Novel mutations of TYK2 leading to divergent clinical phenotypes

TYK2 deficiency is a rare primary immunodeficiency disease caused by loss‐of‐function mutations of TYK2 gene, which is initially proposed as a subset of hyper‐IgE syndrome (HIES). However, accumulating evidence suggests TYK2‐deficient patients do not necessarily present with HIES characteristics, indicating a vacuum of knowledge on the exact roles of TYK2 in human immune system.


| INTRODUC TI ON
The non-receptor tyrosine kinase 2 (TYK2) is a member of the Janus kinase (JAK) family consisting of three additional members (JAK1-3).
Cytokine binding to respective receptor complexes (type I or type II) activates JAKs, which subsequently phosphorylate intracellular receptor chain residues and activate a family of transcription factors termed signal transducers and activators of transcription (STATs, comprised of STAT1-4, STAT5A, STAT5B, and STAT6). 1 Within the JAK-STAT family of proteins, several primary immunodeficiencies have been described, ranging from autosomal reces- To date, there are 13 cases in total in English literature. 5,6 The first TYK2 deficiency patient was reported in 2006, a 22 years old male Japanese, who displayed BCG lymphadenitis, S. aureus infections, and recurrent viral infection, which were attributed to almost abolished responses of patient's cells to IL-23, IL-12, and type I interferon (type I IFN) treatment, respectively. 5 Most interestingly, he also presented with the triad of signs characteristic of HIES: atopic dermatitis, high circulating IgE levels, and recurrent cutaneous staphylococcal infections, which led to the proposal that TYK2 deficiency should be a subset of AR-HIES. However, this notion was challenged lately by a comprehensive study of seven TYK2-deficient patients showing normal IgE levels and absent from atopic dermatitis or cutaneous staphylococcal infection. 6 Later on, both HIES-like and non-HIES TYK2-deficient cases were reported 7-10 , and more confusingly, it has been shown identical TYK2 mutation can result in both HIES-like and non-HIES phenotypes. 7,9 Thus, more cases and further investigations are required to fully understand the nature of TYK2 deficiency.
Here, we present five cases of TYK2 deficiency with novel mutations from five unrelated Chinese families. Briefly, a 2+-yr-old boy (P1) suffered from repeated pneumonia, stomatitis, perilabial herpes, and thrush since the age of 3 months; a 3-yr-old boy (P2) suffered from recurrent respiratory tract infections and diarrhea since the age of 6 months; a 10-yr-old girl (P3) suffered from recurrent respiratory infection since 2 years old and she had a history of refractory eczema; a 5-yr-old boy (P4) suffered from recurrent suppurative otitis media and severe pneumonia and he also had a history of eczema and was highly allergic; and a 5-yr-old boy (P5) suffered from recurrent pneumonia and refractory eczema and he also displayed hypothyroidism. By investigating these cases and comparing with previously reported cases, we aim to uncover a more complete picture of TYK2 deficiency.

| Patients
The Ethics Committee of Children's Hospital of Chongqing Medical University approved the study. Written informed consent for participation in the study were obtained from patients' parents. Blood samples from patients and unaffected relatives were collected for molecular studies, which were performed in accordance with the Declaration of Helsinki.

| Genetic Analyses
Genomic DNA were isolated from peripheral blood samples. Diseasecausing mutations were screened using whole-genome sequencing (WES) or targeted next-generation sequencing (NGS) (MyGenostics, Inc. and Chigene, Inc.). Candidate mutations were confirmed by Sanger sequencing (primer sequences are listed Table S1).

| Western blotting
PBMCs were stimulated with various cytokines for indicated time.
The extracted protein was subjected to 10% SDS-PAGE and transferred onto PVDF membranes (Millipore). The following primary antibodies were used: rabbit anti-phosphorylated STAT1 (Cell Signaling Technology) and rabbit anti-STAT1 (Cell Signaling Technology), rabbit

Key Message
TYK2 deficiency is a rare primary immunodeficiency disease caused by loss-of-function mutations of TYK2 gene.
Due to rarely reported cases, the nature of TYK2 deficiency and the function of TYK2 in human immune system are poorly understood. This study describes five TYK2deficient cases presenting with or without hyper-IgE levels, atopy, and distinct pathogen infection profiles, which are caused by novel TYK2 mutations. Peripheral blood mononuclear cells (PBMCs) from these patients showed heterogeneous responses to various cytokine treatments, including IFNα/β/γ, IL-6, IL-10, IL-12, and IL-23. The homeostasis of lymphocytes is also disrupted. Based on our findings, we propose that TYK2 works as a multi-tasker in orchestrating various cytokine signaling pathways, differentially combined defects which account for the expressed clinical manifestations.

| Flow Cytometry Assay
The antibodies used in flow cytometry assay are shown in Table S4.
The intracellular production of IFNγ was investigated in PBMCs by flow cytometry. PBMCs (2×10 6 cells/ml) were either unstimulated or stimulated with PMA (50 ng/ml) and ionomycin (500ng/ml) for 5 hrs or BCG (MOI =20) or 100 ng/ml BCG +IL-12 for 72 hrs, in 24-well plates. All the samples were treated with 1 μg/ml GolgiPlug (BD) for the last 2 or 6 hrs of culture.
The staining was performed according to the manufacturers' guides. The samples were acquired on a FACSCanto II flow cytometer, and the data were analyzed using FlowJo.

| Real-Time Quantitative PCR
Total RNA was extracted from peripheral blood with the Blood Total RNA Miniprep Kit (Axygen). The cDNA was transcribed by the Transcriptor cDNA Synth. Kit (Roche). Levels of targeted mRNA were determined using a Bio-Rad CFX96 Touch (Bio-Rad) machine.
The results of each gene were determined with the 2 -ΔΔCt method, and data were expressed as fold induction. Primers are listed in Table S2.

| Cycloheximide Chase Assay
This assay is adapted from previous report. 11 Briefly, HEK293T cells were plated and transfected with various WT and mutant expression plasmids. 24 hours later, cycloheximide was added (100ug/ml) and cells were collected at different time points for Western blot detection.

| Statistical analysis
Samples were compared using two-tailed, unpaired Student's t test with GraphPad Prism 7.00 if applicable, unless otherwise stated.

| Clinical courses strongly favor a diagnosis of PID
The patient also displayed secondary hypothyroidism: T3 1.55 nmol/L, FT 33.71 pmol/L, FT4 11.53 pmol/L, TPOAb 177.1 U/mL. All patients were born to non-consanguineous and healthy parents. They were all clinically diagnosed with PID. Detailed description of patients can be found in supplementary information.  Figure 1D).

| Novel mutations identified by WES or targeted NGS lead to TYK2 deficiency
P5 carried a novel homozygous mutation, c.1507C>T, p.R503X (pathogenic) in TYK2 ( Figure 1D). These mutations were all further confirmed by Sanger sequencing ( Figure 1D). According to in-house prediction algorithm developed by the sequencing companies, these mutations were all predicted to be pathogenic to the expression or function of TYK2 protein ( Figure 1B and 1C). To confirm this, we then determined the protein level of TYK2 in patients' PBMCs and we found almost abolished expression of TYK2 ( Figure 1E-1H) in all patients except for P4, whose blood sample was not available.
For P2 and P3, additional sample was available and we then probed TYK2 in their samples with another anti-TYK2 antibody recognizing the C terminus of human TYK2 protein ( Figure 1I). We also determined the mRNA level of TYK2 in P1, P2, and P3, and we found that their mRNA levels were decreased ( Figure 1J).
To further understand and confirm the pathogenicity of these mutations, we performed cycloheximide chase assay to test the protein stability. We found that these mutations indeed impaired the protein stability ( Figure S2). Please note that mutation in P2 has been tested to impair the protein stability previously. 10 Taken together, these five patients from five unrelated families all had TYK2 deficiency.

| TYK2-deficient PBMC showed divergent responses to IFNα/β/γ
Type I IFNs signal through IFNAR1 and IFNAR2, which associate with TYK2 and JAK1, respectively. STAT1 and STAT2 are the major signal transducing STATs in response to type I IFNs. [12][13][14] Nevertheless, also all other STATs can be activated and contribute to responses in a cell type-specific manner. IFNγ signals through IFNGR1 and IFNGR2, which are associated with JAK1 and JAK2, respectively 15, and then, STAT1 is activated. Often STAT3 16,17 and sometimes STAT5 18 are also activated by IFNγ. To further investigate the functional effects of TYK2 deficiency, we tested the responses of PBMC from TYK2deficient patients to type I and type II interferons. For P1, the phosphorylation of STAT1 and STAT3 induced by IFNα and IFNβ was both abolished, while the phosphorylation of STAT1 induced by IFNγ was comparable with that of healthy control (Figure 2A). Reduced STAT1 expression is believed to be partially responsible for reduced phosphorylation of STAT1 in response to type I interferons in TYK2-deficient cells. Consistently, RT-qPCR showed decreased transcription of various ISGs ( Figure 2B). P2 also showed impaired responses to IFNα/β but normal response to IFNγ ( Figure 2C), which was also consistent with ISG expression ( Figure 2D). Taken together, our data indicate that TYK2 deficiency results in divergent responses to IFNα/β/γ.

| TYK2-deficient PBMC showed divergent responses to IL-10 and IL-6
IL-10 signals via IL-10R1 and IL-10R2, which associate with JAK1 and TYK2, respectively. 19 IL-6 can also induce phosphorylation of TYK2. Similarly, we assessed the responses of PBMC from TYK2deficient patients to IL-10 and IL-6. PBMC from P1 showed intact response to IL-6 but abolished response to IL-10 as evidenced by the phosphorylation of STAT3 ( Figure 3A). PBMC from P2 showed significantly impaired responses to both IL-6 and IL-10 as evidenced by the phosphorylation of STAT3 ( Figure 3B), in contrary to P3, whose responses to IL-10 and IL-6 seemed normal as evidenced by the phosphorylation of STAT3 ( Figure 3C) and the transcription of SOCS3 ( Figure 3D), a STAT3 target gene. These data again showed

| TYK2-deficient PBMC showed similar responses to IL-12 and IL-23
TYK2 associates with IL-12Rβ1, a receptor chain that is shared by IL-12 and IL-23. The second receptor chain for both cytokines (IL-12Rβ2 and IL-23R, respectively) associates with JAK2. Biological responses are mainly mediated by STAT4 in response to IL-12 and STAT3 in response to IL-23. 19 We then assessed the responses of PBMC from TYK2-deficient patients to IL-12 and IL-23. In contrast to the responses to IL-6 and IL-10, PBMC from P2 and P3 showed similar impairment in response to IL-12 and IL-23 treatment ( Figure 4A-4D). Consistently, IFNγ production by T cells from P3 was also less compared with healthy control when treated with BCG alone or BCG plus IL-12 ( Figure 4E). Collectively, these data shown so far indicate TYK2 deficiency can lead to divergent clinical phenotypes possibly due to divergent signaling defects.

| TYK2 deficiency disturbs homeostasis of lymphocytes
Cytokine signaling plays essential roles in controlling homeostasis of immune system. 20 The impairment of various cytokines signaling in TYK2-deficient patients could lead to disruption of this balance. P1 had decreased CD4 + T cells, especially the CD4 naïve cells, while his CD4 CM (central memory) and CD4 EM (effect memory) cells were increased. Besides, total B cells of P1 were degressive with elevated naïve B cells and transitional B cells (Table S3). In P2, except for the decline of NK cells, the subsets of T cells and B cells were normal (Table S3). Although the subsets of T cells in P3 were normal, B cells were significantly different from those in normal subjects. The total B cells, memory B cells, and plasmablasts B cells F I G U R E 3 Analysis of cellular response to IL-6 and IL-10 in TYK2deficient patients. A, B, and C, Total STAT3 and pSTAT3 protein level were analyzed by Western blot in P1, P2, P3, and healthy control. The patients' PBMC was stimulated with IL-6 (100ng/ μL) or IL-10 (100ng/μL) for 15 min. Representative images from 2 to 3 independent experiments. D, SOCS3 induction was analyzed by RT-qPCR after 6 h of treatment with 100 ng/ml IL-10, in PBMC from healthy controls and P3. Data are derived from 2 to 3 independent experiments F I G U R E 4 Analysis of cellular response to IL-12 and IL-23 of TYK2-deficient patients. A and B, Total STAT4 and tyrosine-phosphorylated STAT4 (pSTAT4) protein level were analyzed by Western blot in P2, P3, and healthy control. The patients' PBMC were stimulated with IL-12 (100ng/μL) for 15 min. C and D, Total STAT3 and pSTAT3 protein level were analyzed by Western blot in P2, P3, and healthy control. The patients' PBMC were stimulated with IL-23 (100ng/μL) for 15 min. E, Flow cytometry analysis showing intracellular IFNγ production in PBMCs after stimulation in the absence or presence of PMA (50ng/ml) and ionomycin (500ng/ml) for 5 hrs or BCG or BCG and 100 ng/ml IL-12 for 72 hrs. An anti-CD3 Ab was used to identify CD3 + T cells in P3 and healthy control. Representative images from 2 to 3 independent experiments were declined but the naïve B cells and the transitional B cells were higher than normal range (Table S3). Similar to P1, P5 had abnormal T cells; the total T cells, CD4 + T cells, CD8 + T cells were significantly decreased; B cells and NK cells were normal (Table S3).
Further analysis of CD4 + T cell subsets showed that frequencies of Th1 and Th1-like cells were increased in P1 and P2 ( Figure 5A-5F), while frequency of Th2 cells was decreased in P1, P2, and P3 compared with healthy controls (Figure 5A-5F). Th17 cells seemed unaffected in all three patients tested. The frequencies of Tfh cells were elevated in P1 and P3, but normal in P2 ( Figure 5G). We further analyzed the Tfh subsets showing normal Tfr frequency and increased CXCR5 + PD-1 + cells ( Figure 5H-5L and 5N). The proportion of Treg were comparable to that of healthy controls in P2 and P3, but increased in P3 (Figure 5H-5L). processes. Recent studies showed the P1104A TYK2 common variant predisposes host to mycobacteria infection but protects host from autoimmunity both through dampening IL-23 signaling, which also contributed to our understanding of TYK2 function. [21][22][23] In this study, we presented five more cases of TYK2 deficiency and investigated many aspects of the effects TYK2 deficiency brings, which we think would further improve our understanding of TYK2 deficiency.
The first TYK2-deficient patient to be described was Japanese and had HIES as the main clinical feature, associated with numerous intracellular infections. Since then, TYK2 deficiency was considered a subset of AR-HIES. However, newly identified TYK2-deficient patients did not necessarily present with hyper-IgE. A comprehensive comparison of various immunologic defects between the first reported patient and seven patients without hyper-IgE was reported, which proposed that intact IL-6/STAT3 signaling in patients without hyper-IgE might be responsible for this phenotypic difference. 6 As defects in STAT3-mediated signaling play a key role in the development of HIES 24 , IL-6 signaling is TYK2 independent in mice. 25 Another TYK2-deficient patient without HIES was also reported to show normal IL-6 signaling 8 and a TYK2-deficient patient with HIES showing impaired IL-6 signaling was reported 7 (Table 1). Recently, autosomal recessive and dominant mutations in IL6ST, which encodes gp130, a subunit of cytokine receptor for IL-6, IL-11, IL-27, and some others, have been reported in patients with HIES. 26,27 Consistently, our patients, whose PBMC showed normal response to IL-6 treatment in terms of STAT3 activation, did not present with typical hyper-IgE, with P1 showing only mildly elevated IgE and P3 showing low IgE ( Figure 3). However, PBMC from P2 with normal IgE level showed impaired response to IL-6 treatment ( Figure 3). Taken together, these findings could indicate that impaired IL-6 signaling could be responsible for hyper-IgE phenotype as previously reported, but does not necessarily always lead to hyper-IgE.
The viral infections can probably be accounted for by defects of IFNα/β/γ signaling. Similar to the first described TYK2-deficient patient, P1 in our cohort suffered repeated viral infection possibly due to abolished responses to IFNα/β. Although impaired but not abolished IFN signaling does not necessarily lead to severe viral infection as the case of P2 (Table 1)  In summary, we presented five more TYK2 deficiency cases caused by different novel mutations displaying divergent cellular defects and variable clinical phenotypes, which demonstrated that the correlation of cellular defects and clinical phenotypes is far more complicated than previously thought. In view of this, here we propose that TYK2 works as a multi-tasker in orchestrating various cytokines signaling pathways, differentially combined defects of which account for the expressed clinical manifestations. Furthermore, we speculated that the discrepancies in cellular defects could at least partially be explained by different mutational effects to TYK2 protein, which needs to be investigated in future. Nevertheless, our findings might lead us further toward more accurate understanding of TYK2 function in human immune system and a more timely diagnosis of TYK2-deficient patients.

ACK N OWLED G M ENT
We are grateful to all the patients and their families for their continuous cooperation in this study. We thank the members of the laboratory for their technical assistance. We thank doctors and nurses for their generous support to this project.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

AUTH O R CO NTR I B UTI O N
Ge Lv contributed to conceptualization (equal); data curation (lead); formal analysis (lead); investigation (lead); and methodology (lead).
Gan Sun contributed to formal analysis (supporting). Peilin Wu contributed to data curation (supporting). Xiao Du contributed to data curation (equal). Ting Zeng contributed to methodology (supporting). Wen Wen contributed to methodology (supporting). Lina Zhou supervised the study (equal). Tingyan He made formal analysis (supporting). Xiao-dong Zhao acquired funding (lead); made project administration (lead); and involved in writing-review editing (equal).

E TH I C A L A PPROVA L
The Ethics Committee of Children's Hospital of Chongqing Medical University approved the study.