FASEB J. Mp Biomedicals
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.07-104687.
 (The FASEB Journal. 2008;22:2223-2231.)
© 2008 FASEB
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
fj.07-104687v1
22/7/2223    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wakao, H.
Right arrow Articles by Fujita, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wakao, H.
Right arrow Articles by Fujita, H.

In vitro induction of natural killer T cells from embryonic stem cells prepared using somatic cell nuclear transfer

Hiroshi Wakao*,§,1, Rika Wakao{ddagger},2, Sakura Sakata§, Kazuya Iwabuchi{dagger}, Atsushi Oda* and Hiroyoshi Fujita*

* Department of Environmental Biology, School of Medicine, and

{dagger} Division of Immunobiology, Institute for Genetic Medicine, Hokkaido University, Sapporo, Japan; and

{ddagger} Developmental Genetics Group and

§ Immune Regulation Group, Rikagaku Kenkyusho (RIKEN), Research Center for Allergy and Immunology, Yokohama, Japan

1Correspondence: Department of Environmental Biology, School of Medicine, Hokkaido University, N15W7, Sapporo 060-8638, Japan. E-mail: hwakao{at}med.hokudai.ac.jp


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The ectopic expression of the Notch receptor ligand delta-like 1 on stromal cells allows the induction of T cells from embryonic stem cells (ESCs). However, these in vitro-generated T cells are not transplantable because they are too immature to mount an immune response in an immunocompromised animal. We efficiently generated a subset of T cells called invariant natural killer T (iNKT) cells from ESCs derived from peripheral iNKT cells using somatic cell nuclear transfer (ntESCs). These iNKT cells matured autonomously in vivo and exhibited an adjuvant effect accompanying the production of interferon-{gamma} in an antigen-specific manner. This adjuvant effect culminated in the inhibition of inoculated tumor cell growth. Our results indicate that ntESC-derived iNKT cells are transplantable lymphocytes that will be beneficial for the induction of immune tolerance and the treatment of autoimmune diseases, tumors, and infections.—Wakao, H., Wakao, R., Sakata, S., Iwabuchi, K., Oda, A., Fujita, H. In vitro induction of natural killer T cells from embryonic stem cells prepared using somatic cell nuclear transfer.


Key Words: adjuvant effect • in vitro differentiation • transplantable lymphocytes • tumor rejection


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE REALIZATION OF REGENERATIVE medicine requires sufficient amounts of appropriately functional and transplantable cells. Because embryonic stem cells (ESCs) are a pure population of undifferentiated cells that possess pluripotency and indefinite growth yet retain normal karyotypes, the oriented differentiation of monoclonal cells from ESCs is an option that conforms to the above criteria (1) . ESCs cultured in vitro autonomously differentiate and rapidly proliferate to form embryoid bodies consisting of semiorganized germ layer tissues, including mesoderm-derived hematopoietic cells, in the absence of external stimuli (2) . In contrast, ESCs cultured on the bone marrow stromal cell line OP9 allow the generation of myeloid/lymphoid lineages, except T lymphocytes (3 4 5) . These results indicate that external cues play an important role in determining the outcome of cell lineages from ESCs. A recent report demonstrated that delta-like 1(dlk1), a Notch receptor ligand, plays a pivotal role in inducing T lymphocytes from ESCs (6) . Nonetheless, T cells that are induced from ESCs are not fully functional per se and do not result in immune competence; further maturation ex vivo is required to mount an immune response in vivo (6) . Although T lymphocyte derivation from ESCs is attractive from a clinical point of view to remedy diseases related to immune deficiency/dysfunction, functional limitations should be circumvented before applying this technique to regenerative medicine.

Should there exist T lymphocytes that spontaneously mature and function on adoptive transfer by exploiting in vivo environmental cues, such cells would overcome the problem inherent to T lymphocytes that are induced from ESCs and would be clinically relevant. These cells are of prime importance to the realization of cell therapy. T lymphocytes eligible for cell therapy should be innate-like T cells, such as natural killer T (NKT) cells that recognize glycolipid antigens.

NKT cells are T cells that express receptors that are found on natural killer (NK) cells, as well as T cell receptors (TCRs) on the cell surface. A subset of NKT cells, iNKT cells, expresses the invariant TCR{alpha} chain V{alpha}14-J{alpha}18 and plays a pleiotropic role in the immune system. The development of iNKT cells is contingent on glycolipids presented on CD1d, which is a major histocompatibility complex (MHC) class I-related antigen-presenting glycoprotein that is monomorphic. This feature distinguishes iNKT cells from conventional T cells, whose development is restricted by MHC class I and class II molecules that are polymorphic (7 8 9) . The iNKT cells are self-reactive and behave as regulatory T cells by secreting various types of cytokines; this makes them an attractive target for therapeutic intervention to control a variety of autoimmune diseases, tumor growth, and infection (10) . The iNKT cells prevent autoimmunity and inflammation either when activated naturally or when activated by {alpha}-galactosylceramide ({alpha}GalCer), which is a synthetic ligand for iNKT cells, or related compounds in several animal models. The activation of these cells is beneficial for diabetes, encephalomyelitis, and collagen-induced arthritis in murine models (11) . For example, the number of iNKT cells is low in diabetes-prone nonobese diabetic mice, and the adoptive transfer of iNKT cells mitigates this disease (11) . Paradoxically, iNKT cells exert immune-stimulatory functions in several tumor models by suppressing tumor growth (12) . The adoptive transfer of iNKT cells into J{alpha}18–/– mice, which lack iNKT cells because of the deletion of j{alpha}18, restores the ability to inhibit tumor growth (13 , 14) . In addition, iNKT cells appear to participate in the immune responses of infectious agents such as Mycobacterium tuberculosis, Plasmodium yoelii, and Listeria monocytogenes (15) .

Thus, there are potential advantages to using iNKT cells in immune therapy for cancer, autoimmune diseases, and infection. Although knowledge about iNKT cells is steadily accumulating (7 8 9) , much more remains to be learned about these cells to realize their use in immune therapy, in particular, how they exert both immune-stimulatory and immune-suppressive functions. However, the difficulty in studying iNKT cells lies in their extreme rarity in animals, including humans.

We overcame this difficulty by inducing iNKT cells from ESCs. We successfully induced iNKT cells at high efficiency from ESCs that were prepared by somatic cell nuclear transfer (ntESCs) on OP9 cells that ectopically expressed dlk1 (OP9-dlk1). We characterized the cells both in vitro and in vivo.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Reagents
The {alpha}GalCer was obtained from Kirin Brewery Ltd. (Gunma, Japan). Anti-mouse CD4 (L3T4 RM4–5), CD8 (LY-2 53–6.7), TCRβ (H57–597), TCRVβ8 (F23.1), TCRVβ7 (TR310), TCRVβ2 (B20.6), Gr-1 (RB6–8C5), CD19 (1D3), NK1.1 (PK136), CD24 (M1/69), CD44 (IM7), and CD69 (H1.2F3) monoclonal antibodies were obtained from BD PharMingen (San Jose, CA, USA). Murine CD1d dimer XI (BD PharMingen, Franklin Lakes, NJ, USA) was loaded with vehicle or {alpha}GalCer according to the manufacturer’s instructions.

Cell culture
OP9 cells were obtained from the Rikagaku Kenkyusho (RIKEN) Cell Bank (RIKEN Bioresource Center, Tsukuba, Japan). OP9-dlk1 cells were generated by the transduction of a retrovirus expression vector that encoded the human neural growth factor receptor (NGFR) into OP9 cells, followed by the internal-ribosomal entry site (IRES) linked to the murine dlk1 gene. Coculture of OP9 or OP9-dlk1 cells with ESC1 or ntESC4 cells was performed as described previously (6) , except that 5 ng/ml interleukin 15 (IL-15) was added after culture day 10 to OP9-dlk1 for ntESC4. Typically, beginning with 1 x 105 undifferentiated ntESC4 cells, more than 5 x 107 {alpha}GalCer-CD1d-dimer+ cells were obtained from OP9-dlk1 after 3 wk of culture.

Preparation of m-iNKT cells
An equal number of cells from ESC1 and ntESC4 that were cultured on OP9-dlk1 for 18 days was mixed and cultured for a further 3 days on fresh OP9-dlk1. The resulting {alpha}GalCer-CD1d-dimer+ cells were purified through an LS column (Miltenyi Biotec K.K., Tokyo, Japan) and used as m-iNKT cells for intravenous injection to assess adjuvant effects and tumor protection. The purity of the m-iNKT cells was >99%, as determined by flow cytometry.

RT-PCR
Total RNA was prepared from ntESC4 cells harvested at the indicated time points, using an RNeasy kit (Qiagen, Tokyo, Japan), and cDNA was generated using oligo-dT. The sequences of the primers used to detect transcripts of CD3{epsilon}, GATA3, IL-7R{alpha}, Ig{alpha}, and {lambda}5 were described previously (6) . The following primer sets were used: 5'-CCAAgTggAgCAgAgTCCT-3'/5'- CCAAAATgCAgCCTCCCTAA-3' for V{alpha}14-J{alpha}18; 5'-CCCAgAgAACCACAgAAAAAT-3'/5'-TAACCACCCACAATAACAAAT-3' for Rag2; and 5'-CACAggACTAgAACACCTgC-3'/5'-gCTggTgAAAAggACCTCT-3' for hypoxanthine-guanine phosphoribosyltransferase (HPRT).

For rag2 detection, total RNA was extracted from nt-iNKT cells (the purity of the iNKT cells was >92%) using an RNeasy kit and cDNA was synthesized using oligo dT. The iNKT cells were purified from C57BL/6 and reconstituted Rag2–/– mouse liver mononuclear cells by FACS sorting with {alpha}GalCer-CD1d and anti-TCRβ antibody staining (the purity of iNKT cells was >96%). RNA preparation and cDNA synthesis were performed as described above.

Cytokine measurement
Dendritic cells (DCs; 1x105) were prepared from the spleen of J{alpha}18–/– mice using CD11c-magnetic beads (Miltenyi Biotec K.K.) and pretreated with 200 ng/ml {alpha}GalCer for 7 h before incubation with lymphocytes from ESC1 that were cultured for 21 days on OP9-dlk1 (2x105 cells), lymphocytes from ntESC4 that were cultured for 21 days on OP9-dlk1 (2x105 cells), a mixture of lymphocytes from ESC1 and ntESC4 (nonpurified m-iNKT cells, 2x105 cells), or with C57BL/6 spleen cells (2x105 cells). After 72 h of coculture, the concentration of each cytokine was measured. The interleukin 10 (IL-10) concentration was determined using an OptEIA mIL-10 ELISA kit (BD PharMingen). The interferon-{gamma} (IFN-{gamma}), interleukin 4 (IL-4), and interleukin 13 (IL-13) concentrations were measured using Duoset (R&D Systems, Minneapolis, MN, USA).

Assessment of immune adjuvant effects
The experiments were performed essentially as described previously (16) . Briefly, spleen cells from TAP–/– mice were incubated with hypertonic medium in the presence of 10 mg/ml ovalbumin (OVA) and then incubated with hypotonic medium to induce apoptosis. This cell-associated form of OVA was injected intravenously into mice (2x107 cells/mouse) together with {alpha}GalCer [2 µg/mouse; TOG (TAP–/–/OVA/{alpha}GalCer) immunization]. After 7 days, spleen cells were prepared and stimulated with 1 µM OVA257–264 peptide for 6 h or left unstimulated. IFN-{gamma} production was then monitored by intracellular staining.

For reconstitution in J{alpha}18–/–mice (C57BL/6 background), m-iNKT cells (2x106 cells) or donor iNKT cells (2x106 cells) that were prepared from syngeneic (C57BL/6) or allogeneic (BALB/c) strains of mice were injected intravenously. One hour later, TOG immunization was performed as described above. Seven days after the immunization, the recipient spleen cells were challenged with OVA257–264 for 6 h or left untreated, and IFN-{gamma} production was assessed as described above.

Tumor protection assay
The m-iNKT cells (2x106) were adoptively transferred into J{alpha}18–/– mice, followed by TOG immunization as described above. Seven days later, EG7 or EL4 lymphoma cells (2x106) were inoculated subcutaneously into these mice. Tumor growth was monitored by measuring tumor size.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
ntESCs from iNKT cells
Recently, we created cloned mice from peripheral iNKT cells using somatic cell nuclear transfer and established ntESCs (17) . These ntESCs were unique in that both TCR{alpha} and TCRβ were rearranged inframe in the germ line because the donor iNKT cells were derived from the liver of C57BL/6 x 129/Sv mice. We obtained six different ntESC lines, which we analyzed by Southern blot analysis using probes that specifically detect the rearranged TCRV{alpha}14 and TCRβ loci (17) . Bands of 13 and 2.5 kb corresponded to the nonrearranged C57BL/6 and 129 alleles for TCRV{alpha}14, respectively (17) . All ntESCs harbored the 8-kb band that corresponded to the rearranged C57BL/6-TCRV{alpha}14 locus (Fig. 1 , lanes 1–6); in some, the 2.5-kb band was absent, indicating V{alpha}14 locus rearrangement of the other allele (Fig. 1 , lanes 3 and 4). Persistence of the 13-kb-band in these cells most likely reflects the presence of a pseudogene for V{alpha}14 in a C57BL/6 allele (17) . For the TCRβ loci, two bands that differed in size from the control of 10.4 kb were observed clearly in all ntESCs, indicating that both C57BL/6 and 129 alleles possessed a rearranged configuration in the germ line (Fig. 1 , lanes 1–6) (18) . These ntESCs could generate a chimeric mouse on aggregation (data not shown). Among these ntESCs, we focused on one clone, ntESC4, because it showed relatively high chimerism and harbored the V{alpha}14-J{alpha}18 rearrangement at both alleles (data not shown; Fig. 1 , lane 4).


Figure 1
View larger version (61K):
[in this window]
[in a new window]

  		Figure 1. Southern blots of ntESCs from NKT cells. The ntESCs (lanes 1–6) were prepared as described previously (17) . Genomic DNA (5 µg) from these cells was digested with either EcoRI or BamHI and subjected to Southern blot analysis to detect TCRV{alpha}14 and TCRβ rearrangement, respectively (17) . Bands representing the nonrearranged configuration are indicated by arrows. The presence of an 8-kb band or absence of a 2.5-kb band with the TCRV{alpha}14 probe indicates rearrangement. The nonrearranged 10.4-kb band is absent from TCRβ loci in all ntESCs (18) . Lane 4 is ntESC4.

Generation of nt-iNKT cells
We used ntESC4 as a source for in vitro iNKT cell differentiation. We hypothesized that iNKT cells diverge from the pool of conventional T cells, regardless of the fact that they express NK cell markers (19 , 20) and that the inframe V{alpha}14-J{alpha}18 configuration in the germ line may have a strong effect on the development of these cells, as shown in V{alpha}14-J{alpha}18 transgenic mice and the progeny of a cloned mouse from iNKT cells (18 , 21) . Because OP9-dlk1 has been used for successful T cell induction from ESCs, we used this method to induce iNKT cells (6) .

To examine whether ntESC4 cells cultured on OP9 or OP9-dlk1 could be induced to differentiate into B or T cells, respectively, we analyzed, in parallel, the differentiation of ESC1, an ESC line derived from a normal fertilized egg. Although ESC1 and ntESC4 gave rise to erythromyeloid cells expressing Gr-1 or Ter 119 on both OP9 and OP9-dlk1 by the middle stage of differentiation (~2 wk), these erythromyeloid cells steadily decreased, and lymphocytes became predominant at a later stage of differentiation (~3 wk; Fig. 2 A, B). At day 23, both ESC1 and ntESC4 differentiated on control OP9 gave rise to a population of cells expressing CD19, which thus likely belonged to the B cell lineage. In contrast, both ESC1 and ntESC4 failed to generate such cells on OP9-dlk1 (Fig. 2A , left). Furthermore, ESC1 and ntESC4 cultured on control OP9 failed to give rise to T lymphocytes, as judged by the absence of cells expressing TCRβ (data not shown). In contrast, ESC1 gave rise to T lymphocytes mostly consisting of CD4+CD8+ double-positive (DP) T cells on OP9-dlk1, as previously reported (6) . Similarly, CD8+ single-positive (SP) and DP cells were induced from ntESC4 on OP9-dlk1 (Fig. 2A , right panel). These results indicate that ntESC4 has the potential to generate both B and T cell lineages in response to environmental cues.


Figure 2
View larger version (33K):
[in this window]
[in a new window]

  		Figure 2. Generation and characterization of nt-iNKT from ntESC4. A) Generation of B and T cells. ESC1 and ntESC4 cells were cultured on OP9 or OP9-dlk1 cells for 23 days in the presence of the appropriate cytokines. Cells were stained with the indicated monoclonal antibody and analyzed with FACSCalibur. The percentage of cells in each quadrant is shown. Representative data from one of five experiments are shown. B) Generation of iNKT cells from ntESC4. Top: forward scatter (FSC) and side scatter (SSC) profiles of cells generated from ESC1 and ntESC4 cells cultured on OP9-dlk1 for 23 days are shown. The number in the panel indicates the percentage of lymphocytes. Bottom: cells were stained with fluorescent-labeled vehicle-loaded-CD1d dimer (v-CD1d) or {alpha}GalCer-loaded-CD1d dimer ({alpha}GalCer-CD1d) and anti-TCRβ. Lymphocyte-gated cells were analyzed after removing dead cells. The percentage of iNKT ({alpha}GalCer-CD1d dimer+/TCRβ+) cells is shown. C) Allelic exclusion of TCRVβ. Cells in B from ntESC4s were stained with different fluorescent-colored anti-TCRVβ8, 7, 5, or 2, and nt-iNKT cells were gated for analysis of the TCRVβ repertoire. D) Coreceptor expression of nt-iNKT cells. The ntESC4 cells cultured on OP9-dlk1 as described in A were stained with {alpha}GalCer-CD1d dimer, anti-TCRβ, and the indicated monoclonal antibodies. The histogram shows the relative number of cells expressing the indicated marker among the nt-iNKT cells. The number in the figure is the percentage of positive cells (bold line) relative to the isotype control-stained cells (broken line). E) Gene expression analysis of ntESC4 during in vitro culture. The ntESC4 cells cultured on OP9 or OP9-dlk1 were harvested at the indicated times, and cDNA was synthesized as described in Materials and Methods. Threefold serial dilutions of the cDNA were normalized to HPRT and analyzed with gene-specific primers to detect the indicated genes relevant to T and B cell development. F) Cytokine production from nt-iNKT cells on stimulation. Cytokine production was assessed as described in Materials and Methods. In each experiment, the responder cells (cultured on OP9-dlk1 for 21 days or from C57BL/6 spleen) were cocultured with or without stimulator dendritic cells ({alpha}GalCer-pulsed CD11c+ cells from J{alpha}18–/– mice). Responder cells are lymphocytes from ESC1 cells (lanes 1 and 6), whole spleen cells from C57BL/6 (lanes 2 and 7), a mixture of lymphocytes from ESC1 and ntESC4 cultured for a further 3 days as described in Materials and Methods (~30% nt-iNKT cells; lanes 3 and 8), or lymphocytes from ntESC4 (~92% nt-iNKT cells; lanes 4 and 9). Data are expressed as the mean ± SD of four experiments.

We next examined whether T cells derived from ESC1 or ntESC4 contained iNKT. Because {alpha}GalCer binds CD1d, a nonclassical MHC class I-like molecule, and the resulting complex is an agonist that avidly binds to iNKT cells, the presence of iNKT cells was detected using {alpha}GalCer-loaded CD1d-dimer XI as an {alpha}GalCer-CD1d-dimer+/TCRβ+ subset (7) . The T cells generated from ESC1 contained few iNKT cells, whereas T cells induced from ntESC4 on OP9-dlk1 consisted mainly of {alpha}GalCer-CD1d-dimer+/TCRβ+ cells (hereafter, nt-iNKT cells), which occupied up to 92% of the lymphocyte-gated population (Fig. 2B ). Intriguingly, most of these cells appeared to be TCRβhigh cells, whereas T cells derived from ESC1 consisted of cells that expressed a different level of TCRβ, similar to that observed in the thymus (Fig. 2B ), suggesting that the nt-iNKT cells most likely composed a homogenous population of relatively mature cells.

Because the use of TCRVβ for iNKT cells is highly biased toward Vβ8, Vβ7, and Vβ2 (7 8 9) , we analyzed the TCR repertoire using flow cytometry. The nt-iNKT cells showed an allelic exclusion for Vβ8 (Fig. 2C ).

Characterization of nt-iNKT cells
The iNKT cells express coreceptors such as CD44, CD69, and NK1.1, in addition to V{alpha}14-J{alpha}18, and the expression levels of these markers reflect the developmental stage of the cell (22 , 23) . We analyzed the expression of these coreceptors together with CD4, CD8, and CD24 (heat stable antigen, HSA, an immature cell marker) in nt-iNKT cells to assess nt-iNKT cell maturity. Flow cytometry indicated that 20% of nt-iNKT cells expressed CD4 and 45% expressed CD8 (Fig. 2D ). Interestingly, most nt-iNKT cells expressed CD24, although they were CD44low and harbored little CD69 and NK1.1 (Fig. 2D ). Thus, nt-iNKT cells that are generated on OP9-dlk1 represent an early developmental stage of iNKT cells in the thymus (22 , 23) .

To visualize the molecular process that results in lineage restriction by coordinating the expression of lineage-specific genes, we assessed the expression of genes relevant to T and B cell development during the ontogeny of ntESC4 differentiation on control OP9 and OP9-dlk1 cells using semiquantitative RT-PCR (6 , 24) . The transcript for V{alpha}14-J{alpha}18 was first detected on day 12, at which time, factors responsible for T cell development such as Cd3{epsilon} and Gata3 appeared on OP9-dlk1 cells (Fig. 2E ). In contrast, transcripts for {lambda}5 and Ig{alpha}, which are expressed during the early stages of B cell development, were present in control OP9 cells (Fig. 2E ). Transcripts for the interleukin 7 receptor (IL-7R), which is required for the survival and proliferation of lymphocyte progenitors, and transcripts for rag2, which is essential for B and T cell receptor gene rearrangement, were detected when B and T lymphopoiesis first became apparent by flow cytometry (Fig. 2E ). These data mirrored the expression profile of lineage-specific transcription factors observed in ESCs during lymphocyte induction (6) .

The iNKT cells produce both TH-1- and TH-2-type cytokines on stimulation with {alpha}GalCer-pulsed dendritic cells ({alpha}GalCer-DCs) (10 , 11) . We assessed this ability by measuring the production of IFN-{gamma}, IL-4, IL-10, and IL-13 from nt-iNKT cells. Little cytokine was produced when ESC1-derived DP T cells, iNKT cells isolated from mice, or nt-iNKT cells were cultured in the absence of {alpha}GalCer-DCs (Fig. 2F , lanes 1–4). The addition of {alpha}GalCer-DCs to DP T cells from ESC1 produced no cytokine (Fig. 2F , lane 6). In contrast, the culture of whole spleen cells containing iNKT cells or of nt-iNKT cells with {alpha}GalCer-DCs resulted in the production of these cytokines (Fig. 2F , lanes 7–9). It is noteworthy that nt-iNKT cells that were cocultured with ESC1-derived DP T cells for 3 days produced more cytokines than did nt-iNKT cells alone (Fig. 2F , compare lanes 8 and 9).

In vivo maturation of nt-iNKT cells
T cells generated from ESCs on OP9-dlk1 require further maturation using fetal thymic organ culture to mount an immune response in immunocompromised animals on adoptive transfer (6) . In contrast to conventional T cells, iNKT cells are self-reactive and are thought to be positively selected by agonist ligands in the thymus (25) . Accordingly, the expression of CD44 and NK1.1 in iNKT cells appears following their selective enrichment in the thymus on agonist ligand engagement (19 , 22) . Because nt-iNKT cells represent an immature phenotype harboring CD24highCD44lowCD69<–NK1.1<– (Fig. 2D ), we sought to further mature these cells in vivo. Given that CD1d is expressed in many tissues besides the thymus, we hypothesized that nt-iNKT cells could mature in the periphery, and thus we injected these cells intravenously into Rag2–/– mice (26) . Flow cytometric analysis indicated that the nt-iNKT cells rooted in the recipient mice 2 wk after injection, occupying up to 30% of liver mononuclear cells and 0.6% of total spleen cells (Fig. 3 A, top). However, few or no nt-iNKT cells were detected in the thymus (data not shown). The examination of the cell surface markers on nt-iNKT cells from the recipient mice indicated that the expression of CD44, CD69, and NK1.1 was up-regulated concomitant with the decrease in CD24 (HSA), suggesting that the nt-iNKT cells acquired the mature phenotype. However, few CD4+ SP nt-iNKT cells were present, although the percentage of CD8+ SP nt-iNKT cells remained unaltered in the recipient (compare Figs. 2D and 3A ). These results demonstrate that nt-iNKT cells become CD8+ SP or CD4<–CD8<– cells in the recipient. Whereas most iNKT cells that are found in vivo are either CD4+ SP or CD4<–CD8<–, a minute population of CD8+ SP cells exists in the spleen, liver, and lymph nodes (18 , 27) . Thus, our protocol succeeded in inducing the latter two cell types, but failed to induce CD4+ SP iNKT cells.


Figure 3
View larger version (14K):
[in this window]
[in a new window]

  		Figure 3. In vivo maturation of nt-iNKT cells. A) Change in coreceptor expression in nt-iNKT cells. The nt-iNKT cells (1x107) from ntESC4 were cultured for 23 days and then injected intravenously into Rag2–/– mice (C57BL/6 background, obtained from Jittyu-ken, Kanagawa, Japan). Two weeks later, liver mononuclear cells and spleen cells were stained with fluorescently labeled anti-TCRβ, {alpha}GalCer-CD1d dimer, and the indicated monoclonal antibodies. Top: profile and percentage of nt-iNKT cells in the liver and spleen. Bottom: histograms show the relative number of cells expressing the indicated marker among the nt-iNKT cells. The number in the figure gives the percentage of positively stained cells (bold line) relative to the isotype control-stained cells (broken line). One representative experiment of four (3 mice/experiment) is shown. B) Disappearance of the RAG2 transcript. RAG2 transcript was detected using semiquantitative RT-PCR. Threefold serial dilutions of cDNA were normalized to HPRT. Lanes 1–3, iNKT cells purified from C57BL/6 liver mononuclear cells. Lanes 4–6, nt-iNKT cells at culture day 21. Lanes 7–9, nt-iNKT cells purified from adoptively transferred Rag2–/– mouse liver after 2 wk.

To further confirm that nt-iNKT cells mature in the recipient, we examined the expression of Rag2, which is a hallmark of developing T cells. Semiquantitative RT-PCR indicated that Rag2 expression was below the detection limit in nt-iNKT cells that were recovered from the recipient liver (Fig. 3B , lanes 7–9), whereas it was present at the end of culture (Fig. 3B , lanes 4–6). Thus, nt-iNKT cells mature per se in the recipient animal on adoptive transfer, without any extrinsic stimuli.

Adjuvant effects of m-iNKT cells
The {alpha}GalCer acts as an adjuvant for CD8+ T cell-dependent protection against malaria infection when it is immunized with an irradiated sporozoite vaccine (28) . This adjuvant effect is mediated through the maturation of DC by {alpha}GalCer via iNKT cells, which helps to prime antigen-specific TH-1 CD4+ and CD8+ T cell-mediated immunity (16) . We investigated whether nt-iNKT cells have such potential by measuring the antigen-specific production of IFN-{gamma} on immunization with the antigen and {alpha}GalCer, followed by restimulation with the antigen. Because IFN-{gamma} production from nt-iNKT cells was enhanced by mixing ESC1- and ntESC4-derived lymphocytes (Fig. 2F , IFN-{gamma}, compare lanes 8 and 9), we used column-purified {alpha}GalCer-CD1dimer+-iNKT cells from the mixture in the following experiments (hereafter, m-iNKT cells). A cell-associated form of OVA was used as an antigen (29) . Immunization with TOG (the cell-associated form of OVA and {alpha}GalCer) followed by OVA257–264 challenge led to the production of IFN-{gamma} from CD8+ T cells (~1.6%), whereas little IFN-{gamma} was produced in the absence of TOG immunization or OVA257–264 challenge in control mice (Fig. 4 A, WT). The adoptive transfer of syngeneic or allogeneic iNKT cells into J{alpha}18–/– mice resulted in the production of IFN-{gamma} in an OVA257–264-dependent manner, although the latter was not as potent as the former (Fig. 4A , compare syn-iNKT and allo-iNKT). Similarly, the reconstitution of J{alpha}18–/– mice with m-iNKT cells followed by immunization with TOG led to the production of IFN-{gamma} on OVA257–264 challenge (Fig. 4A , m-iNKT). The efficacy of m-iNKT cells in inducing IFN-{gamma} was as potent as that of syngeneic iNKT cells (Fig. 4A ).


Figure 4
View larger version (11K):
[in this window]
[in a new window]

  		Figure 4. Adjuvant effect of m-iNKT cells. A) Induction of IFN-{gamma} on OVA challenge. Control mice (WT) with or without TOG immunization were stimulated with or without OVA257–264 peptide as described in Materials and Methods. J{alpha}18–/– mice reconstituted with the indicated donor cells were immunized with TOG and were restimulated with or without OVA257–264 peptide. The percentage of cells producing IFN-{gamma} is shown. One representative experiment of four is shown. B) Tumor growth inhibition. WT mice with or without TOG immunization received EG7 (left, solid diamond and solid rectangle) or EL4 (right, open diamond and open rectangle), as described in Materials and Methods. Similarly, J{alpha}18–/– mice reconstituted with m-iNKT cells and immunized with TOG were inoculated with EG7 (solid triangle) or EL4 (open triangle). The mean tumor size at days 0, 6, 9, 12, and 15 after tumor inoculation is plotted (n=5). The SD at each time was less than 10%. *EG7 became apparent in one of five mice at day 15.

To further explore the utility of OVA-specific CD8+ T cells in vivo, we assessed the ability of m-iNKT cells to block inoculated tumor growth on adoptive transfer into J{alpha}18–/– mice because OVA-specific CD8+ T cells have been shown to confer resistance against OVA-expressing tumors (16) . We used EG7, which is a murine thymoma that has been transduced to express OVA in a secretory form. TOG immunization prior to tumor inoculation prevented tumor growth as expected, whereas tumor growth occurred in the absence of TOG immunization in control mice (Fig. 4B , left). The inoculation of EG7 cells into J{alpha}18–/– mice that had been reconstituted with m-iNKT cells and immunized with TOG resulted in little tumor growth (Fig. 4B , left). In contrast, the inoculation of OVA-nontransduced parental thymoma EL4 cells led to tumor growth in mice, despite TOG immunization. Similar results were obtained in J{alpha}18–/– mice reconstituted with m-iNKT cells, regardless of TOG immunization (Fig. 4B , right). Thus, m-iNKT cells mount an antigen-specific adjuvant effect in conjunction with {alpha}GalCer in vivo.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Although the main advantage of using ntESCs is that cells that differentiate from these ntESCs show little donor/recipient incompatibility (1) , we demonstrated another feature of ntESCs: the inframe V{alpha}14-J{alpha}18TCR locus in the germ line had a large effect on the development of iNKT cells from ntESCs. The absence of iNKT cells among T lymphocytes induced from ESC1 indicates that V{alpha}14 and J{alpha}18 rearrangement, which determines the identity of iNKT cells, is the result of random recombination. In this regard, the relatively high abundance of these cells in murine thymocytes is most likely caused by the selective proliferation on interaction with CD1d (Fig. 2B ; ref. 19 ). However, the mechanisms by which the inframe V{alpha}14-J{alpha}18TCR locus instructs progenitor cells to become iNKT cells remain to be clarified, regardless of the fact that nascent inframe TCR{alpha} chains are often deleted and subjected to further rearrangement during positive selection. Comparative analyses of the V{alpha}14 and J{alpha}18TCR loci between ntESC4 and normal ESCs during differentiation in terms of chemical modification of the histones, methylation status of DNA, and changes in higher-order chromatin structure would help to elucidate the molecular mechanisms underlying this unique feature.

TH-1- and TH-2-type cytokine production from nt-iNKT cells on {alpha}GalCer-DC stimulation reflects the fact that proper signals emanated from their cognate TCRV{alpha}14-J{alpha}18 (Fig. 2F ). Furthermore, the maturation of nt-iNKT cells in terms of cytokine production (Fig. 2F ) may reflect the fact that the interaction between nt-iNKT cells and DP T cells from ESC1 reinforces cytokine production, especially for IFN-{gamma}, because DP T cells are rich in CD1d (19 , 30 , 31 , and unpublished results). It is noteworthy that the cytokine production profile changes from TH-2 to TH-1 as iNKT cells mature (22) .

Data from the adoptive transfer experiments strongly suggest that no thymic environment is required and that endogenous ligands presented on CD1d outside the thymus are responsible for further maturation of nt-iNKT cells (compare Figs. 2D and 3A ). Such ligands would elicit signals to extinguish rag2 expression. However, whether the up-regulation of CD44 and NK1.1 concomitant with the down-regulation of CD24 is solely dependent on CD1d remains to be determined. Similarly, the disappearance of DP nt-iNKT cells on adoptive transfer into Rag2–/– mice most likely reflects a maturation process, and these cells may represent a transitional state of iNKT cells originating from DP T cells en route to becoming DN and/or CD4SP iNKT cells (19 , 20) .

The failure to generate CD4+ SP T cells from ESCs has been reported previously, and the authors attributed this failure to the absence of MHC class II expression in OP9 (6) . Nonetheless, because the development of iNKT cells is dependent on CD1d, but not on MHC class II, the reason for the failure to generate CD4+SP iNKT cells remains enigmatic. Alternatively, it is possible that CD4+SP iNKT cells originate from DN iNKT cells.

The role of iNKT cells as a mediator of adjuvant effect has been reported; one mechanism underling this phenomenon is the ability of iNKT cells to mature DCs, which is dependent on {alpha}GalCer (16) . Recent studies have described a system in which potent T cell-mediated immunity can be induced against an inoculated tumor that is poorly immunogenic on {alpha}GalCer-loaded live or inactivated tumor cell immunization (32 , 33) . Because iNKT cells play a pivotal role in such tumor immunity, successful application of this technology in clinical cancer therapy requires the intact function of iNKT cells in patients, particularly, an ability to produce IFN-{gamma}. There is, however, no guarantee that these iNKT cells retain such an ability. In this regard, the nt-iNKT cells that were developed here will be more useful because they secrete cytokines such as IFN-{gamma}, which are required to combat tumors in an antigen-dependent manner (Fig. 4A, B ). Although {alpha}GalCer is beneficial for DC maturation and to enhance the immune system, as described above, it induces anergy in iNKT cells, which results in adverse effects in cancer treatments. The iNKT cell anergy induced by {alpha}GalCer in vivo exacerbates cancer metastasis (34) . In this regard, the use of nt-iNKT cells could alleviate such problems. Given that CD1d-dependent NKT cells comprise type I (V{alpha}14-J{alpha}18-expressing cells) and type II (cells expressing TCRs other than V{alpha}14-J{alpha}18) cells and that the former enhances antitumor responses, whereas the latter suppresses these responses, the ability of m-iNKT cells to suppress tumor growth shown here is consistent with their proposed function (35 , 36) . In conjunction with the fact that {alpha}GalCer acts as an adjuvant through m-iNKT cells, we conclude that nt-iNKT cells mirror at least the immune-stimulating function of iNKT cells that are found in vivo and are transplantable.

The advent of nt-iNKT cells will boost research on iNKT cells with regard to understanding their developmental mode at the molecular level and will shed much light on their seemingly contradictory roles in immune responses (10) . The fact that nt-iNKT cells produce both TH-1 and TH-2 cytokines may circumvent a problem inherent to previously reported iNKT cells in V{alpha}14-J{alpha}18 Tg mouse, i.e., that they fail to induce IFN-{gamma} on appropriate stimulation (21) . The quasi-infinite availability of nt-iNKT will also be beneficial to biochemically identify the cellular components that are engaged in the TCR signal emanating from the V{alpha}14-J{alpha}18TCR-CD1d interaction. Furthermore, the generation of CD8+ iNKT cells as shown here will shed much light on their role in the immune system because these cells are extremely rare in wild-type mice. Such new knowledge will open the door to therapeutic applications for conditions and diseases such as transplant tolerance, autoimmune disease, cancer, and infection, in which iNKT cells are thought to play an important role.


   ACKNOWLEDGMENTS
 
We thank H. Fujimoto for FACS sorting, M. Iida and Y. Nagata for technical assistance, Drs. K. Shimizu and S. Fujii for the experiments in Fig. 4 , Dr. Atsuro Ogura (RIKEN, Bioresource Center, Tsukuba, Japan) for ntESCs, Drs. Hiroshi Kawamoto and Kyoko Masuda (RIKEN, Research Center for Allergy and Immunology, Yokohama, Japan) for OP9-dlk1, and RIKEN for lab facilities. This work was supported by RIKEN, and RIKEN reserves all rights on ntESCs.


   FOOTNOTES
 
2 Present address: Pharmaceutical and Medical Devices Agency, 3-3-2 Kasumigaseki, Chiyoda-ku, Tokyo 100-0013, Japan.

Received for publication December 18, 2007. Accepted for publication February 14, 2008.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Keller, G. (2005) Embryonic stem cell differentiation: emergence of a new era in biology and medicine. Genes Dev. 19,1129-1155[Abstract/Free Full Text]
  2. Doetschman, T. C., Eistetter, H., Katz, M., Schmidt, W., Kemler, R. (1985) The in vitro development of blastocyst-derived embryonic stem cell lines: formation of visceral yolk sac, blood islands and myocardium. J. Embryol. Exp. Morphol. 87,27-45[Medline]
  3. Nakano, T., Kodama, H., Honjo, T. (1994) Generation of lymphohematopoietic cells from embryonic stem cells in culture. Science 265,1098-1101[Abstract/Free Full Text]
  4. Nakano, T. (1995) Lymphohematopoietic development from embryonic stem cells in vitro. Sem. Immunol. 7,197-203[CrossRef][Medline]
  5. Cho, S. K., Webber, T. D., Carlyle, J. R., Nakano, T., Lewis, S. M., Zuniga-Pflucker, J. C. (1999) Functional characterization of B lymphocytes generated in vitro from embryonic stem cells. Proc. Natl. Acad. Sci. U. S. A. 96,9797-9802[Abstract/Free Full Text]
  6. Schmitt, T. M., de Pooter, R. F., Gronski, M. A., Cho, S. K., Ohashi, P. S., Zuniga-Pflucker, J. C. (2004) Induction of T cell development and establishment of T cell competence from embryonic stem cells differentiated in vitro. Nat. Immunol. 5,410-417[CrossRef][Medline]
  7. Kronenberg, M. (2005) Toward an understanding of NKT cell biology: progress and paradoxes. Annu. Rev. Immunol. 23,877-900[CrossRef][Medline]
  8. Taniguchi, M., Harada, M., Kojo, S., Nakayama, T., Wakao, H. (2003) The regulatory role of Valpha14 NKT cells in innate and acquired immune response. Annu. Rev. Immunol. 21,483-513[CrossRef][Medline]
  9. Bendelac, A., Rivera, M. N., Park, S. H., Roark, J. H. (1997) Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu. Rev. Immunol. 15,535-562[CrossRef][Medline]
  10. Kronenberg, M., Rudensky, A. (2005) Regulation of immunity by self-reactive T cells. Nature 435,598-604[CrossRef][Medline]
  11. Godfrey, D. I., Kronenberg, M. (2004) Going both ways: immune regulation via CD1d-dependent NKT cells. J. Clin. Invest. 114,1379-1388[CrossRef][Medline]
  12. Smyth, M. J., Crowe, N. Y., Hayakawa, Y., Takeda, K., Yagita, H., Godfrey, D. I. (2002) NKT cells—conductors of tumor immunity?. Curr. Opin. Immunol. 14,165-171[CrossRef][Medline]
  13. Stewart, T. J., Smyth, M. J., Fernando, G. J., Frazer, I. H., Leggatt, G. R. (2003) Inhibition of early tumor growth requires Jalpha18-positive (natural killer T) cells. Cancer Res. 63,3058-3060[Abstract/Free Full Text]
  14. Crowe, N. Y., Smyth, M. J., Godfrey, D. I. (2002) A critical role for natural killer T cells in immunosurveillance of methylcholanthrene-induced sarcomas. J. Exp. Med. 196,119-127[Abstract/Free Full Text]
  15. Godfrey, D. I., Hammond, K. J., Poulton, L. D., Smyth, M. J., Baxter, A. G. (2000) NKT cells: facts, functions and fallacies. Immunol. Today 21,573-583[CrossRef][Medline]
  16. Fujii, S., Shimizu, K., Smith, C., Bonifaz, L., Steinman, R. M. (2003) Activation of natural killer T cells by alpha-galactosylceramide rapidly induces the full maturation of dendritic cells in vivo and thereby acts as an adjuvant for combined CD4 and CD8 T cell immunity to a coadministered protein. J. Exp. Med. 198,267-279[Abstract/Free Full Text]
  17. Inoue, K., Wakao, H., Ogonuki, N., Miki, H., Seino, K., Nambu-Wakao, R., Noda, S., Miyoshi, H., Koseki, H., Taniguchi, M., Ogura, A. (2005) Generation of cloned mice by direct nuclear transfer from natural killer T cells. Curr. Biol. 15,1114-1118[CrossRef][Medline]
  18. Wakao, H., Kawamoto, H., Sakata, S., Inoue, K., Ogura, A., Wakao, R., Oda, A., Fujita, H. (2007) A novel mouse model for invariant NKT cell study. J. Immunol. 179,3888-3895[Abstract/Free Full Text]
  19. Gapin, L., Matsuda, J. L., Surh, C. D., Kronenberg, M. (2001) NKT cells derive from double-positive thymocytes that are positively selected by CD1d. Nat. Immunol. 2,971-978[CrossRef][Medline]
  20. Egawa, T., Eberl, G., Taniuchi, I., Benlagha, K., Geissmann, F., Hennighausen, L., Bendelac, A., Littman, D. R. (2005) Genetic evidence supporting selection of the Valpha 14iNKT cell lineage from double-positive thymocyte precursors. Immunity 22,705-716[CrossRef][Medline]
  21. Bendelac, A., Hunziker, R. D., Lantz, O. (1996) Increased interleukin 4 and immunoglobulin E production in transgenic mice overexpressing NK1 T cells. J. Exp. Med. 184,1285-1293[Abstract/Free Full Text]
  22. Benlagha, K., Kyin, T., Beavis, A., Teyton, L., Bendelac, A. (2002) A thymic precursor to the NK T cell lineage. Science 296,553-555[Abstract/Free Full Text]
  23. Benlagha, K., Wei, D. G., Veiga, J., Teyton, L., Bendelac, A. (2005) Characterization of the early stages of thymic NKT cell development. J. Exp. Med. 202,485-492[Abstract/Free Full Text]
  24. Rothenberg, E. V., Dionne, C. J. (2002) Lineage plasticity and commitment in T-cell development. Immunol. Rev. 187,96-115[CrossRef][Medline]
  25. Baldwin, T. A., Hogquist, K. A., Jameson, S. C. (2004) The fourth way? Harnessing aggressive tendencies in the thymus. J. Immunol. 173,6515-6520[Abstract/Free Full Text]
  26. Bradbury, A., Calabi, F., Milstein, C. (1990) Expression of CD1 in the mouse thymus. Eur. J. Immunol. 20,1831-1836[Medline]
  27. Matsuda, J. L., Naidenko, O. V., Gapin, L., Nakayama, T., Taniguchi, M., Wang, C. R., Koezuka, Y., Kronenberg, M. (2000) Tracking the response of natural killer T cells to a glycolipid antigen using CD1d tetramers. J. Exp. Med. 192,741-754[Abstract/Free Full Text]
  28. Gonzalez-Aseguinolaza, G., Van Kaer, L., Bergmann, C. C., Wilson, J. M., Schmieg, J., Kronenberg, M., Nakayama, T., Taniguchi, M., Koezuka, Y., Tsuji, M. (2002) Natural killer T cell ligand alpha-galactosylceramide enhances protective immunity induced by malaria vaccines. J. Exp. Med. 195,617-624[Abstract/Free Full Text]
  29. Liu, K., Iyoda, T., Saternus, M., Kimura, Y., Inaba, K., Steinman, R. M. (2002) Immune tolerance after delivery of dying cells to dendritic cells in situ. J. Exp. Med. 196,1091-1097[Abstract/Free Full Text]
  30. Coles, M. C., Raulet, D. H. (2000) NK1.1+ T cells in the liver arise in the thymus and are selected by interactions with class I molecules on CD4+CD8+ cells. J. Immunol. 164,2412-2418[Abstract/Free Full Text]
  31. Bendelac, A. (1995) Positive selection of mouse NK1+ T cells by CD1-expressing cortical thymocytes. J. Exp. Med. 182,2091-2096[Abstract/Free Full Text]
  32. Shimizu, K., Kurosawa, Y., Taniguchi, M., Steinman, R. M., Fujii, S. (2007) Cross-presentation of glycolipid from tumor cells loaded with alpha-galactosylceramide leads to potent and long-lived T cell mediated immunity via dendritic cells. J. Exp. Med. 204,2641-2653[Abstract/Free Full Text]
  33. Shimizu, K., Goto, A., Fukui, M., Taniguchi, M., Fujii, S. (2007) Tumor cells loaded with alpha-galactosylceramide induce innate NKT and NK cell-dependent resistance to tumor implantation in mice. J. Immunol. 178,2853-2861[Abstract/Free Full Text]
  34. Parekh, V. V., Wilson, M. T., Olivares-Villagomez, D., Singh, A. K., Wu, L., Wang, C. R., Joyce, S., Van Kaer, L. (2005) Glycolipid antigen induces long-term natural killer T cell anergy in mice. J. Clin. Invest. 115,2572-2583[CrossRef][Medline]
  35. Ambrosino, E., Terabe, M., Halder, R. C., Peng, J., Takaku, S., Miyake, S., Yamamura, T., Kumar, V., Berzofsky, J. A. (2007) Cross-regulation between type I and type II NKT cells in regulating tumor immunity: a new immunoregulatory axis. J. Immunol. 179,5126-5136[Abstract/Free Full Text]
  36. Terabe, M., Berzofsky, J. A. (2007) NKT cells in immunoregulation of tumor immunity: a new immunoregulatory axis. Trends Immunol. 28,491-496[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
fj.07-104687v1
22/7/2223    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wakao, H.
Right arrow Articles by Fujita, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wakao, H.
Right arrow Articles by Fujita, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS