2.1 DOCK8 mutation increases serum IgE levels and decreases DOCK8 protein in T cells

To substitute the alanine residue (encoded by GCC) with an aspartic acid residue (encoded by GAC) at position 1949 of the DOCK8 gene (Transcript: ENSMUST00000025831.8 DOCK8-201), we developed a repair oligonucleotide in conjunction with two single guide RNAs (sgRNAs) specifically targeting exon 45 of the DOCK8 gene. A silent mutation was incorporated at amino acid position 1947 to prevent re-cleavage by Cas9 following homologous recombination. To confirm that the sequence of DOCK8 was changed in exon 45, PCR and sanger sequencing were employed. Sequence alignments were conducted using the basic local alignment search tool (BLAST) search on the NCBI website (Figure 1A,B). To verify that our constructed DOCK8 mutation mouse model could mimic AR-HIES, we examined serum IgE levels by enzyme linked immunosorbent assay (ELISA), which revealed that DOCK8 mutant mice had increased levels of serum IgE (Figure 1C), suggesting that the patient's high IgE clinical phenotype was reproduced in this mouse model. AR loss-of-function mutations in the DOCK8 cause DIDS and the mutations commonly lead to a lack of DOCK8 protein expression.¹³ We found that the DOCK8 mutant mice had a significant decrease in DOCK8 protein expression in CD4⁺ and CD8⁺ T cells (Figure 1D and S1A,B). Therefore, the DOCK8 mutation mouse model was successfully constructed. Finally, although no significant lymphocytic infiltration was observed in DOCK8 mutant mice, we did find that they had larger pLn (Figure 1E,F), which is consistent with observations in DOCK8 Treg-specific deletion mice.²⁶ In summary, the DOCK8 mutation increases serum IgE levels and decreases DOCK8 protein in T cells.

2.2 DOCK8 mutation impairs peripheral T cell homeostasis without affecting thymus T cell development

Reports indicate that DOCK8-deficient patients have reduced numbers of peripheral blood CD4⁺ and CD8⁺ T cells.¹⁴ To further validate the T cell immunophenotype of DOCK8 mutant mice, we analyzed T cell expression in the thymus, spleen, pLn, and mesenteric lymph nodes (mLn) of DOCK8 mutant mice. The results indicated that DOCK8 mutant mice had significantly lower proportions of CD4⁺ and CD8⁺ T cells in the spleen, pLn, and mLn, along with a depletion of splenic CD4⁺ and CD8⁺ T cells, but the pLn and mLn CD4⁺ and CD8⁺ T cell numbers showed no significant difference (Figure 2A–E and Figure S2A). In contrast, no difference was found in the proportion and number of thymus in DOCK8 mutant mice (Figure 2A and Figure S2B,C), suggesting that DOCK8 mutation did not affect the development of thymic T cell.

Subsequently, we examined T cell activation in the DOCK8 mutant mice. The findings revealed that, in DOCK8 mutant mice, the proportion of naive CD44^lowCD62L^high CD8⁺ T cells in the spleen and pLn was distinctly lower, whereas notably higher in the thymus (Figure 2F–H and Figure S2D–I). However, activated CD44^highCD62L^low CD8⁺ T cells in the pLn were significantly increased, whereas no significant difference was seen in the spleen, thymus, and mLn (Figure 2I and S2D–I). Moreover, no significant differences were detected in the proportion of naive and activated CD4⁺ T cells in the spleen and pLn (Figure 2J,K). Additionally, we noted an increase in late apoptotic CD4⁺ T cells in the spleen of DOCK8 mutant mice (Figure 2L and Figure S2J). Finally, we examined the proliferation of CD4⁺ T cells. The results demonstrated reduced proliferation of CD4⁺ T cells in the spleen of DOCK8 mutant mice (Figure 2M and Figure S2K). These results demonstrate that more peripheral T lymphocytes in DOCK8 mutant mice are in an activated state and affect peripheral T cell homeostasis.

To test whether the reduced ratio of CD4⁺ and CD8⁺ T cells in DOCK8 mutant mice was not influenced by the microenvironment of the mice and was intrinsic to the T cells, we constructed a mouse bone marrow chimeric model. To do this, the donor (CD45.2) bone marrow cells were mixed with the bone marrow cells of the CD45.1 recipient mice in a ratio of 1:1. The mixed cells were injected via the tail vein into the irradiated CD45.1 recipient mice (Figure S3A). Furthermore, the proportion of CD4⁺ and CD8⁺ T cells in the spleen and pLn of CD45.2⁺ DOCK8 mutation donor mice was markedly reduced compared to CD45.2⁺ wild type (WT) donor mice (Figure S3B–H), which indicates that the DOCK8 mutation affects the proportion of peripheral T cells intrinsically within T cells, rather than due to abnormal thymic T cell development.

2.3 DOCK8 mutation disturbs the development and ICOS expression of Tregs in the pLn

Tregs are an essential subpopulation of T cells that mediate immune tolerance and regulate immune system homeostasis by maintaining peripheral T cell homeostasis, which inhibits the activation of auto-reactive T cells. Studies have reported that Tregs are reduced and exhibited impaired suppressive activity in DOCK8-deficient patients.²⁷ Therefore, we investigated the effects of DOCK8 mutation on Tregs subpopulations in the DOCK8 mutant mice. The findings revealed that DOCK8 mutant mice showed a significantly lower proportion of CD4⁺CD25⁺ and CD4⁺Foxp3⁺ Tregs in the pLn and a reduced proportion of CD4⁺Foxp3⁺ Tregs in the spleen. However, the number of CD4⁺CD25⁺ and CD4⁺Foxp3⁺ Tregs was not significantly changed, which might be due to the enlargement of the pLn (Figure 3A–F). These studies indicate a possible link between elevated activated T cells and a reduction in Tregs in DOCK8 mutant.

To further investigate the mechanism of Tregs reduction, we first examined Foxp3, a key transcription factor specifically expressed by Tregs, which determines the development and function of Tregs as well as maintains homeostasis and suppresses the immune response effectively.²⁸ The results showed that in the CD4⁺Foxp3⁺ Tregs of DOCK8 mutant mice, there was no distinct difference in the mean fluorescence intensity (MFI) of Foxp3 (Figure 3G), suggesting that the reduction in Tregs was not due to a functional defect in Foxp3. Next, we detected the inhibitory molecules CTLA-4, ICOS, PD-1, and NRP1 that are expressed on the surface of Tregs and enhance their inhibitory function. The results showed that DOCK8 mutant mice exhibited increased MFI of ICOS in pLn CD4⁺CD25⁺ Tregs, decreased MFI of PD-1 in mLn CD4⁺CD25⁺ Tregs, and no significant difference in CTLA-4 and NRP1 in CD4⁺CD25⁺ Tregs (Figure 3H–K). It has been shown that Tregs high expression of ICOS facilitates the maintenance of immune homeostasis.²⁹ These results suggest that DOCK8 mutations disrupt Tregs development and ICOS expression in the pLn, thereby affecting immune system homeostasis.

2.4 DOCK8 mutation promotes Th17 cell differentiation

To investigate the transcriptome changes in DOCK8 mutation, we isolated CD4⁺ T cells for RNA-bulk sequencing. We found that Th17 cell differentiation was significantly enriched in DOCK8 mutant mice (Figure 4A). Also, gene enrichment analysis (GSEA) showed that genes on Th17 cell differentiation were significantly enhanced in CD4⁺ T cells from DOCK8 mutant mice (Figure 4B), suggesting that DOCK8 mutation might promote Th17 cell differentiation. To validate the hypothesis, we detected the levels of IFN-γ, IL-4, and IL-17, which demonstrated that the proportion of IL-17 by CD4⁺ T cells was increased in the spleen and pLn of DOCK8 mutant mice (Figure 4C,D). The results were consistent with those of RNA-bulk sequencing data. Elevated IFN-γ by CD4⁺ T cells in mLn and by CD8⁺ T cells was also observed in DOCK8 mutant mice (Figure 4E–H), whereas no significant difference was detected in IL-4 and IL-2 by CD4⁺ T cells in the spleen and pLn (Figure 4I,J). Furthermore, we also analyzed the RNA-bulk sequencing data for IL-17 producing related genes such as RORC, SOX4, IL1R1, and IL-17F. We detected a significant elevated mRNA level of RORC, IL1R1, and IL-17F in DOCK8 mutant mice by RT-PCR (Figure 4K). These results suggest that DOCK8 mutation promotes Th1 and Th17 cell differentiation.

2.5 DOCK8 mutation CD4⁺ T cells have a Th2 effector fate after lymphocytic choriomeningitis virus (LCMV) infection

Viral infections are frequently observed in DOCK8-deficient patients.¹⁴ To find out the role of DOCK8 mutation in viral pathogenesis, the level of cytokines in CD4⁺ T was examined in the spleens of WT or DOCK8 mutant mice following LCMV infection. The findings revealed that DOCK8 mutant mice had markedly lower proportions of CD4⁺ T cells after LCMV infection (Figure 5A,B). Unsurprisingly, IL-2, IL-4, and IL-17 secreted by DOCK8 mutation CD4⁺ T cells were increased after LCMV infection but IFN-γ levels remained unchanged (Figure 5C). Meanwhile, ELISA experiments demonstrated that the expression of IgE in serum also increased after LCMV infection (Figure 5D). Together, our findings indicate that DOCK8 mutation CD4⁺ T cells are biased toward Th2 effector fate after LCMV infection.

2.6 DOCK8 mutation upregulates the glycolysis level of CD4⁺ T cells associated with Akt/mTOR/S6/HIF-1α pathway

Th1 and Th17 cell differentiation is regulated by mTOR signaling.³⁰ Transcriptomic data showed that GSEA-enriched genes to the mTOR signaling pathway were enriched in CD4⁺ T cells from DOCK8 mutant mice (Figure 6A). Furthermore, we detected the expression of mTOR pathway related molecules. The findings showed that protein levels of pAkt, pmTOR, and pS6 were also markedly elevated in CD4⁺ T cells of DOCK8 mutant mice after TCR activation (Figure 6B), indicating that the DOCK8 mutation upregulated mTOR signaling. Transcriptomic data suggest that starch and sucrose metabolism may be crucial in DOCK8 mutant mice (Figure 4A). mTOR can upregulate HIF-1α to promote glycolysis and induce Th17 cell differentiation.³¹ Furthermore, the protein expression of HIF-1α and pyruvate kinase M2 (PKM2, an important enzyme in the glycolysis process) was significantly elevated in DOCK8 mutation CD4⁺ T cells (Figure 6B). The above results indicate that DOCK8 mutation affects CD4⁺ T cell metabolism.

To examine the role of DOCK8 mutation in CD4⁺ T cell metabolism, the mitochondria morphology and reactive oxygen species (ROS) were observed by staining the PK Mito and dihydroethidium (DHE), respectively. DOCK8 mutation CD4⁺ T cells showed remarkably increased MFI of PK Mito and DHE in both anti-CD3 plus anti-CD28 stimulated and unstimulated (Figure 6C–F). Additionally, extracellular acidification rate (ECAR) revealed that the basal glycolytic capacity of DOCK8 mutation mice in CD4⁺ T cells was elevated by the addition of glucose, the maximal glycolytic capacity of the cells was elevated by the addition of oligomycin, and the glycolytic reserve and non-glycolytic acidification were elevated by the addition of 2-Deoxy-D-glucose (2-DG) (Figure 6G). This suggests that DOCK8 inhibits the CD4⁺ T cell glycolytic pathway. The results of oxygen consumption rate (OCR) showed that basal respiration of DOCK8 mutation CD4⁺ T cells was significantly elevated, the oxygen consumption rate of oxidative phosphorylation was elevated by the addition of oligomycin, and maximum respiration was not significantly different by the addition of FCCP, but spare respiratory capacity and non-mitochondrial oxygen consumption were elevated after the addition of rotenone and antimycin A, indicating that DOCK8 inhibits the oxidative phosphorylation of CD4⁺ T cells (Figure 6H). In conclusion, these results imply that DOCK8 mutation upregulates the glycolysis level of CD4⁺ T cells associated with the Akt/mTOR/S6/HIF-1α pathway.

4.1 Mice

All mice were fed under specific pathogen-free (SPF) conditions with a 12 h light/dark cycle. All experiments were approved by the Animal Experiment Ethics Committee of Tongji Medical College (approval number, 3954).

4.2 Generation of DOCK8 mutation

To convert the GCC codon in exon 45 of the DOCK8 gene (Transcript: ENSMUST00000025831.8 DOCK8-201) from coding for an alanine residue to coding for aspartic acid at position 1949 of DOCK8, we designed a repair oligonucleotide along with two sgRNAs that target exon 45 of the DOCK8 gene. The repair oligonucleotide sequence, with differences from the wild-type C57BL/6 sequence indicated in lowercase and the protospacer sequences of the sgRNAs underlined, is as follows: TGGCCTCACTCTGCGGTTGTCTTCTGTTCAGTTCGTTTTGACTCCGATTGAaGTTGaCATTGAAGATATGAAGAAGAAGACCCTGCAGTTAGCCGTGGCCACTCACCA. The repair oligonucleotide was synthesized by BiOligo Biotechnology Co., Ltd., and the sgRNAs were provided by GenScript. Cas9 mRNA, sgRNAs, and the repair oligonucleotide were injected into mouse embryos as described previously.⁴³ The genotyping of DOCK8^A1949D knock-in mice was performed by sequencing a 346-bp DNA fragment flanking exon 45 using the following pair of PCR primers: forward 5′-CCATCTCTGTGTGACCTGTTCA-3′ and reverse 5′-AACAAACAAGCCCTAACCAAGC-3′.

4.3 Cells preparation

Mouse spleen, thymus, pLn, and mLn were collected and grounded. Cells were suspended in an HBSS solution containing 2% serum and then filtered with a nylon membrane after lysed red blood cells with Red Cell Lysis Buffer to obtain mononuclear cells. Reagents details are presented in Table S1.

4.4 Purification and activation of CD4+ T cell

Purified CD4⁺ T cells were isolated by murine CD4⁺ T cell Isolation Kit. For the activation of CD4⁺ T cells, sorted cells were cultured in 10 µg/mL plate-bound anti-CD3 plus anti-CD28 for 5 h at 37°C as described previously.⁴⁴ Reagents details are presented in Table S1.

4.5 Flow cytometry

For cell surface staining, mononuclear cells (1 × 10⁶) from the thymus, spleen, pLn, and mLn were stained with antibodies on ice for 30 min after incubation with the Fc blocker. For cell intracellular staining, cells (2 × 10⁶) were fixed, permeabilized, and subsequently cultured with antibodies on ice for 30 min. For cytokine analysis, mononuclear cells (2 × 10⁶) were cultured with PMA (50 ng/mL), GolgiStop (1:1000), and lonomycin (1 µM) at 37°C for 5 h. Then cells were collected and stained with antibodies on ice for 30 min. Finally, cells were measured using Attune NxT (Thermo Fisher, AFC2) and analyzed by FlowJo 10 software (TreeStar). Antibodies’ details are presented in Table S1.

4.6 Confocal

Purified CD4⁺ T cells (1.5 × 10⁶) were cultured with 10 µg/mL plate-bound anti-CD3 plus anti-CD28 for 5 h at 37°C. To prepare ROS and PK Mito analysis, the cells were cultured with anti-CD4 and DHE or PK Mito for 15 min at 37°C. Then, the cells were analyzed by a confocal microscope (Nikon). The MFI was analyzed using NIS-elements AR 5.01. Over 50 individual cells were analyzed for confocal analysis. Reagents details are presented in Table S1.

4.7 Western blotting

Sorted and stimulated CD4⁺ T cells (2 × 10⁶) were lysed using RIPA buffer containing cocktail, NaF (1 M), and Na3VO3 (100 Mm). Then, cell lysates were examined by SDS-PAGE, and immunoblotting was performed using antibodies (Bio-Rad). Antibodies details are presented in Table S1.

4.8 Seahorse analysis

Purified CD4⁺ T cells were prestimulated in 24-well plates with 10 µg/mL anti-CD3 plus anti-CD28 for 24 h at 37°C, and then seeded the cells into culture plates at 37°C for 1 h. For ECAR analysis, cells were stimulated with 2 uM Oligomycin, 10 mM Glucose, and 5 mM 2-DG. For OCR analysis, cells were stimulated with 1 µM FCCP, 1.5 µM oligomycin, 1 µM antimycin A, and 500 nM rotenone as described previously.⁴⁵ Finally, cells were examined by the Seahorse XFe24 Cell Metabolism Analyser (Agilent). Reagents details are presented in Table S1.

4.9 ELISA

Levels of IgE in serum were measured using a mouse IgE ELISA Kit. Reagents details are presented in Table S1.

4.10 LCMV infection

Mice were injected intraperitoneally with PBS containing 5 × 10⁵ PFU of LCMV-Armstrong 53b (200 µL). One week later, the mice were executed, and T cells from the spleen were examined. The level of IgE in serum was tested by ELISA (INFINITE 200 PRO, TECAN).

4.11 BM chimeras

BM cells were collected from WT (CD45.1 recipients), DOCK8 mutation (CD45.2, donor), and WT (CD45.2, donor) mice as described previously.⁴⁵ After 8 weeks postinjection, spleen, thymus, and pLn cells were collected and isolated for staining by flow cytometry.

4.12 Transcriptome sequencing

For RNA-seq libraries, VAHTS Universal V8 RNA-seq Library Prep Kit for Illumina was constructed and then sequenced on the BGISEQ platform. Then, low-quality reads were removed by the BGISEQ platform and aligned reads to the mouse reference genome.

4.13 RT-PCR

In brief, total RNA was isolated from CD4⁺ cells (1 × 10⁶) using the trizol reagent and detected using real-time reverse transcription PCR assay as described previously.⁴⁶ The primer sequences refer to Table S2.

4.14 Statistics analysis

Data were presented as mean ± SEM and statistical analyses were presented using Prism 8.0 (GraphPad). When comparing two groups, an unpaired two-tailed Student's t-test was used. Asterisks indicate significant difference: *p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.