The Hedgehog family of secreted intercellular signaling molecules are regulators of patterning and organogenesis during animal development. In this study we provide genetic evidence that Sonic Hedgehog (Shh) has a role in the control of murine T cell development. Analysis of Shh−/− mouse embryos revealed that Shh regulates fetal thymus cellularity and thymocyte differentiation. Shh is necessary for expansion of CD4−CD8− double-negative (DN) thymocytes and for efficient transition from the earliest CD44+CD25− DN population to the subsequent CD44+CD25+ DN population and from DN to CD4+CD8+ double-positive cells.
The thymus, the primary site of T cell production, is formed during fetal development by seeding of the thymic primordium by blood-borne lymphocyte progenitor cells from the fetal liver (1). The development of mature T cells is dependent on interactions between the developing lymphocyte precursors and the thymic stroma, made up of thymic epithelial cells and mesenchyme-derived cells (2). The size of the thymus and the production of T cells are tightly controlled, and thymus size is regulated by factors within the thymus itself (3). Thymus size and the production of mature T cells do not seem to be limited by the constraints of positive selection of the T cell repertoire, so presumably a counting system controlling thymocyte numbers relies on competition for growth factors or intercellular signaling molecules that control thymocyte fate. Many cytokines and their receptors are expressed in the thymus, but mutants in these molecules have often failed to reveal nonredundant roles in T cell development (4). Mutants, however, in IL-7 (5), its receptor components, IL-7Rα (6, 7), the common cytokine receptor γ-chain (γc) 4 (7, 8), and the receptor tyrosine kinase c-Kit (CD117) (9) have reduced thymic cellularity, and mutants deficient in both c-Kit and γc have a complete arrest in thymocyte development (10). In this study we show that mutants of the secreted signaling molecule Sonic Hedgehog (Shh) have reduced thymus cellularity and reduced thymocyte differentiation, demonstrating a role for Shh in the control of normal T cell development.
Thymocyte development has been defined by cell surface expression of developmentally regulated markers: the most immature CD4−CD8− double-negative (DN) thymocytes give rise to CD4+CD8+ double-positive (DP) thymocytes, which give rise to mature CD4+CD8− single-positive (SP) and CD4−CD8+ SP T cells. The DN population can be further subdivided by the expression of CD44 and CD25: CD44+CD25− (DN1) cells differentiate into CD44+CD25+ (DN2) cells, which give rise to CD44−CD25+ (DN3) cells, which finally become the most mature CD44−CD25− (DN4) DN population. The DN4 cells may pass through an intermediate population expressing either coreceptor alone before becoming DP cells. This intermediate population, most commonly expressing CD8, is known as immature single positive (ISP) cells. Progression beyond the DN3 stage is dependent on successful rearrangement of a TCRβ-chain gene and pre-TCR signaling, whereas differentiation from DP to mature SP cell is dependent on the expression and positive selection of an αβTCR (11, 12).
The most immature DN1 cells express CD45 and CD117, but are not fully committed to the T cell lineage. The DN1 population contains cells that can give rise to T, B, NK, and dendritic cells (1, 13, 14). As thymocytes proceed along their program of differentiation they become progressively more committed to the T cell lineage, and DN3 cells that have TCRβ VDJ arrangements are irreversibly committed (12).
The Hedgehog (Hh) family of secreted intercellular signaling molecules are powerful regulators of patterning and organogenesis during animal development (15). Recent studies have revealed that these proteins are also active in self-renewing tissues in the adult, such as the immune system and the hemopoietic system (16, 17). They can function as classical morphogens, inducing distinct dose-dependent cell fates within a target field, and can also regulate cell survival and proliferation (15, 18, 19). There are three mammalian Hh proteins: Sonic Hedgehog (Shh), Indian Hedgehog (Ihh), and Desert Hedgehog (Dhh) (15). Shh is involved in many developmental processes, including left-right asymmetry, and patterning of the brain, spinal cord, and limbs. Shh-deficient embryos are small and die before birth of numerous developmental defects, but Shh+/− mice are fertile and appear normal (20). Both Shh and Ihh are expressed in the mouse thymus. Shh is produced by the thymic epithelium, and Ihh expression is mainly associated with blood vessels in the thymic medulla (16).
Hh proteins share a common, apparently exclusive, signaling pathway. They signal to neighboring cells by binding to their cell surface receptor Patched (Ptc) (21, 22). The cell surface molecule Smoothened (Smo) then transmits the Hh signal into the target cell (23, 24). In the absence of its Hh ligand, Ptc inhibits the ability of Smo to signal (25, 26), but when Hh binds to Ptc, Ptc releases Smo to signal into the cell. At the end of the Hh signaling pathway are members of the Gli family of transcription factors, Gli1, Gli2, and Gli3.
Both Ptc and Smo are expressed by DN thymocytes. The earliest CD44+CD25− DN1 cells that are not committed to the T cell lineage do not express high levels of cell surface Smo. In the next compartment of CD44+CD25+ DN2 cells, Smo expression is up-regulated, whereas in the subsequent DN3 and DN4 populations, both the proportions of positive cells and the levels of expression gradually decline (16).
Our in vitro studies have indicated that Hh signaling is involved in the regulation of differentiation from DN to DP cell. Treatment of fetal thymus organ cultures (FTOC) with recombinant Hh protein arrested thymocyte development at the DN stage, after TCRβ-chain gene rearrangement, whereas treatment of FTOC with a neutralizing anti-Hh Ab accelerated thymocyte differentiation to the DP stage, but did not replace the need for a pre-TCR signal. We showed that pre-TCR signaling down-regulates Smo, demonstrating a direct link between pre-TCR signaling and Hh signaling in the regulation of differentiation of DN thymocytes (16).
These in vitro experiments provided evidence of the effects on thymocyte differentiation of modulating the Hh signal at the end of the period of Smo expression (and presumably active Hh signaling). They did not distinguish between the biological roles of Ihh and Shh or identify earlier actions of Hh signaling on developing thymocytes. Therefore, to provide genetic evidence of a role for Shh in the control of T cell development and to assess its function earlier in thymocyte development, we studied fetal thymocyte development in Shh−/− embryos. In this study we show that thymocyte differentiation was reduced at the transition from DN1 to DN2 cells and at the transition from DN to DP cells in Shh−/− embryos. In addition, thymocyte numbers were greatly reduced at all developmental stages.
Materials and Methods
C57BL/6 and BALB/c mice were purchased from B & K Universal (Sollentuna, Sweden). Shh+/− mice were a gift from Prof. P. Beachy (20). Shh+/− mice were back-crossed onto C57BL/6 mice for at least five generations. Shh+/− embryos and adults were genotyped by PCR as described below, and Shh−/− embryos were typed phenotypically (20). SJL mice were purchased from Harlan Olac (Bichester, U.K.). All mice were bred and maintained in individually ventilated cages at the Central Biomedical Services unit at Imperial College London. Timed mates were performed by mating a male with two females and monitoring the females for plugs. The day the plug was found was designated embryonic day 0.5 (E0.5).
Fetal liver chimeric mice were generated by irradiating SJL mice with 1100 rad and injecting them with 200 μl of E12.5 Shh−/−, Shh+/−, and Shh+/+ fetal liver cells on the following day. Mice were then maintained for an additional 8 wk before analysis.
Flow cytometry and Abs
Thymocyte suspensions were prepared by crushing thymi between two pieces of ground glass. Cells were stained using combinations of the following directly conjugated Abs obtained from BD PharMingen (San Diego, CA): anti-CD44-FITC, anti-CD44-CyChrome, anti-CD25-PE, anti-CD25-FITC, anti-CD4-FITC, anti-CD4-PE, anti-CD4-CyChrome, anti-CD8-CyChrome, anti-CD8-FITC, anti-CD8-PE, anti-CD45.1-PE, anti-CD45.2-FITC, anti-CD3-CyChrome, anti-CD3-PE, and anti-TCRβ-PE, anti-NK1.1-FITC, anti-B220-CyChrome, anti-γδ-biotin, and anti-streptavidin-CyChrome. Cell suspensions were stained with the Abs for 30 min on ice in 50 μl of DMEM (Life Technologies, Gaithersburg, MD) supplemented with 5% FCS and 0.01% sodium azide. Cells were washed in this medium between incubations and before analysis on the FACScan (BD Biosciences, Mountain View, CA). Events were collected in list mode using CellQuest software (BD Biosciences), and data were analyzed using CellQuest Pro software. Live cells were gated according to their forward and side scatter profiles. Data are representative of at least three experiments. Statistical analysis was conducted using an unpaired Student’s t test and StatView software.
Propidium iodide staining was conducted on cells treated with 100 μg/ml RNase (Sigma-Aldrich, St. Louis, MO) and permeabilized in 0.1% Triton X-100 as described previously (27).
Annexin V staining was conducted using an annexin V-FITC apoptosis detection kit (BD PharMingen) according to the manufacturer’s instructions. Before annexin V staining, cells were stained as described above.
The neutralizing mAb 5E1 (directed against Shh) developed by Drs. J. Ericson and T. Jessell (Columbia University, New York, NY) (28) was obtained from the Developmental Studies Hybridoma Bank, developed under the auspices of the National Institute of Child Health and Human Development, and maintained by the University of Iowa (Iowa City, IA). 5E1 was used at a concentration of 5 μg/ml in FTOC.
Fetal thymi were cultured on 8-μm pore size filters (Millipore, Bedford, MA) in AIM-V serum-free medium (Life Technologies) at 37°C and 5% CO2. For wild-type FTOC, BALB/c embryos were used.
Modified human Shh was a gift from Curis (Cambridge, MA). Shh was octylated for high activity. In vitro modification of N-Shh (aa 25–199) with a lipophilic group on the N-terminal cysteine significantly increases the specific activity (>30-fold) of Escherichia coli-derived N-Shh, as measured by activation of Hh signal transduction in cultured cells (29). Octyl N-Shh is a hydrophobically modified version of the Shh signaling protein generated by coupling N-octyl-malemide to the N-terminal cysteine of bacterially derived N-Shh.
DNA for PCR analysis was extracted from embryonic head tissue by digesting tissues in lysis buffer containing 50 mM KCl, 1.5 mM MgCl2, 10 mM Tris-HCl (pH 8.5), 0.01% gelatin, 0.45% Nonidet P-40, 0.45% Tween 20, and 0.5 μg/ml proteinase K (Sigma-Aldrich) in water. DNA (0.5 μg) was used as a template in each PCR. The primers used to amplify the inserted neomycin cassette were: forward, CTGTGCTCGACGTTGTCACTG; and reverse, GATCCCCTCAGAAGAACTCGT (20). Reactions were run on a Stratagene Robocycler (La Jolla, CA) as follows: 5 min at 94°C; 32 cycles of 1 min at 94°C, 1 min at 66°C, and 1 min at 72°C; and 10 min at 72°C. DNA products were resolved on a 2% agarose gel.
Shh regulates thymocyte development from DN1 to DN2
To provide genetic evidence for a role for Shh in the control of T cell development, we studied Shh−/− embryos. We analyzed thymocyte development in Shh−/−, Shh+/−, and Shh+/+ littermates. On E13.5, thymocyte numbers were reduced by between 17- and 6-fold (Fig. 1,A). Given the reduction in cell numbers, the thymocyte population was difficult to identify by side scatter profile. Therefore, we stained cells with anti-CD45, anti-CD25, and anti-CD44 and analyzed thymocyte development by gating on the CD45+ population. The percentage of cells within the live gate that expressed the leukocyte lineage marker CD45 was reduced in the Shh−/− embryos to ∼20% compared with ∼80% in littermate thymi (Fig. 1, B and D). Thus, the E13.5 Shh−/− thymus contained fewer lymphocyte lineage cells than its littermates, suggesting that progenitor cells were either less efficient at colonizing the thymus or were not expanding on entry into the thymus.
We found an increase in the proportion of CD44+CD25− cells and a decrease in the proportion of CD44+CD25+ cells in Shh−/− thymi compared with that of Shh+/− and Shh+/+ littermate thymi (Fig. 1,D). In a typical experiment, 81.9% of Shh−/− thymocytes were in the CD44+CD25− population compared with 35.5 and 38% of thymocytes from Shh+/− and Shh+/+ littermates, respectively. This reduction in CD25+ cells was observed in all litters examined (Fig. 1 C). Interestingly, in wild-type mice the DN2 population express high levels of the Hh signaling molecule Smo at their cell surface (16), and cell surface Smo is up-regulated in response to Hh signaling (30).
The DN1 population is not fully committed to the T cell lineage, but contains cells that are capable of differentiating into T, B, NK, and dendritic cell lineages (1, 31). In mice in which the notch1 gene was inactivated in thymocyte progenitors, there was an arrest in thymocyte development at the CD44+CD25− stage due to a failure of T cell lineage commitment. The CD44+CD25− population that accumulated in these mice was not equivalent to the DN1 population found in wild-type mice, because the cells expressed the B-lineage marker B220 and did not express CD117 (c-Kit) (32).
Therefore, it seemed possible that the accumulation of CD44+CD25− cells in the Shh−/− thymus was also the result of a failure of T lineage commitment, and that these cells were not true DN1 cells. To test this, we analyzed the expression of CD117 and B220 on the CD44+ DN thymocyte population from E13.5 embryos. More than 97% of CD44+ thymocytes from Shh−/− embryos stained positively with anti-CD117, suggesting that this population was equivalent to the most immature thymocyte population found in normal mice (Fig. 2,A). There was no difference in B220 expression on the CD44+CD25−DN population between Shh−/− and littermate thymi (Fig. 2,B). The expressions of NK1.1 and TCRγδ were also very similar in Shh−/− and littermate thymocytes (Fig. 2, C and D).
As we did not find evidence for abnormal lineage commitment in the Shh−/− thymi, the reduction in DN2 cells could either be due to a failure of the DN1 population to differentiate to DN2 or occur because the DN2 cells were failing to survive. To test this we analyzed cell death by annexin V staining. There was no difference in the proportion of apoptotic cells between +/+ and −/− thymi in either the DN1 or DN2 populations (Fig. 2, E and F). Thus, the reduction in DN2 cells occurred because the Shh−/− thymocytes were not efficiently making the transition from DN1 to DN2.
Shh regulates proliferation of DN thymocytes
When we analyzed thymocyte development on E14.5 and E15.5, we found that Shh−/− thymocyte numbers were still greatly reduced in all litters examined (Fig. 3,A). On E14.5 (Fig. 3,B) and E15.5 (data not shown), all DN subsets were present in the Shh−/− thymus. The proportion of Shh−/− thymocytes that expressed CD25 had increased, confirming that CD25+ cells could expand, survive, and accumulate, but was still reduced compared with that in +/+ littermates on E14.5 and E15.5 (Fig. 3 C). On E15.5, CD8+ ISPs and DP cells had not yet appeared (data not shown).
The reduction in thymocyte numbers in the Shh−/− thymus suggests that Shh is involved in signaling for thymocyte proliferation. Therefore, we analyzed the cell cycle status of E14.5 Shh−/− thymocytes. In a typical experiment, propidium iodide staining indicated that 10.5% of Shh−/− thymocytes were in the G2 and S phases of the cell cycle compared with 17 and 20% of Shh+/+ and Shh+/− littermate thymocytes, respectively (Fig. 3 D). Thus, Shh promotes DN thymocyte proliferation.
Reduced production of DP thymocytes in the Shh−/− thymus
On E16.5 the proportion of DP cells was reduced in Shh−/− thymi compared with littermates. In a typical experiment, only 9% of thymocytes were DP in the Shh−/− thymus compared with ∼20% in Shh+/+ and Shh+/− littermates. This decrease was accompanied by a concomitant increase in the proportion of DN cells (Fig. 4,B). Analysis of the DN subsets showed similar proportions of DN3 and DN4 cells in the Shh−/− and littermate thymi (Fig. 4,B). A reduction in thymocyte number and in the DP population was observed in all litters examined (Fig. 4, A–C). The reduction in the percentage of DPs in Shh−/− thymi could be the result of a partial arrest in differentiation at the transition from DN to DP cells or could occur because the DP cells were dying. To test this, we analyzed cell death by annexin V staining. There was no statistically significant difference in the proportion of dying cells in the DP populations from Shh−/− and littermates (Table I). Interestingly, we found increased cell death in the DN4 population (Table I and Fig. 4,D). In a typical experiment, 5.9% of Shh−/− DN4 cells stained positively with annexin V compared with 1.6% in the +/+ DN4 population (Fig. 4 D). These data suggest that Shh provides a survival signal for DN4 cells and are consistent with the reduced production of DP cells.
|.||Control (%) .||Knockout (%) .|
|Annexin V on DN4 cells||1.46 ± 0.25 (n = 9)||4.39 ± 0.64 (n = 4)|
|Annexin V on DP cells||1.73 ± 0.45 (n = 5)||2.30 ± 0.41 (n = 3)|
|γδ cells||4.20 ± 0.33 (n = 11)||5.33 ± 0.80 (n = 3)|
|.||Control (%) .||Knockout (%) .|
|Annexin V on DN4 cells||1.46 ± 0.25 (n = 9)||4.39 ± 0.64 (n = 4)|
|Annexin V on DP cells||1.73 ± 0.45 (n = 5)||2.30 ± 0.41 (n = 3)|
|γδ cells||4.20 ± 0.33 (n = 11)||5.33 ± 0.80 (n = 3)|
The mean percent (± SE) positive staining with annexin V on DN4 and DP cells and anti-TCRγδ on total thymocytes. The difference between the mean percent annexin V staining on DN4 cells between Shh−/− and control littermate thymi is statistically significant by Student’s t test (p = 0.0002), whereas the difference in annexin V staining on DP cells and the difference in anti-TCRγδ staining are not significant. To gate on DN4 cells, staining was carried out as described in Fig. 4. To gate on DP cells, thymocytes were stained with anti-CD4, anti-CD8, and annexin V.
When we examined TCRγδ expression in the E16.5 thymi we found no significant difference between Shh−/− and littermates (Table I).
Shh−/− thymi can produce SP cells in vitro
Shh−/− embryos die before birth (20). Therefore, to study later stages of fetal thymus development, we cultured E14.5 thymi. After 3 days, cell recovery from the Shh−/− cultures was reduced in all experiments (Fig. 5,A). After 6 days, cell number was still greatly reduced in the Shh−/− cultures (Fig. 5,A), but there was no significant difference in the proportion of DP cells among the Shh−/−, Shh+/−, and Shh+/+ cultures (Fig. 5, B and C). Analysis of the DN populations revealed an increase in the proportion of DN1 cells and a decrease in the proportion of CD25+ cells in the Shh−/− FTOC compared with littermate controls (Fig. 5,C), consistent with the composition of DN subsets observed ex vivo (Figs. 1 and 3).
Thus, after the initial lag phase, DP cells were able to accumulate in the Shh−/− cultures. After the 6-day culture period SP cells had appeared in all cultures, and levels of CD3 expression on these cells were similar (Fig. 5 D), indicating that cultures from Shh−/− were able to produce mature SP T cells.
Dose-dependent outcomes of Shh signaling in wild-type FTOC
We have previously shown that treatment of wild-type FTOC with between 0.3 and 1 μg/ml recombinant Shh arrested thymocyte development at the DN stage (16). This arrest was not due to nonspecific toxicity of the recombinant Shh, but was a genuine biological effect, as it could be neutralized by addition of anti-Hh mAb. The analysis of E16.5 Shh−/− thymi suggested, however, that some Shh signaling is necessary at the transition from DN to DP cells (Fig. 4,B). Therefore, we decided to determine whether treatment of FTOC with lower concentrations of Shh could promote thymocyte development. We treated wild-type FTOC for 5 days with 0.0005 μg/ml Shh or 0.5 μg/ml Shh. In a typical experiment treatment with 0.5 μg/ml Shh arrested thymocyte development at the DN stage, and <1% of thymocytes were DP compared with 7.6% in the control cultures, whereas treatment with 0.0005 μg/ml Shh promoted thymocyte development, and 15.2% of cells were DP (Fig. 6,A). Treatment with 0.0005 μg/ml Shh also increased cell recovery relative to the control, and high dose treatment reduced cell recovery (Fig. 6, A–C). Titration of recombinant Shh in FTOC revealed that the dose range from 0.005–0.0005 μg/ml promoted thymocyte development, but this effect was lost in cultures treated with 0.00005 μg/ml (Fig. 6,B). Treatment with 0.5 μg/ml inhibited both the production of DP cells and the overall cell recovery, and this inhibition was neutralized by addition of anti-Hh mAb, confirming that the arrest in thymocyte development was not the result of nonspecific toxicity of the reagent (Fig. 6,B). Thus, in wild-type FTOC, treatment with different concentrations of recombinant Shh can produce distinct outcomes at the transition from DN to DP cells; high-dose Shh treatment arrests thymocyte development, but treatment with a 1000-fold lower dose promotes the production of DP cells (Fig. 6, A–C).
Treatment with Shh promotes thymocyte development in Shh−/− FTOC
As thymocyte cellularity and differentiation were reduced in Shh−/− thymi, we examined whether we could promote thymocyte development in Shh−/− thymus explants by treatment with recombinant Shh. We treated Shh−/− thymus explants for 7 days with 0.5 or 0.0005 μg/ml recombinant Shh. Treatment with 0.5 μg/ml Shh inhibited differentiation from DN to DP cells. In a typical experiment, 5.2% of thymocytes were DP in the treated cultures compared with 36.6% in control cultures (Fig. 6, D and E). Thus, Shh−/− thymocytes were able to respond to a Shh signal in the same way as wild-type thymocytes despite the fact that they had not been exposed to Shh in vivo. Although this high dose of Shh arrested thymocyte differentiation, some activation of the Hh signaling pathway is clearly necessary for normal thymocyte differentiation. When we treated Shh−/− explants with a 1000-fold lower dose of recombinant Shh (0.0005 μg/ml), we found that low dose Shh treatment promoted differentiation to DP cell. After 7 days in culture, Shh−/− thymus explants treated with 0.0005 μg/ml Shh contained 41.6% DP thymocytes and 32.7% DN thymocytes compared with 36.6% DP cells and 43.1% DN cells in untreated control Shh−/− thymus cultures (Fig. 6, D and E). Cell number was also increased in the Shh−/− thymus explants treated with low dose Shh compared with the untreated Shh−/− explants (summarized in Fig. 6,E), so that overall the low dose treatment more than doubled the production of DP cells (summarized in Fig. 6 E).
We repeated these experiments by treating individual Shh−/− thymus lobes with recombinant Shh and comparing them to untreated cultures of the other lobe from the same embryo (summarized in Fig. 6,F, cell production in each culture condition compared with that in the control lobe are shown on a log scale). These cultures confirmed that in all cases low dose Shh treatment increased both cell recovery and the production of DP cells compared with the control lobe from the same embryo. In contrast, high dose Shh treatment inhibited both overall cell production and differentiation to DP cell (Fig. 6 F).
Shh−/− fetal liver cells can reconstitute wild-type irradiated thymus
In the thymus, Shh is produced by the thymic stroma, as evidenced by both RT-PCR from 2′,4′-deoxyguanosine-resistant (stromal fraction) fetal thymus explants and immunohistochemistry of adult thymus sections, in which anti-Shh staining colocalized with anti-cytokeratin staining (16, 33). However, Shh has also been shown to regulate the expansion of primitive human hemopoietic cells in an autocrine manner (17). On E13.5 the number and proportion of hemopoietic (CD45+) cells in the Shh−/− thymus were greatly reduced relative to littermates (Fig. 1). It therefore seemed possible that Shh might also be produced by the earliest thymocyte progenitors and regulate their expansion and differentiation in an autocrine manner. Thus, the reduction in thymocyte development found in the Shh−/− embryos on E13.5 might be the result of a failure of hemopoietic stem cells to expand or enter the thymus or of early thymocyte progenitors to make Shh, rather than be caused by lack of Shh production by the thymic stroma. To test whether Shh−/− fetal liver cells were able to enter and fully repopulate an adult Shh-expressing thymus, we reconstituted irradiated adult wild-type mice with fetal liver cells from Shh−/− embryos. Adult SJL mice (CD45.1+) were irradiated with 1100 rad and injected i.v. with fetal liver cells from E12.5 Shh−/− or littermate embryos (CD45.2+). After 2 mo the thymus had fully reconstituted, and there was no difference in thymus cellularity or thymocyte differentiation between mice reconstituted with Shh−/− or littermate fetal liver cells (Fig. 7, A–D). Staining with anti-CD45.1 (host origin) and anti-CD45.2 (donor origin) indicated that >98% of thymocytes were of donor origin (Fig. 7,A). Three-color flow cytometry using Abs directed against CD4, CD8, and CD45.2 indicated that the Shh−/− fetal liver cells were able to give rise to DN, DP, and SP populations in normal ratios (Fig. 7,B). To analyze the expression of CD25 in DN cells, we stained with anti-CD4, anti-CD8, anti-CD45.1, and anti-CD25 and excluded cells that stained positively with anti-CD4, anti-CD8 and anti-CD45.1. In both Shh−/− reconstituted and control reconstituted thymi, ∼25% of DN cells expressed cell surface CD25 (Fig. 7,C). Comparison of CD3 expression, gated on the CD45.2+ cells, showed no significant difference between Shh−/− reconstituted and control reconstituted thymi (Fig. 7 D). Thus, Shh−/− fetal liver cells were able to reconstitute an irradiated wild-type thymus in which thymic stroma produce Shh (16, 33). This indicates that Shh−/− fetal liver cells were able to survive and home to the thymus and that autocrine Shh production is not essential for their maintenance in the thymus, because they can receive necessary Hh signals from the adult thymic stroma.
This analysis of Shh−/− embryos demonstrates that Shh is important in the expansion and development of DN thymocytes. We have identified two stages at which Shh seems to be necessary for normal thymocyte development (Fig. 7 E). The transition from DN1 to DN2 cell is severely impeded in knockout thymi. The DN3 population then seems to partially recover, but a second bottleneck in development occurs at the transition to DP cell, and there is an increase in cell death of the DN4 population. Shh signaling does not seem to be essential beyond the DP stage of thymocyte development, as after 6 days in culture Shh−/− FTOC were able to partially recover, producing approximately one-fifth the number of SP cells as +/+ littermate thymi. It is possible, however, that there is redundancy between Hh family members in the thymus, and that Ihh functions at later stages of T cell development.
Several other signaling pathways have been shown to be involved in the regulation of differentiation from DN1 to DN2. Both Notch1 (32) and the Wnt signaling pathway (34) are required for this transition, whereas in vitro bone morphogenetic protein 4 arrests differentiation at the DN1 stage (27). These pathways interact with one another in the development of other tissues (35), and it will be interesting to analyze the relationship between these different molecules in the control of thymocyte differentiation.
We have previously shown that in vitro treatment of FTOC with recombinant Shh arrested thymocyte development at the DN3 stage, after TCRβ-chain gene rearrangement, whereas treatment of FTOC with a neutralizing anti-Hh Ab accelerated thymocyte differentiation to the DP stage, but did not replace the need for a pre-TCR signal (16). Therefore, both the reduced differentiation to DP cell in the Shh−/− thymus and the finding that treatment with low dose Shh promoted thymocyte development in vitro seemed surprising. There are at least two possible explanations for these apparent discrepancies between the experimental systems. 1) This analysis of the Shh−/− embryos indicated that Shh signaling is important very early in thymocyte development (before pre-TCR signaling). Thus, the production of DP cells in the Shh−/− thymus might be reduced because fewer thymocytes were reaching the stage at which they could transduce the pre-TCR signal rather than because the pre-TCR signal itself required some input from the Hh signaling pathway. In this case, treatment with low dose Shh might promote thymocyte development at an early stage, before pre-TCR signaling, whereas we have previously shown that neutralization of Hh signaling by treatment with anti-Hh acts after pre-TCR signaling during the transition to DP cell (16).
2) Alternatively, it is possible that different concentrations of Shh produce distinct outcomes at the transition from DN to DP cells. Classical morphogens specify different cell fates in a concentration-dependent manner (36), and different doses of Shh have been shown to produce distinct cellular fates in the developing neural tube (37). It is possible that although addition of a high dose of recombinant Shh arrests thymocyte development at the DN3 stage and counteracts the pre-TCR signal (16), some activation of the Hh pathway is necessary later for progression beyond the DN4 stage or for survival of DN4 thymocytes. In wild-type FTOC neutralization of Hh signaling by Ab treatment accelerated differentiation to DP cell (16), but Ab treatment would presumably greatly lower the dose of available Shh, but not neutralize all molecules. When we treated Shh−/− thymus explants with very low dose recombinant Shh (1/1000th the arresting dose), thymocyte development was promoted, and the production of DP cells was increased. It remains to be determined whether this low dose was acting directly at the transition to DP cell, or if its effects were due to its action on earlier DN cells.
The Shh−/− thymi had greatly reduced cellularity at all developmental stages examined. Shh−/− embryos are smaller than those of their littermates (20), but it seems unlikely that the reduction in thymus cellularity was wholly due to the small size of the embryos. For example, on E13.5, thymocyte numbers were reduced by between 17- and 6-fold compared with those of their littermates (Fig. 1, A and B), but the length of the embryos was 70% that of their littermates, and another hemopoietic organ, fetal liver, contained the same number of cells as in littermates (our unpublished observations). Likewise, the fact that treatment of thymus explants with recombinant Shh is able to increase the production of DP thymocytes in vitro demonstrates that the phenotype of the Shh−/− thymus cannot be simply explained by the general runtiness of the embryo.
In summary, we provide genetic evidence that Shh has a role in the control of thymocyte development in vivo. Shh is involved in the proliferation and efficient development of DN cells and in the regulation of thymus cellularity. Shh, produced by the thymic stroma, provides signals for the transition from DN1 to DN2 cells and promotes the production of DP cells (Fig. 7 E).
We thank Prof. Phillip Beachy for the Shh+/− mice, and Curis for the recombinant Shh protein.
This work was supported by the Wellcome Trust and the Medical Research Council.
Abbreviations used in this paper: γc, common cytokine receptor γ-chain; DN, double negative; E, embryonic day; FTOC, fetal thymus organ culture; Hh, Hedgehog; ISP, immature single positive; Ptc, Patched; Shh, Sonic Hedgehog; Smo, Smoothened; SP, single positive.