All animal experimental procedures and housing have been approved by the Institutional Animal Care and Use Committee (IACUC) of Virginia Polytechnic Institute and State University. In this study, MRL/lpr (MRL/MpJ-Fas^lpr/J, Stock No. 000485), B6/lpr (B6. MRL-Fas^lpr/J, Stock No. 000482), MRL (MRL/MpJ, Stock No.000486), B6 (C57BL/6J, Stock No. 000664) and hCD2-iCre (B6-Cg-Tg (CD2-icre)4Kio/J, Stock No. 008520) mice were purchased from Jackson Laboratory (JAX, Maine, United States) and bred in-house.

Although the MRL/lpr model has been widely used to study human lupus, it has several drawbacks that prevent it from being easy to generate MRL/lpr mice with specific gene deletion. This include that MRL/lpr mice have a mixed genetic background. Consequently, multiple generations of backcross are needed to generate knockout mice in MRL/lpr background. It is even more difficult and time-consuming for generating conditional knockout MRL/lpr mice since the commercially available CRE transgenic mouse stains are in B6 background. We, therefore, generated autoinflammatory-prone B6/lpr mice with conditional deletion of either miR-183C cluster or miR-182 in lymphocytes to investigate the role of miR-183C and miR-182 in lupus in vivo. We bred hCD2-iCre mice with B6/lpr mice in-house to generate CD2-CreB6/lpr mice. The miR-183C^fl/fl and miR-182^fl/fl mice in a B6-129/SvJ mixed genetic background were generated by Dr. David G. Kirsch’s group at Duke University Medical Center. The development and characterization of miR-182^fl/fl mice on a B6-129/SvJ mixed genetic background have been described in detail in a previous report (Sachdeva et al., 2014). An identical strategy was used to generate the miR-183C^fl/fl mice on a B6-129/SvJ mixed genetic background (Supplementary Figure S1) except that the loxP sites flank the entire miR-183, miR-96, miR-182 cluster instead of only miR-182. The miR-183C^fl/fl and miR-182^fl/fl mice in the mixed background were backcrossed with B6 mice for at least eight generations. The miR-183C^fl/fl and miR-182^fl/fl mice in a B6 background were crossbred with B6/lpr to generate miR183C^fl/flB6/lpr and miR-182^fl/flB6/lpr mice, which were then crossbred with CD2-CreB6/lpr mice to generate CD2-CremiR-183C^fl/flB6/lpr (miR183C^−/−B6/lpr) and CD2-CremiR-182^fl/flB6/lpr (miR-182^−/−B6/lpr) mice. The miR183C^−/−B6/lpr and miR-182^−/−B6/lpr mice were bred with miR-183C^fl/flB6/lpr and miR-182^fl/flB6/lpr mice, respectively to obtain experimental miR183C^−/−B6/lpr and miR-182^−/−B6/lpr mice. The littermate miR183C^fl/flB6/lpr and miR-182^fl/flB6/lpr, which do not carry the hCD2-iCre gene, served as wild-type control B6/lpr mice for conditional miR183 and miR-182 knockout B6/lpr mice. Due to the female predominance of lupus, only female mice were used in this study.

All mice were housed in our Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC)-certified animal facility at the Virginia-Maryland College of Veterinary Medicine (VMCVM), Virginia Tech. Mice were fed with a commercial 7013 NIH-31 Modified 6% Mouse/Rat Sterilizable Diet (Harlan Laboratory, Madison, WI, United States) and given water ad libitum. The mice at designated age were euthanized by CO₂ asphyxiation according to the IACUC approved protocol.

Proteinuria was measured by dipstick analysis using Chemistrip-2GP (Roche Diagnostics Corporation, Indianapolis, IN, United States). The semi-quantitative scale was demonstrated as follows: “-”: negative or trace; “+” 30 mg/dl; “++”: 100 mg/dl and “+++”: ≥500 mg/dl.

Peripheral blood was collected every 4 weeks by retro-orbital bleeding following IACUC requirements. Freshly-collected blood was maintained in 4°C refrigerator overnight to clot, and then centrifuged at 1,500× g for 15 min at 4°C to remove clots and collect supernatant as serum. The serum was stored at −80°C freezer.

Whole splenocytes were prepared by following standard laboratory procedures described in detail previously (Lengi et al., 2007; Edwards et al., 2017; Dai et al., 2019; Edwards et al., 2020). Per the manufacturer’s protocol, splenic CD4⁺ cells (T cells) and CD19⁺ cells (B cells) were purified from splenic lymphocytes with anti-CD4 (L3T4) and anti-CD19 MicroBeads (Miltenyi Biotec, San Diego, CA, United States), respectively. The purity of CD4⁺ (93.84% ± 2.03%) and CD19⁺ (91.88% ± 1.81%) cells were confirmed by flow cytometry.

To activate splenic lymphocytes, 2.5 × 10⁶ cells were seeded in 24-well plates and stimulated with 500 ng/ml of Lipopolysaccharide (LPS, from Sigma-Aldrich, St Louis, MO, United States) or anti-mouse CD3ε (plate coated at 5 μg/ml, Bio X cell, Lebanon, NH, United States) plus 2 μg/ml soluble anti-mouse CD28 (Bio X cell) for 24 h h, or 50 ng/ml phorbol 12-myristate 13-acetate (PMA) (Sigma-Aldrich) and 1 μg/ml ionomycin (Sigma-Aldrich) for 6 h. For the induction of IL-17, the splenocytes were stimulated with either non-pathogenic T helper 17 (Th17) stimuli (20 ng/ml IL-6 (BioLegend) plus 3 ng/ml TGFβ 1 (R&D System Inc., Minneapolis, MN, United States) and 1 μg/ml anti-CD3) or pathogenic Th17 stimuli (20 ng/ml IL-6 plus 10 ng/ml IL-1β (BioLegend), 10 ng/ml IL-23 (Invitrogen) and 1 μg/ml anti-CD3) for 72 h. The cell culture supernatants were collected for ELISA assay.

As previously described (Krützfeldt et al., 2005; Dai et al., 2016), the specific antagomirs against certain miR-182, miR-183, and miR-96 were designed based on mature murine miRNA sequences from miRbase (http://www.mirbase.org/) and shown in Table 1. The scramble control antagomir (Stittrich et al., 2010) and specific antagomirs were synthesized by GE Dharmacon (Lafayette, CO, United States). Antagomirs were delivered into splenic lymphocytes using a serum-free Accell siRNA delivery medium (GE Dharmacon) as we reported previously (Dai et al., 2016). Twenty-four hours after antagomir treatment, the cells were stimulated with concanavalin A (Con A; 5 μg/ml) or LPS (500 ng/ml) for 24 h or 48 h, the culture supernatants were collected to measure the cytokine levels by ELISA. The cell pellets were collected for Western blot.

As we reported previously (Dai et al., 2016; Wang Z. et al., 2018; Dai et al., 2019), total RNA containing small RNA was extracted from cells using the miRNeasy Mini Kit (Qiagen) following the manufacturer’s protocol. On-column DNA digestion was performed to remove any residual DNA in the RNA samples during RNA extraction. Concentration and purity of RNA were determined by NANODROP 2000 spectrophotometer (Thermo Fisher Scientific, United States) and samples with the OD260/280 ratio at around 2.0 were used for TaqMan miRNA Assay and RT-qPCR analysis.

As we previously reported (Dai et al., 2010; Dai et al., 2016; Wang Z. et al., 2018; Dai et al., 2019), TaqMan microRNA assays (Applied Biosystems, Waltham, MA, United States) were used to quantify miR-183C miRNAs (miR-182 ID 002599; miR-183 ID 00269; miR-96 ID 000186; snoRNA202 ID 001232, Applied Biosystems) expression levels according to the manufacturer’s instruction. For miRNA analysis, RNA samples were reverse transcribed with TaqMan microRNA reverse transcription kit, followed by quantitative real-time PCR (qPCR) with TaqMan miRNA assay reagent. For RT-qPCR of protein-coding genes, total RNA samples were reverse transcribed using High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Waltham, MA, United States) followed by qPCR using Power SYBR green PCR master mix (Applied Biosystems). The qPCR reactions were performed with a 7,500 Fast Real-Time PCR system (Applied Biosystems, Waltham, MA, United States). The relative expression level of miRNAs and mRNAs was normalized to endogenous small nucleolar RNA 202 (snoRNA202) and house-keeping gene β–actin respectively and calculated relative expression level using the 2^−ΔΔCt method. The primers for quantitative RT-PCR were as follows, Foxo1: (Forward) 5′-CGT​GCC​CTA​CTT​CAA​GGA​TAA​G-3′, (Reverse) 5′-GCA​CTC​GAA​TAA​ACT​TGC​TGT​G-3'; β–actin: (Forward) 5′-CGC​GAG​CAC​AGC​TTC​TT-3′, (Reverse) 5′-GCA​GCG​ATA​TCG​TCA​TCC​AT-3'.

The production level of interferon-gamma (IFNγ), interleukin 6 (IL-6), and interleukin 17 (IL-17) in cell culture supernatant were detected ELISA kits from BioLegend and ThermoFisher Scientific) respectively. Serum levels of IL-6, IFNγ, IL-17, TNF-a, and IL-10 were determined by the SP-X (previously Ciraplex^®) multiplex Chemiluminescent Assay kit (Quanterix, Billerica, MA, United States) as we previously reported (Dai et al., 2019; Edwards et al., 2020). The image of the chemiluminescent array plate was captured with the Aushon Cirascan™ Imaging system (Aushon BioSystems, InC., Billerica, MA, United States) and the image data was processed with Cirasoft software.

Anti-dsDNA autoantibody ELISA was performed to detect serum anti-dsDNA autoantibody production levels as we previously reported (Dai et al., 2010; Edwards et al., 2017). The plates were read at 380 nm with a Versa microplate reader (Molecular Devices).

The relative proportion of subsets of immune cells was quantified by flow cytometric analysis with specific cell surface markers as we previously reported (Dai et al., 2019; Dai et al., 2020). Briefly, the freshly-isolated splenocytes were blocked with purified anti-mouse CD16/CD32 antibody (eBiosciences/ThermoFisher Scientific), and stained with cell surface markers in MACS buffer (Phosphate-buffered saline (PBS, pH 7.2) supplemented with 0.5% bovine serum albumin (BSA) and 2 mM EDTA) for 15–20 min at 4°C. After washing, the stained cells were resuspended in 100 μl MACS buffer for flow cytometry analysis. Anti-CD3-FITC, anti-CD4-eFluor 450, anti-CD8a-Percp-Cy5.5, anti-CD19-Percp-Cy5.5, anti-B220-PE, anti-CD25-Percp/Cy5.5, anti-Foxp3-APC, and anti-GL7-eFluor660 were purchased from eBiosciences/ThermoFisher Scientific. Anti-CXCR5-APC/Cy7, anti-CD138-APC (PE), anti-IgD-PE were obtained from BioLegend, anti-PD1-BV421 was obtained from BD biosciences. FACSAria Flow cytometer (BD Biosciences, San Jose, CA) was used for stained cell profiling. Data were analyzed by FlowJo software (FlowJo LLC, BD Life Sciences).

Western blotting was performed by following the lab-established protocol (Dai et al., 2007; Dai et al., 2009). Briefly, whole cell extracts were prepared by lysing the cell pellets with CelLytic™ M Cell Lysis Reagent (Sigma-Aldrich). Protein lysates were separated by 4%–15% Criterion™ TGX™ Precast sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (Bio-Rad Laboratories, Inc.) and then transferred onto a polyvinylidene difluoride (PVDF) membrane. Membranes were blocked with 5% non-fat milk in PBST, incubated with a primary antibody, and subsequently a secondary horseradish peroxidase-conjugated antibody. After applying ECL plus western blot substrate (GE Healthcare, Cleveland, OH, United States), the blot images were captured with iBright CL1000 Imaging System (ThermoFisher Scientific). The Foxo1 antibody was purchased from Cell Signaling Technology, Inc. United States. β–actin antibody (Santa Cruz Biotechnology) was used as a reference protein.

Renal histopathology and immunofluorescence analysis were performed as we previously reported (Dai et al., 2013; Edwards et al., 2017; Edwards et al., 2020). Frozen section slides and hematoxylin and eosin (H&E) stained slides were prepared by the histopathology laboratory at VMCVM, Virginia Tech. The H&E renal sections were assessed by a board-certified pathologist in a blinded fashion. A grade of 0–4 (0 = perfect, no change; 1 = minimal; 2 = moderate, 3 = marked; 4 = severe) was given to reflect the glomerular, tubular, interstitium, and vessel inflammation and lesions, respectively.

The OCT frozen kidneys were cut to 5 μM sections in the histopathology laboratory at VMCVM. The frozen slide sections were stored at −80°C for immunofluorescent staining. Briefly, the frozen slides were air-dried, fixed with cold acetone for 10 min at room temperature, followed by blocking with PBS containing 1% BSA (Fisher) and anti-mouse CD16/32 for 20–30 min. The slides were then incubated with FITC-conjugated IgG (eBioscience, San Diego, CA, United States) and PE-conjugated Complement Component C3 (Cedarlane, Burlington, NC, United States) for 1–2 h at room temperature in a dark chamber. The stained slides were mounted with coverslips using Prolong Gold anti-fade reagent with DAPI (Invitrogen, Grand Island, NY, United States). The stained kidney sections were assessed for IgG and C3 deposition under a fluorescent microscope. Fiji/ImageJ image processing program (Schindelin et al., 2015) was used to measure the fluorescent intensity of the selected area as we previously described (Edwards et al., 2017; Edwards et al., 2020).

The Foxo1 specific Dicer substrate siRNA (Foxo1 siRNA) and negative control DsiRNA (NC) were purchased from Integrated DNA Technologies (Coralville, IA, United States) and reconstituted with distilled water to a stock concentration at 10 μM. The DsiRNAs were transfected into splenic lymphocytes with Lipofectamine RNAiMAX transfection reagent (ThermoFisher Scientific) per the manufacturer’s protocol as we recently reported (Dai et al., 2020). Forty-eight hours after transfection, the transfected cells were stimulated with anti-mouse CD3ε (plate coated at 5 μg/ml) plus 2 μg/ml soluble anti-mouse CD28 (Bio X cell) for 24 h h, or 50 ng/ml phorbol 12-myristate 13-acetate (PMA) plus 1 μg/ml ionomycin for 6 h. The cell culture supernatants were collected for ELISA assay.

All the values in the graphs were given as means ± SD. Unpaired student t-test and one-way ANOVA with Tukey-Kramer all pair’s comparisons were performed to determine the statistical significance for two or multiple group comparisons, respectively. “*,” “**,” and “***” indicate p < 0.05, p < 0.01, and p < 0.001, respectively. The graphic presentations of data were performed in GraphPad Prism software (v.8.10 for Windows).

We have previously reported a significant upregulation of miR-183C miRNAs in the splenocytes of different murine lupus models (Dai et al., 2010). To determine whether miR-183C miRNAs play a role in regulating lupus-related inflammatory cytokines such as IFNγ and IL-6, we inhibited miR-182 alone or miR-183C miRNAs with specific antagomirs in vitro in splenocytes (Figure 1A). The antagomir-treated cells from MRL (control) and MRL/lpr (lupus-prone) were stimulated with two different stimuli, LPS and Con A, which activated immune cells to elicit cytokine production via distinct signaling pathways. As indicated, the inhibition of miR-182 alone or the miR-183C whole cluster significantly suppressed LPS-induced IFNγ and IL-6 production in both MRL and MRL/lpr splenocytes (Figure 1B). In Con A-stimulated cells, miR-182 and miR-183C inhibition significantly reduced both IFNγ and IL-6 in MRL splenocytes, while in MRL/lpr splenocytes only IL-6 was significantly reduced (Figure 1C). The antagomir treatment did not affect the cell viability (Supplementary Figure S2A). We also noticed that silencing miR-182, but not miR-96 and miR-183 alone, had a similar suppression on IFNγ and IL-6 production as silencing the three miR-183C miRNAs (Supplementary Figure S2B). This suggests that of the three miR-183C miRNAs, miR-182 likely plays a major role in regulating inflammatory cytokine production.

To further investigate the role of upregulated miR-183C cluster miRNAs in vivo in autoinflammation-prone mice, we generated miR-183C^−/−B6/lpr mice with conditional deletion of miR-183C miRNA cluster in both T and B lymphocytes (Figure 2A). As previously reported (de Boer et al., 2003; Siegemund et al., 2015), the hCD2-iCre deleted the floxed-target gene in both T and B lymphocytes. Nevertheless, the deletion efficiency in CD4⁺ T (about 99% reduction) is higher than that in CD19⁺ B (about 90% reduction) cells (Figure 2A). After the mice reached 20 weeks of age, blood samples were collected at 4 weeks intervals to monitor the serum level of anti-double strand DNA (anti-dsDNA) autoantibody in miR-183C^−/−B6/lpr and control miR183C^fl/flB6/lpr mice. The mice were euthanized at 29–32 weeks of age (endpoint) for the experimental analysis. Conditional deletion of miR-183C had no obvious effect on the body weight, spleen weight, and absolute splenocyte counts (Figures 2B–E). Impressively, we found that miR-183C deletion significantly reduced serum anti-dsDNA autoantibody level in miR-183C^−/−B6/lpr mice compared to age-matched miR183C^fl/flB6/lpr control mice (Figure 2F). There was a trend of reduced total IgG in miR-183C^−/−B6/lpr mice (p = 0.054, Figure 2G). There was no difference in the serum IgM levels (Figure 2H).

The in vitro antagomir inhibition data indicated that inhibition miR-182 alone and inhibition of the whole miR-183C had a similar suppressive effect on activation induced inflammatory cytokines production (Figure 1). We, therefore, derived miR-182^−/−B6/lpr to determine whether deletion of miR-182 alone in vivo has a similar effect as the deletion of whole cluster in vivo in B6/lpr mice. As expected, miR-182 was specifically deleted in the splenocytes of miR-182^−/−B6/lpr mice (Figure 3A). There was no change in the expression of miR-96 and miR-183 in the splenocytes of miR-182^−/−B6/lpr mice when compared to control miR182^fl/flB6/lpr mice (Figure 3A). Conditional deletion of miR-182 alone did not affect the body weight (Figure 3B). There was a trend of decreased spleen weight (p = 0.076) and spleen/body weight ratio (p = 0.082) in miR-182^−/−B6/lpr mice when compared to controls (Figures 3C,D). However, there was no obvious change in the absolute splenocytes count in miR-182^−/−B6/lpr compared to control miR-182^fl/flB6/lpr mice (Figure 3E). We only observed a reduction of serum anti-dsDNA levels in miR-182^−/−B6/lpr mice at 28 weeks of age compared to age-matched controls (Figure 3F). Deletion of miR-182 significantly decreased serum IgM levels and demonstrated a trend of reduced total IgG (p = 0.062) in B6/lpr mice (Figures 3G,H).

The effect of miR-183C deletion on the renal function of B6/lpr was determined by examining proteinuria and blood urea nitrogen (BUN) levels. There was no difference in the proteinuria or BUN levels between miR-183C^−/−B6/lpr and control miR-183C^fl/flB6/lpr mice (Figures 4A,B). The renal histopathology analysis also showed no obvious difference in the kidney inflammation and damages in the miR-183C^−/−B6/lpr mice compared to control miR-183C^fl/flB6/lpr mice (Table 2). However, we observed a significant reduction of IgG and complement component C3 immunocomplex deposition in the kidneys of miR-183C^−/−B6/lpr mice compared to age-matched control mice (Figures 4C,D). We also examined the effect of miR-182 deletion on renal pathology and function in B6/lpr mice. Similar to the data with miR-183C deletion, deletion of miR-182 had no effect on renal histopathology, proteinuria levels, serum BUN levels and C3 immunocomplex deposition in the kidneys in B6/lpr mice (Figures 4E,F,H), but significantly reduced IgG immune complex deposition in the kidneys of B6/lpr mice (Figure 4G).

Previous studies have reported that deletion of miR-182 in vivo had no obvious effects on adaptive immune cell development in normal B6 mice (Pucella et al., 2015; Ichiyama and Dong, 2019). Flow cytometric evaluation was performed to determine whether miR-182 and miR-183C deletion affected immune cell composition in B6/lpr mice (Supplementary Figure S3). Consistently, we found that in the autoinflammatory-prone B6/lpr mice, conditional deletion of either miR-182 alone or miR-183C cluster in lymphocytes also did not change the adaptive immune cell development and lymphocyte subset composition in the spleens (Figures 5A,B). The development of CD4⁻CD8⁻CD3⁺B220⁺ double-negative T cells (DN T) is a specific characteristic cell type in lpr lupus mice (Morse et al., 1985; Giese et al., 1993; Hamad et al., 2003). There were no obvious differences in the percentage of DN T cells in the splenocytes between the knockout and control B6/lpr mice (Figures 5A,B). While the previous study has shown that miR-182 regulates Treg cell development during murine EAE development (Wan et al., 2016), we did not observe any notable change of Tregs percentage in the splenocytes in either the miR-183C^−/−B6/lpr or the miR-182^−/−B6/lpr mice compared to their respective controls (Figures 5C,G). Furthermore, we found that conditional deletion of miR-183C or miR-182 did not affect the differentiation of germinal center B cells (GCB), follicular helper T cells (T_FH), or plasma cells in the spleens. The percentages of GCB, T_FH, and plasma cells in the splenocytes of the miR-183C^−/−B6/lpr (Figures 5D–F) and the miR-182^−/−B6/lpr (Figures 5H–J) mice were similar to their respective controls.

We showed that in vitro inhibition of either miR-182 alone or miR-183C significantly suppressed IFNγ and IL-6 production in activated splenocytes (Figure 1). We then checked IFNγ and IL-6 cytokine production in in vitro activated splenocytes from miR-183C^−/−B6/lpr and control miR-183C^fl/flB6/lpr mice. We found that in response to either anti-CD3 plus anti-CD28 or PMA plus ionomycin stimulation, there was a significantly lower IFNγ production in splenocytes from miR-183C^−/−B6/lpr and miR-182^−/−B6/lpr mice compared to splenocytes from their respective control mice (Figures 6A,B). In vivo deletion of miR-183C or miR-182 alone had no effect on IL-6 production in splenocytes in response to anti-CD3 plus anti-CD28 stimulation (Figures 6B,E). There was an increase of IL-6 in PMA plus ionomycin-activated splenocytes from miR-183C^−/−B6/lpr mice, but not from miR-182^−/−B6/lpr compared to their respective controls (Figures 6B,E). A recent study revealed that miR-183C was highly induced by IL-6-STAT3 signaling to promote the differentiation and function of pathogenic Th17 cells (Ichiyama et al., 2016). We found that deletion of miR-183C or miR-182 in vivo did not affect IL-17 production in B6/lpr splenocytes in response to either non-pathogenic (IL-6 plus TGFβ1) or pathogenic (IL-6, IL-23 plus IL-1β) Th17 stimuli (Figures 6C,F). In vitro inhibition of either miR-183C or miR-182 alone suppressed LPS-induced IFNγ and IL-6 production in splenocytes from MRL and MRL/lpr mice (Figure 1). However, we found that deletion of miR-183C in vivo did not affect LPS-induced IFNγ and IL-6 production, and the deletion of miR-182 only inhibited LPS-induced IFNγ, but not IL-6 in splenocytes from B6/lpr mice (data not shown).

Previous studies have shown that miR-183C cluster miRNAs play an immune- and autoimmune-regulatory role through targeting Foxo1 directly (Stittrich et al., 2010; Ichiyama et al., 2016; Wang X. et al., 2018). We found an inverse relationship between the miR-183C miRNAs expression and Foxo1 protein expression in the splenocytes of MRL and MRL/lpr mice (Figure 7A). Consistent with the marked increase of miR-183C miRNAs, there was a significant decrease in the Foxo1 protein expression in MRL/lpr splenocytes. Knocking down miR-182 in vitro alone with antagomir-182 or miR-183C cluster with the mixes of antagomir-182, -96, -183 in MRL/lpr splenocytes increased Foxo1 protein expression (Figure 7B). Further, Foxo1 protein expression was significantly increased in CD4⁺ T cells from miR-182^−/−B6/lpr and miR-183^−/−B6/lpr compared to the CD4⁺ T cells from control mice (Figure 7C). Together, these data demonstrated that Foxo1 is the target of miR-183C miRNAs in splenocytes and T cells of MRL/lpr and B6/lpr mice. Deletion of miR-182 or miR-183C increased Foxo1 expression (Figure 7C) and reduced IFNγ in B6/lpr splenocytes in response to anti-CD3 plus anti-CD28 or PMA plus ionomycin stimulation (Figures 6A,D). To determine whether miR-183C regulates the production of cytokines in splenocytes through targeting Foxo1, we inhibited Foxo1 expression with Foxo1 siRNA in vitro in splenocytes from miR-183C^−/−B6/lpr and miR-182^−/−B6/lpr. We confirmed the significant reduction of Foxo1 mRNA expression level in the Foxo1 siRNA transfected cells (Figure 7D). In Foxo1 siRNA transfected cells, there was a significantly increased IFNγ production in response to anti-CD3 plus anti-CD28 or PMA plus ionomycin stimulation when compared to control siRNA transfected cells (Figures 7E,G). Suppression of Foxo1 in vitro has minimal effect on IL-6 production in activated splenocytes from miR-183C^−/−B6/lpr and miR-182^−/−B6/lpr mice (Figures 7F,H). This is consistent with the finding that deletion of miR-182 or miR-183C in vivo has minimal effect on IL-6 in in vitro activated splenocytes (Figures 6B,E). Together, our data suggested that miR-183C acts through Foxo1 to regulate the production of inflammatory cytokine IFNγ, a key pathogenic cytokine in the lpr lupus model.