A new mouse strain for the analysis of invariant NKT cell function

To the editor

Invariant natural killer T cells (iNKT cells) are a T lymphocyte population with restricted T cell antigen receptor (TCR) diversity. They make up a distinct subset of T lymphocytes that is activated by glycolipid antigens presented by CD1d, a non-classical major histocompatibility complex class I–like molecule. The invariant TCR of iNKT cells consists of a specific α-chain rearrangement, of α-chain variable region 14 and α-chain joining region 18 (V_α14-J_α18), encoded by Tcra gene segments Trav11-Traj18 in mice, and V_α24-J_α18 (TRAV10-TRAJ18) in humans, and a limited β-chain repertoire¹.

iNKT cells are associated with early immune responses and have been reported to exert a crucial influence on a wide variety of immune and inflammatory responses². Mice deficient in iNKT cells provide a key tool for analysis of the role of these cells in various contexts. There are two widely used mouse models of iNKT cell deficiency, but neither provides a completely specific defect. Mice lacking the Cd1d1 and Cd1d2 loci cannot positively select iNKT cells, but they also do not have type II NKT cells³, which are CD1d-reactive T lymphocytes with a more diverse α-chain repertoire⁴. Mice lacking the Traj18 gene segment cannot form the TCRα chain that is essential for iNKT cell development⁵. However, a published study has reported that rearrangements of all the J_α regions upstream of Traj18 were suppressed in the available strain of mice with deletion of the Traj18 gene segment⁶. In this strain, the gene encoding the neomycin-resistance selection marker, driven by the promoter of the phosphoglycerate kinase gene (Pgk-neo^r), is not removed. Therefore, the authors proposed that the transcription of Pgk-neo^r, which is in an orientation opposite to that of the J_α regions, led to the suppression of rearrangement of the majority of J_α regions and thereby caused substantial distortion of the TCRα repertoire.

Here we describe a new mouse strain with deletion of Traj18 created on the C57BL/6 background in which neo^r was deleted along with the Traj18 gene segment. These mice lacked iNKT cells, defined as TCRβ⁺ lymphocytes in spleen and other tissues that reacted with CD1d tetramers loaded with α-galactosyl ceramide (Fig. 1a and Supplementary Fig. 1), and they did not produce interferon-γ (IFN-γ) or interleukin 4 (IL-4) at 2 h or 24 h after injection of α-galactosyl ceramide (Supplementary Fig. 2). Although the minor population of tetramer-positive cells with a V_α10-J_α50 rearrangement would not be affected by deletion of Traj18, we did not find definitive evidence of these cells, which are rare in C57BL/6 mice, by tetramer staining or by PCR (data not shown). To analyze the TCRα repertoire, we isolated CD4⁺CD8⁺ (double-positive) CD69⁻ thymocytes either from C57BL/6 wild-type mice or from Traj18-deficient mice and performed PCR amplification of genes encoding three TCRα chains, V_α14 (Trav11), V_α3 (Trav9) and V_α8 (Trav12), by using a specific forward primer for each V_α-encoding sequence and a reverse primer for sequence encoding the TCRα constant region. We purified and sequenced PCR products. Our results indicated that except for Traj18, the frequency of J_α use for each of these rearranged V_α regions was similar for wild-type mice and Traj18-deficient mice (Fig. 1b and Supplementary Fig. 3a,b). Therefore, the data indicated that the overall TCR repertoire was not perturbed in these mice.

Figure 1.

Traj18-deficient mice lack iNKT cells but have a normal J_α-region repertoire and show significantly fewer eosinophils in an airway hyper-reactivity model. (a) Flow cytometry of total splenocytes isolated from a wild-type mouse (WT) and a Traj18-deficient mouse (Traj18^−/−). Numbers adjacent to outlined areas indicate percent live CD19⁻CD8α⁻ TCRβ⁺ lymphocytes reactive with the CD1d–α-galactosyl ceramide (α-GalCer) tetramer. (b) PCR analysis assessing the frequency of J_α use for Trav9 in CD4⁺CD8⁺ (double-positive) CD69⁻ thymocytes from wild-type mice (458,170 sequences) and Traj18-deficient mice (176,431 sequences). (c) Airway hyper-reactivity in mice sensitized with OVA for 2 weeks and challenged with OVA for 3 consecutive days, assessed by quantification of eosinophils in bronchoalveolar lavage fluid (BALF). Each symbol represents an individual mouse (n = 10 per genotype); small horizontal lines indicate the mean. *P < 0.0001 (Mann-Whitney test). Data are from one experiment (a,c) or are representative of one experiment with one of two mice per genotype analyzed (b).

This new mouse strain should allow investigators to assess the function of iNKT cells specifically, without the complicating factor of the reduction in the diversity of TCR J_α regions. Furthermore, it is unlikely that the type II NKT cells with more diverse TCRs were affected, although we did not analyze this. As a proof of concept, we analyzed airway inflammation and airway hyper-reactivity, which have been shown in mouse models of asthma to require the presence of iNKT cells⁷. We used ovalbumin (OVA)-induced and cockroach antigen–induced models of pulmonary inflammation. For the OVA model, we sensitized mice by intraperitoneal administration of OVA weekly for 2 weeks. At 2 weeks after the second OVA sensitization, we challenged mice intranasally on 3 consecutive days with OVA. We harvested bronchoalveolar lavage fluid 72 h after the final OVA challenge. Our results showed that no airway eosinophilia was induced in Traj18-deficient mice (Fig. 1c), and they also had considerably fewer neutrophils and total leukocytes than their wild-type counterparts had (data not shown). We assessed airway resistance in response to methacholine challenge 48 h after the final OVA challenge and found that Traj18-deficient mice showed significantly less pulmonary resistance than did their wild-type counterparts (Supplementary Fig. 4). For the cockroach antigen model, we challenged mice intranasally with the antigen for 2 weeks on days 1, 3, 6, 8, 10, 13 and 15 and assessed airway resistance in response to methacholine 24 h after the final challenge. In this model as well, Traj18-deficient mice showed significantly less pulmonary resistance than did their wild-type counterparts (Supplementary Fig. 5).

In conclusion, we have generated and characterized a new and potentially useful strain of mice that lack iNKT cells while maintaining an essentially complete TCRα repertoire. We showed that these mice had diminished airway inflammation and resistance, consistent with published studies showing the importance of iNKT cells⁷, but we have also provided definitive evidence that iNKT cells were not only sufficient but also necessary in these contexts. Given other experimental systems in which analysis of mice with deletion of Traj18 has indicated iNKT cells are important, we consider it likely that in many cases, the crucial role of iNKT cells will be confirmed by analysis of the mice of our strain. This will have to be confirmed, however, on a case-by-case basis.

METHODS

Methods and any associated references are available in the online version of the paper.