Identification of alternatively spliced Il7r transcripts in mouse T cells that encode soluble IL-7Rα

All lymphocytes depend on the cytokine IL-7 for their development and homeostasis, and cell surface expression of the IL-7 receptor is a prerequisite for responses to IL-7.¹ Notably, the IL-7 receptor α-chain (IL-7Rα) does not only exist as a membrane-bound form but is also found as a soluble protein in human serum.^2,3 The exact role of soluble IL-7Rα (sIL-7Rα) proteins remains to be determined. However, increased amounts of sIL-7Rα have been associated with an increased risk of inflammation and autoimmunity.^3,4 Mechanistically, sIL-7Rα proteins can be generated by alternative splicing of Il7r gene transcripts that encode IL-7Rα. Curiously, however, such Il7r splice isoforms have only been described in humans and have never been described in mice.⁵ Consequently, the importance of Il7r alternative splice products that are not conserved across species has remained questionable. Here, we report the unexpected identification of an alternatively spliced Il7r transcript in mouse T cells that is distinct in its splice mechanism compared to that of human splice isoforms, but that could potentially produce soluble IL-7Rα proteins. Thus, the generation of IL-7Rα mRNA splice isoforms is a mechanism that is conserved between humans and mice. Assessing its physiological significance will lead to new insights into T cell immunology.

The functional IL-7 receptor is comprised of the common γ-chain (γc, CD132) and the IL-7-proprietary IL-7 receptor α-chain (IL-7Rα, CD127), whose expression is dynamically regulated during lymphocyte differentiation.¹ Both transcriptional and posttranscriptional mechanisms control IL-7Rα expression.⁶ In T cells, transcription factors such as FoxO1 and Runx3 were found to induce IL-7Rα expression, while the nuclear factor Gfi1 suppresses IL-7Rα transcription.⁷ In humans, T cells produce an alternatively spliced form of the Il7r transcript, in which exon 6, which encodes the entire IL-7Rα transmembrane region, is omitted and exon 5 is directly spliced to exon 7 (Fig. 1a).⁵ The resulting protein lacks the transmembrane domain (TD) and is produced as a soluble form that is secreted into serum (Fig. 1a, top). Indeed, sIL-7Rα proteins are found in varying amounts in human serum,² and serum sIL-7Rα levels have been linked with an increased risk for autoimmunity and viral infections.³ In mouse T cells, however, Il7r splice isoforms have not been reported. Therefore, attempts to understand the physiological role of secreted IL-7Rα proteins in animal models have been scarce. Our current study reports the first identification of an alternatively spliced Il7r isoform in mice.

Fig. 1.

Alternative splicing of Il7r in mouse T cells. a Schematic showing the primer design and alternative IL-7Rα pre-mRNA splice isoforms in human and mouse T cells (ED, extracellular domain; TM, transmembrane; ID, intracellular domain). b RT-PCR with primer1/primer2 and primer1*/primer2* identified two Il7r splice isoforms in the cDNA of human and mouse T cells, respectively. c RT-PCR with primer pairs for monkey and pig cDNA amplification demonstrates the presence of Il7r alternative splice products in monkey and pig PBMCs. d Schematic of IL-7Rα mRNA splice isoforms in mouse T cells. Alternative splicing produces an Il7r transcript that retains intron 5, creates an open reading frame shift and replaces the transmembrane and intracellular domains with a new 18-amino acid C-terminal epitope. e cDNA corresponding to Il7r-intron 5+ was tagged with an N-terminal FLAG epitope, and its bacterial expression was induced by IPTG. Whole-cell lysate and purified proteins were solubilized in urea, resolved by SDS-PAGE and stained with Coomassie blue. f Production of sIL-7Rα proteins by transducing Sf9 insect cells with baculoviruses expressing the Il7r-intron 5+ cDNA. Culture supernatants were collected and resolved by SDS-PAGE before immunoblotting with anti-IL-7Rα ED antibodies. g Total RNA from WT and sIL-7RαTg thymocytes was assessed for membrane IL-7Rα (Il7r) and soluble IL-7Rα (Il7r-intron 5+) transcripts by quantitative RT-PCR. The gene expression values were normalized to those of Rpl13 and quantitated in comparison to the values in WT cells. h The abundances of serum sIL-7Rα proteins were determined using commercially available mouse sIL-7Rα ELISA kits (Rockland Immunochemicals). The white, gray, and dark gray circles in column 1 correspond to 0, 1.5, and 3.0 ng/ml recombinant IL-7Rα protein, respectively. Columns 2–15 correspond to sera from the indicated mice. The black circles indicate serum samples. The gray area corresponds to the O.D. value range that was considered to be below the detection limits of the ELISA kit

Using RT-PCR primers that flank either the human or mouse Il7r exon 6, we confirmed the generation of alternatively spliced IL-7Rα pre-mRNA transcripts (Il7r-Δexon 6) exclusively in human T cells (Fig. 1a, b and Supplementary Materials). The 200-bp Il7r-Δexon 6 splice isoform was also found in PBMCs of monkeys and pigs, but it was conspicuously absent in mice (Fig. 1b, c). Curiously, however, we found that mouse T cells expressed a novel 400-bp PCR product that was not present in pig, monkey, or human cells (Fig. 1b, c). Molecular cloning and DNA sequencing identified this PCR product as a novel Il7r splice variant that was specific to mice and contained exon 6 and a part of intron 5 (Fig. 1a, d). Consequently, the RT-PCR products amplified by primers1* and -2* resulted in two distinct bands; the 300-bp product represents the conventional membrane form of IL-7Rα, while the larger 400-bp product corresponds to a novel splice isoform that is only found in mouse immune cells (Fig. 1b). Because the newly identified splice product contains a part of intron 5 (Fig. 1d), we refer to this isoform as “Il7r-intron 5+”.

It was not clear whether Il7r-intron 5+ transcripts encode a protein product; thus, we examined its open reading frame. We found that the continued translation of exon 5 through the intron 5 cDNA would result in the generation of an 18-amino acid neoepitope, followed by a stop codon. As a result, Il7r-intron 5+ encodes a new protein that comprises the entire extracellular domain in addition to an 18-amino acid tail but that lacks a TD (Fig. 1d). Consequently, the product of Il7r-intron 5+ would correspond to a soluble protein of 30 kDa. To test whether the alternatively spliced Il7r isoform can be expressed as a stable protein, we subcloned the Il7r-intron 5+ cDNA into bacterial expression vectors and assessed the IPTG-induced expression of the recombinant protein (Fig. 1e and Supplementary Materials). SDS-PAGE analysis confirmed successful expression of recombinant Il7r-intron 5+ products of the expected size, further bolstering the assumption that mouse T cells can generate alternatively spliced products of Il7r that encode sIL-7Rα proteins.

To determine the presence of sIL-7Rα in mouse serum, we utilized commercially available ELISA kits. Standard curves were established using recombinant mouse IL-7Rα proteins, and we found that the assays can detect IL-7Rα proteins at a low limit of 0.2 ng/ml. The actual assays of wild-type C57BL/6 and BALB/c mouse serum, however, did not reveal any meaningful signals, and we failed to detect sIL-7Rα in serum (Fig. 1h). Hence, we considered two possibilities: either the concentration of serum sIL-7Rα is lower than the detection limit or the Il7r-intron 5+ protein product is not expressed and not secreted. Experimental transfection of the Il7r-intron 5+ cDNA into Sf9 insect cells, however, demonstrated that mouse sIL-7Rα can be successfully expressed and secreted into culture supernatants, as confirmed by immunoblot analysis (Fig. 1f and Supplementary Materials). Encouraged by these results, we attempted to generate transgenic mice that would overexpress sIL-7Rα proteins (sIL-7RαTg), which we achieved by placing the Il7r-intron 5+ cDNA under the control of a human CD2 mini-cassette. In these mice, the transgene is expressed in all T cells, and we confirmed the overexpression of sIL-7Rα mRNA in T cells by qRT-PCR (Fig. 1g). Disappointingly, however, the immune phenotyping of sIL-7RαTg mice did not show any noticeable differences in the cell numbers and frequencies of T cell subsets (Fig. S1), and we still failed to detect any significant amounts of sIL-7Rα proteins in the serum of these mice (Fig. 1h, lane 5). These results support a scenario in which sIL-7Rα mRNA can be robustly expressed, but the protein products are unstable and cannot persist or be secreted into serum. In agreement with this idea, we also failed to detect sIL-7Rα proteins in serum from autoimmune and autoinflammatory mice, including Foxp3-deficient scurfy mice and EAE-induced mice that contain large numbers of inflammatory T cells.^8,9 Thus, the Il7r-intron 5+ alternatively spliced product encodes but likely does not produce soluble forms of the IL-7Rα protein.

We remain committed to establishing methods other than ELISA to detect potential sIL-7Rα proteins in mice. However, even if the protein is detected, the minute amounts of sIL-7Rα protein would place the physiological significance of this product into question. In fact, we consider it unlikely that such small amounts of sIL-7Rα protein could exert any agonistic or antagonistic effects on IL-7Rα signaling or serve as a reservoir for a biologically relevant amount of IL-7.^2,3 Thus, these results prompted us to reassess the physiological implication of the Il7r alternatively spliced product. IL-7 is expressed by stromal cells and not by T cells themselves, and its availability in vivo is very limited.¹⁰ T cells are constantly competing for binding to IL-7 to survive, and the suppression of IL-7Rα expression by IL-7 signaling is a critical mechanism to maximize the in vivo availability of IL-7.⁷ In this regard, the alternative splicing of Il7r pre-mRNA would effectively reduce the amount of membrane IL-7R protein and further limit IL-7 consumption by individual T cells. Consequently, here, we hypothesize that the main purpose of Il7r alternative splicing would be to reduce the abundance of membrane IL-7Rα proteins rather than to generate soluble IL-7Rα proteins.

Altogether, these results indicate that alternative splicing of Il7r is a conserved process in humans and mice, but that mouse T cells utilize a different splicing mechanism to achieve the same goal. The detailed molecular mechanisms that control the alternative splicing of Il7r in T cells upon their activation and differentiation await further investigation.

Author contributions

H.Y.W., Y.J., J.A.S., and C.H. performed experiments, analyzed data, and generated figures. C.H. and J.H.P. directed the study and wrote the paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version of this article (10.1038/s41423-020-0409-8) contains supplementary material.