ABSTRACT
A critical step during intrathymic T-cell development is the transition of CD4+ CD8+ double-positive (DP) cells to the major histocompatibility complex class I (MHC-I)-restricted CD4− CD8+ and MHC-II-restricted CD4+ CD8− single-positive (SP) cell stage. Here, we identify a novel gene that is essential for this process. Through the T-cell phenotype-based screening of N-ethyl-N-nitrosourea (ENU)-induced mutant mice, we established a mouse line in which numbers of CD4 and CD8 SP thymocytes as well as peripheral CD4 and CD8 T cells were dramatically reduced. Using linkage analysis and DNA sequencing, we identified a missense point mutation in a gene, E430004N04Rik (also known as themis), that does not belong to any known gene family. This orphan gene is expressed specifically in DP and SP thymocytes and peripheral T cells, whereas in mutant thymocytes the levels of protein encoded by this gene were drastically reduced. We generated E430004N04Rik-deficient mice, and their phenotype was virtually identical to that of the ENU mutant mice, thereby confirming that this gene is essential for the development of SP thymocytes.
The differentiation step from the double-positive (DP) to single-positive (SP) thymocyte stage is critically regulated by signals originating from the T-cell receptor α/β (TCRα/β) expressed on their surface (3, 5, 16, 17). By using reverse genetic approaches by knocking out or overexpressing various genes that are expected to be involved in TCR signaling, including its ligand major histocompatibility complex molecules and coreceptors CD4 and CD8, the roles of these genes in T-cell development have been investigated intensively (11, 12). However, to identify totally unknown mechanisms in T-cell development, the forward genetic approach is required. N-ethyl-N-nitrosourea (ENU) is a potent mutagen that randomly induces point mutations throughout the genome in a dose-dependent manner, and ENU mutagenesis has been a representative forward genetic strategy (4, 15). We have been screening phenotypes of ENU-mutagenized mice, focusing on defects in T-cell development.
MATERIALS AND METHODS
Mice.C57BL/6 (B6) and B6Ly5.1 congenic mice were purchased from CLEA Japan, Inc. All mice were maintained in the animal facility at the RIKEN Research Center for Allergy and Immunology, and all experiments were done in accordance with institutional guidelines for animal care.
ENU mutagenesis.ENU was administered to male C57BL/6J mice, and their sperm was mated to wild-type eggs and preserved as founder embryos (8, 21) for use in our screening.
Mapping and sequencing.Phenodeviants that showed CD3+ cell reduction were crossed to wild-type C3H/HeJ mice to test for phenotype transmission and for the genetic mapping of the causative genes. Single-nucleotide polymorphism (SNP) mapping was performed as described elsewhere (13). For sequencing, we focused on the ptprk and E430004N04Rik genes, which were among candidate genes listed by PosMed (21) (http://omicspace.riken.jp). cDNA and genomic DNA were amplified by PCR and sequenced using E430004N04Rik complementary primers (sequences are available on request).
Antibodies.The following antibodies were purchased from BD Pharmingen: Ly5.1 (A20), Ly5.2 (104), CD8 (53-6.7), CD4 (H129.19), CD3σ (145-2C11), TCRβ (H57-597), CD24 (M1/69), CD69 (H1.2F3), c-Kit (2B8), Sca-1 (D7), erythroid lineage cells (TER119), Mac-1 (M1/70), Gr-1 (RB6-8C5), CD11c (HL3), B220 (RA3-6B2), NK1.1 (PK136), and CD19 (1D3). TER119, Mac-1, Gr-1, B220, CD19, NK1.1, CD3ε, CD4, and CD8 were used as Lin markers. For immunohistochemistry, the following antibodies were used: anti-FLAG monoclonal antibody (M2; Sigma) and the thymic medullary marker ER-TR5 (donated by W. van Ewijk) as the primary antibody, followed by goat anti-rabbit immunoglobulin G-Texas Red conjugate. 4′,6′-diamidino-2-phenylindole (DAPI) was purchased from Molecular Probes.
Rabbit polyclonal antibodies to themis were raised against a peptide corresponding to mouse themis, amino acids 204 to 220.
NKT cell and regulatory T-cell staining.NKT cells were stained with α-galactosylceramide loaded on a CD1d dimer as previously shown (20). Anti-Foxp3 (FJK-16S) antibody was from eBioscience, and intracellular Foxp3 staining was performed according to the manufacturer's instructions (eBioscience).
Proliferation assay.To obtain CD4SP and CD8SP cells, red blood cells were lysed using a red blood cell lysing buffer (Sigma) followed by the depletion of B220 and Mac-1 cells with magnetic beads. CD4SP cells and CD8SP cells then were sorted by a FACSVantage (Becton & Dickinson) and stained with 6-carboxyfluorescein diacetate succinimidyl ester (CFSE; Wako Pure Chemical Industries, Ltd.). Ninety-six-well flat-bottom plates were incubated with 1 mg/ml 2C11 (anti-CD3ε) and 37.51 (anti-CD28) at 4°C overnight. Cells were washed twice with phosphate-buffered saline (PBS) just before being seeded onto the plates.
RT-PCR.For reverse transcription-PCR (RT-PCR), total RNA of 1 × 106 DP thymocytes was extracted using RNeasy (Qiagen), and cDNA was synthesized by a SuperScript III first-strand synthesis system (Invitrogen). Primers used are the following: β-actin sense, 5′-TCCTGTGGCATCCATGAAACT-3′; β-actin antisense, 5′-GAAGCACTTGCGGTGCACGAT-3′; themis sense, 5′-AATGATGTGGCTGATGTGGA-3′; themis antisense, 5′-AATGATGTGGCTGATGTGGA-3′.
PCR products were subjected to electrophoresis through 1.5% agarose gels and stained with ethidium bromide.
Immunohistochemistry.Tissues were fixed with 4% paraformaldehyde for 15 min, embedded in OCT compound, and snap-frozen in liquid nitrogen. Frozen blocks were cut into serial 5-μm sections and mounted onto Matsunami adhesive silane-coated microscope slides (Matsunami, Osaka, Japan). Sections were incubated with primary antibodies, washed with PBS-0.01% Tween, and then incubated with the appropriate secondary reagents. Nuclei were counterstained with DAPI (Molecular Probes).
RESULTS AND DISCUSSION
To identify autosomal recessive mutants, we have analyzed generation-three offspring mice, in which homozygous deficiency in a particular gene should occur at a 1/16 ratio unless it is embryonic lethal. Among 70 mice in one pedigree, we found 5 mice that had a dramatic reduction in CD4 and CD8 T cells in peripheral blood (Fig. 1A). This phenotype was observed in both sexes and was confirmed to be heritable by the sib-pair mating of mice of the same phenotype. A slight but significant decrease of peripheral T cells was observed in heterozygous mice (Fig. 1B), indicating that this gene gives rise to haploinsufficiency.
Peripheral T cells are reduced in ENU mutant mice. (A) CD4 versus CD8 profiles of peripheral blood mononuclear cells (PBMC) from a representative 12-week-old ENU-induced mutant and wild-type (WT) mouse. (B) Percentages of total CD3+ cells, CD4+ cells, and CD8+ cells in PBMC of wild-type (n = 23), heterozygous (n = 17), and mutant mice (n = 5) are indicated. The data from mice originally were tentatively grouped into normal, heterozygous (hetero), and homozygous groups according to phenotype, and in this figure the data have been regrouped based on themis genotyping. The asterisk indicates a P of <0.001 (t test).
The flow-cytometric analysis of thymocytes from the mutant mice revealed that the number of SP thymocytes, especially the CD4SP cells, was dramatically reduced compared to that of wild-type mice (Fig. 2A and B). TCRβ expression levels on DP and SP thymocytes were indistinguishable between wild-type and mutant mice, but the expression of CD5 was lower at the DP stage in mutant mice and remained low at the SP stage (Fig. 2C). Wild-type SP thymocytes downregulated heat-stable antigen (CD24), reflecting their maturation (2), whereas the heat-stable antigen levels on mutant SP thymocytes remained high. CD69 expression profiles in wild-type SP thymocytes showed two peaks, the brighter peak representing the cells that have just been positively selected (7), whereas in mutant mice only one intermediate-intensity peak was seen. These data collectively indicate that the developmental step from the DP to SP stage that accompanies positive selection (16) is profoundly hampered in the mutant mice. The histological analysis of the thymus revealed that the cortical regions were intact, while the medullary regions were significantly smaller in the mutant mice (Fig. 2D), consistently with the flow cytometry data showing normal numbers of DP thymocytes and reduced numbers of SP thymocytes (19).
Developmental arrest of SP thymocytes in ENU mutant mice. (A and B) Thymocytes from 8-week-old ENU mice and wild-type (WT) C57BL/6 mice were analyzed for the expression of CD4 and CD8. Representative flow-cytometric profile (A) and proportions of CD4/CD8 subsets (means ± standard errors of the means) (B) are shown. (C) The expression of TCRβ, CD5, CD24, and CD69 in double-negative (DN), DP, CD4SP, and CD8SP thymocyte fractions was analyzed for wild-type (blue) and mutant mice (red). (D) Histological analysis of thymi from 8-week-old ENU and B6 mice. Sections were stained with the medullary marker ER-TR5 (red) and DAPI (blue). (E) Representative flow-cytometric profiles of regulatory T-cell and NKT cells in thymi of 8-week-old mice. (F) The development of cells in major lineages other than T lineage in bone marrow was not affected in mutant mice. Bone marrow cells were analyzed with the indicated surface markers. In the bottom panels, profiles of cells gated on the lineage marker-negative (Lin−) fraction of bone marrow cells are shown.
The development of NKT cells in thymus was virtually intact in mutant mice, while that of regulatory T cells looked to be slightly affected (Fig. 2E). Other lineage cells in bone marrow, such as erythroid, myeloid, and B cells, as well as primitive c-kit+ progenitors, were found to be largely normal (Fig. 2F). Based on these phenotypes, we named this mutation SPOTR (for SP thymocyte reduction).
To determine whether the SPOTR phenotype is cell autonomous, we made bone marrow chimeric mice. Wild-type mice transplanted with bone marrow cells of mutant mice after lethal irradiation exhibited a thymic phenotype similar to that of the SPOTR mice, while the reciprocal transplant had no phenotype (Fig. 3A). Thus, the SPOTR phenotype is due to a defect in hematopoietic cells. On the other hand, the existing splenic CD4 and CD8 T cells in the mutant mice responded well to CD3/CD28 stimulation (Fig. 3B), suggesting that the downstream molecules involved in TCR signaling function normally (6). The reduction of SP thymocytes apparently is not due to accelerated apoptosis, because the proportion of annexin V+ propidium iodide− thymocytes was the same in mutant and wild-type mice (data not shown).
Defect in thymocytes but not in thymic stromal cells is responsible for the phenotype of the SPOTR mice. (A) CD4 versus CD8 profiles of thymocytes from bone marrow chimeric mice. Lethally irradiated SPOTR mice (left) or B6Ly5.1 congenic mice (right) were reconstituted with 1 × 106 bone marrow cells from B6Ly5.1 or SPOTR mice, respectively. Thymocytes were harvested 2 months after reconstitution. (B) Proliferative response of mutant and wild-type (WT) T cells. Sorted splenic T cells from mutant and wild-type mice were labeled with CFSE and stimulated with anti-CD3 and anti-CD28 monoclonal antibody. Cells were analyzed by flow cytometry 72 h after stimulation.
To identify the mutation responsible for the SPOTR phenotype, we mated mutant homozygous male mice with C3H/HeJ female mice, and after intercrossing the offspring, the phenotype was screened by the flow-cytometric analysis of peripheral blood T cells. As expected, one-fourth of mice (58 out of 238) exhibited low T-cell numbers (<7.24% CD3-positive cells) (Fig. 4A). Based on genetic linkage analysis using SNP markers (8), the loci responsible for the SPOTR phenotype were predicted to exist in the region around D10SNP2 on chromosome 10, and further genetic mapping indicated that the responsible mutation is localized within 0.78 Mbp of the D10SNP2 marker (Fig. 4A and B and 5A).
Genetic mapping of the SPOTR mutation by SNP analysis. (A) Half of the genetic mapping results, rearranged in accordance with CD3+ cell proportions in peripheral blood, are shown. With SNP analysis between the C57BL/6J and C3H/HeJ strains, a compatible homozygous C57BL/6J SNP pair with a phenotype for low levels of CD3+ cells were found at the D10SNP2 marker on chromosome (Chr.) 10, and the SNP genotype was changed to heterozygote or C3H/HeJ homozygote in accordance with the increase in CD3+ cell numbers. (B) Phenotype and SNP mapping are shown in detail.
Identification of a missense mutation in the SPOTR mouse genome. (A) A total of 254 mice were examined for genetic mapping, and the responsible loci were mapped to the 0.39 centimorgan (cM) region, including D10SNP2 (shown as a red arrow on the magnified chromosome bar). Calculating that 1 cM ≈ 2Mbp, a 0.78-Mbp region was predicted to contain the candidate loci (yellow on chromosome [Chr.] 10). Three genes in this region, ptprk, E430004N04Rik, and 2310057J18Rik, are indicated, with the direction of transcription indicated by the pointed end of the rectangular bars. Pink arrows indicate the approximate analogous region deleted in the LEC rat genome. (B) A missense mutation was found in exon 4 of the E430004N04Rik gene, A1799C, leading to the amino acid substitution T512P. The coding regions of the E430004N04Rik gene are indicated in light green, and the untranslated regions are indicated in black. (C) RT-PCR analysis of ptprκ and themis expression in total RNA prepared from sorted DP thymocytes. The amplification of β-actin cDNA is shown as a control. (D) Western blot analysis of Themis protein in DP thymocytes of wild-type (WT) and mutant mice. Erk1/2 expression is shown as a loading control. (E) Cytochemical analysis for the location of Themis protein. FLAG-tagged themis cDNA was introduced into 293T cells and stained with anti-FLAG monoclonal antibody (red) together with DAPI (blue).
Among candidate genes, we focused on two genes located nearest to the center of the 0.78-Mbp region, ptprκ (encoding a receptor-like protein tyrosine phosphatase type κ; RPTPκ) and E430004N04Rik, since corresponding genes in the rat have been reported to be deleted in the LEC (Long-Evans Cinnamon) rat strain (Fig. 5A). LEC rats are known to have a reduction in CD4 SP T cells in the thymus and periphery, resulting in a defective immune response, a phenotype designated thid (for T helper immunodeficiency), which is inherited in an autosomal recessive manner (14). Two independent studies reported that the large genomic deletion that eliminates ptprκ and RGD1560849 (the rat analog of E430004N04Rik) function causes the thid phenotype. These previous studies proposed that ptprκ, rather than RGD1560849, is responsible for the thid phenotype based on an in vitro functional analysis of these genes (1, 10).
We have sequenced these two genes in SPOTR mice and found a missense mutation in exon 4 of the E430004N04Rik gene, A1779C, resulting in a T512P amino acid substitution (Fig. 5B). This Thr 512 is well conserved among a broad range of Placentalia species (Fig. 6). In contrast, no mutation was detected in the coding region as well as 2 kbp upstream of the first exon of the ptprκ gene (data not shown). These findings suggested that E430004N04Rik is the most likely gene candidate for the SPOTR mutation. In the process of manuscript submission, we noticed that several other groups also have found this gene and named it themis. We decided to use this name.
themis protein orthologs in various Placentalia species. ClustalW protein alignment of themis protein orthologs between mouse, rat (91% identical to mouse), human (80%), chimpanzee (80%), horse (81%), and cow (75%) are shown. Amino acids conserved among all of these proteins are highlighted in dark blue. The location of the point mutation in the SPOTR mouse is indicated by a red rectangle.
While ptprκ is broadly expressed on various tissues, including brain, lung, liver, kidney, and ovary (9), according to the Gene Expression Omnibus profile database, themis is expressed almost exclusively in T cells in thymus and periphery (18). We then examined the expression of ptprκ and themis mRNA in mutant and wild-type littermate thymocytes by quantitative RT-PCR. The level of ptprκ and themis message was comparable between mutant and wild-type mice, indicating that mRNA expression/stability was not affected by the SPOTR mutation (Fig. 5C). However, Themis protein expression was dramatically reduced in SPOTR mouse thymocytes (Fig. 5D).
Themis protein contains 636 amino acids, but we found no homology to known proteins, and there were no recognizable protein domains that might provide clues to Themis function. Histochemical analysis indicated that Themis protein mainly resides in the cytoplasm and not in the nucleus (Fig. 5E). Since T512P amino acid substitution results in decreased Themis protein expression, it is probable that this threonine is a key residue that determines the proper folding and/or stability of the protein.
We next generated themis gene-deficient mice by exchanging the coding region of exon 1 with green fluorescent protein (GFP) (Fig. 7A and B). By analyzing heterozygotes, we found the prominent upregulation of this gene at the transition from the double-negative to DP stage (Fig. 7C). The themis expression level looked to be maintained in CD4SP and CD8SP cells in thymus but reduced in peripheral T cells (Fig. 7C). The homozygous themisgfp/gfp mice had a phenotype very similar to that of the SPOTR mice (Fig. 7D). Confirming the effectiveness of the themis knockout, Themis protein was completely lost in themisgfp/gfp thymocytes, while in the themis+/gfp thymocytes the protein level was reduced by half (Fig. 7E). Since ptprk and themis genes are closely linked, it is possible that the deletion of exon 1 of the themis gene exerts some effects on ptprk gene expression. However, in combination with our finding that the SPOTR mice bearing a single point mutation in the themis gene exhibit a quite similar phenotype, it is almost certain that themis itself is essential for SP cell development.
themis-deficient mice showed a phenotype similar to that of SPOTR mice. (A) themis gene targeting strategy. The coding region of first exon was replaced by GFP and a neomycin (neo) cassette. The open box of exon 1 indicates the 5′ untranslated region. (B) Deletion was screened by PCR using primers shown in panel A. (C) The expression of themis was analyzed by using a GFP reporter mouse. GFP expression profiles in cells of CD4/CD8 thymic fractions from a 6-week old heterozygous mouse are shown. GFP expression profiles in CD3+ cells and B220+ cells from spleen also are shown. (D) CD4 versus CD8 profiles of thymocytes and peripheral blood mononuclear cells (PBMC) from a 4-week-old themis-deficient mouse (themisgfp/gfp), a littermate (themis+/gfp), and a wild-type mouse of the same age (themis+/+) are shown. (E) Western blot analysis of Themis protein in thymocytes from wild-type, heterozygous, and themis-deficient mice.
A decrease in CD4 lineage T cells is a common phenotype seen in LEC rats (14) and themis-deficient mice, but a decrease in CD8 T cells is seen only in the themis-deficient mice. However, since the reduction of CD8 T cells in these mice is not as prominent as the CD4 T-cell defect, it is probable that this phenotype is masked in LEC rats through unknown compensatory mechanisms. We thus presume that the phenotype of themis-deficient mice as well as that of SPOTR mice is identical to the thid phenotype in LEC rats. Although ptprκ has been proposed as the gene responsible for the thid phenotype in LEC rats (1, 10), our data indicate that the deletion of the themis gene alone can bring about the thid phenotype. The possibility is not excluded that each of these genes is responsible of CD4 T-cell development; however, whether ptprκ really is playing a role in T-cell development remains to be proven by making knockout mice.
In summary, through a forward genetics approach, we have identified a novel gene, themis, that is essential for the development of SP thymocytes. Since themis is a totally unknown gene whose functional domains are unknown so far, further studies of the SPOTR mutation will provide insight into the role of this gene in T-cell development.
ACKNOWLEDGMENTS
We thank I. Taniuchi for advice on making the targeting vector, S. Hori for technical assistance in the staining of regulatory T cells, O. Ohara for valuable advice, and Y. Katsura and P. Burrows for the critical reading of the manuscript.
We declare that we have no competing financial interests.
FOOTNOTES
- Received 18 June 2009.
- Returned for modification 3 July 2009.
- Accepted 8 July 2009.
↵▿ Published ahead of print on 20 July 2009.
- American Society for Microbiology
