Self-DNA accumulation as a risk factor for accelerating the pathogenesis of rheumatoid arthritis in elderly individuals

doi:10.21203/rs.3.rs-1827868/v1

Download PDF

Research Article

Self-DNA accumulation as a risk factor for accelerating the pathogenesis of rheumatoid arthritis in elderly individuals

https://doi.org/10.21203/rs.3.rs-1827868/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Ageing is an unavoidable process in humans and a major factor for the increasing risk of various diseases. In the United States, more than 50% of rheumatoid arthritis patients are middle-aged or elderly, but the risk factors and mechanisms by which ageing increases the incidence of rheumatoid arthritis are not known.

It has been suggested that the accumulation of DNA fragments increases the risk of autoimmune diseases, such as systemic lupus erythematosus. DNA fragments are a common nucleic acid metabolite in ageing organisms as well as in the serum of humans and animals with rheumatoid arthritis; therefore, we hypothesize that DNA fragments are one of the factors contributing to the development of rheumatoid arthritis due to ageing.

First, we analysed two in vitro DNA damage response models by using a gene silencing approach and determined that the DNA fragment clearance gene TREX1 can regulate inflammatory factor release in normal cells. Second, after TREX1 expression was knocked down locally or systemically in rats via the Cre-LoxP system and compared with that in AIA (adjuvant-induced arthritis) model rats treated with AAV-TREX1, it was determined that DNA fragments can result in manifestations of arthritis and abnormal activation of the immune system in rats. These results, including the low expression of the TREX1 gene in clinical patient and AIA model samples and the results of immunohistochemical, Western blot, and transcriptome analyses, revealed that the TREX1 gene can regulate cellular senescence-associated secretory phenotype (SASP)-related manifestations and showed that dysregulation of c-Jun and c-Fos, components of the TREX1 transcription factor AP-1, is associated with SASP induction. Finally, it was confirmed in vitro that different causes of decreased c-Fos expression can inhibit TREX1 expression.

These DNA fragments are potent producers of inflammation-releasing mediators, and TREX1 is an effective degrader of DNA fragments; it is also a key gene that regulates cellular immunity and ageing. Therefore, effectively clearing excess DNA fragments from the body and ensuring the health of senescent cells may be a potential prevention strategy for RA.

Rheumatology

Immunology

Molecular Biology

DNA fragment

TREX1

ageing

SASP

rheumatoid arthritis

Rheumatoid arthritis (RA) is a chronic and systemic autoimmune disease with an unclear pathogenesis¹. Current studies have suggested that multiple factors, including ageing, a smoking history, an unfavourable body mass index (BMI), immunogenicity, and genetic polymorphisms, may be associated with the pathogenesis of RA and its refractory problems². Notably, treatment with the current DMARDs is often associated with serious side effects and the development of drug resistance². With the disease progression time of 5 to 10 years and a disability rate of ~ 43.48%³, effective prevention or delay of the onset of RA would be of great benefit to RA patients. Therefore, the identification and characterization of genes related to the onset of RA may lay a solid theoretical foundation for the development of preventive therapeutic approaches for RA in the future.

Ageing is an inevitable process during life, and age-associated chronic diseases are the predominant contributors to global morbidity, hospitalization, health care costs and mortality⁴. Epidemiological studies have identified a variety of age-related diseases, including atherosclerosis, Alzheimer's disease, osteoarthritis, cancer, type 2 diabetes and RA^5–7. Although RA showed a global prevalence of approximately 1%⁸, more than 50% of middle-aged and elderly adults were reported to develop RA in the United States between 2004 and 2014⁹. Thus, studies on the causes of the high incidence of RA in elderly people are necessary to establish preventive strategies for RA.

Studies have shown that metabolic DNA damage and sustained activation of the immune system are important biological activities that drive cellular ageing¹⁰. During the ageing process, DNA fragments accumulate and circulate in the bloodstream to become circulating free DNA (cfDNA)¹¹. Of note, DNA fragmentation is also considered a biomarker for many diseases, including trauma, sepsis, aseptic inflammation, myocardial infarction, stroke, transplant rejection, diabetes mellitus and sickle cell disease¹². In fact, both Aicardi-Goutières syndrome and familial frostbite lupus occur with a common phenotypic feature of failure to properly clear self-DNA or self-RNA fragments, leading to abnormal activation of the immune system¹³. Intriguingly, RA patients treated with disease-modifying anti-rheumatic drugs (DMARDs) were found to have a significant decrease in plasma DNA fragmentation¹⁴, indicating the possible correlation of DNA fragmentation with the pathogenesis of RA. Based on the hypothesis that the increased incidence of RA in middle-aged and elderly patients may be due to age-related accumulation of DNA fragments, determining the origin and promoting the clearance mechanisms of plasma DNA fragments in these patients may open a new avenue for developing prevention and treatment strategies for RA.

In the present study, the accumulation of circulating DNA fragments in the AIA model was confirmed to be triggered by reduced expression of TREX1, which eventually promoted the release of proinflammatory cytokines and generated inflammatory conditions in AIA rats. With validation in conditional knockout rats via the Cre-LoxP system, the current study revealed that knockdown of the 3'->5' DNA exonuclease TREX1 increased the severity of arthritis in AIA rats. In contrast, overexpression of TREX1 by adeno-associated virus (AAV) transduction significantly inhibited the senescence-associated secretory phenotype (SASP) and suppressed arthritic inflammation in AIA rats. Further mechanistic studies revealed that the TREX1 level was decreased in AIA rats and RA-FLSs exhibiting the SASP due to imbalanced levels of c-Jun and c-Fos, the major components of the TREX1 transcription factor AP-1, leading to accumulation of self-DNA fragments, which subsequently activates the systemic immune response. Our findings highlighted the possible pathogenic role of the ageing-related autoimmune response and tissue inflammation in RA and suggested that DNA fragmentation is one of the risk factors for the pathogenesis of RA in elderly individuals.

Cell culture & antibodies

The immortalized RA-FLS (MH7A) was purchased from ATCC. Reagents for cell culture, such as Dulbecco’s modified Eagle’s medium (DMEM), foetal bovine serum (FBS), trypsin-EDTA (0.05%), phosphate-buffered saline (PBS), and 100× penicillin/streptomycin (PS), were obtained from Gibco (Oklahoma, USA). RA fibroblast-like synoviocytes (FLSs) were used between passages three and eight, and the cell type was identified as described in our previously published work. RA-FLSs were cultured in 6-well plates (1×10⁵ cells/well to ensure that the confluence reached 60–80%) in 2 ml of Dulbecco's modified Eagle’s medium (DMEM) supplemented with 10% FBS and 1% PSG at 37 ℃ in a 5% CO₂ humidified environment. The medium was removed and washed once with PBS for subsequent experiments. The anti-TREX1 antibody (diluted 1:1000 for WB and 1:200 for IHC were purchased from Novus Biologicals, USA; the anti-cGAS antibody (1:1000 for WB) was obtained from Taiclone, China Taiwan; the anti-β-actin antibody (1:1000), anti-rabbit secondary antibody (1:2500), and anti-mouse secondary antibody (1:2500) were purchased from Santa Cruz, USA; and flow cytometry antibodies such as APC/Cyanine7 anti-rat CD45 (1:1000), FITC anti-rat CD3 (1:1000), PerCP/Cyanine5.5 anti-rat CD4 (1:1000), APC anti-rat CD8a (1:1000) and PE anti-mouse/rat/human FOXP3 (1:1000) were purchased from Biolegend, USA.

Extraction and assay kits

Total RNA was extracted using a FavorPrepTM Blood/Cultured Cell Total RNA Mini Kit (Favorgen Biotech, China). Genomic DNA was prepared by using a FavorPrep Tissue Genomic DNA Extraction Mini Kit (Favorgen Biotech, China). For Western blotting, signals were visualized by using a SuperSignal West Femto Maximum Sensitivity Substrate Kit (Thermo, USA). Lymphocyte isolation was performed with a Ficoll density gradient centrifugation kit (GE Healthcare, USA). Cell-free DNA was isolated and quantified by using a Dynabeads® SILANE Viral NA Kit (Thermo, USA) and Quant-iT™ PicoGreen ® dsDNA Reagent and Kits (Thermo, USA).

Reagents and services: Lipofectamine™ 3000 (Thermo, USA); TREX1 siRNA (Santa Cruz, USA); cGAS siRNA (Santa Cruz, USA); siRNA control plasmid (Sigma, USA); Mycobacterium tuberculosis (M. tuberculosis; Difco, USA); mineral oil (Sigma, USA); methotrexate (MTX) (LC labs, USA); Cre adeno-associated virus construction (Ubigene, China); TREX1 adeno-associated virus construction (Ubigene, China); dimethyl sulfoxide (DMSO; ACROS, USA); DAPI (Invitrogen, USA); RIPA buffer (10×, Cell Signaling Technology, USA); EDTA-free protease inhibitors, PhosSTOP inhibitor (Roche, Basel, Switzerland); TRIzol (Invitrogen, USA); FastStart Universal SYBR Green Master Rox (Roche Diagnostics, USA); Maxima™ H Minus cDNA Synthesis Master Mix (Thermo, USA); PVDF membranes (Bio-Rad, USA).

Clinical samples

Blood samples were collected from patients with RA, patients with osteoarthritis (OA), and healthy volunteers upon the provision of written informed and voluntary consent at Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangzhou (China), and approval by the research ethics committee (GDREC 2015391H). All of the epidemiological investigations and classification of the volunteers were carried out according to the American College of Rheumatology criteria. The clinically relevant information and data are provided in Table 1.

Table 1

RA, OA, and healthy volunteers’ baseline data
	RA (N = 39)	OA (N = 25)	Healthy (N = 25)
Age (years (mean (range)))	53 (30–73)	57 (30–72)	45 (24–74)
Sex (N (F/M))	24/15	19/6	20/5
Disease duration (years (mean (range))	5 (1 ~ 20)	8 (1 ~ 23)	NA
CRP (mg/L (mean (range)))	20.2 (1.4 ~ 93)	14.5 (3.13 ~ 39.5)	NA
ESR (mm/h (mean (range)))	49.5 (8 ~ 120)	22.3 (8 ~ 47)	NA
RF (> 20 IU/mL)	31/38	NA	NA
CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; RF, rheumatoid factor; NA, not assessed.

DNA fragment preparation & DNA damage model

RA-FLSs were cultured in 6 cm dishes to a confluence of 60–80%. The cells were then exposed to UV light for 5 min, 10 min, and 15 min. The medium was removed and washed twice with PBS for the following experiment. RA-FLSs were harvested for DNA extraction using a DNA extraction kit (Favorgen). DNA fragmentation was performed by ultrasonication for 30 min to shear the DNA into fragments of 500 bp as determined by DNA gel electrophoresis. RA-FLSs cultured in 6-well plates were transfected with 5 µg of DNA fragments for 24–72 h in the absence or presence of Lipofectamine 3000. Furthermore, another group of RA-FLSs was also transfected with different concentrations of DNA fragments (1 µg, 5 µg, and 10 µg) with Lipofectamine 3000 for 24 h. All the medium was removed, and the cells were washed once with PBS before the next step of analysis.

siRNA transfection

The selected RA-FLSs were transfected using a Lipofectamine 3000 kit (Thermo, USA) according to the manufacturer’s protocol. Briefly, after the cells were cultured in a plate overnight to a confluence of 60–80%, the medium was aspirated and washed once with PBS. The TREX1 or cGAS siRNA plasmid for knockdown of TREX1 or cGAS was mixed with 250 µl of Lipofectamine 3000 and incubated for 15 min. Then, this mixture was mixed with Opti-MEM in a total volume of 1 ml and added to a 6-well plate for incubation for 12 h. After incubation for 4 h, another 1 ml of medium was added to the wells and incubated for 24 h. After 48 h, the cells were transfected with DNA fragments for 24 h, 48 h and 72 h. Real-time PCR was used to verify the mRNA expression of TREX1, cGAS, and inflammatory cytokines.

Gene expression analysis by real-time PCR

Total RNA from blood and cultured cells was isolated using the FavorPrep™ Blood/Cultured Cell Total RNA Mini Kit (Favorgen Biotech) according to the manufacturer’s protocol. Total RNA from frozen animal synovium and tissues was extracted with TRIzol (Invitrogen, USA). Approximately 20–40 mg of frozen tissue was mixed with 1 ml of TRIzol reagent and homogenized by at least 3 cycles in a homogenizer. Then, 100 µl of chloroform was added and shaken vigorously for 5 min, and the mixture was centrifuged at 4°C and 12,000 × g for 15 min. The supernatant from the phenol/chloroform extraction step was collected (approximately 400–500 µl) and precipitated with isopropanol (1:1.2 supernatant:isopropanol) for 30 min on ice; this mixture was then centrifuged at 12,000 × g for 15 min. The precipitate was washed with 70–80% cold ethanol. The white pellet was completely dissolved in 50 µl of RNase-free water. A UV spectrophotometer (NanoDrop Technologies, USA) was used to measure the quality and concentration of RNA.

For real-time PCR analysis, 1 µg of RNA was reverse transcribed with Maxima™ H Minus cDNA Synthesis Master Mix (Thermo, USA). The mixture was incubated at 25°C for 10 minutes and at 50°C for 15 minutes, and the reaction was then terminated by heating at 85°C for 5 minutes. The PCR mixture (total 20 µl) comprised 0.5 µl of template DNA, 0.5 µl each of the forward and reverse primers, 0.5 µl of template, 10 µl of SYBR Master Mix (Roche Diagnostics, USA), and 8.5 µl of ddH2O to bring the total volume up to 20 µl. Quantification of gene expression was performed with a ViiA 7 Real-Time PCR System (Applied Biosystems). All mRNA expression data were normalized to the reference gene β-actin using the ΔΔCT method for relative quantification.

Western blotting

RA-FLSs incubated with DNA fragments in the absence or presence of Lipofectamine 3000 were added to RIPA lysis buffer for protein lysate preparation. Proteins in the centrifuged supernatant were then separated by 10% SDS–PAGE. The proteins on the PAGE were transferred to a PVDF membrane, which was probed first with antibodies against TREX1 (1:1000), cGAS (1:1000), and β-actin (1:1000) and then with HRP-conjugated goat anti-rabbit IgG as the secondary antibody (1:5000). The immunoreactive bands were then visualized by ECL with a FluorChem R system (ProteinSimple, America). Quantitative analysis of Western blot signals was performed using ImageJ software.

SD rats used for establishment of the experimental arthritis model (AIA) and treatment

Wild-type male Sprague–Dawley (SD) rats weighing 110–120 g were purchased from SPF Biotechnology Co., Ltd. Conditional knockout (Cko) TREX1^+/- SD rats weighing 80–120 g were generated using CRISPR/Cas9 by Cyagen (USA). Rats were bred and maintained at the State Key Laboratory of Quality Research in Chinese Medicines of Macau University of Science and Technology. The genotypes of TREX1 transgenic rats were identified by PCR of tail genomic DNA using two pairs of primers. The animal study was conducted according to the guidelines of the Institutional Animal Care and Use Committee of Macau University of Science and Technology and approved by the Ethics Committee of Macau University of Science and Technology (protocol code MUSTARE – 003–2020) on 23 Feb 2020. During the experiment, the animals were housed for 1 week on a 12:12 h light-dark cycle at room temperature (22 ± 1°C).

Model 1 (Inflammatory potency of DNA fragment injection in the AIA model)

SD rats were divided into 6 groups (n = 6 ~ 8 rats/group) as follows: (1) healthy control (n = 7), (2) AIA model (n = 8), (3) AIA + DNA (200 µg; n = 7), (4) AIA + DNA (50 µg; n = 7), (5) healthy control + DNA (200 µg; n = 7), and (6) healthy control + DNA (50 µg; n = 7). Mineral oil (Sigma) containing 2.5 mg/ml M. tuberculosis cells (Difco, USA) was ground and rolled intensively until the mixture turned white. DNA fragments were injected into the knee joints of rats, and 0.1 ml of the above mixture was then injected subcutaneously at the base of the tail on Day 0. Arthritis scores and hind paw volumes were evaluated and recorded every 3 days until Day 30.

Model 2 (monitoring proinflammatory cytokine levels in AIA rats injected with DNA fragments)

SD rats were injected with 100 µg of DNA fragments via the tail vein and were then subjected to AIA model establishment. Blood samples were collected at 0 min, 30 min, 2 h, 4 h, 24 h, 48 h, and 72 h for proinflammatory cytokine detection using RT–PCR.

Model 3 (monitoring proinflammatory cytokine levels in AIA rats injected with DNA fragments after the first appearance of hind paw swelling)

SD rats were divided into 4 groups (n = 6 rats/group) as follows: (1) healthy control, (2) AIA model, (3) AIA + DNA (100 µg), and (4) AIA + DNA (50 µg). Mineral oil (Sigma) containing 2.5 mg/ml M. tuberculosis cells (Difco, USA) was ground completely until the mixture turned white, indicating that it was ready for AIA model establishment. The SD rats were then injected with complete adjuvant for arthritis induction (on Day 0). In addition, the AIA rats were t injected with 50 or 100 µg of DNA fragments (sonicated DNA from rat-derived muscle tissue) via the tail vein beginning on Day 0 and continuing every 2 days until the day (Day *) when hind paw swelling first became visible. Blood samples were then collected on day (*) for proinflammatory cytokine detection using RT–PCR.

Model 4 (evaluating the anti-inflammatory effect of TREX1 overexpression in AIA rats)

SD rats were divided into 6 groups (n = 8 rats/group) as follows: (1) healthy control, (2) AIA (adeno-associated virus, AAV-EGFP), (3) AIA + methotrexate (MTX; 7.6 mg/kg/week), (4) AIA + AAV-TREX1 (tail vein injection with 1×10¹¹ PFU), (5) AIA + AAV-TREX1 (tail vein injection with 1×10⁹ PFU), (6) AIA + AAV-TREX1 (joint injection with 2.5×10¹⁰ PFU) and (7) AIA + AAV-TREX1 (joint injection with 2.5×10⁸ PFU). Tail vein injection of AAV-TREX1 (1×10¹¹ PFU) and AAV-TREX1 (1×10⁹ PFU) was conducted 10 days before AIA induction, and joint injection of AAV-TREX1 (2.5×10¹⁰ PFU) and AAV-TREX1 (2.5×10⁸ PFU) was performed 10 days before AIA induction. The SD rats that received AAV injection were subsequently injected subcutaneously with 0.1 ml of the mixture at the base of the tail on Day 0. Arthritis scores and hind paw volumes were evaluated and recorded every 3 days until Day 30.

Joint swelling measurement and clinical scoring

Joint swelling and clinical scores were evaluated daily from the first injection (Day 0 for rats) until the animals were sacrificed. The investigator was blinded to the group allocation when measuring joint swelling. In the AIA model, joint swelling in the rats was evaluated by measuring the volume of the hind paws with a plethysmometer (Ugo Basile, Italy; or Kent Scientific Corp, Connecticut, USA) and calculating the average volume every 3 days. Animals that died during the experimental period were excluded from the analysis. Clinical scores for each hind paw and forepaw of the rats were determined following the standard evaluation process for clinical scoring: the severity of arthritis can be objectively inspected in four paws and is scored for each paw on a scale of 0 ~ 4 according to the arthritis index described in Table 2.

Table 2

Rat arthritis index
Score	Symptom
0	No evidence of swelling and erythema of the joints, including the small joints of the forepaw and the phalangeal joint and the large joints such as the wrist and ankle.
1	Mild swelling and erythema of the ankle joint.
2	Mild swelling and erythema extending to the small joints.
3	Severe swelling and erythema of large joints.
4	Severe swelling and erythema encompassing the small and large joints.

Micro-CT analysis

At the end of the treatment period, the rats were humanly sacrificed, and the left hind paw of each rat was amputated, fixed with 4% PFA, and then scanned using an in vivo Micro-CT scanner (SkyScan 1176, Bruker, Belgium). The following scanning parameters were used to obtain high-quality images of the joints of the rats: 35 µm resolution, 85 kV tube voltage, 385 µA tube current, 65 ms exposure time, rotation step over 360° 0.7, and a 1 mm Al filter. The images were reconstructed using NRecon software (Bruker Micro CT, Belgium). The Micro-CT score was obtained from five disease-related indices for Micro-CT analysis of the calcaneus: bone mineral density, bone volume fraction, cortical mineral density, trabecular number, and total porosity.

Cell-free DNA (cfDNA) extraction and measurement

For extraction of cfDNA in the plasma of rats, peripheral blood samples of rats were collected on Day 30 in EDTA-containing tubes. Then, the blood samples were centrifuged at 400×g for 10 min at 4°C, and the plasma fraction was recentrifuged at 12,000×g for 10 min at 4°C to remove cell debris and was then stored at − 80°C for analysis. cfDNA was extracted from 100 µL of plasma using a Dynabeads® SILANE Viral NA Kit. The concentration of cfDNA was determined with Quant-iT™ PicoGreen ® dsDNA Reagent and Kits.

Haematoxylin and eosin (H&E) staining and immunohistochemistry (IHC) of rat tissue samples

Experimental AIA rats were sacrificed on Day 30, and the synovium and organs, including the brain, heart, kidneys, liver, spleen, thymus, and joint tissues, were collected for weighing and H&E staining. Synovial samples were immediately frozen in liquid nitrogen and stored at -80°C. Organs and joint tissues were fixed with 4% paraformaldehyde for 24 h and were then dehydrated and embedded in paraffin blocks at 60°C. Six-micrometre-thick sections were dehydrated, deparaffinized, and rehydrated according to standard protocols. Synovial, joint, and organ sections were incubated with the anti-TREX1 primary antibody (1:200, Novus Biologicals, USA) at 4°C overnight. After washing, the slides were incubated with streptavidin-conjugated horseradish peroxidase (BioSite Histo Plus (HRP) polymer anti-rabbit kit; Nordic BioSite, Sweden) for 1 h. They were then mounted with FluorSave™ Reagent (Millipore, USA). Images were acquired by light microscopy (Leica DM2500, Germany). Joint and organ sections from AIA rats were subjected to H&E staining. Images were acquired by light microscopy (Leica DM2500, Germany).

Generation of TREX1 conditional knockout (Cko TREX1-/-) SD rats

LoxP-E2-loxP-mediated knockout of Trex1 in SD rats was accomplished by CRISPR/Cas-mediated genome editing. The rat TREX1 gene (GenBank accession number: NM_001024989.1, Gene ID: 100049583) is located on chromosome 8. Two exons have been identified, with the ATG start codon and TAA stop codon in exon 2. Exon 2 was thus selected as the region for conditional knockout. Deletion of exon 2 was expected to result in loss of function of the rat Trex1 gene by inducing a frameshift of downstream exons. The loxP sequence was inserted into the donor vector upstream and downstream of exon 2 by homology-directed repair. gRNA targeting vectors were constructed and confirmed by sequencing, and a donor vector with flanking of homologous arms was also constructed. Cas9 mRNA and gRNA generated by in vitro transcription were co-injected with the donor vector into fertilized eggs to obtain KI rats. The pups were genotyped by PCR, and sequence analysis was then performed.

Genotyping of TREX1-deficient rats

Genomic DNA was isolated from rat tails. Genotyping of TREX1 was performed by polymerase chain reaction (PCR). PCR amplification was performed using primers. Primer sequences: F-AGAGACTCACGGGCTTGTTTGAATC, R- GAGAGTAAGGCAGGCCAGTGGC. Each 25 µL PCR mixture consisted of 4 µL of genomic DNA (~ XX µg), 1 µL of each primer (10 µmol/L), 12.5 µL of 2 × Taq PCR Master Mix (constituents: 0.1 U Taq polymerase/µL, 500 µM each dNTP and PCR buffer), and 6.5 µL of ddH2O (DNase/RNase-free). PCR was performed with an initial denaturation step at 95°C for 5 min, followed by 35 cycles of denaturation at 95°C for 30 s, annealing at 60°C for 30 s and elongation at 72°C for 35 s. The amplified PCR products were subjected to electrophoresis on a 1.0% agarose gel with 0.5 µg/mL ethidium bromide (EB) prior to visualization by ultraviolet light.

Establishment of the AIA model using Cko TREX1^-/- rats

Model 1 (Hind paw joint injection of Cre adeno associated virus into Cko TREX1^-/- rats)

TREX1 conditional knockout (Cko TREX1^−/−) rats were divided into 6 groups (n = 6 ~ 7 rats/group) as follows: (1) Cko TREX1^−/− (healthy control, n = 7), (2) Cko TREX1^−/− (AIA, n = 6), (3) Cko TREX1^−/− (AIA + high concentration (2.5×10¹⁰ PFU) of Cre adeno associated virus (CreH), n = 6), (4) Cko TREX1^−/− (AIA + low concentration (2.5×10⁸ PFU) of Cre adeno associated virus (CreL), n = 6), (5) Cko TREX1^−/− (healthy control + CreH, n = 6), and (6) Cko TREX1^−/− (healthy control + CreL, n = 6). Rats in the three healthy control Cko TREX1^−/− groups and three AIA Cko TREX1^−/− groups were injected in the hind limb joints with nucleic acid-free water or DNA fragments (200 µg), respectively, on Day 0. Approximately 10 days before the Cko TREX1^−/− rats were injected with complete adjuvant for arthritis induction (on Day 0), they were injected in the hind limb joints with Cre adeno associated virus s (2.5 × 10⁸ PFU (CreL) or 2.5 × 10¹⁰ PFU (CreH)) to facilitate knockout of the TREX1 gene.

Model 2 (tail vein injection of Cre adeno associated virus into Cko TREX1-/- rats)

Cko TREX1^−/− rats were divided into 6 groups (n = 6 ~ 12 rats/group) as follows: (1) Cko TREX1^−/− (healthy control, n = 7), (2) Cko TREX1^−/− (AIA, n = 12), (3) Cko TREX1^−/− (AIA + CreH, n = 7), (4) Cko TREX1^−/− (AIA + CreL, n = 8), (5) Cko TREX1^−/− (healthy control + CreH, n = 8), and (6) Cko TREX1^−/− (healthy control + CreL, n = 6). Rats in the three healthy control Cko TREX1^−/− groups and the three AIA Cko TREX1^−/− groups were injected via the tail vein with nucleic acid-free water or DNA fragments (100 µg), respectively, on Day 0. Approximately 10 days before the Cko TREX1^−/− rats were injected with complete adjuvant for arthritis induction (on Day 0), they were injected via the tail vein with Cre adeno associated virus (1 x 10⁹ PFU (CreL) or -1 x 10¹¹ PFU (CreH)) to facilitate knockout of the TREX1 gene.

On Day 30, all rats were sacrificed, and blood, organs and joint tissue were collected for biochemical assays.

Lymphocyte isolation, intracellular cytokine staining, and flow cytometric analysis

Peripheral blood mononuclear cells (PBMCs) were isolated from AIA rats using a standard Ficoll density gradient centrifugation kit (GE Healthcare). T cells were stained as indicated with the following anti-rat CD monoclonal antibodies (mAbs): APC/Cyanine7 CD45, FITC CD3, PerCP/Cyanine5.5 CD4, and APC CD8a. For Treg cell staining, cells were surface stained for CD3, CD4, and CD45; washed, fixed, and permeabilized using a Foxp3 Staining Buffer Set; and then immunostained with PE anti-mouse/rat/human FOXP3 antibodies (all reagents were obtained from BioLegend, Beijing). Cells were acquired using a FACSAria flow cytometer (BD Bioscience) and further analysed using FlowJo v10 software (TreeStar, Inc.).

Statistical analysis

For analysis of two groups, Student's t test was used. One-way analysis of variance was used to analyse differences among 3 or more groups. All results are expressed as the means ± SEMs. GraphPad Prism 9.0 software (GraphPad Software, USA) was used for statistical evaluation of the data. p < 0.05 was considered significant.

3.1 The 3'->5' DNA exonuclease TREX1 is significantly downregulated in RA patients

Ageing is considered an important risk factor for rheumatoid arthritis (RA) ¹⁵via unknown mechanisms. Recent studies have reported that cell-free DNA (cfDNA) is frequently detected in the serum of individuals with advanced age ¹¹ and patients with cancer¹⁶, diabetes¹⁷, systemic lupus erythematosus (SLE)¹⁸, and many other diseases. Therefore, the accumulation of cfDNA or DNA fragments is considered a crucial biological marker for disease diagnosis and pathogenesis. Of note, previous studies reported that the concentration of cfDNA in the serum of RA patients was significantly reduced after treatment with antiarthritic drugs¹⁹, revealing the possible pathogenic role of cfDNA in RA. In this study, we hypothesized that cfDNA-mediated activation of autoimmunity and dysfunction of DNA clearance enzymes may be responsible for the progression and development of RA in elderly individuals. To investigate whether the accumulation of DNA fragments can stimulate an abnormal immune response, we first determined the gene expression levels of TREX1 (a DNA fragment clearance enzyme) and cGAS (a DNA fragment sensor) in peripheral blood samples from healthy volunteers and rheumatoid arthritis (RA) and osteoarthritis (OA) patients. As shown in Fig. 1A & B, the gene expression level of TREX1 was significantly reduced but the level of cGAS was markedly increased in RA patients compared with healthy controls (P < 0.01) and patients with OA (P < 0.01), suggesting that DNA fragment accumulation is closely associated with the pathogenesis of RA in humans. Consistent with the observation that the gene expression of TREX1 is associated with the clearance of cytoplasmic DNA fragments and ageing²⁰, Fig. 1C shows that the transcription levels of both TREX1 and cGAS were highly elevated in blood samples from healthy aged rats (> 12 months old) compared to young control rats (4 weeks old). These findings indicate that the level of cytoplasmic DNA fragments increases with age and that the accumulation of DNA fragments may activate the cGAS signalling cascade. In addition, a high level of TREX1 expression is required for the clearance excessive DNA fragments from aged rats, which could prevent an abnormal immune response and maintain the health status of animals.

3.2 Role of TREX1 and cGAS signalling in DNA fragment-induced release of proinflammatory cytokines

The DNA damage-mediated inflammatory response and the inflammatory role of TREX1 and cGAS-STING signalling in RA-FLSs were determined after induction of a UV-mediated DNA damage response (DDR) in RA-FLSs. Upon UV-induced DNA damage, the transcription levels of TREX1 and the regulators of cGAS-STING signalling were significantly increased, with concomitant increases in the levels of proinflammatory cytokines such as TNF-α, IL-1β, IL-2, IL-6, IL-8, IL-25 and IFN-β in a UV exposure time-dependent manner (Fig. S1-A). Of note, silencing of TREX1 gene expression in RA-FLSs further enhanced the UV-mediated production of these proinflammatory cytokines, whereas knockdown of cGAS significantly suppressed the UV-mediated release of these proinflammatory cytokines. These findings indicate that both TREX1 and cGAS-STING signalling participate in the regulation of DDR-induced inflammatory responses. The UV-mediated DDR can result in the production of DNA fragments that induce multiple intracellular reactions ²¹ and inflammatory responses. Therefore, sheared DNA fragments (sonicated DNA from RA-FLSs) were transfected into RA-FLSs to investigate whether the accumulation of DNA fragments can induce an inflammatory response (Fig. S1-B). As shown in Fig. 1D & E, transfection of DNA fragments into RA-FLSs stimulated the expression of the DNA fragment clearance enzymes TREX1 and cGAS-STING, with concomitant upregulation of proinflammatory cytokines, in a dose-dependent manner. However, it was found that DNA fragments increased the expression of TREX1, cGAS-STING and proinflammatory cytokines only within the first 24 h and that the expression of these activated genes then decreased from 48 to 72 h post-transfection. Therefore, it was postulated that DNA fragment-induced upregulation of TREX1 may facilitate the clearance of the accumulated cytoplasmic DNA fragments and reduce the levels of cGAS-STING and proinflammatory cytokines. Consistent with this hypothesis, the Western blot results also confirmed that the addition of DNA fragments into RA-FLSs increased the expression of the TREX1 and cGAS proteins in a time- and dose-dependent manner (Fig. 2A). To reveal the role of TREX1 and cGAS-STING signalling in the DNA fragment-induced inflammatory response, the TREX1 and cGAS were silenced in RA-FLSs with siRNA prior to challenge with DNA fragments. As shown in Fig. 2B, silencing of TREX1 gene expression activated cGAS-STING signalling and further promoted the release of proinflammatory cytokines in RA-FLSs challenged with DNA fragments, whereas knockdown of cGAS significantly downregulated cGAS-STING signalling and suppressed the DNA fragment-induced release of proinflammatory cytokines. These results confirmed that DNA fragments can stimulate inflammation and that both TREX1 and cGAS play a regulatory role in DNA fragment-mediated inflammatory responses.

3.3 Joint injection of DNA fragments promoted inflammatory responses in AIA rats

The adjuvant-induced arthritis (AIA) model is a well-established model for the study of rheumatoid arthritis (RA). Prior to induction of arthritis in rats, joint injection of sonicated rat muscle DNA was performed in AIA rats to induce inflammatory responses. As shown in Fig. S2A & B, immediate effects, such as severe swelling, erythema and joint rigidity, were identified in the hind paws of AIA rats injected with 200 µg of DNA fragments. The drastic increases in the arthritis score and hind paw volume during Days 9–21 indicated that the arthritic conditions in AIA rats injected with DNA fragments were more severe than those in uninjected AIA rats. Moreover, the quantity of circulating free DNA (cfDNA) in serum was found to be significantly higher in both the uninjected AIA rats and the AIA rats injected with 50 or 200 µg of DNA fragments in comparison with the healthy control animals (Fig. S2C). These findings revealed that a high level of cfDNA is commonly found in rats with AIA. Notably, the gene expression of TREX1 was significantly downregulated but that of the regulators of cGAS-STING signalling and proinflammatory cytokines such as TNF-ɑ, IL-1β, IL-2 and IL-6 was significantly upregulated in the peripheral blood of AIA rats with or without injection of DNA fragments (Fig. S2D). Therefore, DNA fragments may promote the onset of arthritis in the AIA rat model. Based on the observation that the injected DNA fragments may be cleared by TREX1 in the rat model prior to the onset of arthritis (Fig. S2D), consistent with the gradual downregulation of TREX1 expression after the onset of AIA in rats, the hind paw volume and the arthritis scores in the AIA + DNA group were similar to those in the AIA group at Day 30. In addition, the proinflammatory effect of intravenous injection of DNA fragments on healthy rats was confirmed by the upregulation of TREX1 expression, the cGAS-STING signalling cascade and proinflammatory cytokine (TNF-ɑ, IL-1β, IL-2 and IL-6) expression for the first 48 h following injection (Fig. S3). Therefore, tail vein injection of DNA fragments was performed at 48 hour intervals after injection of complete adjuvant for the subsequent experiments (Fig. S4A).

3.4 Intravenous injection of DNA fragments exacerbated paw swelling and bone destruction in AIA rats

Based on the experimental conditions optimized in Fig. S4A, the first symptom of hind paw swelling was observable on Day 12 in AIA rats intravenously injected with 100 µg of DNA fragments. This finding suggested that DNA fragments may trigger inflammatory conditions in AIA rats with reduced levels of TREX1. Although no significant differences in the levels of proinflammatory cytokines were found in the peripheral blood of AIA rats when compared to that of AIA rats injected with DNA fragments (on Day 12) (Figure S4B), Micro-CT analysis revealed a low bone mineral density (BMD), cortical mineral density (TMD), trabecular number, and total porosity in the hind limb joints, indicating bone destruction in AIA rats injected intravenously with DNA fragments (Figure S4C). Apparently, the overall radiological and Micro-CT scores demonstrated significant cartilage and bone destruction in AIA rats injected with 100 µg of DNA fragments when compared to rats treated with an equal volume of nucleic acid-free water or with 50 µg of DNA fragments (Figure S4D). Collectively, these results indicate that DNA fragment challenge may promote bone erosion in AIA rats.

3.5 Joint injection of DNA fragments induced arthritis in TREX-1 conditional knockout (TREX1^Cre) rats

To address the role of TREX1 in the pathogenic mechanism of DNA fragments in the AIA model, Cre-loxP-mediated conditional knockout of TREX1 was then accomplished in SD rats (TREX1^Cre) by CRISPR/Cas-mediated genome editing. One week before AIA model establishment, TREX1^Cre rats were injected with 2 different doses of Cre adeno-associated virus (AAV; Cre-H: 2.5x10¹⁰ PFU, Cre-L: 2.5x10⁸ PFU), or vector control AAV in the joint cavity to knock down TREX1 gene expression. As shown in Fig. 3A-C, severe hind paw swelling, erythema and joint rigidity were found in the hind paws of both TREX1^Cre/+ AIA rats and healthy control TREX1^Cre/+ rats injected with 200 µg of DNA fragments in the hind limb joint cavity. The prompt increases in the arthritis score and hind paw volume further indicated that the arthritic condition was significantly exacerbated in AIA model TREX1^Cre/+ rats in comparison with AIA rats after hind limb joint injection of DNA fragments. Apparently, an arthritic condition was successfully induced in healthy control TREX1^Cre/+ rats by hind limb joint injection of DNA fragments, suggesting that knockdown of TREX1 may be a pathogenic risk factor for the development of RA. Micro-CT analysis further revealed severe joint swelling and bone destruction in DNA fragment-injected AIA model TREX1^Cre/+ and healthy control TREX1^Cre/+ rats by direct comparison of bone mineral density (BMD), cortical mineral density (Ct.TMD), bone volume fraction, trabecular number, and total porosity to those in the healthy control group. Indeed, the AIA model TREX1^Cre/+ rats injected with DNA fragments demonstrated the most severe joint swelling and cartilage erosion compared with healthy control TREX1^Cre/+ rats (Fig. 3C & D). Furthermore, histopathological staining indicated that arthritic manifestations – for example, synovial cell hyperplasia with infiltration of neutrophils, lymphocytes and plasma cells; interstitial fibroblast proliferation accompanied by neovascularization; cartilage thickening in joint tissues; white pulp enlargement and fusion; lymphocyte proliferation in the spleen; thickening of the cortex and an increase in T lymphocytes in the thymus; glomerular oedema; and renal fibrosis – were concomitantly observed in the DNA fragment-injected AIA model TREX1^Cre/−group, AIA model TREX1^Cre/+ group and healthy control TREX1^Cre/+ group relative to the healthy control group (Figs. S5A & S6). In addition, immunohistochemical, protein and gene expression analyses revealed that in DNA fragment-injected AIA model TREX1^Cre/+ rats, the lower the TREX1 expression level, the higher were the levels of proinflammatory factors and metalloproteinases (MMP-1, -3, -9) (Figures S5B & S7). Collectively, these results indicate that knockdown or downregulation of TREX1 might increase the severity of the arthritic condition in control and AIA rats with hind limb joint injection of DNA fragments. Our findings suggest that TREX1 may provide a beneficial therapeutic effect in RA through the clearance of cfDNA in joint tissues.

3.6 DNA fragments promoted T-cell activation and decreased Treg cell populations in TREX1^Cre rats

Systemic immune abnormalities are one of the key pathogenic factors for RA development²². In addition to local genetic abnormalities, aberrant activation of CD4^{+ 23}and CD8^{+ 24} lymphocytes is an important manifestation during RA development. In this study, we evaluated whether DNA fragment injection can stimulate T-cell activation in TREX1^Cre rats. To examine the autoimmune responses induced by DNA fragment challenge, sonicated rat muscle DNA fragments were injected into both healthy control TREX1^Cre/+ rats and AIA model TREX1^Cre/+ rats via the tail vein. Consistent with the above results, more severe hind paw swelling, a significantly higher arthritis score, and severe bone destruction were observed in the hind limb joints of healthy control TREX1^Cre/+ rats and AIA model TREX1^Cre/+ rats injected with DNA fragments (Fig. 4A & B). On Day 30, both CD4⁺ and CD8⁺ T lymphocytes were separated and purified from total peripheral blood lymphocytes with antibodies against CD3⁺. As shown in Fig. 4C, flow cytometric analysis showed that the CD8⁺ T lymphocyte population was significantly increased but the Treg cell population was markedly reduced in AIA model TREX1^Cre/− rats in comparison with healthy control rats. Of note, tail vein injection of DNA fragments further increased the number of CD8⁺ T lymphocytes and decreased the Treg cell population in both AIA model TREX1^Cre/+ (Cre H/L) rats and healthy control TREX1^Cre/+ (Cre H/L) rats compared with healthy control TREX1 wild-type rats (Fig. 4C). We therefore postulated that the increase in circulating DNA fragments due to the decrease in TREX1 expression during RA pathogenesis may abnormally activate CD8⁺ lymphocytes and suppress Treg lymphocytes. The proportion of CD8⁺ lymphocytes is significantly elevated in RA patients, and these cells are capable of secreting a variety of proinflammatory cytokines²⁵. Accordingly, both lower TREX1 gene expression and accumulation of DNA fragments might be the crucial participants in the aberrant activation of immune cells during the pathogenesis of RA.

3.7 Adeno associated virus (AAV)-mediated overexpression of TREX1 suppressed synovial inflammation in AIA rats

To further investigate the role of TREX1 in the pathogenesis of AIA in rats, adeno-associated virus expressing TREX1 (AAV-TREX1) was injected into the joint cavity (2.5 x 10¹⁰ PFU or 2.5 x 10⁸ PFU) or tail vein (1.0 x 10¹¹ PFU or 1.0 x 10⁹ PFU) of rats.

As shown in Fig. 5A, severe swelling, erythema and joint rigidity were found in the hind paws of AIA rats injected with AAV-EGFP (vehicle control). In contrast, similar to MTX-treated rats, AIA rats injected with AAV-TREX1 in either the joint cavity or tail vein showed significant decreases in the arthritis score and hind paw volume, suggesting that the severity of AIA was markedly attenuated in rats with overexpression of TREX1. On Days 9 and 12, significant differences in the arthritis score were observed between AAV-TREX1- and AAV-EGFP-injected AIA rats. Extensive accumulation of DNA fragments in serum was apparent in AAV-EGFP-injected AIA rats in comparison with healthy control rats. Importantly, joint cavity or tail vein injection of AAV-TREX1 markedly reduced the serum concentration of DNA fragments in AIA rats (Fig. 5B), indicating that TREX1 inhibits joint swelling possibly via clearance of DNA fragments. Micro-CT analysis of hind paws further demonstrated the beneficial effects of TREX1 in AIA rats (Fig. 5C), and the protective effects on bone cartilage were consistent with the decreasing trends in the arthritis score and hind paw swelling in AIA rats with TREX1 overexpression. Collectively, these findings indicate that TREX1 might play a suppressive role in the development of arthritis in AIA rats via the clearance of DNA fragments.

3.8 AAV-TREX1 suppressed the immune response in AIA rats via activation of the Treg cell population

Given that that reduction in TREX1 expression induced an arthritic phenotype through DNA fragment accumulation and aberrant T-cell activation in the AIA and TREX1^Cre/+ models, whether TREX1 overexpression can reverse the DNA fragment-induced imbalance in the T lymphocyte ratio was further investigated. First, T lymphocytes were isolated from the peripheral blood of AIA rats with or without injection of AAV-TREX1 on Day 28. As shown in Fig. 5D, flow cytometric analysis showed that while the CD8⁺ T lymphocyte population was significantly increased, the Treg cell population was significantly decreased in the AAV-EGFP-injected AIA group. In contrast, AAV-induced overexpression of TREX1 in AIA rats markedly decreased CD8⁺ T lymphocytes and increased Treg cells (Fig. 5D). Notably, Treg cells were highly increased in AIA rats with tail vein injection of AAV-TREX1, suggesting that the abnormal immune response in AIA rats was suppressed by systemic overexpression of TREX1. Accordingly, TREX1 gene expression and DNA fragments might be potential targets for controlling the abnormal activation of immune cells during the pathogenesis of RA.

3.9 Ageing-related cellular senescence and arthritic conditions contributed to the imbalance in c-Jun/c-Fos transcription factor expression and suppression ofTREX1 expression

As TREX1 is a potential protective gene against the adverse effects of DNA damage¹³, the mechanism underlying its downregulation in RA development was further investigated. Based on the above in vivo experimental data, the SD rats in both the AIA and TREX1^Cre/+ models exhibited the senescence-associated secretory phenotype (SASP), and the arthritis condition and ageing phenotype in these rats were reversed, together with reductions in the expression levels of proinflammatory cytokines (Fig. S8A) and ageing markers ²⁶ such as IL-1β, IL-6, IL-8, IFN-β, p16 and p21 (Fig. S8B and S9A), after MTX treatment or AAV-TREX1 injection. Therefore, decreased expression of TREX1 may be correlated with ageing during the onset of RA. E2F1 ²⁰ and AP-1²⁷ are the key important transcription factors of TREX1. Consistent with this observation, our previous studies showed that both the mRNA and protein levels of E2F1 were significantly reduced in AIA rats and DNA fragment-challenged TREX1^Cre/+ rats (Fig. S9B & C); moreover, those of the AP-1 complex subunit c-Jun were increased but those of c-Fos were significantly decreased in various tissues from the same treatment groups (Fig. S9B). In fact, studies have shown that during ageing or certain disease conditions, the reduction in c-Fos expression can lead to a decrease in TREX1 expression²⁷, suggesting that the downregulation of AP-1 in both AIA rats and DNA fragment-challenged TREX1^Cre/+ rats was due to an imbalance in the c-Jun/c-Fos expression ratio. Interestingly, our data showed that TREX1 expression was significantly increased in healthy SD rats after tail vein injection of DNA fragments and that the gene expression levels of c-Jun and c-Fos were also significantly increased in a time-dependent manner (Fig. S9D). These findings were consistent with the data in Fig. 1C, which shows that increased expression of TREX1 is necessary for the clearance of excessive cfDNA from aged rats and DNA fragment-injected healthy rats. To further show the role of E2F1 in TREX1 expression, E2F1 was overexpressed dose-dependently in RA-FLSs, and the transcript levels of TREX1 and DP-1 (an important activator of the E2F transcription factor) were found to be significantly and gradually increased (Fig. 6A). Concomitantly, both UV-mediated DNA damage (Fig. 6B) and direct challenge with DNA fragments (Fig. 6D) induced the expression of E2F1, DP-1, c-Jun, and c-Fos with the upregulation of TREX1 in RA-FLSs in a time-dependent manner.

However, the increase in TREX1 expression was markedly suppressed in the presence of E2F1 siRNA (Fig. 6C & E), indicating that E2F1 is a key transcription factor of TREX1. To further elaborate on the regulatory role of TREX1 during ageing and RA, the expression of TREX1 and the AP-1 complex components (c-Jun/c-Fos) in both early- and late-passage RA-FLSs challenged with DNA fragments was determined. As shown in Fig. 6F, the expression of TREX1 in late-passage of RA-FLSs was significantly lower than that in early-passage RA-FLSs with or without DNA fragment challenge. Of note, c-Fos expression was induced in early-passage RA-FLSs challenged with DNA fragments but not in late-passage RA-FLSs. Furthermore, siRNA knockdown of c-Fos significantly inhibited DNA fragment-mediated TREX1 expression in early-passage RA-FLSs (Fig. 6G). Taken together, these findings suggest that both in ageing-related cellular senescence and in arthritic conditions, the expression of TREX1 is significantly downregulated via the suppression of its transcription factors, such as E2F1 and AP-1, thereby facilitating the accumulation of cfDNA, which in turn activates the autoimmune response. Accordingly, TREX1 dysregulation and cfDNA accumulation might be the crucial risk factors promoting the progression and development of RA.

Ageing is an inevitable process in humans, and it has been associated with the onset of many diseases, including rheumatoid arthritis (RA)²⁸. RA is a chronic, inflammatory systemic disease of unknown aetiology, predominantly affecting the synovium, that can lead to joint destruction and even disability in severe cases¹. Notably, after the age of 50, the immune system undergoes dramatic changes, loses the ability to regulate immunity and gains proinflammatory functions. Accordingly, immunosenescence is particularly accelerated in patients with RA [²⁹. In fact, the ageing process contributes to cellular stress and eventually promotes apoptotic cell death and DNA fragment release ³⁰. Older individuals have been reported to exhibit higher levels of circulating cell-free DNA (cfDNA)³¹. In addition, previous studies have reported that the concentration of cfDNA in the serum of RA patients was significantly reduced after treatment with antiarthritic drugs¹⁹, revealing the possible pathogenic role of cfDNA in RA. Interestingly, our results demonstrated that the gene expression levels of TREX1 (a DNA fragment clearance enzyme) and cGAS (a DNA fragment sensor) in the peripheral blood samples of RA patients were significantly different from those in healthy volunteers. Accordingly, our studies are the first to reveal the relationship between ageing-mediated cfDNA accumulation, TREX1 downregulation RA pathogenesis.

Here, we first revealed the relationship of RA pathogenesis with ageing by detecting cfDNA and injecting DNA fragments in adjuvant-induced arthritis (AIA) models, classical animal models of RA. CfDNA fragments are well-characterized biomarkers for diseases and conditions such as ageing¹¹, systemic lupus erythematosus ³² and tumours ³³; accordingly, they are one of the key factors in the pathogenesis of RA. Indeed, we successfully demonstrated the proinflammatory effect of DNA fragments in AIA rats and further revealed DNA fragment-induced premature joint swelling in AIA rats. We also first generated rats with Cre-LoxP-mediated conditional knockout of TREX1 (TREX1^Cre+/− rats) and revealed that DNA fragments promoted proinflammatory cytokine release and T-cell activation in rats when the DNA-metabolizing enzyme TREX1 was deleted, which is in line with in vitro studies showing that knockdown of TREX1 promoted the release of proinflammatory cytokines in RA-FLSs challenged with DNA fragments. Most importantly, analysis of serum from the AIA rat models revealed a significant increase in the DNA fragment concentration in serum as well as a marked decrease in TREX1 expression. These findings suggested that DNA fragments may be a potential immunogenic factor for the induction of RA development. On the other hand, adeno-associated virus (AAV)-mediated overexpression of TREX1 in AIA rats was linked to a reduction in serum DNA fragments and a significant improvement in arthritis symptoms, for example, decreases in the arthritis score and hind paw swelling index, improvement in the Micro-CT analysis index, and an increase in Treg cells. Collectively, our data emphasized that TREX1 or DNA fragments may be promising targets for RA treatment.

Accumulating evidence indicates that the secretion of inflammatory mediators is a distinctive feature of ageing³⁴. For example, cell cycle arrest in senescent cells³⁵, the loss of tissue repair capacity³⁶, and the release of proinflammatory and matrix degradation factors ³⁷ are the common manifestations of the senescence-associated secretory phenotype (SASP)³⁸. Notably, senescent cells exhibiting the SASP phenotype can promote various chronic diseases and conditions, including cardiovascular disease, metabolic syndrome, and neurocognitive decline, which are accompanied by the release of growth factors (such as TGF-α and IL-1), expression of transcription factors (such as GATA4 and NF-κB), and release of proinflammatory cytokines (such as IL-6 and IL-8), proteases and matrix metalloproteinases³⁹. Numerous clinical studies of elderly individuals indicated that the levels of several cytokines, especially IL-6 and TNF-α, increase with age even in healthy individuals and in elderly individuals without acute infection; this characteristic is also termed inflammatory ageing⁶. Therefore, systemic inflammation is a developmental process related to ageing and ageing-related diseases. Inflammation is considered part of the normal repair response, and it is essential for the defence against bacterial and viral infections and harmful environmental factors⁴⁰. However, the inflammatory process can also become destructive to organisms when it is prolonged and persistent. The SASP is a proinflammatory phenotype that accompanies mammalian ageing; however, the causative factors of this phenotype are not known⁴¹. IFN-β expression was found to be elevated in our DNA fragment-injected AIA model (Fig. S2); IFN-β is also a factor contributing to the SASP and has been reported to stimulate the DNA damage response (DDR) via the induction of ROS production²⁰, which leads to senescence-like cell cycle arrest in human diploid fibroblasts (HDFs)⁴². IL-6, a common proinflammatory cytokine in RA, is another major factor contributing to the SASP; it can also increase intracellular ROS levels and cause DNA damage⁴³. Importantly, antibodies against IL-6 or its receptor (tocilizumab) have become effective targeted drugs prescribed for the clinical treatment of RA⁴⁴. These findings indicate the strong correlations between ageing and RA. Intriguingly, our study demonstrated that the expression of the cell cycle activator E2F1 ⁴⁵ was reduced but that of cell senescence-related genes such as p16 and p21 ⁴⁶ was elevated in both the AIA and TREX1^Cre/+ rat models, with concomitant elevation of senescence factors such as IL-1β, IL-6, IL-8 and IFN-β, suggesting that the SASP was induced in both AIA and TREX1^Cre/+rats. Accordingly, we hypothesized that cell senescence-induced DNA fragmentation would increase the DDR and trigger positive feedback activation of DNA fragment accumulation, thereby accelerating the development of pathological ageing as well as the pathogenesis of RA. However, whether the downregulation of TREX1 in RA is a cause or a consequence of SASP induction requires further investigation in the future.

AP-1 is one of the transcription factors of TREX1 and is a dimeric complex composed of two important protein subunits, Jun and Fos ⁴⁷. The balance of Jun and Fos expression levels is important for the different transcriptional functions of AP-1, which include cell transformation⁴⁸, angiogenesis⁴⁹, and tumour invasion⁵⁰. It has been shown that the expression of c-FOS is significantly reduced after fibroblast senescence, which leads to a decrease in TREX1 expression²⁷. Consistent with these findings, we demonstrated that AP-1 exhibited elevated expression in wild-type rats after stimulation with DNA fragments, whereas it exhibited lower expression in TREX1Cre rats and AIA rats owing to the imbalance in the c-Jun/c-Fos expression ratio. Accordingly, we speculated that the reason for the high incidence of RA in middle-aged and elderly patients may be due to the imbalance of Jun and Fos in senescent cells, resulting in a reduction in TREX1 expression. Consequently, the decreased expression of TREX1 may contribute to abnormal DNA metabolism and the proinflammatory effect of cfDNA accumulation, eventually driving normal cells to gradually acquire the SASP in inflammatory environments. Interestingly, our results further confirmed that AAV-mediated overexpression of TREX1 markedly reversed the accumulation of cfDNA, suppressed the release of inflammatory cytokines, and ameliorated synovial tissue damage. In summary, we are the first to demonstrate that DNA fragments could be a potential immunogenic factor for the induction of RA development. Downregulation of TREX1 promotes the release of proinflammatory cytokines and possibly contributes to the accumulation of DNA fragments in patients with RA, thereby further aggravating the inflammatory response via the activation of cGAS/STING signalling. Our findings also link autoimmune tissue inflammation to ageing and define effective clearance of DNA fragments as an important strategy for the management and prevention of RA and ageing-related diseases.

Acknowledgements

This research was supported by a FDCT grant from the granted project of Dr. Neher’s Biophysics Laboratory for Innovative Drug Discovery from the Macao Science and Technology Development Fund (Project code: 0033/2019/AFJ; 001/2020/ALC), and Foshan Medicine Dengfeng Project of China (2019–2021)

Author contributions

Vincent Kam Wai Wong, Betty Yuen Kwan Law and Liang Liu conceived and oversaw the project, Weidan Luo and Lijun Yang interpreted all the results and wrote the manuscript. Weidan Luo performed all in vitro experiment and linked clinical information with experimental datasets. Yaling Zeng, Ruilong Zhang, Linlin Song, Baixiong Huang and Xiaoyun Yun performed the AIA model experiment in normal rats. Weidan Luo, Lijun Yang, Yuanqing Qu, Zicong Lin, Xiongfei Xu, Linna Wang, Ruihong Chen and Jiujie Yang performed the TREX1 konckdown model experiment in CKO rats and overexpression animal model by used AAV-TREX1. Jun Lv helped with the H&E staining, immunohistochemistry, interpretation of data and writing of the manuscript. Xi Chen, Wei Zhang, Kaixi Zhang, Huimiao Wang, Xingxia Wang, Liqun Qu, Menghan Liu and Yuping Wang performed collect animal samples. All authors approved the manuscript.

Competing interests: The authors declare no competing interests.

Smolen, J.S., Aletaha, D. & McInnes, I.B. Rheumatoid arthritis. Lancet 388, 2023–2038 (2016).
Wong, V.K.W. Editorial overview: New technologies in 2020: Drug resistance. Current opinion in pharmacology 54, iii-ivi (2020).
Myasoedova, E., Davis, J., Matteson, E.L. & Crowson, C.S. Is the epidemiology of rheumatoid arthritis changing? Results from a population-based incidence study, 1985–2014. Annals of the Rheumatic Diseases 79, 440–444 (2020).
Tchkonia, T., Zhu, Y., van Deursen, J., Campisi, J. & Kirkland, J.L. Cellular senescence and the senescent secretory phenotype: therapeutic opportunities. Journal of Clinical Investigation 123, 966–972 (2013).
Tchkonia, T. & Kirkland, J.L. Aging, Cell Senescence, and Chronic Disease Emerging Therapeutic Strategies. Jama-Journal of the American Medical Association 320, 1319–1320 (2018).
Ferrucci, L. & Fabbri, E. Inflammageing: chronic inflammation in ageing, cardiovascular disease, and frailty. Nature Reviews Cardiology 15, 505–522 (2018).
Jeon, O.H., et al. Local clearance of senescent cells attenuates the development of post-traumatic osteoarthritis and creates a pro-regenerative environment. Nature Medicine 23, 775-+ (2017).
Safiri, S., et al. Global, regional and national burden of rheumatoid arthritis 1990–2017: a systematic analysis of the Global Burden of Disease study 2017. Annals of the Rheumatic Diseases 78, 1463–1471 (2019).
Montero-Melendez, T., et al. Therapeutic senescence via GPCR activation in synovial fibroblasts facilitates resolution of arthritis. Nature Communications 11(2020).
Ou, H.L. & Schumacher, B. DNA damage responses and p53 in the aging process. Blood 131, 488–495 (2018).
Teo, Y.V., et al. Cell-free DNA as a biomarker of aging. Aging Cell 18(2019).
Gravina, S., Sedivy, J.M. & Vijg, J. The dark side of circulating nucleic acids. Aging Cell 15, 398–399 (2016).
Gao, D.X., et al. Activation of cyclic GMP-AMP synthase by self-DNA causes autoimmune diseases. Proceedings of the National Academy of Sciences of the United States of America 112, E5699-E5705 (2015).
Duvvuri, B. & Lood, C. Cell-Free DNA as a Biomarker in Autoimmune Rheumatic Diseases. Frontiers in Immunology 10(2019).
Boots, A.M.H., et al. The influence of ageing on the development and management of rheumatoid arthritis. Nature Reviews Rheumatology 9, 604–613 (2013).
van der Pol, Y. & Mouliere, F. Toward the Early Detection of Cancer by Decoding the Epigenetic and Environmental Fingerprints of Cell-Free DNA. Cancer Cell 36, 350–368 (2019).
Yang, Y., et al. 5-Hydroxymethylcytosines in Circulating Cell-Free DNA Reveal Vascular Complications of Type 2 Diabetes. Clinical Chemistry 65, 1414–1425 (2019).
Hartl, J., et al. Autoantibody-mediated impairment of DNASE1L3 activity in sporadic systemic lupus erythematosus. Journal of Experimental Medicine 218(2021).
Laukova, L., et al. Anti-cytokine therapy and plasma DNA in patients with rheumatoid arthritis. Rheumatology International 38, 1449–1454 (2018).
Takahashi, A., et al. Downregulation of cytoplasmic DNases is implicated in cytoplasmic DNA accumulation and SASP in senescent cells. Nature Communications 9(2018).
Bernard, J.J., Gallo, R.L. & Krutmann, J. Photoimmunology: how ultraviolet radiation affects the immune system. Nature Reviews Immunology 19, 688–701 (2019).
McInnes, I.B. & Schett, G. MECHANISMS OF DISEASE The Pathogenesis of Rheumatoid Arthritis. New England Journal of Medicine 365, 2205–2219 (2011).
Oh, S., Rankin, A.L. & Caton, A.J. CD4 + CD25 + regulatory T cells in autoimmune arthritis. Immunological Reviews 233, 97–111 (2010).
Petrelli, A. & van Wijk, F. CD8(+) T cells in human autoimmune arthritis: the unusual suspects. Nature Reviews Rheumatology 12, 421–428 (2016).
Carvalheiro, H., da Silva, J.A.P. & Souto-Carneiro, M.M. Potential roles for CD8(+) T cells in rheumatoid arthritis. Autoimmunity Reviews 12, 401–409 (2013).
Birch, J. & Gil, J. Senescence and the SASP: many therapeutic avenues. Genes & Development 34, 1565–1576 (2020).
Christmann, M., Tomicic, M.T., Aasland, D., Berdelle, N. & Kaina, B. Three prime exonuclease I (TREX1) is Fos/AP-1 regulated by genotoxic stress and protects against ultraviolet light and benzo(a)pyrene-induced DNA damage. Nucleic Acids Research 38, 6418–6432 (2010).
Borghesan, M., Hoogaars, W.M.H., Varela-Eirin, M., Talma, N. & Demaria, M. A Senescence-Centric View of Aging Implications for Longevity and Disease. Trends in Cell Biology 30, 777–791 (2020).
Weyand, C.M. & Goronzy, J.J. Immunometabolism in early and late stages of rheumatoid arthritis. Nature Reviews Rheumatology 13, 1–11 (2017).
Jylhava, J., et al. Characterization of the role of distinct plasma cell-free DNA species in age-associated inflammation and frailty. Aging Cell 12, 388–397 (2013).
Jylhava, J., et al. Aging is associated with quantitative and qualitative changes in circulating cell-free DNA: The Vitality 90 + study. Mechanisms of Ageing and Development 132, 20–26 (2011).
Xu, Y.L., et al. High levels of circulating cell-free DNA are a biomarker of active SLE. European Journal of Clinical Investigation 48(2018).
Jung, K., Fleischhacker, M. & Rabien, A. Cell-free DNA in the blood as a solid tumor biomarker-A critical appraisal of the literature. Clinica Chimica Acta 411, 1611–1624 (2010).
Lasry, A. & Ben-Neriah, Y. Senescence-associated inflammatory responses: aging and cancer perspectives. Trends in Immunology 36, 217–228 (2015).
Childs, B.G., et al. Senescent cells: an emerging target for diseases of ageing. Nature Reviews Drug Discovery 16, 718–735 (2017).
He, S.H. & Sharpless, N.E. Senescence in Health and Disease. Cell 169, 1000–1011 (2017).
Acosta, J.C., et al. Chemokine signaling via the CXCR2 receptor reinforces senescence. Cell 133, 1006–1018 (2008).
Salama, R., Sadaie, M., Hoare, M. & Narita, M. Cellular senescence and its effector programs. Genes & Development 28, 99–114 (2014).
Loeser, R.F., Collins, J.A. & Diekman, B.O. Ageing and the pathogenesis of osteoarthritis. Nature Reviews Rheumatology 12, 412–420 (2016).
Rathinam, V.A.K. & Chan, F.K.M. Inflammasome, Inflammation, and Tissue Homeostasis. Trends in Molecular Medicine 24, 304–318 (2018).
Herranz, N., et al. mTOR regulates MAPKAPK2 translation to control the senescence-associated secretory phenotype. Nature Cell Biology 17, 1205-+ (2015).
Casella, G., et al. Transcriptome signature of cellular senescence. Nucleic Acids Research 47, 7294–7305 (2019).
Wassmann, S., et al. Interleukin-6 induces oxidative stress and endothelial dysfunction by overexpression of the angiotensin II type 1 receptor. Circulation Research 94, 534–541 (2004).
Nishimoto, N., et al. Mechanisms and pathologic significances in increase in serum interleukin-6 (IL-6) and soluble IL-6 receptor after administration of an anti-IL-6 receptor antibody, tocilizumab, in patients with rheumatoid arthritis and Castleman disease. Blood 112, 3959–3964 (2008).
Emanuele, M.J., et al. Complex Cartography: Regulation of E2F Transcription Factors by Cyclin F and Ubiquitin. Trends in Cell Biology 30, 640–652 (2020).
Barnes, P.J., Baker, J. & Donnelly, L.E. Cellular Senescence as a Mechanism and Target in Chronic Lung Diseases. American Journal of Respiratory and Critical Care Medicine 200, 556–564 (2019).
Karin, M., Liu, Z.G. & Zandi, E. AP-1 function and regulation. Current Opinion in Cell Biology 9, 240–246 (1997).
Maurus, K., et al. The AP-1 transcription factor FOSL1 causes melanocyte reprogramming and transformation. Oncogene 36, 5110–5121 (2017).
Yoshitomi, Y., Ikeda, T., Saito-Takatsuji, H. & Yonekura, H. Emerging Role of AP-1 Transcription Factor JunB in Angiogenesis and Vascular Development. International Journal of Molecular Sciences 22(2021).
Chen, G.L., et al. Fra-2/AP-1 regulates melanoma cell metastasis by downregulating Fam212b. Cell Death and Differentiation 28, 1364–1378 (2021).

SupportingInformation.docx

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Self-DNA accumulation as a risk factor for accelerating the pathogenesis of rheumatoid arthritis in elderly individuals

Archived Versions:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Cell culture & antibodies

Extraction and assay kits

Clinical samples

DNA fragment preparation & DNA damage model

siRNA transfection

Gene expression analysis by real-time PCR

Western blotting

SD rats used for establishment of the experimental arthritis model (AIA) and treatment

Model 1 (Inflammatory potency of DNA fragment injection in the AIA model)

Model 2 (monitoring proinflammatory cytokine levels in AIA rats injected with DNA fragments)

Model 4 (evaluating the anti-inflammatory effect of TREX1 overexpression in AIA rats)

Joint swelling measurement and clinical scoring

Micro-CT analysis

Cell-free DNA (cfDNA) extraction and measurement

Haematoxylin and eosin (H&E) staining and immunohistochemistry (IHC) of rat tissue samples

Generation of TREX1 conditional knockout (Cko TREX1-/-) SD rats

Genotyping of TREX1-deficient rats

Establishment of the AIA model using Cko TREX1-/- rats

Model 1 (Hind paw joint injection of Cre adeno associated virus into Cko TREX1-/- rats)

Model 2 (tail vein injection of Cre adeno associated virus into Cko TREX1-/- rats)

Lymphocyte isolation, intracellular cytokine staining, and flow cytometric analysis

Statistical analysis

Results

3.2 Role of TREX1 and cGAS signalling in DNA fragment-induced release of proinflammatory cytokines

3.3 Joint injection of DNA fragments promoted inflammatory responses in AIA rats

3.4 Intravenous injection of DNA fragments exacerbated paw swelling and bone destruction in AIA rats

3.5 Joint injection of DNA fragments induced arthritis in TREX-1 conditional knockout (TREX1Cre) rats

3.6 DNA fragments promoted T-cell activation and decreased Treg cell populations in TREX1Cre rats

3.7 Adeno associated virus (AAV)-mediated overexpression of TREX1 suppressed synovial inflammation in AIA rats

Discussion

Declarations

References

Supplementary Files

Archived Versions:

Version 1

Establishment of the AIA model using Cko TREX1^-/- rats

Model 1 (Hind paw joint injection of Cre adeno associated virus into Cko TREX1^-/- rats)

3.5 Joint injection of DNA fragments induced arthritis in TREX-1 conditional knockout (TREX1^Cre) rats

3.6 DNA fragments promoted T-cell activation and decreased Treg cell populations in TREX1^Cre rats