Targeting of EIF4EBP1 by miR‐99a‐3p affects the functions of B lymphocytes via autophagy and aggravates SLE disease progression

Abstract Excessive activation of immune cells plays a key role in the pathogenesis of systemic lupus erythematosus (SLE). The regulation of immune cells by miRNAs is a research hotspot. In this study, second‐generation high‐throughput sequencing revealed a reduction in miR‐99a‐3p expression in patients with SLE; however, the specific mechanism underlying this phenomenon remains unclear. After transfection with an miR‐99a‐3p agomir, the proliferation of Ball‐1 cells decreased and the levels of their apoptosis increased. The opposite effects were observed in cells transfected with the miR‐99a‐3p antagomir. Luciferase reporter assay indicated that miR‐99a‐3p directly targeted EIF4EBP1. Rescue experiments confirmed the proposed interaction between miR‐99a‐3p and EIF4EBP1. In vitro, in vivo and clinical investigations further confirmed that the miR‐99a‐3p agomir reduced the expression of EIF4EBP1, LC3B and LAMP‐2A. In the in vivo experiments, serum levels of anti‐nuclear antibodies, double‐stranded DNA, IgE, IgM, IL‐6, IL‐10 and B lymphocyte stimulator were higher in mice from the antagomir group than those in mice from the MRL/lpr group. Furthermore, the protein and mRNA levels of EIF4EBP1, LC3B and LAMP‐2A, the intensity of immunohistochemical staining of EIF4EBP1, LC3B and LAMP‐2A, the urinary protein levels, and the C3 immunofluorescence deposition increased in mice from the antagomir group. The upregulation of miR‐99a‐3p expression protected B cells from EIF4EBP1‐mediated autophagy, whilst the downregulation of miR‐99a‐3p expression induced autophagy via the EIF4EBP1‐mediated regulation of the autophagy signalling pathway in B cells isolated from individuals with SLE. Based on these results, miR‐99a‐3p and EIF4EBP1 may be considered potential targets for SLE treatment.


| INTRODUC TI ON
Systemic lupus erythematosus (SLE) is a type of autoimmunemediated diffuse connective tissue disease that involves multiple systems throughout the body and is characterized by pathological inflammation. 1 Genetic factors influence the clinical phenotype and progression of SLE, whilst environmental factors, such as ultraviolet rays, promote the onset of SLE, 2 which play an important role in stimulating SLE. 3 The age of the onset of SLE in individuals residing in high-altitude areas is lower, and the disease course is shorter. 4 Tibetans comprise the main ethnic group in the Yunnan Diqing Tibetan Autonomous Prefecture; they live in a high-altitude, lowoxygen, dry, very cold environments with strong ultraviolet radiation. They are relatively isolated and typically do not contribute to genetic exchange with the rest of the world. Thus, the epidemiological and transcriptomic characteristics of Tibetan patients with SLE deserve in-depth and systematic investigations.
An increasing number of noncoding single-stranded small RNAs called microRNAs (miRNAs) have been found to play an important role in the pathogenesis of SLE. 5 Previous studies performed by our research group indicated that UVB may mediate the pathogenesis of SLE by decreasing miR-125b-5p expression in patients with SLE and increasing the expression of the target gene UVRAG and the degree of cellular autophagy. 6 We collected venous blood samples from 10 Tibetan patients with SLE and 10 healthy Tibetans. miRNA expression profiles were determined by second-generation high-throughput sequencing after RNA extraction ( Figure 1A). We used RT-qPCR to verify the sequencing results, and miR-99a-3p expression in Tibetan SLE patients was found to be significantly reduced ( Figure 1B). The sequencing results were verified using RT-qPCR, and miR-99a-3p expression was significantly decreased in Tibetan patients with SLE ( Figure 1B). miR-99a-3p is transcribed from the long arm of chromosome 21.
miR-99 is expressed at low levels in a variety of human malignancies.
It participates in the progression of urinary tumours, 7 squamous cell carcinoma, 8 liver cancer 9 and ovarian cancer 10 and has a certain significance in the early diagnosis and determination of tumour stages. 11 Pradhan and Tomankova were the first to demonstrate a reduction in miR-99 expression in patients with SLE. 12 Jin et al. 13 observed the downregulation of miR-99a expression in South Korean patients with SLE using miRNA PCR chip detection. Frangou et al. 14 used cDNA microarrays to compare gene expression in the effector cells and target tissues of patients with SLE and control subjects and found that miR-99a expression was decreased in PBMCs isolated from patients with SLE and was related to the regulation of the type I IFN pathway. However, thus far, no report has documented miR-99a-3p expression in Chinese patients with SLE. Therefore, we explored the role of miR-99a-3p in the pathogenesis of SLE. This study attempts to elucidate the complex mechanisms underlying the pathogenesis of SLE and identifies new targets for SLE treatment.

| Patients and samples
From January to December 2020, we included ten SLE patients of Han ethnicity who were treated at our hospital's outpatient clinic.
We contacted the People's Hospital of Diqing Tibetan Autonomous Both Tibetan and Han patients were diagnosed in accordance with the American Rheumatism Association 1997 classification of SLE and had not consumed chloroquine or hydroxychloroquine in the past three months. With the approval of the hospital ethics committee (PAR-PJ-2020-135), we collected 20 ml of peripheral venous blood from the patients and healthy controls. No significant differences in age or sex were observed between the two groups (p > 0.05, see Table S1). Negative Selection Beads (Thermo, Germany) was added, followed by mixing and further incubation. The test tube was inserted into the MagniSort TM magnetic pole (Thermo, Germany), and the precipitate was centrifuged to collect B lymphocytes (identification of B cells and culture, see Figure S1).

| Isolation of PBMCs and separation and culture of primary B cells
Frozen Ball-1, Jurkat, THP-1 and K562 cells (purchased from the Cell Bank of the Chinese Academy of Sciences) were collected from a liquid nitrogen tank, thawed in a water bath at 37°C and centrifuged; the supernatants were discarded. Next, the cells were resuspended in RPMI-1640 (PM150110)+10% FBS (164210-500)+1% P/S (PB180120) and cultured.

RT-qPCR
TRIzol™ Reagent (15596026, Invitrogen, USA) was used to extract the total RNA. A Transcriptor First Strand cDNA Synthesis Kit (Roche, Switzerland) was used for mRNA, and an All-in-One™ miRNA First-Strand cDNA Synthesis for miRNA Kit 2.0 (GeneCopoeia, USA) was used for miRNA. A total of 1 µg of total RNA was used to synthesize the first-strand cDNA templates for the mRNAs/miRNAs according to the manufacturers' instructions. The reaction conditions for mRNA reverse transcription were as follows: 25°C for 5 min, 42°C for 45 min and 85°C for 5 min. The reaction conditions for miRNA reverse transcription were as follows: 37°C for 60 min and 85°C for 5 min. Reverse transcription was performed using a common PCR machine (Applied Biosystems 2720 Thermal Cycler, Thermo, Germany). BlazeTaq™ SYBR ® Green qPCR Mix 2.0 (GeneCopoeia, USA) was used for mRNA, and an All-in-One miRNA qRT-PCR Detection Kit 2.0 (GeneCopoeia, USA) was used for miRNA, using cDNAs as the templates. GAPDH and U6 served as the internal references. The expression levels of the target genes were detected using the SYBR green method and a real-time fluorescence quantitative PCR instrument (CFX96, Bio-Rad, USA). The PCR amplification conditions were set as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 10 sec, 60°C for 20 s and 72°C for 30 sec.
The fluorescence was recorded, and the CT values were determined; the 2 −△△CT method was used to calculate the relative expression of the target genes. The relevant primers were synthesized by Beijing Kinco Xinye Biotechnology Co., Ltd.; the sequences of these primers are shown in Table S2.

| CCK8 analysis of cell proliferation
Newly cultured Ball-1, Jurkat, THP-1 and K562 cell suspensions (100 µl/well) were seeded into 96-well plates, and the four cell lines were transfected with the miR-99a-3p agomir and antagomir. A total of 10 µl of CCK8 reagent (Life iLab Biotech Co., Shanghai, China) was added to each well on days 1, 2, 3 and 4, followed by incubation for 2 h; the absorbance of the samples was measured at 450 nm using a microplate reader (BioTek, USA).  NxT Flow Cytometer (Thermo, Germany).
The wild-type/mutant 3′-UTR of the target genes was inserted into a luciferase pmirGLO vector by the Wuhan GeneCreate Company.
The miRNA and plasmid (2 µg) were incubated together for 5 min,

| Western blotting
For this experiment, 250 µl of RIPA lysis buffer (containing protease inhibitors, Thermo, Germany) was added to each group of cells, and a BCA protein quantification kit (P0010, Beyotime, China) was used to determine the protein concentration. Total protein (30 µg) was subjected to electrophoresis using SDS-PAGE gels and transferred onto membranes, which were then blocked. The membranes were then incubated with primary antibodies against EIF4EBP1 (1:1000; GTX133182; GeneTex, USA), LAMP-2A (1:1000; EPR4207(2); Abcam, USA), LC3B (1:1000; ab192890; Abcam, USA) and β-Actin (LMAI Bio; Shanghai, China) overnight at 4°C with agitation. The membranes were then warmed to room temperature for 1 h the next day. Next, they were incubated with the secondary antibody (peroxidase-labelled goat anti-rabbit IgG; 1:5000; Sigma, USA) for 30 min. The membrane was exposed to an ECL colour-developing solution (Thermo, Germany), and images were captured with an ECL instrument (Monad, Suzhou, China). Image-Pro Plus 6.0 software was used to analyse the optical density of the bands. The optical density ratio of the target protein to that of the endogenous protein β-actin represents the relative content of the target protein and was calculated to compare the differences in protein expression.

| Joint intervention with the siRNA and antagomir in Ball-1 cells for rescue experiments
siRNA targeting EIF4EBP1 was synthesized by Guangzhou Ruibo Biological Company, and Ball-1 cells were transfected with siEIF-4EBP1-1, siEIF4EBP1-2 and siEIF4EBP1-3. After 48 h, EIF4EBP1 expression in the siEIF4EBP1-1 group was found to be lower than that in the siNC group (for the sequence and screening, see Table S3 and Figure S3). Ball-1 cells were counted and plated, and siRNA EIF4EBP1 and NC were added. Subsequently, the Lipo2000 transfection reagent was added. After 24 h, half of the sample was separated from the siRNA tube, and miR-99a-3p antagomir was added for further culture.
After incubation at 37°C for 48 hours, proliferation, cell cycle progression, apoptosis and Western blotting analyses were performed.

| | Coomassie brilliant blue staining method for the measurement of total protein levels in the urine
An appropriate amount of urine was collected and diluted to a standard volume before the measurement of the total protein contents.
Equal dilutions of urine in PBS were added to 5 ml of Coomassie brilliant blue solution (Xinfan Biological Technology Co., Shanghai, China). The colour changed from red to blue, and absorbance of the samples was measured at 595 nm.

| ELISA for the detection of ANA, dsDNA, IgE, IgM, IL-6, IL-10 and BLyS
A total of 100 µl of an HRP-labelled antibody was added to the wells of the plate provided with the ELISA kit (JL12477-96T, Jiang Lai Bio, China), followed by incubation at 37°C for 60 min. Next, 50 μl of substrates A and B was added to each well, followed by incubation at 37°C in the dark and the addition of a stop solution. The OD value of each well was measured at a wavelength of 450 nm, and the sample concentration was calculated from the absorbance value based on a standard curve.

| HE staining
Longitudinally sectioned kidney tissues from each group were fixed with paraformaldehyde for 24 h and then dehydrated using an as-

| Statistical methods
The data are presented as the means±SDs and were analysed using ANOVA and an LSD t test using the SPSS 23.0 software. A twotailed P value<0.05 was considered significant. The count data were analysed using a chi-square test. GraphPad Prism 6.0 software was used for the statistical analyses of histograms.

| Verification of differential miR-99a-3p expression in patients with SLE
Venous blood from the Tibetan subjects was collected, and the RNA was extracted and subjected to second-generation high-throughput sequencing. RT-qPCR was used to verify the results. The results showed a significant reduction in miR-99a-3p expression in Tibetan patients with SLE ( Figure 1A, 1B).
RT-qPCR was used to detect miR-99a-3p expression in PBMCs

| Confirmation of the target genes
The target genes of miR-99a-3p were predicted using TarBase (http:// mirta rbase.mbc.nctu. edu.tw/php/index.php), miRDB (http://www. Western blotting further verified the expression of target genes in patients with SLE and healthy controls. The EIF4EBP1 and NCAPG levels were significantly increased in patients with SLE ( Figure 3D).

RT-qPCR was used to detect LC3B and LAMP-2A expression in
PBMCs isolated from patients with SLE and healthy controls. Higher LC3B and LAMP-2A expression levels were observed in patients with SLE than those in the healthy control group ( Figure 4A). Western blotting was used to analyse the levels of the autophagy pathway marker proteins LC3B and LAMP-2A in PBMCs isolated from patients with SLE and healthy controls. The ratio of the levels of these two mRNAs was higher in patients with SLE (p = 0.0399, Figure 4B).

Ball-1 cells were transfected, and the expression levels of LC3B
and LAMP-2A were detected using RT-qPCR 48 h after transfection. Transfection with the miR-99a-3p agomir decreased the levels of both LC3B and LAMP-2A in Ball-1 cells, but transfection with the miR-99a-3p antagomir increased the levels of these genes ( Figure 4C).  Figure 4E).

The intensity of IF staining for LC3B expression in Ball-1 and B
cells was weaker in the miR-99a-3p agomir group than in the NC group but stronger in the miR-99a-3p antagomir group than in the NC group, as observed by fluorescence microscopy ( Figure 4F).

| Changes in the functions of Ball-1 cells after a rescue experiment
Western blotting was used to detect the levels of EIF4EBP1, LC3B and LAMP-2A in Ball-1 cells transfected with siNC, siEIF4EBP1 or siEIF4EBP1+antagomir. Lower EIF4EBP1, LC3B and LAMP-2A protein levels were detected in Ball-1 cells transfected with siEIF4EBP1 than those in the siNC group. The EIF4EBP1, LC3B and LAMP-2A proteins were expressed at significantly higher levels in Ball-1 cells transfected with siEIF4EBP1+antagomir than in those transfected with siEIF4EBP1 ( Figure 5A).
After Ball-1 cells were transfected with siEIF4EBP1, the apoptotic rate, as determined by flow cytometry, was significantly higher than that in the siNC group, and the apoptosis rate was significantly lower in cells transfected with siEIF4EBP1+antagomir than that in cells transfected with siEIF4EBP1 ( Figure 5B).
The proliferation of transfected Ball-1 cells in the siEIF-4EBP1 group was lower when observed on days 1, 2, 3 and 4, than that of cells in the siNC group. The proliferation of cells in the siEIF-4EBP1+antagomir group was higher when observed on days 2, 3 and 4, than that of cells in the siEIF4EBP1 group ( Figure 5C).
After transfection with siEIF4EBP1, higher numbers of cells were observed in the G0/G1 and G2/M phases than those in the siNC group, and the number of cells in the S phase was lower than that in the siNC group. After transfection with siEIF4EBP1+antagomir, the number of cells in the G0/G1 phase decreased, and the number of cells in the G2/M and S phases increased compared to that in the siE-IF4EBP1 group ( Figure 5D). Fluorescence microscopy showed that the LC3B IF staining intensity in the siEIF4EBP1 group was lower than that in the siNC group, and the LC3B IF staining intensity was higher in the siEIF4EBP1+antagomir group than that in the siEIF-4EBP1 group ( Figure 5E).

| Disease progression in the mouse MRL/lpr model after experimental intervention with miR-99a-3p under in vivo conditions
The bodyweight of the mice in the four groups increased as the number of feeding weeks increased. One mouse in the agomir group died at 12 w, and the bodyweight of mice in the MRL/Lpr group was higher than that of mice in the C57 group at 10 w, 12 w and 13 w. However, the miR-99a-3p intervention did not significantly alter the bodyweight ( Figure 6A). At 13 w, the hair around the nose and eyes of mice in the MRL/lpr group was shed, and the animals exhibited fewer and slower behavioural responses. The mice from the antagomir group presented alopecia areata around the nose, eyes and forehead, and their behavioural responses were slower.
No difference in the urine protein levels was observed between the four groups at 8 weeks (p > 0.05). However, at 10 w, 12 w and 13 w, the urine protein levels increased significantly in mice from all the groups, except the C57 group. The urine protein levels of mice in the MRL/Lpr and antagomir groups were higher than those of mice in the C57 group, and the urine protein levels in the mice from the agomir group were lower than those in the mice from the C57 group ( Figure 6B). C57 group (p = 0.0184), suggesting that the mice in the MRL/lpr group experienced some glomerular oedema. The glomerular area in the agomir group was smaller than that in the MRL/lpr group (p = 0.2098), and the glomerular area in the antagomir group was greater than that in the MRL/lpr group (p = 0.5888), but these two differences were not significant ( Figure 6C, 6D).
IF staining for C3 deposition was assessed in MRL/lpr mice and C57 mice after the intervention, and C3 deposition was higher in the MRL/lpr group than in the C57 group (p < 0.0001). C3 deposition in the agomir group was lower than that in the MRL/lpr group (p = 0.0002). C3 deposition in the antagomir group was higher than that in the MRL/lpr group (p = 0.0008, Figure 6E, 6F).  MRL/lpr group than in those from the C57 group (p = 0.0293). miR-99a-3p expression in the mice from the agomir group was higher than that in mice from the MRL/lpr group (p = 0.0013); the miR-99a-3p level in mice from the antagomir group was lower than that in the mice from the MRL/lpr group (p = 0.0272). EIF4EBP1, LC3B and LAMP-2A were expressed at higher levels in the mice from the MRL/lpr group than in those from the C57 group. The EIF4EBP1, LC3B and LAMP-2A mRNA expression levels in the mice from the agomir group were lower than that in the mice from the MRL/lpr group. EIF4EBP1, LC3B and LAMP-2A mRNAs were expressed at significantly higher levels in the mice from the antagomir group than in those from the MRL/lpr group ( Figure 7A). reported that miR-155, miR-499a and miR-142 are involved in the pathogenesis and determining the clinical phenotype of SLE. Tao et al. 19 reported that miR-152-3p promotes the Toll-like receptor (TLR)-mediated CD4+ T cell inflammatory response by regulating the DNMT1/MyD88 signalling pathway, which may be considered as a new target for SLE treatment.
In the present study, miR-99a-3p expression was found to be sig- EIF4EBP1 is an inhibitor of translation initiation, and its activity is regulated by preventing the assembly of eIF4E into the eIF4F complex. The role of EIF4EBP1 in SLE pathogenesis has not yet been reported.
Autophagy is a process whereby a cell engulfs its own cytoplasmic proteins or organelles in vesicles that then fuse with lysosomes.
In patients with SLE, autophagy plays a role in the identification and removal of pathogens, antigen processing and antigen presentation.
At the same time, autophagy maintains the stability of the intracellular environment by regulating the levels of cytokines and controlling intracellular energy and metabolism. More autophagosomes were detected in T and B lymphocytes isolated from patients with SLE, and the level of autophagy was significantly increased. If autophagy is abnormal, the presentation and processing of antigens are hindered.
Although the role of autophagy in the pathogenesis of SLE has not yet been completely elucidated, autophagy is closely related to the progression of SLE.
Many studies have confirmed that miRNAs regulate autophagy. 21,22 Abnormal autophagy leads to the accumulation of apoptotic cells and induces the production of autoantibodies. EIF4EBP1 regulates the activity of mTORC1, thereby inducing autophagy, 23 and LC3 is a key protein involved in autophagy. 24  Transfection with miR-99a-3p antagomir exerted the opposite effect, suggesting that miR-99a-3p negatively regulates autophagy.
The trend observed in the changes in EIF4EBP1 expression was consistent with the trend in the changes in the autophagy level, suggesting that EIF4EBP1 exerts a positive regulatory effect on autophagy.
In the present study, a rescue experiment was conducted to confirm the mode of interaction between miR-99a-3p and EIF4EBP1.
Notably, miR-99a-3p affected cell proliferation and apoptosis by modulating the expression of target genes.
BLyS is a newly identified member of the tumour necrosis factor family and is involved in the regulation of B cell proliferation and antibody production. Transgenic mice showing the overexpression of BLyS demonstrate an increased number of B cells, increased serum ANA and dsDNA levels, and immunoglobulin deposition in the kidney. 26 Benlysta was the first inhibitor designed to target BLyS; it binds to soluble BLyS with high affinity and inhibits its activity to achieve disease control. 27 In this study, an in vivo experimental model was constructed using MRL/Lpr mice by injecting a miR-99a-3p agomir or antagomir or a NC into the tail vein of the mice. The levels of ANA, dsDNA, IgE, IgM, IL-6, IL-10 and BLyS in mice from the antagomir group increased significantly, and the SLE disease activity was stronger. In the antagomir group, urinary protein levels and C3 IF deposition in the kidneys were increased, and kidney damage was more serious, which is consistent with results from a previous report. 28 These results indicate that miR-99a-3p inhibition exacerbated SLE progression. However, the IgM levels were higher in mice from the antagomir group than in the mice from the MRL/lpr group. This observation contradicts those of a previous study. 29 This finding may be related to the euthanization of 13-week-old mice in this study; the animals may still have been in the early stages of negative feedback.

| CON CLUS IONS
This study is the first to report that miR-99a-3p, which targets EIF4EBP1, participates in the autophagy signalling pathway and affects the function of B cells, thereby aggravating SLE progression.
This study provides a deeper understanding of the molecular mechanisms underlying the activity of miRNA regulatory networks associated with SLE pathogenesis. The abnormal expression patterns of miR-99a-3p and its target gene EIF4EBP1 observed in the present study are expected to be considered potential therapeutic targets in patients with SLE.

| LI M ITATI O N S
This study had a few limitations. First, most of the experiments in this study were performed using Ball-1 and B cells, but some differences were observed between Ball-1 and B cells derived from healthy controls and B cells obtained from patients with SLE. The results obtained from the B cells used in this study were not ideal.
Second, the rescue experiment did not use gene knockout mice, and thus, verification of the results of the rescue experiment by means of relevant functional experiments is required. Third, due to various restrictions, Tibetan patients were not directly involved in further studies. The number of clinical cases was small, and correlation analysis between clinical indicators and laboratory indicators was not performed. Fourth, the relationship between miR-99a-3p and NCAPG and their roles in the pathogenesis of SLE were not studied.

ACK N OWLED G EM ENTS
We thank Yanlin Zi for help with English language polishing. We thank Zhu Qinghuan for her help with animal experiments.

CO N FLI C T O F I NTE R E S T
The authors have no proprietary interest in any aspect of the study.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data sets generated for this study are included in the article/Supplementary Material.