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

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.

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.

INtRoDUCtIoN
Natural killer (NK) cells and CD8+ T cells, collectively known as cytotoxic lymphocytes (CLs), serve a vital role in immune surveillance by identifying virus-infected and malignant cells and eliminating them through perforin and granzyme-mediated apoptosis (1). The degranulation of CL secretory granules is a critical physiological process to enable perforin and granzyme release into the immune synapse. Bi-allelic mutations in Munc13-4, or the partner proteins Syntaxin-11 or STXBP2, result in defective degranulation and cytotoxicity of NK cells, leading to Familial Haemophagocytic Lymphohisitiocytosis (FHL) (2).
Syntaxin-11 and STXBP2 biochemically interact within the cell (3,4) and are, therefore, likely to function at the same step in secretory granule exocytosis-the docking and fusion of the granules with the plasma membrane. Recent work by Spessott and colleagues confirmed that STXBP2 directly stimulates the Syntaxin-11-mediated fusion of membranes (5).
STXBP2-deficient patient NK cells have severely reduced/ absent degranulation but, surprisingly, and for reasons that are not clearly understood, short-term in vitro treatment of NK cells from STXBP2 (or Syntaxin-11) deficient patients with a low concentration of IL-2 partially or completely restores NK cell degranulation and cytotoxicity (3,4,6), suggesting the existence of a secondary pathway for secretory granule docking. Consistent with these observations, STXBP2/Syntaxin-11-deficient patients develop FHL slightly later than those who harbor mutations in non-redundant proteins, perforin, or Munc13-4 (7,8).
In the current report, we studied NK and T-cell function in an individual with late presentation of FHL due to bi-allelic mutations in STXBP2. Intriguingly, in addition to the expected alterations in STXBP2, we also observed a concomitant reduction in the STXBP1 protein, encoded by the paralog, STXBP1. Further analysis of human NK and T cells demonstrated a functional role for STXBP1 in NK and cytotoxic T-cell cytotoxicity and suggests a unique interplay between STXBP/Munc18 proteins regulating the same exocytic event.

Case RepoRt
A 45-year-old female was admitted to hospital with mild pancytopenia and hepatosplenomegaly. Subsequently, CD3+ T-cell infiltrates were found in the bone marrow, with demonstration of a clonal population by T-cell receptor gene rearrangement analysis suggesting T-cell lymphoma. CHOP (two cycles) and then CHEOP (four cycles) therapy were given with a partial response and an autologous stem cell transplantation was planned 2 weeks after stem cell mobilization with G-CSF (5 µg/kg for 10 days). She developed with rapidly progressive pancytopenia with Hb 90 g/L (reference range 115-165), platelets 8 × 10 9 /L (reference range 150-400) and neutrophils 0.4 × 10 9 /L (reference range 2-7.5), ferritin >16,000 μg/L (reference range 30-150), and sCD25 >60,000 U/mL (reference range <2,400). In this context, FHL was suspected; on further review, it was noted that her brother died in his mid 30s from an illness with some features of FHL. The patient also suffers from focal epilepsy. Subsequent Sanger sequencing of STXBP2 identified two mutations (encoding STXBP2 protein): c.1001C > T (p.P334L) and c.474_483del_insGA (p.C158Wfs*78). While the segregation analysis was not possible, given an apparent familial history of the disease (and the functional results shown below), it is highly likely that the patient had bi-allelic STXBP2 mutations. No mutations were identified in UNC13D, STX11, or PRF1. Monthly Alemtuzumab (Campath) therapy for 6 months partially controlled her symptoms. At the time of writing this report, the patient was 4 months after a T-cell-depleted unrelated donor allograft with no evidence of HLH, but with the complication of multi-drug resistant CMV disease.

t-Cell activation
Peripheral blood mononuclear cells from a healthy donor control and the affected individual were activated with 30 ng/mL of anti-CD3 antibody and 600 U/mL IL-2 for 3 days and then cultured in the presence of 600 U/mL of human rIL-2.

CaR transductions
Amphotropic virus encoding the chimeric antigen receptor anti-erbB2 CD28ζ was produced from the PG13 packaging cells (11) and used to transduce activated T cells following the viral transduction protocol described for KHYG1 cells. CAR-T positive cells were isolated by flow cytometry using an antibody against the surface exposed c-myc epitope and anti-mouse PE-conjugated.

Munc18 Knockdown-primary Human CaR-t Cells
Peripheral blood mononuclear cells 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 negative and CAR-T positive cells were sorted based on the expression of the surface expression of the myc epitope. CAR-T negative CD3 + CD8 + CD4 − cells were immuno-blotted for STXBP1 to determine knockdown. CAR-T positive cells were used to assess CAR-T killing function using 51 Cr release assay, and the effector/target ratio was normalized for % CD3 + CD8 + CD4 − cells; CAR-T + CD4 + cells had marginal (<5%) cytotoxic activity compared to CAR-T + CD3 + CD8 + CD4 − cells.

Degranulation assays
CD107a/Lamp-1 externalization was used to determine NK and T-cell degranulation. Briefly, CAR-T cells were incubated in the presence or absence of MDA-MB-231 Her2 expressing target cells at 1:2 E:T ratio for 3 h at 37°C/5% CO2. NK cells were incubated with K562 targets at 1:2 E:T ratio for 3 h at 37°C/5% CO2 in the absence or presence of 100 U/mL of human IL-2. CD107a externalization was assessed in CD3−CD16+CD56+cells; spontaneous externalization of CD107a was assessed over 3 h in the absence of target cells.

DNa extraction, pCR, and sanger sequencing
Whole venous blood was obtained and genomic DNA extracted using a QIAamp DNA Maxi Kit (Qiagen, Valencia, CA, USA).
Coding exons and splice sites of the STXBP1 gene (Chromosome 9:130, 374, 486-130, 454, 995; NM_003165; ENST00000373302.7) were sequenced. Regions were amplified using gene-specific primers designed to the reference human gene transcript (http:// www.ncbi.nlm.nih.gov/gene). Primer sequences are available upon request. Amplification reactions were cycled using a standard protocol on a Veriti Thermal Cycler (Applied Biosystems, Carlsbad, CA, USA). Bidirectional sequencing of all exons and flanking regions, including splice sites was completed with a BigDyeTM v3.1 Terminator Cycle Sequencing Kit (Applied Biosystems), according to the manufacturer's instructions. Sequencing products were resolved using a 3730xl DNA Analyzer (Applied Biosystems). All sequencing chromatograms were compared to published cDNA sequence; nucleotide changes were detected using Codon Code Aligner (CodonCode Corporation, Dedham, MA).

Cytotoxicity assays
Natural killer (NK) and CAR-T cell killing function was measured using standard chromium ( 51 Cr) release assays, as described previously (13).

ResULts
The patient's NK cell function was assessed using the Lamp-1 externalization assay, a measure of degranulation, and 51 Cr release assays that assesses NK cytotoxicity. Three measurements were made, twice prior to her monthly Alemtuzumab injection ("during therapy") and once 8 weeks after the last Alemtuzumab injection ("post-therapy"). At each of these time points, the patient's naïve NK cells showed marginal activity (Figures 1A-C). Overnight treatment with 100 U/mL IL-2 only partially restored NK cell degranulation and cytotoxicity. In order to determine the functional consequence of STXBP2 mutations on primary human cytotoxic T-cell function, we employed a recombinant chimeric antigen T-cell receptor (CAR) technology. Primary CD3+CD8+ cells from the patient and healthy donors were transduced with an anti-HER2 CAR-T construct, and their activity was assessed against HER2+MDA-MB-231 target cells. In comparison to the controls, the patient cytotoxic T cells had reduced degranulation and cytotoxic activity (Figures 2A,B).
As expected, bi-allelic mutations in STXBP2 resulted in a significant loss of STXBP2 protein relative to the control ( Figure 2C). Consistent with previous reports, we also observed a decrease in the cognate t-SNARE Syntaxin-11, but not cognate Syntaxin-3. Interestingly, we found a significant decrease  in the level of STXBP1 and its cognate t-SNARE Syntaxin-1. No differences in STXBP3 or syntaxin-4 protein levels were observed, suggesting a specific disruption to the STXBP2 and STXBP1 pathway/s ( Figure 2C). Sequencing of the 19 coding exons and splice sites of the STXBP1 gene in our patient did not reveal any pathogenic mutations nor were any single nucleotide polymorphisms detected. To replicate these findings in a recombinant setting, we knocked out STXBP2 in primary human CD8+ T cells using CRISPR/cas9, using three guide RNAs. The results confirmed our original observation, demonstrating significant protein reductions in STX11, STXBP1, and STX1 protein levels as a consequence of STXBP2 loss (Figure 2D). A previous STXBP2/Syntaxin-11 structural study by Hackman et al. (14) demonstrated that, similar to STXBP2, STXBP1 was also capable of binding to Syntaxin-11, albeit at a lower affinity. They speculated that STXBP1 could compensate in the absence of functional STXBP2. Our data suggests that this may not be the case since STXBP1 protein expression appears to be dependant on STXBP2.
To ascertain the role of the STXBP1 protein in CLs function, we used the human NK cell line, KHYG1, which we have previously shown is a suitable model for assessing the NK cell granule death pathway (15). Knockdown of STXBP1 using two different shRNAs resulted in a >90% reduction in STXBP1 protein levels (Figure 3A), and both knockdown cell lines had significantly decreased NK cell-mediated killing of target cells compared with control cell lines ( Figure 3B). We observed no significant loss of function in STXBP2 knockdown NK cells, which was likely due to the presence of IL-2 (600 U/mL) in the culture media ( Figures S1A,B in Supplementary Material). STXBP3 knockdown also had no effect on NK cell function (Figures 3C,D), which demonstrates STXBP isoform specificity in the CL-mediated killing pathway. To substantiate our findings further, we knocked down STXBP1 in primary human CD3+CD8+ T cells that were also transduced with anti-HER2 CAR and demonstrated that the STXBP1-deficient T cells had significantly lower cytotoxic function than control T cells (Figures 3E,F). (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).

DIsCUssIoN
In the current study, we have made the unexpected observation that congenital deficiency of the STXBP2 protein may also affect the expression of STXBP1. Further analysis identified an unsuspected functional role for STXBP1 in secretory granule-mediated NK and T-cell cytotoxicity.
There are certain cellular pathways where single SM isoforms have been shown to regulate discrete trafficking events. Loss of STXBP1, for example, results in a complete block in neurotransmission, despite the presence of both STXBP2 and STXBP3 paralogs in neurons (19). By contrast, both STXBP1 and STXBP2 proteins appear to be important in mast cell degranulation, suggesting that their role may be partly redundant or complementary across the hematopoietic cell lineage (18).
In this study, we have serendipitously uncovered a role for STXBP1 as a regulator of NK and cytotoxic T-cell granule exocytosis. One previous study proposed a similar role for this protein based on its biochemical interaction with Syntaxin-11 in vitro (14). In the current study, STXBP2 deficiency led to reduced levels of Syntaxin-11 as has been previously reported. However we also observed a concomitant protein reduction in STXBP1 and its partner t-SNARE Syntaxin-1, suggesting a greater protein interdependency between these STXBP/ Syntaxin complexes. Our functional analyses revealed that STXBP1 played a role in NK and T-cell cytotoxicity in the presence of IL-2. Since NK and T cells require IL-2 for their growth, we were unable to investigate the role of STXBP1 in the absence of this cytokine. Given the lack of naive NK cell cytotoxicity in the patient described here, we hypothesize that STXBP1 remains "silent" in the absence of IL-2 and only acts in an IL-2 dependent manner to facilitate cytotoxic secretory vesicle docking and target cell killing. Of note, in the current case study, the patient had a significantly reduced level of STXBP1, and this may potentially explain why NK cell function is only partially restored in the presence of IL-2.
STXBP1 plays a non-redundant role in neurotransmission: genetic deletion of STXBP1 results in embryonic lethality in mice (19), and de novo mono-allelic mutations in humans invariably result in epileptic encephalopathy (a rare form of Dravet syndrome) (20). It is, therefore, not surprising that STXBP1 mutations have never been associated with primary immunodeficiency. Importantly, in the current study, the loss of STXBP1 protein was greater than 70% (Figure 2C), strongly suggesting that the observed phenomenon is cell-type specific. Interestingly, a link between STXBP1 and STXBP2 protein expression has been previously described in vitro (21), where loss of STXBP1 resulted in an upregulation of STXBP2 protein, potentially as a compensatory event. By contrast, at least in the current clinical case, the loss of STXBP2/syntaxin-11 also destabilized the STXBP1/syntaxin-1 protein complex, suggesting that there may be a greater interplay between individual Munc18/Syntaxin complexes that act in concert to orchestrate the same exocytic event.

CoNCLUDING ReMaRKs
In the current study, we have identified STXBP1 as an important player in CL function. Although it did not compensate for the loss of STXBP2 in naïve NK cells, STXBP1 appears to play an important role in activated NK and cytotoxic T cells and may be responsible for as much as 50% of their cytotoxic activity. Future studies using a larger cohort of individuals with STXBP2 mutations will determine whether the reduction in STXBP1 protein levels in our STXBP2-deficient FHL case is a more general phenomenon, and will help to unravel the complexity of CL secretory machinery.

etHICs stateMeNt
This study was carried out in accordance with the recommendations of the Peter MacCallum Cancer Centre Human Ethics Committee (approval number 12/73). All subjects gave written informed consent in accordance with the Declaration of Helsinki.
aUtHoR CoNtRIBUtIoNs JAL, TN, LL, KT, HA, PKD, MHK, JAT and IV conducted experiments and/or interpreted the data, and reviewed and revised the manuscript; AM, NJB, AG diagnosed and treated the patient, interpreted the data; JAL, TN, NJB, AG, JAT, IV co-wrote the manuscript; MSH conducted experiments, reviewed manuscript, interpreted the data, and agreed with all aspects of the work; all authors approved the manuscript and agreed with all aspects of the work.