Interleukin-7 supports survival of T-cell receptor-β-expressing CD4 CD8 double-negative thymocytes

Authors


Correspondence: Dr Thomas M. Yankee, Department of Microbiology, Molecular Genetics, and Immunology, University of Kansas Medical Center, 3901 Rainbow Blvd, 3025 WHW – MS 3029, Kansas City, KS 66160, USA. Email: tyankee@kumc.edu

Senior author: Thomas M. Yankee

Summary

Among the milestones that occur during T-cell development in the thymus is the expression of T-cell receptor-β (TCR-β) and the formation of the pre-TCR complex. Signals emanating from the pre-TCR trigger survival, proliferation and differentiation of T-cell precursors. Although the pre-TCR is essential for these cell outcomes, other receptors, such as Notch and CXCR4, also contribute. Whether interleukin-7 (IL-7) participates in promoting the survival or proliferation of pre-TCR-expressing cells is controversial. We used in vitro and in vivo models of T-cell development to examine the function of IL-7 in TCR-β-expressing thymocytes. Culturing TCR-β-expressing CD4 CD8 double-negative thymocytes in an in vitro model of T-cell development revealed that IL-7 reduced the frequency of CD4+ CD8+ double-positive thymocytes at the time of harvest. The mechanism for this change in the percentage of double-positive cells was that IL-7 promoted the survival of thymocytes that had not yet differentiated. By preserving the double-negative population, IL-7 reduced the frequency of double-positive thymocytes. Interleukin-7 was not required for proliferation in the in vitro system. To follow this observation, we examined mice lacking CD127 (IL-7Rα). In addition to the known effect of CD127 deficiency on T-cell development before TCR-β expression, CD127 deficiency also impaired the development of TCR-β-expressing double-negative thymocytes. Specifically, we found that Bcl-2 expression and cell cycle progression were reduced in TCR-β-expressing double-negative thymocytes in mice lacking CD127. We conclude that IL-7 continues to function after TCR-β is expressed by promoting the survival of TCR-β-expressing double-negative thymocytes.

Introduction

T-cell development in the thymus is an ordered process that begins within a population of thymocytes that lack CD4 and CD8 expression. This developmental stage, CD4− CD8 double-negative (DN) thymocytes, can be subdivided into at least five subsets: DN1 (CD44hi CD25), DN2 (CD44hi CD25hi), DN3E (CD44lo CD25hi), DN3L (CD44lo CD25lo) and DN4 (CD44lo CD25).[1, 2] Among the first milestones of T-cell development is the expression of T-cell receptor-β (TCR-β), which can first be detected during the DN3E stage. Once TCR-β is expressed, it combines with pre-Tα, CD3 proteins, and ζ chains and transmits signals that drive survival, proliferation and differentiation (reviewed in refs [3, 4]). The expression of TCR-β and the formation of the pre-TCR are clearly required to advance beyond the DN stage of T-cell development because mice that cannot express TCR-β or signal from the pre-TCR have a complete block in development at the DN3E stage.[5-9]

The pre-TCR is critical for β selection, but other receptor systems are also important, such as Notch and CXCR4.[10-19] We recently showed that DN thymocytes remain responsive to Notch ligation throughout the DN subsets and into the immature single-positive (ISP) CD8+ stage of development.[17] Specifically, Notch is required for survival and proliferation of pre-TCR-expressing thymocytes. CXCR4 also promotes survival of DN3 and DN4 thymocytes and acts as a co-receptor for pre-TCR signalling.[18]

In this report, we tested whether the interleukin-7 (IL-7) receptor could also contribute to the survival or proliferation of TCR-β-expressing DN thymocytes. The IL-7 receptor consists of CD127 (IL-7Rα) and CD132 (γc). The function of IL-7 before β selection is clear; mice lacking IL-7 or CD127 have a dramatic decrease in the absolute number of thymocytes and the block can be traced to the DN1 or DN2 developmental stage.[20-22] Whether IL-7 continues to be required after TCR-β is expressed is controversial.

We propose that IL-7 is required for optimal T-cell development beyond the stage at which TCR-β is expressed. This model is based, in part, on data showing that pre-TCR signals can drive the expression of CD127.[23] Specifically, the pre-TCR signals dependent on the Gads adaptor protein augment CD127 expression.[2] Further, an in vitro thymocyte differentiation assay has been used to show that the addition of IL-7 to TCR-β-expressing DN thymocytes changes the percentages of double-positive (DP) thymocytes obtained at the time of harvest.[2, 24] The mechanism of how IL-7 mediates this effect is unknown.

The goal of our study is to determine whether IL-7 regulates survival or proliferation of TCR-β-expressing DN thymocytes. To accomplish this goal, we used an in vitro differentiation system in which we added or omitted IL-7. We also analysed TCR-β-expressing DN thymocytes from mice lacking CD127. Based on these studies, we conclude that IL-7 promotes the survival of TCR-β+ DN thymocytes by augmenting Bcl-2 expression.

Materials and methods

Mice

CD127/ C57BL/6 mice were purchased from the Jackson Laboratory (Bar Harbor, ME). CD127+/+ and CD127/ C57BL/6 mice were housed under specific pathogen-free conditions with sterile cages, bedding, food and water. All experiments were performed in compliance with the University of Kansas Medical Center Institutional Animal Care and Use Committee. Mice were used between the ages of 3 and 5 weeks.

Antibodies, cell labelling and flow cytometry

Anti-TER119-FITC, anti-CD4-FITC, anti-CD8-FITC, anti-CD44-phycoerythrin-Cy7, anti-CD25-allophycocyanin-Cy7, anti-CD4-HorizonV 450, anti-CD8-Alexa647, anti-CD4-allophycocyanin-Cy7, anti-CD25-FITC, anti-TCR-β-phycoerythrin, anti-CD4-Pacific Blue, anti-TCR-β-allophycocyanin, Annexin V-Horizon V450 and anti-Bcl-2-phycoerythrin were purchased from either BD Biosciences (San Jose, CA) or BioLegend (San Diego, CA).

Surface staining was performed in staining buffer (PBS containing 2% α calf fraction; Hyclone, Waltham, MA) and fixed in 1% paraformaldehyde before analysis. For some samples, CountBright™ absolute counting beads (Life Technologies, Grand Island, NY) were added immediately before running the samples.

Intracellular staining was performed by first completing the surface staining and fixing the cells. Then, cells were washed in staining buffer before permeabilizing in PBS containing 0·1% saponin and 10% α calf fraction for 15 min at room temperature. Cells were stained and washed in saponin buffer at room temperature and fixed in 1% paraformaldehyde before analysis.

For samples labelled with Annexin V-Horizon V450, cells were surface labelled and washed in Annexin V binding buffer [0·01 m HEPES (pH 7·4), 0·14 m NaCl and 2·5 mm CaCl2] before analysis. For cell cycle analysis, thymocytes were labelled with surface and intracellular antibodies, fixed and washed with staining buffer. Then, cells were resuspended in DAPI 1 μg/ml in 0·2% Tween-20 and 2% α calf fraction in PBS. Cells were incubated in the dark for 30 min at room temperature before analysis.

Cells were analysed using a BD LSR II (BD Biosciences). Data were analysed using BD FACSdiva software (BD Biosciences) or FlowJo (Tree Star, Inc., Ashland, OR).

FACS purification

Isolation of DN3E, DN3L and DN4 thymocytes was performed as previously described.[2, 17] Briefly, total thymocytes were depleted using anti-mouse CD4 magnetic particles-DM (BD Biosciences). Remaining cells were surface labelled with anti-TER119, anti-CD4, anti-CD8, anti-CD44 and anti-CD25. Then, cells were gated on TER119  CD4  CD8 thymocytes and DN3E, DN3L and DN4 thymocytes were FACS-purified using a BD FACSAria IIu (BD Biosciences) based on CD44 and CD25 expression.

In vitro differentiation

The in vitro differentiation system was used as described previously.[2, 17, 25] Briefly, 5 × 104 FACS-purified thymocytes were cultured with OP9-DL1 cells or OP9-DL4 cells (generous gifts of Dr Juan-Carlos Zúñiga-Pflücker) [25, 26] in MEM α medium (Invitrogen, Carlsbad, CA) supplemented with 20% fetal bovine serum, penicillin, streptomycin and 5 ng/ml Flt3L (Fms-related tyrosine kinase 3 ligand; PeproTech, Rocky Hill, NJ). Some samples were supplemented with 5 ng/ml IL-7. For some experiments, FACS-purified thymocytes were labelled with 5 μm carboxyfluorescein succinimidyl ester (CFSE) before culturing.

Statistics

Statistics were performed using graphpad Prism software (GraphPad Software, La Jolla, CA). Unless otherwise indicated, the two-tailed Student's t-test was used.

Results

IL-7 reduced the percentages of DP thymocytes in culture

We previously used an in vitro differentiation assay in which we FACS-purified DN3E, DN3L and DN4 thymocytes, cultured the cells with OP9-DL1 cells for 1 or 2 days, and found that IL-7 could alter the percentages of DP thymocytes harvested from each well.[2] OP9-DL1 cells are a bone marrow stromal cell line that expresses Delta-like 1, a Notch ligand. Here, we extended our previous data by increasing the duration of the culture to 88 hr and testing whether culturing thymocytes with OP9-DL4 cells affected thymocyte differentiation. OP9-DL4 cells express Delta-like 4, the Notch ligand most abundantly found in the thymus.[27-29] For each population of cells tested, the percentages of thymocytes that were DP cells at the time of harvest were greater when cultures were deprived of IL-7 than when cultures contained IL-7 (Fig. 1a). This effect was most evident when DN3E cells were seeded into the wells. For wells seeded with DN3L or DN4 thymocytes, most cells had differentiated into DP thymocytes at the time of harvest regardless of the presence of IL-7, making the differences between wells containing IL-7 and lacking IL-7 smaller. There was little difference in the frequency of DP thymocytes between cultures containing OP9-DL1 cells and OP9-DL4 cells, indicating that the Notch ligand did not influence differentiation.

Figure 1.

Interleukin-7 (IL-7) reduces the frequency of double-positive (DP) thymocytes, but enhances the number of thymocytes recovered. DN3E, DN3L, and DN4 thymocytes were co-cultured for 88 hr with OP9-DL1 or OP9-DL4 cells. IL-7 was added to some wells (+IL-7) and omitted from others (−IL-7). (a) Percentages of cells in each well that were DP thymocytes at the end of the culture. Data shown are representative of six independent experiments. (b) The absolute numbers of live thymocytes were calculated for each sample. Each pair of samples connected by a line represents an independent cell isolation divided into wells containing or lacking IL-7. Comparisons between samples containing and lacking IL-7 were calculated using a paired t-test. *= 0·064, **< 0·05, ***< 0·005.

Next, we calculated the number of thymocytes harvested from each well 88 hr after incubation (Fig. 1b). For each subset of thymocytes tested, the absolute number of thymocytes recovered ranged from 2·0-fold greater to 3·6-fold greater when wells contained IL-7, compared with when wells lacked IL-7. The fold change between wells containing and lacking IL-7 was statistically identical regardless of the starting thymocyte population and whether OP9-DL1 or OP9-DL4 cells were used.

The results of these in vitro differentiation assays suggested that IL-7 could promote survival, proliferation, or both. Further, it is possible that IL-7 could block the differentiation of DN thymocytes into DP thymocytes. We previously showed that any cells lacking TCR-β in this study fail to survive, even in the presence of IL-7.[17] Further, approximately 80% of DN3L and DN4 thymocytes express TCR-β. Hence, it is unlikely that the effects of IL-7 in this assay could result from its action on TCR-β DN thymocytes.

IL-7 regulates survival and proliferation of cultured thymocytes

To determine whether IL-7 could regulate proliferation, we labelled DN3E, DN3L and DN4 thymocytes with CFSE before culturing with OP9-DL1 or OP9-DL4 cells. At the conclusion of the experiment, we analysed proliferation among the DN and ISP thymocytes (Fig. 2a) or DP thymocytes (Fig. 2b). Because nearly all the cells divided regardless of the culture conditions, we calculated the percentage of cells that underwent more than three divisions. For DN3E-derived thymocytes, there were no statistically significant differences in the percentages of thymocytes that divided more than three times (Fig. 2c), regardless of whether cells were cultured with OP9-DL1 or OP9-DL4 cells and whether DN/ISP or DP thymocytes were analysed.

Figure 2.

Interleukin7 (IL-7) does not promote proliferation of DN3E, DN3L, or DN4 thymocytes in culture. DN3E, DN3L, and DN4 thymocytes were cultured as described in the legend to Fig. 1, except that the thymocytes were labelled with CFSE before incubation. After culturing for 88 hr, cells were gated on the double-negative (DN) and immature single-positive (ISP) populations (a) or the double-positive (DP) population (b) and analysed using CFSE to assess cell division. The percentages of thymocytes that divided more than three times are shown. Data shown are representative of four independent experiments. (c) Mean ± standard deviation of the data represented by panels (a) (upper panels) and (b) (lower panels).

By contrast, proliferation of DN3L-derived DN and ISP thymocytes was enhanced by IL-7. This observation was only statistically significant when cells were cultured in OP9-DL1 cells (< 0·05). Because only a small number of thymocytes could be harvested when DN3L thymocytes were cultured with OP9-DL4 cells in the absence of IL-7, it was difficult to achieve statistical significance (= 0·10) even though the trend suggested that IL-7 enhanced proliferation (Fig. 2c). In contrast to the DN and ISP populations, IL-7 did not affect the proliferation of DN3L-derived DP thymocytes.

For DN4-derived thymocytes, there were no statistical differences in the percentages of cells that divided more than three times, whether we analysed DN/ISP thymocytes or DP thymocytes. This may be because the only DN4 thymocytes that did not differentiate into DP thymocytes during the course of the experiments ordinarily do not proliferate robustly, either because they are γδ T cells[2] or dying thymocytes.

In conclusion, IL-7 promoted the proliferation of DN and ISP thymocytes derived from DN3L cells. Once the thymocytes differentiated into DP cells, proliferation was independent of IL-7 (Fig. 2b).

Next, we tested whether IL-7 affected cell survival by labelling the thymocytes with Annexin V. As with the proliferation assays, we separated the analysis of DN and ISP thymocytes (Fig. 3a) from that of DP thymocytes (Fig. 3b). When DN3E, DN3Land DN4 thymocytes were co-cultured with OP9-DL1 cells, fewer DN and ISP thymocytes were Annexin V-positive when IL-7 was added to the cultures than when IL-7 was omitted (< 0·05). These trends were also observed with OP9-DL4 cells, but the data did not reach statistical significance with OP9-DL4 cells. When cells were gated on the DP thymocytes, there were no statistically significant differences in the percentages of cells that bound Annexin V between cultures containing or lacking IL-7.

Figure 3.

Interleukin-7 (IL-7) promotes survival of double-negative (DN) and immature single-positive (ISP) thymocytes. DN3E, DN3L, and DN4 thymocytes were co-cultured with OP9-DL1 or OP9-DL4 cells. IL-7 was added to some cultures (+IL-7) and omitted from other cultures (−IL-7). After 60 hr, cells were labelled with anti-CD4, anti-CD8 and Annexin V. The percentages of CD4 (a) or CD4+ (b) thymocytes that were positive for Annexin V are shown for each sample. Representative of three independent experiments.

In summary, our in vitro data indicated that a greater number of thymocytes were recovered from cultures containing IL-7 than cultures lacking IL-7. This increase in cell number was most likely related to the observation that IL-7 promoted survival and possibly proliferation of DN and ISP thymocytes, but not DP thymocytes.

CD127−/− mice have impaired development of TCR-β+ DN thymocytes

Next, we examined T-cell development in CD127−/− mice, focusing on T-cell populations that express TCR-β. As previously reported,[20-22] CD127−/− mice have few thymocytes; on average, CD127+/+ mice contained 107 ± 40 million cells whereas CD127−/− mice contained 1·8 ± 0·72 million cells. Despite the defect in cell number, the distribution of DN, DP and SP thymocytes was similar between CD127+/+ and CD127−/− mice (Fig. 4a). Within the DN population of thymocytes, the percentages of cells in the DN1, DN2, DN3E, DN3L and DN4 gates were markedly different between CD127+/+ and CD127−/− mice (Fig. 4b). As expected, CD127−/− mice had a greater percentage of DN1 and DN2 thymocytes than CD127+/+ mice. On average, 39 ± 19% of CD127−/− DN thymocytes were in the DN1 subset, compared with 12 ± 1·8% of CD127+/+ DN thymocytes (= 0·0003, = 7). As a consequence of the increased DN1 population, the percentages of DN thymocytes that were in the DN3E, DN3L and DN4 subsets were reduced in CD127−/− mice, compared with CD127+/+ mice (< 0·05).

Figure 4.

CD127 is required for T-cell development before T-cell receptor-β (TCR-β) expression. Thymocytes from CD127+/+ and CD127−/− mice were labelled with anti-CD4, anti-CD8, anti-CD44 and anti-CD25. (a) Cells were analysed for CD4 and CD8 expression. Percentages of thymocytes that were DN thymocytes are shown. (b) Double-negative (DN) thymocytes from (a) were analysed for CD44 and CD25 expression. The percentages of cells in each gate are shown below the dot plot. Data shown are representative of seven independent experiments.

Next, we focused on the DN3E, DN3L and DN4 populations and analysed TCR-β expression in CD127+/+ and CD127−/− thymocytes (Fig. 5a). The percentages of DN3E thymocytes that expressed TCR-β in each mouse line were identical. By contrast, a smaller percentage of CD127−/− DN3L and DN4 thymocytes expressed TCR-β than CD127+/+ mice. On average, 67 ± 15% of CD127+/+ DN3L thymocytes expressed TCR-β whereas 32 ± 6·4% of CD127−/− DN3L thymocytes expressed TCR-β (< 0·0001, = 7). Within the DN4 population, 74 ± 6·1% of CD127+/+ cells expressed TCR-β and 53 ± 15% of CD127−/− cells expressed TCR-β (< 0·005, = 7). These data suggested that CD127 plays a role in T-cell development after the expression of TCR-β and is necessary for TCR-β+ thymocytes to progress through the DN3L and DN4 stages.

Figure 5.

CD127 is required for survival and proliferation of T-cell receptor-β-positive (TCR-β+) DN3E, TCR-β+ DN3L and TCR-β+ DN4 thymocytes in vivo. CD127+/+ and CD127−/− thymocytes were labelled with anti-CD4, anti-CD8, anti-CD44, anti-CD25, intracellular anti-TCR-β and either anti-Bcl-2 or DAPI. (a) The percentages of DN3E, DN3L and DN4 thymocytes that expressed TCR-β are shown. (b) TCR-β+ DN3E, TCR-β+ DN3L and TCR-β+ DN4 thymocytes from CD127+/+ mice (dark line) and CD127−/− mice (shaded histogram with dashed outline) were labelled with anti-Bcl-2. Isotype control is shown in the dotted line. Data shown are representative of at least seven mice and three independent experiments. (c) TCR-β+ DN3E, TCR-β+ DN3L, and TCR-β+ DN4 thymocytes from CD127+/+ mice and CD127−/− mice were labelled with DAPI to determine their cell cycle status. The percentages of thymocytes that were in the S, G2 or M phase of the cell cycle are shown for representative data. (d) TCR-β+ DN3E, TCR-β+ DN3L, and TCR-β+ DN4 thymocytes from CD127+/+ mice (+/+) and CD127−/− mice (−/−) were analysed as described in (c). The bars represent the mean ± standard deviation for the each group of mice. *< 0·01.

Survival and proliferation of CD127−/− TCR-β+ DN thymocytes are impaired

A defect in survival or proliferation could account for the decreased percentages of TCR-β+ DN3L and DN4 thymocytes observed in CD127−/− mice. As a marker for survival, we analysed the expression of the pro-survival protein Bcl-2 in TCR-β+ DN thymocyte subsets (Fig. 5b). Bcl-2 expression was reduced, but detectable, in CD127−/− TCR-β+ DN3E thymocytes, as compared with CD127+/+ cells. Bcl-2 was absent in most CD127−/− TCR-β+ DN3L and TCR-β+ DN4 thymocytes, but Bcl-2 was detectable in a subset of CD127−/− TCR-β+ DN3L and DN4 thymocytes.

Finally, we examined the percentages of CD127+/+ and CD127−/− TCR-β+ DN thymocyte subsets that were in the S, G2 or M phase of the cell cycle (Fig. 5c). The percentages of CD127+/+ and CD127−/− TCR-β+ DN3E thymocytes in the S, G2 or M phase of the cell cycle were identical; 43 ± 4·2% of CD127+/+ TCR-β+ DN3E thymocytes and 42 ± 6·6% of CD127−/− TCR-β+ DN3E thymocytes were in the S, G2 or M phase of the cell cycle. However, fewer CD127−/− TCR-β+ DN3L and TCR-β+ DN4 thymocytes were in the S, G2 or M phase of the cell cycle, compared with CD127+/+ cells. On average, 42 ± 9·5% of CD127+/+ TCR-β+ DN3L thymocytes and 19 ± 14% of CD127−/− TCR-β+ DN3L thymocytes were in the S, G2 or M phase of the cell cycle (= 0·0093). For TCR-β+ DN4 thymocytes, 13 ± 6·9% of CD127+/+ cells and 3·4 ± 3·1% of CD127−/− cells were in the S, G2 or M phase of the cell cycle (= 0·0045).

We conclude that CD127 was required for survival of TCR-β+ DN thymocytes by increasing the expression of Bcl-2. In addition, CD127 expression resulted in a greater percentage of TCR-β+ DN thymocytes progressing through the cell cycle.

Discussion

In this manuscript, we used two experimental models, an in vitro model and an in vivo model, to demonstrate that IL-7 remains important for T-cell development beyond the stage at which TCR-β is expressed. In the first model, we purified DN3E, DN3L and DN4 thymocytes and cultured them using the OP9 in vitro differentiation assay. Like our previous data[2] and other reported data,[24, 30, 31] the addition of IL-7 to the culture resulted in a reduced frequency of DP thymocytes (Fig. 1). However, the mechanism for this phenomenon had not been examined previously. We showed that the increased frequency of DP thymocytes is a result of decreased survival of cells that had not yet differentiated into the DP stage (Fig. 3). Further, we demonstrated that the impaired survival is a direct result of the actions of IL-7 on TCR-β+ DN subsets.

Our conclusion that IL-7 directly acts on TCR-β+ DN thymocytes is based on the fact that we were able to FACS-purify the DN subsets that express TCR-β. Approximately 20% of DN3E thymocytes express TCR-β whereas approximately 65–80% of DN3L and DN4 thymocytes express TCR-β.[2] Comparing DN3 and DN4 subsets immediately after harvesting from mice can lead to deceptive results because none of these populations express TCR-β in 100% of the cells. Any outcome from the total population could be derived from the TCR-β or TCR-β+ cells. However, culturing DN3E, DN3L and DN4 thymocytes for several days in vitro eliminates cells that do not express TCR-β;[17] TCR-β cells either undergo apoptosis or express TCR-β during the culture. Hence, any differences observed in cultures containing or lacking IL-7 are most likely caused by the effects of IL-7 on TCR-β+ DN thymocytes.

Adding support for our conclusion that IL-7 promotes the survival of DN and ISP thymocytes in vitro, we found that this effect of IL-7 occurred in cultures containing either OP9-DL1 or OP9-DL4 cells. OP9-DL1 cells are commonly used as a Notch ligand in this in vitro system and DL1 can promote survival, proliferation and differentiation of DN thymocytes.[25] However, DL4 is the Notch ligand most abundant in the thymus and likely to be more physiologically relevant than DL1.[27-29] DL4 binds DN thymocytes more efficiently than DL1,[28] suggesting that the physiological consequences of Notch ligation by DL1 and DL4 may differ. In our studies, IL-7 was required for survival of TCR-β+ DN thymocytes regardless of whether OP9-DL1 cells or OP9-DL4 cells were used to provide the Notch ligand.

A second model that we used to demonstrate the function for IL-7 in TCR-β+ DN thymocytes was the CD127−/− mouse line. As in previous studies,[20-22] the most obvious defects in T-cell development in CD127−/− mice were in the DN1 and DN2 populations. Our data indicate that CD127−/− mice have a second block in T-cell development that occurs after TCR-β is expressed (Fig. 5). Previously published data showed that pre-TCR signalling augments CD127 expression.[2, 23] These results suggest that signalling through the pre-TCR and CD127 combine to promote β selection of DN thymocytes, although whether IL-7 actually functions in TCR-β+ DN thymocytes has been controversial. To directly assess the effects of CD127 deficiency on TCR-β+ DN thymocytes, we electronically gated on these subsets of DN thymocytes and analysed Bcl-2 expression and cell cycle progression (Fig. 5). Compared with CD127+/+ thymocytes, Bcl-2 expression was reduced in CD127−/− TCR-β+ DN3E thymocytes and nearly absent in the TCR-β+ DN3L and TCR-β+ DN4 subsets. The reduction in Bcl-2 expression among TCR-β+ DN subsets is consistent with our in vitro data showing that DN and ISP thymocytes failed to survive without IL-7 (Fig. 3).

Our studies using CD127−/− mice also showed that cell cycle progression of TCR-β+ DN3L and TCR-β+ DN4 thymocytes was reduced in CD127−/− mice, compared with CD127+/+ mice (Fig. 5c). By contrast, cell cycle progression of TCR-β+ DN3E thymocytes was independent of CD127. Consistent with the in vivo cell cycle data, CFSE dilution of thymocytes in the OP9 system was comparable when DN3E thymocytes cultured in the presence or absence of IL-7 (Fig. 2). However, DN3L and DN4 thymocytes cultured in the absence of IL-7 failed to proliferate as robustly as cells cultured in the presence of IL-7, although this trend did not reach statistical significance in all samples. In the in vitro differentiation system, thymocytes that reached the DP stage proliferated independently of IL-7, most likely because most DP thymocytes do not express CD127.

The conclusion that IL-7 could promote Bcl-2 expression in TCR-β+ DN thymocytes is consistent with this function of IL-7 in other thymocytes and T cells. For example, IL-7-deficient splenocytes had reduced levels of Bcl-2 that could be restored by treatment with IL-7.[32] In addition, thymocytes at stages before TCR-β expression are dependent on IL-7 for Bcl-2 expression.[32-34] Furthermore, CD127 and Bcl-2 protein levels both decline in wild-type mice as cells progress through the DN3 and DN4 stages,[2, 35, 36] suggesting a functional link between these proteins.

Whether IL-7 functions after TCR-β is expressed has been a source of controversy in the field of T-cell development. Arguments that IL-7 and CD127 do not function after expression of TCR-β may stem from the fact that CD127 protein levels are highest at the DN2 stage and then decline steadily as the cells proceed through development.[2, 37] This decline in CD127 expression appears necessary for normal T-cell development as transgenic expression of CD127 appears to deprive DN1 and DN2 thymocytes of the IL-7 needed for survival.[38] Despite the progressive decline in CD127 expression throughout early T-cell development, pre-TCR-mediated signalling enhances CD127 expression, suggesting that some IL-7 signalling is important.[2, 23]

Contributing to the controversy are data suggesting that adding IL-7 to total DN4 thymocytes does not activate signal transducer and activator of transcription 5 (STAT5) or promote survival as readily as adding IL-7 to DN1, DN2 and DN3 thymocytes.[34, 39] This reduction in the responsiveness of IL-7 within the DN4 stage can be interpreted to mean that IL-7 does not function at this stage. More likely, the reduction in survival and STAT5 phosphorylation reflects the lower CD127 expression in the DN4 stage, compared with the previous stages. Our data suggest that the survival and STAT5 phosphorylation seen in DN4 thymocytes is important for proper development.

Another model for the function of IL-7 in DN thymocytes is that IL-7 may block differentiation into the DP stage. Several reports, including this paper, have shown that adding IL-7 to the OP9 in vitro differentiation system reduces the percentage of thymocytes harvested that are DP thymocytes.[2, 24, 30, 31] Yu et al.[37] illustrated that IL-7 can inhibit transcription factors that restrict DN thymocyte differentiation. Our conclusion that IL-7 promotes the survival of TCR-β+ DN thymocytes is not mutually exclusive with a model indicating that IL-7 may slow differentiation. In fact, it is quite possible that the function of IL-7 is to promote survival of TCR-β+ DN and ISP thymocytes while slowing differentiation so that proliferation can occur optimally. In the in vitro differentiation system, we consistently found a loss of ISP thymocytes in cultures lacking IL-7 (Fig. 1). This loss of ISP cells may reflect a combination of accelerated differentiation with reduced survival of this population.

To conclude that IL-7 functions in TCR-β-expressing DN thymocyte subsets, these thymocyte populations must encounter IL-7 during the course of their development. Thymocytes enter the thymus at the corticomedullary junction and migrate outward through the cortex as they progress through the DN stages. Because IL-7 promoter activity can be detected in both medullary and cortical thymic epithelial cells,[40-42] it is highly likely that CD127-expressing thymocytes can be activated by IL-7 produced by nearby thymic epithelial cells. Hence, cell localization is not likely to prohibit the biological relevance of IL-7 in TCR-β+ DN thymocytes.

Although our data clearly indicate that signalling through the IL-7 receptor is required for survival of TCR-β+ DN thymocytes, we are unable to eliminate a potential role for thymic stromal lymphopoietin (TSLP) at this stage of development. The IL-7 receptor consists of CD127 and CD132 (γc) while the TSLP receptor consists of CD127 and TSLP receptor (TSLPR). Mice lacking TSLPR have no clear defect in thymocyte number or population distribution.[43, 44] However, mice lacking both CD132 and the TSLPR have a more severe phenotype than mice lacking CD132 alone,[44] suggesting that TSLP can compensate for a lack of IL-7 signalling. Consistent with overlapping functions of IL-7 and TSLP, the addition of TSLP to fetal thymic organ culture can increase the number of thymocytes recovered.[45] Further, transgenic expression of TSLP can rescue T-cell development in IL-7-deficient mice.[46] Hence, it is likely that IL-7 and TSLP can cooperate in promoting survival of TCR-β+ DN thymocytes. Our data using CD127−/− mice are consistent with a role for either IL-7 or TSLP because both cytokines can induce STAT5 phosphorylation and STAT5 promotes Bcl-2 expression.[47-51]

In conclusion, we demonstrated that IL-7 and CD127 continue to function in DN thymocytes beyond the stage at which TCR-β is expressed. Using an in vitro model, we showed that IL-7 promoted the survival of TCR-β+ DN thymocytes, but not DP thymocytes. Then, we used CD127−/− mice to show that CD127−/− TCR-β+ DN subsets had reduced Bcl-2 expression and cell cycle progression.

Acknowledgements

The authors would like to thank Drs Steve Benedict and Marci Chan and members of the Benedict laboratory for helpful discussions. We also thank Dr Benedict for critical review of this manuscript. We also acknowledge the University of Kansas Medical Center Flow Cytometry Core Laboratory. This work was supported by National Institutes of Health COBRE Program of the National Center for Research Resource RR016443 and American Cancer Society Research Scholar Grant 08-182-LIB.

Disclosures

The authors declare no competing financial interests.

Ancillary