Novel Heterozygous Mutations of TNFRSF13B in EBV-associated T/NK Lymphoproliferative Diseases (Ebv-T/nk-lpds)
Xinyue Deng,Tong Ge,Kefeng Shen,Jiachen Wang,Wei Mu,Hui Luo,Jia Gu,Meilan Zhang,Min Xiao
DOI: https://doi.org/10.1097/bs9.0000000000000180
2024-01-01
Blood Science
Abstract:1. INTRODUCTION Epstein-Barr virus (EBV) is able to infect T and/or natural killer (NK) cells and trigger persistent EBV replication and intractable EBV-associated T/NK lymphoproliferative diseases (EBV-T/NK-LPDs) in rare cases,1,2 especially when the functionalities of tonsillar NK cells3 and CD8+ T cells4,5 are impaired. Tumor necrosis factor–like receptors (TNFRs) are part of a superfamily heavily involved in the physiology of immune cells. Mutations in TNFRSF13B (encoding the transmembrane activator and cyclophilin interactor [TACI] protein) were previously reported to be associated with common variable immunodeficiency (CVID),6 indicating the complex imbalance of the immune system caused by TACI deficiency. Increasing evidence also suggests the potential role of TACI in responses of T cells.7,8 Here we report a series of novel heterozygous TNFRSF13B mutations in 6 patients diagnosed with EBV-T/NK-LPDs. In this work, we try to expand the routine panel of genes screened in the patients presenting with EBV-associated proliferative diseases and provide a new possible way to correlate the genetic predisposition, persistent EBV infection, and EBV-T/NK-LPDs etiopathology. 2. CASE REPORTS 2.1. Case presentation From May 2016 to December 2022, 6 patients in our hospital coincidentally met 2 criteria: be diagnosed with EBV-associated hemophagocytic lymphohistiocytosis (EBV-HLH) or chronic active EBV disease of T/NK-cell type (chronic active EBV disease of T/NK-cell type [CAEBV-T/NK]); potentially deleterious alterations in TNFRSF13B with or without other previously confirmed HLH-associated genetic mutations were reported in whole exon sequencing (WES) (Table 1). Routine laboratory examinations, diagnostic imaging, dynamic monitoring of biomarkers for prognosis, and corresponding treatments performed in our hospital are displayed in Table 2, Figure 1A and Supplementary Figure 1, https://links.lww.com/BS/A80. Examinations listed here were performed within 7 days after the first admission or attack, and the test time was as close as enough to the first admission or attack. In total, 4 of 6 patients had met the current diagnostic criteria for HLH and were given HLH-1994 or HLH-2004 regimen. Hemophagocytosis phenomenon in bone marrow was found in 3 of 6 patients (Fig. 1B). Special experiments on EBV infections and the functionalities of CTLs/NK cells are summarized in Figure 1. In all 6 patients, T cells and NK cells were the main targets of EBV infection (Fig. 1C), accompanied by the low NK-cell cytotoxicity and relatively normal expression levels of granzyme B/perforin on CTLs/NK cells (Fig. 1D, E). Chemotherapies were performed in 4 of the 6 patients (Table 2), including Tislelizumab (patient 1), HLH-19949 and DEP regimen (patient 2), HLH-1994 regimen (patient 3), HLH-200410 and L-GMOX regimen (patient 4), and the HLH-1994 plus rituximab (patient 6). Three of 6 patients (patients 2, 3, and 5) had received the hematopoietic stem cell transplantation (HSCT) (Supplementary Table 1, https://links.lww.com/BS/A81) but 2 of them eventually died of uncontrolled HLH. HSCT was also recommended for patient 4 and patient 6 but not finally performed due to personal reasons. Their present status is not available since they were lost to follow-up. Table 1 - Genetic characteristics of patients with TNFRSF13B alterations by whole-exome sequencing. Patient Gender Age Gene Type Exon Frequency cDNA Protein SIFT Polyphen2 Mutation taster MAF (gnomAD) dbSNP 1 F 36 TNFRSF13B Frameshift deletion 3 43.60% c.366delG p.Ser123Valfs*31 - - D - - 2 M 46 TNFRSF13B Frameshift Deletion 2 40.00% c.105delC p.Glu36Lysfs*48 - - D <0.0001 - LYST Missense mutation 34 50.00% c.8624G>A p.Arg2875His D D D 0.000121 rs200353560 3 F 32 TNFRSF13B Missense mutation 2 54.40% c.139T>A p.Cys47Ser D D D <0.0001 rs769182186 NOD2 Missense mutation 4 54.80% c.922C>T p.Leu308Phe D D D - - 4 F 46 TNFRSF13B Missense mutation 2 54.12% c.226G>T p.Gly76Cys D D D <0.0001 rs146436713 5 M 8 TNFRSF13B Missense mutation 2 49.43% c.115T>C p.Tyr39His D D D - - AP3B1 Missense mutation 13 49.80% c.1234T>C p.Tyr412His D D D 0.000014 rs781034104 6 M 16 TNFRSF13B Missense mutation 2 50.40% c.124C>A p.Pro42Thr B B B <0.0001 rs531640813 B = benign, cDNA = complementary DNA, D = deleterious, MAF = minor allele frequency, SIFT = sorting intolerant from tolerant. Table 2 - Clinical and laboratory characteristics* of patients with TNFRSF13B alterations. Patient Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Normal range Gender F M F F M M - Age at onset, y 36 46 32 46 8 16 - Diagnosis EBV-LPD EBV-HLH EBV-HLH; Crohn disease CAEBV-T; HLH CAEBV-NK EBV-HLH; EBV-T/NK-LPD - HLH-2004 criteria No Yes Yes Yes No Yes - Fever, °C 39 40 39 39 39 39 - Lymphadenopathy Yes Yes Yes Yes No Yes - Splenomegaly (thickness, cm) Yes, 5.3 cm Yes, 13.5 cm No Yes, 5.3 cm Yes, NA Yes, 4.3 cm - Hemophagocytosis BM BM - - - BM - EBV-DNA, copy/mL 2.24 × 105 2.11 × 103 4.74 × 104 4.44 × 106 1.71 × 106 1.69 × 106 - Serum EBV, copy/mL 4.94 × 103 <500 4.38 × 105 1.4 × 104 2.06 × 104 2.95 × 107 - Serology EA-lgG + - + NA NA NA - EBVC-IgG + + + NA + NA - EBVC-IgM - - - NA - ± - EBNA-IgG + + + NA + NA - White blood cells, 109/L 1.52↓ 3.4↓ 0.91↓ 1.43↓ 4.25 4.69 3.5–9.5 Neutrophils, 109/L 0.84↓ 2.02 0.94↓ 0.80↓ 2.07 4.33 1.8–6.3 Lymphocytes, 109/L 0.47↓ 0.7↓ 0.16↓ 0.50↓ 0.28↓ - 1.1–3.2 Hemoglobin, g/L 128↓ 77↓ 61↓ 88↓ 104↓ 119↓ 130–175 Platelets, 109/L 116↓ 27↓ 3↓ 192 520 23↓ 125–350 ESR, mm/H - 27↑ 5 - 4 - 0–15 CRP, mg/L 4.7 118.4 152.1 38.6 5.0 128.34 - ALT, U/L 68↑ 11 65↑ 56↑ 83↑ 120.1↑ <41 AST 58↑ 39 141↑ 58↑ 98↑ 310↑ <40 LDH, U/L 222 1519↑ 1627↑ 290↑ 672↑ 2465.3↑ 135–225 PT, s 13.3 16↑ 22.7↑ 15.1↑ 13.8 13.5 11.5–14.5 APTT, s 42.6↑ 48.2↑ 90.9↑ 57.4↑ 49.7↑ 32.4 29–42 Triglycerides, mM/L - 3.91↑ - 1.61 2.83↑ - <1.7 Fibrinogen, g/L 1.69↓ 3.52 0.6↓ 2.15 1.95↓ 2.46 2.0–4.0 Ferritin, μg/L 232.1 3095.8↑ 14323↑ 3242.5↑ 205.1 41776↑ 30–400 sCD25, U/mL 427 >7500↑ >7500↑ - 6049↑ >7500↑ 223–710 IL-1β, pg/mL 5.2↑ <5 6.6↑ - 9.4↑ <5 <5 IL-6, pg/mL <1.50 45.61↑ 209.2↑ 74.07↑ 30.25↑ 3.95 <7.0 IL-8, pg/mL 16 21.5 86↑ - 41.2 128↑ <62 IL-10, pg/mL <5.0 830↑ 635↑ - 82.6↑ 5.5 <9.1 TNF-α, pg/mL 8.9↑ 33.1↑ 18.5↑ - 91.6↑ 10.8↑ <8.1 IgA, g/L 2.26 1.11 - - 2.05 - 0.82–4.53 IgG, g/L 12.8 9.8 - - 22.3↑ - 5.4–15.3 IgM, g/L 0.87 0.51 - - 0.63 - 0.46–3.04 Complement C3, g/L 0.91 0.8 - - 0.93 - 0.65–1.39 Complement C4, g/L 0.46↑ 0.13↓ - - 0.85↑ - 0.16–0.38 Chemotherapy and response Anti-PD-1 antibody (CR) HLH-1994 regimen (NR);DEP (liposomal doxorubicin, etoposide, and methylprednisolone) regimen (NR) HLH-1994 regimen (NR) HLH-2004 regimen (PR);L-GMOX (gemcitabine, oxaliplatin and pegaspargase) regimen plus rituximab (CR) - HLH-1994 regimen plus rituximab (CR) - Allo-HSCT No Yes Yes NA Yes NA - Donor - HLA-identical donor from China's bone marrow bank HLA-identical donor (her brother) - HLA-haploidentical donor (his mother) - - CD34+ cells, ×106/kg - 5.16 6.42 - NA - - Outcome Alive Death (day 93 after transplantation) Death (day 10 after transplantation) NA Alive NA - Allo-HSCT = allogeneic hematopoietic stem cell transplantation, BM = bone marrow, CR = complete response, EA = anti-EBV early antigen, EBNA = EBV nuclear antigen antibody, EBVC = anti-EBV viral capsid antibody, HLA = human leukocyte antigen, NA = not available, NR = no response, PR = partial response.*Examinations listed here were performed within 7 d after the first admission or attack, and the test time was as close as enough to the first admission or attack. Figure 1.: Clinical examinations and immunological and functional phenotypes of patients with TNFRSF13B mutations. (A) Level of EBV-DNA copies in PBMC and in serum, sCD25 (normal range: 223–710 U/mL) and ferritin (normal range: 30–400 μg/L) of patient 2 after diagnosis. (B) Bone marrow biopsy showing a hemophagocytic cell (red arrow) from patient 2. (C) EBV-DNA copies quantification of patients' different cell types (PBMCs [gray], T [green], B [red], and NK [blue] cells) by real-time PCR. (D) Function of NK cells and CTLs of patients with TNFRSF13B mutations. The dashed line shows the lower limit of normal range (normal range of NK cell cytotoxicity: ≥15.11%, degranulation of resting NK cells: ≥5%, degranulation of stimulated NK cells: ≥40%, perforin expression of CTLs: ≥2%, perforin expression of NK cells: ≥84%, granzyme B expression of CTLs: ≥2%, granzyme B expression of NK cells: ≥78%). (E) Flow cytometry analysis of granzyme B-positive and perforin-positive cells in CD8+T cells and NK cells taken from patient 2. CD8+perforin+, CD3-CD56+perforin+, CD8+granzyme B+, and CD3-CD56+ granzyme B+ cells were gated and analyzed using CD3-PerCP Cy5.5, CD8-APC-Cy7, CD56-APC, PE-granzyme B, and FITC-perforin. CTL = cytotoxic T lymphocyte, EBV = Epstein-Barr virus, NK = natural killer, PBMCs = peripheral blood mononuclear cells, PCR = polymerase chain reaction, sCD25 = soluble CD25.2.2. Genetic findings To detect the possible molecular basis of the EBV-T/NK-LPDs, WES was conducted in these patients. Nucleotide substitutions, insertions, and deletions in the whole-exome regions of the human genome were detected. Very rare (minor allele frequency < 0.001), possibly damaging (according to prediction), and disease-related variants were identified as pathogenic genetic aberrations. Genes in the TNF-TNFR superfamily which involves in integrative immune responses were paid particular attention. Sequencing results revealed a series of heterozygous nonsynonymous variants in the TNFRSF13B in 6 patients (Table 1). Four of 6 mutations were located at the cysteine-rich domain (CRD)1 region of transmembrane activator and CAML interactor (TACI, encoded by TNFRSF13B), 1 mutation was located at the CRD2 region which interacts with ligands of TACI, 1 mutation was located at the part between CRD2 and the transmembrane region (Fig. 2A, B). The deletion frameshift mutation in TNFRSF13B carried by patient 1 (c.366delG, p.S123Vfs*31) had not been reported in the OMIM (http://www.omim.org) and ClinVar database (http://www.ncbi.nlm.nih.gov/clinvar), and probably impaired the expression of TACI due to the extremely truncated protein. The position at which the frameshift deletion in patient 2 (c.105delC, p.Glu36Lysfs*48, HGMD CD153877) occurred was reported to be disease-causing at HGMD (https://www.hgmd.cf.ac.uk/ac/index.php). The Sanger sequencing (Fig. 2C) indicated the same TNFRSF13B frameshift deletion of his son. Missense mutations carried by patients 3 (c.139T>A, p.Cys47Ser), 4 (c.226G>T, p.Gly76Cys), 5 (c.115T>C, p.Tyr39His), and 6 (c.124C>A, p.Pro42Thr) were also predicted to be possibly deleterious by the sorting intolerant from tolerant (http://sift.jcvi.org), Polyphen2 (http://genetics.bwh.harvard.edu/pph2), or MutationTaster (https://www.genecascade.org/MutationTaster2021/). The mutation in patient 6 was previously reported in a study on CVID.11 Taken together, all the 6 TNFRSF13B mutations were classified as likely pathogenic.12 Patient 3 carried a heterozygous NOD2 mutation which possibly contributed to her diarrhea similar to Crohn disease. Patient 2 also carried a mutated LYST (c.c.8624G>A, p.Arg2875His) causing nonselective vulnerability to EBV-HLH,13 while patient 5 also carried a heterozygous AP3B1 mutation (c.1234T>C, p.Tyr412His, 49.80%) which contributed to the defects in T/NK-cell degranulation in typical HLH. Apart from patient 2 and patient 5, the other 4 patients lacked a known genetic etiology clearly related to EBV-LPDs.Figure 2.: Genetic phenotypes of the patients with TNFRSF13B mutations. (A) Location of TNFRSF13B alterations in our patients. The numbers below represent amino acid positions. (B) Shown is a 3D model diagram indicating the locations of mutants in the TACI proteins. The figures were prepared via PyMOL (www.pymol.org). Most variants in TNFRSF13B were frameshift and missense variants. The numbers represent amino acid position. (C) Sanger sequencing of area surrounding the frameshift TNFRSF13B mutation (c.105delC, p. Glu36Lysfs*48) in a reference control subject and son of patient 2. TCAI = transmembrane activator and cyclophilin interactor.3. DISCUSSION In most benign cases of EBV infection, the target cells of EBV were B cells and epithelial cells while in some populations susceptible to EBV infections, fatal diseases can occur, indicating the underlying inborn defects of immunity. Impaired innate and/or adaptive response to EBV may lead to uncontrolled EBV replication and the EBV-T/NK-LPDs in the worst case, including the EBV-HLH and CAEBV-T/NK. The EBV infection was controlled by both innate and specific immune cells including NK cells and conventional T cells. Signaling via several members of the TNF-TNFR superfamily supports the CD8+ and CD4+ T-cell functions was reported to involve in EBV-specific responses, including the signals delivered by CD27, CD70, 4-1BB, and DcR3. Mutations in CD27 (TNFRSF7)14-CD7015 pathway and 4-1BB (TNFRSF9)16 were reported to be associated with severe, atypical EBV infections or EBV-related lymphoma by impairing the expansion and activation of CD8+ EBV-specific T cells. In this work, we report a series of novel heterozygous TNFRSF13B mutations in 6 patients diagnosed with EBV-T/NK-LPDs. Functional examinations of NK cells and CTLs revealed significantly low NK cytotoxicity in all 6 patients, indicating weak innate immunity when fighting against EBV. Therefore, it seemed reasonable that the altered immunity to EBV of these 6 patients were the complications of TNFRSF13B deficiency. TACI primarily provides signals for class switch recombination in B cells during T cell–independent antibody response.17,18 It also involves in the physiological processes of follicular helper T (Tfh) cells7 and Th17.8TNFRSF13B mutations were mainly reported in CVID and systemic lupus erythematosus (Supplementary Table 2, https://links.lww.com/BS/A82), indicating a complex imbalance of immunity brought by TACI deficiency.19 The paradox was that the TACI variants were not reported previously to influence NK cells or CTLs,17,20 while the impaired function of these cells was found in the patients. The exact mechanism by which the T or NK cells are infected by EBV has also not been thoroughly investigated. Here we offered 3 speculations on how TNFRSF13B mutations predispose T and/or NK cells to EBV infection according to previous studies: the TACI protein involves in the EBV binding process (similar to the molecule CD21)21; impaired TACI expression or TACI variants influence the susceptibility of myeloid lineage progenitors to EBV infection22; TNFRSF13B mutations impair the activity of T and/or NK cells in controlling EBV replication at the secondary lymphoid organs, resulting in the high load of virus and subsequent infection.23 However, considering of the little expression of TACI on CD16dim NK cells and CTLs (Supplementary Figure 2, https://links.lww.com/BS/A83), their vital role in preventing the EBV infection and eliminating the EBV-infected B cells, and the significantly higher expression of TACI on EBV-transformed B cells compared with the normal B cells (data not shown), this work puts forward the hypothesis that TACI may implicate in the T/NK cell–mediated responses against EBV via the ligand–receptor interactions in EBV-infected B cells. Interestingly, activated NK cells produce soluble B cell activating factor from the TNF family (BAFF), the TACI ligand, when encountering CLL cells. High level of local BAFF protects CLL cells from the NK cell–mediated lysis via unknown mechanisms,24 also suggesting the possible interactions among the TACI-BAFF network, EBV-infected B cells, and T/NK-cell killing capacity. Further experiments on TACI in our laboratory are underway, and are expected to facilitate the discovery of novel targets against atypical EBV infections. Further experiments with TACI in our laboratory are underway and are expected to facilitate the discovery of novel targets against atypical EBV infections. Another concern has been the heterozygous LYST mutation found in patient 2, and the heterozygous AP3B1 mutation found in patient 5. Defects in both of the 2 genes were associated with the initiation of primary HLH.25 Deleterious mutation in LYST and AP3B1 disturbs the vesicle trafficking process in the release of cytolytic particles, interfering the T/NK cell–mediated antiviral responses. However, primary HLH is usually characterized by the homozygous or compound heterozygous mutations, relatively early onset, worse prognosis, and no significant relevance to EBV infection, which was obviously not met in patients 2 and 5. The reasonable theory may be that with the trigger of EBV infection, heterozygous LYST/AP3B1/TNFRSF13B mutation acts as a predisposing allele for EBV-HLH development. The prognostic value of TNFRSF13B mutation is difficult to evaluate due to the very small cohort and the loss to follow-up beyond our control. Judging from the outcome of patients, however, patient 1 who carried only the TNFRSF13B missense mutation suffered the disease with much milder symptoms and better prognosis, compared with patients 2 and 3 who carried both HLH-associated genetic mutations and TNFRSF13B mutation. This interesting trend could also be interpreted by the theory about "predisposing allele." Taken together, our findings expand the range of genes routinely screened for in patients with EBV-T/NK-LPDs by reporting a novel set of TNFRSF13B mutations in 6 patients with EBV-HLH and/or CAEBV-T/NK, outlining the need to investigate the role of TNF-TNFR superfamily members in the pathogenesis of intractable EBV infections. ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (No. 81770211 and No. 82270203 to M.X.). We thank all the faculty and staff in the Clinical and Laboratory Unit of the Department of Hematology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology for their clinical and technical support. AUTHOR CONTRIBUTIONS X.D. and K.S. conceived the project. X.D. and T.G. drafted the manuscript, collected the statistics, and drew the figures. K.S. and W.M. discussed the manuscript. J.W. revised the manuscript. H.L. and J.G. performed part of the experiments. M.X. provided guidance and approved the version to be submitted. All authors contributed to the article and approved the submitted version. ETHICAL APPROVAL The studies involving human participants were reviewed and approved by the Medical Ethics Committee of Tongji Hospital, Tongji Medical College of HUST. The number of ethical approval was TJ-IRB20200706. The ethics committee waived the requirement of written informed consent for participation.