T and NK cell functionality in a patient harboring heterozygous novel BCL11B p.Asp632fsAla∗91 and STX11 p.R129P mutations

Abstract

BCL11B is a transcription factor essential for central nervous system development and T-cell differentiation that regulates numerous genes across various pathways. Heterozygous BCL11B defects can lead to a broad spectrum of phenotypes, including neurological disorders with or without immunological features. STX11 encodes a t-SNARE protein crucial for the final fusion of lytic granules with the plasma membrane of NK-cells and CD8⁺ T-cells. Biallelic mutations in STX11 are linked to familial hemophagocytic lymphohistiocytosis type 4. We analyzed the functional impact of heterozygous BCL11B (p.Asp632fsAla∗91) and STX11 (p.R129P) variants present in a pediatric patient with Evans syndrome and neurodevelopmental delay, without hemophagocytic lymphohistiocytosis presentation. The BCL11B p.Asp632fsAla∗91 variant, positioned to escape nonsense-mediated decay, likely produces a truncated protein losing three zinc finger domains. Given that BCL11B is involved in the activation of IL-2 gene expression, we evaluated this function and found a reduced IL-2 production by the patient's CD4⁺ T-cells. On the other hand, structural analysis of STX11 indicated that the proline substitution at position 129 could disrupt key interactions with Munc18-2. The patient's cells exhibited decreased NK-cell degranulation and cytotoxicity, and diminished CD8⁺ T-cell degranulation compared to healthy donors. In vitro IL-2 treatment restored these functions to healthy donor levels. We also observed a reduced STX11 protein expression in patient PBMCs. We hypothesize that impaired IL-2 secretion caused by the BCL11B mutation may reduce the patient's ability to compensate for STX11 dysfunction, suggesting that the combined effect of BCL11B and STX11 mutations contributes to the observed immune dysfunction and the patient's complex phenotype. Additionally, our findings suggest that abnormal NK- and T-cells function could play a role in the onset of autoimmune disorders.

Highlights

• •NK and T-cell dysfunction linked to autoimmune disease.
• •New BCL11B mutation causes neurodevelopmental delay.
• •BCL11B mutation linked to Evans syndrome.
• •Impaired IL-2 production affects STX11 function.
• •Reduced NK and CD8⁺ T-cell function in patient cells.

1. Introduction

BCL11B is a zinc-finger type transcription factor that plays an essential role in the development of central nervous system and T-cell differentiation by regulating the expression of numerous genes involved in several pathways. Heterozygous defects in the BCL11B gene leading to loss-of-function are associated with a wide spectrum of phenotypes, including neurological disorders with or without immunological features [1]. Studies in mice have shown that BCL11B is essential in T lymphocyte development [2,3]. BCL11B is also indispensable for the T-cells receptor signal-mediated selection of CD4⁺CD8⁺ double-positive thymocytes, following differentiation into CD4^-CD8⁺ and CD4⁺CD8^- single positive thymocytes, which are committed to the cytotoxic and helper lineages, respectively [4]. Furthermore, BCL11B is expressed in CD4⁺ T-cells and its over-expression results in increased IL-2 production following T-cells activation while its silencing with siRNA has the opposite effect [5].

The importance of IL-2 as a key cytokine for T-cells activation and immune function has extensive experimental support. The lack of IL-2 leads to lethal autoimmunity in mice [6] and the failure to produce normal levels of this cytokine upon cell activation is considered a hallmark of T-cells from patients with systemic lupus erythematosus, a chronic autoimmune disease [7].

Natural killer (NK) and CD8⁺ T-cells are essential components of the immune system, responsible for defense in response to viruses and tumor cells by inducing apoptosis of infected or transformed cells. These cells induce apoptosis of target cells through the secretion of cytotoxic granules containing perforin and granzymes, a process called degranulation, or by inducing death receptor-mediated apoptosis via the expression of Fas ligand [8,9]. In the degranulation process, the proper secretion of the granules depends on precise control of vesicular trafficking and exocytosis.

Syntaxin-11 (STX11) functions as a t-SNARE protein crucial for the fusion of lytic granules with the plasma membrane of T- and NK cells [8]. Its interaction with Munc18-2 is essential for this process [10,11]. Biallelic mutations in STX11 cause familial hemophagocytic lymphohistiocytosis type 4, impairing granule secretion and leading to functional defects in cytotoxic NK and CD8⁺ T-cells [12,13]. Although heterozygous mutations have been identified in some patients, their clinical relevance and functional impact remain poorly characterized [14].

Aberrant NK-cells function has been reported in several autoimmune disorders [15], while CD8⁺ T-cells have been shown to recognize self-antigens and induce apoptosis of targeted cells in the pathogenesis of autoimmune diseases [16].

Evans syndrome is a rare autoimmune disorder characterized by the coexistence of autoimmune hemolytic anemia and immune thrombocytopenia, where abnormal immune responses against self-antigens play a crucial role in its development [17].

In this study, we aimed to determine the relevance of BCL11B p.Asp632Alafs∗91 and STX11 p.R129P heterozygous variants present in a pediatric patient with Evans Syndrome and neurodevelopmental delay, without clinical hemophagocytic lymphohistiocytosis presentation. We assessed IL-2 production in CD4⁺ T-cells, analyzed the degranulation capacity of CD8⁺ T-cells, and evaluated the degranulation and cytotoxicity of NK-cells from cells derived from the patient. Functional comparisons were made with cells from both healthy donors and a patient with Chediak-Higashi syndrome (LYST mutation), a disorder with impaired granule secretion [18]. Although LYST, STX11, and BCL11B are not directly linked, using the Chediak-Higashi patient as a control is biologically and clinically relevant. Both LYST and STX11 mutations affect granule exocytosis and protein trafficking, providing insights into their potential interplay in immune and other biological processes. Additionally, due to BCL11B's role as a transcription factor that modulates the expression of multiple genes, we quantified STX11 mRNA expression, as well as the protein levels of STX11, Perforin, and Granzyme B in PBMCs, ensuring that gene expression levels were compared against those of healthy donors. We also investigated Toll-like receptor 4 (TLR4) re-localization in the patient's monocytes [19].

2. Methods

2.1. Patient recruitment and sample collection

Peripheral blood samples were obtained from the patient with Evans Syndrome, her parents, a brother, and a patient with Chediak-Higashi syndrome after informed consent for genetic and functional testing, in accordance with the Declaration of Helsinki. Age-matched healthy donors included in the study as controls provided written consent under a separate ethics protocol for healthy donors. In total, 20 pediatric donors and 10 adult donors were enrolled as controls. Both the patient and healthy donor samples were obtained at Hospital Garrahan, and the study was approved by the internal ethics review board of the hospital (approval code 1872FA4).

2.2. Genetic analysis and massive sequencing (MiSeq)

Gene panel based Next Generation Sequencing was carried out according to established protocols from Hospital Garrahan, using Illumina MiSeq equipment. Genes included in the panels are listed in Supplementary materials Tables I and II

Genomic DNA was extracted from peripheral blood with NucleoSpin Blood kit (Macherey-Nagel), and a minimum of 50 ng gDNA was used for library preparation. Sequencing was performed using "Illumina MiSeq" as 100-bp paired-end reads. DNA reads were mapped to the GRCh37.75 (GRCh37.p13, EnsEMBL release 75 - February 2014) human genome reference using the default parameters of the Burrows-Wheeler Aligner (bio-bwa.sourceforge.net). Technical duplicates were removed using Piccard, and the "HaplotypeCaller" de GATK algorithm was used to call SNPs and indels [20,21]. Filtering and prioritization of variants were carried out using B_Platform and according to minor allele frequency using dbSNP (Single Nucleotide Polymorphism database, July 15, 2019), 1000 Genomes (https://www.internationalgenome.org/, March 10, 2019), dbNSFP (Non-synonymous single-nucleotide variants database, March 20, 2019) [22], ExAC (Exome Aggregation Consortium, Version 1.0, July 10, 2019) [23], and ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/, March 22, 2019). The possible impact of an amino acid substitution was also assessed using PolyPhen2 (Polymorphism Phenotyping, October 10, 2019) [24]. After filtering, the heterozygous STX11 and BCL11B variants were further investigated.

2.3. Sanger sequencing

To confirm MiSeq-detected variants in STX11 and BCL11B, genomic DNA was PCR-amplified using GoTaq polymerase (Promega), and Sanger sequencing was carried out with exon-specific primers. Amplicons were bi-directionally sequenced using the Big Dye Terminator version 1.1 cycle sequencing kit and an Applied Biosystems 3130xl Genetic Analyzer (Life Technologies).

2.4. Functional analysis

2.4.1. Peripheral blood mononuclear cells isolation

Peripheral blood mononuclear cells (PBMC) were obtained by Ficoll-Hypaque density gradient (GE) centrifugation, washed three times, and resuspended in complete culture medium (RPMI containing 2 mM L-glutamine and 10 % fetal bovine serum (FBS); all from Gibco). Cells were used immediately or were cryopreserved in FBS 10 % DMSO for further experiments.

2.4.2. IL-2 production assay

Detection of IL-2 in CD4⁺ T-cells was made in 1 × 10⁶ PBMCs with or without stimulation with 10 ng/mL PMA (Sigma) and 1 μg/mL ionomycin (Sigma) for 4 h. IL-2 staining was performed with “IL-2 secretion assay human PE detection kit” (Miltenyi) according to the manufacturer's instructions, and analyzed by using FACSCalibur (Beckman Coulter) flow cytometry.

2.4.3. Degranulation capacity

Degranulation of NK- or T-cells following in vitro stimulation was assessed by a flow cytometry-based assay as described previously [[25], [26], [27]], with minor modifications. Briefly, PBMCs were plated in 96-well U-bottom plates at 5 × 10⁵ cells/well in the presence of monensin (10 μM) and a phycoerythrin (PE)-labeled monoclonal antibody (MAb) against CD107a (Biolegend). The antiCD107a MAb was present in the medium throughout the stimulation period because CD107a is externalized by NK- or T-cells after degranulation and rapidly internalized again [[28], [29], [30]]. Degranulation was induced by adding K562 target cells (5 × 10⁵ cells/well, effector:target [E:T] ratio = 1:1) for NK-cells or immobilized antiCD3 (100 ng/mL) + soluble antiCD28 (0.5 μg/mL) for T-cells. Stimulation with PMA + Ionomycin was used as positive control and the negative control well received complete culture medium without stimuli. One well for staining control received neither stimuli nor the antiCD107a MAb. Plates were incubated for 2 or 4 h in an incubator (5 % CO2, 37 °C). Cells were harvested and resuspended in phosphate-buffered saline (PBS) containing 0.02 % sodium azide and 0.02 % EDTA in order to dissociate cell-cell aggregates (5 min, room temperature). Then, the cells were washed in PBS with 0.5 % bovine serum albumin and stained with a fluorescein isothiocyanate (FITC)-labeled MAb against CD3 and a PE Cy5-labeled MAb against CD56 to determine the NK-cells population. T-cells were labeled with FITC-labeled MAb against CD3 and PE Cy5-labeled MAb against CD8. Cells were analyzed by flow cytometry. Three readout parameters were measured: (i) the percentage of CD107a⁺ NK or CD8⁺ T-cells among all NK-cells or all CD8⁺ T-cells, (ii) the percentage of CD107a⁺ NK-cells or CD8⁺ T-cells among all PBMCs, and (iii) the mean fluorescence intensity (MFI) of externalized CD107a on CD107a⁺ NK-cells or CD8⁺ T-cells. NK-cells were identified as CD3⁻CD56⁺ events, and CD8⁺ T-cells were identified as CD3⁺CD8⁺ events.

2.4.4. Cytotoxicity assay

NK-cells cytotoxicity activity was measured by using a flow cytometry-based assay as previously described [31]. Briefly, K562 target cells (T) were labeled with CFSE (carboxyfluorescein diacetate succinimidyl ester; Invitrogen, Paisley, United Kingdom) and plated in 96-well U-bottom plates at 8 × 10³ cells/well. Effector NK in PBMCs (E) were added at different ratios (E:T ratio ranged from 7.5:1 to 30:1). Each E:T ratio was tested in duplicate. A control for spontaneous target cell death was done by an additional two wells containing only target cells. The cells were incubated 4 h at 37 °C and 5 % CO2, stained with 7-AAD and analyzed by flow cytometry. In each sample, the percentage of dead target cells was determined as the percentage of 7AAD⁺ events among CFSE⁺ events.

2.4.5. Granzyme and perforin quantitation

1 × 10⁶ PBMC were stimulated for 24 h with the K562 cell line or with immobilized anti-CD3 and soluble anti-CD28. Then, they were cultured with Brefeldin A (10 μg/mL, Sigma-Aldrich) for 4 h. Afterward, PBMC were harvested, washed, and surface staining was performed. To identify NK and CD8⁺ T-cells, anti-CD3, anti-CD8, and anti-CD56 antibodies were used. Following surface staining, cells were fixed and permeabilized for intracellular staining of Granzyme B (GB12 PE-labeled, Invitrogen) and perforin (dG9 PE-labeled, Invitrogen). Cells were analyzed by flow cytometry.

2.4.6. TLR4 relocalization assay

PBMC were treated with LPS from E. coli O111:B4 (1 μg/mL; Sigma-Aldrich) for 1 h. The LPS-containing medium was then completely removed and the cells were incubated in culture medium without LPS for 2 or 4 h. The cells were then washed and immunostained with a PE-conjugated antiTLR4 MAb (InvivoGen), or an irrelevant isotype control MAb for 15 min, and then analyzed by flow cytometry.

2.5. Total RNA extraction, cDNA synthesis and qRT-PCR

Total RNA was extracted from PBMC derived from the pediatric patient with Evans Syndrome and the patient's mother. Age matched healthy donors were used as controls. RNA was extracted using TRIzol reagent (Invitrogen) and cDNA was synthesized by reverse transcription with MMLV RT (Promega). qRT-PCR was performed in a CFX Connect (BioRad) with SYBR Green Master Mix (BioRad) in 20 μL reactions. For the reaction, primers were used at a concentration of 250 nM and the sequences were: GAPDH Fw 5′-ATCACCATCTTCCAGGAGCGA-3′, GAPDH Rv 5′-GGTTCACACCCATGACGAAC-3′, STX11 Fw 5′-CTTCTCGGTTCGCACTCTCG-3′, STX11 Rv 5′-GATGTGGTCCGTCTCGAACA-3′. The cycling program used was 95 °C for 30 s, followed by 40 cycles of 95 °C for 10 s + 60 °C for 30 s. GAPDH was used as a reference gene (housekeeping).

2.6. Immunoblotting

PBMC's whole lysate cells were prepared in RIPA lysis buffer, quantified by Bradford and separated in sodium dodecyl sulfate/polyacrylamide gel electrophoresis (SDS/PAGE) gels 10 % acrylamide/Bis (Bio-Rad, Hercules, CA). Gels were blotted on a nitrocellulose membrane 0.2 μm (Bio-Rad), blocked with PBS containing 5 % milk and 0.05 % Tween-20 for 30 min and incubated with anti-STX11 specific antiserum as described previously [32], rabbit polyclonal antiMunc18-2 (ab103976) (abcam) or mouse monoclonal antiβ-Actin (3700S) (Cell signaling) antibodies. Primary antibodies were detected by using HRP-conjugated secondary anti-species antibodies (Cell Signaling Technologies). Specific bands were visualized by ECL Western Blotting Detection Kit (BIO-RAD) and a luminescent image analyzer (Amersham).

2.7. Bioinformatic structural analysis

The wildtype 3-D protein structure of BCL11B WT and the predicted truncated protein structure were analyzed according to the protein template from AlphaFold [33]. To determine the structural information on STX11 we used the corresponding sequence as derived from uniprot (UniprotID O75558) to perform a BLAST search against the protein structure database PDB which retrieved as the only hit, the structure of Rattus norvegicus STX1A in complex with Munc18-2 (pdbid 3C98). We were able to map the R129P variant in the corresponding structure and compute its effect on protein stability using FoldX [34].

2.8. Statistical analysis

All the statistics data were analyzed by using the software PRISM 7 (GraphPad Software Inc.).

3. Results

3.1. Clinical description

The index case is a 11-year-old female, the second child of non-consanguineous parents, born at term in the context of a controlled pregnancy. Her health remained stable until 10 months of age when she developed severe thrombocytopenia. Initial treatment with intravenous gammaglobulin proved ineffective, leading to her referral to our Immunology service at Garrahan Hospital at the age of 3 years. She presented with autoimmune hemolytic anemia, diagnosed by a Coombs-positive test with the highest value of anti-Rh autoantibodies and initiated treatment with prednisone (2 mg/kg/day). While there was an initial positive response, subsequent relapses occurred, prompting further prescriptions of rituximab and mycophenolate, ultimately resulting in a favorable outcome.

Family history includes hypothyroidism in the maternal grandmother, uncle and aunts, as well as another aunt with rheumatological pathology. Her 18-year-old brother also has hypothyroidism, and a younger brother had congenital nephrotic syndrome and congenital hypothyroidism, which caused his death at around 2 years of age, without clear genetic studies (Fig. 1A).

Fig. 1.

Novel mutations (A) The family pedigree of the patient with BCL11B and STX11 mutations, along with the Sanger sequencing electropherograms of all family members, are shown. In the pedigree, circles represent females, squares represent males, white indicates unaffected individuals, gray represents individuals with autoimmune manifestations, and black denotes the index case (B–C) Schematic representation of STX11 protein (the mutation (p.R129P) is indicated in the Habc domain) o BCL11B protein (the novel variant (p.Asp632fsAla91∗) is indicated between the zinc finger domains 3 and 4) (D)In silico structural models of BCL11B WT (left) or p-Asp632fsAla91∗ (right). Violet: ZNFn1; Blue: ZnFn2; Green: ZnFn3; Red: ZnFn4; Orange: ZnFn5; Yellow: ZnFn6. (E)In silico structural analysis of STX11 p.R129P-STX11 substitution using available STX11:Munc18-2 model. NHabc domain crystallographic structure of STX11 WT (upper left), p.R129P (upper right) and STX11 protein displaying a rich hydrogen bond network with Munc18-2 (down). The d values (distance between hydrogen bonds atoms) are also indicated. The violin plot of the thermodynamic stability analysis is also indicated, each point represents the thermodynamics value of changes in either STX1A crystal or homology models.

The patient's bone marrow examination revealed megakaryocytic and myeloid hyperplasia without lymphoproliferation or infectious events. The patient currently presents molluscum contagiosum and developmental delay. Immunophenotypes from the patient are summarized in Table 1.

Table 1.

Immunological characteristics of the patient with Evans Syndrome and neurodevelopmental delay.

	Patient	Reference Value
Blood count
White blood cells	11.5 × 109/L	5.5-15.0 × 109/L
PMNs	7.1 × 109/L	1.5–8.5 × 109/L
Lymphocytes	3.95 × 109/L	2.00–8.00 × 109/L
Monocytes	0.47 × 109/L	0.30-0.50 × 109/L
Leukocyte subpopulations
T-cells:
CD4+	31.0 %	31.1–47.4 %
CD8+	53.0 %a	16.0–26.9 %
CD4−CD8−	13.1 %	6.7–14.0 %
CD4+CD8+	3.0 %a	0.3–2.5 %
B-cells:
CD19+CD27+	20.0 %	8.1–33.3 %
CD19+CD27+IgD+	13.3 %	3.1–18.0 %
CD19+CD27+IgD−	7.5 %	2.9–17.4 %
NK-cells:
CD56bright	6.6 %	0.8–10.6 %
CD56dimCD16+	25.2 %a	68.0–85.0 %
CD56dimCD16-	6.1 %	3.0–11.0 %
CD56−CD16+	62.0 %a	5.6–12.0 %

The total mononuclear cell count was 3950 cells/μl. T-cells were 78.1 % of total lymphocytes. B-cells were 9 % of total lymphocytes. NK-cells were 6.4 % of total lymphocytes.

^aindicates out of reference value.

Genetic studies for CTLA-4 and STAT3 were conducted due to autoimmune cytopenias and slightly decreased IgA levels, considering a dominant autosomal inheritance pattern in the family. However, no clinically relevant mutations were found in these genes. Moreover CD25 deficiency and autoimmune lymphoproliferative syndrome were ruled out.

In the last year, the patient has been under endocrinological follow-up for suspected hypothyroidism, given her family's clinical history. She currently does not present cytopenias and has been without treatment since 2022.

3.2. Massive genetic evaluation

In 2019, to expand the patient's genetic evaluation, a massive sequencing study, consisting of an in-house developed panel of 153 genes (Supplementary Table I) was performed, which led to the identification of a novel heterozygous STX11 variant, namely: c.386G > C; p.R129P. The combined annotation dependent depletion (CADD) score of the mutation was 22.3, and the polyphen score was possibly damaging with a score of 0.940, suggesting its pathogenicity. The analysis did not reveal any other mutation known to be involved in immunodeficiency or immune dysregulation in the analyzed genes. The mutation was confirmed in the patient using direct Sanger sequencing, and family segregation analysis found that the mother, grandmother, uncle and aunts with autoimmunity were also heterozygous for the variant. As expected the father did not show the mutation. The mutation yields an amino acid substitution in the N-terminal Habc domain of STX11 (Fig. 1B).

3.3. Expanding the genetic evaluation

In late 2022, with the manifestation of neurodevelopmental delay, we decided to expand the genetic evaluation in the patient by conducting a comprehensive sequencing study. This study utilized another in-house developed panel of 331 genes (Supplementary Table II) and resulted in the discovery of a novel heterozygous variant in the last exon of BCL11B gene: c.1895delA; p.Asp632fsAla∗91. Notably, this variant was also found in a heterozygous state in the patient's mother, although it was absent in her brother, uncle and aunts. The BCL11B p.Asp632fsAla∗91 variant was located on the exon 4, before the zinc-finger domain 4 and positioned to escape nonsense-mediated decay (Fig. 1C). Interestingly, the patient has an altered phenotype of NK-cells, with an enrichment of CD16⁺CD56⁻ cells and low CD16⁺CD56^dim subsets (Table 1), similar to her mother (data not shown). Although she has a normal T-cell count, a slight increase in double-positive T-cells was observed in the patient (Table 1) but not in her mother (data not shown). Similar findings in patients with BCL11B mutations have been reported previously [35,36].

3.4. In silico structural analysis of proteins

We utilized the recently introduced AlphaFold algorithm, a computational tool known for its precise prediction of protein structures at the atomic level [33], to predict that the BCL11B p.Asp632fsAla∗91 truncated protein would be deficient in the final three zinc-finger domains. Given that BCL11B acts as a transcriptional activator, the presence of a premature stop codon in the BCL11B sequence likely produces a truncated protein that could impair its ability to bind to target DNA and interact with associated proteins (Fig. 1D).

We also performed in silico structural analysis of p.R129P-STX11 substitution. Although there is no crystallographic structure available for STX11, we were able to use that available for STX1A (a homologous protein from Rattus norvegicus with 98.6 % of sequence identity), which was crystallized in complex with Munc18-2 (PDBID 3c98). Previous works demonstrated that Munc18-2 act as a chaperone of monomeric STX11 as well as an activator for membrane fusion, making their interaction crucial [10,11].

The R129 residue in STX11 (equivalent to K117 in STX1A) is located within a conserved region, constituting part of the initial segment of the second alpha helix within an helix–loop–helix structural motif (Fig. 1E). It forms a clear charged interaction with a glutamic residue from the preceding helix. Consequently, we evaluated how the folding free energy changes in the mutant protein compared to the wild type, bearing in mind that a ΔΔG value of zero indicates no impact on protein stability, a negative value suggests increased stability, and a positive value suggests decreased stability.

While the calculation traditionally involves a specific protein structure, we opted for increased accuracy by conducting calculations across multiple structures. This included utilizing a homologous protein with an available crystal structure, as well as several in silico structural models. As anticipated, thermodynamic analysis revealed a significant destabilization upon introduction of a proline in the corresponding position within both the STX1A crystal and various STX1A homology models. The destabilization was quantified as 5.1 ± 0.6 kcal/mol, surpassing the previously reported threshold of 2 kcal/mol [37]. This is depicted in the violin plot in Fig. 1E, illustrating the dispersion of values obtained for the mutant across several model structures.

Furthermore, the helix–loop–helix motif and the aforementioned residues are integral components of a hydrogen bond network involved in interactions with Munc18-2. Notably, the conserved R126 (R129 in STX1A) forms a charged interaction with Munc18-2 E59 and Q131 (119 in STX1A), along with hydrogen bonds with Munc18-2 T56 (Fig. 1E). These findings strongly suggest potential impairment of the STX11-Munc18-2 interaction, prompting us for a deep analysis on the subject.

3.5. Reduced function in patient T-cells, NK-cells and monocytes compared to healthy donors

To determine the functional impact of the variants found in our patient, we analyzed the IL-2 production of CD4⁺ T-cells, the degranulation capacity of CD8⁺ T-cells, the degranulation and cytotoxicity ability of NK-cells, and the stimulus-dependent transport of TLR4 to the monocyte's plasma membrane.

In terms of CD4⁺ T-cell activation, we noted a significant decrease in IL-2 production by the patient's cells and a minor reduction in the production by the mother's cells compared to pediatric donors (PD) and adult donors (AD), respectively (Fig. 2A). Our results resemble observations described for individuals with BCL11B mutations.

Fig. 2.

Functional analysis (A) T-cells IL-2 production upon stimulation of PBMC with PMA Ionomycin for 4 h. The dot plots show IL-2 production (percentage displayed) by viable CD3⁺CD4⁺ T-cells. (B) Degranulation assay Upper panel: Dot plots show the gating strategy used to identify CD107a surface expression on NK-cells by flow cytometry. Values represent the % of ΔCD107a⁺ NK-cells (%CD107a in stimulated condition - %CD107a in un-stimulated condition). Data are the mean ± SD ∗p < 0.05 for AD vs. mother and AD vs. brother (p = 0.036). ∗∗p < 0.03 for PD vs. patient (p = 0.019) Mann Whitney test. (Right)Bottom panel: Dot plots show the gating strategy used to identify CD107a surface expression on T-cells by flow cytometry. Values represent the % of ΔCD107a⁺ positive CD8⁺ T-cells. Data are the mean ± SD ∗p < 0.05 for PD vs. patient (p = 0.0303). Mann Whitney test. (Right)(C) NK-cells-mediated cytotoxicity assay analyzed by flow cytometry (upper panel). Specific lysis of CFSE-stained cells by NK-cells in adults (bottom left) or pediatrics (bottom right) are shown. Each line represents a single healthy donor. NK cytotoxicity (%) was calculated as cells positive for both CFSE and 7-AAD/total CFSE⁺ cells, after subtracting the spontaneous lysis (%) in negative control. (D)In vitro treatment of NK-cells for 48 h with or without IL-2. Data from patient and healthy donor samples are shown. AD: Healthy Adult Donors. PD: Healthy Pediatric Donors.

As a measure of degranulation we evaluated by flow cytometry the expression of CD107a on CD8⁺ T-cells and NK-cells surface after cell stimulation with antiCD3/CD28 or with K562 human lymphoblastoid cell line, respectively. NK-cells-mediated cytotoxicity was evaluated by flow cytometry after staining target cells with 7AAD and CFSE. We observed a reduction in the degranulation capacity of NK-cells and CD8⁺ T-cells, as well as a decrease in NK-cell cytotoxicity in the patient, her mother (carrying both BCL11B and STX11 variants) and her brother (carrying only STX11 variant) compared to healthy donors (Fig. 2B and C). Nevertheless, these reductions were less severe than those observed for the Chediak-Higashi syndrome cells.

The in vitro treatment of patient and mother NK-cells for 48 h with a low concentration of IL-2 restored the degranulation and cytotoxicity of NK-cells as previously reported for STX11 mutations (Fig. 2D) [16,38].

Finally, given previous findings showing that STX11 knockdown reduces TLR4 stimulus-dependent transport to the plasma membrane in monocytes/macrophages [19], we evaluated TLR4 relocalization in monocytes from the patient, her mother, and her brother. The patient exhibited reduced TLR4 relocalization under stimulation conditions, while no defects were observed in her mother or brother (Supplementary Fig. 1). This functional difference highlights phenotypic variability despite the shared genetic background, suggesting additional contributing factors beyond STX11 and BCL11B variants to the complex phenotype of the patient.

3.6. Similar transcript levels in patient cells and healthy donors cells but reduced STX11 protein levels

Given that BCL11B is a transcription factor that regulates the expression of many genes in T and NK-cells we evaluate the expression of STX11 transcript in cDNA prepared from PBMCs from the patient, her parents, and control cells, and then quantitative real-time PCR specific for STX11 transcript was performed. Our results showed that the level of STX11 mRNA in the patient and mother were similar to healthy donors (Fig. 3A). Similar results of STX11 transcript expression were previously reported in BCL11B KO models [39].

Fig. 3.

mRNA and protein expression (A)Total RNA was extracted from PBMC samples and after cDNA synthesis qRT-PCR was performed to evaluate STX11 mRNA level expression. GAPDH was used as housekeeping. Relative mRNA expression of STX11 transcript was normalized to GAPDH using deltadelta CT method. (B–C). Protein extracts were prepared from PBMC samples. Proteins were separated in a 10 % SDS-PAGE gel and transferred to nitrocellulose membranes. Membranes were blotted with primary antibody against STX11 (B) or Munc18-2 (C) and primary antibody was detected using HRP-conjugated secondary antibody. β-Actin was used as loading control. Cropped immunoblot for STX11 (B) or Munc18-2 (C) protein expression of PBMC's lysates are shown. Densitometry analysis of expressed STX11 content was performed using the Image J program and normalized by the control level. (D) Intracellular staining for perforin or Granzyme B of healthy pediatric donor (PD) or Patient in NK-cells treated with K562 cell line (gated on CD3⁻CD56⁺ cells) or T-cells stimulated with antiCD3/CD28 (gated on CD3⁺CD8⁺ cells). Shaded histograms show marker expression as indicated (mean fluorescence intensity displayed), bold lines show the corresponding isotype control. Data shown are representative of three to four independent experiments.

PBMCs from the patient, her mother (carrying both BCL11B and STX11 variants) and her brother (carrying only STX11 variant) revealed reduced STX11 protein levels (Fig. 3B and Supplementary Fig. 2) with similar Munc18-2 expression when compared to healthy donors (Fig. 3C). Moreover, the expression levels of Granzyme B and perforin, as assessed through intracellular staining, were consistent among healthy donors and the patient in activated NK-cells and CD8⁺ T-cells (Fig. 3D).

4. Discussion

In this study, we investigated the relevance of BCL11B (p.Asp632fsAla∗91) and STX11 (p.R129P) variants observed in a patient clinically diagnosed with Evans syndrome and neurodevelopmental delay.

Regarding the newly discovered BCL11B Asp632fsAla∗91 mutation, positioned to escape nonsense-mediated decay, likely produces a truncated protein that would lack the last three zinc-finger domains and hence affect the protein's function for binding to its target DNA. Functional assessment in CD4⁺ T-cell revealed a significant decrease in IL-2 production in the patient's cells and a minor reduction in the mother's cells, resembling observations in individuals with BCL11B mutations [5]. However, further experiments are needed to elucidate the true molecular mechanisms of action of this variant [1]. The BCL11B Asp632fsAla∗91 mutation also offers a plausible explanation for the neurodevelopmental delays observed in the patient and also contribute to the autoimmune aspects of the condition, as has already been reported for this gene [[40], [41], [42]]. Surprisingly, the mother, who is also a carrier of this variant, does not present the neurological phenotype of the daughter. The absence of neurological symptoms in the mother may be a contribution of genetic background differences that could play a role; making her symptoms relatively mild compared to her child's. Alternatively, she could be mosaic, with the mutation occurring after the central nervous system but before the immune system development, although direct evidence for this has not been obtained.

STX11 functions as a t-SNARE protein critical for cytotoxic granule exocytosis in NK and CD8⁺ T-cells, with low expression in certain brain regions (e.g., cortex, hippocampus) that could suggests possible, yet unexplored, roles in neural vesicle trafficking. However, neurological abnormalities seen in patients with familial hemophagocytic lymphohistiocytosis, including those with STX11 mutations, are generally attributed to the overall disease pathophysiology rather than to STX11 itself [43]. In contrast, the BCL11B mutation identified in our patient is a well-established cause of neurodevelopmental delay, consistent with its essential role in CNS development [[40], [41], [42]]. Thus, the neurological phenotype in this case is more likely to be related to the BCL11B mutation rather than subtle or indirect effects of STX11.

Given that BCL11B is a transcription factor, we measured the expression of STX11 mRNA, as well as the expression of STX11 protein and other proteins involved in T- and NK-cell degranulation such as Munc18-2, Perforin, and Granzyme B. Our studies demonstrated that the expression of STX11 mRNA was similar to that of healthy donors and protein expression of Munc18-2, Perforin, and Granzyme B were not affected. These findings indicate that the BCL11B mutation did not alter the expression of these genes, consistent with previous reports from BCL11B KO models [39].

The observed lower STX11 protein expression in the Western blot, when probed with a well-characterized anti-STX11 polyclonal serum raised against the first 15 amino acids of the STX11 protein [32], can be explained by our in silico analysis of STX11 protein. We observed that the R129 resides in an helix–loop–helix motif crucial for STX11-Munc18-2 interaction. Introduction of a proline in this position is predicted to be highly destabilizing, potentially leading to an unstable or partially misfolded protein prone to degradation [10,11]. Structural analysis suggests that even if present, the p.R129P-STX11 variant would likely impair interaction with Munc18-2, a protein crucial for chaperoning monomeric STX11 and promoting SNARE complex assembly for membrane fusion.

Our patient exhibited lower than normal degranulatory activity of NK-cells and CD8⁺ T-cells, as well as reduced cytotoxic activity in NK-cells. The patient's brother, who does not carry the BCL11B variant, also showed lower in vitro functionality of NK-cells and T-cells, along with reduced STX11 protein levels compared to healthy donors. Similar to our findings, previous studies on mutations affecting the Habc domain of STX11 have shown that heterozygous carriers without disease manifest decreased NK-cell functionality and reduced STX11 protein expression compared to healthy donors [44].

Notably, STX11 has been implicated in TLR4 trafficking by facilitating its transport to the plasma membrane in a SNAP-23-dependent manner [19]. Here we found a reduction in TLR4 relocalization in the patient's monocytes. Previous studies have shown that decreased TLR4 expression on monocytes correlates with Treg abnormalities and diminished IL-10 production, which may exacerbate autoimmune phenotypes, as seen in primary immune thrombocytopenia [45,46]. Nevertheless, in our study, the absence of TLR4 relocalization defects in the mother and brother, who share STX11 variants, suggests that other factors, such as immune activation or autoimmunity, may play a role in this phenomenon. Further studies are necessary to delineate the precise contribution of STX11 to TLR4 trafficking and its potential involvement in the autoimmune phenotype of Evans syndrome.

We speculate that it is plausible that both the BCL11B and STX11 mutations contribute to the complex immunological phenotype of our patient. Our hypothesis is that autoimmune conditions observed in the family members may be linked to the reduction of NK-cells and T-cells functionality. In fact, the lack of IL-2 secretion due to the BCL11B mutation may contribute to the inability in vivo to rescue the effects of the STX11 mutation observed in the patient. Notably, the mother, who also carries the BCL11B mutation and exhibits slightly decreased IL-2 production, does not manifest the same clinical manifestations as the patient, supporting the notion that IL-2 secretion may play a crucial role in modulating the clinical phenotype associated with these genetic variants.

The potential pathogenicity of the heterozygous STX11 p.R129P variant warrants further investigation to clarify its role in disease, if indeed it exists. While this study provides valuable insights using patient-derived cells, including evidence of reduced NK- and T-cell function, diminished IL-2 production, and impaired degranulation capacity, comprehensive functional studies are necessary to determine the variant's impact definitively. In vitro reconstitution assays using STX11-WT and p.R129P constructs in both STX11-deficient and sufficient cell lines would be particularly informative. However, resource constraints and funding limitations have restricted the scope of the current study.

In recent years, novel genes were associated with Evans Syndrome, including STAT3 gain-of-function mutations [47]. Interestingly, mice deficient in STAT3 have increased cytotoxicity and degranulation, and the constitutive activation of STAT3 leads to post-transcriptional regulation of NK-cells [48]. Further studies are needed to investigate the specific molecular mechanisms underlying this dysfunction in post-transcriptional regulation [49]. Moreover, gain-of-function mutations in STAT1 have been shown to result in low cytotoxicity and degranulation in NK-cells [50]. Also CD8⁺ T-cells have been shown to recognize self-antigens and induce apoptosis of targeted cells in the pathogenesis of autoimmune diseases [16].

These results, together with our functional studies, support the idea of an association between NK-cells and T-cells dysfunction and autoimmune disease, such as Evans Syndrome [51], which is a multifactorial autoimmune disease. Our study underscores the importance of comprehensive genetic and functional analyses in understanding the complex interplay between genetic variants and immune dysfunction in individuals with rare disorders, specially autoimmune disorders associated with NK-cells and T-cells dysfunction.

CRediT authorship contribution statement

Lorenzo Erra: Validation, Methodology, Investigation. Ana Colado: Methodology, Investigation. Franco Gino Brunello: Investigation. Emma Prieto: Validation, Investigation. Verónica Goris: Validation, Methodology. Mariana Villa: Investigation, Conceptualization. Matías Oleastro: Investigation. Marcelo Martí: Supervision, Resources, Formal analysis. Roberto Gabriel Pozner: Methodology. Mercedes Borge: Writing – original draft, Supervision, Investigation, Formal analysis, Conceptualization. María Belén Almejun: Writing – review & editing, Writing – original draft, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization.

Consent to participate

Participants, or their parents, signed an informed consent.

Ethics approval

This study was conducted in compliance with the ethical standards of the institutional and national research committees and with the 1964 Helsinki Declaration and its later amendments. Ethical approval for this study was obtained from the Bioethics Committee of Hospital Garrahan under the code 1872FA4.

Data availability statements

All data generated or analyzed during this study are included in this published article.

Funding

This work was supported by grants and fellowships from the Agencia Nacional de Promoción Científica y Tecnológica (ANPCyT) (PICT 2020-SERIEA-01570), the Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) (PIP 2021-2023 – 11220200102208CO), the Ministerio de Ciencia, Tecnología e Innovación (CONVE-2023-100390147-APN-MCT (GeC-EPoF)), the University of Buenos Aires (UBACyT 2020 20020190200067BA) and the Fundación de Ciencias Exactas y Naturales (FUNDACEN) (+4i).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

figs1.

figs2.