IL-7Rα-chain-deficient (IL-7Rα−/−) and common γ chain-deficient (γc−/−) mice both exhibit abnormal thymic and intestinal intraepithelial lymphocyte (IEL) development, but the developmental inhibition is not equivalent. In this report, we assessed whether the defects in T cell development associated with γc−/− mice were due to currently defined γc-dependent cytokines by cross-breeding IL-7Rα−/− mice to mice lacking either IL-2, IL-4, or IL-2Rβ. IL-2/IL-7Rα and IL-4/IL-7Rα double knockout (DKO) mice demonstrated equivalent thymic development to IL-7Rα−/− mice, whereas IL-2Rβ/IL-7Rα DKO mice, which lack IL-2, IL-7, and IL-15 signaling, displayed thymic T cell defects identical to γc−/− mice. Collectively, these data indicate that of the γc-dependent cytokines, only IL-7 and IL-15 contribute to the progression and production of thymic T cells. In the IEL, IL-7Rα−/− mice selectively lack CD8αα TCRγδ cells, whereas IL-2Rβ−/− mice show a significant reduction in all CD8αα cells. IL-2−/− and IL-2/IL-7Rα DKO mice demonstrated a reduction in CD8αα IELs to nearly the same extent as IL-2Rβ−/− mice, indicating that IL-2 functions in CD8αα IEL development. Moreover, IL-2Rβ/IL-7Rα DKO mice lacked nearly all TCR-bearing IEL, again recapitulating the phenotype of γc−/− mice. Thus, these data point to the importance of IL-2, IL-7, and IL-15 as the γc-dependent cytokines essential for IEL development.

The common γ-chain (γc)3 is a shared subunit of the receptor complexes for IL-2, -4, -7, -9, and -15 which consist of γc, a cytokine-specific α-chain, and for IL-2 and IL-15, the shared IL-2Rβ-chain (1). γc−/− mice lack signaling via all five of these cytokines and exhibit severe hypocellularity of both B and T cells and a virtual absence of intraepithelial lymphocytes (IEL) in the intestine, reproductive tract, and other mucosa (2, 3). This phenotype closely mirrors human X-linked severe combined immunodeficiency, which results from mutations in the γc gene (4). By comparison, mice lacking either IL-7, IL-7Rα, or treated with IL-7-specific mAb display similarly reduced B and T cell numbers (5, 6, 7), whereas mice deficient in IL-2, IL-2Rα, IL-2Rβ, IL-4, IL-15Rα, doubly deficient in IL-2 and IL-4, or treated with IL-9-specific mAb have no outward defects in the production of thymus-dependent T cells (8, 9, 10, 11, 12, 13, 14).

Thymic cellularity is dramatically reduced 10- to 25-fold in both IL-7Rα−/− mice and γc−/− mice, whereas the CD4/CD8 thymic subset distribution observed in both of these mice remains essentially the same as in wild-type mice (6). Although many aspects of the thymic developmental defects associated with γc-deficient mice are accounted for by abrogated IL-7/IL-7R signaling, impaired IL-7Rα function does not fully explain the thymic phenotype observed in γc−/− mice. IL-7Rα−/− mice demonstrate an accumulation of triple negative (TN) cells at both the CD44+CD25 stage 1 and the CD44+CD25+ to CD44CD25+ (stages 2–3) transition, whereas γc−/− TN cells are blocked predominantly at the stage 2–3 transition only. Moreover, γc-specific mAb inhibited T cell development in irradiation chimeras reconstituted with IL-7Rα−/− bone marrow precursor cells, implicating one or more γc-dependent cytokines other than IL-7 in conventional thymic T cell development (14, 15).

Regarding extrathymic T cell development, the differences among γc−/− mice, anti-γc mAb-treated irradiation chimeras, and IL-7Rα−/− mice are even more striking and clearly point to a developmental role for at least one other γc-dependent cytokine besides IL-7 (15). The IEL of γc−/− mice are essentially void of all T lymphocytes, whereas IL-7−/− mice and IL-7Rα−/− mice lack only the TCRγδ subset (2, 3, 16, 17). Typically, these thymus-independent TCRγδ cells comprise roughly half the IEL and express CD8α as a homodimer, rather than the CD8αβ coreceptor of thymic-derived CD8 T cells. In normal mice, half of all TCRαβ IEL also express CD8αα and are considered thymus-independent, whereas the remaining IEL express a thymic-derived “conventional” phenotype consisting of TCRαβ CD8αβ and TCRαβ CD4 subsets (18). IL-2Rβ−/− and IL-15Rα−/− mice also display significantly reduced CD8αα IEL of both TCR types (12, 19). Thus, IL-15 appears to play a role in CD8αα IEL development.

In this study we have assessed the potential contribution of γc-dependent cytokines besides IL-7 to conventional thymic and IEL T cell development by cross-breeding either IL-2−/−, IL-2Rβ−/−, or IL-4−/− mice to IL-7Rα−/− mice to obtain multiple double knockout (DKO) strains. Phenotypic analyses of these mice are consistent with an important role for IL-15 in thymic and IEL development. In addition, we have confirmed a role for IL-2, in addition to IL-15, in CD8αα IEL development.

All single gene knockout mice were on the C57BL/6 genetic background and were obtained from The Jackson Laboratories (Bar Harbor, ME), except for γc−/− mice (kindly provided by W. Leonard, National Institutes of Health, Bethesda, MD, and D. Wiest, Fox Chase Cancer Center, Philadelphia, PA) and IL-2Rβ−/− mice (kindly provided by T. Mak, University of Toronto, Ontario, Canada). All mice were maintained under virus-specific Ab free conditions. IL-2/IL-7Rα, IL-2Rβ/IL-7Rα, and IL-4/IL-7Rα DKO mice were generated through the interbreeding of single gene deficient mice for three to five generations to obtain doubly deficient lines. All mice were genotyped by PCR of genomic tail DNA using gene-specific sets of differential primers: one common primer to a nondeleted gene sequence present in both the mutant and wild-type alleles, and an additional primer specific for either an inserted neo sequence found in the mutant allele or a wild-type sequence replaced by recombination in the knockout mouse. PCRs to detect mutant and wild-type alleles were run separately to prevent primer competition. Periodic Southern blot analysis of genomic tail or liver DNA was conducted to confirm PCR genotyping using cDNA (IL-7Rα, IL-2Rβ) or genomic (IL-2, IL-4) probes.

IEL were isolated using a modified procedure described by Poussier and Julius (20). Briefly, small intestine was trimmed of fat and Peyer’s Patches by manual dissection. Fecal contents were flushed with PBS, and longitudinally sectioned 0.5-mm segments were shaken vigorously in 50 ml of RPMI 1640 medium with 100 U/ml penicillin, 100 μg/ml streptomycin, 0.05 mM 2-ME, and 2.5% FCS at 37°C for 30 min to dislodge IELs from the epithelial lining. Supernatants were decanted through gauze and incubated without shaking at 37°C for at least 30 min to kill contaminating epithelial cells. Cell pellets were resuspended in 1.055 g/ml Percoll and layered over 1.085 g/ml Percoll. IEL were collected at the gradient’s interface after spinning 15 min at 1700 rpm and washed three times in HBSS before staining with fluorescent Abs.

Biotinylated-NK1.1 (PK136), FITC-CD4 (GK1.5), FITC-CD25 (3C7), and FITC-CD3 (145-2C11) were prepared in our laboratory. Cychrome-CD8α (53-6.7), cychrome-CD4, cychrome-CD3, FITC-CD8β (53-5.8), biotinylated-CD44 (IM7), biotinylated-TCRαβ (H57-597), biotinylated-TCRγδ (GL3), and PE-streptavidin were purchased from PharMingen (San Diego, CA). Purified mAbs specific for IL-2Rβ (5H4) and IL-7Rα (A7R34; kindly provided by S. Nishikawa, Kyoto University, Kyoto, Japan) were described previously (21, 22). Mouse-adsorbed biotinylated-anti-rat IgG (PharMingen) revealed with PE-streptavidin was used in three-step staining to detect IL-2Rβ and IL-7Rα. All staining was performed for 20 min at 4°C in excess concentrations of Ab. Forward and side scatter axes and propidium iodide exclusion were used to set live lymphocyte gates from whole cell preparations. In some IL-7Rα-deficient mice, we observed an increased fraction of small cells with variable side scatter that partially overlapped the live gate of smaller thymocytes yet did not express CD3, CD4, CD8, c-kit, Thy-1.2, CD5, CD44, or CD25. Because pro/pre-T cells are typically larger than the majority of thymocytes (23), these cells were excluded from the analysis of TN subsets. Between 5 and 10 × 103 gated events were collected per sample on a Becton Dickinson FACScan, and results were quantified using CellQuest software (Becton Dickinson, San Jose, CA). Wild-type C57BL/6 or heterozygous littermates were analyzed for comparison.

The thymic cellularity of adult IL-2/IL-7Rα DKO, IL-4/IL-7Rα DKO, and IL-7Rα−/− mice were all similar (5–10 × 106 cells/mouse; n = 10 mice/group, aged 4–9 wk), with a 10- to 20-fold reduction in thymic cell number. On the other hand, similarly aged IL-2Rβ/IL-7Rα DKO mice (n = 6) and γc−/− mice (n = 3) generally displayed smaller thymi than IL-7Rα−/− mice (<5 × 106 cells/mouse). Moreover, older IL-2Rβ/IL-7Rα DKO mice (n = 2, aged >13 wk) had extremely involuted thymi, which made cell recovery nearly impossible. We have observed this same degree of thymic involution in similarly aged γc−/− mice, whereas complete thymic involution is not observed in IL-7Rα−/− mice until 1–2 mo later (data not shown). These data suggest a role for IL-2Rβ, most likely as a component of the IL-15R, along with IL-7R, but not IL-2 and IL-4, in the production of thymic T cells.

TN thymocytes are divided into four subsets of increasing maturity (23) based on CD44 and CD25 expression. Stage 1 TN cells (CD44+CD25) consist of the earliest defined thymic precursors, with multilymphoid lineage potential, and thymic NK cells. TCRβ, TCRγ, and TCRδ rearrangements commence at TN stage 2 (CD44+CD25+), and both pre-TCRα and TCRβ are expressed on the cell surface at TN stage 3 (CD44CD25+). TN stage 4 (CD44CD25) is the last point before up-regulation of CD4, CD8, and TCRα rearrangements.

Despite the diminished number of thymocytes, the pattern of CD4/CD8 thymic subsets was usually comparable in all the DKO mice tested, including the percentage of TN cells. The pattern of TN subsets, however, in IL-7Rα−/− mice and γc−/− mice typically differed from each other and from normal mice (Fig. 1, Ref. 14). The TN subset distribution of IL-2/IL-7Rα DKO and IL-4/IL-7Rα DKO mice was largely comparable to that seen in IL-7Rα−/− mice (Fig. 1), although CD25 expression was somewhat more diffuse in the DKO mice. All these patterns are characterized by an increased proportion of cells in stage 1 and the stage 2–3 transition. Thus, these data further indicate that IL-2 and IL-4 play no obvious role in thymic T cell development. On the other hand, IL-2Rβ/IL-7Rα DKO mice exhibited a pattern of TN subsets similar to γc−/− mice, with a characteristic block at the stage 2–3 transition and a decrease in stage 1 cells. A similar decrease in stage 1 pro-T cells was also observed in young IL-2Rβ−/− mice of normal thymic cellularity (Fig. 1). Thus, IL-2Rβ/IL-7Rα DKO mice recapitulate the thymic developmental defects of γc−/− mice.

FIGURE 1.

Flow cytometric analysis of CD44/CD25 distribution on thymic TN cells from single and DKO strains. TN cells were analyzed by gating on cychrome-negative thymocytes after staining with cychrome-CD3, cychrome-CD4, cychrome-CD8α, FITC-CD25, and biotinylated-CD44 revealed with PE-streptavidin. A total of 104 cychrome-negative gated events were collected to stage TN cells based on CD44 and CD25 expression. Numbers represent the percentage of total triple negative cells in each quadrant. Data shown are representative of 5–15 mice/group aged 4–9 wk, except for γc−/− mice for which n = 3.

FIGURE 1.

Flow cytometric analysis of CD44/CD25 distribution on thymic TN cells from single and DKO strains. TN cells were analyzed by gating on cychrome-negative thymocytes after staining with cychrome-CD3, cychrome-CD4, cychrome-CD8α, FITC-CD25, and biotinylated-CD44 revealed with PE-streptavidin. A total of 104 cychrome-negative gated events were collected to stage TN cells based on CD44 and CD25 expression. Numbers represent the percentage of total triple negative cells in each quadrant. Data shown are representative of 5–15 mice/group aged 4–9 wk, except for γc−/− mice for which n = 3.

Close modal

Because stage 1 TN cells consist of thymic precursor cells and thymic NK cells, we further characterized the lineage specificity of the stage 1 TN cells that accumulate in IL-7Rα−/− mice. NK1.1 expression was detected on ∼50% of stage 1 cells in wild-type mice, whereas this fraction increased to 75% in IL-7Rα−/− mice. In contrast, the expression of c-kit, a marker of T lineage precursors, was comparable within individual TN cell subsets from IL-7Rα−/− mice and normal mice (data not shown), with a higher fraction of c-kit+ cells at stages 2 and 3 (∼50% of total cells) than at stage 1 or 4 (∼10% each). Thus, a portion of the increase in stage 1 TN cells is due to a higher fraction of thymic NK cells. The expression of both NK1.1 and c-kit in TN cells of IL-2/IL-7Rα DKO and IL-4/IL-7Rα DKO mice were similar to that of IL-7Rα−/− mice (data not shown).

Unexpectedly, weak NK1.1 staining was detected throughout stages 2 and 3 and, to a lesser extent, on stage 4 cells of IL-7Rα−/− mice (Fig. 2,A), but was undetectable on CD4+CD8+ thymocytes. In general, the level of staining at stages 2 and 3 was to 2- to 4-fold higher than control staining in normal mice which exhibited no specific staining for NK1.1 after TN stage 1 (Fig. 2,B). This staining of IL-7Rα−/− TN cells was further judged to be specific by the capacity of unlabeled anti-NK1.1 to inhibit staining by biotinylated-NK1.1/PE-streptavidin (data not shown). This result raises the possibility that normal down-regulation of NK1.1 on developing T cells requires IL-7R signaling. One IL-7Rα−/− mouse, however, with a particularly large thymus (31.2 × 106 cells) demonstrated no NK1.1 staining beyond TN stage 1 (Fig. 2 B). This finding suggests that when there is reasonable compensation for a loss of IL-7R function in vivo, NK1.1 expression on pro-T cells undergoes normal regulation.

FIGURE 2.

Increased expression of NK1.1 on thymic TN cell subsets of IL-7Rα−/− mice vs normal mice. A, TN cells from a 7-wk-old C57BL/6 (dark line) mouse and an age-matched IL-7Rα−/− (shaded region) mouse were gated as in Fig. 1, and staged as indicated following staining with FITC-CD44, FITC-CD25, or both, and biotinylated-anti-NK1.1 mAb revealed with PE-streptavidin. NK1.1 expression on TN cell stages was analyzed by gating on FITC-positive or -negative cells as appropriate. Dotted lines indicate background fluorescence of wild-type cells stained with PE-streptavidin only without primary mAb; background fluorescence of IL-7Rα−/− cells was similar. B, Mean fluorescence intensity (MFI) of biotinylated-anti-NK1.1 staining revealed with PE-streptavidin is displayed for TN subsets of six IL-7Rα−/− mice as the MFI ratio of each IL-7Rα−/− mouse vs a normal (+/+ or +/−) mouse analyzed in the same experiment. Background staining was not subtracted from the experimental samples.

FIGURE 2.

Increased expression of NK1.1 on thymic TN cell subsets of IL-7Rα−/− mice vs normal mice. A, TN cells from a 7-wk-old C57BL/6 (dark line) mouse and an age-matched IL-7Rα−/− (shaded region) mouse were gated as in Fig. 1, and staged as indicated following staining with FITC-CD44, FITC-CD25, or both, and biotinylated-anti-NK1.1 mAb revealed with PE-streptavidin. NK1.1 expression on TN cell stages was analyzed by gating on FITC-positive or -negative cells as appropriate. Dotted lines indicate background fluorescence of wild-type cells stained with PE-streptavidin only without primary mAb; background fluorescence of IL-7Rα−/− cells was similar. B, Mean fluorescence intensity (MFI) of biotinylated-anti-NK1.1 staining revealed with PE-streptavidin is displayed for TN subsets of six IL-7Rα−/− mice as the MFI ratio of each IL-7Rα−/− mouse vs a normal (+/+ or +/−) mouse analyzed in the same experiment. Background staining was not subtracted from the experimental samples.

Close modal

To assess the potential points in TN cell progression dependent upon IL-7 and IL-2Rβ, IL-7Rα and IL-2Rβ expression was measured in pro-T cell subsets of normal C57BL/6 mice. Both these receptor subunits were expressed at high levels on stage 1 TN cells, whereas only IL-7Rα was detected at all TN stages in young mice (Fig. 3). In older mice (>15 wk), expression of IL-7Rα in stages 3 and 4 pre-T cells was usually lower, but still detectable (data not shown). Thus, this pattern of expression is consistent with IL-2Rβ functioning during thymic development at or before stage 1, whereas IL-7Rα may potentially function at all points of TN development.

FIGURE 3.

Expression of IL-7Rα and IL-2Rβ subunits on thymic TN cell subsets of normal mice. TN cells from C57BL/6 mice were analyzed as in Fig. 2,A and subsequently stained with either anti-IL-2Rβ (A) or anti-IL-7Rα (B) mAbs revealed with biotinylated-anti-rat IgG and PE-streptavidin. Receptor subunit expression on TN cell stages was analyzed by gating on FITC-positive or -negative cells as in Fig. 2 A. Solid lines represent specific fluorescence, whereas the dotted lines represent background fluorescence. Data shown are representative of three mice aged 6.5 wk.

FIGURE 3.

Expression of IL-7Rα and IL-2Rβ subunits on thymic TN cell subsets of normal mice. TN cells from C57BL/6 mice were analyzed as in Fig. 2,A and subsequently stained with either anti-IL-2Rβ (A) or anti-IL-7Rα (B) mAbs revealed with biotinylated-anti-rat IgG and PE-streptavidin. Receptor subunit expression on TN cell stages was analyzed by gating on FITC-positive or -negative cells as in Fig. 2 A. Solid lines represent specific fluorescence, whereas the dotted lines represent background fluorescence. Data shown are representative of three mice aged 6.5 wk.

Close modal

As expected based on the phenotype of IL-7Rα−/− and IL-2Rβ−/− mice (6, 10), all DKO animals exhibited a substantially reduced number of splenic B cells, and IL-2Rβ/IL-7Rα DKO mice essentially lacked splenic NK cells (data not shown). However, when compared with IL-7Rα−/− mice, several differences in splenic T cell cellularity were noted (Fig. 4). Both IL-2/IL7Rα and IL-2Rβ/IL-7Rα DKO mice had increased numbers of CD4 T lymphocytes and to a lesser extent CD8 T cells. This result is similar to that reported for older (>4 wk of age) γc−/− mice (2, 3) and suggests that this increase is due to impaired T cell development resulting from a lack of IL-7 signaling and abnormal peripheral CD4 T cell homeostasis due to lack of IL-2 function. Furthermore, IL-4/IL-7Rα DKO mice had a somewhat lower number of splenic CD8 T cells. The reason for this result is currently unknown but might reflect a role for these cytokines in the survival of CD8 T cells (24).

FIGURE 4.

Splenic T lymphocyte cellularity of DKO strains. Whole splenocytes were counted by trypan blue exclusion and stained with mAb to either CD4 (A) or CD8 (B) to determine cellularity of peripheral T cell subsets. Lines represent mean cellularity of each group of mice aged 4–14 wk.

FIGURE 4.

Splenic T lymphocyte cellularity of DKO strains. Whole splenocytes were counted by trypan blue exclusion and stained with mAb to either CD4 (A) or CD8 (B) to determine cellularity of peripheral T cell subsets. Lines represent mean cellularity of each group of mice aged 4–14 wk.

Close modal

IEL cell recovery was consistently reduced in γc−/−, IL-2Rβ−/−, and IL-2Rβ/IL-7Rα DKO mice (Fig. 5), although group variance in IEL cellularity was high in all strains analyzed. Perhaps most striking, the analysis of IL-2Rβ/IL-7Rα DKO mice revealed an IEL phenotype which recapitulated that of γc−/− mice, i.e., a virtual lack of TCR+ IEL (Fig. 5). The few viable cells isolated from the gut preparations of either of these two strains failed to express either TCRαβ, TCRγδ, or CD8α, and most likely represent gut epithelial cell contaminants (data not shown). Thus, the loss of IL-2Rβ and IL-7Rα signaling abrogated IEL development.

FIGURE 5.

IEL cellularity of single and DKO strains. Total IEL isolated from full-length small intestine were counted by trypan blue exclusion and stained with mAb to TCRαβ or TCRγδ to determine absolute numbers of TCR+ IEL by flow cytometric analysis. The portion of TCR cells in each strain was considered to be contaminating epithelial cells. Error bars represent the SD of each TCR+ IEL subset for five mice/group aged 4–9 wk, except for γc−/− mice for which n = 3.

FIGURE 5.

IEL cellularity of single and DKO strains. Total IEL isolated from full-length small intestine were counted by trypan blue exclusion and stained with mAb to TCRαβ or TCRγδ to determine absolute numbers of TCR+ IEL by flow cytometric analysis. The portion of TCR cells in each strain was considered to be contaminating epithelial cells. Error bars represent the SD of each TCR+ IEL subset for five mice/group aged 4–9 wk, except for γc−/− mice for which n = 3.

Close modal

We analyzed the IEL phenotype in more detail for the remaining strains of mice which contained significant numbers of T cells. The IEL subset distribution of IL-4−/− and IL-4/IL-7Rα DKO mice was largely comparable to wild-type C57BL/6 and IL-7Rα−/− mice, respectively, demonstrating that IL-4 does not play an obvious role in the regulation of IEL development (Fig. 5 and Table I). As expected, the number of nonconventional TCRγδ CD8αα IEL was reduced in all mouse strains lacking IL-7R signaling, as well as in IL-2Rβ−/− mice. However, IL-2−/− mice also demonstrated a marked reduction in TCRγδ CD8αα IEL to nearly the same level as IL-2Rβ−/− mice (Figs. 5 and 6,A). Moreover, both IL-2−/− and IL-2/IL-7Rα DKO mice displayed a relative decrease in nonconventional TCRαβ CD8αα cells as compared with wild-type C57BL/6 and IL-7Rα−/− mice (Fig. 6,B and Table I). These data, therefore, indicate a role for IL-2 in the development of CD8αα IEL.

Table I.

Distribution of conventional vs nonconventional TCRαβ IEL in single and double gene knock-out strains

Mouse StrainTotal TCRαβ IEL (× 106)% of Total TCRαβ IELa
CD8αβCD8ααCD4
C57BL/6 1.8 ± 1 32.9 ± 8 41.1 ± 14.1 19.6 ± 8.5 
IL-4−/− 2.5 ± 1.7 39.4 ± 11.9 27.6 ± 2.9 27.0 ± 10.7 
IL-7Rα−/− 3.0 ± 1.2 45.9 ± 16 37.4 ± 13.3 11.8 ± 5.9 
IL-4/IL-7Rα DKO 2.2 ± 1.9 43.5 ± 3.3 44.8 ± 5.9 9.8 ± 3.9 
IL-2Rβ−/− 1.6 ± 1 47.1 ± 10.5 4.5 ± 1.7 48.3 ± 7.7 
IL-2−/− 2.6 ± 2.5 55.5 ± 9.1 10.3 ± 4 32.7 ± 12.3 
IL-2/IL-7Rα DKO 4.6 ± 2.6 60.3 ± 6.7 22.1 ± 7.4 16.3 ± 4.3 
Mouse StrainTotal TCRαβ IEL (× 106)% of Total TCRαβ IELa
CD8αβCD8ααCD4
C57BL/6 1.8 ± 1 32.9 ± 8 41.1 ± 14.1 19.6 ± 8.5 
IL-4−/− 2.5 ± 1.7 39.4 ± 11.9 27.6 ± 2.9 27.0 ± 10.7 
IL-7Rα−/− 3.0 ± 1.2 45.9 ± 16 37.4 ± 13.3 11.8 ± 5.9 
IL-4/IL-7Rα DKO 2.2 ± 1.9 43.5 ± 3.3 44.8 ± 5.9 9.8 ± 3.9 
IL-2Rβ−/− 1.6 ± 1 47.1 ± 10.5 4.5 ± 1.7 48.3 ± 7.7 
IL-2−/− 2.6 ± 2.5 55.5 ± 9.1 10.3 ± 4 32.7 ± 12.3 
IL-2/IL-7Rα DKO 4.6 ± 2.6 60.3 ± 6.7 22.1 ± 7.4 16.3 ± 4.3 
a

Mean cell number or percentage ±SD are shown as determined by flow cytometric analysis of 5–10 mice/group aged 4–9 wk.

FIGURE 6.

Flow cytometric analysis of nonconventional and conventional IEL subset distribution in mice lacking signaling via IL-2, IL-7, and/or IL-2Rβ. A, Whole IEL were stained with mAb to TCRγδ and CD8α to identify decreases in thymus-independent, nonconventional IEL, as compared with wild-type mice. B, Whole IEL were stained with mAb to TCRαβ, CD8α, and CD8β. Viable TCR+ cells were gated and analyzed for nonconventional CD8αα vs conventional CD8αβ subset distribution. Numbers in both panels are the percentages of cells in the indicated quadrants. Data shown are representative of 5–10 mice/group aged 4–9 wk.

FIGURE 6.

Flow cytometric analysis of nonconventional and conventional IEL subset distribution in mice lacking signaling via IL-2, IL-7, and/or IL-2Rβ. A, Whole IEL were stained with mAb to TCRγδ and CD8α to identify decreases in thymus-independent, nonconventional IEL, as compared with wild-type mice. B, Whole IEL were stained with mAb to TCRαβ, CD8α, and CD8β. Viable TCR+ cells were gated and analyzed for nonconventional CD8αα vs conventional CD8αβ subset distribution. Numbers in both panels are the percentages of cells in the indicated quadrants. Data shown are representative of 5–10 mice/group aged 4–9 wk.

Close modal

With respect to thymic-dependent T cell development, we found that IL-2Rβ/IL-7Rα DKO mice recapitulated the phenotype associated with γc−/− mice, demonstrating that impaired T cell development is apparently accounted for by known γc-dependent cytokines. It is highly likely that the dependency on IL-2Rβ-signaling is due to IL-15 rather than IL-2, as IL-2/IL-7Rα DKO mice did not display any thymic defects beyond those associated with IL-7Rα-deficiency.

IL-7 appears to contribute functionally at all stages of TN thymocyte development by providing anti-apoptotic signals (25) and promoting the development of TCRγδ cells (16, 17). Support for this conclusion comes in part from the expression pattern of IL-7Rα, as discussed above, and from the TN cell phenotype of IL-7Rα−/− mice. The relative accumulation of TN cells at the stage 2–3 transition, when TCR rearrangement commences, suggests a role for IL-7 at this point in the developmental scheme. Furthermore, the capacity of γc-specific mAbs to block T cell development from stage 3 and stage 4 pre-T cells upon intrathymic injection (14) most likely reflects a role for IL-7 and not IL-15 at these stages, as IL-2Rβ is not detected on these cells. IL-7Rα may also function at stage 1 to down-regulate NK1.1, as we often observed detectable levels of this protein on stage 2 and 3 TN cells in IL-7Rα−/− single and DKO mice. As in wild-type mice, a relatively high proportion (∼50%) of stage 2/3 TN cells in IL-7Rα−/− mice expressed c-kit, a marker of lymphoid precursors cells, suggesting that this finding represented a true impairment in the down-regulation of NK1.1, rather than committed thymic NK cells with an abnormal phenotype.

IL-15 most likely functions at the earliest points in T cell development, as IL-2Rβ expression was detected only on stage 1 pro-T cells and others have demonstrated expression of this receptor on fetal stem cells (26). The production of many of the TN cells bearing a stage 1 (CD44+CD25) phenotype appears to be dependent on IL-15, as these cells are reduced in IL-2Rβ−/− mice, as well as in γc−/− mice. One role for IL-15 at this early point in T lineage development is in the production of thymic-dependent NK cells (19, 27). The relative accumulation of stage 1 TN cells in IL-7Rα−/− mice is accounted for, at least in part, by functional IL-15 signaling via IL-2Rβ. This increased proportion of stage 1 TN cells primarily represents an increase in NK1.1+ cells. Multilineage precursors have been identified during fetal development which express both NK1.1 and c-kit (28), and as we did not detect an increase in c-kit+ stage 1 TN cells in IL-7Rα−/− mice, it is likely that these TN cells represent thymic NK cells.

A second role for IL-15 is to contribute to the mainstream production of thymic T cells. Two observations support this conclusion. In this study, we consistently detected fewer thymocytes in IL-2β/IL-7Rα DKO mice than in IL-7Rα-deficient mice. We also previously showed that γc-specific mAbs inhibited T cell development in irradiation chimeras reconstituted with bone marrow precursor cells from IL-7Rα−/− mice (14). This latter effect most likely is due to a blockade of IL-15R signaling, because this study demonstrated no role for IL-2 or IL-4 in the production of thymic T cells, and we have previously shown no role for IL-9 (14). The function of IL-15 in thymic T cell development may largely overlap with that of IL-7, because thymocyte number and CD4/CD8 subset distribution is essentially normal in IL-15Rα−/− mice and young IL-2Rβ−/− mice (10, 12). The IL-7R and IL-2Rβ are both expressed early in lymphoid development, as IL-7Rα is expressed on bone marrow lymphoid progenitor cells (29) and, as shown herein, on all subsets of TN cells, permitting a potential overlap in function.

With respect to the role of γc-dependent cytokines in IEL development, two important observations were made. First, our data clearly demonstrate a role for IL-2 in extrathymic T cell development. There was a clear decrease in the proportion of both TCRγδ CD8αα and TCRαβ CD8αα IEL in mice lacking IL-2, while conventional TCRαβ CD8αβ IEL were largely unaffected in mice lacking either IL-2 or IL-15 signaling. Although reductions in TCRγδ and CD8αα IEL have been reported for older IL-2−/− mice (30, 31), these studies only examined colonic T lymphocytes in mice with active autoimmune colitis, rather than the IEL from the small intestine of young, healthy IL-2−/− mice, as in the current study. Our data regarding a role for IL-2 in IEL development are consistent with an additional report which examined the IEL of gnotobiotic IL-2−/− mice (32). These findings seem to contradict a conclusion drawn from IL-2/IL-4 DKO mice which had been reported as having normal numbers of TCRγδ and CD8αα IEL (13). However, this limited analysis of IL-2/IL-4 DKO IEL did not distinguish between conventional and nonconventional IEL subsets, and the low number of TCRγδ and CD8αα IEL reported as “normal” in these mice (16%) may also be interpreted as a decrease in nonconventional T cells, as the IEL of wild-type mice are typically composed of 30–60% TCRγδ cells.

Defects in CD8αα IEL development are obvious in both IL-2−/− mice and IL-15Rα−/− mice, but are most pronounced in IL-2Rβ-deficient animals. Thus, IL-2 and IL-15 appear to function in redundant and overlapping fashion for the development of CD8αα lineage IEL. The contribution of IL-2 may be somewhat dominant over IL-15, as the developmental defect we observed in IL-2−/− mice was somewhat greater than that reported for IL-15Rα−/− mice (12). In any case, there are likely two separate but as yet undefined steps in CD8αα IEL development, wherein IL-2 likely predominates in one and IL-15 in the other, to account for these results. Furthermore, the presence of normal to increased numbers of TCRαβ CD8αα IEL in IL-7Rα−/− mice implies that commitment to CD8αα differentiation occurs before TCR designation. Thus, in the IEL the function of IL-7/IL-7R signaling appears to be limited to the differentiation of TCRγδ cells.

A second striking observation noted in the IEL was the capacity of IL-2Rβ/IL-7Rα DKO mice to recapitulate the T cell defects of γc−/− mice, as was also shown in the thymus. The virtual absence of all IEL subsets in IL-2Rβ/IL-7Rα DKO mice was somewhat remarkable, considering that conventional phenotypic TCRαβ CD8αβ and TCRαβ CD4 IEL are present in normal numbers in IL-2-, IL-2Rβ-, IL-7Rα-, and IL-15Rα-deficient mice (12, 17, 19). Thus, the simplest interpretation of this finding is that an early lymphoid precursor redundantly requires signaling via IL-7Rα or IL-2Rβ to give rise to IEL of conventional phenotype. It is difficult to rule out a role for IL-2 in the IEL at an early stage because IL-2, like IL-15, clearly contributes to IEL development, as discussed above. In addition, IEL of conventional phenotype were also selectively unaffected in anti-γc mAb-treated chimeric mice, suggesting the importance of a cytokine whose activity was poorly blocked by these Abs (15). In this regard, the γc-specific mAbs used in this study were shown to be much more potent inhibitors of IL-7 and IL-15, than IL-2. Ultimately, the determination of whether these precursor cells express functional IL-2R and/or IL-15R should help resolve this issue.

We thank Paul Scibelli for his assistance in Southern blot genotyping of DKO mice.

1

This work was supported by National Institutes of Health Grant CA45957.

3

Abbreviations used in this paper: γc, common γ chain; IEL, intraepithelial lymphocytes; DKO, double knockout; TN, triple negative.

1
He, Y.-W., T. Malek.
1998
. The structure and function of γc-dependent cytokines and receptors: regulation of T lymphocyte development and homeostasis.
Crit. Rev. Immunol.
18
:
503
2
Cao, X., E. Shores, J. Hu-Li, M. Anver, B. Kelsall, S. Russell, J. Drago, M. Noguchi, A. Grinberg, E. Bloom, W. Paul, et al
1995
. Defective lymphoid development in mice lacking expression of the common cytokine receptor γ chain.
Immunity.
2
:
223
3
DiSanto, J., W. Müller, D. Guy-Grand, A. Fischer, K. Rajewsky.
1995
. Lymphoid development in mice with a targeted deletion of the interleukin 2 receptor γ chain.
Proc. Natl. Acad. Sci. USA
92
:
377
4
Uribe, L., K. Weinburg.
1998
. X-linked SCID and other defects of cytokine pathways.
Sem. Hematol.
35
:
299
5
von Freeden-Jeffry, U., P. Vieira, L. Lucian, T. McNeil, S. Burdach, R. Murray.
1995
. Lymphopenia in interleukin (IL)-7 gene-deleted mice identifies IL-7 as a nonredundant cytokine.
J. Exp. Med.
181
:
1519
6
Peschon, J., P. Morrisey, K. Grabstein, F. Ramsdell, E. Maraskovsky, B. Gliniak, L. Park, S. Ziegler, D. Williams, C. Ware, et al
1994
. Early lymphocyte expansion is severely impaired in interleukin 7 receptor-deficient mice.
J. Exp. Med.
180
:
1955
7
Grabstein, K., T. Waldschmidt, F. Finkelman, B. Hess, A. Alpert, N. Boiani, A. Namen, P. Morrissey.
1993
. Inhibition of murine B and T lymphopoiesis in vivo by an anti-interleukin 7 monoclonal antibody.
J. Exp. Med.
178
:
257
8
Schorle, H., T. Holtschke, T. Hünig, A. Schimpl, I. Horak.
1991
. Development and function of T cells in mice rendered interleukin-2 deficient by gene targeting.
Nature
352
:
621
9
Willerford, D., J. Chen, J. Ferry, L. Davidson, A. Ma, F. Alt.
1995
. Interleukin-2 receptor α chain regulates the size and content of the peripheral lymphoid compartment.
Immunity
3
:
521
10
Suzuki, H., T. Kündig, C. Furlonger, A. Wakeham, E. Timms, T. Matsuyama, R. Schmits, J. Simgard, P. Ohashi, H. Griesser, et al
1995
. Deregulated T cell activation and autoimmunity in mice lacking interleukin-2 receptor β.
Science
268
:
1472
11
Kühn, R., K. Rajewsky, W. Müller.
1991
. Generation and analysis of interleukin-4 deficient mice.
Science
254
:
707
12
Lodolce, J., D. Boone, S. Chai, R. Swain, T. Dassopoulos, S. Trettin, A. Ma.
1998
. IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation.
Immunity
9
:
669
13
Sadlack, B., R. Kühn, H. Schorle, K. Rajewsky, W. Müller, I. Horak.
1994
. Development and proliferation of lymphocytes in mice deficient for both interleukins-2 and -4.
Eur. J. Immunol.
24
:
281
14
He, Y.-W., H. Nakajima, W. Leonard, B. Adkins, T. Malek.
1997
. The common γ-chain of cytokine receptors regulates intrathymic T cell development at multiple stages.
J. Immunol.
158
:
2592
15
Malek, T., R. Levy, B. Adkins, Y.-W. He.
1998
. Monoclonal antibodies to the common γ-chain as cytokine receptor antagonists in vivo: effect on intrathymic and intestinal intraepithelial T lymphocyte development.
J. Leukocyte Biol.
63
:
643
16
Moore, T., U. von Freeden-Jeffry, R. Murray, A. Zlotnik.
1996
. Inhibition of γδ T cell development and early thymocyte maturation in IL-7−/− mice.
J. Immunol.
157
:
2366
17
He., Y.-W., T. Malek.
1996
. Interleukin-7 receptor α is essential for the development of γδ+ T cells, but not natural killer cells.
J. Exp. Med.
184
:
289
18
Rocha, B., D. Guy-Grand, P. Vassalli.
1995
. Extrathymic T cell differentiation.
Curr. Opin. Immunol.
7
:
235
19
Suzuki, H., G. Duncan, H. Takimoto, T. Mak.
1997
. Abnormal development of intestinal intraepithelial lymphocytes and peripheral natural killer cells in mice lacking the IL-2 receptor β chain.
J. Exp. Med.
185
:
499
20
Poussier, P., M. Julius.
1997
. Preparation of mononuclear cells from the gut-associated immune system. I. Lefkovits, ed.
Immunology Methods Manual
1509
Academic Press, London.
21
Malek, T., R. Furse, M. Fleming, A. Fadell, Y.-W. He.
1995
. Biochemical identity and characterization of the mouse interleukin-2 receptor β and γc subunits.
J. Interferon Cytokine Res.
15
:
447
22
Sudo, T., N. Satomi, N. Ohno, N. Akiyama, M. Tamakoshi, H. Yoshida, S. Nishikawa.
1993
. Expression and function of the interleukin 7 receptor in murine lymphocytes.
Proc. Natl. Acad. Sci. USA
90
:
9125
23
Godfrey, D., J. Kennedy, T. Suda, A. Zlotnik.
1993
. A developmental pathway involving four phenotypically and functionally distinct subsets of CD3CD4CD8 triple-negative adult mouse thymocytes defined by CD44 and CD25 expression.
J. Immunol.
150
:
4244
24
Vella, A., T. Teague, J. Ihle, J. Kappler, P. Marrack.
1997
. Interleukin 4 (IL-4) or IL-7 prevents the death of resting T cells: Stat6 is probably not required for the effect of IL-4.
J. Exp. Med.
186
:
325
25
von Freeden-Jeffry, U., N. Solvason, M. Howard, R. Murray.
1997
. The earliest T lineage-committed cells depend on IL-7 for Bcl-2 expression and normal cell cycle progression.
Immunity
7
:
147
26
Reya, T., J. Yang-Snyder, E. Rothenberg, S. Carding.
1996
. Regulated expression and function of CD122 (interleukin-2/interleukin-15R-β) during lymphoid development.
Blood
87
:
190
27
Moore, T., A. Zlotnik.
1995
. T-cell lineage commitment and cytokine responses of thymic progenitors.
Blood
86
:
1850
28
Carlyle, J., A. Michie, C. Furlonger, T. Nakano, M. Lenardo, C. Paige, J. Zúñiga-Pflüker.
1997
. Identification of a novel developmental stage marking lineage commitment of progenitor thymocytes.
J. Exp. Med.
186
:
173
29
Kondo, M., I. Weissman, K. Akashi.
1997
. Identification of clonogenic common lymphoid progenitors in mouse bone marrow.
Cell.
91
:
661
30
Sadlack, B., H. Merz, H. Schorle, A. Schimpl, A. Feller, I. Horak.
1993
. Ulcerative colitis-like disease in mice with a disrupted interleukin-2 gene.
Cell
75
:
253
31
Simpson, S., E. Mizoguchi, D. Allen, A. Bhan, C. Terhorst.
1995
. Evidence that CD4+, but not CD8+ T cells are responsible for murine interleukin-2-deficient colitis.
Eur. J. Immunol.
25
:
2618
32
Contractor, N., H. Bassiri, T. Reya, A. Park, D. Baumgart, M. Wasik, S. Emerson, S. Carding.
1998
. Lymhpoid hyperplasia, autoimmunity, and compromised intestinal intraepithelial lymphocyte development in colitis-free gnotobiotic IL-2-defcient mice.
J. Immunol.
160
:
385