Abstract

   Acute lung injury is a serious complication of sepsis with high
   morbidity and mortality. Pyroptosis is a proinflammatory form of
   programmed cell death that leads to immune dysregulation and organ
   dysfunction during sepsis. We previously found that adenosine deaminase
   acting on double-stranded RNA 1 (ADAR1) plays regulatory roles in the
   pathology of sepsis, but the mechanism of ADAR1 in sepsis-induced
   pyroptosis and lung injury remains unclear. Here, we mainly
   investigated the regulatory effects and underlying mechanism of ADAR1
   in sepsis-induced lung injury and pyroptosis of pulmonary macrophages
   through RNA sequencing of clinical samples, caecal ligation and
   puncture (CLP)-induced septic mouse models, and in vitro cellular
   experiments using RAW264.7 cells with lipopolysaccharide (LPS)
   stimulation. The results showed that pyroptosis was activated in
   peripheral blood mononuclear cells (PBMCs) from patients with sepsis.
   In the CLP-induced septic mouse model, pyroptosis was mainly activated
   in pulmonary macrophages. LPS-stimulated RAW264.7 cells showed
   significantly increased activation of the NLRP3 inflammasome. ADAR1 was
   downregulated in PMBCs of patients with sepsis, and overexpression of
   ADAR1 alleviated CLP-induced lung injury and NLRP3 inflammasome
   activation. Mechanistically, the regulatory effects of ADAR1 on
   macrophage pyroptosis were mediated by the miR-21/A20/NLRP3 signalling
   cascade. ADAR1 attenuated sepsis-induced lung injury and hindered the
   activation of pyroptosis in pulmonary macrophages in sepsis through the
   miR-21/A20/NLRP3 axis. Our study highlights the role of ADAR1 in
   protecting pulmonary macrophages against pyroptosis and suggests
   targeting ADAR1/miR-21 signalling as a therapeutic opportunity in
   sepsis-related lung injury.

   Keywords: sepsis, macrophage, ADAR1, A20, pyroptosis, lung injury

Introduction

   Sepsis is acknowledged as one of the main causes of death in the
   emergency room, with high incidence and mortality rates globally over
   many years [47]^1. Sepsis is a multifaceted disease with evolving
   definitions. The latest study states that sepsis is a dysregulated host
   response to infection that can result in life-threatening organ failure
   [48]^2. Sepsis is characterized by increased inflammation as well as
   immune suppression, and the imbalance between them leads to cellular
   dysfunction and even multiple organ failure. Clinically, six organ
   systems, the cardiovascular, respiratory, renal, neurological,
   haematological, and hepatic systems, are frequently examined in
   patients with sepsis [49]^3. Among them, sepsis-induced pulmonary
   dysfunction, which results in tachypnoea and acute respiratory distress
   syndrome (ARDS), is intimately associated with hypoxia and metabolic
   acidosis [50]^2. ARDS is a serious complication of sepsis with high
   morbidity and mortality [51]^4. Although the application of
   lung-protective ventilation has improved the prognosis of ARDS,
   additional research examining the mechanism and pathogenesis of
   sepsis-related acute lung injury (ALI) is also important to develop
   more efficient and novel approaches for ALI treatment [52]^5.

   Numerous cell death processes, such as apoptosis, necrosis, autophagy,
   and pyroptosis, are always activated as sepsis progresses [53]^3.
   Pyroptosis is an inflammatory programmed cell death that is mediated by
   caspases-1/4/5/11 [54]^6^, [55]^7. The Nod-like receptor (NLR) family
   pyrin domain containing 3 (NLRP3) inflammasome mediates caspase-1
   activation and the secretion of the proinflammatory cytokines
   interleukin (IL)-1β and IL-18 [56]^8^-[57]^10. Inflammatory caspases
   cleave the gasdermin D (GSDMD) substrate, which is important for pore
   formation on cell membranes, cytokine release, and ultimately
   pyroptosis [58]^11. Bacterial infection is a crucial incentive for the
   assembly and activation of the inflammasome in the host response
   [59]^12. Pyroptosis is needed for defence against bacterial infection
   to minimize tissue damage during sepsis. However, overactivated
   pyroptosis can result in septic shock, multiple organ dysfunction
   syndrome, or an increased risk of secondary infection [60]^8. Studies
   have revealed that pannexin-1 and P2X7 signalling triggers pyroptosis
   of bone marrow-derived macrophages through LPS-induced activation of
   caspase-11, suggesting possible therapeutic options for patients with
   bacterial sepsis [61]^13. Additionally, Xue et al. found that miR-21
   mediated nuclear factor kappa B (NF-κB) signalling and protein
   A20-mediated NLRP3 inflammasome-regulated caspase-1 activation, both of
   which were positive regulators of LPS-induced sepsis and pyroptosis
   [62]^14. A prospective cohort study revealed increased peripheral blood
   mononuclear cell (PBMC) pyroptosis in patients with sepsis, which was
   related to disease severity and patient mortality; however, the
   mechanisms of pyroptosis in sepsis and its related organ disorders are
   complex, so in-depth exploration of the regulatory molecules and
   mechanisms of septic pyroptosis is important to develop targeted
   interventions and improve clinical outcomes [63]^15^, [64]^16.

   The regulation of cell death is greatly influenced by RNA modification
   [65]^17. The RNA editing enzyme known as adenosine deaminase acting on
   double-stranded RNA 1 (ADAR1) is widely known for converting adenosine
   residues into inosine (A-to-I) in double-stranded RNAs [66]^18. Studies
   have revealed that ADAR1 regulates innate immunity and cell death
   mechanisms such as pyroptosis, apoptosis, necrosis, and PANoptosis
   [67]^19^, [68]^20. These processes are crucial for host survival,
   development, and immunity against infections. Additionally, ADAR1 can
   modify the precursors of miRNA as well as the mature miRNA sequence
   during the production of microRNAs, which suppresses the expression of
   downstream target genes [69]^21. Our previous study demonstrated that
   ADAR1 regulated macrophage polarization by binding to miR-21 precursors
   and editing miR-21 biogenesis [70]^22. ADAR1 was also found to
   alleviate sepsis-induced inflammation and intestinal injury in mice by
   interfering with miR-30a synthesis and altering the miR-30a/SOCS3 axis
   [71]^23. Collectively, ADAR1 regulates the development of sepsis,
   although it is still unknown whether ADAR1 has an impact on the organ
   failure and cell death pathways induced by sepsis. Through clinical
   samples, animal models, and in vitro experiments, we investigated the
   regulatory effects and underlying mechanism of ADAR1 on sepsis-induced
   lung injury and pulmonary macrophage pyroptosis, offering new insights
   and fundamental evidence for the therapeutic intervention of
   sepsis-related lung injury and diseases.

Methods

Ethic statement

   All procedures during the study were approved by the Ethical Committee
   of Xijing Hospital of the Fourth Military Medical University (approved
   number: KY20212172-C-1). The human study complied with the principles
   of the Helsinki Declaration, and all participants signed informed
   consent forms. The experiments relating to mouse models were performed
   according to the Guide for the Care and Use of Laboratory Animals and
   approved by the Institutional Animal Care and Use Committee of the
   Fourth Military Medical University (approved number: 20190305).

Patients and PBMC sample collection

   We mainly enrolled patients with sepsis who presented to the emergency
   department and intensive care unit in Xijing Hospital from 07.2021 to
   07.2022. Patients with sepsis were diagnosed with a SOFA score ≥2,
   combined with at least one infection site, and were managed following
   the newly updated international guidelines [72]^24^, [73]^25. Patients
   who were <18 or >80 years old or pregnant, had other blood diseases,
   cancer, or autoimmune diseases, and had incomplete clinical data were
   excluded. There were 52 patients with sepsis and 36 healthy volunteers
   included in the study. We collected blood samples from these patients
   within 24 h of admission. For PBMC isolation, blood samples were added
   to a Lymphocyte Separation Tube for Human Peripheral Blood (DAKEWE
   BIOTECH, Shenzhen, China) and centrifuged at 400 g and room temperature
   for 20 minutes. Plasma (supernatant) samples were collected and stored
   at -80 °C for further use. PBMC layers were isolated and centrifuged at
   400 g and 4 °C for 10 minutes. The precipitate was added to red blood
   cell lysis buffer (Yeasen, Shanghai, China) and centrifuged again,
   followed by washing with PBS and further centrifugation three times.
   The final precipitate was kept as PBMC samples in liquid nitrogen.

RNA sequencing analysis

   RNA sequencing analysis of PBMCs in the study was conducted by Gene
   Denovo Biotechnology Co. (Guangzhou, China). Four PBMC samples from
   patients with sepsis and 4 from healthy subjects were randomly selected
   to perform mRNA transcriptome sequencing. Briefly, RNA quality was
   detected by an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo
   Alto, CA, USA) and RNase-free agarose gel electrophoresis after total
   RNA extraction. Oligo (dT) beads were used to enrich mRNAs, which were
   fragmented into short fragments and reverse-transcribed into cDNAs.
   Double-strand cDNAs were purified using a QiaQuick PCR extraction kit
   (Qiagen, Venlo, Netherland), end repaired, poly (A) added, and ligated
   to the Illumina sequencing adapter. The products were sequenced using
   an Illumina NovaSeq 6000. Differentially expressed genes (DEGs) were
   screened by the parameters of fold change ≥ 2 and p value ≤ 0.05. The
   Kyoto Encyclopedia of Genes and Genomes (KEGG) database was used to
   perform pathway enrichment analysis of the screened DEGs.

Bioinformatics analysis

   Expression data of ADARs in different cell types were obtained from The
   Human Protein Atlas website ([74]https://www.proteinatlas.org/)
   [75]^26^, [76]^27. Single-cell RNA sequencing data ([77]GSE207651) were
   obtained from the GEO database ([78]https://www.ncbi.nlm.nih.gov/geo/)
   and analysed using the Seurat R package (version 3.0.1). Cell types
   were clustered and annotated as previously described [79]^28.

Luminex assay

   Serum levels of cytokines (IL-1β, IL-6, TNF-α, IL-10, and IL-4) in 29
   patients with sepsis and 6 healthy volunteers were detected by a
   Luminex assay using a Human XL Cytokine Luminex Performance Panel
   Premixed Kit (#FCSTM18-30, R&D Systems, Minneapolis, MN, USA) according
   to the manufacturer's instructions. Outlier data were removed during
   statistical analysis.

Scanning electron microscopy (SEM)

   Cell samples were rinsed with PBS, followed by fixation with electron
   microscope fixation liquid (#CR2202015, Servicebio, Wuhan, China) for
   30 min in the dark. The slides were preserved at 4 °C. After
   dehydration and drying, the samples were sputtered with gold for 30 s
   by a sputter coater and further detected under a scanning electron
   microscope (Hitachi SU8100, Tokyo, Japan).

Animal model establishment

   Male C57BL/6 mice (weight: 20-25 g, age: 8-10 weeks) were purchased
   from the Animal Center of the Fourth Military Medical University.
   Before modelling, all mice were adaptively cultured under normal
   conditions for one week. An animal sepsis model was established through
   caecal ligation and puncture (CLP) as previously described [80]^29.
   Briefly, mice were anaesthetized by inhalation of isoflurane. A midline
   incision (1 cm) on the lower abdomen was made, and then, the free end
   of the caecum was gently removed from the abdominal cavity to carry out
   caecal ligation by a 3/0 silk suture at the middle and two punctures
   with a 21-gauge needle, followed by gently squeezing a small amount of
   faeces to keep the puncture sites open. The caecum was carefully
   returned to the original position, and the abdomen was closed. A
   similar operation without CLP was performed on the mice in the sham
   group. A subcutaneous injection of normal saline (1 ml) was given for
   fluid resuscitation after surgery. A total of 1×10^8 PFU of
   ADAR1-overexpressing adenovirus (dissolved in 200 μl of PBS) was
   administered to septic mice through tail vein injection. Lung tissue
   samples were collected at 0 (sham), 6, 24, 48, and 72 h after surgery.
   Mouse ADAR1 overexpression adenovirus (Ad-O/E-ADAR1,
   [81]NM_001146296.1) was cloned and inserted into the pENT/CMV vector
   and prepared through a commercial service from GenePharma Co. Ltd.
   (Shanghai, China). Subsequent animal experiments were performed at 24 h
   after CLP administration.

Complete blood count

   One hundred microlitres of peripheral blood was taken from the mice and
   transferred into a disposable capillary tube (#KJ002, KangJianMedical
   Apparatus, China). The tube was gently tapped to ensure optimal contact
   between the blood and anticoagulant coating on the inner wall of the
   tube. The tube was allowed to stand at room temperature for 5 minutes
   and then gently tapped again prior to utilizing the fully automated
   blood cell DF-3000Vet Analyzer (Beijing, China).

Determination of bacterial load in blood and lungs

   Blood samples from the orbital cavity and lung tissues of mice in three
   groups, namely, sham, CLP, and CLP+OE-A, were obtained. Blood (100 μl)
   was collected from the orbital cavity of each group of mice and
   transferred into a sterile anticoagulant tube (#G4811-1.5ML,
   Servicebio). The bacterial count was determined by plating 100 μl of
   undiluted blood onto plate count agar medium (#HBPM002, Hopebio,
   China), followed by incubation at 37 °C. Subsequently, the colonies
   were counted. Furthermore, the lungs were excised and maintained under
   sterile conditions. A small volume (100 μl) of tissue homogenate was
   diluted 10-fold in sterilized phosphate-buffered saline (PBS),
   inoculated onto plate count agar medium, and incubated overnight at 37
   °C. Afterwards, the bacterial colony count was determined [82]^30^,
   [83]^31.

Haematoxylin and eosin (HE) staining and scoring

   Lung tissues were embedded in paraffin and cut into 5 μm-thick slices,
   followed by staining with haematoxylin and eosin dyes (Servicebio). The
   slices were observed under an upright microscope (Nikon Eclipse E100,
   Tokyo, Japan). The disease score and lung injury score of tissues were
   blindly evaluated by an experienced pathologist based on six random
   fields of vision of each HE-stained slice according to previously
   published scoring criteria with a slight modification. The disease
   clinical scoring system contained 5 aspects: appearance (4 points),
   behaviour changes at resting state (3 points) and after stimulation (3
   points), respiratory clinical signs (3 points), and dehydration status
   (5 points) [84]^32. Lung injury scoring mainly included 4 aspects:
   pulmonary interstitial oedema, alveolar oedema, inflammatory
   infiltration, and alveolar haemorrhage, with each criterion scored
   according to severity (4 points in total for each, higher points
   indicated more severe damage) [85]^33. The lung injury scores were
   calculated by the sum of these criteria.

Cell culture and transfection

   The murine RAW264.7 macrophage line was purchased from the Cell Culture
   Center, Chinese Academy of Medical Sciences (Beijing, China). The cells
   were cultured in high-glucose DMEM (Servicebio) with 10% foetal bovine
   serum (Zeta Life, Menlo Park, CA, USA) and 1% penicillin‒streptomycin
   (Biosharp, Anhui, China) at 37℃ under 5% CO[2]. RAW264.7 cells (1×10^6)
   were treated with 1 μg/ml lipopolysaccharide (LPS, Corning, Wilmington,
   NC, USA) to mimic the septic environment in vitro (6, 12, and 24 h).
   For cell transfection, RAW264.7 cells were seeded on 60 mm-cell dishes
   and cultured with serum-free Opti-MEM (Gibco, Brooklyn, NY, USA) 1 h
   before transfection when cell confluence reached 70-80%. Then, the
   cells were transfected with ADAR1-overexpression adenovirus
   (GenePharma), ADRA1-siRNA (RiboBio, Guangzhou, China), A20-siRNA
   (RiboBio), miR-21 mimic, miR-21 inhibitor (RiboBio) and their negative
   controls through Zeta transfection reagents (Zeta Life). Further LPS
   treatment was performed 24 h after transfection. The cells were
   collected for further experiments 48 h after transfection.

Immunofluorescence (IF) staining

   Immunofluorescence staining was performed on both lung tissue and
   cellular slices through routine operation. For tissue staining, after
   dewaxing, rehydrating, antigen repairing, and blocking,
   paraffin-embedded tissue slices were incubated with diluted primary
   antibodies of anti-F4/80 (#GB11027, Servicebio), anti-Cytokeratin 7
   (CK7, #GB11225, Servicebio), and/or anti-ADAR1 (#SC-73408, Santa Cruz),
   anti-cleaved-GSDMD (#AF4013, Affinity, Jiangsu, China), TUNEL
   (#GDP1043, Servicebio), anti-iNOS (#GB11119, Servicebio), or anti-CD163
   (#GB11340-1, Servicebio) at 4 °C overnight, followed by incubation with
   their corresponding secondary antibodies at room temperature for 1 h in
   the dark. Between double staining procedures, 488-TSA (Servicebio)
   incubation and reantigen repair were conducted. DAPI (#G1012,
   Servicebio) was used for nuclear counterstaining. The slices were
   mounted with antifluorescence quenching sealing reagents (Boster,
   Wuhan, China) and captured under a fluorescence microscope (Life
   Technologies, Carlsbad, USA). For cellular staining, the primary
   antibodies used were anti-cleaved GSDMD (#AF4013, Affinity) and
   anti-caspase-1 (#GB11383, Servicebio). Mean fluorescence intensity
   (MFI) in a random field of view was measured using ImageJ software.

Preparation and induction of bone marrow-derived macrophages

   Bone marrow-derived macrophages (BMDMs) were prepared following
   previously established protocols [86]^34. BMDMs were stimulated with
   LPS (1 mg/mL) to induce M1 polarization, and after 24 hours, the cells
   were harvested for further experimentation.

Flow cytometry analysis

   Flow cytometry was adopted to assess molecular expression alterations
   in specific cells of human PBMCs and RAW264.7 cells in different
   groups. Human PE-conjugated CD14 antibody (#12-0149-42, eBioscience)
   was used to label monocytes of PBMCs for 30 min on ice in the dark.
   After three washes with FACS buffer, the cells were treated with BD
   Cytofix/Cytoperm Fixation and Permeabilization Solution (BD
   Biosciences) for 30 min and 1× BD Perm/Was buffer for 15 min.
   Anti-cleaved-GSDMD (#ab227821, human, Abcam, Shanghai, China) was
   further incubated with the cells for 30 min, followed by the addition
   of Alexa Fluor™ 488-conjugated goat anti-rabbit IgG (#A11008, Thermo
   Fisher Scientific, Waltham, MA, USA). For murine RAW264.7 cell
   detection, APC-conjugated anti-F4/80 (#MF48005, eBioscience, San Diego,
   CA, USA) was used to label macrophages. FITC-conjugated anti-CD11c
   (#11-0114-81, eBioscience) and PE-conjugated anti-CD206 (#12-2061-80,
   eBioscience) were used to label M1 and M2 polarization. Cell pyroptosis
   was evaluated by anti-cleaved GSDMD (#ab255603, mice, Abcam) and the
   following flow cytometry secondary antibody: Alexa Fluor™
   488-conjugated goat anti-rabbit IgG (#A11008, Thermo Fisher
   Scientific). The cell death states of BMDMs and RAW264.7 cells were
   determined using the FITC Annexin V Apoptosis Detection Kit I (#556547,
   BD Pharmingen, San Diego, CA) according to the manufacturer's
   instructions. In brief, cells (1×10^6) were resuspended in 100 μl of 1×
   binding buffer, followed by the addition of 5 μl of Annexin V-FITC and
   10 μl of PI. Subsequently, the samples were incubated for 30 minutes at
   a temperature of 4 °C in darkness, and the fluorescence signals were
   detected using a COULTER EPICS XL flow cytometer. In addition, for
   determination of the CD4^+ and CD8^+ T-cell subsets within the
   mononuclear cell suspensions prepared from the blood of mice, the
   levels of CD4^+, CD8^+, and CD4^+/CD8^+ were measured using a flow
   cytometer. Mouse APC-conjugated anti-CD19 (#17-0193-82, eBioscience)
   and PE-Cyanine7-conjugated anti-CD3 (#25-0032-82, eBioscience) were
   used to label B and T lymphocytes. T-cell subsets were analysed by
   FITC-conjugated anti-CD4 (#11-0042-83, eBioscience) and PE-conjugated
   anti-CD8 (#12-0081-82, eBioscience) [87]^35. Cell samples were analysed
   using a NovoCyte flow cytometer and NovoExpress software (ACEA
   Biosciences, San Diego, CA, USA).

RNP immunoprecipitation (RIP)

   The association between ADAR1 and miR-21 was studied in our previous
   study. Here, we performed RIP experiments again to verify the binding
   effects of ADAR1 on pre-miR-21 using RAW264.7 cells as previously
   described [88]^22^, [89]^36. A total of 20 million RAW264.7 cells were
   collected and lysed and precipitated with 30 mg of anti-ADAR1
   (#SC-73408, Santa Cruz) or anti-IgG antibodies (#2729, CST) at room
   temperature for 4 h. IP products were then reverse transcribed, and
   quantitative PCR analysis was performed to assess the expression of
   pre-miR-21 and the internal reference.

Dual luciferase reporter assay

   A pmirGLO Dual-Luciferase miRNA Target Expression Vector (GenePharma)
   containing wild-type (WT) or mutant (MUT) 3'UTR of A20 (sequences are
   shown in the figure) was constructed and cotransfected with miR-21-5p
   mimic or mimic-NC into RAW264.7 cells using Lipofectamine 2000 reagent
   (Invitrogen, Carlsbad, CA, USA). Twenty-four hours after transfection,
   the relative luciferase activity of the cells was determined by an
   Infinite M1000 multimode microplate reader (Tecan, Männedorf,
   Switzerland) via a dual-luciferase reporter assay system (Promega,
   Madison, WI, USA).

Quantitative reverse transcription (qRT)-PCR

   Total RNA from human PBMCs, murine lung tissues, or RAW264.7 cells was
   extracted using TRIzol reagent (Invitrogen) and then
   reverse-transcribed into cDNA by a PrimeScript RT Reagent Kit (TaKaRa,
   Beijing, China) according to the manufacturer's instructions. Moreover,
   the miRNA qRT‒PCR Stater Kit (#C107R-1, RiboBio) was used for RT and
   qPCR for miRNA. Further quantitative PCR was performed using cDNA
   templates, primers, PCR Mix (#11141ES60, Yeasen), and SYBR Green Mix
   (#11201ES08, Yeasen). Specific primer pairs were synthesized by Tsingke
   Biotechnology (Beijing, China). The primer sequences are shown in
   [90]Supplementary Table 1. The primer pairs of miR-21 and premiR-21
   were synthesized by RiboBio. GAPDH and U6 RNAs were used as internal
   references for coding genes and miRNAs, respectively. Final relative