Bi-Allelic Mutations in STXBP2 Reveal a Complementary Role for STXBP1 in Cytotoxic Lymphocyte Killing

Abstract

The ability of cytotoxic lymphocytes (CL) to eliminate virus-infected or cancerous target cells through the granule exocytosis death pathway is critical to immune homeostasis. Congenital loss of CL function due to bi-allelic mutations in PRF1, UNC13D, STX11, or STXBP2 leads to a potentially fatal immune dysregulation, familial haemophagocytic lymphohistiocytosis (FHL). This occurs due to the failure of CLs to release functional pore-forming protein perforin and, therefore, inability to kill the target cell. Bi-allelic mutations in partner proteins STXBP2 or STX11 impair CL cytotoxicity due to failed docking/fusion of cytotoxic secretory granules with the plasma membrane. One unique feature of STXBP2- and STX11-deficient patient CLs is that their short-term in vitro treatment with a low concentration of IL-2 partially or completely restores natural killer (NK) cell degranulation and cytotoxicity, suggesting the existence of a secondary, yet unknown, pathway for secretory granule exocytosis. In the current report, we studied NK and T-cell function in an individual with late presentation of FHL due to hypomorphic bi-allelic mutations in STXBP2. Intriguingly, in addition to the expected alterations in the STXBP2 and STX11 proteins, we also observed a concomitant significant reduction in the expression of homologous STXBP1 protein and its partner STX1, which had never been implicated in CL function. Further analysis of human NK and T cells demonstrated a functional role for the STXBP1/STX1 axis in NK and CD8+ T-cell cytotoxicity, where it appears to be responsible for as much as 50% of their cytotoxic activity. This discovery suggests a unique and previously unappreciated interplay between STXBP/Munc proteins regulating the same essential granule exocytosis pathway.

Keywords: Munc18-1; Munc18-2; apoptosis; cytotoxic T cells; cytotoxic lymphocytes; familial haemophagocytic lymphohistiocytosis; immunodeficiency; natural killer cells.

Figure 1

Bi-allelic mutations in STXBP2 impair natural killer (NK) degranulation and cytotoxicity, which is partially restored by IL-2. Peripheral blood mononuclear cells (PBMCs) were isolated from a healthy donor control (Control) and an FHL5 patient and incubated for 16 h in the absence (-IL-2) or presence (+IL-2) of 100 U/mL of human IL-2. (A) PBMCs were incubated for 3 h in the absence (−K562) or presence (+ K652) of K562 target cells. NK degranulation was assessed by measurement of CD107a surface labeling in the CD16+ cell populations. Data are representative of two independent experiments. (B,C) PBMCs taken during therapy (B) or after therapy (C) were incubated with ⁵¹Cr-labeled K562 target cells for 4 h at the indicated effector to target cell ratios (normalized for the% of NK cells). NK cytotoxicity was determined by the release of ⁵¹Cr from target cells. Data are representative of two independent experiments.

Figure 2

STXBP2 mutations inhibit T-cell degranulation and perturb the expression of the STXBP1 isoform. (A) Peripheral blood mononuclear cells (PBMCs) were isolated from a healthy donor control (CAR-T Control) or an FHL5 patient (CAR-T FHL5 Patient) and incubated with 30 ng/mL of anti-CD3 antibody and 600 U/mL of IL-2 for 3 days. Cells were transduced with virus expressing a CAR specific for the Her2 antigen. CAR transduced PBMCs were incubated for 3 h in the absence (−Her2 Targets) or presence (+Her2 Targets) of MDA-MB-231 target cells overexpressing human Her2. CD8+CAR− T-cell degranulation was assessed by measurement of CD107a surface labeling in the CD8+ cell populations. The right panel shows the % of cells labeled for CD107a after incubation with Her2 targets. Data represent the mean ± SEM (n = 3 experiments). (B) CAR-T-expressing Control and FHL5 cells were incubated with ⁵¹Cr-labeled MDA-MB-231 Her2 expressing target cells for 4 h at the indicated effector to target cell ratios (normalized for the % of CD8+ T cells). CAR-T cytotoxicity was determined by the release of ⁵¹Cr from target cells. Data are representative of two independent experiments. The right panel shows the cytotoxicity of FHL5 CD8+ T cells in the absence (CAR−) and presence (CAR+) of CAR expression. Data are representative of two independent experiments. (C) PBMCs were isolated from a healthy donor control (Control) or an FHL5 patient and incubated with 30 ng/mL of anti-CD3 antibody and 600 U/mL of IL-2 for 3 days. CD3+CD8+ T cells were isolated by flow cytometry and whole cell lysates were blotted for STXBP1, STXBP2, STXBP3, STX1, STX3, STX4, STX11, and actin (loading control). Molecular weight standards (KDa) are indicated. The right panel shows the protein expression (% of Control) for each of the proteins in the FHL5 samples after each blot was normalized for actin expression. Data represent the mean ± SD (n = 2–3 experiments). (D) PBMCs were isolated from a healthy donor and incubated with 30 ng/mL of anti-CD3 antibody and 600 U/mL of IL-2 for 3 days. Cells were transduced with virus expressing Cas9 and the sgRNAs #1, #2, and #3. GFP⁺ Cherry⁺ and GFP⁻ Cherry⁻ control (Ctl) cells were isolated by flow cytometry and whole cell lysates were blotted for STXBP1, STXBP2, Syntaxin-11, Syntaxin-1, and actin (loading control). Molecular weight standards (KDa) are indicated. The panel below shows the protein expression (% of Control) for each of the proteins after each blot was normalized for actin expression. Data represent the average of three guides ± SEM (n = 3).

Figure 3

STXBP1 depletion impairs natural killer (NK) cytotoxicity. (A,B) KHYG1 cells were transduced with virus expressing a scrambled control shRNA or two different shRNAs targeting STXBP1 (#1, #2). Cells expressing the shRNAs were sorted based on the expression of the BFP fluorescent reporter. (A) Whole cell lysates were blotted for STXBP1 and actin (loading control). Molecular weight standards (KDa) are indicated. (B) Control and STXBP1 shRNA KHYG1 cell lines were incubated with ⁵¹Cr-labeled K562 target cells for 4 h at the indicated effector to target cell ratios. Data represent the mean ± SEM (n = 3–4 experiments). Data have been normalized to maximal killing observed in the control scrambled shRNA cell line at 10:1 E/T ratio (set at 100%). (C,D) KHYG1 cells were transduced with virus expressing a scrambled control shRNA or a shRNA targeting STXBP3. Cells expressing the shRNAs were sorted based on the expression of the GFP fluorescent reporter. (C) Whole cell lysates were blotted for STXBP3 and actin (loading control). Molecular weight standards (KDa) are indicated. (D) Control and STXBP3 shRNA KHYG1 cell lines were incubated with ⁵¹Cr-labeled K562 target cells for 4 h at the indicated effector to target cell ratios. Data represent the mean ± SEM (n = 7 experiments). Data have been normalized to maximal killing observed in the control scrambled shRNA cell line at 10:1 E/T ratio (set at 100%). (E,F) PBMCs were isolated from a healthy donor control, transduced with virus expressing either scrambled shRNA or STXBP1 shRNA and sorted based on the expression of the BFP reporter. BFP+ cells were subsequently transduced with virus expressing the CAR-T. CAR-T− and CAR-T+ cells were sorted based on the expression of the surface expression of the myc epitope. (E) Whole cell lysates from the CAR-T− cells were blotted for STXBP1 and actin (loading control). Molecular weight standards (KDa) are indicated. (F) CAR-T+ cells were incubated with ⁵¹Cr-labeled MDA-MB-231 Her2 expressing target cells for 4 h at the indicated effector to target cell ratios. CAR-T cytotoxicity was determined by the release of ⁵¹Cr from target cells. Data represent the mean ± SEM (n = 3 experiments).