Brucella-Induced Downregulation of lncRNA Gm28309 Triggers Macrophages Inflammatory Response Through the miR-3068-5p/NF-κB Pathway

Objectives The underlying mechanism of the inflammatory response against Brucellosis caused by Brucella remains poorly understood. This study aimed to determine the role of long non-coding RNAs (lncRNAs) in regulating of inflammatory and anti-Brucella responses. Materials and methods Microarray analysis was performed to detect differentially expressed lncRNAs in THP-1 cells infected with an S2308 Brucella strain. The candidate lncRNAs were screened using bioinformatic analysis and siRNAs; bioinformatic prediction and luciferase reporter assay were also conducted, while inflammatory responses was assessed using RT‐qPCR, western blot, immunofluorescence, ELISA, HE, and immunohistochemistry. Results The lncRNA Gm28309 was identified to be involved in regulating inflammation induced by Brucella. Gm28309, localized in the cytoplasm, was down-expressed in RAW264.7 cells infected with S2308. Overexpression of Gm28309 or inhibition of miR-3068-5p repressed p65 phosphorylation and reduced NLRP3 inflammasome and IL-1β and IL-18 secretion. Mechanistically, Gm28309 acted as a ceRNA of miR-3068-5p to activate NF-κB pathway by targeting κB-Ras2, an inhibitor of NF-κB signaling. Moreover, the number of intracellular Brucella was higher when Gm28309 was overexpressed or when miR-3068-5p or p65 was inhibited. However, these effects were reversed by the miR-3068-5p mimic. Conclusions Our study demonstrates, for the first time, that LncRNAs are involved in regulating immune responses during Brucella infection, and Gm28309, an lncRNA, plays a crucial role in activating NF-κB/NLRP3 inflammasome signaling pathway.


INTRODUCTION
The Gram-negative facultative intracellular bacteria Brucella spp. (B. spp.) cause brucellosis, a systemic infectious zoonotic disease. In animals (1), brucellosis leads to abortion, infertility, and lameness, causing serious economic losses. While in humans, (2) B. spp. causes symptoms such as undulant fever, endocarditis, arthritis, and osteomyelitis. Brucellosis poses a a serious threat for both livestock and human since B. spp. can be transmitted from animals to humansthrough various infection sources, including meat, unpasteurized milk, animal byproducts from infected animals, which lead to economic loss (3) and are public health concerns (4).
It has been reported that the interactions between macrophage cells and Brucella alters the macrophage gene expression profile (5). Thus, the investigation of transcriptional and post-transcriptional changes in macrophages may help to further elucidate the anti-Brucella immune mechanisms. Although Brucella possesses immune subversion mechanisms (6) to evade recognition by toll-like receptors (TLRs) and the immune system activation (7), more than 90% of internalized Brucella are eliminated by the innate immune response (8), indicating that the non-specific inflammatory response against pathogens plays a critical role in controlling Brucella infection. Nucleotide-binding oligomerization domain-like receptors (NLRs) are innate cytosolic receptors that can recognize different pathogenassociated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs) (9). NLRP3 is the best-characterized inflammasome primarily induced by TLR activation, cytokine stimulation, and other signaling molecules (10). A recent study has demonstrated that Brucella ligands activate NLRP3 inflammasomes, leading to infection control (11). The activation of the NLRP3 inflammasome is regulated at the transcriptional level through the NF-kB pathway (12), which induces the synthesis of the proinflammatory interleukin pro-IL-1b and increases NLRP3 expression (13).
Non-coding RNAs (ncRNAs), including long non-coding RNAs (lncRNAs) and microRNAs (miRNAs), regulate various innate and adaptive immune processes (14). lncRNAs are defined as transcripts longer than 200 nucleotides lacking an obvious open reading frame (15). Recently, lncRNAs have been shown to regulate gene expression in various physiological and pathological immune processes (16), through a variety of mechanisms, such as behaving as RNA-binding proteins (RBPs), function as decoys, and microRNA sponges. For instance, the lncRNA, Neat1, stabilizes caspase-1 form and increases inflammasomes assembly by binding to pro-caspase-1 (17); NKILA, an NF-kB-interacting lncRNA, inhibits NF-kB activity to regulate the activation-induced cell death of T cell (18). The most well-studied ncRNAs are miRNAs, which during immune response, typically target mRNA transcripts for inflammasome complexes components to regulate inflammation (19). For example, the microRNA miR-223-3p directly targets two components of the NLRP3 inflammasome, namely, caspase-1, and caspase-8 (20), the miR-145a-5p activates the NLRP3 inflammasome by targeting CD137 and NFATc1 (21). Despite these discoveries, a comprehensive view of the role of ncRNAs in the regulation of host-Brucella interactions is currently lacking. Furthermore, the lncRNAs-mediated molecular mechanisms underlying the innate immune response against Brucella is largely unknown.

Bacterial Strains and Plasmids
Brucella abortus wild-type strain 2308 (S2308) (The Center of Chinese Disease Prevention and Control, Beijing, China) was cultured in tryptic soy broth (TSB) or tryptic soy ager (TSA) (Difco, MI, USA) at 37°C in 5% CO 2 incubator, the Brucella strain was manipulated in a biosafety level 3 laboratory. Escherichia coli DH5a (The Center of Chinese Disease Prevention and Control, Beijing, China) was cultured in Luria-Bertani medium, when appropriate, 100 mg/mL of ampicillin or kanamycin was added to the culture medium. pcDNA3.1 (Youbio,Wuhan, China) and pGL3 plasmid (Promega, Beijing, China), and other constructed plasmids were extracted using Endotoxin-free plasmid extraction kit (TianGen, Beijing, China) for cells transfection.

Cells Culture and Infection
RAW 264.7 and THP-1 cell lines were purchased from Cell Resource Center(Beijing, China) and incubated in the incubator under 37°C and 5% CO 2 conditions. RAW 264.7 cells were cultured in DMEM (Hyclone) supplemented with 10% fetal bovine serum (Gibco). Prior to infection, RAW 264.7 were seeded to proper culture plates at a density of 10 5 cells/mL in complete culture medium without penicillin and streptomycin. THP-1 cells were cultured in RPMI-1640 medium (Hyclone) containing 10% fetal bovine serum (Gibco); THP-1 monocytes were differentiated into macrophages with 200 nM PMA (Beyotime, Shanghai, China) for 24 h. All cells were infected with S2308 strain at a MOI of 50. Culture plates were centrifuged at 350 × g for 5 min at room temperature and incubated at 37°C for 60 min. After washing twice with PBS, the infected cells were incubated for an additional 45 min in the presence of 50 mg/mL of gentamicin to kill extracellular bacteria. Then, the cultures were placed in fresh DMEM containing 25 mg/ mL of gentamicin (defined as time zero) and incubated at 37°C. containing 1×10 6 CFU of S2308. Every week until 4 weeks postinoculation, mice (n = 5 per time point per group) were euthanized and their liver and spleen were removed aseptically and used for histopathological observation and immunohistochemistry experiments, the animal experiments were manipulated in a biosafety level 3 laboratory.

Microarray Analysis of lncRNA and mRNA Expression
Three biological replicates samples of THP-1 cells infected by S2308 at 4 and 24 h were lysed with Trizol reagent (Thermo Fisher Scientific) to extract the total RNA. Microarray hybridization was then conducted according to the standard Arraystar protocols with minor modifications using the Human LncRNA Microarray V3.0 (Arraystar, Rockville, MD, USA). The Agilent Array platform was used for the microarray analysis. Acquired array images were analyzed using the Agilent Feature Extraction software (version 11.0.1.1, Agilent Technologies). Quantile normalization and further data processing were performed using the Agilent GeneSpring GX v11.5 software package (Agilent Technologies). Differentially expressed lncRNAs and mRNAs were identified by performing a Volcano Plot filtering and a Fold Change filtering. Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Omnibus (GO) analysis were used to investigate the possible functions of these lncRNAs. The microarray experiment was performed by Capitalbio Technology, Beijing, China.

Cells Transfection
All plasmids and siRNAs (20 uM/uL) were transfected using the Advanced DNA RNA Transfection Reagent ™ (ZETA LIFE, USA) according to the manufacturer's protocol. Briefly, cells were planted on the cell culture plate one day in advance, the cell confluence degree was up to 60-80% at the time of transfection, plasmid or siRNA was directly mixed with transfection reagent according to 1:1 relationship, then mixed by blowing into a pipette for 10-15 times. Following incubation at room temperature for 10-15 min, the complex was added to the cell culture plates, mix gently, and incubated in the CO 2 incubator for 24 h. All the siRNAs were designed by Gene Pharma (Shanghai, China), transfection efficiency was measured using qRT-PCR. Details are listed in Supplementary Table 1.

Construction of Vectors
We amplified, using PCR, the full-length sequence of Gm23809 and cloned it into the pcDNA3.1 expression vector to obtain a stable Gm23809 overexpression plasmid [Gm23809 (+)]. The wild-type and mutant fragment of Gm28309 were subcloned into the pGL3 plasmid, and the resulting plasmids were referred to as Gm28309-WT and Gm28309-Mut. Likewise, the 3′ UTR sequence of kB-Ras2 mRNA containing either the predicted wild type miR-3068-5p binding site or its mutated form was cloned into the pGL3 plasmid (kB-Ras2-WT and kB-Ras2-Mut, respectively).

Immunofluorescence Analysis
The cells were fixed with 4% paraformaldehyde for 30 min at room temperature, following permeabilization with 0.1% Triton X-100 and blocking with 1% BSA in PBS, cells were incubated with Goat polyclonal to NLRP3 (ab4207, 1:100) overnight at 4°C. Cells were then washed three times with PBS, and incubated with dylight488 for fluorescent labeling of Donkey anti-goat IgG-488 (Thermo Fisher Scientific) for an additional hour at 37°C. Finally, images were observed using confocal microscope (Carl Zeiss 510, German) and analyzed using LAS AF Lite (Leica).

Total, Cytoplasmic, and Nuclear RNA Extraction and Preparation
Cytoplasmic and nuclear RNA fractions were prepared according to published methods (23), with some modifications (24). First, the PBS used in this experiment was supplemented with DEPC, then, cells were collected, washed twice with ice-cold PBS and centrifuged at 1000g for 5 min. The supernatant was removed and the cell pellet was resuspended in 0.3% v/v NP-40-PBS by gently pipetting, and the pellets were placed in ice for 10 min prior to being centrifuged at 1000 × g for an additional 5 min. Then, the supernatant was collected separately as the cytoplasmic fraction, and the pellet was washed twice in ice-cold 0.1% NP-40-PBS, centrifuged at 1000 × g for 5 min, and kept as the nucleus fraction. For both fractions, RNA was extracted using Trizol reagent (Thermo Fisher Scientifific), following the manufacturer's protocol. The RNAs were analyzed using qRT-PCR with primer pairs for Gm23809, Actin as cytoplasm control and U1 as nucleus control. Total RNA and miRNA extraction methods are showed in Supplementary materials and methods, all the primer pairs used in study are listed in Supplementary Table 1.

Proteins Extraction and Western Blot
RAW264.7 cells were collected after 24 h of S2308 infection. Cells were lysed in PIPA buffer (Beyotime, Shanghai, China) supplemented with protease and phosphatase inhibitor cocktail (Pierce, Thermo Fisher Scientifific) for 30 min on ice. Lysates were centrifuged at 14,000 × g for 30 min and the supernatant was kept as protein extract. The protein content in the extracts was quantified using the BCA protein assay kit (Thermo Fisher Scientific). An equal amount about 15 mg of proteins were separated by 12% SDS-PAGE. The integrated density of protein bands was quantified by Image Lab 3.0 software (Bio-Rad, CA, USA) and normalized against b-actin, used as an internal control. Western blotting was performed using the following primary antibodies (Abcam) and dilutions: rabbit polyclonal anti-TGF-b (ab215715, 1:1,000), rabbit polyclonal anti-Pro-caspase-1 (ab179515, 1:1,000), rabbit polyclonal anti-p65 (ab32536, 1:1,000), and rabbit polyclonal anti-NLRP3 (ab214185, 1:1,000). The primary antibody rabbit polyclonal anti-phospho-p65 (3033T, 1:1,000) was acquired from Cell Signaling Technology, and rabbit polyclonal anti-kB-Ras2 (DF2508, 1:1,000) antibody was obtained from Affinity. The quantitive graph of relative protein expression levels of all proteins were analyzed using image J and showed in Supplementary Figures, and b-actin was used as the loading control.

Luciferase Reporter Assay
The wild-type and mutant fragment of Gm28309 were subcloned into the pGL3 plasmid, the resulting plasmids were called Gm28309-WT and Gm28309-Mut. Likewise, the 3′ UTR sequence of kB-Ras2 mRNA containing either the predicted wild type miR-3068-5p binding site or its mutated form was cloned into the pGL3 plasmid (kB-Ras2-WT and kB-Ras2-Mut, respectively). miR-3068-5p mimic, inhibitor or negative control were purchased from Gene Pharma (Shanghai, China).
Cells were co-transfected with the appropriate plasmid and oligomers using the Advanced DNA RNA Transfection Reagent ™ (ZETA LIFE, USA). Luciferase activity was measured 48 h of transfection using the dual-luciferase reporter assay system (Promega, Madison, WI, USA).

Interleukins Detection by ELISA
Culture supernatants were harvested from RAW 264.7 infected with S2308 for 24 h, at an MOI of 100:1 in triplicate wells. Moreover, serum samples were obtained from immunized mice 1, 2, 3, and 4 weeks post-immunization as described previously (26). The presence of IL-1b and IL-18 in samples was detected using an ELISA kit (molbio), according to the manufacturer's instructions.

Statistical Analysis
Statistical analyses were performed using GraphPad Prism 5 software, one-tailed or t-test was used to determine the differences, the data were expressed as mean values ± standard deviation (SD). *p < 0.05, **p < 0.01, and ***p < 0.001 were considered statistically significant. Each treatment was repeated at least three times.

Brucella Infection-Induced Inflammatory Response In Vitro and Vivo
The innate immune system is essential for detection and elimination of bacterial pathogens, inflammatory response caused by Brucella controls infection and be a protective effect in host, inflammasome NLRP3 complex was activated during Brucella infection (11). To study inflammatory response of Brucella infection in vitro, macrophages were infected by S2308 at different time and NLRP3 and Pro-caspase-1 were detected at mRNA (Figures 1A, B) and protein levels ( Figure 1C), our results showed that inflammasome NLRP3 and Pro-caspase-1 were significantly increased, especially at 24 h of infection. Furthermore, in the study in vivo, mice were infected with B. abortus 2308 strain by intraperitoneal injection (dose 1×10 6 cells). In parallel, the control group was injected with the same volume of PBS solution. Peripheral blood was collected weekly until 4 weeks following injection. We then measured the levels of the IL-1b, IL-18 inflammatory interleukins, and TGF-b in the serum using ELISA and found a significant increase in all the examined time points compared with the control group ( Figures  1D-F). Histopathological examination of the liver tissues of the infected mice revealed nuclei dissolution, cell necrosis, and a small number of cells undergoing fatty degeneration. The examination of spleen tissues revealed the presence of endothelial cells in nodules and local tissue congestion ( Figure  1G). Furthermore, IHC assay showed that NLRP3, Caspase-1, and TGF-b were significantly higher in the infected group ( Figure 1H). The results demonstrated inflammatory response were activated during Brucella infection in vivo and vitro.

Differentially Expressed lncRNAs Involved in Inflammatory Responses in THP-1 Cells Infected With Brucella S2308
To evaluate which lncRNAs are differentially expressed during Brucella infection, the total RNA of three biological replicates was extracted from THP-1 cells infected with S2308 at 4 h and 24 h. The lncRNAs were analyzed by microarray, using the Human long non-coding RNA (lncRNA) V 3.0 gene chip. lncRNAs expression change more than two-fold in all the three biological replicates were selected to further study. Compared to the uninfected group (PBS), there were 54 differentially expressed lncRNAs (46 upregulated and 8 downregulated) after 4 h following infection, and 235 differentially expressed lncRNAs (171 upregulated and 64 downregulated) 24 h following infection. The heatmap of partial differentially expressed lncRNAs is shown in Figure 2A.
We further verify the expressions of partial lncRNAs in THP-1 cells infected by S2308 both 4 and 24 h using qRT-PCR, showing that the same tendency with RNA-seq (Supplementary Figure  1A). As the expression of lncRNA after 24 h following infection mimics more closely the physiological changes caused by Brucella (27), all the subsequent analyses were performed on the 24 h dataset. The lncRNAs target genes enriched in pathways associated with inflammation and immune response in host were screened out using Gene Omnibus (GO) ( Figure 2B) and Kyoto Encyclopedia of Genes and Genomes (KEGG) (supplementary Figure 2B). To investigate the possible functions of these lncRNAs, we perform a prediction of potential targets using co-expression analysis, a lncRNA-mRNA correlation network was constructed ( Figure 2C). lncRNAs can act as regulators in innate immunity and antimicrobial defense (28). Therefore, we determined whether the differentially expressed lncRNAs induced by Brucella infection could be inflammatory response and antimicrobial defense regulators. To do this, we primarily focused on lncRNAs enriched in the NODlike receptor signaling pathway and the transforming growth factor b (TGF-b) signaling pathway according to GO and KEGG pathway analysis. The lncRNA-mRNA correlation network for these two pathways is shown in Figure 2D. There were 19 differentially expressed lncRNAs associated with the NOD-like receptor signaling pathway and 23 lncRNAs involved in the TGF-b signaling pathway. We then identified 6 lncRNAs, all of which were involved in inflammation ( Figure 2E): Of the 6 identified lncRNAs, 2 were upregulated, and 4 were downregulated. The expression of these lncRNAs in THP-1 cells was further validated using qRT-PCR ( Figure 2F).

Decrease of the lncRNA P33714 Expression Promotes the Secretion of Pro-Inflammatory Cytokines in THP-1 Cells Infected by S2308
To characterize the role of the six candidate lncRNAs involved in inflammation, we designed three siRNAs for each lncRNA and transfected them into THP-1 cells to detect the expression of the target lncRNA (Supplementary Table 1). siRNAs efficiency was screened using qRT-PCR and the most efficient siRNA for each lncRNA (Supplementary Table 2) was used for subsequent experiments. The selected siRNAs were transfected into THP-1 cells for 24 h, which were subsequently infected by S2308 for 24 h. Laser confocal results shown that the expression of NLRP3 inflammasome in THP-1 cells was higher in P3852-siRNA, P30159-siRNA, and P33714-siRNA than other groups ( Figure  3A). Further, the expression of pro-inflammatory interleukin IL-1b decreased significantly upon the knock-down of P3852, P662, P30159, and P16218, but not P33714 ( Figure 3B). Conversely, IL-18 levels were lower upon knock-down of P662 but higher upon knock-down of P16218, P12873, and P33714 ( Figure 3C). We found that knock-down of P33714 could inhibit Brucella intracellular survival ( Figure 3D), and that the expression of TGF-b was lower for all the knock-downs ( Figure 3E). Collectively when P33714 was knocked-down, IL-1b and IL-18 increased and the number of surviving Brucella in THP-1 cells decreased. Based on these results, we focused on the lncRNA P33714 (NR_037940.1), which is a readthrough of the HOXA10-HOXA9 genes. P33714 is located on chromosome 7p15.2 (from nucleotide 27162438 to 27180261). It consists of 2198 nucleotides and 3 exons. A platform created by Jie Lin (25) contains the orthologs of human and mouse lncRNAs annotated in GENCODE, which allows us to search for the orthologous lncRNA of P33714.In this way, we found the P33714 ortholog in mice, called Gm28309 (Ensembl Gene ID: ENSMUSG00000099521.1). Gm28309 is located on chromosome 2 (from nucleotide 74,683,446 to 74,694,194), and is a 623 transcript forward strand with 3 exons ( Figure 3F). P33714 and Gm28309 expression levels decreased in THP-1 and RAW264.7 cells infected by S2308, respectively. Although they showed a different kinetic expression at 24 h postinfection they showed the same reduced expression levels ( Figure   3G). Therefore, we suspected that Gm28309 may be involved in the activation of inflammation in host cells in response to Brucella infection.

Gm28309 Overexpression Negatively
Regulates the Production of NLRP3 Inflammasome in RAW 264.7 Cells Infected by S2308 Among all the NLR inflammasome complexes, the NLRP3 inflammasome is the best-characterized and is a crucial signaling node controlling the maturation of two proinflammatory interleukins belonging to the (IL)-1 family: IL-1b and IL-18 (9). To further characterize the role of P33714 in inflammation, we overexpressed Gm28309 in RAW264.7 cells and detected, using qRT-PCR, the expression levels of NLRP3 inflammasome components, such as NLRP3 and Caspase-1 in RAW264.7 cells overexpressing Gm28309. Results showed that the expression plasmid for Gm28309 (Gm28309(+) significantly increased Gm28309 expression levels when transfected into RAW264.7 cells, compared with mock-transfected cells and cells transfected with an empty pCDNA3.1 ( Figure 4A).
Overexpression of Gm28309 reduced the formation of the NLRP3 inflammasome, decreasing both the mRNA ( Figures  4B, C) and protein ( Figure 4D) levels of NLRP3 and procaspase-1. Furthermore, mRNA levels of factors activated by the NLRP3 inflammasome (IL-1b and IL-18) were significantly decreased in Gm28309 overexpressed cells, but higher in S2308 infection group (Figures 4E, F), as well as protein secretion levels detected using ELISA ( Figures 4G, H). In addition, we observed an increase in TGF-b expression upon S2308 infection, which was attenuated in cells overexpressing Gm28309 (Figures 4D, I).
Collectively, our results indicate that Gm28309 expression was decreased in RAW264.7 infected by S2308 whereas overexpression of Gm28309 reduces the production of the NLRP3 inflammasome.
Gm28309 Sequesters miR-3068-5p to Activate the NF-kB Signaling Pathway lncRNAs can perform various functions depending on their subcellular location (29). If in nucleus, they are primarily involved in transcription and chromatin remodeling, while in the cytoplasm, they participate in gene regulation by forming complexes with specific proteins (30). We analyzed the cellular localization of Gm28309 by performing qRT-PCR of RAW264.7 subcellular fractions. We found that Gm28309 is preferentially localized in the cytoplasm rather than in the nucleus (Figure 5A), suggesting a potential role for Gm28309 as a competing endogenous RNA (ceRNA). We speculated that Gm28309 might target factors involved in the NF-kB pathway, a well-known inflammationassociated signaling pathway (12). Hence, we performed bioinformatics analysis and we detected a potential targets: miR-3068-5p whose target is kB-Ras2 ( Figure 5B). kB-Ras2 (NKIRAS2) is an IkBb-interacting protein that could inhibit NF-kB activation (31), Moreover, kB-Ras2 knock-out mice show an enhanced IkBbdependent inflammatory response (32). Site-directed mutation at site was used to construct the mutant forms of Gm28309 and kB-Ras2 ( Figure 5B), and the wild-type or mutant forms of Gm28309 and kB-Ras2 were cloned to pGL3 plasmid respectively. The result of luciferase assays showed a decrease of luciferase activity in the presence of miR-3068-5p overexpression, while their mutant forms were not affected ( Figures 5C, D). Moreover, qRT-PCR results showed that, in RAW264.7 cells infected with S2308, the decreased expression of Gm28309 is accompanied by an increased expression of miR-3068-5p and the NF-kB p65 subunit; in contrast, Gm28309 overexpression decreased levels of miR-3068-5p and p65 levels ( Figures 5E, F), but increased expression of kB-Ras2 (Figures 5G,  H). Furthermore, when we treated RAW264.7 cells with an miR-3068-5p inhibitor, we observed a decrease in p65 phosphorylation levels ( Figures 4F, H) but a significant increase in kB-Ras2 ( Figures  5G, H), compared with that in cells infected by S2308. These results indicated that, in normal conditions, Gm28309 act as a ceRNA by binding miR-3068-p5 molecules, preventing kB-Ras2 mRNA degradation and ultimately the suppression of the NF-kB pathway. Upon Brucella infection, the decreased levels of Gm28309 enable the degradation of kB-Ras2 mRNA by releasing miR-3068-p5 and the activation of the NF-kB pathway.
Brucella Activates the NLRP3 Inflammasome Through the Gm28309-miR3068-5p-NF-kB Regulatory Axis In Vitro We further explored the relationship between miR-3068-5p and NF-kB/NLRP3 inflammasome signaling pathway using a specific small-molecule p65 inhibitor, JSH-23. Data showed that the overexpression of miR-3068-5p could significantly increase the expression of p65 and components of the NLRP3 inflammasome, as determined using qRT-PCR (Figures 6 A-C) and western blot ( Figure 6D). However, cells treated with 300 mM-specific p65 inhibitor JSH-23 for 24 h (33) was observed a decrease in the expression of p65 and components of the NLRP3 inflammasome. Moreover, we found that the expression of the NLRP3 inflammasome was higher when miR-3068-5p was overexpressed in the presence of the p65 inhibitor both at the mRNA level (Figures 6B, C) and the protein levels ( Figure 6D), as well as the secretion of IL-1b and IL-18 ( Figures 6E-H). Furthermore, the number of intracellular Brucella was higher when p65 was inhibited ( Figure  6I). Therefore, our results demonstrated that Gm28309 has an important role in the NF-kB/NLRP3 inflammasome pathway through its binding to miR-3068-5p.

DISCUSSION
In this study, we found that the expression profiles of lncRNA and mRNA in macrophages infected with B. abortus for 4 and 24 h were significantly altered compared to normal cells. This observation was consistent with what has been previously in human macrophages infected by Mycobacterium tuberculosis (34) and Listeria monocytogenes (35). Moreover, it is worth noting that more lncRNAs were differentially expressed after 24 h of infection compared to 4 h of infection (235 vs. 54, respectively). These results are consistent with the idea that Brucella, which is an intracellular pathogen, use the IV secretion system (T4SS) to deliver bacterial effectors that modulate of the expression of lncRNAs and mRNAs, to alter host functions (36), thereby escaping neutralization and replicate in the Brucella containing vacuole (BCV) (37). Recognition of Brucella PAMPs by different TLRs (38) triggers a signaling cascade to activate NF-kB factors (39), as well as the NLRP3 inflammasome, to control the initial response against Brucella (40). The activation of NF-kB is has previously been reported to be correlated with an effective inflammatory response to B. abortus infections (41). The role of lncRNAs and miRNAs during the pathogenesis of Brucella infection remains poorly understood. In this study, we identified a lncRNA, P33714, as a regulator of the inflammatory response. The expression of P33714 and its murine ortholog Gm28309 expression is decreased in THP-1 and RAW264.7 cells infected by S2308. Moreover, we discovered that Gm28309 negatively regulates the production of NLRP3 inflammasome in RAW 264.7 cells. Collectively, our results suggest a key role for lncRNA Gm28309 in the control of the host response against Brucella infection.
It has previously been reported that the expression of miRNAs is significantly altered in RAW264.7 macrophage cells (42), CD4+ T cells (43), and CD8+ T cells (44) upon Brucella infection. We found that Gm28309 was primarily localized in the cytoplasm, suggesting a role as a miRNA sponge. Bioinformatics analysis revealed that Gm28309 has predicted binding sites for miR-3068-5p and kB-Ras2. kB-Ras2 is an indirect inhibitor of the NF-kB transcription factor through its binding to IkB proteins. The binding between kB-Ras2 and IkB decrease IkB rate of degradation, blocking NF-kB activation (32). Mechanically, our results suggest a model in which Gm28309 acts as a ceRNA to sequester miR-3068-5p, blocking the degradation of kB-Ras2, which in turn inhibits the NF-kB pathway (Figure 7).
TGF-b is a pleiotropic cytokine that is involved in osteoarthritis (45). Inhibition of TGF-b signaling protects adult knee joints in mice against the development of osteoarthritis (46). Interestingly, arthritis is the classical brucellosis symptom, both in humans and animals (47). Consistent with this, our data showed that the expression of TGF-b was higher following Brucella infection, both in vivo and in vitro. We found that TGF-b is decreased while Gm28309 was overexpressed during Brucella infection, therefore, the downregulation of Gm28309 upon Brucella recognition by TLRs promotes the production of TGF-b to activate the NF-kB signaling and trigger the inflammatory response (Figure 7). Further investigations are warranted to elucidate the molecular mechanism underlying Gm28309 negative regulation of TGF-b expression.
Collectively, our results suggest a model in which the recognition of Brucella by TLRs located on the host cell membrane represses the expression of lncRNA Gm28309. The decrease of Gm28309 has two effects, both resulting in the activation of the NF-kB pathway: On one side, expression of TGF-b is promoted, leading to a cascade of TAK1 and IKK kinases promoting p65 phosphorylation; while on the other, miR-3068-5p, which is normally sponged by Gm28309, is released into the cytoplasm, where it targets kB-Ras2 for degradation, thereby inducing p65 phosphorylation. The activation of the NF-kB pathway promotes the assembly of the

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The animal study was reviewed and approved by the Animal Health Committee of Shihezi University.