Alternative Splicing of the Inhibitory Immune Checkpoint Receptor SLAMF6 Generates a Dominant Positive Form, Boosting T-cell Effector Functions

The SLAM family of immune receptors (SFR) comprises nine genes and is considered part of the CD2 cluster of the immunoglobulin superfamily (1). Most members of this group are self-binding receptors engaged in homotypic interactions; they cluster at the immune synapse (2) and serve as collectivity sensors (3). It is characteristic of SFRs that at least one unique immunoreceptor tyrosine-based switch motif (ITSM) is present in their cytoplasmic portion (4). The amino acid sequence motif TxYxxV/I can recruit the SH2-homology domain-containing adaptor proteins SAP and EAT2 (5, 6), or protein phosphatases SHP1, SHP2, and SHIP (7). The capacity of SFRs to acquire two types of adaptors, one that recruits kinases and one that dephosphorylates signaling molecules, led to the idea that these receptors have a dual function (2). The duality concept was corroborated by studies of inborn X-linked lymphoproliferative disease (XLP), in which a mutation in SAP leads to severe immune dysregulation (8).

In contrast to the paradigm of a bidirectional modulatory role of SFRs, we recently reported that genetic deletion of Slamf6 in murine CD8⁺ T cells significantly enhances their functional capacity (9). We showed that the SLAMF6 knockout (KO) T cells demonstrate a global improvement of effector function, including cytokine release, cytotoxicity, and posttransfer persistence. Interestingly, activated SLAMF6 KO cells express low to null levels of SAP protein, whereas SHP1 is intact. We concluded that SLAMF6 is an obligatory negative checkpoint receptor and suggested that the acquisition of SAP is required for the inhibitory effect. In contrast, we saw no indication that SHP1 is necessary for SLAMF6-mediated coinhibition.

The main limitation to extrapolating from the functional data generated with murine SLAMF6 (also known as Ly108) to the human receptor relates to a different splicing pattern. The two murine SLAMF6 splice variants occur in the cytoplasmic region. One splice variant (Ly108–1) contains one tyrosine-based motif, and the other (Ly108–2) bears two motifs (10). The splice variant with one tyrosine-based motif is associated with a propensity to develop autoimmune systemic lupus erythematosus (11). In contrast to mouse SLAMF6, splicing of the human SLAMF6 transcript occurs in the extracellular part of the receptor: the canonical sequence includes eight exons, the isoform SLAMF6^Δ17–65 lacks part of exon2, and SLAMF6^Δ18–128 is an isoform in which the entire of exon2 is skipped.

The fact that the human SLAMF6 ectodomain, the critical part for signal initiation, appears in molecular configurations that do not exist in mice prompted us to study human SLAMF6 splice variants. Surprisingly, we found that although the canonical SLAMF6 acted as a coinhibitory receptor, SLAMF6^Δ17–65, which existed as a protein in T cells, functioned as a strong costimulatory agonist. By producing T cells that did not express the canonical SLAMF6, we were able to study in-depth the molecular events promoted by the shorter isoform. On the basis of those data, we developed a method based on splice-switching antisense oligonucleotides (ASO), that can be implemented in the future for adoptive cell therapy of melanoma.

pcDNA3.1⁺/C-(K)DYK-SLAMF6 transcript isoforms were purchased from Genscript (OHu04772, OHu04774, OHu4776, and the empty vector).

Plasmids for expressing each SLAMF6 isoform were produced according to the following procedure: the Flag tag was inserted by fusion PCR between the signal sequence and the rest of the open reading frame (ORF) of the SLAMF6 isoform as described here: the two fused PCR sequences were amplified in SensQuest lab cycler machine (Danyel Biotech) with the following pairs of primers:

• VAR-5: TATATTCTAGACCATGTTGTGGCTGTTCCAATCG

• VAR1–3: TATATGAATTCTTACACGACATTGTCAAGGGC

Each PCR sequence was cut with XbaI (NEB, catalog no. R0145T) + EcoRI (NEB, catalog no. R0101T) and then cloned into NheI (NEB, catalog no. R0131M)/EcoRI-cut lentiviral vector pLL3.7 (Addgene catalog no. 11795). Sequencing results indicated that the genes were cloned as designed, and no nonspecific mutations were present (Supplementary Fig. S1).

For the control empty vector, the pLL3.7 vector was cut with NheI + EcoRI and treated with Klenow fragment enzyme followed by self-ligation. The sequencing result indicated that the EGFP from the vector was deleted.

pSpCas9(BB)-2A-GFP plasmids for SLAMF6 KO were produced according to the Zhang protocol (https://www.addgene.org/crispr/zhang/) including all the reagents purchased from Addgene, using the following single-guide RNA (sgRNA) guides:

• Construct 1: Fw: CACCGAGAATCCCGTTCACCATCAA; Rev: AAACTTGATGGTGAACGGGATTCT

• Construct 2: Fw: CACCGAGAAGACAAACAGGAGCGATGTT; Rev: AAACAACATCGCTCCTGTTTGTCTTCT.

For flow cytometry, cells were labeled with the following reagents: anti-SLAMF6 (NT-7), anti-CD45RA (HI100), anti-CCR7 (G043H7), anti-CD62 L (DREG-56), anti-CD8 (RPA-T8), anti-TNFα (Mab11), anti-CD244 (C1.7), anti-CD48 (BJ40), anti-CD229 (HLy-9.1.25), anti-CD84 (CD84.1.21), anti-CD4 (RPA-T4) were all obtained from BioLegend. Anti-SLAMF6 (REA) was from Miltenyi Biotec, anti-IFNγ (4S.B3) from Biogems, anti-FLAG (F3165) from Sigma-Aldrich, and anti-CD150 [A12(7D4)] from eBioscience.

For immunoblotting, proteins were detected with anti-SLAMF6 (goat, AF1908, R&D Systems), anti-SAP (rat, 1A9, BioLegend), anti-SHP1 (rabbit, Y476, Abcam), and anti-β actin (mouse, sc-47778, Santa Cruz Biotechnology). Secondary antibodies—goat anti-rabbit (111–035–003), donkey anti-rat (712–035–150), goat anti-mouse (115–035–146), and donkey anti-goat (705–035–003)—were all from Jackson ImmunoResearch.

Soluble ectodomain (se)SLAMF6, a soluble polypeptide consisting of the ectodomain sequence of SLAMF6, was purchased from Novoprotein (today BonOpus; catalog no. C387). seSLAMF6^Δ17–65, a soluble polypeptide consisting of the ectodomain sequence of SLAMF6^Δ17–65 isoform, was a custom protein production by BonOpus. Protein sequence: MLWLFQSLLFVFCFGPVPHETKSPEIHVTNPKQGKRLNFTQSYSLQLSNLKMEDTGSYRAQISTK TSAKLSSYTLRILRQLRNIQVTNHSQLFQNMTCELHLTCSVEDADDNVSFRWEALGNTLSSQPNL TVSWDPRISSEQDYTCIAENAVSNLSFSVSAQKLCEDVKIQYTDTKMHHHHHH *.

The cell line 526mel (HLA-A2⁺/MART-1⁺gp100⁺, created in 1987) was a gift from M. Parkhurst in 2004 (Surgery Branch, NCI, NIH, Rockville, MD). The cells were cultured in complete RPMI medium: RPMI1640 (Gibco, catalog no. 21875–034) supplemented with 10% heat-inactivated FCS (Gibco, catalog no. 12657–029), 2 mmol/L l-glutamine (Gibco, catalog no. 25030–024), and 1% combined antibiotics—penicillin/streptomycin (Gibco, catalog no. 15140–122). All lines were regularly tested and were Mycoplasma free. Cells were authenticated once a year using flow cytometry markers. Cells grew in culture for a week (2–3 passages) before use.

526mel human melanoma cell lines were transfected with the pcDNA3.1⁺/C-(K)DYK-neomycin-SLAMF6 transcript isoforms using Lipofectamine 2000 (catalog no. 11668027, Thermo Fisher Scientific, according to the manufacturer instructions). Neomycin (Gibco, catalog no. 11811–031)-resistant melanoma cells were subcloned in melanoma medium, as described above for 7 days in concentration of 1.2 mg/mL, and the stably transfected cells were labeled with SLAMF6 Ab and sorted in an ARIA-III sorter as described below (Flow cytometry section). Cells transfected with the empty vector were subcloned using neomycin resistance.

The Hek293 cell line was a gift from the Surgery Branch of NCI, NIH (Rockville, MD) in 2004. The cells were cultured in DMEM (Gibco, catalog no. 41965–039) supplemented with 10% heat-inactivated FCS, 2 mmol/L l-glutamine, and 1% combined antibiotics as described above. The line was regularly tested and was Mycoplasma free. Cells grew in culture for a week (2–3 passages) before use.

Hek293 cells were transfected with the pLL3.7-FLAG-SLAMF6 transcript isoforms or the empty control vector using Lipofectamine 2000 (Thermo Fisher Scientific). Neomycin-resistant Hek293 cells were subcloned, and the stably transfected cells were labeled with SLAMF6 Ab and sorted in an ARIA-III sorter. Cells transfected with the empty vector were subcloned using neomycin resistance. The protcols for transfection and subcloning were as above (in Aberrant SLAMF6 splice variant expression on melanoma cells).

The Jurkat cell line was purchased from ATCC in 2014. The cells were cultured in RPMI1640 supplemented with 10% heat-inactivated FCS, 2 mmol/L l-glutamine, and combined antibiotics. The line was regularly tested and was Mycoplasma free. The cells are authenticated once a year using flow cytometry markers. Cells grew in culture for a week (2 passages) before use.

SLAMF6 KO Jurkat cells were generated using CRISPR-Cas–mediated genome editing, using the pSpCas9(BB)-2A-GFP plasmid with the sgRNA guides described above. Jurkat cells were washed twice in RPMI medium without supplements and resuspended in 10 × 10⁶ cells/mL RPMI. 5 × 10⁶ Jurkat cells with 5 μg pSpCas9(BB)-2A-GFP SLAMF6-CRISPR plasmid were electroporated in Bio-Rad 0.4-cm cuvettes using an ECM 630 Electro Cell manipulator (BTX Harvard apparatus) at 260 V, 975 μF, 1575 Ω. After electroporation, cells were immediately seeded into complete RPMI medium. Forty-eight hours after transfection, cells expressing GFP were selected by sorting (ARIA-III sorter). Cells lacking human SLAMF6 were selected by single-cell sorting in 96 wells using SLAMF6 Ab and cultured for several weeks until single-cell colonies grew (without antibiotic selection). The genomic DNA from each colony was sequenced to confirm the mutation.

The T2 cell line was a kind gift from the Surgery Branch in 2004. The cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated fetal calf serum (FCS), 2 mmol/L L-glutamine, and combined antibiotics. The line was regularly tested and was mycoplasma free. The cells are authenticated once a year using flow cytometry markers. Cells grew in culture for a week (2 passages) before use.

T2 cells expressing perturbated SLAMF6 isoforms were generated using CRISPR-Cas–mediated genome editing, using the pSpCas9(BB)-2A-GFP plasmid with the sgRNA, as described above (in SLAMF6 KO Jurkat cells).

Peripheral blood mononuclear cells (PBMC) were purified from buffy coats of healthy donors (Hadassah Blood Bank) or purified from samples taken from patients undergoing ICB treatment after being signed on a written informed consent (Helsinki approval 383–23.12.05), the study was performed after approval by Haddasah institutional review board (IRB). Blood samples from patients with metastatic melanoma were obtained at three time points: once before starting immune checkpoint blockade (ICB) treatment and twice during the treatment, three months apart. The fresh cells were sorted into four CD8⁺ or CD4⁺ subpopulations using an ARIA-III sorter, based on staining for CD8 or CD4, CCR7, and CD45-RA markers (staining procedure is detailed below in flow cytometry section). For each subset, RNA was extracted, and quantitative RT-PCR for SLAMF6 isoforms was performed, as described below (in RT-PCR and qRT-PCR). All patients gave their informed consent to participate in this study.

Fresh tumor specimens taken from resected metastases of patients with melanoma at the point the tumor was evaluated as stage III melanoma at diagnosis were used to release tumor-infiltrating lymphocytes (TIL) using a microculture assay (based on the Helsinki approval no. 395–16.09.05, Use of tumor cell lines or mononuclear cells from peripheral blood or tissue samples or serum from cancer patients, collected in the past, for immunologic research, vaccination, and immune monitoring studies; ref. 12). Human lymphocytes were cultured in complete medium consisting of: RPMI1640 supplemented with 10% heat-inactivated human AB serum (Valley Biomedical Inc., catalog no. HP1022), 2 mmol/L l-glutamine, 1 mmol/L sodium pyruvate (Gibco, catalog no. 11360–039), 1% nonessential amino acids (Gibco, catalog no. 11140–035), 25 mmol/L HEPES (pH 7.4; Biological Industries, catalog no. 03–025–113), 50 μmol/L 2-ME (Gibco, catalog no. 31350–010), and combined antibiotics as described above. The complete medium was supplemented with 6,000 IU/mL recombinant human IL2 (rhIL2, Chiron). Cells grew in culture for 1 day before use.

Fourteen days after TIL culture, cells were washed with PBS, resuspended in PBS supplemented with 0.5% BSA, and stained with FITC-conjugated HLA-A*0201/MART_126–135 (Immudex, catalog no. WB2162) or HLA-A*0201/gp100_209–217 dextramer (Immudex, catalog no. WB2158) for 30 minutes at 4°C. Lymphocytes were then incubated with allophycocyanin-conjugated mouse anti-human CD8 (RPA-T8, eBioscience, catalog no. 17–0088–42) for an additional 30 minutes at 4°C and washed. CD8⁺ T lymphocytes, positively stained by dextramer, were sorted with a BD FACS-Aria and directly cloned at one or two cells per well in 96-well plates in the presence of anti-CD3 (30 ng/mL, OKT3, eBioscience, catalog no. 14–0037–82), rhIL2 (6,000 IU/mL), and 4 Gy–irradiated 5 × 10⁴ allogeneic PBMCs as feeder cells. Five days later, rhIL2 (6,000 IU/mL) was added and renewed every 2 days. On day 14, the clones were assayed for IFNγ secretion (as described below in Intracellular cytokine staining) in a peptide-specific manner following their coincubation with T2 cells pulsed with MART-1_26–35 or gp100_209–217 [both commercially synthesized and purified (>95%) by reverse-phase HPLC by Biomer Technology] by ELISA (R&D Systems). The MART-1_26–35 or gp100_209–217–reactive clones were further expanded in a second-round exposure to anti-CD3 (30 ng/mL), and rhIL2 (6,000 IU/mL) in the presence of a 50-fold excess of irradiated feeder cells.

Nude (athymic Foxn1^−/−) mice were purchased from Harlan Laboratories.

RNA was isolated from cells using the GenElute Mammalian Total RNA kit (Sigma-Aldrich, catalog no. RTN70) according to the manufacturer's protocol (for more than 4 × 10⁶ cells) or Quick-RNA MicroPrep kit (Zymo Research, catalog no. D7001) according to the manufacturer's protocol (for fewer than 1 × 0⁶ cells). RNA was then transcribed to cDNA using the qScript cDNA Synthesis kit (Quantabio, catalog no. 95047–500) according to the manufacturer's instructions. RT-PCR was performed using 100 ng cDNA in 50 μL reaction in the SensQuest Lab Cycler Machine (Danyel Biotech). The products were then run on 1.5% agarose gel. RT-PCT was performed using the primers:

• hSlamf6 Fw: GCGGAAAGCATGTTGTGGCTG; hSlamf6 Rev: GGAGACAGTGAGGTTTGGCTG. hSh2d1a Fw: AGCTATTTGCTGAGGGACAG; hSh2d1a Rev: TTATGTACCCCAGGTGCTGT.

• hGapdh Fw: AGAACGGGAAGCTTGTCATC; hGapdh Rev: TTGATTTTGGAGGGATCTCG used for noramalization.

• mSlamf6 Fw: AGCTTATGAAAGAATGGCTGTC; mSlamf6 Rev: GATAGTGACATTTGGTCCTCC.

qRT-PCR was performed in triplicates using 2 μL of template in 10 μL reaction in the CFX384 real-time system (Bio-Rad) using Bio-Rad CFX Manager 3.0 software. Signals were quantified using Image Studio Lite software (LI-COR Biosciences). To determine gene ezpression the calculation of 2^–ΔΔC_t was used. qRT-PCR was performed using the primers: hSlamf6 Fw: CTGTTCCAATCGCTCCTGTT; hSlamf6 Rev: GGGGTTAAGCTGCTTTGTGA. hSLAMF6^Δ17–65 Fw: CTGTTCCAATCGCTCCTGTT; hSLAMF6^Δ17–65 Rev: CAGGGAGTAGGACTGGGTGA. hIl2 Fw: AAGTTTTACATGCCCAAGAAGG; hIl2 Rev: GATATTGCTGATTAAGTCCC. hTcf7 Fw: CCAAGTACTATGAGCTGGCC; hTcf7 Rev: CCTCGACCGCCTCTTCTTC. hRunx3 Fw: TCATGAAGAACCAGGTGGCC; hRunx3 Rev: ATGGTCAGGGTGAAACTCTT. hTox Fw: TTTGACGGTGAGAACATGTA; hTox Rev: GAATGTTGAAGTCTTCACTTT. hEomes Fw: TCTTCTTGGATAGAGACACC; hEomes Rev: GCCTTCGCTTACAAGCACTG. hTbx21 Fw: AACACGCATATCTTTACTTT; hTbx21 Rev: TCAATTTTCAGCTGAGTAAT. hBcl6 Fw: TGGCCTGTTCTATAGCATCT; hBcl6 Rev: TACATGAAGTCCAGGAGGAT. hId2 Fw: GTGAGGTCCGTTAGGAAAAA; hId2 Rev: GTTCATGTTGTATAGCAGGCT. hC-jun Fw: ATCAAGGCGGAGAGGAAGCG; hC-jun Rev: TGAGCATGTTGGCCGTGGAC. hGata3 Fw: TGTGGGCTCTACTACAAGCTTCAC; hGata3 Rev: GCTAGACATTTTTCGGTTTCTGGT. hC-fos Fw: CTGGCGTTGTGAAGACCAT; hC-fos Rev: TCCCTTCGGATTCTCCTTTT. hHprt Fw: GAGGATTTGGAAAGGGTGTTT; hHprt Rev: CATCTCGAGCAAGACGTTCA used for normalization.

TILs (1 × 10⁵) were cocultured for 6 hours at 37°C at a 1:1 ratio with the indicated target melanoma cells. After 2 hours, brefeldin A (BioLegend, catalog no. 420601) was added (1:1,000). After incubation, the cells were washed twice with PBS and stained with anti-CD8 (RPA-T8, BioLegend, catalog no. 301049) for 30 minutes at room temperature. Following fixation and permeabilization (eBioscience, catalog no. 88–8824–00 according to the protocol at https://www.thermofisher.com/il/en/home/references/protocols/cell-and-tissue-analysis/protocols/staining-intracellular-antigens-flow-cytometry.html), intracellular IFNγ and TNFα were labeled with anti-IFNγ and anti-TNFα antibodies (BioLegend), respectively, for 30 minutes at room temperature. Cells were washed with permeabilization buffer, resuspended in FACS buffer, and subjected to flow cytometry (as described below in Flow cytometry section).

1 × 10⁵ Jurkat cells (either WT or SLAMF6^Δ17–65 cells) were activated with 200 ng/mL PMA (Sigma-Aldrich, catalog no. P1585) and 300 ng/mL ionomycin (Sigma-Aldrich, catalog no. I0634) for 48 hours. At the end of the activation, the conditioned medium was collected, and IL2 or IFNγ secretion was detected by ELISA [R&D Systems, catalog no. DY202 (for IL2), or catalog no. DY285 (for IFNγ), according to the manufacturer's instructions].

Human lymphocytes (1 × 10⁵) were cocultured overnight at a 1:1 ratio with the indicated target cells. Conditioned medium was collected, and IFNγ secretion was detected by ELISA (R&D Systems, catalog no. DY285, according to the manufacturer's instructions).

siRNAs against SLAMF6, SHP1, and SAP were purchased from QIAGEN [siSLAMF6 (1) – Hs_SLAMF6_1 FlexiTube siRNA SI00147252, siSLAMF6 (2) – Hs_SLAMF6_2 FlexiTube siRNA SI00147259, siSAP (1) – Hs_SH2D1A_4 FlexiTube siRNA SI00036568, siSAP (2) – Hs_SH2D1A_3 FlexiTube siRNA SI00036561, siSHP1 (1) – Hs_PTPN6_5 FlexiTube siRNA SI02658726, siSHP1 (2) – Hs_PTPN6_6 FlexiTube siRNA SI02658733]. The cells were washed twice, resuspended in RPMI1640 medium without supplements, and then electroporated in Bio-Rad 0.2-cm cuvettes containing 20 μL of siRNA without further reagents using ECM 630 Electro Cell manipulator at 250 V, 300 μF, 1,000 Ω. After electroporation, the cells were immediately seeded into complete RPMI medium. Twenty-four hours postelectroporation, the cells were activated and evaluated as described above.

ASOs were designed to target the splicing site in exon2; the oligos were purchased from Microsynth, synthesized with a full (all nucleotides) phosphorothioate (PS) backbone, and in which each ribose 2′-hydroxyl was modified to 2′-methoxyethyl (2′-MOE). Oligos sequences are as follows: Scrambled ASO: 5′ UGACCGAAAAGUCAUCUCAA 3′. ASO: 5′ GGGUACUAUGAAGGCAAGAG 3′.

Jurkat cells were collected, washed twice, resuspended in RPMI1640 medium without supplements, and then electroporated in Bio-Rad 0.2-cm cuvettes containing 15 μL ASO without further supplements using ECM 630 Electro Cell manipulator at 250 V, 300 μF, 1,000 Ω. After electroporation, the cells were immediately seeded into complete RPMI medium and cultured for 24 hours before analysis.

ASOs were electroporated into PBMCs using the Human T cell nucleofector kit according to the manufacturer's directions (VPA-1002, Lonza).

209TIL cells (described above in Tumor-infiltrating lymphocytes) were electroporated according to the protocol (13). All cells were evaluated 24 hours postelectroporation.

Cells were lysed using RIPA buffer and protein concentrations were measured using Bradford quantification (Bio-Rad, catalog no. 5000205). Equal concentrations of lysates were resuspended in SDS sample buffer [250 mmol/L Tris-HCl (pH 6.8), 5% w/v SDS, 50% glycerol, and 0.06% w/v bromophenol blue] for 5 minutes at 95°C. Proteins were separated by SDS PAGE and transferred to a polyvinylidene difluoride membrane. Membranes, blocked with 1% milk, were incubated with primary antibodies (as listed in the Antibodies section) overnight at 4°C, followed by incubation with horseradish peroxidase (HRP)–conjugated secondary antibodies for 1 hour at room temperature (Jackson ImmunoResearch Laboratories, as listed in the Antibodies section). Signals were detected by enhanced chemiluminescence reagents (Clarity Western ECL Substrate, Bio-Rad, catalog no. 1705061) and quantified using Image Studio Lite software.

Cells were incubated with increasing concentrations from 25 to 100 μg/mL seSLAMF6 for 30 minutes on ice, washed twice, and labeled with goat anti-SLAMF6 (R&D Systems, AF1908), and a secondary Ab (donkey anti-goat, Jackson ImmunoResearch, catalog no. 705–035–003) on ice. Antibody binding was measured by flow cytometry.

seSLAMF6 and seSLAMF6^Δ17–65 were biotinylated using EZ Link Sulfo NHS Biotin (Thermo Fisher Scientific, catalog no. 21217) for 30 minutes at room temperature. Biotinylated proteins were collected by 15-minute centrifugation at 1,700 rcf in an Amicon Ultra-4 Centrifugal Filter Unit (Millipore) in 2 mL PBS.

A plate was coated overnight with 0.5 μmol/L seSLAMF6 or seSLAMF6^Δ17–65 in coating buffer (8.4 g NaHCO₃, 3.56 g Na₂CO₃ in 1 L DDW, pH 9.5), 4°C. The next day, the plate was washed and blocked with blocking buffer (1%BSA in PBS) for 1 hour at room temperature, and then biotinylated proteins (seSLAMF6 or seSLAMF6^Δ17–65) were added for 2 hours at room temperature. Streptavidin-HRP (R&D Systems, catalog no. 890803) was added for 1 hour at room temperature. Signals were detected by the addition of TMB/E substrate (Millipore, catalog no. ES001).

Cells were stained with antibodies at room temperature for 25 minutes, according to the manufacturer's instructions, washed, and analyzed using a CytoFlex (Beckman Coulter). For flow cytometry–based sorting, cells were stained with antibodies for 30 minutes on ice water, washed, and kept on ice until sorting. The sorting was performed in an ARIA-III sorter, and flow cytometry analysis for all experiments was performed using FCS Express 5 flow research edition (De Novo software).

Statistical significance was determined by unpaired t test (two-tailed with equal SD) using Prism software (GraphPad). A P value of <0.05 was considered statistically significant. Analysis of more than two groups was performed using a one-way ANOVA test. *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001. For each experiment, the number of replicates and the statistical test used are stated in the corresponding figure legend.

Datasets with the following accession numbers were downloaded from NCBI SRA: SRR7588547, SRR7588548, SRR7588549, SRR7588550, SRR7588559, SRR7588560, SRR7588561, SRR7588562, SRR7588567, SRR7588568, SRR7588573, SRR7588574, SRR7588579, SRR7588580. All from the Bioproject: PRJNA482654.

Poly-A/T stretches and Illumina adapters were trimmed from the reads using cutadapt; resulting reads shorter than 30-bp were discarded. Reads were mapped to the H. sapiens reference genome hg38 using STAR (14), supplied with gene annotations downloaded from RefSeq (and with EndToEnd option and outFilterMismatchNoverLmax set to 0.04). Isoform expression was analyzed with the RSEM version 1.3.0. The pipeline was run using snakemake. The genome browser JBrowse was used (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4830012/).

WT and SLAMF6^Δ17–65 Jurkat cells activated (0, 6, 12, 24, and 36 hours) in 96 flat bottom plates in the presence or absence of PMA (200 ng/mL) and ionomycin (300 ng/mL) were lysed in 100 μL lysis buffer containing TCL buffer (Qiagen, catalog no. 1031576) containing 1% β-mercaptoethanol. After lysis, plates were immediately frozen and stored at −80°C until samples were processed. Libraries were created using the Smart-Seq2 protocol as described previously (15), with minor modifications for bulk populations. For RNA purification, 10 μL of cell lysates for each sample was used. Forty microliters of Agencourt RNAClean XP SPRI beads (Beckman Coulter, catalog no. A63987) was added to the cell lysates. After 10 minutes of incubation at room temperature, plates were placed on a DynaMag 96 side skirted magnet (Invitrogen, 12027) for 5 minutes, followed by supernatant removal and 2 cycles of 75% ethanol washes. After ethanol removal, plates were incubated for 15 minutes or until beads started to crack, and 4 μL of Mix-I [containing: 1 μL (10 μmol/L) RT primer (DNA oligo) 5′–AAGCAGTGGTATCAACGCAGAGTACT30VN-3′; 1 μL (10 mmol/L) dNTPs; 1 μL (10%, 4U/μL) recombinant RNase inhibitor; and 1 μL nuclease-free water] was added and mixed in each well, following incubation at 72°C for 3 minutes. Next, 7 μL of Mix-II [containing: 0.75 μL nuclease-free water; 2 μL 5X Maxima RT buffer; 2 μL (5 mol/L) betaine; 0.9 μL (100 mmol/L) MgCl₂; 1 μL (10 μmol/L) TSO primer (RNA oligo) 5′-AAGCAGTGGTATCAACGCAGAGTACATrGrG+G-3′; 0.25 μL (40U/μL) recombinant RNase inhibitor; and 0.1 μL (200 U/μL) of Maxima H Minus Reverse Transcriptase], was added and mixed, and reverse transcription (RT) was performed at 50°C for 90 minutes, followed by 5-minute incubation at 85°C. After RT incubation, 14 μL of Mix-III [containing: 1 μL nuclease-free water; 0.5 μL (10 μmol/L) ISPCR primer (DNA oligo) 5′-AAGCAGTGGTATCAACGCAGAGT-3′; and 12.5 μL 2X KAPA HiFi HotStart ReadyMix] was added for the cDNA amplification step, performed at 98°C for 3 minutes, followed by 12 cycles at 98°C for 15 seconds, 67°C for 20 seconds, and 72°C for 6 minutes with a final extension at 72°C for 5 minutes. Next, two cycles for removal of primer dimer residues were performed by adding 20 μL of Agencourt AMPureXP SPRI beads (Beckman Coulter, catalog no. A63881), incubation of plates for 5 minutes at room temperature, followed by 5 minutes of incubation on a DynaMag 96 side skirted magnet, 2 cycles of 75% ethanol washes, and resuspension of dried beads with 50 μL of TE buffer (Tekanova, catalog no. T0228). After generating the cDNA, quality control was performed to evaluate (i) the concentration of each sample using the Qubit dsDNA high sensitivity assay kit (Life Technologies, catalog no. Q32854) and (ii) size distribution using the High Sensitivity DNA BioAnalyzer Kit (Agilent, catalog no. 5067–4626). Libraries were constructed using the Nextera XT Library Prep kit (Illumina, catalog no. FC-131–1096). Combined libraries were sequenced on a NextSeq 500 sequencer (Illumina) using the 75 cycles kit, with paired-end 38-base-reads and dual barcoding.

FASTQ files were aligned to the NCBI Human Reference Genome Build GRCh37 (hg19) using STAR (14). Principal component analysis (PCA) was performed using the DeSeq2 plot PCA function.

Differential expression was performed using DeSeq2 results function, with log fold change (lfc)threshold = 0.25 and AllHypothesis = “greaterAbs”. The Benjamini–Hochberg procedure corrected P values. Significant genes were chosen if they had adjusted P values <0.05. Differential expression was calculated either comparing each time point to its corresponding sample type at 0 hours or as differentially expressed genes between WT and Jurkat-SLAMF6^Δ17–65 samples for each time point. Expression values were normalized via the DESeq2 DESeq function. Next, each value was subtracted by the mean of its gene expression in all the samples. Immune-related genes were selected from the RNA-sequencing results based on a unified list of genes created from the Immunogenetic Related Information Source (IRIS) list and the MAPK/NFKB Network list (https://www.innatedb.com/redirect.do?go=resourcesGeneLists). Sashimi plots were generated with the IGV visualization tool (16).

Genes that were highly expressed compared with 0 hour for each cell type were analyzed by GeneAnalytics. The scores were calculated as the −log₂(P) (http://geneanalytics.genecards.org).

The cytotoxic score was calculated by using a method similar to that of Tirosh and colleagues (17) with the same naïve and cytotoxic genes chosen. The value for each gene was calculated as the average expression of three replicates of a sample type at a specific time point.

The processed gene expression data are deposited at Gene Expression Omnibus (GEO; accession number: GSE160817).

209TIL cells electroporated with scrambled ASO or ASO1 were washed and mixed at a 1:1 ratio (1 × 10⁶ cells each) with 526mel and immediately injected subcutaneously into the back of 8- to 9-week-old female nude (athymic Foxn1^−/−) mice (n = 7). Tumor size was measured in two perpendicular diameters three times per week. Mice were sacrificed when tumors reached a 15-mm diameter in one dimension or when the lesion necrotized. Tumor volume was calculated as L (length) × W (width)² × 0.5.

Animal studies were approved by the Institutional Review Board - Authority for biological and biomedical models, Hebrew University, Jerusalem, Israel (MD-20–16104–5).

The human SLAMF6 gene was cloned in 2001 and mapped to the SLAM gene cluster on Chr1q22 in 2002 (18, 19). The gene encodes a 331 amino acid sequence with four splice isoforms recorded on RefSeq (20). Isoform 1 is the canonical sequence composed of an immunoglobulin core with variable and constant domains (NCBI RefSeq: NM_001184714.2). Isoform 3 includes an alternative acceptor site, which consists of a 3′ alternative splicing of exon2, lacking amino acids 17–65 of the variable region (NM_001184715.2), referred to in this study as SLAMF6^Δ17–65. Isoform 4 is the result of exon2 removal, which leads to a protein missing the whole variable domain, referred to here as SLAMF6^ΔExon2 (NM_001184716.2; Fig. 1A and B). The transmembrane and intracellular domains of these three isoforms are identical. Using primers that complement sequences upstream and downstream of exon2 splice sites, we identified three bands, matching isoforms 1, 3, and 4, as referenced in RefSeq, by their correct size, in RNA extracted from PBMCs, Jurkat cells, and CD8⁺ TILs (Fig. 1C). We sequenced the bands to confirm their identity.

To confirm the existence of these isoforms, we analyzed a public RNA-sequencing bioproject (accession no. PRJNA482654; ref. 21). Using the RSEM package for quantifying isoform abundance, we identified all SLAMF6 RNA isoforms (Fig. 1D). We failed to identify exon2 splicing in the murine analogue of SLAMF6, Ly108, nor did we find data on murine isoforms in Ly108 extracellular domain from genome browsers (Supplementary Fig. S2A). In activated CD8⁺ T cells, SLAMF6 isoform 1, always expressed at higher levels compared with the shorter isoforms (ratio >1), was even further upregulated in effector and effector memory (EM) cells compared with naïve or central memory (CM) cells (Fig. 1E; Supplementary Fig. S2B and S2C). Flow cytometry confirmed that the expression of full-length SLAMF6 protein increased in parallel (Fig. 1F and G).

To systematically test for isoform expression during activation, we used RNA from 48-hour–activated Jurkat cells (Fig. 1H) and identified a general increase in SLAMF6 transcripts. We also tested SLAMF6 splicing in CD4⁺ T cells (Supplementary Fig. S2D–S2F). The results showed that the highest expression of SLAMF6 was found in EM and CM cells, but the ratio between the canonical isoform and the shorter isoforms remains >1 in all subsets, as in CD8⁺ T cells.

In conclusion, we found that the SLAMF6 splicing ratio varied in the CD8⁺ and CD4⁺ T-cell subsets, but, in all cases, the canonical SLAMF6 isoform was the dominant isoform expressed.

SLAMF6, a self-binding receptor, is expressed on hematopoietic cell lineages (1) but not on nonhematopoietic cells. To evaluate the self-binding properties of SLAMF6 isoforms in trans, we took advantage of the receptor's absence in somatic cells. We ectopically expressed isoform-1 (full-length), isoform-3 (Δ17–65), and isoform-4 (ΔExon2) in a melanoma cell line (Supplementary Fig. S3A and S3B). We cocultured the SLAMF6-expressing 526mel-SLAMF6 (full-length), 526mel-Δ17–65 and 526mel-ΔExon2 with cognate CD8⁺ TILs. Compared with the original melanoma cells, TILs cocultured with melanoma cells expressing canonical SLAMF6 secreted significantly less IFNγ (P < 0.001), whereas TILs cocultured with melanoma cells expressing SLAMF6^Δ17–65 secreted higher levels of IFNγ (P < 0.05). Aberrant expression of SLAMF6^ΔExon2, which lacks the binding interface of the receptor, had no effect on IFNγ secretion in the coculture assays (Fig. 2A). When measuring intracellular production of IFNγ and TNFα, we noted that the population of double-positive TILs, expressing both cytokines, was enriched when cocultured with melanoma cells expressing SLAMF6^Δ17–65 (Fig. 2B and C).

To support this observation, we established another model for trans activation using TAP-deficient T2 cells, which naturally express SLAMF6. Using CRISPR-Cas9–directed mutations of the SLAMF6 exon2 splicing site, we generated a single-cell clone with a higher expression of SLAMF6^Δ17–65 compared with the canonical isoform (Fig. 2D and E). When pulsed with a cognate melanoma peptide identified by three TILs, TILs cocultured with the T2 cells expressing higher levels of SLAMF6^Δ17–65 secreted more IFNγ than the TILs cocultured wild-type (WT) T2 cells (Fig. 2F).

Previous resports have shown that canonical SLAMF6 serves as an inducer of T-cell exhaustion and acts as an inhibitory checkpoint for T cells (9, 11, 22). The results presented here with the full-length SLAMF6 support this role. Surprisingly, SLAMF6^Δ17–65 seemed to be a counterbalance to full-length SLAMF6, acting as a costimulator in trans.

Next, we asked whether the isoforms pair together and if SLAMF6^Δ17–65 can cross-bind the canonical receptor.

To test the possibility that the canonical SLAMF6 associates with the extracellular domains of both the full-length molecule and of SLAMF6^Δ17–65, we used a binding competition assay with HEK293 cells ectopically expressing one isoform of SLAMF6 (Fig. 2G) and incubated the cells in the presence of seSLAMF6 in increasing concentrations to compete with a blocking antibody of SLAMF6, detectable by flow cytometry. The assay showed that seSLAMF6 could bind both isoforms expressed on HEK293 cells. Also, a quantitative ELISA was performed; seSLAMF6 and seSLAMF6^Δ17–65 served either as a plate-bound bait or as a secondary biotinylated protein detectable with HRP-conjugated streptavidin (Fig. 2H). The self-binding capacity of seSLAMF6^Δ17–65 was the strongest, whereas the self-binding capacity of seSLAMF6 was the weakest. The pairing of seSLAMF6 and seSLAMF6^Δ17–65 was stronger than SLAMF6 homotypic binding.

We concluded from these experiments that the ectodomains of SLAMF6 and SLAMF6^Δ17–65 cross-react, with the latter displaying a more robust binding capacity.

The opposing comodulatory effects of the canonical SLAMF6 and SLAMF6^Δ17–65 when expressed on target cells as ligands, the first inhibiting T cells, and the latter costimulating, raised the question of the contribution of each isoform when acting as a receptor. To address this issue, we generated T cells expressing only the short isoforms of SLAMF6 using CRISPR-Cas9 targeting exon2 in the part that is skipped-out in Δ17–65 and Δexon2 (Fig. 3A). Thus, only a transcript in which the mutated segment is skipped by alternative splicing will be appropriately translated. For stable transfectants, we used a single-cell colony generated from the Jurkat T cell line following CRISPR-Cas 9 editing, in which a frameshift mutation leading to a stop-codon occurred, and name the cells Jurkat-SLAMF6^Δ17–65 cells. Of note, SLAMF6^Δexon2 that showed no effect in trans, was not deleted, as there is no possibility to remove this isoform without eradicating SLAMF6^Δ17–65.

Using RNA sequencing (detailed below, activated SLAMF6^Δ17–65 Jurkat cells acquire a cytotoxic gene expression program), we identified the skipping of junction 2, which is the splicing site for the canonical SLAMF6. In accordance, we observed an increased prevalence of the alternative junction, joining together the codons encoding for amino acids 16 and 66 (Supplementary Fig. S4A). Quantitative analysis of SLAMF6 isoform expression in the cells showed that following CRISPR-Cas 9 editing, the transcript coding for the canonical SLAMF6 is decreased, probably due to the process of nonsense-mediated decay (Supplementary Fig. S4B; ref. 23).

Using antibodies targeting the variable and constant regions of SLAMF6 (Fig. 3B), we showed that the edited Jurkat cells did not express the canonical SLAMF6 receptor but had detectable levels of SLAMF6^Δ17–65 protein (Fig. 3C and D), which was expressed on the cell membrane (Fig. 3E). The expression of other SLAM family members remained unchanged (Supplementary Fig. S4C).

After verifying that SLAMF6^Δ17–65 exists as a viable protein on the cell surface, we evaluated the functional capacity of Jurkat-SLAMF6^Δ17–65 cells. Cytokine secretion, measured using IL2 transcript and ELISA following 2-day activation, showed a 3-fold increase in Jurkat-SLAMF6^Δ17–65 cells compared with the parental line (Fig. 3F and G; Supplementary Fig. S4D and S4E). Of note, Jurkat cells missing all SLAMF6 isoforms did not show an increased level of IL2, indicating that the splice isoform was acting on its own and not via the absence of the canonical SLAMF6 inhibitory receptor. The positive effect of SLAMF6^Δ17–65 was obliterated by siRNA designed to knock down the gene (Fig. 3H; Supplementary Fig. S4F). Although activated Jurkat cells did not secrete detectable levels of IFNγ, Jurkat-SLAMF6^Δ17–65 cells did produce IFNγ (Supplementary Fig. S4G), attesting for polyfunctionality.

To generate a comprehensive transcription map of Jurkat-SLAMF6^Δ17–65 cells, longitudinal RNA sequencing was performed following activation with PMA and ionomycin. We noted that the splice isoform induced some distinct traits in the lymphocytes (Fig. 4A). In a PCA to map cell populations in two-dimensional space (Fig. 4B), WT and Jurkat-SLAMF6^Δ17–65 cells were observed to cluster closely before activation but were substantially separated after activation. Shortly after activation, we noted that the expression of SLAMF7 was highly and constitutively expressed in the Jurkat-SLAMF6^Δ17–65 cells, at all time points; at several time points, SLAMF1 was also expressed. SLAMF7 is typically expressed in activated NK and CD8 T cells, but not in Jurkat or CD4 T cells. The de novo transcription of SLAMF7 was accompanied by a major rise in GZMB and NKG2D (KLRK1), delineating a transition from a CD4 to an NK/CD8 profile. A high NKG2D expression at the protein level was also noted in 72-hour–activated Jurkat-SLAMF6^Δ17–65 cells compared with WT (Fig. 4C). When combined with the increased expression of TNF, IFNG, IL2, CD30LG, and LIGHT, this profile gives rise to an improved cytotoxic score of the Jurkat-SLAMF6^Δ17–65 cells (Fig. 4D; ref. 17).

Analysis of differentially expressed genes coupled with GO term enrichment analysis was used to compare transcriptional programs of the lymphocytes (Fig. 4E and F). Jurkat-SLAMF6^Δ17–65 cells showed more substantial usage of NFkB and ERK signaling with a higher susceptibility to apoptosis than WT Jurkat cells. Lastly, Jurkat-SLAMF6^Δ17–65 cells had an intriguing transcription factor profile distinct from that of the parental line (Fig. 4A and G). Levels of TBX21, RUNX3, C-JUN, and TCF7, which typify effector T-cell subsets, were significantly higher in activated Jurkat-SLAMF6^Δ17–65 cells than in WT cells. In comparison, already at 6 hours, the parental line expressed high levels of EGR3, an immune response–restraining transcription factor, and this factor remained highly expressed until 24h; EGR3 had hardly any expression in Jurkat-SLAMF6^Δ17–65 cells in the early time points. The expression of TOX, a key regulator of the exhausted state, decreased in both cell populations, but the reduction was more substantial for Jurkat-SLAMF6^Δ17–65 cells.

In summary, the deletion in CD4 Jurkat cells of full-length SLAMF6, together with preferential expression of the SLAMF6 isoforms SLAMF6^Δ17–65/SLAMF6^Δexon2, induces a cytotoxic gene transcription program accompanied by an array of cytokines and cytolytic genes overexpressed in the modified T cells.

The ITSMs of SLAMF6 serve as possible docking sites for several intracellular adaptors necessary for signal generation. Using our RNA-sequencing analysis, we tested the expression level of these adaptors in WT Jurkat, and in Jurkat-SLAMF6^Δ17–65 cells before and during activation. While EAT2 (SH2D1B) is not expressed at all, SHIP1 (INPP5D) and SHIP2 (INPPL1) expressions are much lower than SAP (SH2D1A), SHP1 (PTPN6), and, SHP2 (PTPN11; Fig. 5A). We focused on SAP and SHP1 as the most probable adaptors involved in SLAMF6 signal modifications. SAP, based on previous reports, plays a central role in the collaboration of B and T cells and is considered critical for their activation through SLAMF6 and other SFRs (24, 25). It was, therefore, surprising to find that the SAP transcript was reduced in Jurkat-SLAMF6^Δ17–65 cells (Fig. 5B; at 6 hours of activation, P = 0.007, and 36 hours of activation, P < 0.0001), with similar results at the protein level (Fig. 5C). XLP is a fatal immune dysregulation syndrome caused by a mutation creating nonfunctioning SAP and characterized by unremitting Epstein–Barr virus infections, dysgammaglobulinemia, and lymphoma (26). PBMCs obtained from two XLP patients with SAP mutations displayed a very high IFNγ secretion of 9,000 pg/mL, compared with 250 pg/mL secreted by PBMCs from healthy donors tested in the same experiment (Fig. 5D). Indirectly, the XLP patient data show that SAP depletion augments cytokine secretion, hinting that the high cytokine production we saw in Jurkat-SLAMF6^Δ17–65 cells might be connected to SAP depletion. To investigate the role of SAP and SHP1 in WT Jurkat and Jurkat-SLAMF6^Δ17–65 cells, we knocked down each adaptor with siRNAs and measured IL2 secretion postactivation (Fig. 5E–H). Silencing SAP in WT Jurkat cells did not affect IL2 secretion, whereas silencing SAP in Jurkat-SLAMF6^Δ17–65 cells resulted in higher IL2 secretion than in the baseline.

Silencing SHP1 curtailed IL2 secretion in Jurkat-SLAMF6^Δ17–65 cells while augmenting IL2 in WT cells, suggesting that SHP1 contributes to the agonistic activity of the SLAMF6^Δ17–65 isoform, but also to the antagonistic effect of the canonical SLAMF6 in WT cells. Coupling of SHP1, a protein phosphatase, to SLAMF6 has previously been associated with reducing T-cell help to B cells (27). Its effect on the Jurkat-SLAMF6^Δ17–65 cells draws attention to the dependence of signal-propagating molecules on both cell-type context.

Thus, although the cytoplasmic tail of SLAMF6^Δ17–65 remains unchanged, its modulatory role reduces SAP expression and requires SHP1 collaboration, generating outcomes opposite to those shown for the full-length SLAMF6.

ICB that reverses T-cell dysfunction changed the prospects for patients with melanoma. PD-1 blockers, in particular, have a significant impact on T-cell functionality (28, 29). We, therefore, evaluated how ICB therapy would affect the ratio of the SLAMF6 isoforms. To address this question, PBMCs were obtained from seven patients with metastatic melanoma before and during ICB therapy. CD8⁺ T-cell subsets were separated, and SLAMF6 isoforms were detected using RT-PCR (Fig. 6A). In healthy donors, isoform expression was measured at several time points and found to be consistently unchanged. In four patients with melanoma, no change in isoform expression was noted after ICB treatment, with one patient achieving a complete response. Two of the four patients did not suffer from any adverse effects (AE) from the drugs; one had thyroiditis (a widespread, relatively easy AE). One had chronic arthritis, emerging after a long course of treatment. In the other three patients, a higher expression of SLAMF6^Δ17–65 was recorded after treatment, in effector and EM CD8⁺ T-cell subsets (Fig. 6B and C). Interestingly, these patients with higher SLAMF6^Δ17–65 expression suffered severe AEs that necessitated treatment disruption: one had grade IV liver toxicity on a combination of CTLA4 and PD-1 inhibitors (the patient is now in complete remission), and two had grade III diarrhea (one attained partial response but then progressed; the other received preventive treatment). Of these, patient 3 (Fig. 6D), who received a combination of CTLA4 and PD-1 inhibitors, was switched to PD-1 monotherapy due to the toxicity. The treatment change was associated with a reciprocal decrease in SLAMF6^Δ17–65 in effector CD8⁺ T-cell populations. The connection between higher SLAMF6^Δ17–65 expression and treatment toxicity may suggest that ICB induces splicing events in lymphocytes.

To investigate the role of SLAMF6^Δ17–65 in healthy T cells, we developed a method to modify SLAMF6 splicing in genetically nonmanipulated cells. ASOs, which are short, single-stranded DNA molecules that interact with mRNA, were used as the molecular modifier for this purpose. ASOs were initially developed to prevent the translation of genes that lead to pathological RNA or protein gain of function, such as those occurring in muscular degenerative diseases (30). An ASO that tightly anneals to an exonic target may prevent the splicing machinery from joining together two exons while removing the intervening intron (Fig. 7A; ref. 31). The SLAMF6^Δ17–65 isoform is typically expressed in T cells at a very low level, compared with the canonical isoform (Fig. 1D). To augment the proportion of the SLAMF6^Δ17–65 isoform, we designed a splice-switching ASO with RNA oligonucleotides synthesized with a full phosphorothioate backbone and in which each ribose-2′-hydroxyl was modified to 2′-methoxyethyl. After electroporation into Jurkat cells, the designed ASO changed the splicing of SLAMF6 in favor of SLAMF6^Δ17–65 (Fig. 7B; Supplementary Fig. S5A and S5B). Higher IL2 secretion was induced by this change (Fig. 7C). Transcription factor expression in the ASO-transfected cells was almost identical to that found in the Jurkat-SLAMF6^Δ17–65 cells, characterized by increased TBX21, C-JUN, and RUNX3, and decreased TOX levels (Fig. 7D). In primary T cells, activated ASO-modified lymphocytes showed higher IFNγ secretion compared with nonmodified cells (Fig. 7E). Finally, we implanted human melanoma cells mixed with splicing-modified and nonmodified cognate TILs (209TIL) into nude athymic mice (Fig. 7F and G). Mice injected with 209TIL that underwent ASO-induced SLAMF6^Δ17–65 preferential expression showed improved tumor control (Fig. 7H), maintaining their superior effect for as long as 30 days after tumor injection (Fig. 7I and J). These results demonstrate the feasibility of using ASOs in adoptive cell therapy (ACT) and emphasize the power of splicing modification to produce T cells with improved antitumor characteristics.

Alternative splicing (AS) is a process by which one gene encodes more than one protein. The outcome of AS is a set of proteins derived from the same DNA sequence but having diverse and even contradictory functions. In cancer research, in the last decade, thousands of splice variants have been identified resulting from mutations in splice sites and regulatory elements or an erratic splicing process (32). We investigated AS of the SLAMF6 immune receptor and evaluated the role of this process in the immune response to cancer.

Using the Immunological Genome Project transcriptional data, Ergun and colleagues showed that over 70% of the genes expressed in immune cells undergo AS and that exon patterns cluster according to lineage differentiation and cell cycling (33). The type of stimuli dictates varied isoform usage in activated monocytes (34) and splicing pattern shifts in activated macrophages (35), dendritic cells exposed to influenza (36), and lymphocytes (33). The Porgador group has reported that decidual NK cells in healthy pregnancies display diverse NK receptor splicing compared with cells isolated from miscarriages (37). This report is among the earliest to link AS of immune receptors to a change of function.

A well-known example of AS that determines a functional state is surface phosphatase CD45. The classical marker of naïve T cells, CD45RA, is the longer variant of CD45, which de-phosphorylates inhibitory tyrosine kinases at a slower rate than CD45RO, an isoform lacking exons 4–6. CD45RO defines effector and memory T lymphocytes. The fast activation of the latter subsets is attributed to the loss of the CD45RA exon and consequently enhanced function (38).

Here we report that the AS of an immune receptor, SLAMF6, generates two opposing modulatory activities. The mode of action of the shorter isoform differs from the canonical variant at all levels investigated, including transcriptional events, signal generation, and modulatory function. The canonical SLAMF6 is a constitutively expressed self-binding receptor whose expression increases in activated T cells, which inhibits their function, acting as a type of rheostat. The shorter variant, SLAMF6^Δ17–65, acts as a dominant-positive costimulatory receptor that strongly enhances T-cell activation and, in our model system, augments an antimelanoma response. The change in function that the splicing process induces in SLAMF6 is profound and perhaps unique. The transcriptional landscape of SLAMF6^Δ17–65 is governed by SLAMF7, GZMB, and NKG2D, which confer cytotoxic activity, and by the transcription factors TBX21 and RUNX3. These features are encountered in helper and exhausted progenitor phenotypes, cell states that develop in activated T cells and characterize subsets critical for an effective immune response (39–41). Thus, although SLAMF6 is an identifier of exhausted T cell precursors, it is also a key player in T-cell activation.

The work presented here suggests that SLAMF6^Δ17–65, not the canonical SLAMF6, is the effector trait driver.

The patient data (Fig. 6), which show that the SLAMF6 splicing ratio may shift during treatment, support our conclusion that SLAMF6^Δ17–65 is involved in the effect of PD-1 and CTLA4 blocking antibodies. Although we only tested a small number of patients, we did note some connection between a stronger global immune response (represented as toxicity) and a change in SLAMF6 splicing. This connection might be explained by a nonexhausted gene expression program in the treated cells affecting also SLAMF6 splicing. This hypothesis should be studied in future work.

The signal generation following SLAMF6 transactivation depends, among others, on the recruitment of SAP. The depletion of SAP protein noted in Jurkat-SLAMF6^Δ17–65 cells is central to its agonistic effect but unrelated to its cytoplasmic sequence, which was not altered. The absence of functional SAP led to a considerable amount of IFNγ secretion in T cells from XLP patients, and likewise, SAP silencing enhanced IL2 secretion in Jurkat-SLAMF6^Δ17–65 cells. Therefore, it is reasonable to assume that SAP recruitment is inseparable from the SLAMF6 rheostatic effect. SAP depletion may hamper other regulatory SFRs, mainly SLAMF4 (42).

In contrast, SHP1, which presumably displaces SAP, participates in the agonistic effect, as shown in the silencing experiments (Fig. 5H). Because inhibitory tyrosine kinases are among the targets of SHP1, we propose that by inactivating them, SHP1 mediates the SLAMF6^Δ17–65 agonistic effect. This mechanism was attributed to NK cell education via SLAMF6 against hematopoietic targets (43). Other intracellular signaling partners, such as SHP2, might also be part of the process; a deeper understanding of the downstream signaling is required for follow-up studies. Altogether, SLAMF6 splice-switched T cells generate a strong, integrated initial stimulus. The cohesive molecular program that ensues initiates an array of cytotoxicity genes and directs the T cells toward a cytotoxic phenotype, even if they originate in the CD4⁺ T-cell lineage, as the Jurkat cell data show.

The functional advantage that heightened SLAMF6^Δ17–65 confers to antitumor CD8⁺ T cells is an asset for adoptive cellular therapies (ACT). Splice-switching oligonucleotide drugs attract attention based on success in clinical trials for Duchenne muscular dystrophy and spinal muscular atrophy (44–46). Chemical modification of nucleic acids, with improved base-pairing affinity and specificity and increased resistance to nucleases, has increased the utility of ASO in the clinic (47). Although a significant problem in genetic diseases is penetrance to the affected tissues, ex vivo modification of T cells before their adoptive transfer avoids most of these drawbacks. The encouraging tumor growth inhibition achieved by the anti-melanoma TILs in which the ratio between SLAMF6 isoforms had been switched highlights the potential of this approach as a novel method for ACT (Fig. 7H).

In summary, the immune system faces a significant challenge in confronting disease, as it must mount a rapid, effective response while maintaining homeostatic regulation to prevent collateral damage. The example of SLAMF6 demonstrates how one receptor gene achieves this balance via a yin-yang relationship. SLAMF6 encodes two tightly coordinated isoforms, one acting as a dominant-positive counterpart of its canonical partner. The danger entwined in the agonistic SLAMF6^Δ17–65 isoform is kept in check by the coexpression of the canonical inhibitory receptor, and vice versa. This balance, however, may be disrupted by therapeutic intervention, as seen with the rise of the shorter transcript in patients experiencing ICB toxicity. ASOs that switch splicing to favor the SLAMF6^Δ17–65 isoform yield T cells with augmented tumor control. Judicious use of such targeted ASOs may therefore constitute a new strategy to improve cell-based immunotherapies.

E. Hajaj reports grants from Dr. Miriam and Sheldon G. Adelson Medical Research Foundation (AMRF), Canadian Institutes of Health Research (CIHR), Melanoma Research Alliance (MRA), Israel Cancer Research Fund (ICRF), International Development Research Centre (IDRC), Israel Science Foundation (ISF), The Azrieli Foundation, Deutsche Forschungsgemeinschaft (DFG), Rosetrees Trust, and Perlstein Family Fund during the conduct of the study, as well as a patent for Nucleic Acid Agents Modulating SLAMF6 Isoforms pending. J. Cohen reports personal fees from Roche Pharmaceuticals, Novartis, MSD, Bristol-Myers Squibb, Medison Pharma, and Sanofi Aventis Israel Ltd. outside the submitted work. S. Frankenburg reports grants from Dr. Miriam and Sheldon G. Adelson Medical Research Foundation (AMRF), Canadian Institutes of Health Research (CIHR), Melanoma Research Alliance (MRA), Israel Cancer Research Fund (ICRF), International Development Research Centre (IDRC), Israel Science Foundation (ISF), the Azrieli Foundation, Deutsche Forschungsgemeinschaft (DFG), Rosetrees Trust, and Perlstein Family Fund during the conduct of the study. N. Hacohen reports other support from BioNTech and Related Sciences outside the submitted work. T. Peretz reports grants from Dr. Miriam and Sheldon G. Adelson Medical Research Foundation (AMRF), Canadian Institutes of Health Research (CIHR), Melanoma Research Alliance (MRA), Israel Cancer Research Fund (ICRF), International Development Research Centre (IDRC), Israel Science Foundation (ISF), The Azrieli Foundation, Deutsche Forschungsgemeinschaft (DFG), Rosetrees Trust, and Perlstein Family Fund during the conduct of the study. G. Eisenberg reports grants from Dr. Miriam and Sheldon G. Adelson Medical Research Foundation (AMRF), Canadian Institutes of Health Research (CIHR), Melanoma Research Alliance (MRA), Israel Cancer Research Fund (ICRF), International Development Research Centre (IDRC), Israel Science Foundation (ISF), The Azrieli Foundation, Deutsche Forschungsgemeinschaft (DFG), Rosetrees Trust, and Perlstein Family Fund during the conduct of the study, as well as patent WO2020261265 pending. M. Lotem reports grants from Dr. Miriam and Sheldon G. Adelson Medical Research Foundation (AMRF), Canadian Institutes of Health Research (CIHR), Melanoma Research Alliance (MRA), Israel Cancer Research Fund (ICRF), International Development Research Centre (IDRC), Israel Science Foundation (ISF), The Azrieli Foundation, Deutsche Forschungsgemeinschaft (DFG), Rosetrees Trust, and Perlstein Family Fund during the conduct of the study; grants from Dr. Miriam and Sheldon G. Adelson Medical Research Foundation (AMRF), and personal fees from Oncohost and Merck outside the submitted work; and patent 160119 pending. No disclosures were reported by the other authors.

E. Hajaj: Conceptualization, data curation, formal analysis, methodology, writing–original draft, writing–review and editing. E. Zisman: Data curation, formal analysis. S. Tzaban: Data curation, formal analysis. S. Merims: Conceptualization, supervision. J. Cohen: Conceptualization, supervision. S. Klein: Conceptualization, data curation. S. Frankenburg: Writing–original draft, writing–review and editing. M. Sade-Feldman: Data curation, formal analysis. Y. Tabach: Conceptualization, supervision. K. Yizhak: Supervision, methodology. A. Navon: Conceptualization, investigation. P. Stepensky: Resources, data curation. N. Hacohen: Conceptualization, resources. T. Peretz: Supervision, writing–review and editing. A. Veillette: Conceptualization, methodology. R. Karni: Conceptualization, formal analysis, supervision. G. Eisenberg: Conceptualization, data curation, validation. M. Lotem: Conceptualization, supervision, writing–original draft, writing–review and editing.

The work of E. Hajaj, E. Zisman, S. Tzaban, G. Eisenberg, S. Klein, S. Frankenburg, S. Merims, J. Cohen, and M. Lotem was supported by research grants from the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation (AMRF), the Canadian Institutes of Health Research (CIHR), the Melanoma Research Alliance (MRA), the Israel Cancer Research Fund (ICRF), the International Development Research Centre (IDRC), the Israel Science Foundation (ISF), The Azrieli Foundation, Deutsche Forschungsgemeinschaft (DFG), the Rosetrees Trust, and the Perlstein Family Fund. The authors wish to acknowledge the devoted technical work of Inna Ben-David, Anna Kuznetz, and Yael Gelfand. They thank Eli Pikarsky and Ofer Mandelboim for helpful discussions and Karen Pepper for editing the manuscript. They thank the G-INCPM, Weizmann Institute of Science, for data analysis. The authors are very grateful to Stewart and Maggie Greenberg for their long-standing support.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.