast, MHC-independent NK recognition mechanisms are less well characterized. In this study, we investigated the role of NKR-P1B:Clr-b (Klrb1:Clec2d) interactions in determining the outcome of murine hematopoietic cell transplantation in vivo. Using a competitive transplant assay, we show that Clr-b2/2 bone marrow (BM) cells were selectively rejected by wild-type B6 recipients, to a similar extent as H-2Db2/2 MHC-I–deficient BM cells. Selective rejection of Clr-b2/2 BM cells was mitigated by NK depletion of recipient mice. Competitive rejection of Clr-b2/2 BM cells also occurred in allogeneic transplant recipients, where it was reversed by selective depletion of NKR-P1Bhi NK cells, leaving the remaining NKR-P1Blo NK subset and MHC-I– dependent missing-self recognition intact. Moreover, competitive rejection of Clr-b2/2 hematopoietic cells was abrogated in Nkrp1b-deficient recipients, which lack the receptor for Clr-b. Of interest, similar to MHC-I–deficient NK cells, Clr-b2/2 NK cells were hyporesponsive to both NK1.1 (NKR-P1C)–stimulated and IL-12/18 cytokine–primed IFN-g production. These findings support a unique and nonredundant role for NKR-P1B:Clr-b interactions in missing-self recognition of normal hematopoietic cells and suggest that optimal BM transplant success relies on MHC-independent tolerance mechanisms. These findings provide a model for human NKR-P1A:LLT1 (KLRB1:CLEC2D) interactions in human hematopoietic cell transplants. The Journal of Immunology, 2015, 194: 2909–2918.

B

one marrow (BM) transplantation models have vital importance in studying immunological tolerance and clinical therapies for hematopoietic disorders and cancer.

*Department of Immunology, Sunnybrook Research Institute, University of Toronto, Toronto, Ontario M4N 3M5, Canada; †Department of Biochemistry, Microbiology, and Immunology, University of Ottawa, Ottawa, Ontario K1H 8M5, Canada; ‡ Prince Henry’s Institute, Monash Medical Centre, Clayton, Victoria 3168, Australia; and xDepartment of Biochemistry and Molecular Biology, Monash University, Clayton, Victoria 3168, Australia 1

P.C. and O.A.A. contributed equally to this work.

2

A.P.M. and J.R.C. contributed equally to this work.

Received for publication June 13, 2014. Accepted for publication December 26, 2014. This work was supported by an operating grant from the Canadian Institutes of Health Research (CIHR 86630; to A.P.M. and J.R.C.), an Early Researcher Award from the Ontario Ministry of Research and Innovation (to J.R.C.), and support from the Victorian Government Operational Infrastructure Support Program (to M.T.G.). A.P.M. holds a Canada Research Chair in Innate Pathogen Resistance. J.R.C. held a Canadian Institutes of Health Research New Investigator Award and holds an Investigator in the Pathogenesis of Infectious Disease Award from the Burroughs Wellcome Fund. Address correspondence and reprint requests to Prof. James R. Carlyle or Prof. Andrew P. Makrigiannis, Department of Immunology, University of Toronto, Sunnybrook Research Institute, 2075 Bayview Avenue (S-236), Toronto, ON M4N 3M5, Canada (J.R.C.) or Department of Biochemistry, Microbiology, and Immunology, University of Ottawa, 451 Smyth Road, Ottawa, ON K1H 8M5, Canada (A.P.M.). E-mail addresses: [email protected] (J.R.C.) or [email protected] (A.P.M.) The online version of this article contains supplemental material. Abbreviations used in this article: BM, bone marrow; LAK, lymphokine-activated killer; NIH, National Institutes of Health; ODN, oligodinucleotide; poly I:C, polyinosinic-polycytidylic acid; qRT-PCR, quantitative RT-PCR; RCMV-E, rat CMV-English; WT, wild-type. Copyright Ó 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00 www.jimmunol.org/cgi/doi/10.4049/jimmunol.1401523

Standard mixed BM chimeras are used to evaluate the development and function of the immune system, where donor BM cells are genetically manipulated and engrafted into syngeneic, allogeneic, or immunocompromised recipients, including NOD/SCID (1) or RAG/gc (2) mice. Clinically, cellular transplantation has expanded from a treatment for immunodeficient and chemotherapy patients to adoptive cell-mediated immune therapy for combatting malignant cells. Adoptive therapies take advantage of the graft-versustumor effect (3), a regimen inspired by clinical graft-versus-host disease (4). Hematopoietic transplant success depends upon the minimization of immunological graft rejection and the maximization of donor stem cell reconstitution. Although transplantation of healthy hematopoietic cells could be beneficial to patients, their immune systems actively prevent efficient engraftment of allogeneic transplants. A major obstacle preventing BM transplant reconstitution is the presence of host NK cell activity. The importance of NK cells in BM graft rejection surfaced when “hybrid resistance” was reported. Hybrid resistance occurs when F1 hybrid offspring of two MHC-disparate mouse strains reject or are resistant to BM grafts derived from either parent (5–8). The mechanism behind hybrid resistance is postulated to stem from “missing-self” recognition of one of two codominantly expressed parental MHC-I haplotypes (e.g., H-2b or H-2d alleles in [C57BL/6xBALB/c]F1 mice), which are independently sensed by F1 NK cell subsets expressing nonoverlapping inhibitory self–MHC-I–specific Ly49 receptor repertoires. During development, F1 NK cells are “educated” to express a Ly49 receptor repertoire specific for, and tolerant to, both self– MHC-I haplotypes (9–11); however, because NK receptor expression is variegated, at least four overarching categories can be

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

Peter Chen,*,1 Oscar A. Aguilar,*,1 Mir Munir A. Rahim,† David S. J. Allan,* Jason H. Fine,* Christina L. Kirkham,* Jaehun Ma,* Miho Tanaka,* Megan M. Tu,† Andrew Wight,† Vicky Kartsogiannis,‡ Matthew T. Gillespie,‡,x Andrew P. Makrigiannis,†,2 and James R. Carlyle*,2

2910

ROLE OF Clr-b IN HEMATOPOIETIC TRANSPLANT REJECTION

Materials and Methods Cells BWZ.36 and P815 cells were obtained from Drs. Nilabh Shastri and David Raulet (University of California, Berkeley, Berkeley, CA). BWZ.CD3z/ NKR-P1B, BWZ.Clr-b cells were generated previously (22, 24). Cells were cultured in complete DMEM–high glucose, supplemented with 2 mM glutamine, 100 U/ml penicillin, 100 mg/ml streptomycin, 50 mg/ml gentamicin, 110 mg/ml sodium pyruvate, 50 mM 2-ME, 10 mM HEPES, and 10–20% FCS. Mouse tissues were processed into single-cell suspensions using 40-mm filters (BD Biosciences), ACK Lysing Buffer (Life Technologies), and/or Lympholyte (Cedarlane). Solid organs were processed using gentleMACS (Miltenyi Biotec).

Animals B6, B6.CD45.1, and B6.b2m2/2 mice were purchased from The Jackson Laboratory; B6.H-2Kb2/2Db2/2, B6.H-2Kb2/2, and B6.H-2Db2/2 mice from Taconic; and National Institutes of Health (NIH) Swiss mice from National Cancer Institute Frederick (25); B6.Clr-b2/2 (Ocil2/2; Clec2d2/2) mice were described previously (26). B6.Nkrp1bB62/2 mice were generated as described (27). The B6.Clr-b2/2 genetic background was verified from genomic tail DNA via the Illumina Mouse Medium Density Linkage Panel Whole-Genome Genotyping Service (Hospital for Sick Children, Toronto, ON, Canada). B6.Clr-b2/2b2m2/2 and B6.CD45.1/CD45.2 double-congenic

mice were generated in-house. All animals were maintained according to approved protocols at Sunnybrook Research Institute or the Donnelly Centre, University of Toronto, Toronto, ON, Canada.

RT-PCR and quantitative RT-PCR Tissue RNA was isolated using an RNA Purification Kit (Invitrogen) and reverse transcribed using oligo-dT primer and a SuperScript III cDNA Synthesis Kit (Invitrogen) (22, 23). Semiquantitative RT-PCR was performed using 50–125 ng of template cDNA, ExpandPLUS High Fidelity enzyme (Roche), and primers in Table I. PCR products were cloned (pcDNA3.1/TOPO; Invitrogen) and sequenced. Real-time quantitative RTPCR (qRT-PCR) was performed on a CFX-96 Real-Time PCR Detection System (Bio-Rad) using 50–100 ng of template cDNA, SsoFast EvaGreen Supermix (Bio-Rad), and primers in Table I. Data were analyzed using CFX Manager software.

Abs Abs were purchased from eBioscience, BD Pharmingen, Sigma-Aldrich, or BioLegend, except Clr-b mAb, 4A6 (22), and NKR-P1BB6 mAb, 2D12 (28). The mAb clones are as follows: NKG2D (MI6); Ly49C/I/F/H (14B11); Ly49C/I (5E6); Ly49H (3D10); Ly49I (YLI-90); Ly49D (4E5); Ly-6G/Gr-1 (RB6-8C5); FLAG (M2); CD19 (ID3); H-2Kb (AF6-88.5); H-2Db (KH95); CD25 (3C7); NKG2A (16a11), NKG2A/C/E (20d5); CD244.2/2B4 (eBio244F4); CD11c (N418); NKp46 (29A1.4); CD94 (18d3); CD44 (1M7); CD4 (GK1.5); CD11b (M1/70); CD8a (53-6.7); CD3ε (145-[11C]); CD45.1 (A20); NK1.1/NKR-P1CB6 (PK136); CD45.2 (104); IFN-g (XMG1.2); and FcgRII/III (2.4G2). 2D12 purified mAb and hybridoma were generous gifts from Dr. Koho Iizuka (University of Minnesota, MN). Subcloned 2D12 hybridoma supernatant was purified using protein A resin (Pierce) and biotinylated in-house by the hybridoma facility at Sunnybrook Research Institute.

Flow cytometry Cells were FcR blocked on ice for 10 min using 2.4G2 supernatant, stained with primary mAb for 25 min, and washed and stained for 20 min with secondary streptavidin conjugates. Intracellular staining, fixation, and permeabilization used Cytofix/Cytoperm (BD Biosciences), as well as IFN-g Ab (30 min in 13 Perm/Wash). Cells were analyzed using an LSRII (BD Biosciences) and FlowJo software (TreeStar). DAPI exclusion was used for live cell viability.

Competitive BM rejection assay Donor cells from various animals were singly or combinatorially labeled using PKH26, PKH67 (Sigma-Aldrich), and/or CellVue Maroon (eBioscience) (Table II), washed, counted, and stoichiometrically pooled prior to transplant. Recipient mice were sublethally irradiated (6.5 Gy; Mark I Cssource irradiator; J.L. Shepherd & Associates) on day (–1). Untreated mice or NK-depleted cohorts [injected i.p. with 150–200 mg PK136 on day (–2)] were injected i.v. with 10–21 3 106 labeled cells on day (0). Recipient spleens were harvested 18–48 h post transplant and analyzed by flow cytometry. Some mice were treated with polyinosinic-polycytidylic acid (poly I:C) (24 h prior) or CpG-B oligodinucleotides (ODN) (6 h prior), where noted, before transplant. For NKR-P1BB6–deficient recipients, WT and Clr-b2/2 donor cells were labeled with 5 mM or 0.5 mM CFSE in lieu of PKH dyes. Percent engraftment of experimental donor cells relative to control WT or syngeneic donor cells was determined as follows: [(% donor)/ (% control)]input/[(% donor)/(% control)]output 3 100%. Percent Rejection = 100% 2 % Engraftment.

Reporter cell assays BWZ cells were retrovirally transduced with CD3z/NKR-P1 constructs and sorted for GFP+ cells, as described (22, 24). Reporter cells (105 per well) were incubated with serial 3-fold titrations of stimulator cells overnight. Control reporter cells were stimulated with 10 ng/ml PMA and 0.5 mM ionomycin. Cells were washed, resuspended in 100 ml of 13 CPRG buffer (90 mg/L chlorophenol-red-b-D-galactopyranoside [Roche], 9 mM MgCl2, 0.1% NP-40, in PBS), incubated, then analyzed using a microplate reader (Varioskan) using OD595–655. DOD values represent specific BWZ. CD3z/NKR-P1B129 responses to stimulators following subtraction of parental BWZ responses to stimulators.

Chromium release assay Cytotoxicity assays were performed as described (22). Splenic lymphokineactivated killer (LAK) effector cells were cultured in 10% complete RPMI

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

described: NK subsets expressing an inhibitory Ly49 specific for either H-2d or H-2b alleles, a subset expressing Ly49 specific for both H-2d and H-2b alleles, and a subset expressing neither class of Ly49 receptor. Thus, parental (H-2b/b or H-2d/d) BM cells transplanted into an F1 (H-2b/d) host are perceived to be “missing” at least one self MHC-I allele by NK cell subsets rendered tolerant to non-overlapping allelic specificities. At the molecular level, F1 inhibitory Ly49 receptors that recognize only H-2b or H-2d alleles are not engaged; hence, missing-self recognition ensues, leading to selective rejection of parental BM grafts. Hybrid resistance is one of the earliest phenomena delineating a unique role for NK cells in determining BM transplant rejection (8, 12, 13). Further evidence came from observations in which allogeneic BM cells engraft in NK-depleted hosts (14–16) yet are rejected by C.B-17/ SCID recipients (17, 18) (which lack B and T lymphocytes but possess intact NK cell activity) (19). Nonetheless, MHC-independent NK recognition mechanisms have recently emerged, including the NKR-P1B:Clr-b receptor– ligand pair. In rats, viral evasion of this system determines the outcome of viral titers during rat CMV-English infection (RCMV-E; Mhv-8) (20, 21). Briefly, RCMV-infected cells lose surface expression of host Clr-b–like rat Clec2d11, yet RCMV-E encodes an NKR-P1B–specific decoy, RCTL, that functionally replaces Clec2d11 and inhibits NK cells. In addition, mouse tumor lines and stressed cells also lose expression of Clr-b (22, 23), thereby augmenting their susceptibility to NK cells. These findings broaden the significance of MHC-independent NK cell immunosurveillance in the discrimination of pathological versus healthy target cells. In this study, we investigated whether this system plays a nonredundant role in normal self–nonself discrimination during autologous and allogeneic hematopoietic transplants. Specifically, we assessed the selective rejection of Clr-b2/2 BM cells in comparison with wild-type (WT) and various MHC-I–deficient BM cells, using a competitive in vivo transplantation assay. We show that Clr-b2/2 BM cells are selectively and acutely rejected by B6 and allogeneic recipients in an NK cell–dependent manner, mediated specifically by the NKR-P1Bhi NK cell subset. Furthermore, Clr-b2/2 NK cells are functionally hyporesponsive, similar to MHC-I–deficient NK cells. These results support the importance of MHC-independent mechanisms in regulating hematopoietic transplant rejection and NK cell function in vivo.

The Journal of Immunology

2911

Table I. RT-PCR and qRT-PCR primers Primer Name

TAG TCC CAC TGA CGA CTC ATC TCC GCC GAG AGC AAT ctc gag ATG gcg gcc gc AGC TCC TCA AGG GGA GAT TGT GCA GTG TGA AGG GGT AGA GCC ACG CTG GGA AGC

AGG CAG CCC GC TCT GTG CAG GCC A CCT TCT GCC GA GCC AGC CCC GG AGT GCT GCA AAG GTT GA CTA TCC GGG ATT TAC AAT TA GCT CTG AGA TGT GTG GGT TCC GTG CCT TT CCA GCC TCG TC CGT TGA TGG CAA CA GAC AAC TGC GTT G CCA ACT TCT GCA C

1640 plus 2500 U/ml human rIL-2 (Proleukin; Novartis) for a total of 6 d. On day 4, these cells were FACS sorted for NKp46+ and NKR-P1B+ expression and used as effectors in 51Cr-release assays. Target cells (107/ml) were incubated with 100 mCi Na251CrO4 (PerkinElmer) in FCS at 37˚C for 1 h, washed, and plated (round-bottom) in serial titrations with effector cells (3–5 3 105 per microliter) at 37˚C for 4 h. Supernatants (100 ml) were transferred to scintillation plates (PerkinElmer), dried overnight, and counted using a Top Count NXT Microplate Scintillation Counter (Packard Instrument Company). Percent specific lysis values were calculated relative to maximum release (2% Nonidet P-40) and spontaneous release (media) values.

IFN-g assays Splenocytes were cultured in 96-well Immulon plates, with either platebound PK136 mAb or 5 ng/ml IL-18 (R&D Systems) plus IL-12 (PeproTech), for 5 h in the presence of GolgiStop and GolgiPlug (BD Biosciences). Cells were then stained for surface markers, fixed, permeabilized, and stained intracellularly for IFN-g.

Results

Phenotypic and functional characterization of Clr-b2/2 mice B6.Ocil/Clr-b2/2 (hereafter, Clr-b2/2) mice exhibit a mild defect in osteoclast development and function (26). These animals were

FIGURE 1. Phenotypic confirmation of the Clr-b2/2 mouse genotype. (A) Semiquantitative RT-PCR analysis of Clr-b transcript expression using primers specific for exons-2 through -4 for the coding sequence of Clr-b relative to GAPDH as an internal loading control. The smaller fragment observed for Clr-b2/2 mice corresponds to a transcript deleted for exon-3, which creates a frame-shift resulting in the absence of a functional ectodomain. (B) Flow cytometric analysis of Clr-b surface expression using 4A6 anti–Clr-b mAb on cells derived from WT B6 versus Clr-b2/2 mice. Shown are BM, lymph node (LN), spleen (SP), thymus (TH), and liver (LV). Viable cells were gated by forward/side scatter, followed by DAPI exclusion. FMO, fluorescence minus one; RCN, relative cell number. (C) BWZ.CD3z/NKR-P1B reporter cell analysis of Clr-b ligand function on ex vivo BM, spleen, and liver cells from WT B6 (d) or Clr-b2/2 mice. Clr-b2/2 cells (s) lack NKR-P1B–specific ligand function. Each data point represents the average value of three replicates (6SEM).

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

Clr-b-F(e2) Clr-b-R(e4) GAPDH-F GAPDH-R FL-Clr-b-F(e1) FL-Clr-b-R(e5) qClr-bF qClr-bR qGAPDH-F qGAPDH-R qTBP-F qTBP-R

Sequence (59–39)

generated directly on a B6 background (verified by Illumina genotyping) by targeted deletion of exon-3 of the Clec2d gene; this leads to a frame-shifted transcript and the absence of a functional ectodomain (26). We confirmed the deficiency of these Clr-b2/2 animals via three approaches: 1) RT-PCR analysis of Clr-b transcripts, using multiple primers (Table I); 2) flow cytometric analysis of Clr-b surface protein, using 4A6 mAb (22, 29); and 3) reporter analysis of Clr-b ligand function, using BWZ.CD3z/ NKR-P1B cells (22, 29). Mutant Clr-b transcripts were detected by RT-PCR and qRT-PCR in all Clr-b2/2 tissues analyzed (Fig. 1A and data not shown). Most tissues containing hematopoietic cells expressed Clr-b transcripts, with nearly undetectable levels in brain (data not shown). However, as expected, transcripts from Clr-b2/2 mice were smaller than WT Clr-b transcripts, and sequencing of Clr-b amplification products confirmed exon-3 deletion in Clr-b2/2 mice. Nonetheless, transcriptional regulation of the mutant Clrb (Ocil/Clec2d) gene appears to be intact in Clr-b2/2 mice. To confirm a lack of Clr-b surface protein, we evaluated Clr-b expression on cells from B6 and Clr-b2/2 tissues ex vivo, using Clr-b mAb, 4A6 (22). As shown in Fig. 1B, Clr-b2/2 cells lacked detectable Clr-b surface protein, in turn validating the lack of cross-reactivity of 4A6 mAb with other Clr family members. To confirm a loss of Clr-b ligand function, ex vivo Clr-b2/2 cells were used as stimulator cells in BWZ.CD3z/NKR-P1B reporter cell assays. Here, the cognate NKR-P1B receptor recognized intact Clr-b ligand on WT cells, but not Clr-b2/2 cells (Fig. 1C). In turn, the absence of BWZ.CD3z/NKR-P1B reporter stimulation by Clr-b2/2 cells confirms the specificity of NKR-P1B for Clr-b, and suggests that Clr-b is the only ligand recognized by NKR-P1B on normal B6 cells ex vivo. Taken together, these data demonstrate a lack of Clr-b surface protein and a loss of NKR-P1B ligand function on Clr-b2/2 cells.

2912

ROLE OF Clr-b IN HEMATOPOIETIC TRANSPLANT REJECTION

Steady-state Clr-b2/2 hematopoietic characterization

subset-restricted expression of NKR-P1B was examined in greater detail with respect to that of the self–MHC-I–specific B6 strain NK cell receptors Ly49C/I and NKG2A, on splenic and BM NK cells from WT and Clr-b2/2 mice (Fig. 2C, 2D; Supplemental Fig. 1C, 1D). Of note, NKR-P1B was enriched on self–MHC-I– specific (“educated”) NK cells expressing either Ly49C/I or NKG2A. Moreover, gating on NKR-P1B– NK cells revealed an enrichment of the classical “hyporesponsive” subset lacking expression of Ly49C/I and NKG2A; notably, however, a significant fraction of this subset did express NKR-P1B (∼30–50%), suggesting these NK cells are indeed educated and not entirely hyporesponsive, at least to MHC-independent signals. Only subtle differences were noted in the NK repertoires between WT and Clr-b2/2 mice; specifically, fewer NKR-P1B+ cells among the hyporesponsive Ly49C–/I–/NKG2A– subset, and a corresponding increase among the Ly49C/I+ subsets, were observed among Clr-b2/2 NK cells (Fig. 2C, 2D; Supplemental Fig. 1C, 1D). Similar observations were observed for NK subsets grouped according to CD94 (i.e., NKG2A/C/E) in lieu of NKG2A (Supplemental Fig. 2A–D). Interestingly, NKR-P1B+ NK cells were almost entirely 2B4+, suggesting a cosegregation of NK cells responsive to MHCindependent ligands (Supplemental Fig. 2E–H). Notably, trends toward higher expression of inhibitory NK receptors (e.g., NKR-P1B, Ly49) and lower expression of stimu-

FIGURE 2. Characterization of steady-state NK cell subsets in the Clr-b2/2 mouse. (A) NK1.1+CD3– splenocytes harvested from either B6 or Clr-b2/2 animals were gated and assessed for NK cell subset frequency and receptor expression levels. (B) Quantitation of NK cell receptor expression levels from splenic NK cells. All comparisons between B6 and Clr-b2/2 mice were analyzed with the two-tailed t test. In all histograms, empty bars represent B6 and solid bars represent Clr-b2/2 NK cells. *p , 0.05, **p , 0.01. (C) Subset-restricted expression of NKG2A and Ly49C/I on splenic NK cells (gated NK1.1+/ CD3–) from B6 and Clr-b2/2 mice, further analyzed for NKR-P1B levels (histogram overlays, below) for each indicated subset. Numbers represent cell percentages (upper histogram gate, B6; lower gate, Clr-b2/2); shaded gray histograms, B6; black lines, Clr-b2/2. (D) Alternative gating strategy based upon NKR-P1B levels (left plot) for subsetting of splenic NK cells from B6 and Clr-b2/2 mice, showing NKG2A and Ly49C/I expression on NKR-P1B+/2 subsets (right plots). All data are representative of at least three independent mice from each genotype. DN, double-negative, hyporesponsive subset; DP, doublepositive, educated, total SP + DP subsets; MFI, median fluorescence intensity; SP, single-positive.

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

We next characterized Clr-b2/2 mice in comparison with WT and MHC-I–deficient animals, such as B6.b2m2/2 mice (30); although these latter mice are deficient in CD8+ T cells (31), their NK cells are functional (32), albeit hyporesponsive (33, 34). Because Clr-b expression resembles that of MHC-I, we investigated the influence of Clr-b absence during hematopoietic development. Various Clr-b2/2 tissues were compared with age- and sexmatched WT tissues for both lymphoid and myeloid cell composition. A nearly identical leukocyte distribution (neutrophils, macrophages, dendritic cells, B cells, T cells, NKT cells, and NK cells) was observed (data not shown). This result might be expected, as Clr-b is the only known ligand for NKR-P1B (also known as NKR-P1D in B6 mice) (22, 28). We further characterized the Clr-b2/2 NK cell subset composition and receptor repertoire. Of interest, NKR-P1BB6 expression levels (i.e., NKR-P1D, assessed using 2D12 mAb) (28) were significantly higher on Clr-b2/2 NK cells versus WT NK cells (Fig. 2A, 2B; Supplemental Figs. 1–3). This was expected because elevated Ly49 levels are observed on NK cells from MHC-I–deficient mice (35). Somewhat unexpectedly, NKp46 levels were significantly lower on Clr-b2/2 NK cells versus WT NK cells; similar trends were also observed for other stimulatory receptors (Fig. 2A 2B; Supplemental Figs. 1A, 1B, 1E, 1F, 3). Next, the

The Journal of Immunology latory receptors (e.g., NKp46, NKR-P1C, 2B4, NKG2D, CD94+/ NKG2A–) were observed among NK cells from Clr-b2/2 mice (Supplemental Fig. 3 and data not shown). These shifts in signal balance suggest that Clr-b2/2 NK cells may be hyporesponsive; conversely, WT NK cells may selectively reject Clr-b2/2 hematopoietic grafts owing to diminished inhibition via NKR-P1B. Acute NK cell–mediated rejection of Clr-b2/2 hematopoietic grafts in WT recipients

various MHC-I–deficient BM donors were rejected differentially (Supplemental Fig. 4), similar to observations reported in previous studies (36). Notably, Clr-b2/2 BM cells were selectively rejected in WT recipients, roughly comparable to single MHC-I–deficient H-2Db2/2 BM cells (Fig. 3B; Supplemental Fig. 4). Moreover, NK depletion reversed the rejection of Clr-b2/2 and MHC-I–deficient BM cells (Fig. 3B). In addition, Clr-b2/2 and MHC-I–deficient splenocyte rejections mirrored the patterns observed for BM rejections, perhaps with a slightly weaker contribution of Clr-b versus H-2Db (Fig. 3C). Thus, Clr-b2/2 hematopoietic cells are selectively rejected in WT mice in an NK-dependent manner, similar in magnitude or slightly weaker than single MHC-I–deficient (H-2Db2/2) cells. Notably, Clr family members other than Clr-b may play similar roles in transplant tolerance. Acute rejection of allogeneic Clr-b2/2 BM cells mediated by NKR-P1Bhi NK cells To assess whether selective rejection of Clr-b2/2 BM cells was mediated by the NKR-P1B+ NK cell subset, we used allogeneic NIH Swiss recipients, in which PK136 mAb treatment selectively depletes the NKR-P1Bhi/+ subset, leaving the remaining NKRP1Blo/– NK cells intact (37–39). Allogeneic (B6 strain) mixtures of WT, Clr-b2/2, and b2m2/2 BM cells were again stoichiometrically transplanted into cohorts that were either untreated or PK136 treated. In this case, depletion of NKR-P1Bhi/+ NK cells is expected to abrogate selective rejection of Clr-b2/2 versus WT (B6) BM cells, whereas b2m2/2 BM cells should be rejected by the remaining NKR-P1Blo/– NK cells. Indeed, in these allogeneic recipients, both Clr-b2/2 and b2m2/2 BM cells were again se-

FIGURE 3. Clr-b2/2 BM cells (BMC) are acutely rejected in an NK cell–dependent manner by WT B6 host mice. (A) Rejection of Clr-b2/2 BM cells from B6 mice 48 h after transplantation. Mice shown in the figure were treated with poly(I:C) or PK136 mAb 24 h before BM transplantation. Symbols represent data from individual mice, and small horizontal bars represent mean values. Statistics were analyzed via an unpaired t test. (B) Rejection of Clr-b2/2 and various MHC-I–deficient BM cells 48 h after transplantation. Recipient mice were either untreated or treated with PK136 mAb 24 h prior to BM transplantation. Histograms show the average rejection of three recipients. Data are representative of five independent transplantation experiments. (C) Rejection of Clr-b2/2 and various MHC-I–deficient splenocytes 48 h after splenocyte infusion. Recipient mice were either untreated or treated with PK136 mAb 24 h prior to splenocyte transplant. Histograms show the average rejection of three recipients. Data are representative of two independent experiments. Data from (B) and (C) were analyzed via two-way ANOVA; the rejection phenotypes between NK-sufficient and NK-deficient recipients were assessed by the Bonferroni posttest. *p , 0.05, **p , 0.01, ***p , 0.001.

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

To assess MHC-independent missing-self rejection of Clr-b2/2 donor cells, we established short-term (18–48 h) competitive transplant assays using equal mixtures of fluorophore-labeled (or congenically marked) Clr-b2/2 (CD45.2) versus WT (or CD45.1) donor hematopoietic cells into sublethally irradiated WT (or CD45.1/2) recipient mice. NK cell activity was assessed using NK-depleted (PK136-treated) versus untreated or NK-primed (polyI:C-treated) cohorts. As shown in Fig. 3A, Clr-b2/2 BM cells, relative to WT donor cells, were acutely and selectively rejected by WT recipients. This rejection was mitigated in NKdepleted mice (Fig. 3A; p = 0.0013). Thus, Clr-b plays a nonredundant MHC-independent role in the recognition of normal BM cells by NK cells. To establish a rejection hierarchy, donor Clr-b2/2 BM cells were competitively transplanted with WT and MHC-deficient (H-2Kb2/2, H-2Db2/2, H-2Kb2/2Db2/2, b2m2/2) BM cells. Donor cohorts were labeled using different combinations of three fluorescent dyes, then transplanted stoichiometrically into untreated or NK-depleted recipients (see Materials and Methods; Table II; Supplemental Fig. 4). Following transplantation (18–48 h), the

2913

2914

ROLE OF Clr-b IN HEMATOPOIETIC TRANSPLANT REJECTION

Table II. Dye label combinations used for competitive hematopoietic transplants Genotype 2/2

b2m Clr-b2/2 C57BL/6 H-2Db2/2 H-2Kb2/2 H-2Kb2/2 H-2Db2/2

PKH67

PKH26

CellVue Maroon

+ – – + – +

– + – + + –

– – + – + +

Hematopoietic cells from the indicated mouse genotypes were singly or combinatorially labeled using the dye patterns indicated, then stoichiometrically pooled prior to transplant. Labeled cells were analyzed via flow cytometry.

Rejection of Clr-b2/2 hematopoietic cells is abrogated in NKR-P1BB62/2 recipients To further confirm the specificity of rejection in syngeneic mice, we transplanted Clr-b2/2 hematopoietic cells into NKR-P1BB62/2 recipients (27). Notably, Clr-b2/2 BM cells and splenocytes were selectively rejected in WT (Nkrp1bB6+/+) littermates, whereas rejection was mitigated in Nkrp1bB62/2 recipients, which are fully NK sufficient but lack the cognate NKR-P1B receptor (27)

NKR-P1B function in the absence of Clr-b during NK cell education Because we observed elevated NKR-P1B levels on Clr-b2/2 NK cells versus WT NK cells, we investigated NKR-P1B receptor function in the absence of Clr-b during NK cell education. In this study, tumor variants (BWZ.Clr-b+ and parental BWZ– cells) were used as targets for sorted NKR-P1B+ WT or Clr-b2/2 splenic LAK effectors in 51Cr-release assays. As shown in Fig. 5A, surface Clr-b on BWZ.Clr-b target cells inhibited cytotoxicity mediated by both WT and Clr-b2/2 LAK effectors, relative to control BWZ– targets, an effect blocked using Clr-b mAb (4A6), but not isotype control mAb. We next performed redirected inhibition assays using NKRP1BB6 mAb (2D12) (28) to ligate the receptor on WT or Clr-b2/2 NK-LAK effector cells (sorted NKp46+NKR-P1BB6+) via the FcR on P815 target cells. Notably, P815 targets were susceptible to NKLAK cells from both genotypes; however, 2D12 mAb (but not isotype control mAb) decreased cytotoxicity of P815 targets via NKR-P1B–mediated redirected inhibition on both WT and Clr-b2/2 effectors (Fig. 5B). Notably, redirected inhibition was not observed using sorted NKR-P1B– NK-LAK effectors (data not shown). These data confirm the intact inhibitory function of NKR-P1B, even during Clr-b2/2 NK cell education, and suggest that Clr-b2/2 NK cells may possess altered responsiveness to some tumor targets.

FIGURE 4. Acute rejection of Clr-b2/2 BM cells (BMC) in allogeneic, but not Nkrp1b-deficient syngeneic, recipient mice. Rejection of (A) Clr-b2/2 BM cells and (B) b2m2/2 BM cells by NIH Swiss recipients, 48 h after transplantation. Mice were either untreated or treated with PK136 mAb 24 h prior to BM transplantation. Symbols represent data from individual mice; small horizontal bars represent mean values (6SEM). Data were analyzed via two-way ANOVA; the BM rejection phenotypes between untreated and NK1.1 (NKR-P1BSw)–depleted recipients were assessed by the Bonferroni posttest. (C) Rejection of Clr-b2/2 hematopoietic cells [(HPC), splenocytes, BM cells] by WT and Nkrp1b-deficient B6 recipients, 48 h after transplantation. Mice were treated with 150 mg polyI:C 24 h prior to transplantation. (D) Rejection of Clr-b2/2 BM cells by WT and Nkrp1b-deficient recipients, 18 h after transplantation. Mice were treated with either 150 mg polyI:C (i.p.) or 5 mg CpG ODN (i.v.) 24 h or 6 h prior to transplantation. *p , 0.05, **p , 0.01, ***p , 0.001.

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

lectively rejected (versus B6 WT cells), whereas elimination of the NKR-P1Bhi/+ NK subset (PK136 depletion) reversed the rejection of Clr-b2/2 donor cells (Fig. 4A; p , 0.05), but not b2m2/2 BM cells (Fig. 4B). This finding suggests that rejection of Clr-b2/2 BM cells is mediated specifically by cognate NKR-P1Bhi/+ NK cells, in the absence of which, Clr-b2/2 BM grafts are tolerated. In contrast, residual NKR-P1Blo/– NK cells in PK136-treated NIH Swiss recipients are sufficient to mediate rejection of b2m2/2 BM cells.

(Fig. 4C). Importantly, rejection of Clr-b2/2 hematopoietic cells was not observed in Nkrp1b B62/2 recipients even following in vivo NK cell priming using polyI:C or CpG-B ODN (Fig. 4D), which induce high levels of innate cytokines and augment Clr-b2/2 rejection in WT recipients. These results confirm the cognate interaction between NKR-P1BB6 and Clr-b, and further support the allelic nature of the NKR-P1B/D receptors.

The Journal of Immunology

Clr-b2/2 NK cells are hyporesponsive to cytokine priming and NKR-P1C stimulation To assess whether Clr-b2/2 NK cells display a generalized hyporesponsive phenotype, similar to that of MHC-I–deficient NK cells, we stimulated splenic NK cells from WT and Clr-b2/2 mice, using plate-bound NK1.1 (NKR-P1CB6, PK136) mAb or cytokines, then measured intracellular IFN-g production. Clr-b2/2 NK cells were hyporesponsive to NKR-P1CB6 mAb-mediated crosslinking (Fig. 6A), perhaps in part owing to reduced NK1.1 expression (see Supplemental Fig. 3). Similar results were observed in response to IL-12/IL-18 cytokine priming (Fig. 6C). Collectively, relative to WT NK cells, Clr-b2/2 NK cells display a hyporesponsive phenotype to NK1.1 stimulation (0.7-fold) and IL-12 cytokine priming (0.6-fold), similar to the phenotype observed for MHC-I–deficient NK cells. Of interest, differences between the NKR-P1Bhi/+ and NKR-P1Blo/– NK cell subsets were not observed (Fig. 6B, 6D), suggestive of a generalized hyporesponsiveness of Clr-b2/2 NK cells, one that is not NKR-P1B subset restricted. Clrb2/2 NK cells are self-tolerant despite differential NKR-P1F ligand expression To examine self-tolerance, Clr-b2/2 mice were used as recipients of WT, Clr-b2/2, b2m2/2, and b2m2/2Clr-b2/2 BM cells. Because the recipient genotype was being varied, results were plotted as

percent engraftment relative to autologous (identical) BM cells. As a control, WT mice transplanted with the same BM mixtures are shown in Fig. 7A (plotted relative to WT grafts). As predicted by intact NKR-P1B function in Clr-b2/2 recipients, WT BM cells engrafted slightly better than Clr-b2/2 BM cells (autologous graft), whereas b2m2/2 BM cells were still rejected (Fig. 7B). Of interest, b2m2/2 BM cells engrafted slightly better than Clr-b2/2b2m2/2 BM cells in Clr-b2/2 recipients (Fig. 7B). Thus, MHC-I–dependent and MHC-independent NK cell tolerance mechanisms are nonredundant and regulated by distinct yet overlapping ligand expression. Because Clr-b is one of many Clr family members, Clr-b2/2 cells may possess differential expression of other Clr ligands, which may affect the balance of recognition by other NKR-P1 or NK cell receptors. Notably, Clr-b2/2 mice display decreased NKp46 and NKR-P1C levels, and NKR-P1F/G share overlapping Clr ligand specificities (22, 28). To assess this, ex vivo WT or Clr-b 2/2 splenocytes were used as stimulator cells for BWZ.CD3z/NKR-P1(A,C,F,G) reporter cells. Of interest, Clr-b2/2 cells significantly stimulated BWZ.CD3z/NKR-P1F reporter cells, relative to WT stimulators (Fig. 7C). In contrast, no stimulation was observed for BWZ.CD3z/NKR-P1(A,C,G) reporters (data not shown). This finding suggests that Clr-b2/2 cells may display an alternative surface Clr ligand for NKR-P1F. Because Clr-g transcripts are expressed in hematopoietic cells (29, 40–43), antagonistic Clr-b/g heterodimers are possible; alternatively, the absence of Clr-b could unmask another Clr ligand, such as Clr-c (NKRP1F, but not NKR-P1G ligand), in turn modulating NKR-P1F receptor expression or function during NK cell development and education. The same may be true for NKR-P1C, which is slightly downregulated on Clr-b2/2 NK cells; however, NKR-P1C ligands remain unknown to date.

Discussion The importance of NK cells in determining the outcome of BM transplantation is highlighted by hybrid resistance experiments (5– 8) and BM allograft acceptance in NK-depleted, but not NKsufficient, SCID mice (14–19). These findings have since been explained by missing-self recognition of MHC-I molecules, which are highly polymorphic. MHC-I polymorphisms in turn have influenced the convergent evolution of two highly diversified receptor families (Ly49 in mice, KIR in humans), which also exhibit multiple polymorphisms at both the allelic and gene content levels (44). In contrast, fewer yet significant allelic polymorphisms exist in the mouse Nkrp1:Clr system (30, 44), in which the overall haplotype structure appears to be conserved (44, 45). Interestingly, among the Clr family, Clr-b possesses an expression pattern similar to that of MHC-I, yet demonstrates no allelic polymorphisms to date between evaluated mouse strains (B6, BALB, 129) (22). Because loss of Clr-b has been documented in tumor, infected, and stressed cell pathological states (21, 23), we investigated the role of this MHC-independent missing-self recognition mechanism in the recognition of normal, healthy cells, using Clr-b– deficient mice and a competitive in vivo BM transplantation model (26, 41, 42, 46). The Ocil/Clr-b2/2 mice were previously generated by targeted deletion of exon-3 of Clec2d and display a mild defect in osteogenesis (26). This mutation results in a smaller frame-shifted but expressed transcript. Notably, data in our laboratory reveal that some tumor lines possess endogenous exon-3–deleted Clr-b transcripts (C.L. Kirkham, unpublished observations), suggesting alternative splicing may occur naturally or in cancer. We observed that Clr-b2/2 (but not Clr-b+/2) cells lack surface Clr-b protein (detected by 4A6 mAb) and functional NKR-P1B ligand (assessed by BWZ.CD3z/NKR-P1B reporters). These data suggest

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

FIGURE 5. The NKR-P1B receptor is functional on Clr-b2/2 NK cells. (A) 51Cr-release cytotoxicity assays using BWZ (Clr-b–) and BWZ.Clr-b (Clr-b+) cells as targets for sorted NKR-P1B+ WT B6 or Clr-b2/2 splenic LAK effector cells in the presence of either isotype control mAb (rIgM, top) or Clr-b mAb (4A6, bottom). (B) 51Cr-release cytotoxicity assays using P815 (FcR+) cells loaded with NKR-P1B mAb (2D12) or isotype control mAb (mIgG2a) as targets for sorted NKR-P1B+ splenic LAK cells in Ab-induced redirected inhibition assays. Data are shown as mean 6 SEM. All data are representative of at least three independent experiments. Experiments were analyzed using ANOVA with Bonferroni posttest. *p , 0.05, **p , 0.01, ***p , 0.001.

2915

2916

ROLE OF Clr-b IN HEMATOPOIETIC TRANSPLANT REJECTION

that Clr-b is the only constitutive ligand recognized by NKR-P1B. However, other developmentally regulated or activation-induced ligands or antagonists may exist. The steady-state immune and hematopoietic composition of Clr-b2/2 mice showed no major differences. In addition, we observed few significant differences in NK cell receptor expression. However, significantly higher expression of the inhibitory NKR-P1B receptor was observed on BM, splenic, and hepatic NK cells from Clr-b2/2 animals, along with significantly lower expression of stimulatory NKp46 receptor. Other trends were observed toward decreased expression of stimulatory NK receptors (NKR-P1C, 2B4, NKG2D, CD94/NKG2C/E) and perhaps slightly increased expression of inhibitory NK receptors (Ly49, CD94/NKG2A). These changes in balanced signaling are expected to elicit hyporesponsiveness. Elevated Ly49 receptor expression has been observed on NK cells from b2m2/2 mice, which lack MHC-I ligands (35). Mechanistically, elevated NKR-P1B expression is likely posttranslational and due to a lack of receptor ligation or internalization following interaction with cognate ligand (on NK cells in cis or on neighboring cells in trans) (47, 48). Furthermore, increased NKR-P1B on Clr-b2/2 NK cells may occur during education, as developing NK cells integrate stimulatory and inhibitory signals. To achieve balance, developing Clr-b2/2 NK cells may augment inhibitory NKR-P1B expression and/or decrease that of stimulatory receptors. As highlighted above, Clr-b2/2 NK cells exhibit reduced NKp46 receptor levels. NKp46 is reported to recognize an unknown self-ligand (49); thus, lowering NKp46 expression may reduce the net threshold of activation to compensate during NK cell education. Acute rejection of MHC-I–deficient hematopoietic grafts by NK cells has been widely documented (32, 36). We developed a competitive assay to assess Clr-b2/2 graft rejection relative to control WT and MHC-I–deficient grafts in a single mouse. This assay employs three fluorophores with distinct spectra to enable differential labeling of seven donor populations. This approach is indexed and internally controlled, and inherently conserves recipient animals and reagents, in addition to reducing variability, as competitive rejections of all donors occur simultaneously. As

expected, MHC-I–deficient BM grafts (b2m2/2, Kb2/2Db2/2, Kb2/2, Db2/2) were rejected to different degrees by WT NK cells. Notably, b2m2/2 and Kb2/2Db2/2 grafts were rejected vigorously because they lack expression of all MHC-I proteins and Qa-1b; in addition, b2m2/2 mice lack expression of CD1d and many other

FIGURE 7. In vivo NK cell tolerance and differential NKR-P1F ligand expression in Clr-b2/2 mice. (A) Percent engraftment of donor B6 WT, Clr-b2/2, b2m2/2, and b2m2/2Clr-b2/2 BM cells (BMC) relative to autologous (WT) donor grafts in B6 WT recipients (n = 5). (B) Percent engraftment of donor B6 WT, Clr-b2/2, b2m2/2, and b2m2/2Clr-b2/2 BM cells relative to autologous (Clr-b2/2) donor grafts in Clr-b2/2 recipients (n = 5). (C) BWZ reporter cells bearing CD3z/NKR-P1F fusion receptors were mixed with titrated doses of ex vivo splenocyte stimulator cells harvested from either B6 (d) or Clr-b2/2 (s) mice.

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

FIGURE 6. Clr-b2/2 splenic NK cells are hyporesponsive. (A) B6 (d) and Clr-b2/2 (s) splenocytes were stimulated with plate-bound anti-NK1.1, and the frequencies of IFN-g–producing NK cells were determined by flow cytometry. (B) Frequencies of IFN-g–producing NK cells among the NKR-P1Bhi/+ (closed symbols) and NKR-P1Blo/– (open symbols) subsets from B6 (d, s) and Clr-b2/2 (n, N) mice upon anti-NK1.1 plate-bound stimulation. (C) Splenocytes from B6 (d) and Clr-b2/2 (s) mice were stimulated in vitro with various amounts of IL-12 in the presence of IL-18 (5 ng/ml). The frequencies of IFN-g–producing NK cells are shown. (D) Frequencies of IFN-g–producing NK cells among the NKR-P1Bhi/+ (closed symbols) and NKR-P1Blo/– (open symbols) subsets from B6 (d, s) and Clr-b2/2 (n, N) mice upon IL-12/IL-18 stimulation. In all panels, frequencies of IFN-g–producing NK cells were determined by gating on NKp46+CD3– cells, followed by gating on IFN-g+ splenocytes. In (A)–(D), n = 3–4 6 SEM. *p , 0.05, **p , 0.01, ***p , 0.001, calculated by two-way ANOVA with Bonferroni correction.

The Journal of Immunology

tion of other Clr (e.g., Clr-c/d/g dimers, via NKR-P1F). The Clr homo/heterodimerization model may further our understanding about the molecular interactions that occur at the NK cell:target synapse during cellular pathological states, via integration of the missing-self and induced-self recognition axes. It remains to be determined if homologous receptor–ligand interactions in humans, including NKR-P1A:LLT1 (encoded by KLRB1:CLEC2D) polymorphisms, play similar roles in NK cell recognition and rejection of normal syngeneic and allogeneic hematopoietic cell grafts.

Acknowledgments We thank Drs. Koho Iizuka and Wayne M. Yokoyama for generously providing the 2D12 purified mAb and hybridoma.

Disclosures The authors have no financial conflicts of interest.

References 1. Prochazka, M., H. R. Gaskins, L. D. Shultz, and E. H. Leiter. 1992. The nonobese diabetic scid mouse: model for spontaneous thymomagenesis associated with immunodeficiency. Proc. Natl. Acad. Sci. USA 89: 3290–3294. 2. Goldman, J. M., R. P. Gale, M. M. Horowitz, J. C. Biggs, R. E. Champlin, E. Gluckman, R. G. Hoffmann, S. J. Jacobsen, A. M. Marmont, P. B. McGlave, et al. 1988. Bone marrow transplantation for chronic myelogenous leukemia in chronic phase. Increased risk for relapse associated with T-cell depletion. Ann. Intern. Med. 108: 806–814. 3. Asai, O., D. L. Longo, Z. G. Tian, R. L. Hornung, D. D. Taub, F. W. Ruscetti, and W. J. Murphy. 1998. Suppression of graft-versus-host disease and amplification of graft-versus-tumor effects by activated natural killer cells after allogeneic bone marrow transplantation. J. Clin. Invest. 101: 1835–1842. 4. Ferrara, J. L. M., H. J ., and Deeg. 1991. Graft-versus-host disease. N. Engl. J. Med. 324:667–674. 5. Cudkowicz, G., and J. H. Stimpfling. 1964. Induction of immunity and of unresponsiveness to parental marrow grafts in adult F-1 hybrid mice. Nature 204: 450–453. 6. Cudkowicz, G., and J. H. Stimpfling. 1964. Hybrid resistance to parental marrow grafts: association with the K region of H-2. Science 144: 1339–1340. 7. Cudkowicz, G., and J. H. Stimpfling. 1964. Deficient growth of C57bl marrow cells transplanted in F1 hybrid mice: association with the histocompatibility-2 locus. Immunology 7: 291–306. 8. Kumar, V., T. George, Y. Y. A. Yu, J. Liu, and M. Bennett. 1997. Role of murine NK cells and their receptors in hybrid resistance. Curr. Opin. Immunol. 9: 52–56. 9. Raulet, D. H., W. Held, I. Correa, J. R. Dorfman, M. F. Wu, and L. Corral. 1997. Specificity, tolerance and developmental regulation of natural killer cells defined by expression of class I-specific Ly49 receptors. Immunol. Rev. 155: 41–52. 10. Raulet, D. H., and R. E. Vance. 2006. Self-tolerance of natural killer cells. Nat. Rev. Immunol. 6: 520–531. 11. Raulet, D. H., R. E. Vance, and C. W. McMahon. 2001. Regulation of the natural killer cell receptor repertoire. Annu. Rev. Immunol. 19: 291–330. 12. Bennett, M. 1987. Biology and genetics of hybrid resistance. Adv. Immunol. 41: 333–445. 13. Waterfall, M., L. S. Rayfield, and L. Brent. 1984. Abrogation of resistance to bone marrow transplantation by induction of specific tolerance in natural killer cells? Nature 311: 663–665. 14. Murphy, W. J., V. Kumar, and M. Bennett. 1987. Rejection of bone marrow allografts by mice with severe combined immune deficiency (SCID). Evidence that natural killer cells can mediate the specificity of marrow graft rejection. J. Exp. Med. 165: 1212–1217. 15. Murphy, W. J., V. Kumar, and M. Bennett. 1987. Acute rejection of murine bone marrow allografts by natural killer cells and T cells. Differences in kinetics and target antigens recognized. J. Exp. Med. 166: 1499–1509. 16. Tiberghien, P., D. L. Longo, J. W. Wine, W. G. Alvord, and C. W. Reynolds. 1990. Anti-asialo GM1 antiserum treatment of lethally irradiated recipients before bone marrow transplantation: evidence that recipient natural killer depletion enhances survival, engraftment, and hematopoietic recovery. Blood 76: 1419–1430. 17. Bosma, G. C., R. P. Custer, and M. J. Bosma. 1983. A severe combined immunodeficiency mutation in the mouse. Nature 301: 527–530. 18. Fulop, G. M., and R. A. Phillips. 1990. The scid mutation in mice causes a general defect in DNA repair. Nature 347: 479–482. 19. Dorshkind, K., S. B. Pollack, M. J. Bosma, and R. A. Phillips. 1985. Natural killer (NK) cells are present in mice with severe combined immunodeficiency (scid). J. Immunol. 134: 3798–3801. 20. Ettinger, J., H. Geyer, A. Nitsche, A. Zimmermann, W. Brune, G. R. Sandford, G. S. Hayward, and S. Voigt. 2012. Complete genome sequence of the English isolate of rat cytomegalovirus (Murid herpesvirus 8). J. Virol. 86: 13838. 21. Voigt, S., A. Mesci, J. Ettinger, J. H. Fine, P. Chen, W. Chou, and J. R. Carlyle. 2007. Cytomegalovirus evasion of innate immunity by subversion of the NKRP1B:Clr-b missing-self axis. Immunity 26: 617–627.

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

cell surface molecules (50). Therefore, these transplanted grafts promote disinhibition of both Ly49+ and NKG2/CD94+ NK subsets, which synergize to mediate rejection (51). Functionally relevant B6 NK subsets include those expressing Ly49C/I and NKG2A/CD94 (52). In contrast, single MHC-I–deficient grafts (Kb2/2, Db2/2) exhibited reduced rejection compared with double-deficient Kb2/2Db2/2 grafts, as shown previously (36). The hierarchically comparable rejection of Clr-b2/2 BM cells to H-2Db2/2 BM cells was quite striking because both H-2Db and Clr-b are single loci, and ∼60% of NK cells express NKR-P1B (48). In contrast, many overlapping NK subsets are simultaneously disinhibited by full MHC-I–deficient grafts, resulting in robust rejection relative to disinhibition of the NKR-P1Bhi/+ NK subset alone (which accounts for ∼30–40% of acute BM graft rejection). Data generated using allogeneic NIH Swiss recipients further strengthen this claim. b2m2/2 grafts were rejected ∼95%, whereas Clr-b2/2 grafts were rejected up to ∼50%. This difference is likely attributed to enhanced recognition of Clr-b by NKR-P1BSw versus NKR-P1BB6, compounded with allogeneic Ly49:MHC-I interactions. Importantly, selective depletion of NKR-P1Bhi/+ NK cells reversed rejection of Clr-b2/2 grafts, as did genetic NKR-P1B2/2 deficiency, whereas b2m2/2 graft rejection was not affected. These findings support an important and nonredundant role for MHC-independent missing-self recognition of Clr-b by NKRP1B+ NK cells in hematopoietic transplants. The absence of donor MHC-I, Clr-b, or both results in poorer prognosis for BM transplant engraftment in WT recipients. In addition, Clr-b2/2 recipients revealed a trend toward increased engraftment of WT versus autologous Clr-b2/2 grafts, as well as b2m2/2versus Clr-b2/2 b2m2/2 double-deficient grafts. Because Clr-b2/2 NK cells exhibit elevated NKR-P1B levels, donor Clr-b may provide a static inhibitory signal to Clr-b2/2 recipient NK cells, conferring increased engraftment. The cytotoxicity assay results demonstrated that NKR-P1B is indeed functional and inhibitory on Clr-b2/2 NK cells. In addition, the IFN-g stimulation assays demonstrated that Clr-b2/2 NK cells (both NKR-P1Bhi/lo subsets) are generally hyporesponsive to NKR-P1C crosslinking and IL-12 cytokine priming. This generalized hyporesponsiveness may be due to differential NK cell education, which may in turn impact stimulatory receptor expression (NKp46 etc.). Finally, previous unpublished data suggest that Clr-b may heterodimerize with other Clr, such that loss of surface Clr-b may promote expression of alternative Clr dimers. We showed that NKR-P1F, a putative costimulatory NK receptor, displays enhanced recognition of ex vivo Clr-b2/2 hematopoietic cells. Thus, missing Clr-b may promote enhanced dimerization of NKR-P1F ligands (e.g., Clr-c, d, g) (29). This induced-self recognition of Clr dimers by NKR-P1F+ NK cells may also partially explain the increased rejection of Clr-b2/2b2m2/2 double-deficient versus b2m2/2 grafts in Clr-b2/2 recipients. Notably, NKR-P1F/G share some overlapping (and some distinct) ligand specificities, suggestive of balanced or integrated recognition. Alternatively, it is also possible that Clr family members may engage in both cis/ trans interactions with their cognate receptors (or one another), similar to what has been shown for MHC-I ligands and Ly49 receptors (53–56). These results suggest a model in which Clr-b expression may modulate surface expression of other Clr ligands. Under normal conditions, Clr-b may form both homodimers, which inhibit NK cells via NKR-P1B, and heterodimers with other Clr, which may not be recognized. Loss of Clr-b during stress, transformation, or infection may promote both missing-self recognition of Clr-b homodimers (via NKR-P1B), as well as induced-self recogni-

2917

2918

ROLE OF Clr-b IN HEMATOPOIETIC TRANSPLANT REJECTION 38. Carlyle, J. R., A. Martin, A. Mehra, L. Attisano, F. W. Tsui, and J. C. Zu´n˜igaPfl€ucker. 1999. Mouse NKR-P1B, a novel NK1.1 antigen with inhibitory function. J. Immunol. 162: 5917–5923. 39. Kung, S. K. P., R. C. Su, J. Shannon, and R. G. Miller. 1999. The NKR-P1B gene product is an inhibitory receptor on SJL/J NK cells. J. Immunol. 162: 5876– 5887. 40. Plougastel, B., C. Dubbelde, and W. M. Yokoyama. 2001. Cloning of Clr, a new family of lectin-like genes localized between mouse Nkrp1a and Cd69. Immunogenetics 53: 209–214. 41. Zhou, H., V. Kartsogiannis, J. M. Quinn, C. Ly, C. Gange, J. Elliott, K. W. Ng, and M. T. Gillespie. 2002. Osteoclast inhibitory lectin, a family of new osteoclast inhibitors. J. Biol. Chem. 277: 48808–48815. 42. Tian, W., R. Nunez, S. Cheng, Y. Ding, J. Tumang, C. Lyddane, C. Roman, and H. C. Liou. 2005. C-type lectin OCILRP2/Clr-g and its ligand NKRP1f costimulate T cell proliferation and IL-2 production. Cell. Immunol. 234: 39–53. 43. Tian, W., B. Feng, and H. C. Liou. 2005. Silencing OCILRP2 leads to intrinsic defects in T cells in response to antigenic stimulation. Cell. Immunol. 235: 72– 84. 44. Carlyle, J. R., A. Mesci, J. H. Fine, P. Chen, S. Be´langer, L. H. Tai, and A. P. Makrigiannis. 2008. Evolution of the Ly49 and Nkrp1 recognition systems. Semin. Immunol. 20: 321–330. 45. Higuchi, D. A., P. Cahan, J. Gao, S. T. Ferris, J. Poursine-Laurent, T. A. Graubert, and W. M. Yokoyama. 2010. Structural variation of the mouse natural killer gene complex. Genes Immun. 11: 637–648. 46. Zhou, H., V. Kartsogiannis, Y. S. Hu, J. Elliott, J. M. Quinn, W. J. McKinstry, M. T. Gillespie, and K. W. Ng. 2001. A novel osteoblast-derived C-type lectin that inhibits osteoclast formation. J. Biol. Chem. 276: 14916–14923. 47. Ciechanover, A., A. L. Schwartz, and H. F. Lodish. 1983. Sorting and recycling of cell surface receptors and endocytosed ligands: the asialoglycoprotein and transferrin receptors. J. Cell. Biochem. 23: 107–130. 48. Aust, J. G., F. Gays, K. M. Mickiewicz, E. Buchanan, and C. G. Brooks. 2009. The expression and function of the NKRP1 receptor family in C57BL/6 mice. J. Immunol. 183: 106–116. 49. Gur, C., A. Porgador, M. Elboim, R. Gazit, S. Mizrahi, N. Stern-Ginossar, H. Achdout, H. Ghadially, Y. Dor, T. Nir, et al. 2010. The activating receptor NKp46 is essential for the development of type 1 diabetes. Nat. Immunol. 11: 121–128. 50. Bilsland, C. A., and C. Milstein. 1991. The identification of the beta 2-microglobulin binding antigen encoded by the human CD1D gene. Eur. J. Immunol. 21: 71–78. 51. Vance, R. E., J. R. Kraft, J. D. Altman, P. E. Jensen, and D. H. Raulet. 1998. Mouse CD94/NKG2A is a natural killer cell receptor for the nonclassical major histocompatibility complex (MHC) class I molecule Qa-1(b). J. Exp. Med. 188: 1841–1848. 52. Orr, M. T., and L. L. Lanier. 2011. Inhibitory Ly49 receptors on mouse natural killer cells. Curr. Top. Microbiol. Immunol. 350: 67–87. 53. Ito, D., Y. M. Iizuka, M. P. Katepalli, and K. Iizuka. 2009. Essential role of the Ly49A stalk region for immunological synapse formation and signaling. Proc. Natl. Acad. Sci. USA 106: 11264–11269. 54. Doucey, M. A., L. Scarpellino, J. Zimmer, P. Guillaume, I. F. Luescher, C. Bron, and W. Held. 2004. Cis association of Ly49A with MHC class I restricts natural killer cell inhibition. Nat. Immunol. 5: 328–336. 55. Held, W., and R. A. Mariuzza. 2008. Cis interactions of immunoreceptors with MHC and non-MHC ligands. Nat. Rev. Immunol. 8: 269–278. 56. Back, J., E. L. Malchiodi, S. Cho, L. Scarpellino, P. Schneider, M. C. Kerzic, R. A. Mariuzza, and W. Held. 2009. Distinct conformations of Ly49 natural killer cell receptors mediate MHC class I recognition in trans and cis. Immunity 31: 598–608.

Downloaded from http://www.jimmunol.org/ at State University of New York @ Stony Brook on March 16, 2015

22. Carlyle, J. R., A. M. Jamieson, S. Gasser, C. S. Clingan, H. Arase, and D. H. Raulet. 2004. Missing self-recognition of Ocil/Clr-b by inhibitory NKR-P1 natural killer cell receptors. Proc. Natl. Acad. Sci. USA 101: 3527–3532. 23. Fine, J. H., P. Chen, A. Mesci, D. S. Allan, S. Gasser, D. H. Raulet, and J. R. Carlyle. 2010. Chemotherapy-induced genotoxic stress promotes sensitivity to natural killer cell cytotoxicity by enabling missing-self recognition. Cancer Res. 70: 7102–7113. 24. Chen, P., A. Mesci, and J. R. Carlyle. 2011. CELLISA: reporter cell-based immunization and screening of hybridomas specific for cell surface antigens. Methods Mol. Biol. 748: 209–225. 25. Pe´rarnau, B., M. F. Saron, B. Reina San Martin, N. Bervas, H. Ong, M. J. Soloski, A. G. Smith, J. M. Ure, J. E. Gairin, and F. A. Lemonnier. 1999. Single H2Kb, H2Db and double H2KbDb knockout mice: peripheral CD8+ T cell repertoire and anti-lymphocytic choriomeningitis virus cytolytic responses. Eur. J. Immunol. 29: 1243–1252. 26. Kartsogiannis, V., N. A. Sims, J. M. Quinn, C. Ly, M. Cipetic, I. J. Poulton, E. C. Walker, H. Saleh, N. E. McGregor, M. E. Wallace, et al. 2008. Osteoclast inhibitory lectin, an immune cell product that is required for normal bone physiology in vivo. J. Biol. Chem. 283: 30850–30860. 27. Rahim, M. M. A., P. Chen, A. N. Mottashed, A. B. Mahmoud, M. J. Thomas, Q. Zhu, C. G. Brooks, V. Kartsogiannis, M. T. Gillespie, J. R. Carlyle, et al. The NKR-P1B:Clr-b recognition system is a negative regulator of innate immune responses. Blood. In press. 28. Iizuka, K., O. V. Naidenko, B. F. Plougastel, D. H. Fremont, and W. M. Yokoyama. 2003. Genetically linked C-type lectin-related ligands for the NKRP1 family of natural killer cell receptors. Nat. Immunol. 4: 801–807. 29. Chen, P., S. Be´langer, O. A. Aguilar, Q. Zhang, A. St-Laurent, M. M. A. Rahim, A. P. Makrigiannis, and J. R. Carlyle. 2011. Analysis of the mouse 129-strain Nkrp1-Clr gene cluster reveals conservation of genomic organization and functional receptor-ligand interactions despite significant allelic polymorphism. Immunogenetics 63: 627–640. 30. Koller, B. H., P. Marrack, J. W. Kappler, and O. Smithies. 1990. Normal development of mice deficient in beta 2M, MHC class I proteins, and CD8+ T cells. Science 248: 1227–1230. 31. Zijlstra, M., M. Bix, N. E. Simister, J. M. Loring, D. H. Raulet, and R. Jaenisch. 1990. Beta 2-microglobulin deficient mice lack CD4-8+ cytolytic T cells. Nature 344: 742–746. 32. Bix, M., N. S. Liao, M. Zijlstra, J. Loring, R. Jaenisch, and D. Raulet. 1991. Rejection of class I MHC-deficient haemopoietic cells by irradiated MHCmatched mice. Nature 349: 329–331. 33. Sun, K., M. Alvarez, E. Ames, I. Barao, M. Chen, D. L. Longo, D. Redelman, and W. J. Murphy. 2012. Mouse NK cell-mediated rejection of bone marrow allografts exhibits patterns consistent with Ly49 subset licensing. Blood 119: 1590–1598. 34. Wu, M. F., and D. H. Raulet. 1997. Class I-deficient hemopoietic cells and nonhemopoietic cells dominantly induce unresponsiveness of natural killer cells to class I-deficient bone marrow cell grafts. J. Immunol. 158: 1628–1633. 35. Roth, C., J. R. Carlyle, H. Takizawa, and D. H. Raulet. 2000. Clonal acquisition of inhibitory Ly49 receptors on developing NK cells is successively restricted and regulated by stromal class I MHC. Immunity 13: 143–153. 36. Patel, R., S. Be´langer, L. H. Tai, A. D. Troke, and A. P. Makrigiannis. 2010. Effect of Ly49 haplotype variance on NK cell function and education. J. Immunol. 185: 4783–4792. 37. Carlyle, J. R., A. Mesci, B. Ljutic, S. Belanger, L. H. Tai, E. Rousselle, A. D. Troke, M. F. Proteau, and A. P. Makrigiannis. 2006. Molecular and genetic basis for strain-dependent NK1.1 alloreactivity of mouse NK cells. J. Immunol. 176: 7511–7524.