Two recently published datasets from the NCBI GEO (Edgar et al., 2002) contain gene expression profiles obtained from the labial glands and CD4⁺ T cells of SS patients and healthy donors. Specifically, we analyzed 18 patients with SS and 18 healthy controls in the salivary gland dataset (accession number: GSE94510) (Tasaki et al., 2017) and 17 SS patients and 15 healthy controls in the CD4⁺-T cell dataset (accession number: GSE143153) (Joachims et al., 2020). These datasets provide an invaluable resource to further explore the molecular mechanisms that underpin SS. The expression data were processed using the IOBR package in R for conversion into TPM format (Zeng et al., 2021). Additionally, Gene ontology-biological process (GO-BP) gene sets were obtained using the msigdbr package (Liberzon et al., 2015), (Subramanian et al., 2005) as described (Yin et al., 2020), (Yin et al., 2022a), (Yin et al., 2022b). The corresponding web links are provided in Supplementary File S1.

This study was approved by the Ethics Committee of the Shanghai Ninth People’s Hospital affiliated to the Shanghai Jiao Tong University School of Medicine (Approval IDs: SH9H-2019-T159-2 and SH9H-2021-TK69-1). All human participants provided written informed consent. The selected SS patients met the criteria of the American-European Consensus Group for SS (Bruserud, 2001) and had not received any immunosuppressive or immunomodulatory drugs before sample collection. All donors were middle-aged females aged 30–60 years old. Peripheral blood was collected from each donor (10 mL) and peripheral blood mononuclear cells (PBMCs) were isolated as described (Fu et al., 2020). Then, human CD4⁺ T cells were isolated from the PBMCs by the Human CD4⁺ T Cell Enrichment Kit (19,052, Stemcell Technologies) according to the manufacturer’s instructions. After suspension of PBMCs obtained with a 37 μm cell strainer (Falcon), a cell suspension of approximately 5 × 10^7 cells/mL was prepared in PBS containing 2% (v/v) fetal bovine serum and 1 mM EDTA. Subsequently, 50 μL/mL of Enrichment Cocktail was added and the cells were cultured for 10 min. A magnetic bead solution (100 μL/mL) was then added to the mixture for 5 min, after which the round-bottom test tube was placed in a magnet. The cell suspension was emptied from the tube after 5 min and used for subsequent experiments.

Female C57BL/6 mice and NOD/ShiLtj mice were purchased from the Model Animal Research Center of Nanjing University (China), and were cared for in accordance with the criteria outlined in the Guide for the Care and Use of Medical Laboratory Animals (National Research Council, 2011). BI8626 (HY-120204, MCE, 44 mg/kg) and vehicle [40% (v/v) PEG300, 5% (v/v) Tween-80, 45% (v/v) saline] were administered to the experimental group and control group, respectively. At 8 weeks of age, mice were intraperitoneally injected with 3.6 mg/mL BI8626 twice or four times a week for 4 weeks. In the control group, mice were administered an intraperitoneal injection of vehicle (200–250 μL) twice weekly for 4 weeks. Once weekly, salivary flow rate (SFR) was measured in mice by an intraperitoneal injection of pilocarpine (5 mg/kg, HY-B0726, MCE) to induce saliva production. A pipette tip was placed in the mouth of the animals, and saliva was collected for 15 min to calculate the SFR. Following cervical dislocation, the bilateral salivary gland tissues of the mice (12 weeks) were obtained. Hematoxylin & eosin (H&E) staining was performed on one side of each mouse’s salivary gland tissue, and the severity of lymphocyte infiltration in the gland tissues was evaluated using a classical scoring system (Chisholm et al., 1970), (Gu et al., 2018). An aggregate of more than fifty lymphocytes in a random field (4 mm²) of the salivary gland tissue was defined as a lymphocytic focus, according to the score system detailed in Table 1. Five mice per group were used as biological replicates in a study designed to assess the presence of lymphocytic foci in salivary gland tissue. Tissue samples were minced and then digested with 1 mg/mL collagenase type 4 and DNase I for 45 min. The resulting suspension was then passed through a 70 μm mesh nylon strainer (Falcon) and rinsed with PBS containing 2% (v/v) fetal bovine serum (FBS, Gibco).

Following centrifugation, cells were resuspended in 3 mL PBS supplemented with 37% (v/v) Percoll, and carefully layered on top of 3 mL of PBS containing 70% (v/v) Percoll. An additional 3 mL of PBS containing 30% (v/v) Percoll was overlaid, and the sample was centrifuged at 800 g for 20 min. The cell layer was collected, resuspended in PBS, and further purified. Cells were labeled with PE-Cy7 anti-mouse CD45 (#561868, BD Biosciences), FITC anti-mouse CD3 (#100204, BD Biosciences), PE anti-mouse CD4 (#100512, BD Biosciences), and APC anti-mouse CD8 (#100711, BD Biosciences) for 30 min. Subsequently, the fluorescence intensity was measured using a Beckman CytoFlex S.

Spleens were carefully ground in a 70 μm mesh nylon strainer (Falcon) and rinsed with PBS containing 2% (v/v) FBS to obtain a single-cell suspension. We incubated the centrifuged cell masses with 5 mL erythrocyte lysate on ice for 2 min, followed by a second centrifugation. Subsequently, total CD4⁺ T cells were isolated from the cell suspension using a Mouse CD4⁺ T Cell Positive Selection Kit (18,952, Stemcell Technologies). We suspended cell masses in 1 × 10^8 cells/mL PBS containing 2% (v/v) FBS and 1 mM EDTA and then added 50 μL/mL rat serum and 50 μL/mL Selection Cocktail for 5 min. Subsequently, 30 μL/mL magnetic bead solution was added for 3 min, followed by 1.5 mL PBS containing 2% FBS and 1 mM EDTA. The round-bottom test tube containing the mixture was then placed in a magnet for 3 min, then the cells were washed twice with the previously described PBS solution. The purified cells and magnetic beads were left in the test tube for further use. The purified cells were then cultured in Rosewell Park Memorial Institute (RPMI) 1640 medium (Gibco) supplemented with 10% (v/v) FBS and 1% (v/v) penicillin/streptomycin (HyClone), and stimulated with a combination of 5 μg/mL plate-bound anti-CD3ε and anti-CD28 antibodies (BD Biosciences). Chemical inhibitors included BI8626 (HY-120204, MCE), N-Butyldeoxynojirimycin (NB-DNJ, UGCG inhibitor, HY-17020, MCE), and DIDS (HY-D0086, MCE).

Using the highly stable cell tracer 5-(and-6)-Carboxyfluorescein Diacetate Succinimidyl Ester (CFSE)-based cell proliferation kit (C34554, Thermo Fisher Scientific), we assessed the degree of T cell proliferation. We incubated CD4⁺ T cells in suspension with 2 μM CFSE solution in the dark for 10 min. To stop the staining, 10 mL of medium containing 10% (v/v) FBS was added to the cell suspension and incubated on ice for a further 10 min in the dark. After staining, the cells were used in subsequent experiments. We then monitored the fluorescence intensity using a Beckman CytoFlex to detect changes in cell proliferation.

Analysis of the cell cycle phase distribution was conducted using propidium iodide (PI) staining of DNA content. Cells were first washed with pre-cooled PBS and centrifuged, then fixed with 70% ethanol at −20°C for 2 h. After removing the ethanol, cells were washed with PBS and mixed with a working solution containing 25 μL PI solution and 2.5 μL RNase. Staining was done at 37°C for 30 min, followed by incubation at 4°C for an additional 30 min. Cell clumps were removed from the sample by filtration through a 70 μm nylon screen and DNA quantification was conducted using a Beckman CytoFlex.

T cells were seeded into a confocal-private plate using Cell-Tak cell and tissue adhesive (Corning), and then fixed with 4% (w/v) paraformaldehyde. Intracellular cholesterol was fluorescently stained using a cholesterol cell-based detection assay kit (Cayman) for 30 min at 4°C in the dark. Images of stained cells were acquired on an ECLIPSE Ts2R laser-scanning confocal microscope (Nikon). Total cholesterol was extracted and determined using a total cholesterol detection kit (Applygen Technologies), while cellular protein content was measured with an Enhanced BCA Protein Assay Kit (Beyotime).

Cells were fixed with 4% (w/v) polyformaldehyde for 30 min at room temperature, and then incubated with a biotin-labeled cholera toxin B subunit conjugated with FITC (C1655, Sigma-Aldrich). Subsequent analysis by the Beckman CytoFlex S enabled the characterization of lipid rafts.

Splenocytes were first stimulated with 1 μg/mL α-CD3 antibody for up to 90 min at 37°C. Cells were then fixed with 4% paraformaldehyde and permeabilizated in ice-cold 100% methanol. Subsequently, antibodies against phosphorylated Zap70 (683705, Biolegend) were used for immunostaining. This approach allowed us to elucidate the dynamics of TCR signaling.

Cholesterol efflux detection was based on Zhao et al. (Zhao et al., 2022). Purified murine CD4⁺ T cells from the relevant groups were labeled with a medium containing 2.5 mM BODIPY-cholesterol (HY-125746, MCE), obtained by complexing the sterols (unlabeled cholesterol and BODIPY-cholesterol, 8:2) with methyl-β-cyclodextrin (HY-101461, MCE) at a molar ratio of 1:15. Labeling was carried out by incubating the cells in labeling medium supplemented with 0.2% fatty acid-free BSA and 2 μg/mL Sandoz58-035 (ACAT inhibitor), at 37°C and 5% CO₂ for 60 min. After washing twice with normal culture medium and centrifuging at 350 g for 5 min, the cells were resuspended in culture medium for a 1 h equilibration. ApoAI (10 μg/mL, A0722, Sigma-Aldrich) or HDL (50 μg/mL, SAE0054, Sigma-Aldrich) was then added to the cell suspension to initiate cholesterol efflux. After a four-hour incubation period, efflux was stopped by centrifuging at 350 g for 5 min to remove the supernatant-containing acceptors. Samples treated without ApoAI or HDL after equilibration were used as the baseline for efflux. The mean fluorescence intensity (MFI) of BODIPY-cholesterol on T cells was subsequently measured by flow cytometry. The level of cholesterol efflux was then calculated as follows: (baseline MFI-remaining MFI)/baseline MFI×100.

Protein extraction was performed using a Pierce immunoprecipitation (IP) lysis buffer (Thermo Fisher) supplemented with a Protease Inhibitor Cocktail (Thermo Fisher). Samples were separated by SDS-PAGE gel electrophoresis and transferred to a PVDF membrane (Millipore). The membrane was blocked with Quick Block Buffer (Beyotime Biotechnology) for 15 min and then treated overnight with primary antibodies against Cyclin-dependent kinase 2 (CDK2, 1:1000, CST), Cyclin-dependent kinase 4 (CDK4, 1:1000, CST), Hydroxymethylglutaryl-CoA reductase (HMGCR, 1:1000, CST), Squalene epoxidase (SQLE, 1:1000, CST), HMGCS1 (1:1000, CST), ATP-binding cassette transporter A1 (ABCA1, 1:1000, Abcam), ATP-binding cassette transporter G1 (ABCG1, 1:1000, Abcam), Scavenging Receptor SR-BI (SR-BI, 1:1000, Abcam), HUWE1 (1:1000, Abcam) and ACTIN (1:1000, CST) at 4°C. After incubation with the primary antibody, the membrane was allowed to return to room temperature over 1 h. Subsequently, the membrane was washed three times with 1×TBST buffer for 10 min each time. Horseradish peroxidase (HRP)-labeled goat anti-rabbit IgG (H + L) (Beyotime) or horseradish peroxidase (HRP)-labeled goat anti-mouse IgG (H + L) (Beyotime) was then diluted with 1×TBST buffer and incubated with the membrane at room temperature for 2 h, followed by three washes with 1×TBST. Finally, Immobilon Western Chemiluminescent HRP Substrate (Millipore) was used. The membrane was then blotted on filter paper and incubatd in full contact with luminescent working liquid at room temperature for 1 min. The results of the WB analysis were visualized and quantified using an Amersham Imager 600.

Total RNA was extracted from T cells with TRIzol Reagent (TaKaRa) according to the manufacturer’s protocol (Yin et al., 2022a). Chloroform (0.2 mL) was added to 1 mL of cell lysate and mixed by shaking. The mixture was then placed on ice for 15 min, centrifuged, and the uppermost clear phase was collected. An equal volume of isopropyl alcohol was then added to the mixture and the mixture was allowed to stand for 10 min before a second centrifugation. The liquid in the tube was discarded, and 0.6 mL of pre-cooled 75% (v/v) ethanol was added to wash the RNA precipitates on the tube wall. This process was repeated twice, after which the ethanol was discarded. When the edges of the RNA precipitate became clear, it was dissolved in 20 μL DEPC water and the concentration and purity of the dissolved precipitate were measured using a Biochrom NanoVue Plus. All steps were carried out on ice. Subsequently, 1000 ng of total RNA was reverse-transcripted to cDNA using Takara PrimeScript RT reagent kits (TaKaRa), which was then used for RT-qPCR on a LightCycler96 Instrument (Roche). The primer sequences are listed in Table 2, and ACTB/Actb was used as the internal mRNA control. Experiments were repeated in triplicate, and the relative RNA expression rates were calculated using the 2^-△△Ct method.

Huwe1-shRNA lentivirus (shRNA-Huwe1-01, shRNA- Huwe1-02 and shRNA- Huwe1-03) and normal control lentivirus were designed and synthesized by Hanbio (Shanghai). These shRNAs were transfected into murine CD4⁺ T cells according to the manufacturer’s recommended protocol (Li et al., 2018).

Statistical analyses were performed using GraphPad Prism 8.0 software. Data are presented as means±standard deviations (SDs). Student’s t tests were used to compare groups, with p < 0.05 indicating a statistically significant difference. A one-way analysis of variance followed by Tukey’s multiple comparison post-hoc tests was employed for multiple comparisons.

Salivary glands from SS patients were characterized by extensive lymphocyte infiltration, which was also present in the samples we collected (Figures 1A–C). We further isolated CD4⁺ T cells from the peripheral blood of SS patients and conducted PCR verification. We found higher expression of HUWE1 in CD4⁺ T cells of the SS group (Figure 1D). Furthermore, immunohistochemical and immunofluorescence staining were performed on labial gland samples from SS patients. Immunohistochemical results showed a possible overlap between the CD4 positive area and the HUWE1 positive area (Supplementary File S2A). Immunofluorescence further confirmed an increased presence of CD4⁺ T cells in the positive HUWE1 region, as compared to the negative HUWE1 region (Figure 1E; Supplementary File S2B). In our analysis of the gene microarray dataset (GSE143153) from the GEO database, we identified an increased expression of HUWE1 in CD4⁺ T cells derived from labial glands of SS patients, compared to healthy controls (Figure 1F). These latter findings provide an external validation of the observed upregulation of HUWE1 in SS. However, no significant correlation was observed between HUWE1 expression and clinical characteristics, such as age and sex (Figure 1F). Using splenic CD4⁺ T cells from mice, we examined the expression of ubiquitin ligase genes in activated cells by gene sequencing (Yin et al., 2022c). As shown in Figure 1G, Huwe1 was highly expressed in activated murine CD4⁺ T cells. The results of PCR assays showed that Huwe1 expression was significantly upregulated in CD4⁺ T cells after activation (Figure 1H). Immunofluorescence staining also showed that HUWE1 expression was significantly increased in activated versus resting CD4⁺ T-cells (Figure 1I). This further supports the hypothesis that HUWE1 may play an important role in CD4⁺T cell infiltration foci.

We obtained sequence data of labial salivary gland tissues (GSE94510) from SS patients from the NCBI GEO database to define the functions of HUWE1. These data provide insight into the role of HUWE1 in SS pathogenesis (Edgar et al., 2002). We interrogated eighteen SS samples divided into two groups based on the HUWE1 expression level, and performed a functional enrichment analysis to infer biological functions of DEGs between the two groups. Functional analysis revealed a role for HUWE1 in T-cell receptor signaling pathway, lymphocyte homeostasis, and protein polyubiquitination (p-value < 0.05, Figure 2A). These findings indicated that HUWE1 is likely to be involved in the regulation of T-cell activation. We then tested the effect of HUWE1 on CD4⁺ T cell activation by treating murine splenic CD4⁺ T cells with the selective HUWE1 inhibitor BI8626 (Peter et al., 2014) and sh-Huwe1 (Supplementary Files S2C, D). Murine CD4⁺ T-cell proliferation was inhibited by BI8626 and sh-Huwe1, as shown in CFSE proliferation assays (Figure 2B). Following anti-αCD3/CD28 stimulation, CD4⁺ T cells treated with BI8626 or sh-Huwe1 exhibited a decreased frequency of blasts and a significantly reduced median size, as measured by the forward scatter area (FSC) on the flow cytometric analyses, compared to the control group (Figures 2C, D). Huwe1-deficient CD4⁺ T cells also exhibited reduced granularity as evidenced by a marked decrease in side scatter area (SSC). In addition, no significant change in overall apoptosis level of cells was observed across different treatment groups (Supplementary File S2E). To further investigate cell cycle changes influenced by cell proliferation, flow cytometry was used to analyze cell cycle distributions in cells treated with BI8626 or sh-Huwe1. Results showed an increased number of cells in the S phase in the BI8626 or sh-Huwe1 treatment groups (Supplementary File S2F). This was accompanied by a significant reduction in the expression of CDK2, a key regulator of S phase progression (Supplementary File S2G). To investigate the underlying mechanism, we examined T-cell activation. Analysis of three activation biomarkers-CD69, CD25, and IL-2 in CD4⁺ T cells following treatment with BI8626 revealed a non-significant decrease in the CD69 positive rate and Cd69 mRNA expression. Mean fluorescence intensity (MFI) remained largely unchanged, indicating that the majority of CD4⁺ T cells treated were still able to respond to early activation signals (Figure 2E). The MFIs of CD25 and IL-2 were significantly reduced (Figures 2F, G). (Shi et al., 2016) CD25 is an intermediate marker of T cell activation and is essential for their proliferation following TCR engagement (Boyman and Sprent, 2012), (Ross and Cantrell, 2018). In that case, the reduced expression of both IL-2 cytokine and its receptor, CD25, likely contribute to defective proliferation observed in the BI8626 treatment group. Similarly, sh-Huwe1 did not inhibit CD69, but inhibited the expression of CD25 and IL-2 in these cells (Supplementary File S3A). Next, we detected the expression of phosphorylated ZAP (pZAP70), an important tyrosine kinase along the TCR signaling pathway, in CD4⁺ T cells upon activation of anti-CD3α (Figure 2H). Remarkably, we observed a decrease in the MFI of pZAP70 in both the BI8626 and sh-Huwe1 groups, indicating a weakening of TCR signaling. These findings suggest that inhibition of Huwe1 can impede the activation and proliferation of CD4⁺ T cells.

The composition and arrangement of lipids in the plasma membrane dictate its “lipid order”, an essential factor for the correct localization of signaling proteins during the formation of immune synapses. Glycosphingolipids and cholesterol are the two important components of lipid order, and any alteration in their relative abundance and arrangement can significantly affect membrane protein interactions and signal transduction (Kidani et al., 2013), (Armstrong et al., 2010), (Bensinger et al., 2008).

We measured glycosphingolipid expression through fluorescently conjugated Cholera Toxin Subunit Beta (CTB), a well-established surrogate glycosphingolipid marker (Lencer and Tsai, 2003). After 24 h of αCD3/CD28 stimulation, CD4⁺ T cells treated with BI8626 showed no significant change in CTB fluorescence, indicating no alteration in glycosphingolipid concentration upon early activation (Figure 2I). In contrast, upon treatment with NB-DNJ (UDP-glucose ceramide glucosyltransferase inhibitor) to blockade glycosphingolipid synthesis, CTB fluorescence was significantly downregulated. This expression of glycosphingolipid remained unchanged until 48 h after activation (Figure 2I), suggesting that sphingolipid may not be directly associated with the inhibition of the early activation of BI8626 treated T cells. To further investigate this phenomenon, we used Fillipin III staining to quantify the cholesterol content of murine CD4⁺ T cells activated by anti-CD3 plus anti-CD28 antibodies. We found that both BI8626 and sh-Huwe1 treatment decreased the amount of cholesterol (Figure 2J), which was further confirmed by total cholesterol detection (Figure 2K). Together, these findings indicate that inhibition of cholesterol content may be a potential mechanism underlying the suppression of CD4⁺ T-cell activation observed with BI8626/sh-Huwe1 treatment.

Exogenously supplementing CD4⁺ T cells with soluble cholesterol (10μM, Sigma Aldrich), a methyl-β-cyclodextrin (MBCD)-conjugated form of the compound, can ameliorate defects in lipid homeostasis and improve cell survival (Hong and Amara, 2010), (Idowu and Hagenbuch, 2022), (Wilfahrt et al., 2021), (Marty-Roix et al., 2016), (Kidani et al., 2013). Interestingly, cholesterol was found to slightly rescue the blasting ratio, cell size, and cell granularity of CD4⁺ T cells treated by BI8626 (Figure 3A). Cholesterol had a certain effect on cell granularity in the sh-Huwe1 group, but did not influence the blasting ratio and cell size (Figure 3B). There was no significant difference in the CD69 MFI between BI8626/sh-Huwe1 treatment group and the control group, while the addition of cholesterol stimulated the upregulation of CD69 in the BI8626 treatment group (Figures 3C, D). Furthermore, the exogenous supplementation of soluble cholesterol was found to attenuate the suppressive impact of BI8626/sh-Huwe1 on CD25 expression (Figures 3E, F). This approach also showed an obvious rescue effect on IL-2 expression of BI8626 treated CD4⁺ T cells (Figures 3G, H). Importantly, the exogenous administration of cholesterol improved p-ZAP70 levels after cell activation, indicating its potential to counteract the inhibitory effects of BI8626/sh-Huwe1 on TCR signaling intensity (Figure 3I). In addition, cholesterol also increased the proportion of proliferating CD4⁺ T cells (Figure 3J; Supplementary Files S3B, C). Thus, we posit that HUWE1/Huwe1 inhibition-mediated regulation of CD4⁺ T cell activation and proliferation is intricately linked to cholesterol, highlighting the importance of this metabolite.

Based on our previously published sequencing data (Yin et al., 2022c), we found that the expression of key genes involved in de novo cholesterol synthesis was significantly increased after the activation of murine CD4⁺ T cells (Figure 4A). To confirm whether the inhibition of HUWE1/Huwe1 affects the cholesterol synthesis pathway, we detected the expression of several relevant genes. Intriguingly, we found that BI8626 and sh-Huwe1 treatments did not inhibit the mRNA levels of Hmgcr, Hmgcs1, or Sqle (Figure 4B). WB analysis revealed a significant increase in the protein levels of HMGCR in CD4⁺ T cells treated with BI8626/sh-Huwe1 (Figures 4C, D). This may be caused by the combination of upregulation of HMGCR transcription and increased stability of HMGCR due to sterol deficiency (Leichner et al., 2011). The expression of other key rate-limiting enzymes involved in cholesterol biosynthesis remained relatively stable upon Huwe1 inhibition (Figures 4C, D). Combined with the fact that the cellular cholesterol content of BI8626/sh-Huwe1 group was upregulated in response to the CD3/CD28 stimulation (Figure 2K), we believe that Huwe1 inhibition does not affect cholesterol biosynthesis. Therefore, we tested the cholesterol efflux ability of CD4⁺ T cells. Both BI8626 and sh-Huwe1 treatments dramatically enhanced apo AI-mediated and HDL-mediated cholesterol efflux from CD4⁺ T cells, as evidenced by the results shown in Figures 4E, F. These findings suggest that the inhibition of Huwe1 may stimulate active cholesterol efflux, ultimately reducing intracellular cholesterol levels. It is known that activated T cells typically inhibit cholesterol efflux to maintain stable intracellular cholesterol levels, thereby facilitating further proliferation and effector differentiation (Michaels et al., 2021). Additionally, we also found that the mRNA expression levels of key cholesterol efflux transporters ABCA1, ABCG1, and NR1H3 in CD4⁺T cells of SS patients were significantly lower than those from healthy controls (Supplementary File S3D), suggesting that a weakening of cholesterol efflux function of CD4⁺ T cells is involved in the pathophysiology of SS.

We performed transcriptome sequencing on spleen CD4⁺ T cells from various groups of mice and found that the expression profiles of BI8626 and sh-Huwe1 groups were similar (Figures 4G–I). Notably, our results reveal a significant upregulation in mRNA expression of Abca1 upon treatment with BI8626/sh-Huwe1 (Figure 4J). However, our analysis did not reveal any statistically significant differences in the expression of Abcg1 and Scarb1 under the tested experimental conditions. We utilized RT-qPCR to verify the expression of several cholesterol efflux-related genes in murine CD4⁺ T cells. Specifically, we observed a significant upregulation in the mRNA expression of Abca1 in both BI8626 and sh-Huwe1 groups, while there was no difference in the expression of Abcg1 and Scarb1 (Figure 4K). Since the main function of HUWE1 is to regulate ubiquitinosomal degradation of proteins, we performed mass spectrometry (MS) to identify the proteins expressed in CD4⁺T cells. The fragment mass tolerance measured by MS conformed to a normal distribution (Figure 4L). Unfortunately, the abundances of ABCA1, ABCG1 and SR-B1 in murine CD4⁺ T cells were not high enough to be detected by the mass spectrometer (Supplementary File S4). Our WB results showed that both BI8626 and sh-Huwe1 could significantly upregulate the expression of ABCA1 and SR-B1 proteins (Figures 4M, N), suggesting that E3 ubiquitin ligase HUWE1 might regulate the stability of ABCA1 and SR-B1.

Activated CD4⁺ T cells treated with cycloheximide (CHX) for 8 h showed profoundly reduced expression of ABCA1 and SR-BI proteins, suggesting that both may be subject to proteasome degradation in early activation (Supplementary File S5A). However, the levels of ABCA1 and SR-BI proteins gradually recovered over time, potentially due to a weakening of inhibition by CHX. No significant change in ABCG1 expression was observed after 8 h of treatment. To further understand the degradation of these proteins during early activation, protein levels were measured within 4 h and revealed a gradual decrease in ABCA1 and SR-BI expression, while ABCG1 expression remained unchanged (Figure 5A). Treatment of cells with the proteasome inhibitor MG-132 revealed that ABCA1 protein levels increased over time (Figure 5B), indicating that proteasome-mediated degradation is responsible for the reduction in ABCA1 expression. In contrast, no significant change in SR-BI levels was observed, suggesting that its decreased protein levels are likely due to other degradation mechanisms. Co-immunoprecipitation assays in activated CD4⁺ T cells revealed an interaction between HUWE1 and ABCA1 (Figure 5C). Although we also tested for a potential interaction between HUWE1 and ABCG1/SR-BI, we found no evidence of this in our analysis (Supplementary File S5B). Subsequently, a ubiquitin pull-down assay in activated CD4⁺ T cells showed that knocking-down Huwe1 decreased the ubiquitination of ABCA1 (Figure 5D). Collectively, these results suggest that HUWE1 interacts with ABCA1, diminishing its stability by mediating its ubiquitination-driven degradation, thus suppressing ABCA1 protein levels.

To confirm that HUWE1 depends on ABCA1 to regulate activation and proliferation, we treated CD4⁺ T cells with DIDS, a selective inhibitor of ABCA1. We observed that DIDS markedly increased the proliferation rate of CD4⁺ T cells treated with BI8626/sh-Huwe1 (Figure 5E). The blasting ratio and cell size of CD4⁺ T cells treated by BI8626 were rescued by DIDS (Figure 5F). In addition, DIDS also restored the expression of CD25, which had been reduced by BI8626/sh-Huwe1 (Figures 5G, H), suggesting that HUWE1 regulates T-cell activation through controlling the stability of ABCA1.

To assess the potential impact of targeting HUWE1 on SS, we conducted in vivo experiments using NOD/ShiLtj mice, a validated SS animal model. Nod/ShiLtj mice showed a large number of lymphocyte infiltrates in the submandibular gland, with CD4⁺ T cells forming the majority (Supplementary File S5C). Such locally infiltrated CD4⁺ T cells are usually overactivated (Fu et al., 2020). The disease phenotype in NOD/ShiLtj mice, including abnormal exocrine gland function and pathological changes, is analogous to the clinical manifestations of SS. To test the effect of BI8626 on thymocyte development, we evaluated the proportions of the major thymocyte populations (double-positive, CD4 single-positive, and CD8 single-positive cells) and found no significant changes (Figures 6A–D). No significant alteration in the proportions of single-positive CD4 and CD8 T cells in the spleen was observed following BI8626 treatment (Figure 6E). In contrast, we observed a slight decrease in the proportion of CD4⁺ T cells in the cervical lymph nodes (Figure 6F), indicating that local inflammation may have been ameliorated. Analysis of CD44 and CD62L expression revealed comparable frequencies of naïve (CD44⁻CD62L⁺), central memory (CD44⁺CD62L⁺), effector memory (CD44⁺CD62L⁻) and effector (CD44⁻CD62L⁻) T-cell subsets between control and BI8626-treated mice (Figures 6G, H). These findings suggest that BI8626 does not significantly alter the composition of the peripheral T-cell compartment. We examined SFR in NOD/ShiLtj mice with and without treatment on a weekly basis. Submandibular glands were isolated for HE staining, as previously described (Fu et al., 2020). In the control group, SFR decreased significantly over time; however, this reduction was partially counteracted in the BI8626-treated group (four times a week) (Figure 6I). This suggests that BI8626 treatments may ameliorate SS progression. Furthermore, histological analysis through HE staining showed a lower lymphocyte infiltration in the BI8626-treated (four times a week) group compared to the controls (Figure 6J). Immunohistochemical staining revealed a marked reduction in CD4⁺ T cells within the salivary gland lymphocyte infiltration foci of BI8626-treated NOD/ShiLtj mice, compared to the control group (Figure 6J). This was accompanied by a decrease in both CD4⁺ and CD8⁺ T cells (Figures 6K, L). Moreover, BI8626 treatment resulted in a significant decrease in the number of proliferating cells (PCNA positive) within the lymphocyte infiltration foci (Figure 6J), indicating a suppression of CD4⁺ T-cell proliferation. Collectively, these findings demonstrate that BI8626 suppresses CD4⁺ T-cell accumulation and proliferation in the submandibular glands of NOD/ShiLtj mice, providing a mechanistic explanation for the observed attenuation of symptoms. Additionally, we compared the serum concentrations of TNFα, IL-17, IFN-γ, and IL-6 between the control and BI8626-treated (four times a week) groups, and found that TNFα was decreased in the latter group (Figure 6M). While the resulting infiltration score was not significantly different (Figure 6N), the lymphocyte focus number in both BI8626-treated groups was significantly reduced compared to the controls (Figure 6O). Collectively, these results demonstrate that BI8626 treatments may be beneficial in ameliorating SS progression. These results suggest that HUWE1 inhibition can reduce disease progression by downregulating the activity and numbers of CD4⁺ T cells. Thus, our findings provide a rationale for the therapeutic potential of BI8626 in SS.