Cavβ1 regulates T cell expansion and apoptosis independently of voltage-gated Ca2+ channel function

Abstract

TCR stimulation triggers Ca²⁺ signals that are critical for T cell function and immunity. Several pore-forming α and auxiliary β subunits of voltage-gated Ca²⁺ channels (VGCC) were reported in T cells, but their mechanism of activation remains elusive and their contribution to Ca²⁺ signaling in T cells is controversial. We here identify Ca_Vβ1, encoded by Cacnb1, as a regulator of T cell function. Cacnb1 deletion enhances apoptosis and impairs the clonal expansion of T cells after lymphocytic choriomeningitis virus (LCMV) infection. By contrast, Cacnb1 is dispensable for T cell proliferation, cytokine production and Ca²⁺ signaling. Using patch clamp electrophysiology and Ca²⁺ recordings, we are unable to detect voltage-gated Ca²⁺ currents or Ca²⁺ influx in human and mouse T cells upon depolarization with or without prior TCR stimulation. mRNAs of several VGCC α1 subunits are detectable in human (Ca_V3.3, Ca_V3.2) and mouse (Ca_V2.1) T cells, but they lack transcription of many 5' exons, likely resulting in N-terminally truncated and non-functional proteins. Our findings demonstrate that although Ca_Vβ1 regulates T cell function, these effects are independent of VGCC channel activity.

Fig. 1. shRNA screen identifies Cacnb1 as a regulator of antiviral T cell responses.

A In vivo ion channels and transporters (ICT) screen. CD4⁺ T cells from SMARTA mice were transduced with a pooled shRNA library targeting 223 ICTs (1342 shRNAs including control shRNAs), enriched by cell sorting for transduced (Ametrine⁺) cells and injected into host mice. 7 days after infection with LCMV^ARM, CD4⁺CD45⁺Amt⁺ donor T cells were isolated from host spleens and analyzed by next generation sequencing (NGS) for the depletion or enrichment of shRNAs. B Scores of depleted shRNAs and their p values calculated based on the negative-binomial model using the MAGeCK software package. Cacnb1 (blue dot), other ICTs (gray) and positive controls (red) are indicated. Shown are the pooled data from three independent screens. C, D mRNA expression of Cavβ subunits in mouse (C) and human (D) T cells compared to other immune cells based on data from ImmGen and Fantom5^, databases. Each column represents a different type of immune cell. Heatmaps represent % raw min-max expression for each gene. NK natural killer, NKT natural killer T, Mo monocyte, B B cell, DC dendritic cell, E eosinophil, M macrophage, N neutrophil, LC Langerhans cell. E, F Absolute mRNA expression of auxiliary β, α2δ and γ subunits of VGCCs in mouse (E) and human (F) T cells and reference tissues. Mouse CD4⁺ and CD8⁺ T cells were left unstimulated (−) or stimulated for 12 or 24 h with anti-CD3 + CD28 antibodies. Human CD4⁺ T cells were left unstimulated (−) or stimulated for 6 h with anti-CD3 + CD28 antibodies. mRNA expression was analyzed by RNA sequencing. HD represents the averaged data from three individual healthy donors (HD) and a patient with a STIM1 null mutation (STIM1^null). RNA-Seq for mouse T cells and for human and mouse heart, skeletal muscle, brain, frontal cortex, and retina were extracted from GEO datasets (Supplementary Table 2).

Fig. 2. Deletion of Cavβ1 impairs viability of CD4⁺ T cells and their expansion after viral infection.

A mRNA expression of Cacnb1 in CD4⁺ T cells of LSL-Cas9; Cd4Cre mice transduced with control sgRNA (sgCtrl) and sgRNA targeting Cacnb1. mRNA levels were measured in transduced (Ametrine⁺) T cells by qPCR at day 3 post-transduction. Rlp32 was used as housekeeping control. sgCacnb1 samples were normalized to sgCtrl. B Representative Western blot (left) and quantification (right) of Cavβ1 protein in CD4⁺ T cells transduced with sgCtrl or sgCacnb1. After 4–5 days, Cavβ1 was detected using a monoclonal antibody recognizing aa 19-34 in the N-terminus of Cavβ1. Data in (A, B) are the mean ± SEM of n = 3 mice from independent experiments. C–E CD4⁺ T cells from LSL-Cas9; Cd4Cre mice were transduced with sgCtrl or sgCacnb1 and at day 4 restimulated with anti-CD3 + CD28. C Cell counts shown as the ratio of sgCacnb1 / sgCtrl transduced T cells normalized to non-transduced T cells. D Representative flow cytometry plots (left) and quantification (right) of CFSE dilution at 1 and 3 days after re-stimulation. E Representative flow cytometry plots (left) and quantification (right) of apoptosis measured by annexin V staining at 3 days after re-stimulation. Data in (C–E) are the mean ± SEM of n = 6 mice (in C, D) and n = 8 mice (in E). F Adoptive transfer of CD4⁺ T cells from SMARTA LSL-Cas9; Cd4Cre mice that has been transduced with sgCtrl or sgCacnb1 followed by LCMV^ARM infection. Transduced donor T cells were mixed at 1:1 ratio before injection. At day 7 post-infection, the ratios of sgCacnb1/sgCtrl T cells (and sgCtrl/sgCtrl) were analyzed. Representative flow cytometry plots (bottom left) and quantification (bottom right) of T cell ratios. Data are the mean ± SEM from n = 3 independent experiments pooled from n = 3 donor SMARTA; LSL-Cas9; Cd4Cre mice and n = 10 WT host mice per group. Statistical analysis was conducted by two-tailed, unpaired Student’s t test. **p < 0.01, ***p < 0.001.

Fig. 3. Cavβ1 is not required for Ca²⁺ influx and cytokine production by T cells.

CD4⁺ T cells of LSL-Cas9; Cd4Cre mice were transduced with sgCtrl, sgCacnb1 or sgStim1. A, B After 3 days, Amt⁺ T cells were enriched by cell sorting, recovered for one day in medium containing IL-2 and IL-7 and analyzed. Cytosolic Ca²⁺ levels were measured following stimulation of T cells by anti-CD3 (TCR) cross-linking and ionomycin (Iono) in Ringer’s solution containing 2 mM Ca²⁺. Averaged Ca²⁺ traces (A) and quantification (B) of the area under the curve (AUC) in the time periods indicated by the dotted lines. C, D Cytokine production by CD4⁺ T cells was measured at day 4 after transduction and restimulation for 6 h with PMA and ionomycin. Representative contour plots (C) and quantification (D) of IL-2⁺, TNF⁺ and IFN-γ⁺ CD4⁺ T cells. Data in (A, B, D) are the mean ± SEM of n = 7 independent experiments performed in duplicates. Statistical analysis by two-tailed, unpaired Student’s t test. ***p < 0.001.

Fig. 4. Depolarization of T cells fails to evoke Ca²⁺ influx and Ca²⁺ currents.

A Cytosolic Ca²⁺ levels in mouse CD4⁺ T cells. T cells were stimulated with anti-CD3 + CD28, cultured for 3 days and exposed to 60 mM (top) and 150 mM (bottom) KCl followed by stimulation with ionomycin (Iono). B Cytosolic Ca²⁺ levels in human T cells from a healthy donor (HD) cultured for 10 days in vitro, exposed to 60 mM KCl and stimulated with ionomycin. Averaged Ca²⁺ traces (left) and quantification (right) of the mean F340/F380 ratio during the time periods indicated by dotted lines. Data shown are the mean ± SEM of n = 3–4 (in A) and n = 7 (in B) independent experiments conducted in duplicates. C–H No voltage-gated Ca²⁺ currents and signals in human T cells. C Membrane currents in HD T cells were recorded in 110 mM Ba²⁺ in response to voltage steps from −60 to +60 mV for 200 ms from a holding potential of −80 mV. Current traces were leak-subtracted using the P/8 method with steps from −100 mV. D Current-voltage (I–V) plots (top) and [Ca²⁺]_i concentrations (bottom) measured using Indo-1 in the same cell stimulated in the presence of 20 mM extracellular Ca²⁺ using the voltage protocol shown in (C). E [Ca²⁺]_i was measured in Indo-1 loaded HD T cells held at −80 mV for 20–30 s to establish the baseline [Ca²⁺]_i followed by application of 40-50 depolarizing steps to +10 mV every 1 s. F, G Simultaneous measurements of [Ca²⁺]_i and I_CRAC. F T cells pretreated with TG were exposed to a step-ramp voltage protocol comprising a −100 mV step (50 ms) followed by a ramp from −100 to +100 mV (50 ms) every 1 s. The holding potential was +60 mV to prevent Ca²⁺ influx during the interpulse interval. G Representative I–V plot typical of I_CRAC recorded during the −100 mV pulse from the experiments shown in (F). H [Ca²⁺]_i rises (left) and current densities (right) in response to either depolarizing steps (+10 mV) or TG treatment. For details see Methods. Data in (C–H) represent the mean ± SEM from n = 5–6 cells per condition. Statistical analysis by two-tailed, unpaired Student’s t test. **p < 0.01, ***p < 0.001.

Fig. 5. Lack of voltage-dependent Ca²⁺ influx and Ca²⁺ current in T cells after simultaneous TCR stimulation.

A, B Cytosolic Ca²⁺ levels in CD4⁺ T cells isolated from WT C57BL/6 mice. T cells were stimulated by anti-CD3 (TCR) crosslinking followed at 600 sec by exposure to Ringer’s solution containing either 4.5 mM (left), 60 mM (middle) or 150 mM (right) KCl. Stimulation with ionomycin (Iono) at 1000 s was used as positive control. A Averaged Ca²⁺ traces and (B) quantification of the area under the curve (AUC) during the indicated recording periods. Data represent the mean ± SEM of n = 5 experiments conducted in duplicates. C, D No voltage-activated Ca²⁺ currents are detectable in naïve (C) and expanded (D) mouse T cells. The membrane voltage was stepped from −80 to +60 mV in increments of 10 mV for 50 ms from a holding potential of −70 mV. Currents were leak-subtracted by collecting traces for the same voltage steps in the presence of 100 µM LaCl₃. The current-voltage plot is shown on the right in each case. E, F Voltage-activated K⁺ currents (Kv1.3) in naïve (E) and activated (F) mouse T cells. Kv currents were elicited by depolarizing steps from −100 and +100 mV in increments of 20 mV from a holding potential of −70 mV. The right plot shows the I–V relationship of the recorded Kv currents. Currents were leak-subtracted using the P/8 method. G, H Depolarization fails to induce voltage-gated Ca²⁺ current in activated T cells. G Human T cells from a healthy donor (HD) were stimulated with anti-CD3 antibodies (OKT3) and membrane currents were recorded in extracellular Ringer’s solution containing 110 mM Ba²⁺. Displayed are the raw current traces (without leak subtraction) in response to depolarizing voltage stimuli from −60 to +60 mV in steps of 10 mV for 200 ms from a holding potential of −80 mV. To identify VGCC currents, OKT3-stimulated HD T cells were left untreated or treated with 8 μM nimodipine. H I–V plot of cells shown in (G) in the absence (black circles) or presence (open circles) of nimodipine. Data in (C–H) are representative of the following number of cells analyzed: n = 16 (C), n = 14 (D), n = 5 (E), n = 9 (F), n = 3 (G, H). Statistical analysis in (B) by two-tailed, unpaired Student’s t test. **p < 0.01, ***p < 0.001.

Fig. 6. Lack of ORAI1 or STIM1 does not induce voltage-gated Ca²⁺ current or Ca²⁺ influx in T cells.

A–C Perforated patch recordings of human T cells from a patient with a loss-of-function (LOF) mutation (p.R91W) in ORAI1 (ORAI1^LOF). T cells were left untreated (B) or stimulated with OKT3 (C) for 5–25 min prior to measurements. To record Ca²⁺ current and cytosolic Ca²⁺ levels, T cells were stepped from −60 to +60 mV for 200 ms from a holding potential of −80 mV. Displayed are membrane currents (top), I–V plots (middle) and [Ca²⁺]_i traces (bottom) measured simultaneously in the same cells in 20 mM Ca²⁺ solution. Currents were leak-subtracted using the P/8 method. Data shown in (B, C) are representative of n = 7 and n = 3 cells, respectively. D, E T cells of the ORAI1^LOF patient were loaded with Indo-1 and either left unstimulated (D) or stimulated with OKT3 (E). T cells were stepped to +10 mV for 200 ms every second from a holding potential of −80 mV. Ca²⁺ traces in (D, E) are representative of n = 5 and n = 4 cells, respectively. F Quantification of [Ca²⁺]_i at +10 mV in unstimulated and OKT3-stimulated T cells and after treatment with 5 μM ionomycin (Iono). Δ[Ca²⁺]_i was measured as the difference between the [Ca²⁺]_i prior to and at the end of 30 depolarization pulses. Data shown are the mean ± SEM from n = 4–5 cells. G Cytosolic Ca²⁺ levels in human T cells from a patient with a STIM1 c.497 + 776 A > G null mutation (STIM1^null). T cells were cultured for 10 days in vitro, loaded with Fura-2 and depolarized with Ringer’s solution containing 60 mM KCl. Averaged Ca²⁺ traces (left) and quantification (right) of the peak F340/F380 ratios during the time periods indicated by dotted lines. Data are the mean ± SEM from n = 5 independent experiments. (Note that Ca²⁺ traces of the HD T cells are the same as those shown in Fig. 4B; HD T cells were analyzed together with STIM1^null T cells and are shown for comparison). H, I Cytosolic Ca²⁺ levels in CD4⁺ T cells from wildtype (WT) and Stim1^fl/flCd4Cre mice. T cells were activated for 3 days with anti-CD3/CD28 and then depolarized with Ringer’s solution containing 60 mM or 150 mM KCl. H Averaged Ca²⁺ traces and (I) quantification of the mean (left) and AUC (right) of F340/F380 ratios during the indicated time periods. Data represent the mean ± SEM from n = 5 independent experiments. Statistical analysis by two-tailed Mann–Whitney U test. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 7. Lack of full-length transcripts of VGCC α₁ subunits in T cells.

A, B Absolute mRNA expression levels of α₁ subunits of VGCCs in human (A) and mouse (B) T cells and reference tissues. A CD4⁺ T cells were left unstimulated (−) or stimulated for 6 h with anti-CD3 + CD28 antibodies. HD represents the averaged data from three individual healthy donors (HD) and a patient with a STIM1 null mutation (c.497 + 776 A > G; STIM1^null). B Mouse CD4⁺ and CD8⁺ T cells were left unstimulated (−) or stimulated for 12 or 24 h with anti-CD3 + CD28 antibodies. VGCC expression in T cells in (A, B) was analyzed by RNA sequencing and compared to that in human and mouse heart, skeletal muscle, frontal cortex, whole brain and retina using published datasets (Supplementary Table 2). C–F Exon usage of human CACNA1I (Cav3.3) (in C, D) and CACNA1H (Cav3.2) (in E, F) in CD4⁺ T cells and frontal cortex. C, E Normalized mRNA expression in unstimulated T cells from n = 3 HDs (averaged; red) and frontal cortex (blue) per exon. D, F Transcript levels (as RPKM, reads per kb of transcript, per million mapped reads) in frontal cortex and CD4⁺ T cells from an individual HD superimposed on exons 12–15 (for CACNA1I in D) and exons 12–14 (for CACNA1H in F). In T cells, exon 12 and exon 13 are the first transcribed exons of CACNA1I and CACNA1H, respectively. Green boxes and nucleotide sequences indicate predicted alternative transcriptional start sites (TSS). G Exon usage of Cacna1a (Cav2.1) in mouse CD4⁺ T cells (red) and brain (blue). Normalized Cacna1a mRNA expression per exon and exon-intron structure. H Cacna1a transcript levels (as RPKM) in brain and CD4⁺ T cells from mice superimposed on exons 32–36. In T cells, exon 34 is the first transcribed exon. Green boxes and nucleotide sequences indicate predicted alternative TSS. I Membrane topology model of the predicted N-terminally truncated Cav3.3 protein in T cells lacking channel domain I and part of domain II. AID α₁ interacting domain, ABP α₁ binding pocket.