|Home | About | Journals | Submit | Contact Us | Français|
The tyrosine kinase Fyn has been implicated as playing an important role in the generation of both stimulatory and inhibitory signaling events induced by TCR engagement. To assess the role of Fyn for antigen-driven negative selection and regulatory T cell (Treg) development, which are both dependent on the strength and nature of TCR signaling, we generated mice that co-express the transgenes for ovalbumin and the OT-II TCR, which recognizes a peptide from ovalbumin. In mice expressing both transgenes, negative selection, Treg development in the thymus, and the number of Treg in the periphery were each unaffected by ablation of Fyn. Moreover, fyn-/- Treg were functional, as assessed in vitro. We further tested the role of Fyn for the adaptor function of c-Cbl, using mice containing a point mutation in c-Cbl that abolishes its E3 ubiquitin ligase function but maintains its adaptor function. The functional and signaling properties of this mutant c-Cbl were unaltered in fyn-/- thymocytes. Combined, these data indicate that Fyn was not required for: the induction of central tolerance by negative selection; the adaptor protein role of c-Cbl; orthe normal development and function of Treg.
The fate of developing T cells within the thymus is controlled by the strength and nature of signals from the pre-TCR, and from the TCR and its co-receptors, CD4 and CD8 . Signaling though the TCR is responsible for positive and negative selection , and determines if a thymocyte will enter the differentiation pathway leading to the formation of conventional CD4 or CD8 T cells, NKT cells, or Treg [2-4]. Two Src-family kinases, Fyn and Lck are involved in the initiation of TCR-mediated signaling [5, 6]. Fyn is directly associated at low stochiometery with the TCR ζ-chain [7, 8], whereas Lck is associated with the cytoplasmic domains of CD4 and CD8 . Following ligation of the TCR, both Fyn and Lck phosphorylate immunoreceptor-tyrosine based activation motifs (ITAM) within the TCR ζ-chain and CD3 leading to the recruitment and activation of ZAP-70, the recruitment and phosphorylation of LAT (linker for activation of T cells), and the formation of a multicomponent signaling complex that propagates the TCR signal and ultimately determines T cell behavior [5, 6].
Interestingly, Fyn has been implicated not only in initiation and propagation of the TCR signal but also in down-regulation of this signal by phosphorylating proteins that limit TCR signaling . In particular, Fyn interacts with c-Cbl, which has a dual role in TCR signaling. C-Cbl can attenuate the TCR signal though its RING finger E3 ubiquitin ligase domain, by which it catalyses the ubiquitinylation of the TCR, Src family kinase members and other signaling proteins, targeting them for internalization and degradation . Conversely, phosphorylation of c-Cbl on Y731 by Fyn allows c-Cbl to act as an adapter protein for phosphoinositide-3 kinase (PI-3 kinase), thereby enhancing TCR signaling [12, 13]. The phosphorylation of c-Cbl, leading to PI-3 kinase recruitment is thought to be dependent on Fyn [11, 12].
Although development of conventional CD4+ and CD8+ T cells does not appear to be greatly affected in Fyn-deficient mice (fyn-/-), there is evidence to suggest that TCR-mediated signaling is altered in fyn-/- thymocytes. In response to stimulation with anti-CD3 antibodies, fyn-/- thymocytes do not flux calcium, have diminished phosphorylation of various signaling molecules including ZAP-70 and LAT, and do not proliferate as well as WT thymocytes [14-16]. In addition, negative selection in the thymus in response to the endogenous-super-antigen, Mls-1a is decreased in fyn-/- mice . As there are important differences between anti-CD3 antibodies, super-antigens, and peptide-major histocompatibility (MHC) complexes as ligands for the TCR, particularly in regard to the involvement of the co-receptors CD4 and CD8 , we wished to determine the effect of Fyn-deficiency on negative selection and thymic development of CD4+ T cells in vivo under more physiological conditions. For this purpose we generated fyn-/- mice expressing the transgene-encoded OT-II TCR, which recognizes a peptide from chicken egg albumin (Ovap323-339) in the context of the I-Ab class-II MHC molecule . The development of OT-II TCR transgenic (OT-II+) fyn-/- thymocytes was then assessed in mice that expressed ovalbumin in the thymus and periphery under the control of the rat insulin promoter (RIP). Here, we report that both positive and negative selection of OT-II+CD4+ T cells were not altered by ablation of Fyn, and that development and function of Treg were also unaffected when compared to WT. These results reveal a high level of redundancy between Fyn and other proteins kinases for T cell function.
To determine if Fyn-deficiency affected either positive or negative selection of MHC-class II-restricted thymocytes, we compared T cell development in OT-II wild type (WT) mice and fyn-/- mice, in the presence or absence of transgenic ovalbumin. In the thymus of OT-II WT mice and OT-II fyn-/- mice, the frequency of CD4-CD8- (double negative, DN), CD4+CD8+ (double positive, DP), CD4+CD8- (CD4 single positive, SP), and CD4-CD8+ thymocytes (Supplementary Figure 1) and the number of thymocytes (Figure 1A) were similar. Both OT-II WT mice and OT-II fyn-/- mice had an increased frequency of CD4+ single positive (SP) thymocytes (mean ± standard deviation (SD): 18.1 ± 3.9% n=7, and 15 ± 2.5% n=6, respectively, student T test p=0.15) when compared to non-transgenic controls (mean ± SD: 5.6 ± 0.31% SD n=4 WT mice; and 5.67 ± 0.48% n=4 G15 fyn-/- mice). These results indicate that positive selection of TCR-transgenic thymocytes occurs normally in OT-II fyn-/- mice, and concurs with previous studies on positive selection of thymocytes expressing either CD4- or CD8-selecting TCR transgenes in fyn-/- animals [19-21]. In contrast, in the thymi of RIP-Ova OT-II WT mice and RIP-Ova OT-II fyn-/- mice there was a marked decrease in the frequency of CD4+ SP thymocytes (mean ± SD: 2.8 ±1.1% n=6 and 3.24± 1.3% n=4, respectively, student T test p=0.6), and a corresponding increase in the other thymocyte sub-populations when compared to OT-II mice that did not express ovalbumin (Supplementary Figure 1). In addition, both RIP-Ova OT-II fyn-/- mice and RIP-Ova OT-II WT mice showed a significant and similar decline in the total number of thymocytes when compared to controls (Figure 1A), which was primarily due to a loss of OT-II+CD4+CD8- thymocytes, as assessed by the expression of the α-chain of the OT-II TCR, Vα2 (Figure 1B). Both RIP-Ova OT-II WT mice and RIP-Ova OT-II fyn-/- mice had a nearly complete reduction in the number of OT-II+ CD4 SP thymocytes when compared to non-Ova expressing OT-II+ controls. The mean percentage deletion of OT-II+CD4+CD8- thymocytes was 94% (± 4.0% SD n=5) in RIP-Ova OT-II WT mice and 88% (± 6.9% SD n=5) in RIP-Ova OT-II fyn-/- mice. The modest difference seen between RIP-Ova OT-II WT mice and RIP-Ova OT-II fyn-/- animals was not statistically significant (student T test p=0.24). The lack of a defect in negative selection in OT-II+ transgenic fyn-/- mice is in contrast to the observation that fyn-/- mice have a defect in thymic negative selection induced by the super-antigen, Mls-1a . However, challenge of fyn-/- mice with the super-antigen Staphylococcal enterotoxin A , or the development of H-Y TCR transgenic Fyn-deficient CD8+ T cells in male mice also failed to reveal defects in negative selection . We have now extended these observations to negative selection of antigen-specific MHC class II-restricted thymocytes. Thus, the role of Fyn in negative selection by Mls-1a appears to be atypical and our data combined with that of others [15, 19] demonstrate that negative selection occurs normally in fyn-/- mice.
To determine if the expression of cell surface molecules sensitive to antigenic stimulation were altered in the absence of Fyn, we used flow cytometry to measure the expression on thymocytes of the α- and β-chains of the OT-II TCR (Vα2 and Vβ5), CD69, and CD5. On DN and DP thymocytes there was no significant difference in the expression of these cell surface markers when comparing OT-II WT mice and OT-II fyn-/- mice, irrespective of the presence of the RIP-Ova transgene (data not shown). However, when the Vα2+CD4+ SP thymocytes were examined, the expression of both Vα2 and Vβ5 was down-regulated on thymocytes from RIP-Ova OT-II WT mice when compared to OT-II WT mice (Figure 1C and data not shown, the mean frequency of CD4 SP thymocytes that expressed Vα2 was 94.9 ± 2.8% SD n=8 for OT-II WT mice and 66 ± 21% SD n=5 for RIP-Ova OT-II WT mice), and a similar down-regulation of OT-II TCR expression was also observed in CD4 SP thymocytes from RIP-Ova OT-II fyn-/- mice (Figure 1D and data not shown, the mean frequency of CD4 SP thymocytes that expressed Vα2 was 94.4 ± 2.8% SD n=7 for OT-II fyn-/- mice and 61 ± 27% SD n=4 for RIP-Ova OT-II fyn-/- mice). Moreover, the expression of CD5 was higher on Vα2+CD4 SP thymocytes from both RIP-Ova OT-II WT and fyn-/- mice when compared to non-ovalubmin expressing controls (Figure 1E and F, respectively). The changes in cell surface marker expression observed indicate that OT-II+ thymocytes in RIP-Ova mice had encountered antigen in the thymus. In particular, the upregulation of CD5 is consistent with the OTII+ thymocytes having received a strong TCR signal following recognition of the neo self-antigen ovalbumin. Down-regulation of the TCR and up-regulation of CD5 are both known to modulate TCR signaling in a manner that reduces the magnitude of that signal . These changes occurred to a similar degree in both Fyn-expressing and Fyn-deficient transgenic thymocytes suggesting that the loss of Fyn did not significantly alter the strength of TCR signaling in CD4 SP thymocytes responding to specific antigen/MHC-II complexes.
Previous work has demonstrated that some α/β TCR expressing thymocytes adopt a CD4+ regulatory T cell fate following self-antigen recognition in the thymus [23, 24]. To assess if Fyn-deficiency altered the development of these cells, we determined the frequency of Foxp3+CD4+ thymocytes in WT and fyn-/-mice in the presence and absence of the OT-II TCR and RIP-Ova transgenes. The frequency of Foxp3+CD4 SP thymocytes was somewhat higher in non-transgenic G15 fyn-/- mice when compared to C57BL/6 mice (mean percentage of Foxp3+CD4 SP thymocytes ± SD: 4.4 ± 0.48% n=4 and 3.6 ± 0.2% n=4, respectively; student T test p = 0.04), however this difference was not observed when comparing OT-II TCR transgenic WT and fyn-/- mice. OT-II WT and fyn-/- mice had a lower frequency of Foxp3+Vα2+ CD4 SP thymocytes compared to the percentage of Foxp3+ CD4 SP thymocytes in non-transgenic mice, and this frequency did not significantly differ between thymocytes from OT-II fyn-/- and OT-II WT mice (Figure 1G and H. Mean percentage of Foxp3+Vα2+CD4 SP thymocytes ± SD; 0.51 ± 0.25% n=8, and 0.52 ± 0.19% n=8, respectively, student T test p = 0.3). The low frequency of Foxp3+CD4 SP thymocytes in the OT-II mice presumably reflected a low level of self-reactivity of OT-II+ thymocytes in the absence of ovalbumin. Upon inclusion of the RIP-Ova transgene, there was a large increase in the proportion and absolute numbers of Vα2+CD4 SP thymocytes that were Foxp3+, as has been seen in several other TCR and antigen transgenic mice [3, 23-25]. The increase in Foxp3+ Treg was similar in OT-II WT and fyn-/- mice (Figure 1G and H. Mean percentage of Foxp3+Vα2+CD4 SP thymocytes ± SD; 9.98 ± 0.47% n=3, and 9.88 ± 3.7% n=4, respectively, student T test p = 0.4). These data clearly show that the thymic generation of self-reactive Foxp3+ Treg, at least with the OT-II TCR, was unaffected by a deficiency of Fyn.
The lack of a substantial effect of Fyn-deficiency thymocyte development was surprising given evidence for a unique role for Fyn in phosphorylating several key regulators of TCR signaling [7, 13, 26-28]. One such regulator is c-Cbl. To test the requirement for Fyn for the adaptor function of c-Cbl, we took a genetic approach. Mice have been generated in which the c-Cbl adaptor function is intact but the E3 ubiquitin ligase function is lacking. These mice carry an allele of the c-Cbl gene carrying a point mutation that inactivates the c-Cbl RING finger, combined with a total loss of function allele (c-CblA/-). The c-CblA/- mice have a rapid loss of thymic cellularity that correlates with elevated cell surface expression of the TCR, CD5 and CD69 molecules in DP thymocytes (Figure 2A-C and ). Thymocytes from c-CblA/- mice also have enhanced intracellular signaling consistent with the loss of the inhibitory function of c-Cbl combined with retention of its positive adapter protein role. It is thought that stronger TCR signaling resulting from altered c-Cbl function leads to increased negative selection, explaining the reduction in thymic cellularity and the altered cell surface phenotype of c-CblA/- thymocytes . Analysis of signaling pathways in these thymocytes indicated that there is hyper-induction of the PI-3 kinase/Akt pathway in c-CblA/- thymocytes. As Fyn has been reported to phosphorylate c-Cbl on Y731, we assessed whether the high rate of thymocyte apoptosis in c-CblA/- mice was dependent on Fyn by crossing these c-Cbl alleles to the fyn-/- mice. Analysis of 4-5 week old fyn-/-c-CblA/- mice revealed that Fyn-deficiency failed to restore thymocyte numbers in the c-CblA/- mice. Thymocyte numbers were 9.0 ± 2.4 × 107 in fyn-/-c-CblA/- mice and 26.7 ± 3.3 × 107 in their fyn-/-c-Cbl+/- littermates, (n=2), compared with 6.7 ± 1.6 × 107 in c-CblA/- and 27.4 ± 3.9 × 107 in c-Cbl+/- littermates (n=6-7). Consistent with this observation, the expression of the TCR, CD5 and CD69 were similarly elevated on fyn-/-c-CblA/- DP thymocytes and on c-CblA/- thymocytes (Figure 2A-F). Moreover, Fyn-deficiency did not reduce the extent of c-Cbl hyperphosphorylation and increased phosphorylation of Akt on activation sites in these thymocytes (Figure 2G). Analysis of cell lysates from anti-CD3 stimulated c-Cbl-/- thymocytes showed that the 120kD band was correctly identified as c-Cbl (data not shown and ). Thus, the phosphorylation of c-Cbl leading to recruitment and activation of PI-3 kinase was not compromised by ablation of the expression of Fyn, suggesting that in the absence of Fyn another protein tyrosine kinase is able to phosphorylate c-Cbl. Previous studies have indicated that Yes or Syk can also phosphorylate c-Cbl , and these kinases may provide a level of redundancy allowing for the phosphorylation of c-Cbl in thymocytes lacking Fyn. In contrast, decreased phosphorylation of c-Cbl has been reported in fyn-/- splenic T cells compared to WT T cells . This difference correlates with the expression pattern of Syk, which is not expressed in most splenic T cells, but is expressed in DP thymocytes , consistent with the possibility that Syk provides redundancy with Fyn for phosphorylation of c-Cbl in thymocytes.
Although the majority of OT-II+CD4+ T cells were deleted in the thymus of RIP-Ova OT-II WT and fyn-/- mice, some TCR transgenic CD4+ T cells escaped negative selection and could be identified in the periphery, albeit in low numbers (Figure 3). In RIP-Ova OT-II WT mice and RIP-Ova OT-II fyn-/- mice, the total number of splenocytes was reduced when compared to their non-Ova expressing OT-II TCR transgenic counterparts (Figure 3A), largely due to a reduction in the number of splenic OT-II+CD4+ T cells (Figure 3B). The loss of Vα2+CD4+ T cells was similar in both RIP-Ova OT-II WT mice and RIP-Ova OT-II fyn-/- mice (Figure 3B). Peripheral CD4+ T cells from both RIP-Ova OT-II WT mice and RIP-Ova OT-II fyn-/- mice had phenotypic changes that were consistent with down-regulation of the TCR signal, including the reduced expression of Vα2 and Vβ5 and increased expression of CD5 on Vα2+CD4+ T cells similar to that observed in the thymus (Figure 1 and data not shown). In addition, the early activation marker, CD69, was expressed by less than 10% of the OT-II+CD4+ T cells from RIP-Ova OT-II WT mice or from RIP-Ova OT-II fyn-/- mice (data not shown). The similarity in the expression of the cell surface molecules examined on Vα2+CD4+ T cells from RIP-Ova OT-II WT and fyn-/- mice indicates that a Fyn-deficiency did not have a discernible effect on the phenotype of these cells, and suggests that the response of OT-II+CD4+ T cells following recognition of Ovap323-339/MHC-II complexes was not significantly altered in the absence of Fyn kinase. Moreover, the low level of expression of CD69 on OT-II+CD4+ T cells in the RIP-Ova OT-II mice despite the continued presence of antigen in the periphery suggests that there was not an ongoing immune response to Ova in either WT or fyn-/- mice.
Expression of the RIP-Ova transgene led to a large increase in the frequency and absolute number of Foxp3+OT-II+ T cells, and there was no evidence of an effect of a Fyn-deficiency on the increase in the frequency or number of these Treg (Figure 3C and mean number of splenic Foxp3+OT-II+CD4+ T cells ± SD: 10.5 ± 3.7 × 106, n=4 RIP-Ova OT-II WT mice and 12.9 ± 6.0 SD × 106 n=4 RIP-Ova OT-II fyn-/- mice, student T test p = 0.5). In contrast, the numbers were significantly different between OT-II WT mice and RIP-Ova OT-II WT mice or between OT-II fyn-/- mice and RIP-Ova OT-II fyn-/- mice (Figure 3C. Mean number of splenic Foxp3+OT-II+CD4+ T cells ± SD: 1.4 ± 0.6 × 106 n=6 OT-II WT mice and 2.2 ± 1.5 × 106 n=6 OT-II fyn-/- mice. Student T test p values = 0.0008 or 0.006 when comparing RIP-Ova OT-II WT and OT-II WT, or RIP-Ova OT-II fyn-/- and OT-II fyn-/- mice, respectively). In addition, there were no significant differences in the numbers of splenic Foxp3+CD4+ T cells between WT and fyn-/- mice (Mean number of splenic Foxp3+CD4+ T cells ± SD: 7.1 ± 0.8 × 106 cells n=4 WT mice and 10.8 ± 2.6 × 106 cells n=4 fyn-/- mice; student T test p=0.1. Figure 3D). These data suggest that many of the OT-II+ thymocytes in RIP-Ova OT-II WT mice and in RIP-Ova OT-II fyn-/- mice develop into Treg and are exported to the periphery as Foxp3+ Treg, and that this process is little affected by the expression of Fyn.
We next wanted to assess whether Treg from fyn-/-mice were functionally similar to their WT counterparts. To address this question, the activity of Treg was assessed using an in vitro system previously described by Thornton and Shevach (2000) in which the ability of Treg to inhibit anti-CD3 induced proliferation of naïve T cells is assessed . Naive T cells isolated from non-transgenic WT mice or fyn-/- mice were stimulated with soluble anti-CD3, in the presence of γ-irradiated splenocytes and variable numbers of WT or fyn-/- Treg. Non-transgenic T cells were used in these experiments, as it was difficult to obtain sufficient numbers of Treg from RIP-Ova OT-II mice. WT and Fyn-deficient naïve T cells exhibited similar levels of 3H-thymidine incorporation in this assay (Figure 3E and F). The addition of Treg to these cultures resulted in the inhibition of 3H-TdR incorporation. The extent of inhibition of naïve T cell proliferation was dependent on the ratio of naïve:regulatory T cells present in the culture, with the most pronounced inhibition (85-90%) occurring at a ratio of naïve:regulatory T cell of 1:1 (Figure 3E and F). Similar results were obtained by co-culturing WT naïve T cells with Treg from either WT mice or fyn-/- mice (Figure 3E), or conversely by co-culturing fyn-/- naïve T cells with Treg from WT mice or fyn-/- mice (Figure 3F). These results demonstrate that fyn-/- naïve T cells are as susceptible to the suppressive activity of Treg as naïve WT T cells and that fyn-/- Treg are as capable as their WT counterparts at inhibiting naïve T cell proliferation in vitro. These data demonstrate that signaling though Fyn is not required for Treg function in vitro, and suggest that fyn-/- Treg would be capable of participating in the maintenance of peripheral tolerance. However, the efficacy of fyn-/- Treg and their ability to maintain peripheral tolerance in vivo remain to be assessed.
The intracellular protein tyrosine kinase Fyn has been implicated as being important for T cell function due to its involvement in early TCR signaling events following antigen recognition [5, 6], and its role in signaling by a subset of CD2-family members via the adaptor SAP [33, 34]. In the experiments presented here, we have addressed the question of whether Fyn has non-redundant functions in TCR signaling reactions that are important for establishment of tolerance to self in the CD4+ T cell compartment. We examined the effect of Fyn-deficiency on negative selection and regulatory T cell fate choice by OT-II TCR transgenic T cells in the absence or presence of a transgene expressing its cognate antigen, ovalbumin. Our data show that Fyn-deficiency does not alter either of these two key mechanisms of central T cell tolerance. In addition, Fyn-deficiency was shown to be dispensible for the adaptor function of c-Cbl in thymocytes, in contrast to earlier reports that this function was dependent on Fyn. Finally, Fyn-deficient Treg effectively suppressed the in vitro proliferation of naïve effector CD4+ T cells from WT or fyn-/- animals, which suggests that a Fyn-deficiency may not impair this arm of peripheral tolerance. These data indicate that TCR signaling in class II MHC-restricted TCR α/β thymocytes and in peripheral Treg is sufficiently flexible to be able to compensate for the lack of signaling though Fyn, even at the level of individual signaling targets such as c-Cbl.
Thymocyte and peripheral T cell populations were assessed in sex- and age-matched mice at 6-7 weeks of age. Experiments using c-Cbl mutant mice were done in sex-and age-matched fyn-/-c-Cbl+/- and fyn-/-c-CblA/- littermates at 4-5 weeks of age. Assessment of Treg activity was done using cells from sex- and age-matched mice at 8-10 weeks of age.
C57BL/6 mice were originally purchased from Jackson Laboratory (Bar Harbor, ME). C57BL/6 fyn-/- mice were generated by crossing mice expressing the inactivated fyn-allele , on a mixed C57BL6/129S7/Sv background to C57BL/6 mice for 15 generations (G15) and then to homozygosity for the inactivated fyn-allele. C57BL/6 mice expressing the OT-II TCR transgene  were provided by Dr Mark Anderson (University of California, San Francisco). To generate OT-II fyn-/- mice, G15 fyn-/- female mice were crossed to the OT-II males. RIP-Ova C57BL/6 mice , were a gift from Dr M. Anderson and were used to generate RIP-Ova OT-II fyn-/- mice and fyn+/+ counterparts. Fyn-deficient mice were bred to c-CblC379A (c-CblA/+) and to c-Cbl-/- mice  to generate fyn-/-c-CblA/+ and fyn-/-c-Cbl-/- mice. These mice were then mated to generate fyn-/-c-Cbl+/- and fyn-/-c-CblA/- littermates.
With the exception of the c-Cbl mutant mice, all mice were housed and maintained within the specific pathogen-free animal facility at UCSF under conditions that meet institutional animal care and use committee (IACUC) and National Institutes of Health guidelines. Fyn-/-c-CblA/- and fyn-/-c-cbl+/- mice were housed at animal facilities at the University of Western Australia in accordance with Animal Ethics Committee approval.
The genotype of the RIP-Ova transgenic mice and fyn-/- mice used in these experiments was confirmed by PCR. The OT-II TCR transgene is carried on the Y chromosome so all males have the TCR transgene, the expression of the OT-II TCR was periodically confirmed by flow cytometry based on the high frequency of CD4+ T cells that expressed Vα2. The genotype of fyn-/-c-Cbl+/- and fyn-/-c-CblA/- mice was confirmed by PCR as previously described .
Anti-CD3 (145-2C11 or 500A2), anti-CD4 (RM4-5 or GK1.5), anti-CD8α (53-6.7), anti-TCRβ (H57-597), anti-Vα2 (B20.1), anti-Vβ5 (MR9-4), anti-CD69 (H1.2FE), anti-CD25 (2A3), anti-CD5 (53-7.3), anti-CD62L (MEL-14), anti-CD16/CD32 (2.4G2), and isotype controls for rat IgG2a κ (R35-95), mouse IgG1 κ (MOPC-31C), Armenian hamster IgG1 κ (A19-3) unconjugated or conjugated to biotin, fluorescein isothiocyanate (FITC), allophycocyanin (APC), phycoerythrin (PE), PE-Cy7, PerCp-Cy5.5 as needed were purchased from Becton Dickinson PharMingen (San Diego, CA). Anti-Fyn (FYN3) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and anti-c-Cbl from BD Transduction Laboratories (San Jose, CA). Anti-phospho-Akt (S473) (193H12) and anti-Akt1 (2H10) were purchased from Cell Signaling Technology (Danvers, MA). The mouse monoclonal anti-phosphotyrosine (4G10) antibody was a gift from Dr Brian Druker (Oregon Health and Science University, Portland).
Cells from thymus or spleen were prepared for analysis or purification by flow cytometry as previously described . For intracellular staining for Foxp3, thymocytes or splenocytes were prepared as previously described  and stained with anti-CD4-APC, anti-CD8-PECy7 and anti-Vα2-PE, or with anti-CD4-PECy7, anti-CD25-APC and anti-Vα2-PE, respectively. Intracellular cellular staining for Foxp3 was done using the FITC anti-mouse/rat Foxp3 Staining set from eBioscience (#71-5775 San Diego, CA), as per the manufacturer's instructions.
Calculation of the percentage deletion of thymocytes
The percentage deletion was calculated by the following formula, % deletion = 100 − [(number of Vα2+CD4 SP thymocytes from RIP-Ova OT-II mice / the mean for the number of Vα2+CD4 SP thymocytes from OT-II mice) ×100].
Thymocytes were incubated with biotinylated antibodies against CD3 (500A2) and CD4 (GK1.5) and stimulated by streptavidin cross-linking at 37°C for 5 min. Cells were lyzed and whole cell lysates were analyzed by immunoblotting with antibodies as previously described .
Naïve and regulatory CD4+ T cells were prepared from the lymph nodes (cervical, axillary, brachial and inguinal) and spleen of OT-II TCR-transgenic WT mice or OT-II fyn-/- mice. Cells were prepared as previously described , stained with anti-CD4-APC, anti-CD25-PE and anti-CD62L-FITC and purified using a MoFlo high performance cell sorter (Dakocytomation Fort Collins, CO). For naïve CD4+ T cells, sorting gates were set to include live CD25-CD62LhiCD4+ lymphocytes, whereas for CD4+ Treg, gates were set to include live CD25+CD62LhiCD4+ lymphocytes. Typical purity for both naïve CD4+ T cells and regulatory CD4+ T cells within the range of 89-95% (data not shown).
Naïve CD4+ T cells (5×104cells/well) were co-cultured with regulatory CD4+ T cells (at varying cell density as indicated) with 0.5μg/ml soluble anti-CD3 and γ-irradiated C57BL/6 splenocytes (5×104cells/well) in supplemented RPMI-1640 medium as previously described  in 96-well U-bottom tissue culture plates (Corning Costar, Lowell, MA.). Proliferation was assessed by [methyl-3H]-Thymidine (3H-TdR) (Amersham, Little Chalfont, UK) incorporation, as previously described .
The authors wish to acknowledge the assistance of Shiloh Martin and Barbara Sheer for help in preparing mouse tissues, members of Dr Abbas's laboratory for technical advice and reagents, and Mark Lanett for assistance.
This work was supported by The Joseph Marino Research Award from the Crohn's & Colitis Foundation of America (to A.A.M.) and National Institutes of Health grants PO1 AI-35297 (to A.L.D.). National Health and Medical Research Council (NHMRC) Project Grant 458539 (to W.L. and C.T.) and NHMRC Research Fellowship 572505 (to W.L.).
Conflict of interest: The authors declare no financial or commercial conflict of interest.