T R A N S P L A N T A T I O N A N D C E L L U L A R E N G I N E E R I N G

Adoptive therapy by transfusing expanded donor murine naturalkiller T cells can suppress acute graft-versus-host disease in

allogeneic bone marrow transplantation_2395 407..417

Jie Yang, Li Gao, Yan Liu, Yana Ren, Rufeng Xie, Huahua Fan, and Kaicheng Qian

BACKGROUND: Invariant natural killer T cells (iNKTcells) may suppress graft-versus-host disease (GVHD)after allogeneic transplantation. The purpose of thisstudy was to investigate the therapeutic potential ofiNKT cells from major histocompatibility complex(MHC)-mismatched donors for preventing GVHD afterallogeneic bone marrow transplantation (BMT).STUDY DESIGN AND METHODS: In vitro, mouseiNKT cells were expanded with a-galactosylceramideand interleukin (IL)-2 treatment. In the NKT-treatedgroup, lethally irradiated DBA/2(H-2Kd) mice were adop-tively transferred with expanded iNKT, bone marrow(BM), and spleen cells (SCs) from C57BL/6 (H-2Kb)mice. Recipients in the control group were transferredonly BM and SCs. The two groups were compared insurvival, weight, histopathologic specimens, and serumcytokine analysis.RESULTS: In the iNKT-treated group, 80% of mice sur-vived past Day 60 after BMT, but all died within 38 daysin the control group. The mice treated with iNKT did notexhibit signs of GVHD after Day 42 except for a changein fur color. There were higher IL-4 levels by Day 7 inserum of mice that received iNKT compared to thosewithout iNKT treatment, while the interferon-g levelsshowed no significant difference between two groups.Levels of IL-2 and IL-5 increased by Day 21 only iniNKT-treated mice.CONCLUSION: The results suggest that donor iNKTcells could alleviate GVHD symptoms and prolong sur-vival after MHC-mismatched allogeneic BMT, whichmay be associated with the maintenance in IL-4 levels.These findings indicate that the therapy based on iNKTcells from MHC-mismatched donors has great potentialin protection against GVHD after allogeneic hematopoi-etic stem cell transplantation.

Allogeneic hematopoietic stem cell transplanta-tion (HSCT) is an important curative therapyfor hematologic malignancies and inheriteddisorders of blood cells.1 However, graft-versus-

host disease (GVHD), especially acute GVHD (aGVHD), isknown to be an intractable and severe obstacle to success-ful allogeneic HSCT.2,3 To diminish the risk of aGVHD,patients receive immunosuppressive agents that impairT-cell functions or allografts that have been depleted ofT cells.4 Unfortunately, this immunosuppression hasnegative consequences, including increasing the risk ofinfections and diminishing T-cell-mediated graft-versus-leukemia effect.5,6 Although the pathophysiology ofaGVHD is complex and controversial, it is known that Th1polarization of donor T cells plays a role predominantly ininducing the “cytokine storm” that is seen in severalmodels of aGVHD, whereas Th2 polarization primarilysuppresses inflammatory cascades and reduces aGVHD.7,8

Therefore, cell therapy that transfers suitable cells to regu-late the Th1/Th2 balance in vivo would be the most effec-tive means to suppress aGVHD with the fewer side effects.

ABBREVIATIONS: a-GalCer = a-galactosylceramide;

aGVHD = acute GVHD; BM = bone marrow;

HSCT = hematopoietic stem cell transplantation; iNKT

cell(s) = invariant natural killer T cell(s); mIL = recombinant

mouse interleukin; MoAb = monoclonal antibody; NK

cell(s) = natural killer cell(s); SC = spleen cells; TCR = T-cell

receptor; Treg = regulatory T cell; Va14iNKT cell(s) = natural

killer T cell(s) with an invariant Va14-Ja18 TCR rearrangement

in mice.

From the Blood Engineering Laboratory, Shanghai Blood

Center, Shanghai, China.

Address reprint requests to: Huahua Fan, Blood Engineering

Laboratory, Shanghai Blood Center, Room 701, 1191 Hong Qiao

Road, Shanghai 200051, China; e-mail: fhh021@hotmail.com.

The first authors contributed equally to this work.

Received for publication March 17, 2009; revision received

July 9, 2009, and accepted July 10, 2009.

doi: 10.1111/j.1537-2995.2009.02395.x

TRANSFUSION 2010;50:407-417.

Volume 50, February 2010 TRANSFUSION 407

Among the different kinds of immunomodulatingcells, invariant natural killer T cells (iNKT cells) are a dis-tinct subset of lymphocytes characterized by expression ofsurface markers of natural killer (NK) cells together with asemi-invariant T-cell receptor (TCR).9-11 Mouse iNKT cellswith TCR Va14-Ja18 (Va14iNKT cells) can be specificallyactivated by a synthetic glycolipid, a-galactosylceramide(a-GalCer), in a nonclassical major histocompatibilitycomplex (MHC) Class I molecule CD1d-restrictedmanner.12-14 Va14iNKT cells could influence the Th1/Th2balance, mainly via rapid secretion of large amounts ofTh1 (such as interferon [IFN]-g) and Th2 (interleukin[IL]-4, IL-10, and IL-13) cytokines.15 It is reported thatiNKT cells play a critical role in various immuneresponses.16,17 Thus, iNKT cell therapy has great potentialfor allogeneic transplantation.18 It has been demonstratedthat both donor bone marrow (BM)-derived19 and host-residual20,21 NK1.1+ NKT cells can inhibit aGVHD. Further-more, several groups have reported that adoptive transferof a-GalCer, a Va14iNKT cell–specific ligand, could modu-late aGVHD and prolong survival.22 Nevertheless, the fre-quency of human NKT cells with an invariant Va24-JaQTCR rearrangement paired with Vb11 TCR (Va24iNKTcells) is very low (<0.5%) in peripheral blood mononuclearcells.11 Furthermore, it has been reported that the numberof NKT cells in recipients of HSCT with aGVHD is lowcompared to those without aGVHD, and in some caseshost NKT could not be expanded with in vivo administra-tion of a-GalCer.23 Thus, we hypothesized that the alloge-neic NKT cells from healthy donors have a wider clinicalapplication.

Our aim is to investigate the therapeutic potential ofiNKT cells from MHC-mismatched donors for the pre-vention of aGVHD after allogeneic bone marrow trans-plantation (BMT). In this report, we provide directevidence that adoptively transferred in vitro expandedVa14iNKT from MHC-mismatched donors could sup-press aGVHD in a mouse model of allogeneic BMT. Thiseffect may be dependent on IL-4, as demonstrated byusing the serum cytokine analyses at different timepoints after BMT.

MATERIALS AND METHODS

MiceWild-type male DBA/2(H-2Kd) mice were purchased fromthe Shanghai Laboratory Animal Center of the ChineseAcademy of Science (Shanghai, China) and allowed toreach a minimum weight of 25 g/animal before initiationof host conditioning. C57BL/6 (H-2Kb) male donor micewere also purchased from Shanghai Laboratory AnimalCenter of the Chinese Academy of Science. All mice,between 6 and 12 weeks of age, were maintained in a spe-cific pathogen-free environment and received drinkingwater supplemented with gentamycin.

GVHD modelTen- to 12-week-old DBA/2 mice (recipients) were giventotal body lethal irradiation (800 cGy) from a gamma raysource and injected with 3 ¥ 107 BM cells and 1 ¥ 107 spleencells (SCs) from C57BL/6 mice (donors) via the tail veinafter 24 hr. Donor cells were resuspended in 0.1 mL ofphosphate-buffered saline (PBS; Gibco, Grand Island, NY),and the same amount of PBS was injected as the control.

In vitro expansion of Va14iNKT cellsIn vitro expansion of Va14iNKT cells was performed aspreviously described.24 Briefly, SC suspensions (2 ¥ 106

cells/mL) from C57BL/6 mice were cultured with a-GalCer(50 ng/mL, Kirin Brewery Company, Tokyo, Japan) andrecombinant mouse interleukin-2 (mIL-2, 300 IU/mL,R&D Company, Minneapolis, MN) in RPMI 1640 (Gibco)supplemented with 10% fetal calf serum (Gibco), penicillin(50 U/mL, Sigma, St Louis, MO), streptomycin (50 mg/mL,Sigma) in a 37°C, 5% CO2 incubator. Va14iNKT cells wereharvested after 7 days.

Flow cytometryThe following monoclonal antibodies (MoAbs) were used:CD3-fluorescein isothiocyanate (FITC), CD4-FITC, CD8-phycoerythrin (PE), CD25-PE, and NK1.1-PE (BD Bio-sciences PharMingen, San Jose, CA). FITC-conjugatedCD1d/a-GalCer tetramer (BD Biosciences PharMingen)was prepared according to the manufacturer’s specifica-tions. Isotype controls were run in parallel. Immunofluo-rescence staining was performed according to standardprocedures.25 Cell populations were phenotyped with thepanel of MoAbs and analyzed on a flow cytometer(FACScan, EPICS-ELITE, Coulter, Miami, FL). Dead cellswere gated out on the basis of their light scatter properties.

Molecular analysis of TCR-a transcriptTotal RNA was extracted from 1 ¥ 106 NKT cells accordingto the manufacturer’s protocol (Tri Reagent LS, Invitro-gen, Carlsbad, CA) and reverse transcribed. The tran-scribed cDNA was subjected to PCR amplification usingthe primer pair (5′-CTAAGCACAGCACGCTGCACA-3′,Va14; and 5′-TGGCGTTGGTCTCTTTGAAG-3′, C a) or thepair for b-actin (5′-GTGGGGCGCCCCAGGCACCA-3′ and5′-CTTCCTTAATGTCACGCACGATTTC-3′) under the fol-lowing conditions: 94°C for 60 seconds, 60°C for 60seconds, and 72°C for 60 seconds for 40 cycles.

In vitro cytokine production by expandedVa14iNKT cellsFor in vitro cytokine production assay, Va14iNKT cellswere stimulated for 4 hours at 37°C with 20 ng/mL phyto-

YANG ET AL.

408 TRANSFUSION Volume 50, February 2010

hemagglutinin (Sigma) and 1 mg/mL ionomycin (Sigma)in a complete medium containing intracellular transportinhibitor monensin (10 mg/mL, Sigma). After beingwashed with PBS, the cells were permeabilized withfluorescence-activated cell sorting (FACS) permeabilizingsolution (BD Biosciences PharMingen) for 10 minutes atroom temperature. These cells were resuspended in stain-ing buffer and stained for 30 minutes in the dark withPE-conjugated anti-IFN-g and anti-IL-4 MoAbs (BD Bio-sciences PharMingen). Unbound Abs were removed bytwo washes with staining buffer. Finally, the cells wereanalyzed by flow cytometry as described above.

Adoptive transfer of Va14iNKT cellsFor the Va14iNKT adoptive transfer, recipient micereceived 1 ¥ 107 donor iNKT cells via the tail vein immedi-ately after BM and splenocyte transplantation.

Monitoring of GVHD and survivalSurvival and appearance were monitored every other dayand body weight was measured weekly. GVHD wasassessed using a scoring system that summed changes infive clinical variables:26 weight loss, posture, activity, furtexture, and skin integrity (maximum index, 10). Meanbody weights of surviving mice in each group were deter-mined on Day 60.

ChimerismSplenocytes from euthanized mice on Day 15 posttrans-plant were stained with H-2Kb-PE or H-2Kd-PE MoAbs (BDBiosciences PharMingen). Immunofluorescence stainingwas performed according to standard procedures.25 Cellswere analyzed by flow cytometry as described above.

HistopathologyHistopathologic specimens from the liver and intestine ofhost mice were obtained on Day 7 or 42 after transplanta-tion if the host survived, fixed in 10% formalin, and embed-ded in paraffin blocks. Four- to 5-mm sections were stainedwith H&E using standard protocols. Microscopic imageswere obtained using a microscope (Eclipse E1000M, Nikon,Tokyo, Japan) with a SPOT room temperature digitalcamera and acquisition software (Diagnostic Instruments)with final magnification (objective, ¥20/0.45 numericalaperture) provided with each figure. Image processing wasperformed with a computer program (Photoshop CS,Adobe Systems, San Jose, CA), with standard adjustmentsof brightness, contrast, and color balance to the entireimage. A semiquantitative scoring system was used toassess the following abnormalities known to be associatedwith GVHD, as previously described:27 0, normal; 0.5, focal

and rare; 1.0, focal and mild; 2.0, diffuse and mild; 3.0,diffuse and moderate; and 4.0, diffuse and severe.

Serum cytokinesSerum was obtained from surviving recipient mice onDays 3, 7, and 21 posttransplant and stored at –80°C. Thelevels of IL-2, IL-4, IL-5, tumor necrosis factor (TNF)-a,and IFN-g were detected simultaneously using the mouseTh1/Th2 cytokine cytometric bead array kit (BD PharM-ingen) according to the manufacturer’s specifications.Cytokine bead staining was analyzed by flow cytometryand data were compiled with BD Biosciences cytometricbead array software. A total of three samples were ana-lyzed in each group.

Statistical analysisResults are expressed as mean � SD. Statistical differencesin animal survival were analyzed with the log-rank test.Differences in weight of hosts and in serum cytokine levelsbetween transplanted groups were analyzed with com-puter statistical software (Statistica 6.0, StatSoft, Inc.,Tulsa, OK). For all tests, a p value of less than 0.05 wasconsidered significant.

RESULTS

Establishment of aGVHD model miceIn the prototype transplantation from wild-type C57BL/6to wild-type DBA/2, all of the recipients showed lethalGVHD as judged by typical features, including diarrhea,weight loss, hunched back, and hair loss. Histologic exami-nations also confirmed GVHD in representative mice. OnDay 15 the expression of MHC I (H-2Kb/d) in the normalC57BL/6, normal DBA/2, and DBA/2 mice that receivedBM from C57BL/6 were checked by FACS. The FACS resultsindicated that after allo-BMT the expression of MHC I inDBA/2 changed and splenocytes expressed H-2Kb com-pletely, just as in normal C57BL/6 mice (Fig. 1). In thissetting, we confirmed that full donor chimerism wasachieved in all of the recipients. All of the transplantationexperiments were performed six times independently andwe then integrated all of the results from individual experi-ments to avoid experiment-to-experiment variation.

Expansion of mouse Va14iNKT cells in vitroTo obtain a large number of Va14iNKT cells, SCs originat-ing from C57BL/6 mice were cultured with a-GalCer andmIL-2 for 7 days as previously reported.14 As shown inFig. 2A, 2.18% of freshly isolated SCs comprised Va14iNKTcells (CD1d/a-GalCer tetramer-positive cells). Afterexpansion in culture, the percentage and absolute

DONOR Va14INKT CELLS CAN SUPPRESS aGVHD IN ALLOGENEIC BMT

Volume 50, February 2010 TRANSFUSION 409

number of Va14iNKT cells increased to 82.45% and37-fold, respectively. In addition, most Va14iNKT cellswere CD4+ and CD4-CD8- double-negative subsets. Halfof CD1d/a-GalCer tetramer-positive cells were NK1.1+ Tcells, and approximately 50% of expanded Va14iNKT cellsexpressed CD25, the activated phenotype (Fig. 2B). InFig. 2C, electrophoresed PCR products from in vitroexpanded iNKT cells have been shown. It was confirmedthat the in vitro expanded cells are Va14iNKT cells. Theseresults indicate that Va14iNKT cells originated from thespleen were expanded effectively under the costimulationof the a-GalCer and mIL-2 in vitro.

Adoptive transfer of in vitro expanded donorVa14iNKT cells reduces symptoms of aGVHD andprolonged survival time of aGVHD miceWe examined whether the adoptive transfer of in vitroexpanded donor Va14iNKT cells, in which MHC I mol-ecules were mismatched completely with host’s, couldameliorate clinical symptoms and histologic findings ofaGVHD. This was accomplished by summing changes infive clinical variables of aGVHD and analyzing histologyof liver and intestine tissue specimens. It was found thatthe severity of aGVHD in mice receiving only BM and SC(aGVHD mice) was greater than in those transferred withadditional Va14iNKT cells (Va14iNKT-treated mice),including hair loss, hunched back, swollen face, and diar-rhea (Fig. 3A; p < 0.001; log-rank test). Figure 3B shows

the mean serial body weights of the groups in Fig. 3A. TheVa14iNKT-treated mice lost more weight at earlier timepoints than the aGVHD mice and then recovered fromthis weight loss and did not exhibit other signs of GVHDexcept for development of white fur (from gray) afterDay 42.

Meanwhile, hosts that received only BM andVa14iNKT cells survived at least 120 days without clinicalsigns of GVHD, but fur also turned white in these animals.The aGVHD mice showed severe changes typical of GVHDon microscopic examination of livers obtained at autopsyon Day 7 posttransplant, including the presence ofremarkable hepatic lymphocyte infiltration surroundingthe portal area and local putrescence. Characteristicchanges of GVHD were also observed in intestines includ-ing crypts atrophy associated with apoptosis of cryptcells and severe inflammatory cell infiltration. In contrast,very little infiltration and no putrescence were detectedfrom the liver and intestine specimens obtainedfrom Va14iNKT-treated mice after 7 days, and therewere similar findings in those that received only BMand Va14iNKT cells. The mean GVHD scores of the groupgiven only BM and SC was significantly increased(p < 0.05) compared with the group given iNKT with BMand SC in both tissues examined (Figs. 4A and 4B). GVHDsymptoms were further reduced in the iNKT-treated miceon Day 42. The mean histopathologic GVHD scores of theliver and intestine of the group given iNKT-treatment andthe group without are shown in Fig. 4C.

Fig. 1. Expression of MHC I (H-2Kb/d) in the normal C57BL/6 (A), normal DBA/2 (B), and DBA/2 mice that received BM from

C57BL/6. On Day 15 after BMT, the expression of MHC I (H-2Kb/d) in the DBA/2 hosts that received BM from C57BL/6 were checked

by FACS. The results indicated that the splenocytes of normal C57BL/6 mice expressed H-2Kb (A). However, DBA/2 mice expressed

H-2Kd (B). After allo-BMT the expression of MHC I in DBA/2 changed, and splenocytes expressed H-2Kb completely, just as normal

C57BL/6 (C).

YANG ET AL.

410 TRANSFUSION Volume 50, February 2010

To obtain more direct evidence that Va14iNKT cellssuppressed GVHD, survival rates of aGVHD mice withtransferred Va14iNKT cells and of controls were analyzed.In the group receiving only BM and SC, half of the mice diedby Day 20 and 100% died by Day 38 post-BMT. With theadditional iNKT treatment, 83.33% of the mice survivedby Day 60, and more than 60% of the mice survived byDay 120. In addition, all hosts that received only BMand Va14iNKT cells survived at least 120 days (Fig. 5).Therefore, these results indicate that the adoptive transfer

of donor Va14iNKT cells can alleviate symptoms ofaGVHD and markedly prolonged survival time in aGVHDmice.

In vitro expanded donor Va14iNKT cells cansuppress GVHD by maintenance of the amounts ofIL-4 productionTo determine the relationship in Th1/Th2 cytokine andaGVHD development, first, the levels of IFN-g and IL-4

Fig. 2. Phenotypic identification of the in vitro expanded Va14iNKT cells by FACS. (A) CD1d/a-GalCer tetramer expression of fresh

and in vitro expanded Va14iNKT after 7 days in culture. Cultured with a-GalCer and mIL-2 treatment for 7 days, Va14iNKT cells

originated from spleen were efficiently induced and expanded. (B) The surface marker expression of the in vitro expanded

Va14iNKT. Most Va14iNKT cells were CD4+ and CD4-CD8- double-negative subsets. Half of CD1d/a-GalCer tetramer-positive cells

were NK1.1+ T cells. Approximately 50% of expanded Va14iNKT cells expressed the activated phenotype (CD25+). (C) Electrophore-

sed PCR products from in vitro expanded iNKT cells using Va14-specific 5� primer and Ca-specific 3� primer (Lane 2) and those

using b-actin–specific primer (Lane 1).

DONOR Va14INKT CELLS CAN SUPPRESS aGVHD IN ALLOGENEIC BMT

Volume 50, February 2010 TRANSFUSION 411

expression of Va14iNKT cells after invitro stimulation were tested by flowcytometry. Consistent with previousreports, in vitro expanded Va14iNKTcells could produce large amounts ofIFN-g and IL-4 rapidly (Fig. 6A). Levelsof cytokines in the serum of the micewith or without NKT treatment wereexamined. It was obvious that the pro-duction of IFN-g and TNF-a in serumquickly increased in both groups andthere was no marked differencebetween them in the early phase post-BMT. Although the levels of IL-4 alsogrew in the early stage after transplanta-tion, there was a significant decrease inthe production of IL-4 on Day 7 afterBMT in the aGVHD group. Comparedwith this group, the levels of IL-4 werehigher in mice with NKT treatment onDay 7 after BMT and remained equiva-lent later (Fig. 6B). Meanwhile, thelevels of IL-2 and IL-5 increased quicklyon Day 3 and reduced significantly onDay 7 in both groups, but in the iNKT-treated mice the production of IL-2 andIL-5 increased again at Day 21 (Fig. 6C).These findings suggest that in vitroexpanded Va14iNKT cells played acrucial role in reducing the severity ofaGVHD, and this reduction is associatedwith a delayed maintenance in serumIL-4 levels and polarizing Th responsetoward Th2 in the early phase after allo-geneic BMT.

Fig. 3. Adoptive transfer of in vitro expanded donor Va14iNKT cells reduced symp-

toms of aGVHD. (A) Mean clinical GVHD score and SEM of the DBA/2 mice receiving

allo-BM and SC from C57BL/6 mice ( , aGVHD mice, n = 11), mice receiving

Va14iNKT with allo-BM and SC (�, Va14iNKT-treated mice, n = 6), and mice receiv-

ing allo-BM and Va14iNKT (�, n = 6). The GVHD score was determined from the sur-

viving mice. From the 7th day to 35th day after transplantation, the GVHD score was

significantly lower in the Va14iNKT-treated mice (p<0.001). (B) Mean body weight

(�SEM) of groups shown in A at serial time points after transplantation (+). Analysis

was stopped when only one host remained in the group.

�Fig. 4. Histopathologic analysis of specimens from the liver and intestine of host mice showed adoptive transfer of in vitro

expanded donor Va14iNKT cells reduced symptoms of aGVHD. (A) Microscopic examination of H&E-stained sections of liver at

autopsy in DBA/2 hosts treated with 800 cGy total body irradiation followed by allo-BM, SC, and NKT cell transfer. Evidence of

remarkable hepatic lymphocyte infiltration (,) surrounding the portal area and local putrescence (¥) is shown in the liver speci-

men of the aGVHD group on the seventh day after transfer. Hosts that received Va14iNKT with BM and SC showed little lymphocyte

infiltration and no putrescence in the liver sections upon autopsy on Day 7 and infiltration was further reduced on Day 42. Similar

results were seen in those that received only BM and Va14iNKT cells. Normal liver sections from normal mice are shown for com-

parison. Liver specimens are shown at ¥200. (B) Microscopic examination of H&E-stained sections of intestine from the same

groups as A. Evidence of crypts atrophy (,) associated with apoptosis of crypt cells and severe inflammatory cell infiltration (¥) is

shown with separation of crypts by inflammation and lifting off of crypts from the basement membrane in the aGVHD group on

Day 7. There are moderate lymphocyte infiltrations and no crypt abscess in the intestine from the Va14iNKT-treated mice on Day 7.

Further, infiltration was reduced on Day 42. Normal intestine sections from normal mice are shown for comparison. Intestine

specimens are shown at ¥200. (C) Mean (�SEM) histopathologic GVHD scores of liver and intestine from the DBA/2 mice receiving

allo-BM and SC from C57BL/6 mice (�), mice receiving Va14iNKT with allo-BM and SC ( ), and mice receiving allo-BM and

Va14iNKT ( ). *Differences (p < 0.05) between two groups were significant; n = 3 for all groups.

YANG ET AL.

412 TRANSFUSION Volume 50, February 2010

DISCUSSION

Many studies have suggested that invariant NKT cells playa significant role in immune regulation, including autoim-munity, tumor immunity, and infection.16,17 In transplan-

tation immunity, it is also thought that iNKT cells areimportant in the induction of allograft tolerance.28,29

Some reports have suggested that both donorBM-derived and host residual NKT cells (NK1.1+or DX5+ Tcells) may suppress aGVHD. These NKT cell populations

DONOR Va14INKT CELLS CAN SUPPRESS aGVHD IN ALLOGENEIC BMT

Volume 50, February 2010 TRANSFUSION 413

should overlap with the iNKT cell population that wedescribed here. Although we used donor spleen–derivedVa14iNKT cells and myeloablative DBA/2 mice as recipi-ents, we also obtained similar results. Therefore, the effec-tive attenuation of GVHD and markedly prolonged survivalof aGVHD mice by adoptive transfer of in vitro expandeddonor Va14iNKT cells described here is consistent withprevious reports. Furthermore, it was reported that donorBM Type II (non-Va14Ja18 CD1d-restricted) NKT cells alsocould suppress GVHD by producing IFN-g and IL-4.30,31

In our aGVHD model DBA/2 mice with gray fur werechosen as the host, differing from other research efforts, inwhich BALB/C mice with white fur were used. Interest-ingly, we found that the host mice that received iNTK cellsdid not exhibit other signs of GVHD except for a change infur color on Day 40 after allogeneic BMT. It may be a cluethat adoptive transfer of iNKT cells can reduce aGVHD butalso transform aGVHD to a chronic form of GVHD.

As already known, Va14iNKT cells could producelarge amounts of Th1 (such as IFN-g) and Th2 cytokine(IL-4, IL-10, and IL-13) rapidly after stimulation in vitro. Todetermine how NKT cells reduce GVHD, we examinedvarious cytokines in the serum of mice with or withoutNKT treatment based on the aGVHD model. From ourresults, it is obvious that the levels of IL-4 in serum werehigher in mice with iNKT treatment than in those without

treatment 7 days after BMT, but high levels of IFN-g werefound in both groups and there were no significant differ-ences between them in early phases. Inflammatory cytok-ines have been shown to be important effect molecules ofaGVHD. Th2 cytokine responses subsequently inhibitedinflammatory cytokine cascades and reduced morbidityand mortality of aGVHD, as previously described.32 Thesefindings are consistent with our observation that IL-4 pro-duction is critical for the protective effects of NKT cellsagainst aGVHD in the early time after allogeneic BMT. As aresult of only a single injection of Va14iNKT in our study,signs of the suppression of iNKT cells, such as higher IL-4secretion, primarily appeared at the early stage posttrans-plantation. Therefore, iNKT-based therapy could beimproved by modulating the numbers and subsets ofiNKT cells, injection frequency, and so on. Current strate-gies against aGVHD are induction of a Th2-dominantcytokine profile before the onset of obvious aGVHD.33

Some investigators have reported that stimulation of hostNKT cells by injecting OCH, the analog of a-GalCer, polar-ized donor T cells toward Th2 cytokine secretion. Thisresulted in marked reduction of serum IFN-g levels afterBMT and protection against aGVHD.34,35

In addition, we found that the serum levels of IL-2 andIL-5, the primary cytokines produced by Th2 cells,increased after Day 21 posttransplantation. In addition,inflammatory cytokines IFN-g and TNF-a decreasedrapidly in the NKT-treated group by Day 21, while the pro-duction of IL-4 remained equivalent across both groups.In this regard, we think that the tolerant state is estab-lished with iNKT treatment in vivo. We hypothesize that itmay be related to regulatory CD4+CD25+ T cells, whichneed large amounts of IL-2 when induced and can alsoinhibit GVHD.36,37 The recent report38 that demonstratedhost NKT cells induce an IL-4-dependent expansion ofdonor regulatory T cells supports our hypothesis.

In vitro, Va14iNKT cells can kill several tumors andthe antitumor effects are enhanced through their activa-tion by a-GalCer stimulation (data not shown). On theother hand, activation of iNKT cells substantially influ-ences function of other cell types, particularly DC, NKcells, CD4 Th1 cells, and CD8 cytotoxic T cells, all contrib-uting to antitumor immune responses.39-41 We believe thatadoptive transfer of NKT cell therapy combined with allo-geneic HSCT may be beneficial for cancer patients,because of NKT cell function for prevention of GVHD andantitumor effects.

Fig. 5. Adoptive transfer of in vitro expanded donor Va14iNKT

cells prolonged survival time of aGVHD mice. The DBA/2 host

survival of the DBA/2 mice receiving allo-BM and SC from

C57BL/6 mice ( , aGVHD mice, n = 11), the mice receiving

Va14iNKT with allo-BM and SC (�, Va14iNKT-treated mice,

n = 6), and mice receiving allo-BM and Va14iNKT (�, n = 6).

Va14iNKT-treated mice survived longer than aGVHD mice. All

of the hosts that received only BM and iNKT cells survived

through Day 60.

�Fig. 6. In vitro expanded donor Va14iNKT cells can suppress GVHD by maintaining the amounts of IL-4 production. (A) The pro-

duction of IFN-g and IL-4 of the Va14iNKT cells after stimulation in vitro checked by FACS. After stimulation, expanded iNKT cells

could produce large amounts of IFN-g and IL-4 cytokine rapidly. (B) Comparison of mean serum levels (�SEM) of IFN-g, IL-4,

TNF-a, IL-2, and IL-5 on Days 3 and 7 in the aGVHD group ( ) versus NKT-treated group ( ). *Significant differences (p < 0.05)

between both groups; n = 3 for all groups. (C) The changes of cytokines in serum at a time series time in the NKT-treated group. (�)

3 days; ( ) 7 days; ( ) 21 days. *Significant differences (p < 0.05) between two time points; n = 3 for all groups.

YANG ET AL.

414 TRANSFUSION Volume 50, February 2010

DONOR Va14INKT CELLS CAN SUPPRESS aGVHD IN ALLOGENEIC BMT

Volume 50, February 2010 TRANSFUSION 415

In conclusion, adoptive transfer of in vitro expandeddonor iNKT can protect against aGVHD after allogeneicBMT by regulating the amounts of IL-4 production andpolarizing Th response toward Th2. Our findings indicatethe potential of therapeutic use of iNKT cells for the pre-vention or treatment of aGVHD.

ACKNOWLEDGMENTS

We thank Kirin Brewery Company for providing synthetic

a-GalCer. We thank the Department of Animal Science in Shang-

hai Jiaotong University School of Medicine for the laboratory

animal husbandry.

CONFLICT OF INTEREST

None of the authors have a financial conflict of interest.

REFERENCES

1. Shlomchik WD. Graft-versus-host disease. Nat Rev

Immunol 2007;7:340-52.

2. Korngold B, Sprent J. Lethal graft-versus-host disease

after bone marrow transplantation across minor

histocompatibility barriers in mice. Prevention by

removing mature T cells from marrow. J. Exp. Med. 1978;

148:1687.

3. Sun Y, Tawara I, Toubai T, Reddy P. Pathophysiology of

acute graft-versus-host disease: recent advances. Transl

Res. 2007;150:197-214. Epub 2007 July 3.

4. Messina C, Faraci M, de Fazio V, Dini G, Calò MP, Calore E;

EBMT Paediatric Working Party. Prevention and treatment

of acute GvHD. Bone Marrow Transplant 2008;41 Suppl

2:S65-70.

5. Bleakley M, Riddell SR. Molecules and mechanisms of the

graft-versus-leukaemia effect. Nat Rev Cancer 2004;4:371-

80.

6. Kolb HJ. Graft-versus-leukemia effects of transplantation

and donor lymphocytes. Blood 2008;112:4371-83.

7. Dickinson AM, Holler E. Polymorphisms of cytokine and

innate immunity genes and GVHD. Best Pract Res Clin

Haematol 2008;21:149-64.

8. Hill GR, Ferrara JL. The primacy of the gastrointestinal

tract as a target organ of acute graft-versus-host disease:

rationale for the use of cytokine shields in allogeneic bone

marrow transplantation. Blood 2000;95:2754-9.

9. Godfrey DI, Hammond KJ, Poulton LD, Smyth MJ, Baxter

AG. NKT cells: facts, functions and fallacies. Immunol

Today 2000;21:573-83.

10. Godfrey DI, MacDonald HR, Kronenberg M, Smyth MJ, Van

Kaer L. NKT cells: what’s in a name? Nat Rev Immunol

2004;4:231-7.

11. Bendelac A, Savage PB, Teyton L. The biology of NKT cells.

Annu Rev Immunol 2007;25:297-336.

12. Wilson MT, Singh AK, Van Kaer L. Immunotherapy with

ligands of natural killer T cells. Trends Mol Med 2002;8:

225-31.

13. Tsuji M. Glycolipids and phospholipids as natural

CD1d-binding NKT cell ligands. Cell Mol Life Sci 2006;63:

1889-98.

14. Silk JD, Salio M, Brown J, Jones EY, Cerundolo V. Structural

and functional aspects of lipid binding by CD1 molecules.

Annu Rev Cell Dev Biol 2008;24:369-95.

15. Dickinson AM, Holler E. Polymorphisms of cytokine and

innate immunity genes and GVHD. Best Pract Res Clin

Haematol. 2008;21:149-64.

16. Taniguchi M, Harada M, Kojo S, Nakayama T, Wakao H.

The regulatory role of Valpha14 NKT cells in innate and

acquired immune response. Annu Rev Immunol 2003;21:

483-513.

17. Matsuda JL, Mallevaey T, Scott-Browne J, Gapin L. CD1d-

restricted iNKT cells, the “Swiss-Army knife” of the

immune system. Curr Opin Immunol 2008;20:358-68.

18. Cerundolo V, Silk JD, Masri SH, Salio M. Harnessing invari-

ant NKT cells in vaccination strategies. Nat Rev Immunol

2009;9:28-38.

19. Zeng D, Lewis D, Dejbakhsh-Jones S, Lan F, García-Ojeda

M, Sibley R, Strober S. Bone marrow NK1.1- and NK1.1+ T

cells reciprocally regulate acute graft versus host disease.

J Exp Med 1999;189:1073.

20. Haraguchi K, Takahashi T, Matsumoto A, Asai T, Kanda Y,

Kurokawa M, Ogawa S, Oda H, Taniguchi M, Hirai H, Chiba

S. Host-residual invariant NK T cells attenuate graft-

versus-host immunity. J Immunol2005;175:1320-8.

21. Hashimoto D, Asakura S, Miyake S, Yamamura T, Van Kaer

L, Liu C, Tanimoto M, Teshima T. Stimulation of host NKT

cells by synthetic glycolipid regulates acute graft-versus-

host disease by inducing Th2 polarization of donor T cells.

J Immunol 2005;174:551-6.

22. Pillai AB, George TI, Dutt S, Teo P, Strober S. Host NKT

cells can prevent graft-versus-host Dis and permit graft

antitumor activity after bone marrow transplantation.

J Immunol 2007;178:6242-51.

23. Nieda M, Okai M, Tazbirkova A, Lin H, Yamaura A, Ide K,

Abraham R, Juji T, Macfarlane DJ, Nicol AJ. Therapeutic

activation of Va24+Vb11+ NKT cells in human subjects

results in highly coordinated secondary activation of

acquired and innate immunity. Blood 2004;103:383-9.

24. Watarai H, Nakagawa R, Omori-Miyake M, Dashtsoodol N,

Taniguchi M. Methods for detection, isolation and culture

of mouse and human invariant NKT cells. Nat Protoc 2008;

3:70-8.

25. Haraguchi K, Takahashi T, Hiruma K, Kanda Y, Tanaka Y,

Ogawa S, Chiba S, Miura O, Sakamaki H, Hirai H. Recovery

of Valpha24+ NKT cells after hematopoietic stem cell trans-

plantation. Bone Marrow Transplant 2004;34:595-602.

26. Cooke KR, Kobzik L, Martin TR, Brewer J, Delmonte J Jr,

Crawford JM, Ferrara JL. An experimental model of idio-

pathic pneumonia syndrome after bone marrow

YANG ET AL.

416 TRANSFUSION Volume 50, February 2010

transplantation: I. The roles of minor H antigens and

endotoxin. Blood 1996;88:3230-9.

27. Kaplan DH, Anderson BE, McNiff JM, Jain D, Shlomchik MJ,

Shlomchik WD. Target antigens determine graft-versus-

host disease phenotype. J Immunol 2004;173:5467-75.

28. Seino K, Fukao K, Muramoto K, Yanagisawa K, Takada Y,

Kakuta S, Iwakura Y, Van Kaer L, Takeda K, Nakayama T,

Taniguchi M, Bashuda H, Yagita H, Okumura K. Require-

ment for natural killer T (NKT) cells in the induction of

allograft tolerance. Proc Natl Acad Sci U S A 2001;98:2577-

81.

29. Chargui J, Hase T, Wada S, Naganuma T, Yoshimura R.

NKT cells as nonspecific immune-regulator inducing toler-

ance in mouse model transplantation. Transpl Proc 2001;

33:3833-4.

30. Oh SJ, Kim JH, Min CK, Chung DH. Role of type II NKT

cells in the suppression of graft-versus-host disease. Crit

Rev Immunol 2008;28:249-67.

31. Kim JH, Choi EY, Chung DH. Donor bone marrow type II

(non-Valpha14Jalpha18 CD1d-restricted) NKT cells sup-

press graft-versus-host disease by producing IFN-gamma

and IL-4. J Immunol 2007;179;6579-87.

32. Krenger W, Snyder KM, Byon JC, Falzarano G, Ferrara JL.

Polarized type 2 alloreactive CD4+ and CD8+ donor T cells

fail to induce experimental acute graft-versus-host disease.

J. Immunol 1995;155:585-93.

33. Goker H, Haznedaroglu IC, Chao NJ. Acute graft-vs-host

disease: pathobiology and management. Exp Hematol

2001;29:259-77.

34. Miyamoto K, Miyake S, Yamamura T. A synthetic glycolipid

prevents autoimmune encephalomyelitis by inducing

TH2 bias of natural killer T cells. Nature 2001;413:

531-4.

35. Oki S, Chiba A, Yamamura T, Miyake S. The clinical impli-

cation and molecular mechanism of preferential IL-4 pro-

duction by modified glycolipid stimulated NKT cells. J Clin

Invest 2004;113:1631.

36. Sakaguchi S, Yamaguchi T, Nomura T, Ono M. Regulatory

T cells and immune tolerance. Cell 2008;133:775-87.

37. Vignali DA, Collison LW, Workman CJ. How regulatory T

cells work. Nat Rev Immunol 2008;8:523-32.

38. Pillai AB, George TI, Dutt S, Strober S. Host natural killer T

cells induce an interleukin-4-dependent expansion of

donor CD4+CD25+Foxp3+ T regulatory cells that

protects against graft-versus-host disease. Blood 2009;113:

4458-67.

39. Nicol A, Nieda M, Koezuka Y, Porcelli S, Suzuki K,

Tadokoro K, Durrant S, Juji T. Human invariant valpha24+natural killer T cells activated by alpha-galactosylceramide

(KRN7000) have cytotoxic anti-tumour activity through

mechanisms distinct from T cells and natural killer cells.

Immunology 2000;99:229-34.

40. Ambrosino E, Berzofsky JA, Terabe M. Regulation of tumor

immunity: the role of NKT cells. Expert Opin Biol Ther

2008;8:725-34.

41. Motohashi S, Nakayama T. Clinical applications of natural

killer T cell-based immunotherapy for cancer. Cancer Sci

2008;99:638-45. Epub 2008 Feb 21.

DONOR Va14INKT CELLS CAN SUPPRESS aGVHD IN ALLOGENEIC BMT

Volume 50, February 2010 TRANSFUSION 417