EM, ET, LG, CS, and OL conceived and designed the experiments. EM, ET, LD, LG, HL, CT, VP, AD, ICM, FT, SC, GV, and CS performed the experiments. EM, ET, LD, LG, HL, CT, ICM, CS, and OL analyzed the data. SL and OL contributed reagents/materials/analysis tools. CS and OL wrote the paper.
OL laboratory has received funding from Innate-Pharma, which co-owns the filled patent for the 3C10 antibody. EM has received salary through this funding.
Mucosal-associated invariant T (MAIT) cells display two evolutionarily conserved features: an invariant T cell receptor (TCR)α (iTCRα) chain and restriction by the nonpolymorphic class Ib major histocompatibility complex (MHC) molecule, MHC-related molecule 1 (MR1). MR1 expression on thymus epithelial cells is not necessary for MAIT cell development but their accumulation in the gut requires MR1 expressing B cells and commensal flora. MAIT cell development is poorly known, as these cells have not been found in the thymus so far. Herein, complementary human and mouse experiments using an anti-humanVα7.2 antibody and MAIT cell-specific iTCRα and TCRβ transgenic mice in different genetic backgrounds show that MAIT cell development is a stepwise process, with an intra-thymic selection followed by peripheral expansion. Mouse MAIT cells are selected in an MR1-dependent manner both in fetal thymic organ culture and in double iTCRα and TCRβ transgenic RAG knockout mice. In the latter mice, MAIT cells do not expand in the periphery unless B cells are added back by adoptive transfer, showing that B cells are not required for the initial thymic selection step but for the peripheral accumulation. In humans, contrary to natural killer T (NKT) cells, MAIT cells display a naïve phenotype in the thymus as well as in cord blood where they are in low numbers. After birth, MAIT cells acquire a memory phenotype and expand dramatically, up to 1%–4% of blood T cells. Finally, in contrast with NKT cells, human MAIT cell development is independent of the molecular adaptor SAP. Interestingly, mouse MAIT cells display a naïve phenotype and do not express the ZBTB16 transcription factor, which, in contrast, is expressed by NKT cells and the memory human MAIT cells found in the periphery after birth. In conclusion, MAIT cells are selected by MR1 in the thymus on a non-B non-T hematopoietic cell, and acquire a memory phenotype and expand in the periphery in a process dependent both upon B cells and the bacterial flora. Thus, their development follows a unique pattern at the crossroad of NKT and γδ T cells.
White blood cells, or lymphocytes, play an important role in defending the body from infection and disease. T lymphocytes come in many varieties with diverse functions. Mucosal-associated invariant T (MAIT) cells constitute a subset of unconventional T lymphocytes, characterized by their invariant T cell receptor (TCR)α chain and their requirement for the nonpolymorphic class Ib (MHC) molecule, MR1. MAIT cells are extremely abundant in human blood and mucosae. Contrary to mainstream T cells, their development requires B cells and commensal microbial flora. To shed light on the little-understood MAIT cells, we used new tools, including an antibody that we recently developed to detect human MAIT cells, and we were able to show that MAIT cell development is a stepwise process, with an intra-thymic selection followed by peripheral expansion. We show that thymic selection is MR1 dependent but requires neither B cells nor the commensal flora, which are both necessary for the expansion in the periphery. In contrast with the other evolutionarily conserved invariant subset, the natural killer T (NKT) cells, we found that MAIT cells exit the thymus as “naïve” cells before becoming antigen-experienced memory cells and expanding in number to represent a significant 1%–4% of peripheral T cells in human blood. In mice, we found that MAIT cells remain naïve and do not expand substantially. We conclude that MAIT cell development follows a unique scheme, where, unlike NKT cells, MAIT cell selection and expansion are uncoupled events that are mediated by distinct cell types in different compartments.
Mucosal-associated invariant T cells, the most abundant invariant T cell subset in humans, arise via a distinct developmental pathway that represents a hybrid of that seen for NKT and γδ T cells, two other unconventional T cell subsets.
Unconventional T cells include several T cell receptor (TCR)αβ+ major histocompatibility complex (MHC) class Ib restricted, as well as TCRγδ+ subsets in both mice and humans [
Among unconventional T cells, only two subsets display both a TCR and selecting MHC class Ib molecules highly conserved between species, the NKT cells and the mucosal-associated invariant T (MAIT) cells (reviewed in [
For NKT cells, the molecular basis of their specific development begins to be unraveled: contrary to mainstream T cells and γδ T cells, NKT cell development requires the integrity of the SLAM/SAP/fyn pathway with the involvement of homotypic interactions of SLAM receptor family members and CD1d expression on CD4+/CD8+ (DP) thymocytes [
Little is currently known about MAIT cell development. They are absent from
Herein, we investigated MAIT cell development in both humans and mice. In humans, we used an anti-Vα7.2 monoclonal antibody (mAb) generated in our laboratory to track normal unmanipulated MAIT cells. MAIT cells are rare and display a naïve phenotype both in human thymus and cord blood, while they are abundant and show a homogeneous memory/effector-like phenotype in the blood of all healthy subjects tested. In mice, the over-expression of one of the TCR chains was sufficient to increase the frequency of MAIT cells in central and peripheral lymphoid organs in an MR1-dependent manner. Fetal thymic organ culture (FTOC) and phenotypic studies of the thymus in these mice demonstrated that MAIT cells are selected in the thymus in the absence of B cells and bacterial products. However, most mouse MAIT cells remained naïve in the periphery. The complementary data obtained from humans and mice indicate that MAIT cells are exported from the thymus as naïve cells and subsequently become memory and expand, following interactions with B cells and the commensal flora, in an SAP-independent manner. Thus, MAIT cells use a developmental pathway distinct from both known unconventional T cell subsets and mainstream T cells, showing that they represent a unique subset with its own features.
As MAIT cells are defined by the use of an invariant iVα7.2-Jα33 TCRα chain, we developed a monoclonal antibody (3C10) recognizing the TCRα Vα7.2 segment by immunizing Balb/c mice with a recombinant Vα7.2-Jα33/Vβ13 recombinant protein (see
(A) Representative 3C10 and CD161 staining gated on αβTCR T cells in DN (+CD8αα), CD8β, and CD4 populations. Percentages of 3C10+CD161+ and 3C10+CD161− are boxed.
(B) Percentage of 3C10+CD161+ cells in indicated DN, CD8β, and CD4 populations (left panel); percentage of DN, CD8β, and CD4 3C10+CD161+ cells present in the αβTCR T cells (right panel) both estimated by FACS analysis on 104 healthy blood samples. Each dot represents a donor.
(C) Representative FACS staining of 3C10+CD161+ and 3C10+CD161− cells in DN, CD8β, and CD4 populations (left, center, and right panels, respectively): expression of CD45RO gated on αβTCR T cells (upper panels) and expression of CD27 versus CD45RA gated on MAIT cells (middle panels) and mainstream T cells (lower panels).
The high numbers of DN+CD8αα and CD8βMAIT cells were similar in 104 blood donors, whereas the CD4 subset contained only one-tenth as many such cells (
We used the aforementioned staining strategy to assess the presence of MAIT cells in the human thymus, and observed some 3C10+ DN or single-positive CD8β and CD4 T cells (
(A) Representative 3C10 and CD161 staining gated on mature CD3hi T cells in DN, CD8β, and CD4 populations in the thymus of a 5-y-old patient, upper panel. The corresponding blood analysis is shown in the lower panel. Percentages are indicated in the quadrants.
(B) Quantitative PCR analysis of iVα7.2-Jα33 segment expression normalized to Cα gene expression, on sorted 3C10+CD161+ DN and 3C10+CD161− DN, CD8β, or CD4 populations in the thymus of three different patients (aged 8 months, 1 y, and 8 y). 3C10+CD161+ DN and 3C10+CD161− CD4 fractions from PBMC were used as positive and negative controls respectively (left panel). The 100% value was defined as that obtained with the 3C10+CD161+ DN sample from the PBMC, which contains 100 % MAIT cells, whereas the background was defined as the value for 3C10+CD161− CD4 cells, which have no canonical sequence. The values for the different samples were normalized with respect to these controls.
(C) Representative staining of CD27 versus CD45RA gated on DN MAIT cells in thymus and PBMC (upper and lower left panels, respectively). CD45RO expression by MAIT cells (histogram) in thymus and PBMCs. Representative of three independent experiments.
MAIT Cells Ontogeny in Mice Transgenic for the iVα19 TCRα, MAIT Cell Vβ6 TCRβ Chains or Both
Mice have very few MAIT cells and the lack of an antibody specific for the invariant mVα19 TCRα chain precludes the direct analysis of MAIT cells in vivo or ex vivo. We therefore generated mice expressing an invariant mVα19-Jα33 TCRα chain transgene (Tg). A detailed analysis of these mice and their phenotype are provided as supporting information (see
The forced transgenic expression of a TCRα chain greatly modifies T cell ontogeny [
To further increase the proportion of MAIT cells and to decrease the unwanted TCR specificities due to the endogenous TCRβ chains, we crossed the TCRα Tg and TCRβ Tg mice together in the presence or absence of MR1. The thymuses of the resulting mice were a little smaller than those of wt, in both backgrounds (53 mean [M] ± 26, ± standard deviation [SD],
(A) Representative HSA and TCRβ staining (upper panel) and CD8 and CD4 staining gated on mature thymocytes (lower panel) in iVα19-Jα33/Vβ6 transgenic mice in a MR1+/+ Cα−/− or MR1−/− Cα−/− background (left and right panel, respectively). Percentages of mature T cells (TCRβhiHSAlo) and intermediate T cells (TCRβhiHSAhi) are boxed, (upper panel). Representative of six mice in each group.
(B) Representative CD8 and CD4 staining gated on mature T cells in C57Bl/6 and iVα19-Jα33 transgenic mice in a MR1+/+ or MR1−/− Cα−/− background in FTOC.
(C) Frequency of Vβ6 and Vβ8 expression in mature T cells (TCRβhi HSAlo) in DN, CD8, and CD4 subsets from FTOC of iVα19-Jα33 transgenic Cα−/− mice in the presence (gray circle) or not of (white circle) MR1background. Each dot represents an individual fetal thymus. Representative of two independent experiments. *,
In all the mice studied above, the mature thymocytes were in very small numbers and could be recirculating T cells generated elsewhere. We therefore studied FTOC from iVα19 Tg Cα−/− mice in the presence or not of MR1 to demonstrate formally the intra-thymic development of MAIT cells. Negligible numbers of CD4 T cells were generated whereas the proportion of CD8 and DN T cells was very high among the mature (TCRβhi/HSAlo) thymocytes (
We have previously shown that accumulation of MAIT cells in the periphery depends on the presence of B cells [
(A) Representative HSA and TCRβ staining (upper panel) and CD8 and CD4 staining gated on mature thymocytes (lower panel) in iVα19-Jα33/Vβ6 transgenic mice in a MR1+/+ or MR1−/− RAG−/− background (left and right panel, respectively). Percentages of mature and intermediate T cells are boxed (upper panel). Representative of four mice in each group.
(B) Representative Vβ6 and CD19 staining gated on MLN lymphocytes from iVα19/Vβ6 double transgenic RAG−/− mice on a MR1+/+ (upper panels) or MR1−/− (lower panels) background without (left panels) or 14 d after MR1− or MR1+ CD3ε−/− splenocyte transfer. Representative of three to ten mice per group.
(C) Absolute numbers of T lymphocytes in the MLN of the different groups analyzed: MR1+/+ (filled circles) and MR1−/− (open circles) iVα19/Vβ6 double Tg RAG−/− mice, without or after MR1− or MR1+ CD3ε−/− splenocyte transfer. *,
The periphery of MR1−/− RAG−/− TCRαβ double Tg mice was almost completely devoid of mature T cells (1.5 ± 0.5 × 103, m ± standard error [SE],
Therefore, in mice where MAIT cells are selected in the thymus but do not expand in the periphery, the adoptive transfer of MR1+ B cells is sufficient to promote their accumulation. Interestingly, in all cases MAIT cells acquired a memory (CD44hi) phenotype after transfer of B cells (unpublished data). The acquisition of a memory phenotype might be related to a “homeostatic” expansion due to the sudden provision of selecting niches. In any cases, although it could be argued that the splenocyte mixture we injected contains both B cells and other cell types, such as dendritic cells or macrophages, the latter are already present in the host mice before transfer and are obviously not able to induce the peripheral accumulation of MAIT cells. We can therefore formally conclude that B cells are necessary to allow MAIT cells accumulation in the MLN of these monoclonal Tg mice, either through survival, expansion, and/or addressing of MAIT cells to the intestinal territory.
The accumulation of murine MAIT cells in the intestinal compartment also requires the presence of the commensal flora [
(A) Representative 3C10 and CD161 staining gated on αβTCR T cells in DN (+CD8αα) and CD8β populations in human cord blood (left and right panel, respectively). The percentages of MAIT cells (3C10+CD161+) and mainstream T cells (3C10−CD161−) are boxed.
(B) Representative CD45RO expression on DN and CD8β MAIT cells (left and right panels, respectively) in human cord blood and adult blood (upper panel). Expression of CD27 versus CD45RA is gated on MAIT cells (middle panel) or on mainstream T cells (lower panel).
(C) Representative CD8 and CD4 staining of MLN from iVα19/Vβ6 double transgenic Cα−/−/TAP−/−/Ii−/− either on MR1+/+ (left panels) or MR1−/− (right panels) in upper panel; lower panel expression of CD44 gated on DN, CD8, and CD4 αβTCR T.
(D) Quantitative PCR analysis of iVα7.2-Jα33 segment expression normalized to Cα gene expression on sorted CD44int and CD44hi DN cells from the MLN of Vβ6 transgenic mice (left panel) and of TAP−/−/Ii−/− mice (right panel) on MR1+/+ or MR1−/− background as noted.
We next addressed the question of the naïve/memory status of mouse MAIT cells, which are in much lower number than human MAIT cells. Indeed, in the different Tg mice studied above, the DN and CD8 mature thymocytes are mostly naïve (CD44lo/CD122lo), while the low numbers of CD4 T cells are constantly CD44hi (
To further address the question of the naïve/memory phenotype of MAIT cells, we sorted CD44hi and CD44lo/int T cell subsets from the MLN of TAP−/−/Ii−/− (to enrich in MAIT cells) and Vβ6 Tg mice on a MR1+ or MR1−/− background and quantified the amount of iVα19 transcripts. In both strain of mice, we found that the iVα19 transcripts largely segregated in the CD44lo/int fraction, confirming that wt polyclonal MAIT cells indeed display a naïve phenotype in mice (
Altogether, these results indicate that murine MAIT cells, like their human counterparts, are selected by MR1 in the thymus without acquiring a memory phenotype. However, they remain naïve in the periphery, even in the presence of B cells and the commensal flora. The difference in number and phenotype of MAIT cells between humans and mice might then be related to the absence of peripheral expansion in the latter.
Contrary to mainstream T cells, NKT cells express the ZBTB16 transcription factor from the first stage of their differentiation in the thymus. In mice deficient for this transcription factor, NKT cells remain naïve, do not expand, and do not colonize the effector organs such as the liver but are found in small numbers in the LN [
(A) Quantitative PCR analysis of zbtb16 expression normalized to GAPDH gene expression on sorted CD8 and NKT cells from C57BL/6 mice (left and grey) and from sorted CD8 and DN cells from iVα19-Jα33/Vβ6 transgenic mice in a MR1+/+Cα−/− or MR1−/−Cα−/− background (right and black and white).
(B) Representative CD8β and CD4 staining on gated αβTCR T cells (left panels); 3C10 and CD161 staining gated on DN+CD8αα αβTCR T cells (right panels) from a SAP deficient patient (lower panels) and a control subject (upper panels).
(C) Percentage of specific NKT cells (Vα24) and MAIT (3C10) cells in control subjects and SAP deficient patients gated on DNCD161+CD3+TCRγδ− T cells; each dot represents an individual.
(D) Quantitative PCR analysis of zbtb16 expression normalized to GAPDH gene expression on sorted mainstream CD4 and CD8 T cells and DN and CD8 MAIT specific cells from control subject and from sorted DN and CD8 MAIT specific cells from SAP deficient patient.
Although both NKT cells and MAIT cells are selected on hematopoietic cells, the former subset expands and acquires a memory phenotype within the thymus, whereas MAIT cells seemingly show a more conventional selection process. We sought to investigate the molecular basis for this divergence by studying the SAP-dependency of MAIT cells development. Indeed, CD1d-restricted NKT cells ontogeny involves signaling through the SAP/Fyn/NF-κB signaling pathway, which is triggered by homotypic interactions between SLAM family molecules expressed both on developing NKT cells and selecting cortical thymocytes [
MAIT cells constitute a new subset of MHC class Ib-restricted T lymphocytes conserved among mammals. In this study, we provide for the first time, to our knowledge, phenotypic data on human MAIT cells. We show that they can be tracked by costaining with an anti-Vα7 antibody (3C10) and CD161, allowing us to assess their frequency in the peripheral blood of healthy subjects. Blood MAIT cells are numerous (at least one order of magnitude higher than NKT cells), in accordance with our previous estimates [
Two other groups have generated iVα19 TCRα Tg mice [
We have previously shown that MAIT cells accumulate in the MLN and the gut LP. However, MAIT do not accumulate in sufficient numbers in the thymus of wt mice to be detectable by sensitive reverse transcription (RT)-PCR methods, precluding the possibility to discriminate between intra-thymic selection per se and peripheral expansion. Moreover, it has been suggested recently that the development of some unconventional intra-epithelial lymphocyte (IEL) T cell populations requires a functional thymus but actually takes place in the gut mucosa [
Our data clearly demonstrate that B cells are not the selecting cells in the thymus, but are necessary for the accumulation of MAIT cells in the MLN and LP, by promoting either expansion, survival, and/or addressing to the gut compartment. We have previously shown that MAIT cells are selected on non-T hematopoietic cells [
Recent data from Bendelac and coworkers and from our own lab show that MAIT cells and NKT cells exclusively share the expression of the transcription factor PZLF (ZBTB16) [
The absence of ZBTB16 expression by mouse MAIT cells could be related to the cleanness of the animal facilities, which would not provide the ligand or inflammatory context necessary for MAIT cell expansion. Alternatively, one key genetic component may be missing in mice because of the genetic bottleneck that laboratory mouse strains have been through without selection pressure by the putative function mediated by the MAIT cells. Finally, how ZBTB16 expression is acquired in the thymus by NKT cells and in the periphery by human MAIT cells is an open question. It is probably related to the context of the interactions between the iTCR and the selecting element.
The acquisition of a memory phenotype by human MAIT cells after birth is most likely linked to the colonization of the gut (and other mucosae) with the commensal flora. We speculate that B cells provide the link between the gut bacterial flora and MAIT cells. Direct or indirect (via epithelial cells or dendritic cells [
In conclusion, we show here that besides their similarities, MAIT cell ontogeny is clearly different from NKT cells. Postselection NKT thymocytes already display innate-like properties, with a high frequency and a memory phenotype, whereas postselection MAIT cells are still naïve and need contacts with both B cells and bacteria to expand, acquire a memory phenotype in human, and accumulate in the gut. In this respect, they may be compared to Vδ2+ T cells, which appear in blood as naïve but become memory soon after birth [
CD3ε-deficient B6 mice, Cα deficient mice (N9 to C57Bl/6 [B6]) and Ii−/− (B6/129) mice were obtained from the CNRS CDTA central animal facility (Orléans, France). TAP−/− mice in a B6/129 background were obtained from the Jackson laboratory. Cα−/−, Ii−/−, and TAP−/− mice were intercrossed to obtain double and triple deficient mice. MR1−/− mice have been described elsewhere [
The canonical TCRα chain and the associated Vβ13 TCRβ chains were cloned (ET and OL), from an iVα7.2-Jα33 T cell clone (J.F. Davodeau, ET, M. Bonneville, and OL, unpublished data). Using this template, L. Teyton (SCRI) generated a recombinant heterodimeric protein, which was used to immunize Balb/c mice. After fusion with SP2/0, hybridoma specifically staining a high proportion of CD3+ DN and a proportion of CD8α but few CD4 T cells were selected and cloned. The 3C10 hybridoma was chosen for further characterization because, according to quantitative PCR, the Vα7.2-Jα33 TCRα chain segregated with the 3C10 positive fraction in DN T cells. When transfectants expressing different mouse TCRα and TCRβ chains and a chimeric human iVα7.2-mouse Cα were stained with the 3C10 supernatant, only transfectants harboring the Vα7.2 segment displayed positive staining demonstrating that 3C10 was not an anti-TCRβ chain antibody (unpublished data). Quantitative PCR analysis of the 3C10+ and 3C10− fractions using primers for all Vα segments, demonstrated the absence of significant cross-reactivity between the 3C10 mAb and other Vα segments (unpublished data). The 3C10 antibody was biotinylated and detected with streptavidin PE-Cy7 (BD Pharmingen).
Cell suspensions were prepared from thymus, spleen, peripheral or mesenteric lymph nodes, and gut LPL as previously described [
Flow cytometry was performed with directly conjugated antibodies (BD Pharmingen) according to standard techniques with analysis on FACS Aria and LSRII flow cytometers (Becton Dickinson). DAPI and a 405-nm excitation were used to exclude dead cells. The following antibodies, mostly from BD Pharmingen or eBiosciences, were used in mice: anti-CD45.2-Fitc (104), anti-CD3ε-PC7 (145–2C11), anti-CD5-APC (Ly-1), anti-Vβ6-PE (RR4–7), anti-CD44-PE or APC (IM-7), anti-CD45.1-PE (A20), anti-βTCR-PC5 (H57–597), anti-CD19-PE, Fitc or APC-Cy7 (1D3), anti-CD8α-APC-Cy7 (Ly-2), and anti-CD4-PE-Texas Red (L3T4).
For human stainings, the following were used: anti-CD4-APC-Cy7 (RP4-T4), anti-CD3ε-Alexa 700 (UCHT1), anti-TCRγδ-PC5 (IMMU510), anti-CD8β-PE Texas Red (2ST8.5H7), anti-CD45RO-Fitc (UCHL1), anti-CD45RA-PE (HI100), anti-CD27-Fitc (M-T271), anti-CCR7-PE (150503), anti-CD62L-PE (Dreg 56), anti-CXCR6-PE (56811), anti-CD161-APC (DX12), and anti-CD8α-PE Alexa700 (RPA-T8).
Fragments of human thymuses were operating residues from children undergoing cardiac surgery. Thymuses were cut into small pieces in cold 0.5% BSA PBS. The resulting cell suspension was centrifuged to exclude aggregates. The supernatant was recovered and the cells were washed again with 0.5% BSA PBS. Thymocytes were then counted and labeled.
Blood samples were obtained from healthy donors from the blood bank in accordance with institutional regulations.
Spleens from CD3ε−/− MR1+ mice were harvested and 107 splenocytes were IV injected into MR1+ or MR1− TCRαβ double Tg Rag−/− mice. 2 wk later, the blood, spleen, and MLN were harvested and analyzed by FACS.
iVα19 Tg Cα−/− MR1−/− or MR1+/+ female mice were caged one night with the respective genotype male mice. Pregnant female mice were sacrificed at day 14. Uterine corns were harvested in CO2-independent medium supplemented with 5% FCS and penicillin/streptomycin (Invitrogen). Fetal thymuses were harvested and cultured in 300 μl of IMDM medium supplemented with 10% FCS, 50 μM beta-mercaptoethanol, 10 mM Hepes, 1 mM Sodium pyruvate, 2 mM L-glutamine, and penicillin/streptomycin in transwell plates from Costar (0.4 μm, 12 wells). The medium was changed every 3 d over a period of 6–8 d.
Molecular biology methods: RNA extraction, reverse transcription, TCR primers, PCR methods, and quantitative polyclonal PCR and polyclonal sequencing were carried out as previously described [
Sequence of Vα7.2-Cα amplicons obtained after polyclonal sequencing, with a specific Vα7.2 primer, of αβTCR T cells sorted from either 3C10+CD161high or 3C10+CD161lo cells in DN, CD8β, and CD4 subsets. Representative of four independent experiments.
(866 KB PDF)
Representative staining of MAIT cells (3C10+CD161high) and mainstream T cells (3C10−CD161lo) gated on CD4− αβTCR T cells in PBMCs (upper panel). Representative CD8α/CD8β staining gated on mainstream CD4− T cell and CD4− MAIT cells (left and right middle panels, respectively). CD8α and CD8β expression by CD4− mainstream T cells or CD4− MAIT cells (left and right lower panels, respectively). Percentages of mainstream T and MAIT cells are boxed, upper panel. Representative of three independent experiments.
(98 KB PDF)
(A) Absolute number of mature T cells in lymphoid organs of C57Bl/6 (black square) and iVα19-Jα33 transgenic mice in a Cα−/− (gray circle) or Cα−/−TAP−/−Ii−/− (black circle) background. Mononuclear cells from thymus (a), spleen (b), peripheral lymph nodes (PLN) (c), MLN (d), and gut LP (e) were harvested as described, counted, and the percentage of TCRβ+ cells was determined by FACS analysis. Each dot represents an individual mouse.
(B) Frequency of Vβ6 (left panel) and Vβ8 (right panel) in DN, CD8, and CD4 populations, analyzed by FACS analysis, in αβTCR T cells in the MLN (upper panel) and LP (lower panel) of C57Bl/6 and iVα19-Jα33 transgenic mice in a Cα−/− or Cα−/−TAP−/−Ii−/− background.
(C) Vβ6 (left panel) and Vβ8 (right panel) bias in transgenic mice in DN, CD8, and CD4 populations on different backgrounds, analyzed by FACS, in mature thymocytes of C57Bl/6 and iVα19-Jα33 transgenic mice on a Cα−/− or Cα−/−MR1−/− (white circle) background. Each dot represents an individual mouse.
(D) Quantitative PCR analysis of iVα19-Jα33 segment expression normalized with respect to Cα gene expression on sorted DN, CD8, or CD4 T cells from thymocytes or MLN lymphocytes of MR1+ or MR1−/− C57Bl/6 mice. Each dot represents an individual mouse.
(149 KB PDF)
Representative CD8/CD4 staining gated on total thymus (upper panels) and on mature thymocytes (TCRβhigh HSAlow) (lower panels) of C57Bl/6 and iVα19-Jα33 transgenic mice on a MR1+/+ or MR1−/− Cα−/− background either (left, middle, and right panels, respectively). percentages of CD4, CD8, and DN T cells populations are boxed. Representative of ten independent experiments.
(71 KB PDF)
Frequency of Vβ6 (left panel) and Vβ8 (right panel) in DN, CD8, and CD4 populations, analyzed by FACS, in αβTCR T cells from the MLN of C57Bl/6 (black square) and iVα19-Jα33 transgenic mice on a Cα−/− (gray circle) or Cα−/−MR1−/− (white circle) background; upper panels. Frequency of Vβ6 (left panel) and Vβ8 (right panel) in DN, CD8, and CD4 populations, analyzed by FACS, in αβTCR T cells from the MLN of C57Bl/6 (black square) and Cα−/−TAP−/−Ii−/− (black circle) or Cα−/−TAP−/−Ii−/−MR1−/− (white circle) background; lower panels. Each dot represents an individual mouse.
(76 KB PDF)
Sequence of Vα7.2-Jα33 amplicons obtained after polyclonal sequencing, with a specific Vα7.2 primer, of 3C10+CD161hi DN and 3C10+CD161lo CD8 αβTCR T cells sorted from cord blood. Representative of two separate experiments.
(229 KB PDF)
Representative staining of CD44 expression on DN, CD8, and CD4 gated on mature T cells (TCRhiHSAlo) from the thymus left part; and from TCRhi T cells from MLN right part, in C57BL/6 and Vβ6, iVα7.2-Jα33/ Cα−/− and in iVα19-Jα33/Vβ6/ Cα−/− transgenic mice. Shaded grey histogram is on an MR1+/+ background while black line is on an MR1−/− background.
(323 KB PDF)
(A) Absolute numbers of mature T cells from the thymus and the MLN of iVα19-Jα33/Vβ6/Cα−/−/TAP−/−/Ii−/− transgenic mice on a MR1+/+ (black diamond) or MR1−/− (white diamond) background.
(B) Representative CD44 and a pool of endogenVβ staining of DN, CD8, and CD4 gated on TCRβhi T cells from MLN of C57BL/6 (upper panel) and iVα19-Jα33/Vβ6/ Cα−/−/TAP−/− /Ii−/− transgenic mice on a MR1+/+ (middle panel) or MR1−/− (lower panel) background.
(188 KB PDF)
(68 KB DOC)
We wish to thank M. Garcia, C. Billerit, S. Laigneau, and I. Grandjean for managing the mouse colonies in Paris; Z. Maciorowski and C. Guerin for cell sorting; D. Mendes Da Cruz and M. Dussiot for thymus samples; C. Picard for immunodeficient patient samples; L. Teyton for the recombinant iVα7.2/Vβ13 protein; and J.F. Davodeau and M. Bonneville for growing the iVα7.2 clone. We thank M. Bonneville and K. Benlagha for discussions and reviewing the manuscript.
fetal thymic organ culture
lamina propria
mucosal-associated invariant T
major histocompatibility complex
mesenteric lymph node
MHC-related molecule 1
natural killer T
T cell receptor
wild type