Long noncoding RNA SNHG16 regulates TLR4-mediated autophagy and NETosis formation in alveolar hemorrhage associated with systemic lupus erythematosus

Background Dysregulated long noncoding RNA (lncRNA) expression with increased apoptosis has been demonstrated in systemic lupus erythematosus (SLE) patients with alveolar hemorrhage (AH). SNHG16, a lncRNA, can enhance pulmonary inflammation by sponging microRNAs, and upregulate toll-like receptor 4 (TLR4) expression via stabilizing its mRNAs. TRAF6, a TLR4 downstream signal transducer, can induce autophagy and NETosis formation. In this study, we investigated whether SNHG16 could regulate TLR4-mediated autophagy and NETosis formation in SLE-associated AH. Methods Expression of SNHG16, TLR4 and TRAF6 and cell death processes were examined in lung tissues and peripheral blood (PB) leukocytes from AH patients associated with SLE and other autoimmune diseases, and in the lungs and spleen from a pristane-induced C57BL/6 mouse AH model. SNHG16-overexpressed or -silenced alveolar and myelocytic cells were stimulated with lipopolysaccharide (LPS), a TLR4 agonist, for analyzing autophagy and NETosis, respectively. Pristane-injected mice received the intra-pulmonary delivery of lentivirus (LV)-SNHG16 for overexpression and prophylactic/therapeutic infusion of short hairpin RNA (shRNA) targeting SNHG16 to evaluate the effects on AH. Renal SNHG16 expression was also examined in lupus nephritis (LN) patients and a pristane-induced BALB/c mouse LN model. Results Up-regulated SNHG16, TLR4 and TRAF6 expression with increased autophagy and NETosis was demonstrated in the SLE-AH lungs. In such patients, up-regulated SNHG16, TLR4 and TRAF6 expression was found in PB mononuclear cells with increased autophagy and in PB neutrophils with increased NETosis. There were up-regulated TLR4 expression and increased LPS-induced autophagy and NETosis in SNHG16-overexpressed cells, while down-regulated TLR4 expression and decreased LPS-induced autophagy and NETosis in SNHG16-silenced cells. Pristane-injected lung tissues had up-regulated SNHG16, TLR4/TRAF6 levels and increased in situ autophagy and NETosis formation. Intra-pulmonary LV-SNHG16 delivery enhanced AH through up-regulating TLR4/TRAF6 expression with increased cell death processes, while intra-pulmonary prophylactic and early therapeutic sh-SNHG16 delivery suppressed AH by down-regulating TLR4/TRAF6 expression with reduced such processes. In addition, there was decreased renal SNHG16 expression in LN patients and mice. Conclusions Our results demonstrate that lncRNA SNHG16 regulates TLR4-mediated autophagy and NETosis formation in the human and mouse AH lungs, and provide a therapeutic potential of intra-pulmonary delivery of shRNA targeting SNHG16 in this SLE-related lethal manifestation. Supplementary Information The online version contains supplementary material available at 10.1186/s12929-023-00969-5.


Background
Systemic lupus erythematosus (SLE) with a loss of immune tolerance to autoantigens, has overproduced autoantibodies, dysregulated cytokines milieu and defective T-cell subpopulations [1,2].Its crucial pathogenic mechanism is an imbalance between accelerated cell death and disposal of death-related materials [3].Immune complexes (IC) formed by pre-existing autoantibodies with released autoantigens from cell death processes can deposit in different organs, leading to visceral inflammation.Lupus nephritis (LN) is the most common cause of disease morbidity with IC depositing in glomerular basement membranes [4].Alveolar hemorrhage (AH), a fatal respiratory emergency, in SLE is mediated by accumulation of IC in pulmonary capillary walls, causing c with intra-alveolar assembly of red blood cells (RBCs) [5].AH has an up to 10% occurrence in SLE, and an association with higher disease activity than other clinical manifestations such as LN.
Noncoding RNAs (ncRNAs) can be classified as small ncRNAs and long ncRNAs (lncRNAs) [6].MicroRNAs (miRNAs), small ncRNAs, have complementary interaction with messenger RNAs (mRNAs) at their target sites in the 3′ untranslated region, i.e., miRNAs recognition element (MRE) [7], regulating the expression of proteins involved in various cell death processes [8][9][10].LncRNAs modulate physiological responses through interacting with intra-cellular molecules [11], and function as competing endogenous RNAs (ceRNAs) to de-repress the activity of other transcript RNAs by competing for the same miRNAs [12,13].By the interaction of lncRNAs with miRNAs, the regulation of gene expression can be extended to a complicated network-based pattern.Interestingly, a novel therapeutic approach in SLE is to utilize small interfering RNAs (siRNAs) or short hairpin RNAs (shRNAs) to target miRNAs or cytoplasmic lncRNAs, and small-molecule compounds capable of de-stabilizing putative ncRNAs [14].Our earlier experiments demonstrated that intra-pulmonary delivery of shRNA targeting lincRNA-p21, a pro-apoptotic lncRNA presenting in the cytoplasm, could reduce hemorrhagic frequencies through decreasing apoptosis formation in the SLE-AH lungs [15].
Small nucleolar RNAs with exon sequences, named small nucleolar RNA host gene (SNHG), can present in the cytoplasm to serve as ceRNAs to up-regulate the translation of target mRNAs [16,17].Increased SNHG16 expression has been demonstrated in interstitial lung disease and acute pulmonary infection, down-regulating the expression of specific miRNAs to enhance inflammation in the lungs [18][19][20][21].TRAF6, a toll-like receptor 4 (TLR4) downstream signaling molecule, is a potent inducer of various cell death processes [22][23][24].Silencing SNHG16 expression can alleviate apoptosis and autophagy induced by lipopolysaccharide (LPS), a TLR4 agonist, in lung cells by sponging particular miRNAs [19,20,25,26].SNHG16 acts as a ceRNA to upregulate TLR4 mRNA expression via targeting miR-15a/16 [27].This lncRNA can recruit EIF4A3 to stabilize TLR4 mRNA, while its knockdown and overexpression can reduce and enhance the expression of TLR4, respectively [28].MiR-146a, a ceRNA target of SNHG16, regulates TLR4-mediated cell death by targeting TLR4 and its downstream TRAF6 [19,24,29].In our previous studies analyzing the SLE-associated AH lungs [24], down-regulated miR-146a expression was shown to enhance apoptosis and neutrophil extracellular traps (NETs) formation through reversing its targeting effects on TAF9b, a p53 co-activator/stabilizer, and TRAF6, a signal transducer activating IL-8 production, respectively.Altogether, these findings raise a possibility that SNHG16 participates in the SLE-associated AH pathogenesis via regulating the TLR4-mediated cell death formation.
In this study, SNHG16, TLR4 and TRAF6 expression and cell death processes including apoptosis, autophagy and NETs formation were examined in lung tissues and peripheral blood (PB) leukocytes from AH patients associated with SLE and other autoimmune diseases as well as in the lungs and spleen from a pristane-induced C57BL/6 AH mouse model.The expression of NEAT1, another lncRNA participating in SLE disease activity and sharing the same target miR-146a with SNHG16 [30], was also evaluated in SLE-AH patients and a mouse model.SNHG16-overexpressed or -silenced LPS-stimulated alveolar and myelocytic cells were analyzed for autophagy and NETs formation, respectively.SNHG16overexpressed or -silenced alveolar cells were also stimulated with doxorubicin (Dox) to assess their apoptosis formation.Pristane-injected mice received the intrapulmonary delivery of lentivirus (LV) vectors carrying SNHG16 for overexpression, and prophylactic/therapeutic infusion of shRNA targeting SNHG16 to evaluate their

Patients and controls
SLE patients, 55 females and 7 males aged 20 to 62 years (36.2 ± 10.2), fulfilling the American College of Rheumatology revised criteria [31], and their age/sex-matched healthy controls (HC) were enrolled with an approval from the Institutional Review Board of National Cheng Kung University Hospital (No. A-ER-108-455 and No. B-ER-111-400).SLE disease activity index 2000 (SLE-DAI-2K) scores were used for assessing their disease activity [32].LN was diagnosed by histopathological and/ or laboratory evidences, while AH was defined as newonset pulmonary infiltrates, overt drop in hemoglobulin (Hb) and other clinical/laboratory findings [33].Control groups for AH included age/sex-matched HC, LN (only nephritis) and Nil (neither nephritis nor AH).Blood and urine samples were obtained from SLE patients and HC.Lung specimens were from SLE-associated AH with controls from non-inflammatory pneumothorax (PTX).Renal specimens were from LN with controls from uninvolved parts of renal cell carcinoma.In addition, blood samples and lung specimens were from AH patients related to other autoimmune diseases, including antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV), anti-phospholipid syndrome (APS) and IgA vasculitis (IgAV).A summarized table (Additional file 2: Table S1) shows the clinical profiles and therapeutic modalities of AH in different patient groups including SLE, AAV, APS and IgAV.

Cell lines
The cell lines used in this study were 293T cells (human embryonic kidney cells, American Type Culture Collection, ATCC, Manassas, VA, USA), MLE-12 cells (mouse alveolar epithelial cells, ATCC) and HL-60 cells (human promyelocytic cells, ATCC).All were cultured with 1 × 10 6 cells/mL in 3.5 cm dish in a humidified incubator with 5% CO 2 at 37 °C.

Cell purification
PB mononuclear cells (PBMCs) were purified by Ficoll-Paque PLUS (GE-Healthcare, Chicago, IL, USA) gradient centrifugation.PB neutrophils (PBNs) were obtained by Polymorphprep ™ (AXIS-SHIELD PoC AS, Oslo, Norway) gradient separation for the granulocyte fraction with the removal of RBCs by hypotonic lysis.PB monocytes (PBMs) were isolated from strongly adherent cells by incubating PBMCs in tissue culture-treated dishes for 90 min at 37 °C.The purity of human neutrophils and monocytes was more than 90 and 80% by flow cytometric analysis of surface markers CD11b and CD14, respectively, in this study.Urine sediment cells (USCs) were purified by centrifugation of fresh urine specimens at 3000g for 30 min under 4 °C.In addition, CD3-positive T cells, CD19-positive B cells and CD14-positive monocytes in PBMCs as well as miR-146a-overexpressed and SNHG16-silenecd green fluorescence protein (GFP)-positive cells were sorted by a Moflo XDP cell sorter (Beckman Coulter, Mountain View, CA, USA).
Wild-type 8-week-old female C57BL/6 Jackson National Applied Research Laboratories (JNarl) mice were purchased from National Laboratory Animal Center, (Taipei, Taiwan), and housed under specific pathogen-free conditions with free access to food and water on a 12 h/12 h light-dark cycle at the Laboratory Animal Center, College of Medicine, National Cheng Kung University.Mice were housed for 1 week before starting the investigation.Animal experiments were approved by the Institutional Animal Care and Use Committee, National Cheng Kung University (No. 109034 and No. 112067), and performed according to the guidelines.Mice received intraperitoneal injection of 3% thioglycollate medium (Difco, Detroit, MI, USA), followed by phosphate-buffered saline (PBS) injection 24 h later.Lavage fluid was centrifuged for 10 min at 200g at room temperature (RT) to collect cell pellets.Neutrophils were further isolated by Percoll gradient solution (Sigma-Aldrich, St. Louis, MO, USA).The purity of mouse neutrophils was up to 95% by flow cytometric analysis of the surface marker Ly-6G in this study [24].

Construction of LV-based shRNA targeting mouse SNHG16
Four shRNA sequences were designed as follows, including three targeting mouse SNHG16 with a NC targeting luciferase, an enzyme catalyzing insect luciferin.

Generation of clustered regularly interspaced short palindromic repeat (CRISPR) -Cas13d LV vector targeting human SNHG16
Guide RNA sequences targeting human SNHG16 were designed with a NC targeting mCherry, a monomeric red fluorescent protein.A 2.7 kp stuffer was removed from pAll-EF1a-CasRx-2A-EGFP, a RNA-targeting CRISPR-Cas13d LV vector (Biomedical Translation Research Center, Academia Sinica) with a GFP domain [24], by BsmBI for cloning of guide sequences and NC after an overnight ligation to create CRISPR-CasRX-SNHG16 and -NC, respectively.Sub-confluent 293T cells were transfected with CRISPR-CasRX-SNHG16 or -NC for 48 h at 37 °C.GFP-positive cells were sorted and examined by qRT-PCR analyses.Guide sequences 5ʹ-TTA GAG GAA CAA TTA GCA GCAGA-3ʹ targeting SNHG16 with the highest silencing efficacy and 5ʹ-CGC CGC CGT CCT CGA AGT TCATC-3′ targeting mCherry as a NC were chosen for further experiments in this study.

Construction of LV vectors carrying miR-146a
Recombinant LV vectors with the pre-miRNA expression construct containing pre-miR-146a or pre-miRNA scramble NC (System Biosciences, Palo Alto, CA, USA) were produced by transfecting sub-confluent 293T cells, along with packaging/envelope plasmids psPAX2/ pMD2.G under the calcium phosphate precipitation.The LV-miR-146a was chosen for further experiments based on the results of qRT-PCR analyses on the transfectants.LV-miR-146a or LV-miR-scrambled (LV-miR-scr, as a NC) vectors were harvested and concentrated by ultracentrifugation with the determined viral titers in TU.

Generation of LV-based shRNA targeting miR-146a
The shRNA sequences targeting miR-146a were designed.A 1.9 kb stuffer was removed from pLKO.1puro by AgeI and EcoRI for cloning shRNA sequences.To obtain recombinant LV vectors, the created pLKO.1sh-miR-146avectors were transfected into sub-confluent 293T cells, along with packaging psPAX2 and envelope pMD2.G plasmids by calcium phosphate precipitation to obtain LV-sh-miR-146a.Based on the results of RT-PCR analyses on miR-146a levels in LV-sh-miR-146a-transfected transfectants, the shRNA sequence sense 5ʹ-AGT GTC AGA CCT CTG AAA TTA-3ʹ, and antisense 5ʹ-TAA TTT CAG AGG TCT GAC ACT TTT TT-3ʹ was chosen for further experiments.LV-sh-luciferase was used as a NC.Stable LV-sh-miR-146-or LV-sh-luciferase-transfected transfectants were created under puromycin selection process in sorted CRISPR-CasRX-SNHG16-transfected HL-60 cells.
All mRNA levels were normalized to GAPDH by ΔCt method.For analyzing miR-146a and miR-17 levels, total RNAs were reverse transcribed by a TaqMan MicroRNA reverse transcriptase kit (Applied Biosystems) in Smart Cycler (Cepheid, Sunnyvale, CA, USA).Quantitative levels of miR-146a and miR-17 were analyzed with RNU6B small RNA (Applied Biosystems) as an endogenous control [24].The average levels of human HC cells or control tissues, neutrophils from healthy individuals, mouse cells or control tissues on day 0, and PBMs, mouse neutrophils, MLE-12 cells or HL-60 cells without stimulation were determined as 100%.

Dox-induced apoptosis in MLE-12 cells
MLE-12 cells were seeded with 1 × 10 6 cells/mL in 3.5 cm dish in the presence of different concentrations of Dox (TTY Biopharm, Taipei, Taiwan) for 24 h at 37 °C.After stimulation, cells were stained with PE-Annexin V (BD Pharmingen, San Diego, CA, USA) and 7-amino-actinomycin D (BD Pharmingen).Annexin V-positive and 7-amino-actinomycin D-negative cells were as apoptosis, and average apoptotic percentages without stimulation were as apoptotic cell ratio 1.0 [15,35].MLE-12 cells were also stained by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) detection cocktail (Promega, Madison, WI, USA) with cell nuclei counterstained by DAPI (Sigma-Aldrich), and observed under a confocal microscope.After stimulation, these cells were subjected to qRT-PCR analyses for expression of p53 and Bax, or treated by lysis buffer for immunoblot assay with primary antibodies against p53 (Santa Cruz, Santa Cruz, CA, USA) and Bax (Abcam, Cambridge, UK).Their culture supernatants were assessed for HMGB1 concentrations by ELISA (LSBio, Seattle, WA, USA).In addition, PBMCs were stained with PE-Annexin V and 7-aminoactinomycin D with average apoptotic percentages in HC as apoptotic cell ratio 1.0.

NETs formation in PBNs, HL-60 cells and mouse neutrophils
PBNs (5 × 10 5 cells/mL) were allowed to adhere to a poly-l-lysine (Sigma-Aldrich) coated 24-well plate in the presence 3 μg/mL LPS (Pseudomonas aeruginosa 10, Sigma-Aldrich), for 4 h at 37 °C.HL-60 cells were cultured with 10 6 cells/mL in 3.5 cm dish in the presence of 1.25% dimethyl sulfoxide (DMSO, Sigma-Aldrich) for 5 days at 37 °C to induce a differentiated status, dHL-60 cells.These cells were cultured in the presence of 500 ng/ mL LPS with serum-free X-VIVO 15 medium (Lonza, Basel, Switzerland) for 4 h at 37 °C.Mouse neutrophils (10 6 cells/mL) were cultured in a 3.5 cm dish for 30 min at 37 °C, while attached neutrophils were incubated under the same condition with different concentrations of HMGB1 (Atlantis Bioscience, Singapore, Republic of Singapore) or 3 μg/mL LPS for 4 h.After culture, aforementioned cells were stained with Sytox Green (Thermo Fisher Scientific) to detect DNAs under a fluorescence microscope.Their morphology was categorized into lobulated neutrophils, de-lobulated neutrophils, diffused NETs and spread NETs [24].Furthermore, these cells were subjected to qRT-PCR analyses.Their culture supernatants and cells lysates were measured by enzyme-linked immunosorbent assay (ELISA) for citrullinated histone 3 (CitH3) and protein arginine deiminase 4 (PAD4) concentrations (Cayman, Ann Arbor, MI, USA), respectively.

NETs formation in SNHG16-overexpressed and -silenced HL-60 cells
HL-60 cells (10 6 cells/mL) in 3.5 cm dish were transfected with SFFV or SFFV-SNHG16 for 48 h at 37 °C in the presence of polybrene.These cells were incubated with puromycin to select SNHG16-overexpressed transfectants.Furthermore, HL-60 cells were transfected with CRISPR-CasRX-NC or -SNHG16 for 48 h at 37 °C in the presence of polybrene, and GFP-positive cells were sorted to obtain SNHG16-silenced cells confirmed by qRT-PCR analyses.SNHG16-overexpressed or -silenced cells were cultured in the presence of 1.25% DMSO for 5 days to induce dHL-60 cells, and then stimulated with 500 ng/ mL LPS with serum-free X-VIVO 15 medium for 4 h at 37 °C.Such cells were stained with Sytox Green to detect DNA morphology under fluorescence microscopy.The culture supernatants and cells lysates were measured by ELISA for CitH3 and PAD4 concentrations, respectively.

LPS-induced autophagy in MLE-12 cells
MLE-12 cells were seeded with 1 × 10 6 cells/mL in 3.5 cm dish in the presence of different concentrations of LPS for 4 h at 37 °C.After stimulation, these cells were subjected to qRT-PCR analyses for LC3 and Beclin-1 expression.Alternatively, after overnight stimulation with 50 μg/mL LPS, MLE-12 cells treated by lysis buffer were subjected to immunoblot assay with primary antibodies against LC3 (Cell Signaling, Danvers, MA, USA) and Beclin-1 (Cell Signaling).MLE-12 cells treated with 1 µM rapamycin (Rapa, Sigma-Aldrich) were served as a PC.In addition, PBMs (5 × 10 5 cells/mL) were cultured in 24-well plate in the presence 50 μg/mL LPS for 4 h at 37 °C, and subjected to qRT-PCR analyses for LC3/Beclin-1 expression.

NETs formation in miR-146a-silenced HL-60 cells
Sorted CasRX-SNHG16-transfected HL-60 cells were transfected with sh-miR-146a or sh-luciferase and underwent puromycin selection to create stable transfectants.Selected transfectants were cultured in the presence of 1.25% DMSO for 5 days to induce dHL-60 cells, and further received 500 ng/mL LPS stimulation with serumfree X-VIVO 15 medium for 4 h at 37 °C.These cells were stained with Sytox Green to detect DNA morphology, and culture supernatants and cells lysates were measured for CitH3 and PAD4 concentrations, respectively.

Pristane-induced mouse AH or LN model
Female 8-week-old C57BL/6 JNarl mice received intraperitoneal injection of 0.5 mL pristane to induce AH, while their controls were injected with 0.5 mL PBS [15].They were sacrificed on day 0, 4, 9 and 14 to obtain the lungs and spleen.Their blood samples on day 14 were measured for RBC numbers, Hb levels and hematocrit (Hct) by a blood cell analyzer.Their serum samples on day 14 were examined for anti-RNP levels with a mouse anti-RNP ELISA kit (Alpha Diagnosis, San Antonio, TX, USA).
Female 8-week-old BALB/c JNarl mice from National Laboratory Animal Center, (Taipei, Taiwan), received intraperitoneal injection of 0.5 mL pristane and PBS to induce LN and serve as controls, respectively [35].They were sacrificed at month 0, 1, 3, 5 and 6 to obtain the kidneys.Urine specimens were collected for measuring proteinuria by test strips (Arkray, Edina, MN, USA) with results determined by a urine chemistry analyzer at month 0, 1, 3, 5 and 6.The serum samples were examined for anti-dsDNA and anti-RNP ELISA kits (Alpha Diagnosis) at month 0, 1, 3, 5 and 6.

Histopathological, TUNEL and immunofluorescence staining
Removed lung tissues were fixed in 10% buffered formalin overnight, and embedded in paraffin.Lung tissues were cut into 5 µm sections, and stained with hematoxylin and eosin (H&E).Paraffin-embedded sections were de-paraffinized in xylene, dehydrated in ethanol and rehydrated in distilled water.To determine glomerulonephritis (GN), mouse renal tissues were analyzed by Periodic acid-Schiff staining [35].For TUNEL staining, de-paraffinized mouse or human lung sections were treated by proteinase K to reactivate antigens, re-fixed by 4% formaldehyde, incubated with equilibrate buffer, and finally labelled by TUNEL detection cocktail [15].Cell nuclei were counterstained with DAPI.For detecting the expression of CitH3 or LC3, de-paraffinized mouse or human lung sections were stained with anti-CitH3 or anti-LC3 antibodies, followed by Alexa Fluor 488-conjugated antibodies (Thermo Fisher Scientific), while cell nuclei were counterstained with Hoechst 33258 [24].
Their fluorescence was detected by a confocal microscope.Cells with positive TUNEL, colocalized CitH3 with DNAs or cytoplasmic LC3 were determined by averaging the number from 3 fields of positively stained cells with the highest density in each section.

Statistical analyses
Data are expressed as the mean ± standard deviation (SD).Numerical data between patients and HC or between different patient groups were analyzed by Mann-Whitney U test.Correlation analysis was performed by Spearman correlation coefficient test.For comparing SNHG16, TLR4 and TRAF6 levels in PBMC or PBN from SLE and HC, and SNHG16, TLR4 and TRAF6 levels in PBMC or PBN from SLE and SLEDAI-2K, multivariable analysis adjusted for age/sex or plus medications were performed by SAS software 9.4 version (SAS Institute Inc, Cary, NC, USA).Hemorrhage frequencies in the lungs between different mouse groups were compared by Fisher's exact test.Differences in other analyses were determined by Student's t test.P values less than 0.05 were considered significant in this study with symbols as * for p < 0.05, **p < 0.01 and *** for p < 0.001.

Up-regulated SNHG16, TLR4 and TRAF6 expression with increased apoptosis, autophagy and NETs formation in SLE-AH lung tissues
Representative H&E-stained lung tissues from a PTX control patient and 3 AH patients with underlying SLE, IgAV or AAV were shown in Fig. 1a.The AH lungs were examined for the expression of apoptosis, autophagy and NETs formation.In Figs.1b, 2a, higher numbers of TUNEL-positive cells were found in lung tissues from SLE-AH patients than from PTX controls (40.3 ± 6.1 versus 1.4 ± 1.1, p = 0.036).In Fig. 1c, 2a, colocalized expression of CitH3 with DNAs, in favor of NETosis, was identified in the SLE-AH lungs but not in the PTX In particular, binding of ANCA to myeloperoxidase expressed on the membrane of pulmonary neutrophils could be responsible for distinct NETosis in AAV-associated AH lung tissues (Figs.1c, 2a) [36].In Fig. 1d, 2a, cytoplasmic LC3-positive cells, suggesting autophagy formation, were identified in SLE-AH but not in PTX (Fig. 2a, 7.7 ± 3.1 versus 0 ± 0, p = 0.018), IgAV-AH or AAV-AH lung tissues.
Collectively, these findings suggested up-regulated SNHG16 expression with a synchronously increased TLR4 and TRAF6 levels to enhance apoptosis, autophagy and NETs formation in the SLE-associated AH lungs.
Furthermore, there were significant differences by using multivariable analyses adjusted for age and sex for the comparison of SNHG16, TLR4 or TRAF6 levels between SLE and HC, while a significant positive correlation was found by using multivariable analyses adjusted for age, sex and medications for the comparison of SNHG16, TLR4 or TRAF6 levels with activity scores (Additional file 2: Table S2).
In addition, PBMC subpopulations were examined for the expression of SNHG16 in sorted CD3-positive T cells, CD19-positive B cells and CD14-positive monocytes from each healthy individual (Additional file 1: Fig. S2b).In comparison with the average expression levels of SNHG16 in neutrophils from 3 healthy individuals, there were no differences in T cells, whereas lower levels were found in monocytes and B cells (For neutrophils, 100.0 ± 2.0%, for T cells, 85.1 ± 14.0%, for monocytes, 36.3.0 ± 16.3%, p = 0.002, for B cells, 30.1 ± 6.4%, p = 0.002).We also examined the actual cell number of PB neutrophils, T cells, monocytes and B cells in each healthy individual (Additional file 1: Fig. S2b).There were higher circulating numbers of neutrophils, followed by T cells, monocytes and B cells in each healthy individual.
We further examined the expression of NEAT1, another lncRNA involved in the SLE activity and also a ceRNA targeting miR-146a [30], in PBMCs from SLE patients.Higher levels were found in SLE patients than in HC (Additional file 1: Fig. S4b, 253.8 ± 428.8% versus 100.0 ± 96.9%, p = 0.011).A positive correlation was found between NEAT1 levels and activity scores (Additional file 1: Fig. S4c, r = 0.306, p = 0.016).For NEAT1 expression in different patient groups and HC, AH had higher levels than HC (Additional file 1: Fig. S4d, p = 0.038), whereas no differences were found between SLE-AH and other AH, LN or Nil.Although both lncR-NAs had up-regulated expression in SLE patients and acted as ceRNAs targeting miR-146a, SNHG16 rather than NEAT1 appeared to be involved in the pathogenesis of AH manifestation.
Taken together, these results implicated that up-regulated SNHG16 levels with a concurrent increase in the expression of TLR4 and TRAF6 in PBMCs can participate in SLE activity, resulting in the AH manifestation.
In addition, there were significant differences by using multivariable analyses adjusted for age and sex for the comparison of SNHG16, TLR4 or TRAF6 levels between SLE and HC, while a significant positive correlation was found by using multivariable analyses adjusted for age, sex and medications for the comparison of SNHG16, TLR4 or TRAF6 levels with activity scores (Additional file 2: Table S3).
Furthermore, purified PBNs were stimulated with LPS to induce NETosis by observing DNA morphology with NETs formation and measuring CitH3 and PAD4 levels.For spread NETs formation, AH had higher percentages than LN, Nil or HC (Fig. 5l, p = 0.016 for LN, Nil or HC).For CitH3 or PAD4 production, AH had higher levels than LN, Nil or HC (Fig. 5m, for CitH3, p = 0.032 for LN, p = 0.016 for Nil or HC; for PAD4, p = 0.016 for LN, Nil or HC).
Altogether, increased cell death processes including apoptosis, autophagy and NETosis were found in PB leukocytes from SLE patients, particularly in those with the AH manifestation.
As a whole, these results suggested that modulating SNHG16 expression could control the Dox-inducedapoptosis via damaging DNAs to trigger a p53-dependent process in alveolar cells.

TLR4-mediated NETs formation in mouse neutrophils by LPS or HMGB1 stimulation
In PBNs from SLE-AH patients, there was up-regulated SNHG16 expression, while LPS stimulation could enhance NETosis as shown by increased spread NETs formation and CitH3/PAD4 production.On day 4, 9 and 14 after the pristane induction, neutrophils were isolated 24 h later from thioglycolate-injected mice.There was up-regulated expression of SNHG16, TLR4 and TRAF6 as well as down-regulated miR-146a levels since day 4 (Additional file 1: Fig. S9a, on day 4, for SNHG16, p < 0.001, for TLR4, p = 0.002, for TRAF6, p = 0.034, for miR-146a, p = 0.004), but no differences in the NEAT1 expression (Additional file 1: Fig. S4h).
In sum, our experimental data revealed that, in myelocytic cells, a TLR4/TRAF6 axis-mediated NETs formation could be regulated by SNHG16.
Parallelly increased expression of SNHG16 and TLR4 was found in human and mouse AH lung tissues.Downregulated TLR4 levels were shown in miR-146a-overexpressed MLE-12 cells as compared with control cells (Additional file 1: Fig. S8c, p < 0.001), while LV-mediated transfection of SNHG16 in such cells could reverse the targeting effect on TLR4 expression (Additional file 1: Fig. S8c, p = 0.001).These results indicated that SNHG16 could act as a ceRNA to upregulate the expression of TLR4, a proven target mRNA of miR-146a [29], in alveolar cells.
Interestingly, SNHG16 has been reported to facilitate cell autophagy via sponging miR-542-3p to upregulate ATG5, a molecule participating in the formation of autophagosome [40].A dose/time-dependent increase in ATG5 expression was shown in alveolar cell culture under LPS stimulation (Additional file 1: Fig. S11a), while up-regulated ATG5 levels were found in pristaneinduced AH lung tissues since day 4 (Additional file 1: Fig. S11b, day 4, p < 0.001).There were increased and decreased pulmonary ATG5 expression in SNHG16overexpressed and -silenced pristine-injected mice on day 14, respectively (Additional file 1: Fig. S11c, for overexpressed, p = 0.046, for silenced, p = 0.043), suggesting that, in addition to LC3 and Beclin-1, SNHG16 could regulate the autophagic process through modulating the ATG5 expression in the AH lungs.
Notably, except the lungs, the examinations after scarification in pristane-injected mice receiving intra-tracheal delivery of LV-SNHG16, sh-SNHG16 or their control vectors on day 14 revealed no morphological abnormalities in other visceral organs, indicating no evident extra-pulmonary global impact related to the intra-pulmonary administration of these LV vectors.
Figure 12 is a schematic summary of the latest evidence and aforementioned experimental results in this study.Under the specific intra-pulmonary cytokine milieu with increased IL-6 expression in the SLE-AH lungs, upregulated SNHG16 levels in alveolar cells and resident neutrophils can enhance TLR4 expression via stabilizing its mRNA expression and sponging miRNAs that can target Fig.11 AH in pristane-injected mice receiving intra-pulmonary sh-SNHG16 delivery on different days.Hemorrhagic frequencies on day 14 in pristane-injected mice receiving a day − 1 prophylactic, d day 4 early therapeutic and g day 8 late therapeutic intra-pulmonary delivery of sh-luciferase or sh-SNHG16.RBC numbers, Hb levels and Hct on day 14 in mice receiving b day − 1 prophylactic, e day 4 early therapeutic and h day 8 late therapeutic intra-pulmonary delivery of sh-luciferase or sh-SNHG16.Pulmonary SNHG16 levels on day 14 in mice receiving c day − 1 prophylactic, f day 4 early therapeutic and i day 8 late therapeutic intra-pulmonary delivery of sh-luciferase or sh-SNHG16.Values are mean ± SD.Horizontal lines are mean values.Mouse numbers per group, 16 in a, b, d, e, g, h, 8 in c, f, i it, e.g.miR-146a [27][28][29].TRAF6, a miR-146a target molecule, is a potent inducer of autophagy and NETosis [23,24].SNHG16 can regulate the TLR4/TRAF6 axis and activate downstream genes to generate pro-inflammatory cytokines like IL-8, a potent inducer of NETosis [24].Under higher levels of reactive oxygen species (ROS), the autophagic process is initiated through the inactivation of mTOR pathway [41].

Discussion
Based on the morphology, triggers and involved pathways, cell death processes can be classified into various categories such as apoptosis, autophagy and NETosis [42].Apoptosis, a programmed cell death causing cellular self-destruction, controls cell turnover, embryonic development and body growth to maintain homeostasis [43].Autophagy characterized by autophagosomes to digest unused components and damaged organelles through fusing with lysosomes, allows for recycling to provide nutrients to promote cell survival [44].NETosis, a regulated necrotic death pathway in neutrophils, is a specialized process with the formation of NETs [45].All of above death processes have been shown to be involved in the disease onset and flare of SLE [3].In particular, by using genome-wide association studies, autophagy-related genes like ATG5 were found to be associated with disease susceptibility in SLE, while risk alleles of variants locating in PRDM1-ATG5 gene region were positively correlated with ATG5 expression in B cells from lupus patients [46].In this study, AH patients had higher p53 levels, apoptotic cell ratios and Beclin-1 levels in their PBMCs than those from other patient groups.By LPS stimulation, in comparison with other patient groups, AH patients had higher Beclin-1 levels in PBMs and greater NETs percentages and CitH3/PAD4 concentrations in PBNs.These results indicated that such patients have more severe cell death in PB leukocytes than other patient groups in SLE.Furthermore, in the SLE-AH lungs, there were higher TUNEL-positive apoptotic cell numbers and up-regulated levels of p53/Bax, increased NETs formation with colocalized CitH3/DNAs and elevated PAD4 levels, and the presence of autophagy with cytoplasmic LC3-positive cells and up-regulated LC3/Beclin-1 expression.Altogether, the aforementioned findings implicate that, in active lupus patients, apoptosis, autophagy and NETosis take place in their lungs, providing nuclear autoantigens for IC formation with deposition in pulmonary capillary wall, ultimately resulting in capillaritis and AH.
Interestingly, distinct NETs formation was also observed in AAV-related AH lung tissues in this study.Unlike SLE, AAV is an autoimmune disease lacking IC formation, while ANCA binding to the membrane-bound myeloperoxidase on resident neutrophils can bring about intra-pulmonary NETosis in such patients [36].Similar to SLE, IgAV is another IC-mediated autoimmune disorder with pulmonary capillaritis and AH [47].Nevertheless, cell death processes were not identified in IgAV-AH lung tissues.Exogenous antigens derived from microorganisms or medications rather than endogenous nuclear autoantigens released from cell death are responsible for the IC formation in IgAV.Despite no available APS-AH lung tissues for evaluating cell death processes in this investigation, such a respiratory manifestation is known as a non-thrombotic inflammatory process unrelated to the IC-mediated pathogenesis [33].Moreover, there were no differences in SNHG16, TLR4 or TRAF6 levels in PBMCs between control subjects and non-SLE AH from AAV, IgAV and APS patients.Altogether, these results suggest complex molecular mechanisms involving in the immune-mediated AH with distinct dissimilarities among different underlying autoimmune diseases.
In SLE patients, elevated circulating levels of proinflammatory cytokines like TNF-α, IL-6 and IL-8 are positively correlated with the activity scores [48].Upregulated expression of specific lncRNAs has been demonstrated to be in parallel with SLE disease activity [49], while the expression of cellular lncRNAs is highly responsive to the surrounding cytokine milieu [50].Our earlier experiments revealed increased lincRNA-p21 expression in PBMCs positively correlating with the activity scores, and a dose-dependent up-regulated lincRNA-p21 expression in the culture of T cells under the stimulation of TNF-α [35].Furthermore, increased lincRNA-p21 levels with in situ apoptosis were identified in SLE-AH lung tissues [15].In this study, we identified higher levels of SNHG16 in PBMCs or PBNs with a positive correlation with disease activity in SLE, while active lupus patients had increased cell death processes in their PB leukocytes.Since a crucial pathogenic mechanism in SLE is accelerated cell death to release nuclear autoantigens for IC formation, upregulated SNHG16 expression in circulating leukocytes during the active disease appears to be involved in the death processes.In the presence of IL-6, there was a dose-dependent increase in SNHG16 expression in the cultured alveolar cells and neutrophils.Furthermore, up-regulated IL-6 and SNHG16 levels with increased in situ apoptosis, autophagy and NETosis were identified in both human and mouse AH lung tissues.Taken together, these data propose that, during the active disease in SLE patients, increased circulating levels of IL-6 can upregulate the expression of SNHG16 to enhance cell death processes.
The SLE pathogenesis-related innate immune receptors include endosomal TLR7/9 binding nucleic acids and cell membrane-bound TLR2/4 binding DAMP molecules like HMGB1, further contributing to the enhanced autoimmune responses [51,52].The contribution of membrane-bound TLR4 to the development of autoimmunity has been consistently demonstrated in transgenic lupus mouse models [53].Mice overexpressed with TLR4 were shown to generate the SLE-like autoimmunity [54], while upregulated TLR4 levels were found in pulmonary and splenic tissues from pristane-induced lupus mice in this study.In active SLE patients, there were increased circulating HMGB1 concentrations and upregulated TLR4 in PBMCs [52,55].In our investigation, SNHG16 and TLR4 expression in PB leukocytes were in a positive correlation, while their levels were higher in AH patients than other patient groups.Furthermore, there were increased pulmonary HGMB1 expression and upregulated SNHG16 and TLR4 levels in lung tissues from SLE-AH patients.A positive correlation existed between TLR4 and SNHG16 expression in PB leukocytes from SLE patients.Up-regulated SNHG16 expression and TLR4-mediated NETs formation were shown in mouse neutrophils under HMGB1 or LPS stimulation.There was enhanced LPS-induced autophagy in SNHG16-overexpressed and reduced autophagy in SNHG16-silenced alveolar cells.Increased LPS-induced NETosis was found in SNHG16-overexpressed myelocytic cells, while there was reduced NETosis in SNHG16-silenced cells.LVmediated transfection of SNHG16 in miR-146a overexpressed alveolar cells could reverse the targeting effect of miR-146a on TLR4 expression, indicating a ceRNA role of SNHG16 to upregulate TLR4 expression via targeting miR-146a.Intra-pulmonary overexpressing and silencing the expression of SNHG16 in pristane-induced AH mice could up-regulate and down-regulate TLR4 expression with enhanced and reduced cell death processes, respectively.Collectively, these discoveries display that, up-regulated pulmonary SNHG16 expression in active SLE can enhance TLR4-mediated autophagy and NETs formation, leading to the AH manifestation.
In vitro pristane-induced apoptosis in T cells could be blocked by inhibiting the activity of caspase 9 rather than caspase 8, implying the involvement of a p53related intrinsic apoptosis pathway [56].Furthermore, in pristane-induced C57BL/6 strain lpr mice with a Fas mutation, there were no reduced AH frequencies with preserved sensitivity of in vitro pristane-induced cell apoptosis, suggesting the independence of Fas-related extrinsic apoptosis pathway in this model [57].In this study, up-regulated SNHG16 and p53/Bax and downregulated miR-146a expression were demonstrated in alveolar cells stimulated with Dox, a DNA damage inducer to trigger the p53-dependent apoptosis [15], and in lung tissues of pristane-induced AH mice with enhanced in situ apoptosis.There were increased Doxinduced apoptosis in SNHG16-overexpressed cells with up-regulated p53/Bax and down-regulated miR-146a expression, and reduced apoptosis in SNHG16-silenced cells with down-regulated p53/Bax and up-regulated miR-146a expression.Furthermore, AH was enhanced by intra-pulmonary LV-SNHG16 delivery through increasing p53/Bax and down-regulating miR-146a expression to enhance apoptosis, while it was suppressed by sh-SNHG16 delivery through decreasing p53/Bax and up-regulating miR-146a levels to reduce apoptosis.These experimental results revealed that SNHG16 can regulate the p53-related intrinsic apoptosis pathway by sponging miR-146a to reverse its targeting effects on TAF9b, a p53 co-activator/stabilizer [24].Interestingly, TRAF6, a miR-146a target molecule, is able to induce the extrinsic apoptosis pathway by interacting and activating caspase 8 [22].Although the pristane-induced AH mouse model is independent of extrinsic apoptosis pathway, such an apoptotic process has been demonstrated to be involved in the pathogenesis of disease activity in SLE patients [3].Further studies should examine whether SNHG16 can modulate the TLR4-mediated apoptosis by using LPS to induce the extrinsic apoptotic process.
Clinically, AH can be classified into non-immune and immune-mediated etiologies, while the latter category includes post-transplantation, drug-induced vasculitis and autoimmune diseases [5].In particular, despite intensive respiratory support, SLE-associated AH has an average mortality rate around 40% based on accumulated statistics from large-scale series since 2010s [5,15].In addition to the routine use of high-dose corticosteroids or plus pulse methylprednisolone, current medications such as cyclophosphamide and intravenous immunoglobulin are associated with unsatisfying responses in SLE-AH patients [58].In spite of an extensive application in refractory cases, therapeutic plasmapheresis lacks a beneficent impact on the overall fatality [59].An improved survival has been observed in applying rituximab (RTX), a B-cell depleting antibody, to reduce the formation of IC [60].Nevertheless, this biologic agent can harm antiviral humoral immunity and cause a difficulty in cleaning the invasion of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) [61].RTX use in SLE has been reported to be associated with poor vaccination efficacy, prolong hospitalization and death outcomes during the pandemic SARS-CoV-2 infection [62].Indeed, there is an imperative need to develop effective and secure therapeutics in controlling the disease activity to avoid mortality in SLE-associated AH.
Current investigation on the inhibition of NETosis pathway or NETs degrading drugs can provide potential therapeutic approach for life-threatening complications in autoimmune diseases, such as the AH manifestation in SLE patients [63].Available SLE medications including cyclosporin, hydroxychloroquine and tacrolimus can target NETs components or interfere with the formation of NETs; however, there are no recently approved anti-NETs drugs to treat lupus patients [64].The anti-NETs approach has demonstrated therapeutic efficacy in animal models of SLE.In addition to sh-SNHG16 intra-tracheal delivery in our investigation, DNase-1 inhalation to reduce NETs could suppress AH manifestation in the pristaneinduced mouse model, while the injection of Cl-amidine, a PADs inhibitor, protected MRL/lpr lupus mice from NETs-mediated visceral injury [63].Nevertheless, considering the risk of systemic infections in NETs-depleted individuals, further studies are necessary for the development of effective clinical compounds able to safely treat such patients [63,64].
In this study, intra-pulmonary delivery of shRNA targeting SNHG16 could reduce AH through inhibiting TLR4-mediated autophagy and NETs formation in the lungs of a mouse model.By intra-tracheal infusion of LV-sh-SNHG16, lower pulmonary SNHG16 levels were observed without differences in splenic expression, indicating no extra-pulmonary leakage of infused LV vectors.Besides, there were no undesired effects outside the lungs as shown by no morphological abnormalities in other visceral organs after the scarification of mice.Although safety always is a crucial issue for applying LV vectors into clinical practice, the intra-pulmonary delivery route can evade the adverse impacts on non-target organs through the systemic administration.Infusion of activated recombinant factor VII via an intra-pulmonary route by nebulizer or bronchoscopy has been shown to complete hemostasis and improve AH without adverse effects in SLE-AH patients [65,66].Interestingly, RNA therapeutics are emerging treatment modalities with 17 approved clinical products, including 2 macromolecular mRNA drugs and 15 oligonucleotide drugs with one aptamer, 4 siR-NAs and 10 antisense oligonucleotides [67].In particular, despite not through the intra-pulmonary administration, there are ongoing siRNA therapeutic trials focusing on lung malignancies with favorable development prospects [67,68].Notably, our experimental results implicated a therapeutic potential in the SLE-associated AH lungs by intra-pulmonary infusion of shRNA targeting SNHG16 to reduce in situ cell death processes.

Conclusions
Our experimental results demonstrate firstly that lncRNA SNHG16 regulates TLR4-mediated autophagy and NETs formation in the human and mouse AH lungs, and provide a potential of intra-pulmonary delivery of shRNA targeting SNHG16 to treat such a lethal manifestation in SLE patients.S1.Clinical profiles and therapeutic modality of AH in different patient groups.Table S2.Linear regression for SNHG16, TLR4 and TRAF6 levels in PBMCs from SLE and HC.Table S3.Linear regression for SNHG16, TLR4 and TRAF6 levels in PBNs from SLE and HC.

Fig. 1
Fig. 1 Increased apoptosis, autophagy and NETs formation in SLE-associated AH lung tissues.a From top to low, representative histopathology from a PTX control and AH patients from SLE, IgAV and AAV.In PTX, normal alveoli.In SLE, AH with blood in alveoli.In IgAV, AH with interstitial lymphoplasmatic cells and fibrosis.In AAV, AH with fibroblastic plugs in alveoli and alveolar ducts.H&E staining, scale bar = 100 µm and magnification ×100.b From top to low, representative TUNEL IF staining (green) from a PTX control and AH patients from SLE, IgAV and AAV.Cell nuclei counterstained with DAPI (blue).Scale bar = 25 µm, magnification ×400.c From top to low, representative CitH3 IF staining (green) from a PTX control and AH patients from SLE, IgAV and AAV.Cell nuclei counterstained with Hoechst 33258 (blue).Scale bar = 12.5 µm, magnification ×800.d From top to low, representative LC3 IF staining (green) from a PTX control and AH patients from SLE, IgAV and AAV.Arrows pointing cells with positive cytoplasmic LC3 staining.Cell nuclei counterstained with Hoechst 33258 (blue).Scale bar = 10 µm, magnification ×1000 for SLE.Scale bar = 12.5 µm, magnification ×800 for PTX, IgAV and AAV

Fig. 2
Fig. 2 Up-regulated cell death processes and pulmonary expression of SNHG16, TLR4 and TRAF6 in SLE-associated AH lung tissues.a Quantification of cell numbers with positive TUNEL, colocalized CitH3 and DNAs, and cytoplasmic LC3 in lung tissues from PTX controls and AH patients from SLE, IgAV and AAV.Expression levels of b SNHG16, c TLR4, d TRAF6, e miR-146a and f NEAT1 in lung tissues from PTX controls and AH patients from SLE, IgAV and AAV.Expression levels of g p53, h Bax, i HGMB1, j PAD4, k LC3, l Beclin-1, m mTOR, n p62, o IL-6, p IL-8, q IFN-α and r MX-1 in lung tissues from PTX controls and SLE-AH patients.Lung sample numbers, n = 5 for PTX, n = 3 for SLE, n = 1 for IgAV, n = 1 for AAV.Values are mean ± SD.Horizontal lines are mean values.*p < 0.05

Fig. 3
Fig. 3 Up-regulated SNHG16, TLR4 and TRAF6 expression in PBMCs from SLE-AH patients.a SNHG16, e TLR4 and i TRAF6 levels in PBMCs from HC and SLE patients.A positive correlation between PBMC b SNHG16, f TLR4 and j TRAF6 levels and SLEDAI-2K activity scores.c SNHG16, g TLR4 and k TRAF6 levels in PBMCs from HC, Nil, LN, SLE-AH and other AH.A positive correlation between SNHG16 and d miR-146a, h TLR4 or l TRAF6 levels in PBMCs from SLE patients.Values are mean ± SD.Horizontal lines are mean values.Patient numbers, n = 62 for SLE, n = 7 for Nil, LN, SLE-AH, n = 6 for other AH.*p < 0.05, **p < 0.01, ***p < 0.001

Fig. 4
Fig. 4 Increased autophagy formation in PBMCs from SLE-AH patients.a LC3, d Beclin-1, g mTOR and j p62 levels in PBMCs from SLE patients and HC.A positive correlation between b LC3 or e Beclin-1 levels in PBMCs from SLE patients and SLEDAI-2K activity scores.No correlation between h mTOR levels in PBMCs from SLE patients and SLEDAI-2K activity scores.c LC3, f Beclin-1 and i mTOR levels in PBMCs from HC, Nil, LN and AH patients.k LC3 and l Beclin-1 levels in LPS-stimulated PBMs from HC, Nil, LN and AH patients.Values are mean ± SD.Horizontal lines are mean values.Patient numbers, PBMCs, n = 62 for SLE, n = 7 for Nil, LN, AH.Patient numbers, PBMs, n = 5 for Nil, LN, n = 3 for AH.*p < 0.05, **p < 0.01, ***p < 0.001

Fig. 5
Fig. 5 Up-regulated SNHG16, TLR4 and TRAF6 expression and increased NETs formation in PBNs from SLE-AH patients.a SNHG16, d TLR4 and g TRAF6 levels in PBNs from SLE patients and HC.b SNHG16, e TLR4 and h TRAF6 levels in PBNs from HC, Nil, LN and AH patients.A positive correlation between c SNHG16, f TLR4 and i TRAF6 levels in PBNs from SLE patients and SLEDAI-2K activity scores.A positive correlation between SNHG16 and j TLR4 or k TRAF6 levels in PBNs from SLE patients.PBNs from HC, Nil, LN and AH patients stimulated with LPS to detect DNAs morphology and measure CitH3/PAD4 levels.l, Left, representative photographs from a HC and an AH patient.Scale bar = 50 µm, magnification ×200.Right, quantification of NETs formation in HC, Nil, LN and AH patients.m Quantification of CitH3 (left) and PAD4 (right) levels in HC, Nil, LN and AH patients.Values are mean ± SD.Horizontal lines are mean values.Patient numbers, n = 15 for SLE, n = 5 for Nil, n = 5 for LN, n = 4 for AH.*p < 0.05, **p < 0.01

Fig. 9
Fig. 9 AH enhanced by intra-pulmonary SNHG16 delivery through increasing cell death processes in pristane-injected mice.a Left, representative gross and histopathological photographs in the mouse lungs with no, partial and complete hemorrhage.Scale bar = 100 µm, magnification ×100.Right, hemorrhagic frequencies in pristane-injected mice receiving intra-pulmonary delivery of SFFV or SFFV-SNHG16 on day 14.b RBC numbers, Hb levels and Hct on day 14 in mice receiving delivery of SFFV or SFFV-SNHG16.c Anti-RNP titers on day 14 in serum samples from SFFV-or SFFV-SNHG16-treated mice.d Left, pulmonary and right, splenic SNHG16 levels on day 14 from mice receiving delivery of SFFV or SFFV-SNHG16.e Left, representative TUNEL IF staining (green) in lung tissues from SFFV-or SFFV-SNHG16-treated mice.Scale bar = 20 µm, magnification ×400.Right, quantification of TUNEL-positive cell numbers in lung tissues.f Immunoblot assay (left) with signal intensity quantitation analysis (right) of pulmonary CitH3 and PAD4 expression from mice receiving delivery of SFFV or SFFV-SNHG16 on day 14.g Left, representative CitH3 IF staining (green) with cell nuclei counterstained with Hoechst 33258 (blue).Scale bar = 12.5 µm, magnification ×800.Right, quantification of cell numbers with positive colocalized CitH3 and DNAs in lung tissues.h Representative LC3 IF staining (green) with cell nuclei counterstained with Hoechst 33258 (blue).Scale bar = 10 µm, magnification ×1000.Right, quantification of cell numbers with positive cytoplasmic LC3 in lung tissues.i From left to right, pulmonary TLR4, TRAF6, miR-146a, IL-6, IL-8, IFN-α, MX-1, p53, Bax, LC3 and Beclin-1 levels on day 0, 4, 9 and 14 from mice receiving delivery of SFFV or SFFV-SNHG16.Values are mean ± SD.Horizontal lines are mean values.Mouse numbers per group, 16 in a, b, 8 in c, d, e, g, h, i, 4 in f.All results in figure were representative of 2 independent experiments with similar findings.*p < 0.05, **p < 0.01, ***p < 0.001

Fig. 10
Fig. 10 AH suppressed by intra-pulmonary prophylactic sh-SNHG16 delivery through reducing cell death processes in pristane-injected mice.a Left, representative gross and histopathological photographs in the mouse lungs with no, partial and complete hemorrhage.Scale bar = 100 µm, magnification ×100.Right, hemorrhagic frequencies on day 14 in pristane-injected mice receiving intra-pulmonary delivery of sh-luciferase or sh-SNHG16.b RBC numbers, Hb levels and Hct on day 14 in mice receiving delivery of sh-luciferase or sh-SNHG16.c Anti-RNP titers on day 14 in serum samples from mice receiving delivery of sh-luciferase or sh-SNHG16.d Pulmonary (left) and splenic (right) SNHG16 levels on day 14 from mice receiving delivery of sh-luciferase or sh-SNHG16.e Left, representative TUNEL IF staining (green) in lung tissues from sh-luciferase-and sh-SNHG16-treated mice.Scale bar = 25 µm, magnification ×400.Right, quantification of TUNEL-positive cell numbers in lung tissues.f Left, representative CitH3 IF staining (green) with cell nuclei counterstained with Hoechst 33258 (blue).Scale bar = 12.5 µm, magnification ×800.Right, quantification of cell numbers with positive colocalized CitH3 and DNAs in lung tissues.g Immunoblot assay (left) with signal intensity quantitation analysis (right) of pulmonary CitH3 and PAD4 expression on day 14 from mice receiving delivery of sh-luciferase or sh-SNHG16.h Left, representative LC3 IF staining (green) with cell nuclei counterstained with Hoechst 33258 (blue).Scale bar = 10 µm, magnification ×1000.Right, quantification of cell numbers with positive cytoplasmic LC3 in lung tissues.i From left to right, pulmonary TLR4, TRAF6, miR-146a, IL-6, IL-8, IFN-α, MX-1, p53.Bax, LC3 and Beclin-1 levels on day 0, 4, 9 and 14 from mice receiving delivery of sh-luciferase or sh-SNHG16.Values are mean ± SD.Horizontal lines are mean values.Mouse numbers per group, 16 in a, b, 8 in c, d, e, f, h, i, 4 in g.All results in figure were representative of 2 independent experiments with similar findings.*p < 0.05, **p < 0.01, ***p < 0.001

.
Fig.11AH in pristane-injected mice receiving intra-pulmonary sh-SNHG16 delivery on different days.Hemorrhagic frequencies on day 14 in pristane-injected mice receiving a day − 1 prophylactic, d day 4 early therapeutic and g day 8 late therapeutic intra-pulmonary delivery of sh-luciferase or sh-SNHG16.RBC numbers, Hb levels and Hct on day 14 in mice receiving b day − 1 prophylactic, e day 4 early therapeutic and h day 8 late therapeutic intra-pulmonary delivery of sh-luciferase or sh-SNHG16.Pulmonary SNHG16 levels on day 14 in mice receiving c day − 1 prophylactic, f day 4 early therapeutic and i day 8 late therapeutic intra-pulmonary delivery of sh-luciferase or sh-SNHG16.Values are mean ± SD.Horizontal lines are mean values.Mouse numbers per group,16 in a, b, d, e, g, h, 8 in c, f, i.All results in figure were representative of 2 independent experiments with similar findings.*p < 0.05, **p < 0.01, ***p < 0.001

Fig. 12 A
Fig.12A schematic summary of the latest evidence and experimental results in this study.Under the specific intra-pulmonary cytokine milieu with increased IL-6 expression in SLE-AH lung tissues, upregulated SNHG16 levels in alveolar cells and resident neutrophils can enhance TLR4 expression via stabilizing its mRNA expression and sponging miRNAs that can target it, e.g.miR-146a.TRAF6, a miR-146a target molecule, is a potent inducer of autophagy and NETosis.TRAF6, a miR-146a target molecule, is an inducer of autophagy and NETosis.SNHG16 can regulate the TLR4/ TRAF6 axis and activate downstream genes to generate pro-inflammatory cytokines like IL-8, a potent inducer of NETosis.Under high ROS levels, the autophagic process is initiated through the inactivation of mTOR pathway.Figure is created with BioRender.com

B
cells healthy individual No. 1, No. 2 and No. 3. Values are mean ± SD.Fig. S3.Expression of miR-146a in PBMCs and PBNs from SLE patients.a MiR-146a levels in PBMCs from SLE patients and HC.b A negative correlation between miR-146a levels in PBMCs and SLEDAI-2K activity scores.c MiR-146a levels in PBMCs from HC, Nil.LN, SLE-AH and other AH patients.A negative correlation between miR-146a and d TLR4, e TRAF6 and f NEAT1 levels in PBMCs from SLE patients.g MiR-146a levels in PBNs from SLE patients and HC.h A negative correlation between miR-146a levels in PBNs and SLEDAI-2K activity scores.i MiR-146a levels in PBNs from HC, Nil.LN and AH patients.Values are mean ± SD.Horizontal lines are mean values.Patient numbers, n = 62 for PBMCs from SLE, 15 for PBNs from SLE, 7 for PBMCs from Nil, LN and SLE-AH, 6 for PBMCs from other AH, 5 for PBNs from Nil, LN, 4 for PBN from AH. *p < 0.05.**p < 0.01, ***p < 0.001.Fig. S4.Expression of NEAT1 in SLE patients and pristane-injected mice.NEAT1 levels in a PBMCs and e PBNs from SLE patients and HC.b A positive correlation and f no correlation between NEAT1 levels in PBMCs and PBNs from SLE patients, respectively, and SLEDAI-2K activity scores.d No correlation between NEAT1 and TLR4 levels in PBMCs from SLE patients.NEAT1 levels in c PBMCs and g PBNs from HC, Nil, LN and AH patients.h NEAT1 levels in thioglycolate-induced mouse neutrophils from pristane-induced AH mice.NEAT1 levels in i lung and j spleen tissues from pristane-induced AH mice.Values are mean ± SD.Horizontal lines are mean values.Patient numbers, n = 62 for PBMCs from SLE, 15 for PBNs from SLE, 7 for PBMCs from Nil, LN, SLE-AH, 6 for PBMCs from other AH, 5 for PBNS from Nil, LN, 4 for PBNs from AH. 4 mice per group in h. 5 mice per group in i, j.All results in Fig. S4h, S4i and S4j were representative of 2 independent experiments with similar findings.* p < 0.05.Fig. S5.p53 levels and apoptotic cell ratios in PBMCs from SLE patients.a Left, p53 levels in SLE patients and HC.Right, p53 levels in HC, Nil, LN and AH.b Left, apoptotic cell ratios in SLE patients and HC.Right, apoptotic cell ratios in HC, Nil, LN and AH.Values are mean ± SD.Horizontal lines are mean values.Patient numbers, n = 30 for SLE, n = 5 for Nil, n = 5 for LN, n = 3 for AH.*p < 0.05, **p < 0.01, ***p < 0.001.Fig. S6.Increased autophagy formation by immunoblot assay in PBMCs from SLE patients.a Representative immunoblot assay for Beclin-1, LC and mTOR in PBMCs from SLE patients and HC.b Signal intensity quantitation analysis for Beclin-1, LC and mTOR in PBMCs from SLE patients and HC.Values are mean ± SD.Patient numbers, n = 3 for representative immunoblot assay, n = 10 for signal intensity quantitation analysis.* p < 0.05.Fig. S7.Apoptosis formation in Dox-stimulated MLE-12 cells regulated by SNHG16.a SNHG16 expression in MLE-12 cells stimulated with various concentrations of IL-6.b Left, representative photographs of TUNEL IF staining (green) in MLE-12 cells stimulated with 1 μM Dox.Cell nuclei counterstained with DAPI (blue).Scale bar = 10 µm, magnification ×1000.Middle, quantification of TUNEL-positive cell percentages in mock and stimulation with 1 μM Dox.Right, apoptotic cell ratios in MLE-12 cells stimulated with various concentration of Dox.c HMGB1 levels in culture supernatants of MLE-12 cells stimulated with various concentration of Dox.d From left to right, expression of SNHG16, TRAF6, p53, Bax and miR-146a levels by qRT-PCR analyses in MLE-12 cells stimulated with various concentration of Dox. e Representative immunoblot assay of p53/Bax (top) and TRAF6 (low) expression in MLE-12 cells stimulated with various concentration of Dox.f MiR-17 and miR-146a levels in SNHG16-overexpressed and -silenced MLE-12 cells.g Left, SNHG16 expression in SNHG16-overexpressed MLE-12 transfectants.Right, from left to right, levels of p53, Bax and SNHG16 and apoptotic cell ratios in 1 μM Dox-stimulated SNHG16-overexpressed MLE-12 transfectants.h Left, SNHG16 expression in SNHG16-silenced MLE-12 transfectants.Right, from left to right, levels of p53, Bax and SNHG16 and apoptotic cell ratios in 1 μM Dox-stimulated SNHG16silenced MLE-12 transfectants.Values are mean ± SD. Results in Fig. S7a to S7f were representative of 3 independent experiments, and in Fig. S7g and S7h were representative of 2 independent experiments with similar findings.*p < 0.05, **p < 0.01, ***p < 0.001.Fig. S8.TLR4 expression in SNHG16-overexpressed/silenced and miR-146a-overexpressed MLE-12 cells.a TLR4 mRNA levels in SNHG16-overexpressed (left) and -silenced MLE-12 transfectants (right).b Representative immunoblot assay of TLR4 levels in SNHG16-overexpressed (left) and -silenced MLE-12 transfectants (right).c Left, miR-146a levels in miR-146a-overexpressed MLE-12 transfectants.Right, TLR4 levels in miR-146a-overexpressed MLE-12 transfectants and in such cells overexpressed with SNHG16.Values are mean ± SD. Results in Fig. S8a and S8b were representative of 3 independent experiments, and in Fig. S8c were representative of 2 independent experiments with similar findings.**p < 0.01, ***p < 0.001.Fig. S9.Involvement of SNHG16 in TLR4-mediated NETs formation in mouse neutrophils.a Expression of SNHG16, TLR4, TRAF6 and miR-146a in thioglycolate-induced neutrophils from pristane-injected mice on day 4, day 9 and day 14.b SNHG16 expression in naïve mouse neutrophils stimulated with various concentrations of IL-6 or 3 μg/mL LPS.c Naïve mouse neutrophils stimulated with various concentrations of HMGB1 or LPS.Left, representative photographs of NETs morphology from naïve mouse neutrophils under 300 ng/mL HMGB1 or 3 μg/mL LPS stimulation.Scale bar = 30 µm, magnification ×400.Middle, quantification of NETs morphology with diffuse/spread NETs percentages, Right, CitH3 concentrations, SNHG16, TRAF6 and miR-146a levels in naïve mouse neutrophils stimulated with various concentrations of HMGB1 or 3 μg/mL LPS.Values are mean ± SD. 4 mice per group in a.All results in Fig. S9 were representative of 2 independent experiments with similar findings.**p < 0.01, ***p < 0.001.Fig. S10.TLR4 expression in SNHG16-overexpressed/silenced HL-60 cells and reversed reduction in NETosis in miR-146a-silenced CasRX-SNHG16-transfected HL-60 cells.a TLR4 mRNA levels inSNHG16-overexpressed (left) and -silenced HL-60 cells (right).b Representative immunoblot assay of TLR4 levels in SNHG16-overexpressed (left) and -silenced HL-60 cells (right).c MiR-17 and miR-146a levels in SNHG16-overexpressed and -silenced HL-60 cells.d Left, miR-146a levels in miR-146a-silenced CasRX-SNHG16-transfected HL-60 cells.Right, quantification of NETs formation percentages (left), CitH3 levels (middle), and PAD4 levels (right) in miR-146a-silenced CasRx-SNGH16-transfected dHL-60 transfectants stimulated with 500 ng/mL LPS for 4 h.Values are mean ± SD. Results in Fig. S10a to S10c were representative of 3 independent experiments, and in Fig. S10d were representative of 2 independent experiments with similar findings.*p < 0.05, **p < 0.01, ***p < 0.001.Fig. S11.ATG5 expression in LPS-stimulated MLE-12 cells and lung tissues from pristane-induced AH mice.a ATG5 levels in MLE-12 cells under the stimulation of different LPS concentrations for 4 h (left) and 50 µg/mL LPS for different time (right).b ATG5 levels in lung tissues from pristane-induced or saline-injected mice.c ATG5 levels in lung tissues from pristane-induced mice receiving SFFV/SFFV-SNHG16 or sh-luciferase/ sh-SNHG16 intra-pulmonary delivery.Values are mean ± SD. 5 mice per group in b. 8 mice per group in c. Results in Fig. S11a were representative of 3 independent experiments, and in Fig. S11b and S11c were representative of 2 independent experiments with similar findings.*p < 0.05, **p < 0.01, ***p < 0.001.Fig. S12.SNHG16 and NEAT1 expression in PBMCs, USCs and biopsied renal tissues from LN patients.a SNHG16 and d NEAT1 levels in PBMCs from HC, Nil and LN patients.b SNHG16 and e NEAT1 levels in USCs from HC, Nil and LN patients.c SNHG16 and f NEAT1 levels in kidney tissues from control and LN.Values are mean ± SD.Horizontal lines are mean values.Patient numbers, for PBMC and USC, n = 15 for Nil, n = 15 for LN, for renal tissue, n = 5 for control, n = 5 for LN.*p < 0.05, ***p < 0.001.Fig. S13.SNHG16 expression in a mouse LN model.a Serial measurement of proteinuria levels in mice at month 0, 1, 3, 5 and 6.Serial measurement of anti-RNP.b and anti-dsDNA titers.c at month 0, 1, 3, 5 and 6. d Periodic acid-Schiff staining of renal glomeruli at month 6 after saline (left) or pristane injection (right).Arrows indicating normal glomeruli (left) or a glomerulus with GN formation (right).Scale bar = 10 µm, magnification ×400.e Kinetic expression of SNHG16 in the kidneys from saline-and pristane-injected mice at month 0, 1, 3, 5 and 6.Values are mean ± SD.All results in Fig. S13 were representative of 2 independent experiments with similar findings.5 mice per group in Fig. S11.*p < 0.05, **p < 0.01, ***p < 0.001.Additional file 2: Table