Inbred strain characteristics impact the NKT cell repertoire

Abstract

Natural killer T (NKT) cells are primed lymphocytes that rapidly secrete cytokines and can directly kill cancerous cells. Given the critical role NKT cells play in cancer immune surveillance, we sought to investigate the effect of mutations in BRCA-1, specifically a conditional deletion of exon 11, on type I iNKT cell development. We observed a significant reduction in iNKT cells in both primary lymphoid and peripheral organs in BRCA-1 mutant mice, compared to wildtype C57BL/6. However, the original BRCA-1 mutant strain was on a mixed background containing FVB/N. We determined that strain differences, rather than mutations in BRCA-1 led to the observed loss in NKT cells. Importantly, we found that while FVB/N mice lack Vβ8, there was striking increase in the total number of thymic type I, CD1d-α-Galactosylceramide (α-GalCer) tetramer positive NKT cells and skewing of the NKT cell population to NKT2, compared to C57BL/6 mice. Collectively, our data demonstrate the profound effect genetics can have on NKT cell subset differentiation.

Introduction

Invariant natural killer T (NKT) cells are an integral component of the immune system. NKT cells recognize lipid antigens presented in the context of CD1d, a non-polymorphic major histocompatibility complex class I-like antigen-presenting molecule (1). NKT cells possess an invariant T-cell antigen receptor (TCR) made up of a single invariant TCRα chain (Vα14-Jα18 in mice, Vα24-Jα18 in humans) in combination with specific TCRβ chains (Vβ8.2, 7 or 2 in mice, Vβ11 in humans) to engage CD1d (2–5). Upon activation, NKT cells directly kill infected or malignant cells, as well as rapidly produce cytokines that stimulate other immune cells, thus they are potent effector cells (6–9). While NKT cells comprise a relatively small population of T cells, their unique effector functions establish them as an important regulatory cell population that can guide the development of immune responses. However, NKT cell number and activity are reduced in multiple cancer types and in chronic infections, therefore their development and effector functions are being widely investigated in numerous mouse models (10–12).

NKT cells develop in the thymus with other T lymphocytes, but they diverge when they reach the double positive stage. Instead of being selected on thymic epithelial cells, they are selected by other double positive thymocytes (13). This selection event is dependent on CD1d:TCR engagement and SLAM:SLAM homotypic interactions which start the NKT cell development program by upregulating the transcription factors Egr2 and PLZF (14–18). Similar to conventional T helper subsets, NKT cells have 3 subsets termed NKT1, NKT2, and NKT17. NKT1 cells express T-bet and primarily secrete IFN-γ, NKT2 cells express high levels of GATA3 and PLZF and secrete Th2-type cytokines such as IL-4 and IL-13; whereas, NKT17 express intermediate levels of PLZF, are RORγt+ and secrete IL-17 (19–21). Despite effector differentiation occurring during thymic development, significant plasticity in cytokine production has been demonstrated after stimulation (22). This could contribute to the discovery of additional NKT cell subsets, including IL-9 producing NKT cells at mucosal surfaces, NKTFH (follicular helper) cells expressing Bcl-6 and providing help to B cells, and NKT10 cells expressing Nfil3 (E4BP4), not PLZF, and secreting IL-10 (23–26). NKT cell subsets are thought to differentiate in the thymus and have effects on the conventional T cells developing around them (27).

In this study we initially sought to investigate the role of the Breast Cancer Susceptibility gene 1 (BRCA1) on NKT cell development. BRCA1 encodes a tumor suppressor protein that functions, in part, as a caretaker gene in preserving chromosomal stability (28). Mutations in BRCA1 account for the majority of hereditary breast cancer and predominantly result in truncation of the BRCA-1 protein (29–31). Due to the importance of BRCA-1 in DNA damage repair, complete knockout is embryonically lethal and studies investigating the effects of BRCA-1 mutation have largely used conditional deletions or mutations (32). Previous studies have shown that conditional deletion of BRCA-1 exons 5 and 6 under the Lck promoter resulted in disruption of T cell lineage development due to increased apoptosis and decreased proliferation (33). Backcrossing this conditional mutant onto backgrounds that suppressed apoptosis (Bcl-2 overexpression) or the DNA damage response (p53 KO) restored the T cell lineage (33). Another group observed normal T cell development when BRCA-1 exon 11 was conditionally excised under the MMTV promoter (34). In this model, when Cre is expressed by activation of the MMTV promoter, exon 11 is excised, producing a shortened form of BRCA-1 (35). Although MMTV (mouse mammary tumor virus) suggests restricted expression to the mammary tissue, this promoter is leaky and results in excision of exon 11 in most tissues, including lymphoid tissue (35–37). Human BRCA-1 mutations are frequently located in exon 11, thus, we used this model to examine NKT cell development (31). This model had been backcrossed onto C57BL/6 and then maintained for several years by inbreeding. While T cell development appeared normal, we observed a significant decrease in and phenotypic alteration of NKT cells in the thymus and in the periphery in these BRCA-1 mutant mice, compared to C57BL/6. In order to thoroughly characterize the mechanisms accounting for the loss of NKT cells, we reestablished the model in order to have BRCA-1+/+ littermate controls. Surprisingly, the phenotype was lost when we compared F3 wildtype and BRCA mutant littermates. Given that the MMTV-Cre strain was originally established on the FVB/N background and that transgenic mice are frequently generated on this background, we assessed the NKT cell population in FVB/N mice. Although FVB/N mice lack Vβ8, we observed a striking increase in the total number of NKT cells and skewing of the NKT cell population to NKT2. Importantly, our data demonstrate the profound effect strain crossing can have on NKT cell subset differentiation and further suggest that outcomes observed when investigating different diseases in specific mouse strains may be differentially affected by the immune cell composition and subsequent skewing of the cytokine profile.

Materials and Methods

Mice

Male and female C57BL/6J, FVB/N, and MMTV-Cre mice were either bred inhouse or purchased from The Jackson Laboratory (Bar Harbor, ME). The original MMTV-Cre-BRCA-1^fl/fl mice were kindly provided by Dr. Laundette Jones (University of Maryland, Baltimore). The genetic background check was conducted by DartMouse Lab (Lebanon, NH). BRCA-1^fl/fl mice that had been backcrossed onto C57BL/6 for 10 generations (01XC8 STOCK Brca1tm2Cxd) were generously provided by Dr. Andre Nussenzweig (NIH/NCI) and maintained in specific pathogen-free facilities at the University of Maryland, Baltimore. BRCA-1^fl/fl mice were bred with MMTV-Cre mice purchased from The Jackson Laboratory (Tg(MMTV-cre)4Mam/J) and progeny were backcrossed onto BRCA-1^fl/fl twice to produce BRCA-1^Δ11/Δ11 experimental animals and BRCA-1^fl/fl littermate controls. C57BL/6 CD45.1+, CD45.2+ mice were kindly provided by Nevil J. Singh (University of Maryland, Baltimore). All mice were age and sex-matched, and were used between 4–8 weeks of age. Data were similar between male and female mice. All experiments were performed in accordance with procedures approved by the University of Maryland School of Medicine animal care and use committee.

Isolation of mononuclear cells

Thymus, spleen, and liver were harvested from mice, processed into single cell suspensions as described previously, and analyzed by flow cytometry (38).

Thymocyte Proliferation Assay

Splenocytes were harvested from C57BL/6 CD45.1+, CD45.2+ mice and loaded with vehicle or 10ng/mL α-GC overnight in 48-well plates. Thymocytes were harvested from experimental and littermate control animals. Thymic NKT cells were enriched by depletion using α-CD8α (clone 53–6.7, BioLegend at 5μg/mL) and α-CD24 (clone M1/69, BioLegend at 10μg/mL) and Sheep α-Rat IgG Dynabeads (Thermo Fisher). Enriched NKT cells were loaded with 1μM CTV for 10 minutes at 37C. Antigen-loaded splenocytes were harvested and washed. Enriched NKT cells and antigen-loaded feeder cells were plated in a 48 well plate at a 1:1 ratio with approximately 2×10⁶ cells/well and co-cultured for 48 and 72 hours. Cells were harvested for analysis by flow cytometry.

Flow Cytometry

Cells were stained in staining buffer: PBS containing 0.5% BSA and 2 mM EDTA. For surface staining, cells were resuspended in staining buffer with FcBlock (BioLegend, clone 93) and incubated at room temperature for 15 minutes. Surface stain mixture, potentially including CD45.1 (BioLegend, clone A20), CD45.2 (BioLegend, clone 104), CD3ε (BioLegend, clone 145–2C11), TCRβ (BioLegend, clone H57–597), CD8β (Thermo Fisher, clone H35–17.2), CD24 (BioLegend, clone M1/69), CD69 (BioLegend, clone H1.2f3), CD28 (BioLegend, clone 37.51), CD44 (BioLegend, clone IM7), NK1.1 (BioLegend, clone PK136), Vβ7 (BioLegend, clone TR310), Vβ2 (BioLegend, clone B20.6), Vβ8 (BioLegend, clone KJ16–133.18), Unloaded or PBS-57:CD1d Tetramer (NIH Tetramer Core Facility), and fixable live-dead (Thermo Fisher), was added to each well and incubated at 4C for 1.5 hours. Surface stained cells were washed twice with 200μL of staining buffer and resuspended in 200μL staining buffer. ICS cells were fixed in BD cytofix/cytoperm for 20 minutes on ice followed by Thermo Fisher FoxP3 fix/perm for 2 hours on ice and then permeabilized by 2 washes with Thermo Fisher FoxP3 perm wash. ICS samples were resuspended in FcBlock in perm wash for 10 minutes on ice and then ICS stain mixture, including PLZF (Thermo Fisher, clone Mags.21f7), Egr2 (Thermo Fisher, clone Erongr2), and RORγt (Thermo Fisher, clone Afkjs-9), was added to each well and incubated for 1 hour on ice. ICS samples were washed twice with perm wash and once with staining buffer then resuspended in 150μL staining buffer. Samples were analyzed at the UMGCCC Flow Cytometry Core on the BD Canto II, BD LSR II, or Cytek Aurora and data analysis was performed using FCS Express 6 Flow Research Edition by de Novo Software. For Figure 1, the gating strategy was as follows: Lymphocytes were selected on an FSC-A × SSC-A plot, NKT cells were gated on an α-GalCer Tetramer × CD3 plot. For Figures 2–4, the gating strategy was as follows: Singlets were selected using a FSC-H × FSC-A plot, Live cells were isolated using a FSC-A × Live/Dead plot, CD45.2 single positive cells were gated on using a CD45.1 × CD45.2 plot, α-GalCer Tetramer, TCRβ double positive cells were selected using a α-GalCer Tetramer × TCRβ plot. NKT cell expression of NK1.1, CD44, CTV, CD28, CD69, PLZF, and RORγt were examined using dot plots and/or histograms. Prism GraphPad was used for graphing data and performing statistical analyses.

Figure 1:

The NKT cell compartment is reduced and phenotypically altered in BRCA-1 mutant mice compared to C57BL/6 mice. (a) Thymocytes, splenocytes and liver mononuclear cells were harvested from 2-month old C57BL/6 wildtype (WT, n=2) and BRCA-1 mutant (MMTV-Cre;BRCA-1^fl/fl, n=4) mice. Cells were stained with α-CD3ε and α-GalCer loaded tetramer. (b) Average percentage of NKT cells and (c) absolute cell numbers were calculated using cell counts and gating percentages. Data from one representative experiment of >5 is shown. (d) Thymic cells were stained with α-CD3ε and α-GalCer-loaded CD1d tetramer to gate on total NKT cells and then the developmental stages were stratified by staining with α-CD44 and α-NK1.1. (e) Average percentage of stage 1, 2, and 3 NKT cells and (f) absolute cell numbers were calculated using cell counts and gating percentages. (g) DartMouse strain determination and (h) SNP mapping. Statistical significance was determined by t test. p<0.05

Figure 2:

NKT cells from BRCA-1 mutant mice are the same as littermate controls. (a) BRCA-1 fl/fl and Δ11/Δ11 mice were bred over the course of three backcrosses. Thymic cells were extracted from 1–2 month-old mice (n=2/group) and were stained for CD24-, CD1dTet+ NKT cells (b) and then stratified into stages by expression of NK1.1 and CD44 (c) and subsets by expression of PLZF and RORγt (d). The percentage (e) and absolute cell count (f) of CD24- NKT cells. The percentage (g) and absolute cell count (h) of stage 1, stage 2, and stage 3 NKT cells. The percentage (i) and absolute cell count (j) of NKT1, NKT2, and NKT17 cells. (k) Thymic NKT cells were stimulated for 48 hours with vehicle or α-GalCer loaded splenocytes and then stained for NKT cells (TCRβ+, CD1dTet+) and proliferation was assessed by CTV dilution. The percentage of NKT cells in each division bin for 48 (l) and 72 (m) hours. Statistical significance was determined by t test.

Figure 4:

FVB/N NKT cells do not express Vβ8, have lower CD1d expression, and higher Egr2 expression. (a) Thymic NKT cells from 1–2 month old FVB/N mice (n=3) were stained for β chain expression. (b) CD1d expression was examined on double positive thymocytes from 1–2 month old FVB/N (n=3) and C57BL/6 (n=3) mice and MFI was graphed (c). C57BL/6 and FVB/N NKT cells were stratified by subtype and then Egr2 expression was examined (d). (e) Geometric MFI of Egr2 expression. Statistical significance was determined by t test *** p<0.001, ** p<0.01 and * p<0.05.

Statistical analyses

Two-tailed student’s t test, one-way analysis of variance (ANOVA) or two-way ANOVA were used as appropriate. Specific experimental groups were compared with controls using the Bonferroni posttest. p value less than 0.05 was considered significant. All analyses were performed using Prism 5.02 by GraphPad (La Jolla, CA). *** p<0.001, ** p<0.01 and * p<0.05.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Results

NKT cells are reduced and phenotypically altered in BRCA-1 mutant mice compared to C57BL/6

Exon 11 of BRCA-1 ensures genomic integrity by transducing signals to arrest the cell cycle and initiate DNA repair (29). Here, we hypothesized that BRCA-1, through its regulation of c-Myc (39), affects NKT cell development. To determine if mutations in BRCA-1 have an effect on NKT cell development, we utilized MMTV-Cre;BRCA-1^fl/fl mice, a well-characterized, clinically relevant model (40), and compared their NKT cell populations to those in C57BL/6 mice. When we examined the thymus, liver, and spleens of BRCA-1 mutant mice, we found that the NKT cell population was significantly reduced (Fig 1A, B, C), compared to wildtype control mice. Typically, NKT cells make up 30–40% of the T cells in the liver of C57BL/6 mice and the NKT cells make up 1–2% of the mononuclear population in the spleen. Strikingly, there was a ~60% decrease in the thymus, and 50% reduction in the liver. Similarly, the population in the spleen was only 0.42%, which represents a 50% reduction in NKT cells. These data demonstrate that MMTV-Cre;BRCA-1^fl/fl mice have a significantly reduced population of NKT cells compared to C57BL/6 mice.

BRCA-1 mutant mice have reduced populations of NKT cells prior to tumorigenesis in primary lymphoid organs (thymus) and secondary lymphoid organs (spleen and liver) (Fig 1A). The reductions in the thymic population led us to examine NKT cell development. NKT cell developmental stages can be discerned based on expression of CD44 (stage 2) and NK1.1 (stage 3) (41). We found that BRCA-1 mutant NKT cells are significantly restrained in stage 2 of development with very few progressing to stage 3 (Fig 1D, E, F). Studies from the Hogquist lab have indicated NKT cell development can be affected by strain differences (27). In order to confirm that the MMTV-Cre;BRCA-1^fl/fl mice were on the C57BL/6 background, tail snips were sent to DartMouse. It was determined that the mice were only ~40% C57BL/6 (Fig 1G). Analyses of individual single nucleotide polymorphisms indicated that there were large regions that were homozygous non-B6-like (Fig 1H). In order to minimize this confounding factor, we reestablished the model using BRCA-1^fl/fl mice that had been backcrossed onto C57BL/6 for 10 generations. These mice were crossed with MMTV-Cre mice for two generations in order to obtain mice with BRCA mutations (BRCA-1^Δ11/Δ11) and the proper littermate controls (BRCA-1^fl/fl) (Fig 2A).

NKT cell levels are similar between BRCA-1^Δ11/Δ11 and BRCA-1^fl/fl littermate controls

When we compared mice with BRCA-1 mutations (BRCA-1^Δ11/Δ11) to BRCA-1^fl/fl littermate controls, we detected no significant differences in percentage or absolute number of NKT cells (Fig 2B, E, F). Additionally, the drastic alterations in the stages of NKT cell development observed in the original mouse model were no longer present (Fig 2C, G, H). Moreover, when we assessed thymic NKT cell subset differentiation based on PLZF and ROR-γt expression, we found no significant differences between NKT1, NKT2, or NKT17 cells (Fig 2D, I, J). NKT cells are known to undergo high levels of proliferation during development (42). Thus, we sought to investigate if there were differences in the proliferative potential in BRCA-1 mutant mice. In this assay, splenocytes loaded with vehicle or α-GalCer were used as feeder cells and co-cultured with pre-enriched thymic NKT cells. α-GalCer stimulation resulted in similar levels of expansion in both groups at 48 and 72 hours (Fig 2K, L, M). These data suggest that strain differences accounted for the variations in the NKT cell population observed in the original mouse model, rather than functional BRCA-1 expression.

FVB/N mice have higher levels of NKT cell levels compared to C57BL/6 and MMTV-Cre background mice

MMTV-Cre mice were initially developed on the FVB/N background. Thus, we compared thymic NKT cell populations in C57BL/6 mice to FVB/N and the BRCA-1 mice crossed with MMTV-Cre. While the MMTV-Cre background mice had significantly reduced NKT cells compared to C57BL/6, FVB/N mice had significantly higher levels of NKT cells compared to the other strains (Fig 3A, D, E).

Figure 3:

FVB/N mice have higher levels of NKT cell levels compared to C57BL/6 and MMTV-Cre background mice. Thymic cells from mice 1–2 months of age were stained for CD24-, CD1dTet+ NKT cells (a) and then stratified into stages by expression of NK1.1 and CD44 (b) and subsets by expression of PLZF and RORγt (c). The percentage (d) and absolute cell count (e) of CD24- NKT cells. The percentage (f) and absolute cell count (g) of stage 1, stage 2, and stage 3 NKT cells. The percentage (h) and absolute cell count (i) of NKT1, NKT2, and NKT17 cells. C57BL/6 (n=4), FVB/N (n=4), BRCA-1fl/fl (n=7), BRCA-1Δ11/Δ11 (n=4). Statistical significance was determined by One-way ANOVA and Bonferroni post-test *** p<0.001, ** p<0.01 and * p<0.05.

When we stratified the developmental stages based on expression of CD44 and NK1.1 (Fig 3B, F, G), we found that MMTV-Cre background mice had a significantly higher proportion of cells in stage 2 than C57BL/6, although the majority of cells were still in stage 3. In contrast to C57BL/6, FVB/N NKT cells were primarily in stage 2 of development (Fig 3B, F, G) – similar to the original BRCA-1 mutant mouse model (Fig 1 D, E, F). We examined NKT cell subset differentiation based on transcription factor expression and found that MMTV-Cre background mice did not have significantly different subset differentiation from C57BL/6 mice (Fig 3C, H, I). Conversely, FVB/N mice had a significantly higher percentage of NKT2 and NKT17, compared to C57BL/6 and MMTV-Cre background mice (Fig 3C, H). Although FVB/N mice had a significantly lower percentage of NKT1 cells, they had a significantly higher number of NKT1, NKT2, and NKT17 due to the large size of the thymus and overall greater cellularity (Fig 3I).

FVB/N NKT cells do not express Vβ8, have lower CD1d expression, and higher Egr2 expression

The increased population of NKT cells in FVB/N mice is striking, given that they have been reported to be Vβ8 deficient (43). In agreement, we did not detect Vβ8.1/2 expression; however, the mice express Vβ7 and Vβ2 (Fig 4A), which are important for invariant, type 1 NKT cells. It has been previously reported that decreased expression of CD1d in CD1d +/− mice increases usage of Vβ7 in mature NKT cells (44). Interestingly, we found that FVB/N double positive thymocytes expressed significantly less CD1d (Fig 4B, C) – potentially facilitating the selection of Vβ7+ NKT cells. There is an established correlation between TCR signal strength, Egr2 expression, and subset differentiation (45). We assessed Egr2 expression in FVB/N and C57BL/6 mice (Fig 4D, E). In agreement with prior literature, NKT2 and NKT17 had higher Egr2 expression than NKT1. Strikingly, Egr2 expression was significantly higher in FVB/N mice in both NKT2 and NKT17 cells (Fig 4E).

Discussion

Our results are in agreement with other groups that have demonstrated in a panel of inbred mouse strains that NKT cell numbers were highly variable (46). In one study that examined six different commonly used inbred strains of mice (C57BL/6, NOD, DBA2, CBA, 129, and Balb/c), it was determined that mice of a similar age showed variability in the frequency of total thymic iNKT cells (27). In another study (47), examining NKT cells in NOD mice it was found that the decrease in NKT cells was primarily due to higher levels of CD1d expression on DP thymocytes, which was mapped to a region on chromosome 13. The authors sought to further determine whether the inverse relationship between CD1d expression levels on DP thymocytes and the frequency of thymic iNKT cell existed in other mouse strains. In these studies a panel of inbred strains, CBA/J, BALB/cJ, 129/SvInJ, A/J, FVB/J, SJL/J, C57BL6/J,DBA2/J, and ICR/HaJ mice were examined, and they identified a significant negative correlation (P=0.024) between the CD1d expression level and the frequency of thymic iNKT cells (47). In fact, several studies have investigated the genetic basis for iNKT cell defects that contribute to T1D development in NOD mice (48, 49). A recent study generated a series of B6.129 congenic lines, and showed that there were two genetic elements that regulate iNKT cell cytokine production in response to αGalCer, specifically they demonstrated a role for the Slam gene family and Fcgr3 (50). Herein, we present evidence demonstrating that the phenotype observed in the original BRCA-1 mutant mice was due to an inbred strain characteristic, most likely FVB/N. Moreover, we show that NKT cells from mice on an MMTV-Cre background (F3) are significantly different from C57BL/6 and FVB/N. Specifically, we detected differences in the overall percentage of NKT cells; there were fewer in MMTV-Cre background and more in FVB/N, compared to C57BL/6. In addition, FVB/N mice have significantly altered NKT cell subset differentiation. There was an increase in NKT2 and NKT17, and decreased NKT1 compared to C57BL/6. These differences in NKT cell subset differentiation suggested that there may also be differences in developmental signalling cues during selection perhaps due to alterations in the expression of endogenous antigens or in the cytokine milieu.

TCR composition has been shown to affect NKT cell selection, development, and differentiation (51). In a previous study investigating the mechanisms accounting for the Vβ8, Vβ7, and Vβ2 bias in murine NKT cells, it was found that α-GalCer was recognized by a broader set of Vβ chains, including the traditional Vβ8, Vβ7, and Vβ2, but also Vβ6, Vβ9, Vβ10, and Vβ14 (52). FVB/N mice lack Vβ8 which usually composes 50–60% of C57BL/6 NKT cells (53). It is thought that Vβ8, Vβ7, and Vβ2 form the optimal TCR for recognition and selection of endogenous ligands. However, based on our data showing that FVB/N have a much higher number of α-GalCer tetramer+ NKT cells, and only 70–80% are Vβ7 and Vβ2, there may be other Vβ chains that are potentially useful for the recognition of lipid antigens. We also observed decreased CD1d expression on double positive thymocytes. In C57BL/6 mice, decreased CD1d altered Vβ usage to favor Vβ7 (44). Notably, the mice used in our studies were not co-housed, thus the impact of the microbiota on the differences observed between the groups is unknown.

Recently the Hogquist lab published a study demonstrating that some NKT cells may remain in an undifferentiated precursor stage (NKTp) that leaves the thymus and differentiates in the periphery (54). Each immature stage is marked by a vigorous, c-Myc-driven round of expansion that is necessary to produce mature, stage 3 NKT cells in the periphery (55). We were initially interested in exploring the dynamics of this c-Myc driven expansion due to the relationship between BRCA-1 and c-Myc (39). However, genetic differences in the activity of proteins involved intracellular signalling have been shown to impact NKT differentiation. A recent study by the Kronenberg and Mallevaey groups examined the role of src homology 2 domain– containing phosphatase 1 (Shp1) on NKT cell development and function using a T cell–specific Shp1 deletion (56). Although Shp1fl/fl CD4-cre mice had normal numbers of NKT cells, they observed a cell-intrinsic bias toward iNKT2 and iNKT17 cells in the thymus. Mechanistically, it was found that Shp1 regulated NKT cell proliferation in response to cytokines, specifically IL-2, IL-7, and IL-15. TCR signalling has also been demonstrated to affect NKT differentiation with increased TCR signalling results in increased Egr2 expression – thus NKT2 cells have the highest Egr2 expression. Concomitantly, decreased Zap70 activity led to a decrease in Egr2 expression and an increase in NKT1 populations (45). We examined Egr2 expression and found that FVB/N NKT2 and NKT17 cells had increased Egr2 expression compared to their C57BL/6 counterparts. These data suggest that genetic differences that lead to changes in response to cytokines or TCR signalling may also dictate NKT cell development and differentiation.

Importantly, our results demonstrate the necessity of utilizing littermate controls when investigating immune subsets in genetically manipulated mice. This is particularly important in the field of cancer immunotherapy – where cell types can be hailed as cancer permissive or cancer killing depending on the mouse model used. Genetically engineered cancer models are powerful tools for studying oncology and more recently cancer immunosurveillance and immunotherapy (57). FVB/N mice, specifically, were commonly used for producing transgenic mouse models due to their large pro-nuclei and large litter size (58). Unfortunately, these models are plagued by not only strain crosses to incorporate additional mutations and enhance tumorigenesis, but also maintenance by inbreeding that allows drift away from control animals. For extensively inbred and backcrossed models, re-designing the breeding scheme to produce littermate controls is imperative.