key: cord-0331495-ejrdhlr3
authors: Giordano, Luca; Gregory, Alyssa D.; Verdaguer, Mireia Perez; Ware, Sarah A.; Harvey, Hayley; DeVallance, Evan; Brzoska, Tomasz; Sundd, Prithu; Zhang, Yingze; Sciurba, Frank C.; Shapiro, Steven D.; Kaufman, Brett A.
title: Extracellular release of mitochondrial DNA is triggered by cigarette smoke and is detected in COPD
date: 2021-10-04
journal: bioRxiv
DOI: 10.1101/2021.10.04.462069
sha: cd97734c039eadbc6601f2d6161ec47acd563dcb
doc_id: 331495
cord_uid: ejrdhlr3

Chronic obstructive pulmonary disease (COPD) is characterized by continuous and irreversible inflammation frequently caused by persistent exposure to toxic inhalants such as cigarette smoke (CS). CS may trigger mitochondrial DNA (mtDNA) extrusion into the cytosol, extracellular space, or foster its transfer by extracellular vesicles (EVs). The present study aimed to elucidate whether mtDNA is released upon CS exposure and in COPD. We measured cell-free mtDNA (cf-mtDNA) in the plasma of former smokers affected by COPD, in the serum of mice that developed CS-induced emphysema, and in the extracellular milieu of human bronchial epithelial cells exposed to cigarette smoke extract (CSE). Further, we characterized cells exposed to sublethal and lethal doses of CSE by measuring mitochondrial membrane potential and dynamics, superoxide production and oxidative stress, cell cycle progression, and cytokine expression. Patients with COPD and mice that developed emphysema showed increased levels of cf-mtDNA. In cell culture, exposure to a sublethal dose of CSE decreased mitochondrial membrane potential, increased superoxide production and oxidative damage, dysregulated mitochondrial dynamics, and triggered mtDNA release in extracellular vesicles. The release of mtDNA into the extracellular milieu occurred concomitantly with increased expression of DNase III, DNA-sensing receptors (cGAS, NLRP3), proinflammatory cytokines (IL-1β, IL-6, IL-8, IL-18, CXCL2), and markers of senescence (p16, p21). Exposure to a lethal dose of CSE preferentially induced mtDNA and nuclear DNA release in cell debris. Our findings demonstrate that CS-induced stress triggers mtDNA release and is associated with COPD, supporting cf-mtDNA as a novel signaling response to CS exposure.

Necroptosis has been recently emphasized among several maladaptive 187 pathways involved in COPD pathogenesis (7, 21), including senescence (22), 188 autophagy (23, 24), and apoptosis (24). Necroptosis is a caspase-independent cell 189 death process that promotes a controlled cell membrane lysis and may facilitate DNA 190 release into the extracellular space (25). From lung tissue collected from COPD 191 patients or donors without signs of obstruction, we measured protein levels of the 192 upstream regulators of necroptosis, Receptor-Interacting serine/threonine Protein 193 Kinase 1 and 3 (RIPK1 and RIPK3, respectively). We did not observe a significant 194 change in RIPK1 between the two groups. However, we measured a two-fold increase 195 of RIPK3 in the COPD group compared to the control ( Figure To test the cf-mtDNA levels in an animal model, we exposed mice to CS for six 202 months and quantified the lung damage by measuring the mean alveolar chord length. 203

As expected, we found enlarged alveolar airspace (emphysema) in CS-exposed mice 204 compared to the room air-exposed group ( Figure 1C-D) . In the same cohort of mice, 205 we measured cf-mtDNA and cf-nDNA levels in the serum. CS-exposed mice showed 206 almost three-fold higher cf-mtDNA than the mice exposed to room air ( Figure 1E) . 207

The cf-nDNA levels followed a similar trend but did not reach statistical significance 208 ( Figure 1F ). The necroptotic RIPK1 and RIPK3 proteins were higher in the lung 209 lysates of CS-exposed mice than room air controls ( Figure E2D-F To identify molecular mechanisms underlying our in vivo observations, we employed 216 an in vitro cell model to study CS effects. Specifically, human bronchial epithelial 217 cells (BEAS-2B) were exposed to increasing doses of CSE for 24 hours. 218

Sulforhodamine B assay revealed that CSE toxicity is dose-dependent (Figure 2A) . 219

Increased cell death was observed at 20% and 30% CSE, as evidenced by the rounded 220 and shrinking cell morphology. Cells incubated with 10% CSE were spared this 221 severe effect with only a 22% decrease in cell density compared to unexposed cells 222 (Figure 2A-B) . To test whether cell death was related to the genotoxic effect of CSE, 223 we measured de novo genomic DNA synthesis by 5-ethynyl-2'-deoxyuridine (EdU) 224 incorporation during S-phase. DNA synthesis was almost absent (75.8% decline 225 relative to unexposed) in cells exposed to 20% CSE, whereas cells exposed to 10% 226 CSE showed only a 17.5% decline ( Figure 2C ). Because DNA double-strand breaks 227 (DSB) induce permanent growth arrest and cell death, we measured the number of 228 p53-binding protein 1 (53BP1) foci, a protein involved in DSB signaling and repair 229 (26). Cells exposed to 20% CSE showed more than double the foci per nucleus than 230 the control (Figure 2D-E) . On the contrary, 10% CSE promoted a selective 231 recruitment of 53BP1, as indicated by the increased volume of the foci ( Figure 2D, F To test whether DNA is released into the extracellular milieu of bronchial epithelial 239 cells, we exposed cells to increasing doses of CSE for 24 hours, and from the growth 240 medium, we collected extracellular vesicles (EVs) and cell debris. Nanoparticle 241 tracking analysis of the EVs showed no variation in the abundance or size of the 242 particles among the control, 5, and 10% CSE preparations. In contrast, the EVs from 243 20% CSE preparations were significantly more abundant (1.9x10 9 versus 9.31x10 8 244 particles per ml) and slightly larger than the control (150.5 nm versus 120. patients and mice exposed to CS ( Figure E2) . 297

Because we observed replication inhibition in cells exposed to CSE ( Figure  298 2C), we next tested whether markers of senescence were activated. First, we analyzed 299 cell senescence by β-galactosidase (β-gal) staining. Cells exposed to 5-20% CSE 300

showed an increased number of β-galactosidase positive cells than unexposed cells 301 ( Figure 4H -I), with 10% showing peak levels. To confirm replicative senescence in 302 the 10% cells, we measured the proteins levels of the Cyclin-dependent Kinase 303

Inhibitor 1A (p21) as a marker of cell cycle arrest. Western blotting showed a 2.8-fold 304 increase of p21 in cells exposed to 10% CSE compared to control ( Figure 4J-K) . 305

Both p21 and Cyclin-dependent Kinase Inhibitor 2A (p16) mRNAs were also 306 increased ( Figure 4L ) demonstrating a transcriptional response to block cell cycle 307 progression in cells exposed to 10% CSE. These results suggest that exposure to a 308 sublethal dose of CSE induced mtDNA release in EVs and promotes cellular 309 senescence without activation of necroptosis. 310 14 311

Mitochondrial dynamics are altered in cells exposed to a sublethal dose of CSE 312 and in COPD lung tissue 313

Perturbations to mitochondrial dynamics may also favor mtDNA release (27). To test 314 whether the mitochondrial structure was altered, we performed morphometric 315 analysis, using TOMM20 immunostaining to visualize the mitochondrial network. 316

Morphometry showed that the number of mitochondria per cell did not change at 10% 317 CSE ( Figure 5A-B) , in agreement with the observation that intracellular mtDNA 318 content (mtDNA/nDNA) and TFAM (whose levels usually correlate with mtDNA 319 abundance) were also unchanged ( Figure E3F-H) . However, the total mitochondrial 320 area and perimeter per cell, and the mean area, perimeter, aspect ratio, and form factor 321 per mitochondrion were increased ( Figure 5A, C-H) . These results indicate that 322 exposure to a sublethal dose of CSE promotes mitochondrial enlargement and 323 elongation. 324

To enhance our understanding of this process, we quantified protein markers 325 of mitochondrial fusion, including the long-form of Optic Atrophy 1 (L-OPA1), 326

Mitofusin-1 and -2 (MFN1 and MFN2), which are involved in tethering the inner and 327 outer mitochondrial membranes (31). L-OPA1, MFN1, and MFN2 showed a slight 328 but significant decrease in cells exposed to 10% CSE ( Figure 5I -J, L-M) whereas no 329 changes were observed for the short form of OPA1 (S-OPA1, Figure 5K) . Similarly, 330 MFN1 and MFN2 proteins were decreased in COPD lung samples compared to the 331 lung tissue of donors without signs of obstruction, confirming the dysregulated 332 mitochondrial dynamics in the pathological tissues ( Figure 5N-P) . These findings 333 reveal that alteration of the mitochondrial dynamics is a relevant process in cells 334 exposed to CS and in COPD lung tissue, and could contribute to mtDNA release. 335 showed an almost 50% increase in cells exposed to 10% CSE compared to unexposed 345 cells ( Figure 6A) , suggesting the activation of a cytosolic mechanism to digest 346 mislocalized DNA. 347

To identify possible receptors for mislocalized mtDNA, we evaluated the 348 expression of DSRs mainly localized to the cytoplasm that could trigger an 349 inflammatory response caused by CS exposure, specifically cGAS and NLRP3. 350 Furthermore, we measured TLR4, which is involved in mtDNA fragmentation (15), 351 and whose deficiency causes age-dependent emphysema (33). cGAS, NLRP3, and 352 TLR4 expression increased in cells incubated with 10% CSE compared to unexposed 353 cells ( Figure 6B) inflammatory cell infiltration and alveolar enlargement. In cells exposed to 10% CSE, 375 IL-1 and IL-6 were strongly upregulated while IL-18 was increased to a lesser extent 376 ( Figure 6D ). Overall, these data indicate that exposure to a sublethal CSE dose 377 induces the transcription of cytokines and chemokines, suggesting an active role for 378 bronchial epithelial cells in recruiting neutrophils and macrophages. 379 380 Discussion 381 COPD is responsible for three million deaths worldwide each year, is caused mostly 382 by cigarette smoking, and has no resolutive therapeutics (1, 2). Excessive airway 383 inflammation and remodeling remain even after several years of smoking cessation, 384 suggesting autoimmunity as a significant driver of the ongoing processes (3, 4). In the 385 17 last decade, mislocalized mtDNA has been shown to promote inflammatory signaling 386 (11) (12) (13) (14) . In this study, we wanted to determine whether CS exposure triggers mtDNA 387 release, and whether it is detected in COPD. Measuring extracellular mtDNA and 388 understanding its role may identify novel therapeutic targets for smokers and COPD 389 patients. 390

Our data revealed elevated levels of cf-mtDNA in the plasma of former 391 smokers with COPD ( Figure 1A-B) and in the serum of mice with CS-induced 392 enlarged airspace (Figure 1C-F; supplemental discussion_1) . To understand the 393 mechanism of its release, we exposed BEAS-2B cells to a sublethal dose of (10%) 394 CSE (Figure 2A-B) , and observed decreased mitochondrial membrane potential, 395 increased oxidative stress (Figure 4A-G) , and altered mitochondrial dynamics 396 ( Figure 5A-M) . These changes occurred concomitantly with replicative senescence, 397 as demonstrated by the expression of senescent markers and cell cycle inhibition 398 ( Figure 2C, 4H-L) . Noteworthy, we described two ways of mtDNA extrusion, by 399

EVs and cell debris (Figure 3B-E) . EVs showed a relative increase in mtDNA 400 content over nDNA with increasing CSE doses, while cell debris showed a relative 401 decrease ( Figure 3D, G) , suggesting a different paradigm of release and signaling. 402

Necroptosis has been shown as a driving mechanism of cell death caused by 403 exposure to high doses of (16-20%) CSE, and in the lung of COPD patients (7, 21). 404

The upregulation of RIPK3 in COPD lungs and RIPK1 and RIPK3 in CS-exposed 405 murine lungs ( Figure E2 ) suggest that necroptosis is a possible mechanism of 406 mtDNA release. However, RIPK1 and RIPK3 were not upregulated in cells exposed 407 to a sublethal (10%) dose of CSE ( Figure E3 ). This is consistent with a recent study 408 that used the same cell line, and in which necroptosis was not driven by upregulation 409 In our in vitro model (Figure 7) , CSE exposure induced nDNA release 425 predominantly by cell debris (Figure 3C, F) , supporting the idea that it is released by 426 dead cells, as observed in neutrophils (40). On the contrary, mtDNA was released by 427 cell debris and EVs ( Figure 3B, E) , trended with decreased mitochondrial membrane, 428 increased superoxide, and protein oxidation (Figure 4) . In cells exposed to a sublethal 429 dose of CSE, the increased mtDNA in the EVs did not influence the intracellular 430 mtDNA content and TFAM levels ( Figure E3E-H) , suggesting that the amount of cf-431 mtDNA released is insufficient to cause detectable depletion among the hundreds of 432 mtDNA copies per cell. Furthermore, at the same sublethal dose of (10%) CSE, 35% 433 of these cells were senescent ( Figure 4I) . Senescence has been shown to increase the 434 mitochondrial mass (41, 42) and would be expected to offset cellular decreases of 435 19 mtDNA content. 436

Interestingly, BEAS-2B cells have been shown to increase exosome release 437 when exposed to CSE. This process is driven by the thiol-reactive properties of CSE, 438 especially acrolein, which may deplete the free thiol group of glutathione (43). 439

Oxidative stress may also alter mitochondrial dynamics, which is one of the 440 mechanisms believed to exacerbate mtDNA packaging into EVs (27). In this regard, 441 we found decreased MFN1, MFN2, and L-OPA in BEAS-2B exposed to a sublethal 442 dose of CSE and in emphysematous human lung tissues ( Figure 5I In conclusion, we demonstrated that CS exposure triggers extracellular 497 mtDNA release in BEAS-2B cells and in a mouse model of emphysema, and we 498 showed that cf-mtDNA levels are elevated in former smokers with COPD. 499

Understanding the mechanism of mtDNA extrusion, and its extracellular role may 500 identify novel therapeutic targets for smokers and COPD patients. Bowler for editorial support. 508

Author disclosures are available with the text of this article at www.atsjournals.org. 509 510

Projections of global mortality and burden of disease from 2002 to 512 2030

Global strategy for the diagnosis, management, and prevention of 515 chronic obstructive lung disease 2017 report. Gold executive summary

Immunologic aspects of chronic obstructive pulmonary 518 disease

The cytokine network in chronic obstructive pulmonary disease

Prolonged cigarette smoke exposure alters mitochondrial 523 structure and function in airway epithelial cells

Cigarette smoke-induced blockade of the mitochondrial respiratory 526 chain switches lung epithelial cell apoptosis into necrosis

Mitophagy-dependent necroptosis contributes to the pathogenesis 530 of copd

Mitochondrial iron chelation ameliorates cigarette smoke-533 induced bronchitis and emphysema in mice

Alternative oxidase attenuates cigarette 536 smoke-induced lung dysfunction and tissue damage

Mitochondrial transfer of induced pluripotent stem cell-derived mesenchymal stem cells to airway 539 epithelial cells attenuates cigarette smoke-induced damage

Mitochondrial dna in innate immune responses and inflammatory 542 pathology

Innate immunity and tolerance toward mitochondria

Autophagy proteins regulate innate immune responses by inhibiting the 547 release of mitochondrial dna mediated by the nalp3 inflammasome

Mitochondrial dna that escapes from autophagy causes inflammation 550 and heart failure

Cold-inducible rna-binding protein through tlr4 signaling induces mitochondrial dna fragmentation and 553 regulates macrophage cell death after trauma

Mitochondrial dna damage is more extensive and persists longer 557 than nuclear dna damage in human cells following oxidative stress

Mitochondrial 560 genome damage associated with cigarette smoking

Oxidative dna damage in lung tissue from patients with copd is clustered in functionally significant 563 sequences

Elevated plasma level of pentraxin 3 is associated with emphysema and mortality in 566 smokers

Rip3-dependent necroptosis contributes to the pathogenesis of chronic obstructive 569 pulmonary disease

Targeting p16-induced senescence prevents cigarette smoke-induced 572 emphysema by promoting igf1/akt1 signaling in mice

Cigarette smoke-574 induced autophagy impairment accelerates lung aging, copd-emphysema exacerbations and 575 pathogenesis

Autophagy protein microtubule-associated protein 1 light chain-3b (lc3b) activates extrinsic 578 apoptosis during cigarette smoke-induced emphysema

The molecular machinery of 581 regulated cell death

Push back to respond better: Regulatory inhibition of the dna double-583 strand break response

Mechanisms of mitochondrial dna escape and its 585 relationship with different metabolic diseases

Redox control of protein degradation

P62/sqstm1 -steering the cell through health and disease

Insufficient autophagy promotes bronchial epithelial cell senescence in chronic 593 obstructive pulmonary disease

Mitochondrial fusion: Reaching the end of mitofusin's tether

Dna sensing by the cgas-sting pathway in health and 597 disease

Toll-like receptor 4 deficiency causes pulmonary 599 emphysema

The neutrophil in chronic obstructive pulmonary disease

Differences in interleukin-8 and tumor 603 necrosis factor-alpha in induced sputum from patients with chronic obstructive pulmonary disease or 604 asthma

Smoke extract stimulates lung epithelial cells to release neutrophil and monocyte chemotactic activity

Prognostic utility of admission cell-free dna levels in patients with chronic obstructive pulmonary 610 disease exacerbations

Association of plasma mitochondrial dna with copd severity and 613 progression in the spiromics cohort

Association of urine mitochondrial dna with clinical 616 measures of copd in the spiromics cohort

Cigarette smoke-induced damage-619 associated molecular pattern release from necrotic neutrophils triggers proinflammatory mediator 620 release

Shanley 622 DP. Dynamic modelling of pathways to cellular senescence reveals strategies for targeted interventions

Mitochondria in cell senescence: Is 625 mitophagy the weakest link?

Cigarette smoke extract induced exosome 628 release is mediated by depletion of exofacial thiols and can be inhibited by thiol-antioxidants

During autophagy mitochondria elongate, are spared 631 from degradation and sustain cell viability

Tubular network formation 633 protects mitochondria from autophagosomal degradation during nutrient starvation

A network of macrophages 637 supports mitochondrial homeostasis in the heart

Interleukin-1β 639 induces mtdna release to activate innate immune signaling via cgas-sting

Mitochondrial dna stress signalling protects the nuclear genome

Deoxyribonuclease 1 reduces pathogenic effects of cigarette smoke 646 exposure in the lung