A Novel Compound Heterozygous Mutation in the IL12RB1 Gene Causes Susceptibility To Mycobacterium Tilburgii Infection

Abstract

Mendelian susceptibility to mycobacterial disease (MSMD) is a rare clinical syndrome that is characterized by selective vulnerability to intracellular pathogens. Deficiency in IL12RB1 is the most common type of MSMD but the heterogeneity of its clinical Manifestation Makes precise diagnosis difficult. Here, we report a previously healthy 29 year-old woman who had suffered from disseminated infection with Mycobacterium tilburgii, which is a rare, unculturable environmental mycobacteria, for over 2 years. We used whole exome sequencing to detect a novel compound heterozygous variant in the IL12RB1 gene. Immunological analysis of the patient’s peripheral lymphocytes showed a barely detectable level of IL-12Rβ1, a reduced population of follicular helper T (Tfh) cells and impaired production of IFN-γ in response to IL-12/IL-23 stimulation. Metagenomic next-generation sequencing was used to identify the causative pathogen and to analyze drug susceptibility. The infection was contained by a combination of anti-mycobacterial drugs and IFN-γ supplementary treatment. An RNA-seq analysis, using follow-up blood samples, revealed the limited success of these treatments over 6 months. Our findings support the screening for inherited immunological problems in patients with difficult-to-treat mycobacterial infections. The suboptimal response to prolonged anti-mycobacterial drugs and IFN-γ supplementation warrants the development of novel therapeutic strategies for MSMD patients.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10875-025-01930-x.

Introduction

Mendelian susceptibility to mycobacterial disease (MSMD) is a rare congenital genetic condition [1] that was first exposed in 1996 with the identification of an IFNGR1 gene mutation as the hereditary cause of newborn fatal Bacillus Calmette–Guérin (BCG) infection [2]. According to the IUIS Phenotypic Classification, MSMD is classified as “defects in intrinsic and innate immunity” [3]. Three forms of inheritance are associated with the disease: X-linked recessive [4], autosomal dominant [5] and autosomal recessive [2]. Currently, 22 genes are recognized as being associated with MSMD. These genes are CCR2, CYBB, IFNG, IFNGR1, IFNGR2, IL12B, IL12RB1, IL12RB2, IL23 IRF1, IRF8, ISG15, JAK1, MCTS1, NEMO, RORC, SPPL2A, STAT1, TBX21, TYK2, USP18 and ZNFX1 [6, 7]. Defects in these genes render affected individuals selectively vulnerable to a range of intracellular pathogens, mainly by diminishing Interferon-Gamma (IFN-γ) production, response to IFN-γ, or both [8–10]. Forty-six distinct genetic reasons can be classified according to various routes of inheritance [11]. However, only 50% of individuals obtain a definitive diagnosis [12]. In addition, neutralizing autoantibodies against cytokines, such as interferon-γ (IFN-γ), have been implicated as a cause of the IEI phenotype [13]. Despite the absence of germline mutations, IFN-γ autoantibody-mediated disease still mimics the clinical phenotype of MSMD. In particular, IFN-γ neutralizing autoantibodies can explain up to 5% of MSMD cases [14].

The IL12RB1 gene is one of the most prevalent causative genes of MSMD [8]. Increased susceptibility to multiple pathogens, such as Salmonella, Candida and Mycobacteria, has been reported in patients with IL12RB1 deficiency [15] and the infection spectrum is broader, when compared to MSMD caused by other gene defects [15]. Currently, it is difficult to develop a sensitive clinical diagnostic criteria for IL12RB1 deficiency because of the highly variable and non-specific symptoms and the incomplete clinical penetrance [16]. The prognosis for patients with IL12RB1 deficiency is heterogeneous [15]. Long-term antimicrobials are the main treatment and they are used in combination with surgery for those with recurrent or drug-resistant infection [17]. High-dose IFN-γ replacement has been shown to be a promising treatment in some studies [18–20]. However, the optimal diagnosis and treatment strategy for IL12RB1 deficiency remains to be developed.

Here, we report the case of a young woman who presented with abdominal pain, multifocal lymphadenopathy and an initial diagnosis of disseminated Mycobacterium tilburgii infection. Genetic analysis later identified a novel compound heterozygous IL12RB1 mutation. Treatment response was followed up using RNA sequencing. To the best of our knowledge, this is the first Chinese case of a patient with IL12RB1 deficiency who was infected with M. tilburgii.

Results

Case Presentation

The patient was a 29-year-old, previously healthy Chinese woman from a nonconsanguineous family. Since November 2022, she had experienced recurrent dull upper abdominal pain. Her laboratory tests showed no obvious abnormalities apart from a slightly increased C-reactive protein concentration of 26.8 mg/L. Abdominal imaging revealed multiple lymphadenopathies in the retroperitoneum and mesenteric root (the largest was 41 × 28 mm), and splenomegaly (113 × 53 mm). In the following 2 months, the pain recurred and she developed diarrhea and a low-grade fever of 37.7 °C. Positron emission tomography-computed tomography indicated multiple enlarged lymph nodes in the II-V areas of the bilateral neck, right axilla, abdomen and retroperitoneum, with increased standard uptake (Fig. 1A). An abdominal lymph node puncture was performed and malignant disease was ruled out. Surprisingly, a peripheral blood sample used for metagenomic next-generation sequencing (mNGS) revealed the presence of M. tilburgii (1,593 reads), which is a rare pathogen. Hence, disseminated nontuberculous mycobacterial (NTM) infection was diagnosed.

Fig. 1.

Clinical imaging data of the patient. (A) PET/CT showed that the patient had multiple lymphadenopathy. (B) Abdominal enhanced CT scans were performed on March 8 2023 (up) and May 19, 2023 (down), and indicated that there were still multiple enlarged lymph nodes in the retroperitoneum and mesentery, some of which were slightly larger than the previous scan, and splenomegaly. (C) Gastroenterostomy showed antral gastritis (congestive exudative type, moderate) with bile reflux, colonoscopy showed duodenal mucosal lesions and terminal ileal mucosal lesions. The results of gastroenteroscopy in 2024 (right) showed that the villi of the descending duodenum were shortened, coarse, and milky white; the terminal ileum mucosa showed cobblestone, granular, and nodular changes. The lesions have partially improved compared to the initial stage of the disease

The patient started treatment with clarithromycin, ethambutol, levofloxacin/moxifloxacin, rifampicin and amikacin. Her symptoms were relieved after treatment and the erythrocyte sedimentation rate and C-reactive protein levels decreased. However, she still had intermittent abdominal pain and follow-up abdominal enhanced computed tomography scans revealed that the splenomegaly and multifocal necrotic lymphadenopathy in the mesentery and retroperitoneum remained, and in some cases were marginally larger than on the previous scans (Fig. 1B). Multiple fecal acidfast staining on the 30th of May and the 27th of June 2023 were positive, whilst mycobacterial cultures were negative. A gastrointestinal endoscopy revealed duodenal and terminal ileum mucosal lesions (Fig. 1C). Biopsy histopathology showed a large number of foamy tissue cells had infiltrated the lamina propria of the mucosa. Acid-fast staining showed intracellular, red, slightly curved rods, which suggested the possibility of mycobacterial infection. Mycobacterium tilburgii was detected by mNGS (406,417 reads). Repeated peripheral blood sample mNGS analyses showed a decreased read count of M. tilburgii (15 reads), in comparison with the previous test. Follow-up fecal acid-fast staining remained positive. The patient’s current antibiotic regimen is moxifloxacin, azithromycin, rifampin, amikacin, ethambutol and linezolid.

We traced the patient’s initial endoscopy report (February 2022), which indicated multiple mucosal lesions in the duodenum, whilst the pathological results suggested xanthoma. This indicated that the patient had already been infected for 9 months before the onset of symptoms.

Identification of a Novel Pathogenic Compound Heterozygous Mutation in IL12RB1

Mycobacterium tilburgii is a rare NTM with low virulence, with reports previously limited to immunocompromised individuals [21]. Our patient was negative for an HIV serology test but a detailed medical history revealed BCG-osis following vaccination, which resolved spontaneously during infancy, and recurrent fevers during Childhood. Her father reported a history of tuberculosis at the age of 40, while her mother had suffered from lung diseases and recurrent fevers since childhood. Primary immunodeficiency, particularly MSMD, was therefore suspected. Whole exome sequencing (WES) was conducted and three rare heterozygous non-synonymous variants were identified in IL12RB1 in the patient. Sanger sequencing of the family then confirmed that one nonsense c.1561 C > T (NM_005535: p.R521X) variant and one missense c.271G > A (p.A91T) variant were inherited from her mother, whilst a nonsense c.847 C > T (p.R283X) variant was inherited from her father (Fig. 2A). In accordance with the ClinVar database, both nonsense variants were classified as stopgain pathogenic variants (p.R521X, CADD score: 41; p.R283X, CADD score: 28.5), while the missense p.A91T (CADD score:7.03) variant was recorded as benign. Thus, IL12RB1 deficiency, caused by the R521X/R283X compound heterozygous mutation, was suspected in this patient.

Fig. 2.

The loss-of-function variant and efficacy of recombinant IFN-γ treatment. (A) Family pedigree of the patient and Sanger sequencing results for the novel compound heterozygous mutation in the family. (B) Level of IL12Rβ1 protein on the cell surface of the patient and her parents (C) The IFN-γ production of PBMCs from six healthy controls and the patient was measured under the same conditions, which included no stimulation (NS), BCG lysate alone, IL-12, IL-23, or a combination. The positive control was with PMA/Inomycin. (D) Volcano plot of differential gene expression before and after 6 months of interferon treatment, with upregulated genes in red and downregulated genes in green. Interferon-related genes are marked in the figure. (E) KEGG pathway enrichment analysis of differentially expressed genes obtained from samples before and 6 months after interferon treatment. Top pathways included the IL-17 signaling pathway and osteoclast differentiation

The Compound Heterozygous IL12RB1 Mutation Abolished IL-12Rβ1 Protein Expression and Impaired IL-12/IL-23 Induced IFN-γ Production

Since IL12RB1 deficiency and AIGA can both lead to increased vulnerability to NTM, we conducted a series of immunological assessments to confirm the pathogenicity of the compound heterozygous IL12RB1 mutation. We first assessed the IL-12Rβ1 protein levels using flow cytometry and observed a barely detectable level of IL-12Rβ1 in peripheral lymphocytes from the patient. This level was significantly lower than in her parents (Fig. 2B) and supported the hypothesis that the R521X/R283X compound heterozygous mutation disrupts the production of IL-12Rβ1. To investigate how the IL12RB1 mutation affects the production of IFN-γ and the IL-12/23- IFN-γ axis, we stimulated the patient’s peripheral lymphocytes with IL-12 and IL-23. The results revealed severely compromised IFN-γ secretion by the patient’s lymphocytes in response to both cytokines, either with or without the addition of BCG lysate, when compared with healthy controls. The patient’s PBMCs’ capacity to generate IFN-γ upon PMA stimulation was noticeably less than that of healthy controls (Fig. 2C). Together, these results suggested that the R521X/R283X compound heterozygous mutation can be interpreted as a loss-of-function variant that interferes with the downstream signaling of IL-12/IL-23 and IFN-γ production by malforming the receptor complex.

As alterations in immune cells have been reported in previous IL12RB1 deficiency cases [22], immunophenotyping was also performed using the patient’s peripheral blood. Results showed a slight decrease in total CD4 + T cell count but a significantly decreased population of Tfh cells. The absolute counts of Th1 and Th17 in patients were close to the lower limit of the reference range. The numbers of CD4 + effector T cells and CD8 + effector T cells were low, though in the normal range. And the CD4/8 ratio (0.95) inverted. The number of B cells was slightly reduced, and its differentiation and activation status were altered, as the proportion of transitional B cells and several memory B cell subsets were simultaneously elevated. As for innate immune cells, only a decrease in dendritic cells was observed (Table 1).

Table 1.

Deep immunophenotype analysis of the patient

Proportion (%)	Patient (Reference range#)	Absolute count	Patient (Reference range#)
T cells
γδT/T	13.55% (2.79–23.07%)	γδT	133.54 (42–417)
Vγ9+/γδT	92.73% (12.12–95.04%)	Vγ9 + γδT	122.90(8-341)
αβT/T	86.32% (76.56–97.13%)	αβT	844.17(786–3088)
CD4+/CD3 + T	46.93% (52.21–76.15%)	CD4 + T	396.21(425–1612)
CD8+/CD3 + T	49.19% (19.97–44.50%)	CD8 + T	415.24(296–1374)
Th1/CD4 + T	17.15% (7.96–20.79%)	Th1	67.96(65–233)
Th2/CD4 + T	20.61% (11.15–24.41%)	Th2	81.65(77–238)
Th17/CD4 + T	9.01% (7.91–19.70%)	Th17	35.70(55–175)
Tfh/CD4 + T	1.98% (5.25–10.42%)	Tfh	7.84(36–132)
Treg/CD4 + T	7.11% (5.65–13.26%)	Treg	28.17(39–214)
CD28+/CD4 + T	98.46% (83.97–99.97%)	CD28 + CD4 + T	390.11(401–1604)
CD38+/CD4 + T	71.30% (25.82–73.18%)	CD38 + CD4 + T	282.52(183–992)
HLA-DR+/CD4 + T	13.62% (6.60-21.45%)	HLA-DR + CD4 + T	53.95(45–246)
Naïve/CD4 + T	53.19% (14.53–63.55%)	Naïve CD4 + T	210.75(103–947)
Central memory/CD4 + T	24.45% (17.84–39.62%)	Central memory CD4 + T	96.87(118–375)
Effector memory/CD4 + T	20.57% (12.73–49.68%)	Effector memory CD4 + T	81.50(105–373)
Effector/CD4 + T	1.79% (0.61–10.23%)	Effector CD4 + T	7.10(5–68)
ICOS-PD-1+/CD4 + T	16.89% (8.89–36.06%)	ICOS-PD-1 + CD4 + T	66.93(71–225)
ICOS + PD-1+/CD4 + T	2.97% (1.15–10.84%)	ICOS + PD-1 + CD4 + T	11.78(9–74)
ICOS + PD-1-/CD4 + T	0.97% (0.76–14.84%)	ICOS + PD-1- CD4 + T	3.84(6-119)
ICOS-PD-1-/CD4 + T	79.16% (58.45–87.35%)	ICOS-PD-1- CD4 + T	313.66(291–1311)
CD28+/CD8 + T	73.96% (36.35–95.95%)	CD28 + CD8 + T	307.09(122–978)
CD38+/CD8 + T	89.28% (17.71–55.13%)	CD38 + CD8 + T	370.71(75–605)
HLA-DR+/CD8 + T	40.30% (10.12–54.47%)	HLA-DR + CD8 + T	167.33(36–273)
Naïve/CD8 + T	55.01% (12.98–66.59%)	Naïve CD8 + T	228.41(53–690)
Central memory/CD8 + T	2.15% (0.81–6.28%)	Central memory CD8 + T	8.92(3–28)
Effector memory/CD8 + T	19.18% (8.38–32.75%)	Effector memory CD8 + T	79.64(30–213)
Effector/CD8 + T	23.66% (11.05–67.77%)	Effector CD8 + T	98.26(58–489)
ICOS-PD-1+/CD8 + T	10.59% (8.38–32.75%)	ICOS-PD-1 + CD8 + T	43.97(32–138)
ICOS + PD-1+/CD8 + T	0.60% (0.12–3.23%)	ICOS + PD-1 + CD8 + T	2.51(1–14)
ICOS + PD-1-/CD8 + T	0.51% (0.15–2.49%)	ICOS + PD-1- CD8 + T	2.14(1–11)
ICOS-PD-1-/CD8 + T	88.29% (64.45–90.98%)	ICOS-PD-1- CD8 + T	366.62(246–1250)
B cells
B cell/LYM	10.59% (9.23–22.94%)	B cell	127.78(157–762)
Early transitional B/B	8.49% (1.10–5.66%)	Early transitional B	10.85(2–20)
Late transitional B/B	70.63%(70.39–91.01%)	Late transitional B	90.25(131–558)
Transitional B/B	16.70%(0.33–4.99%)	Transitional B	21.34(1–25)
Naïve B/B	76.42%(51.29–89.88%)	Naïve B	97.65(124–493)
Margin zone B/B	7.64%(2.02–9.35%)	Margin zone B	9.76(5–45)
Memory B/B	12.30%(2.51–23.46%)	Memory B	15.72(12–132)
Switched B/B	4.88%(1.14–15.17%)	Switched B	6.23(6–60)
IgM memory B/B	4.22%(0.67–3.98%)	IgM memory B	5.40(2–16)
IgG memory B/B	5.03%(0.12–2.59%)	IgG memory B	6.43(0–6)
Atypical memory B/B	3.96%(0.58–3.86%)	Atypical memory B	5.06(1–18)
Plasmablasts/B	0.89%(0.31–3.71%)	Plasmablasts	1.14(1–11)
Innate immune cells
CD14 + Monocyte/non-TBNK	97.56%(80.18–96.12%)	CD14 + Monocyte	264.73(210–1160)
Classical monocyte/monocyte	81.10%(43.05–93.96%)	Classical monocyte	214.69(184–543)
Non-classical monocyte/monocyte	11.85%(4.95–56.70%)	Non-classical monocyte	31.36(15–658)
MDSC/monocyte	6.16%(0.11–4.47%)	MDSC	16.30(0–16)
Dendritic cells/non-TBNK	0.24%(1.72–13.77%)	Dendritic cells	1.22(7–85)
pDCs/DC	7.02%(2.03–26.14%)	pDCs	0.09(0–6)
mDCs/DC	40.35%(1.58–51.57%)	mDCs	0.49(1–9)
Basophil/non-TBNK	0.12%(0.01–2.56%)	Basophil	0.58(0–12)
NK cell/LYM	4.49%(1.61–20.43%)	NK cell	54.23(40–542)
CD56 bright CD16-/NK	18.85%(0.79–17.50%)	CD56 bright CD16- NK	10.22(2–34)
CD56 dim CD16+/NK	67.39%(62.48–93.58%)	CD56 dim CD16 + NK	36.54(25–477)
CD56-CD16+/NK	2.76%(0.84–7.55%)	CD56-CD16 + NK	1.50(1–21)
CD56 dim CD16-/NK	11.00%(2.40–14.90%)	CD56 dim CD16- NK	5.96(6–18)

#Reference ranges were established using a small in-house cohort of healthy adults

Phenocopy of primary immunodeficiency was also taken into consideration. We used ELISA to measure the amount of anti-IFN-γ autoantibody (AIGA) in the patient’s blood and obtained a positive result, with an OD of 0.83 (cut-off is 0.30). However, the patient had a valid IFN-γ release assay (T-SPOT.TB) result, with the positive control having more than 20 spots forming cells. These results suggested the co-existence of MSMD and AIGA associated adult-onset immunodeficiency (AOID).

Molecular Drug Resistance Analysis of M. tilburgii using mNGS

The patient exhibited a partial response to anti-mycobacterial treatment, as her abdominal lesions did not progress but follow-up microbacterial tests continued to be positive. Given the unculturable nature of M. tilburgii, knowledge on its drug susceptibility features is very limited [23]. We therefore conducted a molecular drug resistance analysis using mNGS data from her colon biopsy, prior to anti-mycobacterial treatment. We used an online integrative database (https://card.mcmaster.ca/home) and excluded synonymous mutations [24, 25]. Only mutations with a coverage over 15x were considered valid. We identified mutations reported to be associated with resistance to rifamycin, isoniazidlike antibiotic, mupirocin-like antibiotic, macrolide and lincosamide antibiotic in the genome of M. tilburgii (Table 2).

Table 2.

Drug sensitivity results based on tissue mNGS results

Gene	Drug class	Reads
Bifidobacterium adolescentis rpoB mutants conferring resistance to rifampicin	rifamycin antibiotic	15
efpA	rifamycin antibiotic; isoniazid-like antibiotic	28
Bifidobacterium bifidum ileS conferring resistance to mupirocin	mupirocin-like antibiotic	29
tlrC	macrolide antibiotic; lincosamide antibiotic	39
RbpA	rifamycin antibiotic	47
rpoB2	rifamycin antibiotic	174

Adjustment of Regimen after the Definitive Diagnosis and Performance of RNA-seq on follow-up Samples

In January 2024, we initiated recombinant human IFN-γ (rhIFN-γ) replacement therapy, with a dosage of 1 million units, by subcutaneous injection every other day, to accelerate infection resolution. However, 3 months later, due to changing drug accessibility and side effects, the dosage was reduced to 1 million units twice a week. To monitor the presence and induction of AIGA, we conducted a follow-up assessment of AIGA during the cytokine therapy period. The results revealed a decreased titer of AIGA, as the OD level dropped from 0.83 to 0.34, with the infection gradually being controlled.

To better investigate the effect of rhIFN-γ treatment, we conducted bulk transcriptome sequencing using a series of samples, which included blood collected prior to treatment, 1-month and 6-month follow-up samples. No significant alterations in gene expression were observed in the first month. However, after 6 months of rhIFN-γ treatment, alterations in the expression of a number of genes were noticed (Fig. 2D). A Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis showed that the differentially expressed genes were enriched in the IL-17 pathway and osteoclast differentiation (Fig. 2E).

Methods

Ethics Statement

Written informed consent was obtained from the patient and her parents. The study received approval from the institutional ethics review board of Huashan Hospital, Fudan University, and was conducted in accordance with the principles outlined in the Declaration of Helsinki.

Whole Exome Sequencing and Sanger Sequencing

Whole exome sequencing was carried out on an Illumina HiSeq X Ten platform (2GB, 2 × 150 bps paired-ends, 400x sequencing depth) using genomic DNA extracted from the whole blood of the patient. A bioinformatics analysis was performed to detect rare sequence variants using the following databases: 1000 Genomes Project database, dbSNP, Exome Aggregation Consortium and Genome Aggregation database. And we used the Genome Reference Consortium Human Build 37 (NCBI GRCh37) as the reference genome. Functional prediction of variants was performed using the following bioinformatic tools: CADD, REVEL, SIFT, PolyPhen-2 and MutationTaster. Sanger sequencing was used to confirm the variant of the IL12RB1 gene using genomic DNA extracted from the whole blood of the patient and her parents, using the following primers:

271G-F: 5’ CAGTGCATGCTGGGTAAATAAC 3’;

271G-R: 5’ TCCCAGGACTCAGGGTTGC 3’;

847 C-F: 5’ CACGTTGTAGGCAGCACCC 3’;

847 C-R: 5’ GTCTGCCTATGGGATGATGAGTG 3’;

1561 C-F: 5’ GAGTTGTAGAGCTGCAGGGTCAC 3’;

1561 C-R: 5’ CCTGGGTGACAGAATGAAACTC 3’.

Deep Immunophenotyping of Leukocytes using Spectral and Flow Cytometry. Flow Cytometry Detection of Cell Surface IL-12Rβ1

We obtained the patient’s whole blood and stained the peripheral blood using flow cytometry to assess the patient’s lymphocyte subsets. The table in the appendix lists specific antibodies. We assessed the level of IL-12Rβ1 in peripheral blood mononuclear cells (PBMC).

Cytokine Detection in Serum from Simulated PBMC Using ELISA

We plated 5 × 10^5 cells per well in 96-well U-bottomed plates. Cells were stimulated with IL-12 (5 ng/ml, R&D Systems), IL-23 (20 ng/ml, R&D Systems) or BCG lysate (5 ug/ml) for 48 h, followed by 50 ng/mL of PMA and 1 µM ionomycin for 5 h. We detected IFN-γ using the ELISA MAX™ Deluxe Set Human IFN-γ (BioLegend).

RNA-seq Analysis

We collected whole blood samples from the patient before treatment, 1 month and 6 months after treatment. The detection and quantification of RNA were performed using Qubit4.0 (thermofisher, USA). The PE150 sequencing was performed using Illumina’s Novasek platform. All FASTQ files passed quality control and were aligned with the reference genome using STAR and FeatureCount. We used StringTie software for RNA quantification. The software used for differential expression was edgeR, based on the negative binomial distribution model. The differentially expressed gene were selected using FDR < 0.05 and |log2FC|≥1. Cluster Profiler was used for the Gene Ontology (GO) and KEGG pathway enrichment analyses. Fisher’s exact test was used for calculations. The corrected p-value threshold was set at 0.05.

Discussion

There have only been 16 confirmed cases of infection since M. tilburgii was originally discovered in 1995 [21, 26]. This bacteria has recently been verified as a member of the M. simiae complex [27]. Most patients with this infection suffer from intracellular immunodeficiency, mainly HIV. Other factors include idiopathic CD4⁺ T lymphocytopenia, steroid use and IL12RB1 deficiency [28–33]. Accurate drug sensitivity data have been impossible to obtain because the bacteria cannot be cultivated. In early instances of the infection, various therapeutic plans were used, dependent on specific situations. The initial treatment regimen was usually based on the M. avium complex, which consists of clarithromycin and ethambutol, combined with rifampicin and/or quinolones. In certain cases, surgery was necessary [23, 29]. Rescue medicine therapy options included aminoglycosides, quinolones or linezolid. Treatment was occasionally coupled with adjuvant immunotherapy, using interleukin-2 or IFN-γ [32, 34, 35]. We examined the drug resistance of this bacterium using tissue mNGS. This is a different method of analysis than that used in the Resistance Gene Identifier version 5.0.5 database [27]. Consistency of the mNGS method with conventional drug resistance tests is unknown and requires further validation. Therefore, the molecular drug resistance analysis of M. tilburgii can currently only provide very limited information.

The IL12RB1 gene is the most prevalent pathogenic gene known to cause MSMD, with more than 200 families identified as having a pathogenic variant in this gene. The clinical manifestations of variants in this gene are highly variable [15, 36]. This report discusses a novel compound heterozygous mutation that results in nonsensemediated decay of the transcript and subsequent loss-of-function. In previous large cohorts, 78% of probands vaccinated with BCG acquired BCG disease [15]. Our patient had suspected BCG-itis in her childhood and she developed a disseminated infection with low-virulent mycobacteria in adulthood. A possible explanation is that the penetrance of the disease increases with age [15]. In previous reports, the infection spectrum of patients with IL12RB1 deficiency is wider than that of patients with variants in other pathogenic genes. In addition to mycobacteria, Salmonella, chronic mucocutaneous candidiasis (CMC) and Leishmania have also been reported. Autoimmune manifestation and inflammatory bowel disease (IBD) may also coexist. It is clear that genetic heterogeneity is responsible for the heterogeneous clinical phenotypes of infectious diseases.

AOID, mediated by anti-IFN-γ autoantibodies (AIGA), is thought to be an autoimmune phenotypic replication of innate genetic defects in the IL-12/IFN-γ axis, with a high occurrence in Asians [37]. Disseminated opportunistic infection is the most prevalent characteristic of AIGA, with mycobacterial infection accounting for approximately 85% of cases [38–40]. The production of AIGA is believed to be genetically predisposed and the HLA class II alleles, DRB1*16:02–DQB1*05:02 and HLA-DRB1*15:02–DQB1*05:01, may be strongly associated with the progress of AIGA. However, the penetrance in carriers is low [41–43] and the mechanism has been unknown until now [44]. In contrast with MSMD, most of the 600 reported cases were adults (mean age 55 years, with approximately 87.6% of patients aged 40–87 years) [45]. Throughout our study, the patient’s anti-interferon antibody titer fluctuated. However, since neutralization assays were not performed in this case, it cannot be assumed that the detected AIGAs were functionally relevant or contributed to the patient’s disease, but one point was definitive: the application of exogenous cytokines did not lead to the production of neutralizing antibodies [12]. Given the similarities between the two diseases, the differential diagnosis of the two should be considered when encountering complicated and refractory mycobacterial infections.

As the primary CD4⁺ T cell subset that assists B cells in generating antibody responses, Tfh are crucial for humoral immunity against mycobacteria. The Tfh cells, together with B cells, form the germinal center in which the B cells will differentiate into either memory B cells or long-lived plasma cells [46]. Nathalie et al. found that IL12RB1 deficient individuals suffer from substantially less circulating memory Tfh and memory B cells, which causes hypoimmunoglobulinemia. The main mechanism behind this is the association between the IL-12-STAT4 axis and the development and functions of Tfh [22]. We detected a reduction in Tfh in our patient but her immunoglobulin levels were high and her plasma cells showed no discernible abnormalities. One reasonable explanation was that less IL-12-induced IFN-γ production results in a qualitative change in the Tfh cytokine repertoire, with reduced IFN-γ mediated suppression of B cell responses [47]. There was no doubt that the deficiency of IFN-γ was the primary trigger of the patient’s continuous infection. However, in patients with hereditary IL-12Rβ1 loss, type 1 immunity has been reported to retain some function [48]. Diana et al. found that patients with IL12RB1 deficiency had significantly fewer circulating T follicular regulatory cells and produced higher levels of anti-actin autoantibodies, in vivo [49]. We did not monitor this subpopulation of cells at the time, but there was no increase in autoantibodies. IFN-γ-producing cell subsets have been more thoroughly categorized in earlier investigations. Following PMA stimulation, the proportion of cells generating IFN-γ in Mucosal-Associated Invariant T cell, Vδ2 + γδ T, NK cells, and CD4 + T cells in IL-12Rβ1-/-patients did not appear to be appreciably different. However, upon PMA stimulation, our patient’s supernatant IFN-γ levels were noticeably lower than those of healthy individuals. This might be because the mutation is heterogeneous, producing a range of outcomes [50].Overall, inborn errors of immunity caused by IL12RB1 deficiency are highly heterogeneous and it is difficult to define the disease by a fixed immune phenotype.

Hematopoietic stem cell transplantation (HSCT) is thought to be the most promising therapeutic option [51]. However, due to various limitations, HSCT is not the first choice of treatment. For patients with IL12RB1 deficiency, exogenous IFN-γ, in combination with antibiotics, can be used during the active phase of the disease to clear the mycobacteria [15, 18, 52]. Recently, the use of pegylated IFN-α2b in patients with an IL12RB1 defect has been reported to complement antimicrobial therapy and improve the quality of treatment [53] but clinical outcomes vary between patients [15]. In our case, analysis of the data after 6 months of interferon treatment showed limited results, in contrast to when used to treat IFNG deficiency [54]. The IL-17 A protein is a cytokine that protects against mycobacteria infection in the host [55]. Research has shown that IL-17 F and IL-17 A can trigger the recall response and effector function of Vγ2Vδ2 T lymphocytes specific to tuberculosis phosphoantigens in nonhuman primates, following Mycobacterium tuberculosis infection and BCG vaccination [56]. During our study, the information we acquired fell short of our expectations and we assumed that not all patients with IL12RΒ1 deficiency will benefit from interferon therapy. It is possible that the credibility of the results was diminished by the limitation of samples and we may need to prolong the treatment period [57]. The symptoms of the patient described are now controlled well. Clinicians should consider that, even though there are numerous treatments available for this type of disease, each one has certain inherent drawbacks that limit the potential for broader usage.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

M.Q. Qian, J.Y. Zhou, P.D. Chen: conceptualization, investigation, formal analysis, writing-original draft; X.C. Chen, N.Jiang: investigation, formal analysis; H.X. Xu, Y.X. Yang, Q.L. Yang, F.R. Zhou, X. Lin: data collection, investigation; T. Wang: clinical management and data collection of the kindred; Q.L. Ruan, L.Y. Shao: conceptualization, methodology, resource, writing-review & editing; W.H. Zhang participated in project administration.

Funding

This work was supported by the National Natural Science Foundation of China (grant #82271794 to Q.L. R) and Clinical Research Project of Huashan Hospital, Fudan University.

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Ethics Approval

This study was performed in line with the principles of the Declaration of Helsinki. Written informed consent was obtained from the patient and her parents.

Consent To Participate

Informed consent was obtained from all individual participants included in the study.

Consent To Publish

The authors affirm that human research participants provided informed consent for publication of the images in Figures.

Competing Interests

The authors declare no competing interests.