A Conserved Long Noncoding RNA, GAPLINC, Modulates The Immune Response During Endotoxic Shock

To learn more info plz contact david.wan@wecistanche.com

Significance

Inflammation has largely been studied in the context of protein-coding genes. Recent studies have uncovered IncRNAs as important regulators of immunity. The functional characterization of these genes remains an active area of research. In this study, we identify GAPLINC as a functionally conserved IncRNA between humans and mice.GAPLINC depletion results in enhanced expression of immune response genes that are direct NF-kB targets. Astoundingly, we observe that Gaplinc knockout mice show resistance to LPS-induced endotoxic shock and find that basal expression of inflammatory genes prevents dot formation to protect against multiorgan failure and death. These findings have implications in the treatment of sepsis, in which new therapies targeting IncRNAs can contribute valuable information in understanding inflammation and improving patient outcomes.

Click here to learn more about Cistanche

FG 1. identification and characterization of macrophage-specific IncRNA GAPLINC. (A) A schematic for macrophage differentiation in vitro using primary human celeb or immortalized THP-1 cells, Isolated monocytes from human P8MCs are differentiated into macrophages using recombinant magnon stimulating factor. PHP-1 cells are differentiated into macrophages by treatment with PMA (100 nM). (8)RNA-Seg analysis on macrophages differentiated from monocytes isolated from human PBMCs (n = 4 donors). Results are represented in a volcano plot. GAPUNC (shown in red) is the most up-regulated and RNA (>1,000-fok). (C) RNA-Seq analysis of GAPLNC expression during monocyte to macrophage differentiation for the indicated time points.GAPLINC expression is represented as copies per cell (FPKM. (D) Heat map represents gene expression from a custom Nanostring panel, which shows the top 10 differentially expressed InciNA comparing monocytes to macrophages in primary human cells and THP-1 cells. Data from Nanotring were performed in biological duplicates. (E) A schematic for granulocyte-monocyte progenitor cells that gives rise to two distinct populations; 1)MDDCs and 2) MDMs. UCSC genome browser track displays RNA-Seq reads from monocytes, macrophages, and dendritic cells at the GAPLINC locus. () qPCR analysis of RNAs purified from nuclear (white) and cytoplasmic (gray) fractions in MDMs(G qPCR analysis of Russ isolated from different polysome fractions of MDM systems (10 to 50% sucrose gradient, SW41, 40 K rpm, for ~1.5 h).

Recent studies have identified thousands of long noncoding RNAs (ncRNAs)in mammalian genomes that regulate gene expression in different biological processes. Although IncRNAs have been identified in a variety of immune cells and implicated in immune response, the biological function and mechanism of the majority remain unexplored, especially in sepsis. Here, we identify a role for a hdRNA—gastric adenocarcinoma predictive long intergenic noncoding RNA(GAPLINC)—previously characterized for its role in cancer, now in the context of innate immunity, macrophages, and LPS-induced endotoxic shock. Transcriptome analysis of macrophages from humans and mice reveals that GAPLINC is a con-served IndRNA that is highly expressed following macrophage differentiation. Upon inflammatory activation, GAPLINC is rapidly down-regulated. Macrophages depleted of GAPLINC display enhanced expression of inflammatory genes at baseline, while over-expression of GAPLINC suppresses this response. Consistent with GAPLINC-depleted cels, Gaplinc knockout mice display enhanced basal levels of inflammatory genes and show resistance to LPS-induced endotoxic shock. Mechanistically, survival is linked to increased levels of nuclear NF-xB in Gaplinc knockout mice that drive basal expression of target genes typically only activated following inflammatory stimulation. We show that this activation of immune response genes prior to LPS challenge leads to de-areas blood clot formation, which protects Gaplinc knockout mice from multiorgan failure and death, Together, our results identify a previously unknown function for GAPLINC as a negative regulator of inflammation and uncover a key role for this lncRNA in modulating endotoxic shock.

Fig.2. GAPLINC is a negative regulator of inflammation and dependent on Nf-sB signaling. Monocytes isolated from human PBMCs and differentiated into macrophages were transfected with control or GAPLINC siRNA. (A)RNA-Seqanalysis on GAPLINC kd or control siRNA MDMs. RNA-Seq was performed in biological duplicates. Results are represented in a volcano plot. Significantly up-regulated genes with fold change ≥2 are boxed in red. (6) GO-Term analysis on significantly up-regulated genes. (C)Heat map represents gene expression of top immune-related genes up-regulated upon GAPLINC kd. Data from RNA-Seq was performed in biological duplicates. (D)GAPLINC isoforms in MDMs were determined by Nanopore-based R2C2 sequencing. Data from Nanopore sequencing was performed in biological duplicates (raw data are available at SRAunder Bloproject PRINA639136). (E)A table representing read counts and percent of each GAPLINC isoform. (F)Bidirectional vector expressing GFP-Zetocin on one side and GAPLINC on the other side. (G and H) qPCR analysis of GAPLINC expression in THP-1 cells expressing ectopic GAPLINC or empty vector control. Levels of IL6 were quantified following stimulation with LPS (200 nal mL) for 6 h; data were pooled from three independent experiments.*P<0.05. () RNA-Seq analysis of MDMs stimulated with LPS(200 ng/mL) for the indicated time points. Data from RNA-Seq was performed in biological duplicates. GAlPLINC expression is represented as copies per cell (FPKM), U qPCR analysis of GAPLINC in MDMs （n = 3）pretreated with DMSO or BAY-7082（10μM）, followed by LPS stimulation （200 ngmL） for 6 ht data were pooled from three independent experiments.*P<0.05. (K)ATACseq analysis of monocytes and macrophages, untreated and treated with LPS (200 ng/mL) for 1,6, and 18h.UCSC browser track displays ATAC-Seq reads at the GAPLINC locus.

sepsis is a life-threatening illness caused by an overreaction of the body to the presence of infection, which can rapidly lead to multiorgan failure and death. The immune system is essential in providing protection against infection; however, uncontrolled activation can have serious consequences for the host. According to the Centers for Disease Control and Prevention, one in three patients who die in a hospital has sepsis(1). and yet we still do not understand the underlying molecular mechanisms that lead to fatality. Clinical options for the treatment of sepsis are limited to the delivery of fluid, antibiotics, and supportive care and have remained largely unchanged for decades. Though early diagnosis and rapid treatment have improved sepsis outcomes(2), there is a critical need to develop new therapies. Although gene expression studies have been performed to examine potential therapeutic targets for sepsis, these targets remain largely uncharacterized(3). We have identified a long noncoding RNA (lncRNA) with roles in controlling the immune response and endotoxic shock that provides avenues for novel drug development to target sepsis.

Result

Fig. 3. GAPLINCis conserved in mice and regulates response to endotoxic shock. (4) GAPLINC is conserved in synteny. GAPLINC is located on Chr18in humans and on Chr17 in mice, between protein-coding genes Dlgap1 and Gift. Dlgapil is not expressed in macrophages. (B)MCA shows the distribution of Gaplinc levels in various immune cell types (BM). (C) qPCR analysis of Gaplinc expression in B cells and BMDMs; these data (mean ± SD) are representative of three independent experiments, (D) qPCR analysis of Gaping expression in BMDMEstimulated withLPS (200 na/mL) for 6h; these data (mean ±SD) are representative of three independent experiments. (E) Schematic of Gaplinc locus before and after CRISPRCas9 mediated deletion. Dashed lines indicate the approximate region of deletion. Gel represents PCR amplification of genomic data. Amplicon lengths are compared for WT and Gaping KO mice, (F) qPCR analysis of Gaplinc expression in WT and Gaplinc-KO BMDMs using a combination of primers to detect Exon1, Exon2, and exon-spanning regions of the Gaplinc transcript; these data (mean ± SD) are representative of three independent experiments. (G)RNA-Seq analysis in BMDMs from WT and Gaplinc KO mice (n =3). Results are represented in a volcano plot. Significantly up-regulated genes with a fold change ≥2 are shown in red. (H) GO-Term analysis on significantly up-regulated genes. (I and J) Survival data of WT and Gaplinc KO mice are shown in response to the E. coli LPS challenge(5 mg/kg/mice) (n =6 to 10). The statistical test of differences was counted using the log-rank (Mantel-Cox) test. "…P< 0.001. Changes in body temperature of WT and Gapline KO mice were recorded at the indicated time points. (k) Survival data of WT and Gaplinc KO mice are shown in response to the E. coli LPS challenge(20 mg/kg/mice)(n=10). The statistical test of differences was calculated using the log-rank(Mantel-Cox)test.***P<0.001.

As lncRNA expression can regulate the immune response by affecting immune cell differentiation and their respective function(4-7), we wanted to investigate the role of ncRNAs in macrophages. Macrophages are important innate immune cells that can be derived from monocytes and are critical for pathogen recognition through the use of Toll-like receptors (TLRs).Upon activation. TLRs initiate complex signaling pathways that activate key transcription factors such as NF-kB.leading to the transcription of hundreds of immune response genes(8).

To identify IncRNAs involved in macrophage differentiation and function, we conducted RNA sequencing(RNA-Seq)in both human primary monocyte-derived macrophages(MDMs)and the monocytic cell line THP-1s(Fig. lA and SI Appendix. Fig. S1). We identified gastric adenocarcinoma predictive long intergenic noncoding RNA (GAPLINC) as the most up-regulated IncRNA during monocyte to macrophage differentiation (Fig. 1B). GAPLINC levels were detectable by day 1 and increased to ~300 copies per cell(Fig 1C). Using RNA-profiling technology (nCounter, Nanostring), we validated GAPLINC as one of the top 10 mRNAs expressed in differentiated primary MDMs and THP-1s (Fig1D). We also confirmed that GAPLINC is highly expressed in MDMs but not expressed in the closely related cell type monocyte-derived dendritic cells (MDDCs)(Fig. 1E), suggesting that expression of GAPLINCis cell-type specific. By performing a cell fractionation experiment and measuring GAPLINC levels in the cytoplasmic and nuclear compartments of macrophages using qPCR, we found GAPLINC is predominantly localized in the cytosol when com-pared to CD14, a cytoplasmic mRNA, and NEAT1, a nuclear IncRNA (Fig. 1F). This is consistent with findings in cancer cells, in which GAPLINC is mainly localized to the cytoplasm(9). To ensure GAPLINC is noncoding, we performed polysome pro-filing, a method used to analyze whether a gene is actively translated into protein. In contrast to CD14, neither NEATI nor GAPLINC were found in the high-polysome fraction, suggesting that GAPLINC is not translated (Fig 1G),

Fig.4. Gaplnc KO up-regulates IRGs and shows an increased level of p65in the nude at baseline. (A) Cytokine levels in serum of WT and Gaplinc KO mice at basal.n=4 to 7,*P<0.05.(B and C)RNA-Seg analysis in BMDMs from WTand Gaplinx KO mice stimulated with LPS (200naiml)for6h (n=3). The23aenes up-regulated in Gap/inc KO-only condition are compared to WT and Gapfinc-KO BMDMs stimulated with LPS. The resulting fold change upon LPS stimulation is shown for WT and Gapinc-KO BMDMs.Genes are ranked according to their fold change in WT. (D) Coagulation parameters assessed for WT and Gapinc KO mice challenged IP.with E. coils(5mg/kg/mice)(n =4 to 5).Plasma was collected 18h post-LPSiniection. aPTT was measured. (E)Genes upregulated non siRNA KD of human GAPLNC in MDMs are compared to genes up-regulated upon CRSPRCas9 knockout of mouse Gaplincin BMDMs (fold change ≥1.5). Up-regulated genes overlapping in both humans and mice are shown in the middle. (F) Western blot of kB-gin WT and Gaplinc-KO BMDMs at the indicated time points following stimulation with LPS (200 ngmL); these data are representative of three independent experiments,(G) Western blot of p65 in WT and Gaplinc-KO BMDMs (n = 3) at basal.**P<0.01. (H Western blot of p65 in the nuclear fraction of WT and Gaplinc-KO BMDMs (n = 3).

Next, we investigated the effect of GAPLINC silencing on differentiating macrophages. As GAPLINC levels increased during differentiation, we hypothesized that GAPLINC knock-down (KD) would impact genes involved in differentiation. Using pooled small interfering RNA(siRNA), we achieved~55 to 65%kd of GAPLINC in primary MDMs(SIAppendix, Fig. S2). We performed RNA-Seq and identified a number of genes that were dysregulated upon GAPLINC kd (Fig. 24). We confirmed the top hits using Nanostring (SI Appendix, Fig. S3). Gene Ontology (GO) enrichment analysis showed that immune response genes were significantly overrepresented in genes up-regulated upon GAPLINC kd and not genes involved in macrophage differentiation, contrary to our original hypothesis. (Fig. 2B). Notably, differentially expressed genes in GAPLINC-KD cells include proinflammatory cytokines and chemokines(IL6, CXCL10, and TNFSF10), IFN-stimulated genes(ISGs)(IFIT2 and RSAD2), and guanylate-binding proteins (GBPs)(GBP3 and GBP5)(Fig. 2C and SI Appendix. Fig. S4). To verify that macrophage differentiation was unaffected by GAPLINC kd. levels of CD11B.CD16 and CD14 were measured in control and signaling-treated macrophages and found to be similar(SI Appendix. Fig S5). Next, we used an additional kd approach utilizing CRISPR inhibition (CRISPRi)to target the transcriptional start site by GAPLINCusing three different guide RNAs (gRNAs) in THP-1 cells stimulated with phorbol 12-myristate 13-acetate (PMA). We achieved over 80% kd of GAPLINCand again observe increased expression of the proinflammatory cytokine IL6 when GAPLINC is knocked down at baseline(SI Appendix. Fig. S6 A and B). Interestingly, following lipopolysaccharide (LPS)activation, the levels of IL6 were similar in the control versus GAPLINC kd THP-1 cells (SIAppendix, Fig. S6C). Since the baseline levels differ, the overall fold induction of IL6 is reduced when GAPLINC is removed. (S7 Appendix, Fig. S6D).

As GAPLINC kd resulted in the up-regulation of immune response genes, we generated a THP-1 cell line overexpressing GAPLINC to determine if it mediates the opposite effect. Using long-read sequencing data. we identified the dominant isoforms of GAPLINCexpressed in MDMs(Fig. 2D). The most abundant isoform, which matches the Reference Sequence (RefSeq) gene annotation, was incorporated into our construct (Fig. 2E). We utilized a lentiviral vector containing a bidirectional promoter to drive green fluorescent protein(GFP)/Zeocin and GAPLINC in parallel (Fig. 2F). We confirmed using qPCR that GAPLINC was stably expressed in THP-1s compared to control (Fig. 2G). Overexpression of GAPLINC reduced IL6 at the RNA level compared to control following stimulation with LPS, a component of gram-negative bacteria (Fig. 2H). These observations suggest that GAPLINC acts as a negative regulator of the inflammatory response.

Next, we examined GAPLINC levels in primary MDMs in response to inflammation. Upon LPS stimulation, we found that GAPLINC was rapidly down-regulated (Fig. 2I). Additionally, using qPCR we show that GAPLINC is also down-regulated following activation with a variety of TLR ligands(SI Appendix, Fig. S7A). Induction of positive-control inflammatory genes TNF-a, IL6, and CCL5 was confirmed (SI Appendix, Fig. S7B). These data suggest that GAPLINC expression must be reduced following stimulation in order for optimal inflammatory gene induction to occur. To evaluate the role of NF-kB in controlling the expression of GAPLINC, MDMs were pretreated with dimethyl sulfoxide (DMSO) or BAY11-7082, an NF-B inhibitor,

followed by stimulation with LPS for 6 h.In BAY11-7082-treated MDMs.the down-regulation of GAPLINC was impaired relative to control(Fig. 2/). which suggests the regulation of GAPLINC is dependent on NF-KB signaling. Induction of positive-control inflammatory genes TNF-α and IL6 upon LPS stimulation was confirmed (SIAppendix, Fig, S7C).To understand how GAPLINC is being regulated, we utilized Assay for Transposase-Accessible Chromatin using sequencing(ATAC-seq)in MDMs to assess chromatin accessibility of the GAPLINC locus. We found that GAPLINC is actively transcribed in resting macro-phages but tightly shut down following LPS stimulation (Fig. 2K), suggesting that GAPLINC is regulated at the level of transcription.

To explore the conservation of GAPLINC between humans and mice. we sought to identify syntenic loci, in which genes are positionally conserved between the same two protein-coding genes. followed by an assessment for functional conservation. Here, we show that Gaplinc is positionally conserved, locating a transcript between genes Dlgap1 and TgifI (Fig.3A).To confirm cell-type specificity, we utilized the Mouse Cell Atlas (MCA)

(10) to assess transcript levels across immune cells in the bone marrow(BM) and found it highest expressed in macrophages, with lower levels of expression in neutrophil progenitors (Fig. 3B). Next, to determine if expression patterns were conserved during macrophage differentiation in both humans and mice, we performed RNA-Seq comparing BM cells to bone marrow-derived macrophages(BMDMs), completed de novo transcript assembly, and found mouse Gaplinc levels increased following differentiation(SI Appendix, Fig. S8A). We validated this by qPCR, comparing Gaplinc levels in BM cells to BMDMs (Fig. 3C). Comparable to human GAPLINC. mouse Gaplinc is rapidly down-regulated in LPS-stimulated BMDMs(Fig.3D). As a control, we confirmed the induction of inflammatory genes The-a and I6(SI Appendix. Fig. S8B). Mouse Gaplinc is also rapidly down-regulated following activation with various TLR ligands. Induction of the positive-control genes ll6 and Ccl5 were also confirmed(SI Appendix, Fig. S8C).

Using CRISPR, we generated a Gaplinc knockout (KO)mouse in which Exon I and the majority of the first intron were removed (Fig. 3E). Chaplin KO mice bred normally and displayed no obvious developmental defects. The deletion of Gapline was confirmed by PCR amplification of genomic DNA in wild-type(WT) and Gaplinc KO mice, with amplicon sizes ~1,300 bp for WT and~500 bp for Gaplinc KO mice(Fig.3E), The full genotyping strategy to confirm WT and Gaplinc KOs is highlighted in SI Appendix, Fig. S9. We also confirmed Gaplinc deficiency in BMDMs using qPCR (Fig. 3F). To assess whether Gaplinc deficiency affected differentiation, we stained WT and Gaplinc-KO BMDMs for CD11B and F4/80; staining patterns were similar, suggesting normal macrophage differentiation (SI Appendix, Fig. S104). To assess whether Gaplinc KO disrupted macrophage function, we compared phagocytosis activity in WT and Gaplinc-KO BMDMs and found no differences(SI Appendix, Fig.S10B).

To assess the global impact of Gaplinc deficiency in macrophages, we performed RNA-Seq on WT and Gaplinc-KO BMDMs; both were untreated and LPS stimulated for 6 h. In Gaplinc-KO BMDMs, the expression of 23 genes (shown in red. Fig.3G) was significantly up-regulated at baseline. These genes include proinflammatory cytokines and chemokines (Il6.l1-a. I1-8,and Cxd10),ISGs(Ifitlbl1), GBP-family members(Gbp5 and Gbp0),and cell-surface markers specific for activated macrophages(Cd69).GO analysis confirmed that genes involved in the immune response are overrepresented in Gaplinc-KO BMDMs at baseline(Fig. 3H). These genes are typically up-regulated following LPS stimulation(SI Appendix. Fig. S11). Similar to data obtained from our human studies, these data suggest that GAPLINC is functionally conserved across species to basally control the expression of inflammatory-response genes (IRGs). There was no major difference in the genes that were up-regulated following LPS stimulation between the WT and Gaplinc-KO BMDMs (Dataset S1).

Numerous studies have shown the ability of IncRNAs to regulate the transcription of neighboring genes(11). As such, we explored potential cis-regulatory roles for Gaplinc as its neighboring gene, Tgif1, has been previously implicated in modulating macrophage activation (12). Using our RNA-Seq data, we confirmed that in GaplincKO BMDMs, the expression of Tgifl relative to T cells was not affected (SIAppendir.Fig.S124). We further confirmed by qPCR that Tgifl was unaltered in Gaplinc-KO BMDMs(SI Appendix, Fig. S12B). Next, we explored the possibility that disruption of the Gaplinc locus could remove an important regulatory element, such as an enhancer. We utilized ATAC-seq data from WT BMDMs to assess transcriptionally active regions in the Gaplinc locus(13). We did not identify signals within the Gaplinc deletion region, only those corresponding to the Gaplinc promoter(SI Appendix. Fig.S13). Collectively, these data suggest that up-regulation of IRGs upon Gaplinc deficiency is not due to effects on neighboring genes or removal of a regulatory element.

As Gaplinc KO in BMDMs up-regulates the expression of IRGs under basal conditions, we next wanted to challenge Gaplinc KO animals in vivo to observe differences in host response. We employed an"endotoxic shock" model. in which Escherichia coli LPSwas intraperitoneally (i.p.) injected into WT and Gaplinc KO mice to measure differences in survival. At a dose of 5 mg/kg/mouse. WT mice showed 0% survival after 2 d (Fig.30. However,100% of Gaplinc KO mice survived, sg-getting that Gaplinc KO mice have resistance to LPS-induced endotoxic shock(Fig. 31). Notably, significant temperature differences between WT and Gaplinc-deficient mice are observed (Fig 3). Even when utilizing a much higher dose of LPS of 20 mg/kg, the Gqplinc-deficient mice display a significant survival advantage(Fig. 3K). Collectively, these data show that Gaplinc plays an important role in regulating the immune response in vivo.

From our human studies, we know that GAPLINC can regu-late immune genes at baseline. While there is no difference in cytokine expression following inflammatory activation, there is a difference in the magnitude of the response since the baseline levels differ, Therefore, to better understand these survival differences in vivo, we assessed for changes in cytokine expression at baseline. We utilized a multiplexed cytokine array to simultaneously measure biomarkers associated with the immune re-sponse, sepsis, and cancer. At baseline, key immune genes including MDC. MIP-1a.IL-13. IL-5. and M-CSF was significantly elevated in the serum of Gqplinc KO mice compared to WT(Fig.44). These cytokines are implicated in cell recruitment: however, using flow cytometry, we confirmed the percentage of neutrophils, T cells, B cells, eosinophils, monocytes, and macrophages were comparable between WT and Gaplinc KOmice at baseline(SI Appendix, Fig. S14). Interestingly, these percentages were also comparable to post-LPS challenges at the 6 h and 18 h time points(SI Appendix, Figs. S15 and S16). Furthermore, serum from WT and Gaplinc KO mice tested for clinical features of sepsis, including lactate and C-reactive protein(CRP)(S7 Appendix, Fig. S17), showed no differences

As increased levels of MDC or IL-13(Fig.44) have been previously characterized to play protective roles against endo-toxic shock through modulation of proinflammatory cytokines (14,15), we next examined for differences in proinflammatory cytokine levels, including l6, Il1-α, l1-β, and Cxcl10, following inflammatory activation. While WT and Gaplinc KO levels are the same at both the transcript (Fig. 4B) and protein level(SI Appendix, Fig.S18A-C), the magnitude of change is much lower (Fig. 4C), suggesting that this reduced fold change may play a role in preventing the susceptibility of Gaplinc KO mice to LPS. induced endotoxic shock.

Along with the rampant production of cytokines. another clinical aspect of endotoxic shock that can lead to mortality is the formation of blood clots in the smaller vessels, leading to a multiorgan failure(16.17). To address this, we analyzed the serum of LPS ip.-treated mice to assess differences in coagulation. Using an activated partial thromboplastin time(a) assay that measures time to clot formation, we find that WT mice show a significantly prolonged time compared to Gaplinc KO mice upon LPS challenge(Fig. 4D). Prolonged aPTT times suggest WT mice have already undergone coagulation and depleted key coagulation factors, such that at the time of testing, initiating clotting, and measuring, time to clot formation is increased. The data suggest that differences in survival upon LPS challenge are due to WT mice undergoing increased clotting, leading to eventual organ failure and death.

Next, we tried to mechanistically understand how Gaplinc is mediating this effect. First, we confirmed its localization and analyzed Gaplinc levels in both the cytoplasmic and nuclear compartments of BMDMs by qPCR. Similar to human GAPLINC. mouse Gaplinc was predominantly cytoplasmic(ST Appendix, Fig.S19). Next, we performed a modified Comprehensive Identification of RNA-binding proteins(ChIRP), coupled with mass spectrometry and small RNA sequencing, as GAPLINC has been previously shown to interact with micro RNAs(18). We did not observe interactions with previously identified targets (SIAppendix, Fig. S20).

As we could not identify a direct binding partner, we focused on the conserved function between humans and mice, specifically the conserved genes impacted in our human GAPLINC kd and mouse Gaplinc KO studies, the majority of which are NF-kB regulators (Fig. 4E). The most abundant form of NF-KB activated by LPS is the p65:p50 heterodimer, and seven out of the nine conserved genes are direct p65 targets(19). We used native RNA immunoprecipitation (RIP) and showed that there is no direct interaction between Gaplinc and p65(S7 Appendix, Fig. S21).In the classical NF-kB signaling pathway, the p6:p50 subunits are located in the cytosol and bound to inhibitory I-kB-αin resting cells. Upon activation, I-KB-α is degraded, allowing p65:p50 to translocate into the nucleus and activate target genes (20). In WT and Gaplinc-KO BMDMs, we measured the degradation of I-kB-α using Western blot and found no differences (Fig.4F). Next, we compared total p65 levels in BMDMs. In our RNA-Seg data,p65(RelA) transcript levels in both the basal and LPSstimulatedconditions for WT and Gaplinc-KO BMDMs are comparable(SI Appendix, Fig. S22). However, when we compared total p65 protein levels by Western blot, we found significantly increased p65levels in Gaplinc-KOcells compared to WT (Fig. 4G). This suggests that Gaplinc regulates total p65 levels at the translational level. Additionally, we assessed p65 levels in the cytoplasmic and nuclear compartments and found that nuclear p65 is more abundant in Gaplinc-deficient cells compared to WT (Fig. 4H). Combined, these data suggest a mechanistic role for GAPLINC in priming the activation of critical IRGs.

Discussion

Here, we have identified GAPLINC as a conserved cytosolically expressed IncRNA in humans and mice that functions to control inflammatory gene expression through regulation of NF-xB. In comparison to reported cytosolic IncRNAs that impact translational efficiency by modulating protein expression or ribosome assembly (21-23), we find Gapline is cytosolically contributing to p65 translational ability; however, the mechanism of interaction, either direct or indirect, remains to be determined. Our findings provide insights into how a functionally conserved IncRNA regulates the immune response and provide avenues of investigation for the development of therapeutics for endotoxic shock.