Female MRL/lpr mice were purchased from SLC Inc (Japan). These animals were maintained under specific pathogen-free conditions from 7–16 weeks of age. C57BL/6 mice were purchased from OrientBio (Korea). All procedures of animal research were performed in accordance with the Laboratory Animals Welfare Act, the Guide for the Care and Use of Laboratory Animals and the Guidelines and Policies for Rodent experiment provided by the Institutional Animal Care and Use Committee (IACUC) of the School of Medicine, The Catholic University of Korea (approval numbers: CUMS-2018-0341-01 and 2018-0236-01). Baricitinib (MyBioSource) was administered to MRL/lpr mice twice daily at 10 mg/kg in 0.5% methyl cellulose (Sigma) from 8 weeks of age for a total of 8 weeks. Body weights were checked every two weeks.

Mice sera were diluted at 1:100,000 for immunoglobulin (Ig) G, 1:50,000 for IgG2a, 1:1,000 for anti-double strand DNA (dsDNA) IgG, 1:10 for IL-6 and IL-17, and 1:20 for B cell activating factor (BAFF) respectively. Mouse total IgG quantitation set and mouse total IgG2a quantitation set were acquired from Bethyl laboratory. Mouse anti-dsDNA IgG quantitation kit was purchased from Alpha Diagnostics. Mouse IL-6 ELISA Duoset, mouse IL-17 ELISA Duoset, and mouse BAFF ELISA Duoset were obtained from R&D systems. Albumin and creatinine levels in urine samples of mice at 16 weeks of age were analyzed using mouse albumin ELISA set (Bethyl lab) and creatinine parameter assay kit (R&D systems), respectively. All tests were performed following the manufacturer’s instructions.

Spleens and kidneys were minced and filtered through a 40-μm cell strainer (Falcon) to prepare single-cell suspensions. Red blood cells (RBCs) were removed with ammonium-chloride-potassium (ACK) lysis buffer. For intracellular staining, cells were stimulated with 25 ng/mL phorbol 12-myristate 13-acetate (PMA, Sigma) and 250 ng/mL ionomycin (Sigma) with monensin-containing Golgistop (BD Biosciences) for five hours. Cells were harvested and stained with a eFluor 780-Fixable viability dye (eBioscience) to exclude dead cells for analysis. Cells were stained with Pacific blue anti-CD90.2, PerCP-Cy5.5 anti-CD4, PE anti-CD8a, APC anti-CD19, FITC anti-CXCR5, BV605 anti-CD279 (PD-1), BV605 anti-CD44, Alexa 488 anti-CD62L (Biolegend), APC anti-CD25, PE anti-Foxp3 antibodies (eBioscience), and PE anti-CD138 antibodies (BD Biosciences). Flow cytometric analysis was performed using BD LSRFortessa (BD Biosciences). Data were analyzed using BD FACSDiva (version 10, BD Biosciences).

Mice were anesthetized using a zoletil-xylazine cocktail during perfusion with phosphate-buffered saline (PBS). Kidney was vertically dissected and fixed with 10% formalin. Kidney sections (3 μm in thickness) were stained with Periodic acid-Schiff (PAS) stain (Sigma). Kidney pathology was evaluated using a lupus nephritis classification system as described in a previous study (16). Immunohistochemistry was performed using a Vectastain ABC kit (Vector). Tissue sections were incubated with anti-nephrin (Progen), anti-synaptopodin (Abcam), anti-podocin (Abcam), or isotype control antibodies (Abcam) at 4°C overnight. Staining was developed using 3,3′-diaminobenzidine chromogen (Dako). Sections were counterstained with hematoxylin QS (Vector).

Tissue sections were stained with antibodies to IgG (Bethyl lab) and nephrin (Abcam) for 16 hours at 4°C followed by staining with Alexa594 conjugated anti-rabbit IgG, Alexa488 conjugated anti-guinea pig IgG, and Alex594 conjugated anti-phalloidin (Invitrogen). Nuclei were stained with 4’, 6-diamidino-2-phenylindole dihydrochloride (DAPI, Invitrogen). Isotype-control staining was conducted via probing with rabbit IgG rather than with primary antibodies. Confocal images were acquired using an LSM 700 confocal microscope (Zeiss).

Mice spleens were removed and prepared for single cell suspension using glass teasing slides and a 40-μm cell strainer (BD Biosciences). RBCs were lysed with ACK lysis buffer (Thermo). Splenocytes were cultured with RPMI1640 media (Gibco) supplemented with 10% fetal bovine serum (AB Frontier).

CD19+ B cells from C57BL/6 mice and MRL/lpr mice were obtained by magnetic activated cell sorting using CD19 microbeads (Miltenyi Biotec). The purity was over 95% as demonstrated by flowcytometry. CD19+ B cells were stimulated with 1 μg/mL F(ab′)2-goat anti-mouse IgM antibodies (eBioscience), 250 μg/mL soluble CD40 ligand (sCD40L) and 100 ng/mL recombinant mouse IL-4 (Peprotech) for 2 or 5 days with or without baricitinib pre-treatment. RPMI1640 medium supplemented with a 10% FBS was used for their culture.

Primary mouse glomeruli were harvested using magnetic isolation protocols. Briefly, C57BL/6 mice that had received Dynabead M-450 (Invitrogen) perfusions were euthanized. Their kidneys were dissected and digested with type I collagenase (Gibco) for 30 minutes at 37°C. They were then passed through a 100-μm cell strainer. The supernatant was discarded and the cell pellet was resuspended in 2 mL of Hanks balanced salt solution (HBSS, Gibco). Dynabead-trapped glomeruli were isolated using a magnetic particle concentrator and washed with HBSS. For podocyte primary culture, freshly isolated glomeruli were plated onto 60 mm dishes coated with collagen at 37°C in DMEM/F12 (Gibco) supplemented with 5% fetal bovine serum (Gibco), 0.5% Insulin-Transferrin-Selenium liquid (Gibco), and 1% penicillin–streptomycin (Gibco). The outgrowth of podocytes started between day 2 and 3. Seven days later, glomeruli and outgrowth cells were detached from the plate using a 0.25% trypsin-EDTA solution (Gibco). Trypsinized cells were strained by passing a 40-mm cell strainer and plated onto collagen-coated dishes. Podocytes of passages 1 were used in all experiments. Mouse podocyte cell line E11 was acquired from Cell Line Service (Germany). These cells were maintained with RPMI1640 media supplemented with 10% FBS at 33°C in a humidified atmosphere with 5% CO2. To differentiate podocytes, cells were transferred to a non-permissive condition at 37°C for 12–14 days.

Total RNA was collected using an RNAiso Plus reagent (Takara). Up to 1–2 µg of total RNA was converted to complementary DNA (cDNA), using a Transcriptor First-Strand cDNA Synthesis kit (Roche). A LightCycler 96 instrument (Roche) was used for PCR amplification and analysis. All reactions were performed with SYBR Green I Master, according to the manufacturer’s instructions. Primers were designed with a web tool from GenScript (http://www.genscript.com). Primer sequences are as follows (forward and reverse, respectively): Actb (beta-actin), 5′ - GAAATCGTGCGTGACATCAAAG - 3′ and 5′ - TGTAGTTTCATGGATGCCACAG - 3′; Gapdh, 5′ - CAGCAACTCCCACTCTTCCAC - 3′ and 5′ - TGGTCCAGGGTTTCTTACTC - 3′; Nphs1 (nephrin), 5′ - ACTACGCCCTCTTCAAATGCA - 3′ and 5′ - TCGAGGGCCTCATACCTGAT - 3′; Synpo (synaptopodin), 5′ - TATCAACCAGAACCCGTC - 3′ and 5′ - AATCAAGTGTGCCATTCG - 3′; Nphs2 (podocin), 5′ - CCATCTGGTTCTGCATAAAGG - 3′ and 5′ - CCAGGACCTTTGGCTCTTC - 3′; Aicda, 5′ - CGTGGTGAAGAGGAGAGATA - 3′ and 5′ - CAGTCTGAGATGTAGCGTAG -3′; Bcl6, 5′ - CCCTGTGAAATCTGTGGCAC - 3′ and 5′ - ACACGCGGTATTGCACCTTG - 3′; Xbp1, 5′ - GCAAGTGGTGGATTTGGAAG - 3′ and 5′ - CCTCTGGAACCTCGTCAG - 3′; Irf4, 5′ - ACCAGTCACACCCAGAAATC - 3′ and 5′ - GGGCACAAGCATAAAAGGTTC - 3′; Ifit1, 5′ - CTGAAATGCCAAGTAGCAAGG - 3′ and 5′ - CCAAAGGCACAGACATAAGGA - 3′; Isg15, 5′ - AGTGCTCCAGGACGGTCTTA - 3′ and 5′ - TCGCTGCAGTTCTGTACCAC - 3′; Mx1, 5′ - GATCCGACTTCACTTCCAGA -3′ and 5′ - CATCTCAGTGGTAGTCAACC- 3′. All mRNA expression levels were normalized to the expression of Actb or Gapdh. Relative fold induction was calculated using 2 - (ΔCq) or 2 - (ΔΔCq), where ΔΔCq was ΔCq (stimulated) - ΔCq(unstimulated), ΔCq was Cq (target) - Cq (Actb or Gapdh), and Cq was the cycle at which the threshold was crossed. PCR product quality was monitored using post-PCR melting curve analysis.

Cytosolic proteins were extracted using RIPA buffer containing Halt protease/phosphatase inhibitor cocktail (Thermo). Membrane fraction of cell lysates were extracted using Mem-PER Plus membrane protein extraction kit (Thermo). For immunoblotting, proteins (15–30 μg for each sample) were separated by 10–12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), transferred onto polyvinylidene fluoride membranes (Biorad) and probed with the following antibodies: anti-pSTAT1_Y701, anti-STAT1, anti-pSTAT3_Y705, anti-pSTAT3_S727, anti-STAT3, anti-p-JAK2, anti-JAK2 (Cell signaling technology), anti-nephrin (Progen), anti-podocin, anti-Na/K ATPase (Abcam) and anti-β-actin (Sigma). Subsequently, membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG or goat anti-mouse IgG (Thermo). Reactive proteins on membranes were visualized using a SuperSignal West Pico Chemiluminescent substrate (Thermo). These membranes were then exposed to an Amersham Imager 600 (GE Healthcare).

All statistical analyses were performed using GraphPad Prism (version 8, GraphPad Software). Statistical significance was determined by the Mann Whitney test, or by one-way analysis of variance with Tukey multiple comparison test. Statistically significant difference was considered when p value was less 0.05.

MRL/lpr mice presented lupus-like features such as organomegaly and renal inflammation with seropositivity of multiple autoantibodies from the age of 12 weeks (17). To determine whether baricitinib could suppress lupus-mimicking phenotypes in these mice, we administered vehicles with or without baricitinib to 8-week-old MRL/lpr mice for 8 weeks. Baricitinib-treated mice showed higher body weights during in vivo experiments ( Figure 1A ) from 8-week-old to 16-week-old. Baricitinib treatment resulted in lower degrees of splenomegaly and lymphadenopathies at the age of 16 weeks ( Figure 1B and Supplementary Figure 1 ) compared to control (vehicle treatment). Weights of spleens and cervical lymph nodes from baricitinib-treated mice were significantly lower than those of controls ( Figure 1C ). Circulating IgG, IgG2a, and anti-dsDNA IgG levels measured from sera of mice were significantly lower in the baricitinib-treated group than in the vehicle control group ( Figure 1D ). Serum levels of lupus-related pro-inflammatory cytokines including IL-6, BAFF, and IL-17 were markedly decreased by baricitinib treatment ( Figure 1E ). These results suggest that the administration of baricitinib to MRL/lpr mice can ameliorate the lupus-like phenotypes by suppressing systemic autoimmunity.

MRL/lpr mice with systemic autoimmune problem showed abnormalities of T and B cells, resembling human SLE (18). Using flow cytometric analysis, we measured absolute counts and proportions of T cell and B cell subsets among splenocytes from baricinitib-treated or vehicle-only-treated mice. Total T cells (CD90.2+) and CD8+ T cells were significantly decreased in 16-week-old mice treated with baricitinib than in control mice treated with the vehicle ( Figure 2A ). DNT cells (CD4-CD8-) are related to SLE pathogenesis and activities of lupus-prone mice and SLE patients (19, 20). Their numbers increased according to the disease severity in this mouse model. Baricitinib reduced the numbers of DNT cells in MRL/lpr mice at the age with the disease fully blown whereas total B cells (CD19+) were not influenced. Intriguingly, among various T cell subsets, central memory CD8+ T cells (CD44+CD62L+, Supplementary Figure 2 ) were the most remarkably reduced subpopulation after baricitinib treatment ( Figure 2B ). Their roles have not been fully defined in the pathogenesis of SLE. Furthermore, the proportion of regulatory T (Treg) cells (CD4+CXCR5-CD25+FOXP3+) and follicular regulatory T (Tfr) cells, which can modulate germinal center (GC) reactions, were significantly increased in splenocytes from baricitinib-treated mice while the absolute numbers of Treg cells and Tfr cells, and both the absolute numbers and the proportion of follicular helper T (Tfh) cells showed no significant differences ( Figure 2C and Supplementary Figure 3 ) after baricitinib treatment. Collectively, these results indicate that baricitinib can modulate abnormal expansion of lupus-pathogenic T cell populations by reducing DNT cells while increasing Treg and Tfr cells.

Deposition of immune complex and in-situ infiltration of various immune cells can cause profound inflammation in renal tissues of MRL/lpr mice. Such renal inflammation can also occur in SLE patients, frequently resulting in compromised renal function. Therefore, renal inflammation is regarded as the main cause of mortality of patients with SLE. The objective of this study was to evaluate whether administration of baricitinib could reduce renal inflammation in a lupus-animal model. Kidney tissues were acquired from 16-week-old MRL/lpr mice with or without baricitinib treatment. They were then assessed for the degree of inflammation. Histopathologic findings showed markedly reduced inflammation in the glomerulus, tubules, and perivascular area of kidney tissue sections from baricitinib-treated mice ( Figures 3A, B ). Urine albumin/creatinine ratio was also remarkably decreased in mice treated with baricitinib, indicating reduced proteinuria and renal damages ( Figure 3C ). Confocal microscopic images demonstrated decreases of IgG deposition in renal tissues from baricitinib-treated mice compared to control (vehicle-treated) mice ( Figure 3D ). In addition, Treg cells in murine kidneys were increased by baricitinib treatment ( Figure 3E and Supplementary Figure 4 ).

Podocytes are specialized epithelial cells that can maintain selective filtering functions of renal glomerulus due to their unique morphologic features known as foot processes and slit diaphragms. Renal inflammation in patients with LN can result in decreased expression of functional proteins in podocytes, thereby affecting the structural integrities of these specialized cellular barriers (21). We investigated whether baricitinib could restore podocyte functions by increasing impaired expression of intra- and extra-cellular key proteins in podocytes of lupus-prone mice. At first, decreased expression levels of nephrin and podocin as elemental proteins of slit diaphragm and synaptopodin as the main regulator of actin cytoskeleton of foot process in glomeruli from patients with LN were confirmed by comparing with healthy controls ( Figure 4A ). We then assessed expression levels of these functional proteins in podocytes from MRL/lpr mice treated with or without baricitinib. Immunohistological results demonstrated that all podocyte-related proteins expressed in renal glomeruli from baricitinib-treated mice at the age of 16 weeks were remarkably increased ( Figure 4B ). Expression levels of mRNA for nephrin (Nphs1), podocin (Nphs2) and synaptopodin (Synpo) measured from isolated glomeruli ( Figure 4C ) of baricitinib-treated mice were also significantly elevated than those from the vehicle-only-treated mice ( Figure 4D ). All these results indicate that baricitinib has therapeutic effects on renal inflammation in MRL/lpr mice. In addition, improved SLE features of mice after treatment with baricitinib were due to its ability to restore functions of specialized glomerular cells, podocytes, in targeted tissues.

Cytokines related to B cell differentiation and maturation exert their functions mainly via the JAK/STAT pathway (13, 22, 23). Therefore, we determined whether the administration of baricitinib could repress the expression of genes related to differentiation of naïve B cells. CD19+ B cells were isolated from spleens of C57BL/6 mice ( Figures 5A–D ) and MRL/lpr mice ( Supplementary Figure 5 ) using microbeads. Their purity was validated by flow cytometric analysis ( Figure 5A ). These cells were treated by graded doses of baricitinib and stimulated under the same conditions, inducing B cell maturation with anti-mouse IgM antibodies, sCD40L, and IL-4 for 2 or 5 days, respectively. After two days of stimulation, the expression level of Aicda was found to be significantly suppressed by baricitinib in a dose-dependent manner, whereas that of Bcl6 was increased ( Figure 5B ). Expression levels of Xbp1 and Irf4 were also remarkably decreased after 5 days of stimulation ( Figure 5C ). In addition, the amount of IgG production measured from supernatants under the same conditions was significantly reduced depending on increasing doses of baricitinib after both two days and five days of stimulation, respectively ( Figure 5D ). The similar tendency was also replicated in the results of the in vitro experiments using B cells from MRL/lpr mice ( Supplementary Figure 5 ).

Blockade of the JAK/STAT pathway in B cells was assessed by measuring expression levels of phosphorylated STAT proteins using immunoblot. Baricitinib treatment effectively suppressed the expression of phosphorylated STAT1 (pSTAT1) and STAT3 (pSTAT3), both tyrosine 705 (pSTAT3_Y705) and serine 727 (pSTAT3_S727) of pSTAT3, respectively, in naïve B cells from C57BL/6 mice after stimulation with type I and II IFNs ( Figure 5E ). These results were consistent with results using CD19+ B cells acquired from splenocytes of MRL/lpr mice at the age of 16 weeks and treated with the same manner as previously described ( Figure 5F ). Overall, baricitinib effectively repressed signals of the JAK/STAT pathway in B cells both in vitro and in vivo. These effects on naïve B cells could result in suppression of B cell differentiation and immunoglobulin production that are critical for the pathogenesis of systemic autoimmunity in SLE.

Previously, we have demonstrated that baricitinib could restore expression levels of functional proteins in podocytes and their mRNA levels in murine kidneys. These proteins are essential for maintaining cytoskeletal structures of podocytes and glomerular filtration functions (21). Therefore, we investigated whether baricitinib treatment could restore the actin cytoskeleton of podocytes disrupted by inflammation. Immortalized murine podocyte cell line in permissive condition was stimulated by IFNs with or without baricitinib. Inflammatory stimulus induced disintegration of cytoskeletal structures of podocytes as shown in confocal microscopic images. However, baricitinib treatment recovered it ( Figure 6A ). Expression levels of IFN-stimulated genes (ISGs) and intracellular proteins related to the JAK/STAT pathway including phosphorylated JAK2 (pJAK2), pSTAT1, and pSTAT3_Y705 in differentiated podocytes were significantly down-regulated by treatment with baricitinib, whereas those of functional and structural proteins such as nephrin and podocin were increased by baricitinib ( Figures 6B, C ).

In addition to immortalized podocytes, primary podocytes were also isolated from glomeruli of C57BL/6 mice as described in Materials and Methods. These primary podocytes presented reduced expression of nephrin and distorted skeletal structures after IL-6 stimulation. However, such morphological abnormalities were ameliorated by treatment with baricitinib ( Figure 6D ). Furthermore, baricitinib treatment effectively suppressed phosphorylation of JAK2, STAT3, and STAT1 in these primary podocytes in a dose-dependent manner ( Figure 6E ). These in vitro results suggest that baricitinib could stabilize structural abnormalities of podocytes caused by inflammation related to the JAK/STAT pathway.