Long Noncoding Transcriptome in Chronic Obstructive Pulmonary Disease

Abstract

Chronic airway inflammation from recurring exposures to noxious environmental stimuli results in a progressive and irreversible airflow limitation and the lung parenchymal damage that characterizes chronic obstructive pulmonary disease (COPD). The large variability observed in the onset and progression of COPD is primarily driven by complex gene–environment interactions. The transcriptomic and epigenetic memory potential of lung epithelial and innate immune cells drive responses, such as mucus hyperreactivity and airway remodeling, that are tightly regulated by various molecular mechanisms, for which several candidate susceptibility genes have been described. However, the recently described noncoding RNA species, in particular the long noncoding RNAs, may also have an important role in modulating pulmonary responses to chronic inhalation of toxic substances and the development of COPD. This review outlines the features of long noncoding RNAs that have been implicated in regulating the airway inflammatory responses to cigarette smoke exposure and their possible association with COPD pathogenesis. As COPD continues to debilitate the increasingly aging population and contribute to higher morbidity and mortality rates worldwide, the search for better biomarkers and alternative therapeutic options is pivotal.

The association between exposure to inhaled toxicants and the development of pulmonary structural abnormalities is well known. Several environment–gene interactions have been described and it is known that the effects of specific inhalant type, dose, and duration are influenced by the genetic variants in several airway inflammatory response pathways (1). Furthermore, recurring exposures to inhaled toxicants shape pulmonary immune responses by creating a memory-based acquired immunity that plays a pivotal role in determining the outcome of subsequent exposures. These repetitive exposures, or so-called “experiments of nature,” have contributed to the evolution of memory-dependent and host-beneficial trained responses that entail transcriptomic, epigenetic, and metabolomic remodeling (2, 3). Recurrent exposures to cigarette smoke (CS), the most common factor associated with the development of chronic obstructive pulmonary disease (COPD) in developed countries, leads to an accelerated-aging phenotype characterized by a progressive degeneration of lung structure and function and a reduced capacity to mount inflammatory responses to environmental stresses and injury (4, 5). COPD is therefore considered a condition of accelerated lung aging because it involves many of the changes that are observed in the lungs of elderly individuals, such as lung function decline, gas trapping, reduced lung elasticity, enlargement of the alveolar spaces, immunosenescence, and elevated basal oxidative stress and inflammation, together referred to as inflammaging (6, 7).

COPD is a protean/systemic disease, and several attempts have been made to classify distinct subgroups based on pathophysiologic derangements, response to treatment, and disease progression. Genome-wide association studies have described over 20 genetic loci that are convincingly associated with COPD affection status, and additional loci have demonstrated an association with several COPD-related phenotypes, such as chronic bronchitis, emphysema, and hypoxemia (8). However, a comprehensive understanding of all other epigenetic and genetic underpinnings that lead to specific disease phenotypes is still needed to enable the development of targeted pharmacological therapies for this condition (9).

Recent advances in next-generation sequencing have revealed that only a fraction of transcription processes across the human genome are associated with the protein-coding genes, with at least four times more noncoding than coding transcripts (10). These noncoding RNAs (ncRNAs), which include microRNAs (miRNAs) and long ncRNAs (lncRNAs) among others, have now been shown to play critical roles in various cellular functions, development, and diseases (11–14). In particular, the lncRNAs occur in large quantities, and their biological functions in maintaining cellular and tissue homeostasis and regulating inflammation are just beginning to be understood (14, 15). Here, we review the findings reported to date in these novel field of research, and outline the potential role of lncRNAs in the processes of aging, response to CS, and pathogenesis of COPD.

LncRNA Biology

With the advent of modern sequencing techniques, it became evident that less than 2% of the total human genome encodes for proteins and the rest is transcribed into ncRNAs. The specific roles of all individual ncRNAs are still unclear, but it is becoming increasingly evident that ncRNAs are key players in regulating protein expression at both the transcriptional and post-transcriptional levels. Consequently, aberrant ncRNA levels can lead to dysfunctional regulation and ultimately contribute to the pathogenesis of various diseases (16).

The ncRNA species have mostly been classified based on size into short ncRNAs (sncRNAs) and lncRNAs, with an additional class known as housekeeping RNAs. sncRNAs encompass molecules such as miRNAs, P-element Induced WImpy (testis in Drosophila) (PIWI)-interacting RNAs (piRNAs), endogenous siRNAs, and promotor-associated RNAs (pRNAs). Housekeeping RNAs encompass ribosomal RNAs (rRNAs), transfer RNAs (tRNAs), small nuclear RNAs (snRNAs), and small nucleolar RNAs (snoRNAs). The miRNAs have been the most extensively studied, and there is growing evidence that they may be at the center of aberrant immune-mediated inflammatory responses and oncogenesis. In particular, the overexpression or underexpression of miRNAs has been implicated in the development of lung cancer and various other respiratory diseases. For example, reduced expression of the let-7 family of miRNAs has been implicated in both lung cancer and COPD pathogenesis (17).

The lncRNAs are defined as a functionally diverse set of transcripts that are >200 nt in size and do not encode for functional proteins. These molecules include the long intergenic ncRNAs (lincRNAs), natural antisense transcripts (NATs), intronic lncRNA, pseudogenic transcripts, circular RNAs, long enhancer ncRNAs (eRNAs), and transcribed ultraconserved regions (T-UCRs), and other lncRNA species identified in last few years (18). Several biological functions have been attributed to the lncRNAs associated with CS exposure and COPD, as summarized in Table 1. For a more comprehensive list of lncRNAs, see Table E1 in the data supplement. These functions primarily include alteration of protein localization, mRNA decay, chromatin modification, alternative splicing, modulation of protein activity, and transcription initiation/repression. lncRNAs also act as either inducers and restrictors of inflammation, and participate in multiple levels of immune cell development and response to pathogens (18). As mentioned above, the knowledge gap regarding the role of these molecules is large, as only a small portion of lncRNAs have been characterized.

Table 1.

Select Long Noncoding RNAs Associated with Cigarette Smoke Exposure and Chronic Obstructive Pulmonary Disease

lncRNA/Gene Symbol	Target(s)	Biological Functions	Analysis	Study Model	Reference
UCSC_3382_2454	CYP1B1	Metabolic pathways	Microarray	Lung tissues from healthy individuals and smokers with/without COPD	19
RNA175876|ENST00000554946	NF-κB activating protein 1	Cell signaling
RNA43329|UCSC_1041_4013, RNA45078|UCSC_3181_2530, and RNA53055|H-InvDB_314_668	IG gene
RNA37093 |ENCODE-1963-473, RNA175499|ENST00000544591, RNA48255|UCSC-7110-1509, RNA174930|ENST00000508732, RNA47218|UCSC_5826_1803, RNA53748|H-InvDB_1025_432, RNA44480|UCSC_2447_2885, RNA39398|RefSeq_1374_2314, RNA44021|UCSC_1880_3276, and RNA39240|RefSeq_1208_2498	—	—
TUG1	α-SMA and fibronectins	Cell proliferation	Microarray	Lung tissues from patients with COPD	20
SAL-RNA1, SAL-RNA2, and SAL-RNA3	SIRT1/FoxO3a, p53, p21	Cellular senescence		Lung tissues from patients with COPD	21
LINC00599, LINC01362, LINC00824, LINC01624, RP11-563D10.1, RP11-98G13.1, and AC004791.2	—	—	RNA-seq	Whole-blood cells from former and current smokers with COPD	22
LINC00987, A2M-AS1, and LINC00612	A2M	Protease inhibitor	RNA-seq	Whole-blood cells from nonsmokers and smokers with COPD	23
HLA-DQB1-AS1	ADRB1	Adrenergic receptor involved in cardiac muscle contractility and bronchomotor tone
LOC101928100 and LINC00861	RORA	Nuclear receptor; involved in metabolism and inflammation
RORA-AS1	TGF-β3	Implicated in lung development, ion homeostasis, cell adhesion, and extracellular matrix production
ENST00000502883.1/RP11-499E18.1	CXCL16	Expressed by T cells, B cells, macrophages, and dendritic cells; chemoattractant for T-helper type 1 cells	Microarray	PBMCs in smokers and subjects with COPD	25
NR_026891.1/FLJ10038	CXCL16
HIT000648516	HMOX1	Induced by inflammation and oxidative stress; regulates cellular homeostasis and the Nrf2-Keap 1 pathway
XR_429541.1	SLA2	Inhibits T-cell activation; highly expressed in hematopoietic cells
ENST00000597550.1/CTD-2245F17.3	SIGLEC14	Expressed in innate immune cells; involved in cell adhesion
ANRIL	TNF-α, IL-1β, IL-17A, LTB-4	Inflammatory cytokine		Plasma from patients with acute exacerbations of COPD and stable patients with COPD	29
ENST00000447867 and NR-026690	RAPGEF3	Multifunctional protein; intracellular sensors of cAMP; nucleotide exchange factors	Microarray	CD4+T cells from patients with COPD	30
NR_102714	UCHL1	Deubiquitinating enzyme regulation	RNA-seq	CS-induced COPD mouse model and 16HBE cells	33
Fantom3_D330021G15	IL1RL1	Toll-like receptor; induced by proinflammatory cytokines
Fantom3_D830009E10	GGT5	Enzyme involved in biosynthesis
AK076311 and UC007COI.2	CCR10	Inflammation-associated biomarker	Microarray	Mouse model of CS exposure	36
ENSMUST00000181247	CD177

Definition of abbreviations: α-SMA = α-smooth muscle actin; COPD = chronic obstructive pulmonary disease; CS = cigarette smoke; lncRNA = long noncoding RNA; PBMC = peripheral blood mononuclear cell; RNA-seq = RNA sequencing.

Smoke Exposure Altered Lung Tissue lncRNAs

In a pioneering study analyzing the genome-wide expression of lncRNAs, RNA was extracted from lung tissue from nonsmokers without COPD, smokers without COPD, and smokers with COPD, and subjected to a comparative microarray analysis (19). The participants (all of whom had lung cancer) included three nonsmokers without COPD, five smokers without COPD, and five smokers with COPD. A microarray analysis of 39,253 distinct lncRNAs showed differential expression of a large number of lncRNAs in patients with COPD independently of smoking status. Compared with smokers without COPD, 120 lncRNAs were overexpressed and 43 underexpressed in patients with COPD, indicating that smoking alters the expression of lncRNAs, and also that these lncRNAs may be associated with changes in key pathogenic processes of COPD caused by tobacco smoke exposure. As such, several of these lncRNAs are likely involved in regulating the expression of cellular mediators and signaling pathways that are linked to the altered immune system in COPD. However, the existence of many of these lncRNAs has been inferred from very limited expressed sequence tags numbers, and their relevance should therefore be viewed cautiously pending experimental validation. In addition, the use of whole-lung homogenates makes it extremely difficult to confidently tease out local lncRNA expression levels from systemic ones. Most importantly, the patients with COPD underwent diverse treatment regimens (two patients reportedly used inhaled corticosteroids and β-adrenergic agonists, and three used theophylline). This complicates interpretation of the results, as these compounds could also modulate noncoding transcriptomes and other cellular responses.

In an attempt to identify novel biomarkers for the diagnosis and treatment of COPD, Tang and colleagues performed a microarray analysis of patients with COPD and identified several differentially expressed lncRNAs compared with healthy control subjects, with n = 3 in each group (20). A subsequent genomic locus correlation analysis between these lncRNAs and protein-coding genes suggested that most of the differentially expressed lncRNAs in COPD lung tissue were intergenic lncRNAs (e.g., TUG1 [taurine upregulated gene 1] lncRNA). Although 20 patients from each group were used for histopathological and other validation analyses, no information was provided about the medication history of the patients with COPD, which might very well alter lncRNA expression. Nonetheless, the results from in vitro silencing of lncRNA TUG1 strongly support the notion that this lncRNA plays a role in COPD pathogenesis. In a related study, lung parenchymal tissues obtained from 22 patients without COPD and 12 with COPD were examined to elucidate the role of lncRNAs in lung tissue repair and maintenance of alveolar homeostasis (21). The expression of senescence-associated lncRNAs (SAL-RNAs), specifically SAL-RNA2 and SAL-RNA3, was significantly increased in COPD lung tissues, along with upregulation of p53 and p21 levels, compared with control tissues. However, both the SAL-RNA1 levels and the SIRT1/FoxO3a pathway were significantly downregulated. As such, these studies suggest that faulty regulation of lncRNAs in type II airway epithelial cell senescence may well contribute to the pathogenesis of COPD (21).

Smoke Exposure Altered lncRNAs in Hematopoietic Tissues

A whole-blood RNA sequencing (RNA-seq) analysis of 229 current and 286 former smokers revealed that 171 genes, including seven lncRNAs (LINC00599, LINC01362, LINC00824, LINC01624, RP11-563D10.1, RP11-98G13.1, and AC004791.2), were differentially expressed between current and former smokers (22). This was further supported by RNA-seq of five smokers with COPD, seven nonsmokers with COPD, and five healthy nonsmokers, and by quantitative PCR–based validation in five additional subjects (two smokers with COPD, two nonsmokers with COPD, and one healthy nonsmoker). However, the sex disparity among the analyzed groups may represent a major confounding factor in the reported observations, as all of the smokers with COPD were male, whereas four of the healthy nonsmokers and five of the nonsmokers with COPD were female.

Similarly, in another high-throughput sequencing analysis, Qian and colleagues examined lncRNAs, miRNAs, and signaling pathways in smoking and nonsmoking patients with COPD (23). Blood samples were collected from five healthy subjects, five nonsmoking patients with COPD, and five smoking patients with COPD. In all, the authors identified 96 differentially expressed lncRNAs in nonsmoking patients, 44 lncRNAs in smoking patients with COPD, and 15 lncRNAs common to nonsmoking and smoking patients with COPD. Based on an interaction network analysis, they found that Let-7-ADRB1-HLA-DQB1-AS1 may play a key role in the pathogenesis of smoking-associated COPD, and that miR-218–5p/miR15a-RORA-LOC101928100/LINC00861 and miR-218–5p/miR15a-TGF-β3-RORA-AS1 interactions may be involved in the pathogenesis of nonsmoking COPD. Interestingly, their data suggest that miR-122–5p-A2M-LINC00987/A2M-AS1 (Alpha-2-macroglobulin protein antisense RNA 1)/linc0061 interactions may play key roles in COPD irrespective of smoking status. In contrast to the aforementioned study (22), this analysis was well controlled for sex, race, and lifetime smoke exposure, and other confounders as described in the COPDGene study (24). However, one minor limitation of this work, as acknowledged by the authors, is the fact that the methods used did not control for confounding results due to different cell subpopulation distributions among the blood samples.

In another study, Qu and colleagues determined the expression of lncRNA and mRNA by analyzing a microarray of peripheral blood mononuclear cell (PBMC) RNAs from 20 healthy nonsmokers, 17 smokers without airflow limitation, and 14 patients with COPD (25). They identified 158 differentially expressed lncRNAs and selected five lncRNAs (NR_026891.1, ENST00000502883.1, HIT000648516, XR_429541.1, and ENST00000597550.1) for further validation based on their dysregulation in COPD. The regulatory role of ENST00000502883.1 in CXCL16 expression, and consequently an indirect role in PBMC recruitment, was confirmed. In addition, a gene ontology (GO) enrichment analysis showed that leukocyte migration, immune response, and apoptosis (cellular responses previously reported to be involved in the pathogenesis of COPD) were the main processes associated with the predicted targets of these lncRNAs. However, the authors poorly addressed the sex distribution in their cohort (a recurrent confounding variable in studies of this nature), which consisted of mostly male subjects (specifically, 42 males and nine females [eight healthy nonsmokers and one smoker with COPD]). It is well established that there is a differential susceptibility to tobacco exposure, with female smokers being generally more susceptible to developing COPD (26).

The lncRNA ANRIL (antisense ncRNA in the INK4 locus) is a well-characterized functional lncRNA that has been associated with the development and prognosis of chronic inflammatory diseases. It has been shown to repress the NF-κB pathway by inducing CARD-8 (caspase recruitment domain-containing protein 8) and downregulating TGF-β1 expression (27, 28). Recently, Ge and colleagues examined lncRNA ANRIL expression in peripheral blood cells and associated inflammatory cytokines in plasma from 136 patients with acute exacerbations of COPD (AECOPD), 138 stable patients with COPD, and 140 healthy control subjects (29). They observed lower ANRIL expression in patients with AECOPD as compared with both stable patients with COPD and healthy individuals. Furthermore, ANRIL expression was negatively correlated with proinflammatory cytokines in patients with AECOPD and stable patients with COPD, and with Global Initiative for Chronic Obstructive Lung Disease stage in patients with AECOPD, but not in stable patients with COPD.

More recently, Qi and colleagues conducted a study to elucidate the lncRNA profiles in CD4⁺ T cells from patients with COPD (30). CD4⁺ T cells isolated from 56 patients with AECOPD, 56 patients with stable COPD, and 35 healthy control subjects were used in this study. A microarray analysis using five age-matched samples from each group revealed that four lncRNAs were significantly upregulated in the AECOPD group compared with the stable COPD and control groups. A qRT-PCR analysis validated these four lncRNAs in 50 AECOPD, 50 stable COPD, and 30 control samples, and two lncRNAs (ENST00000447867 and NR-026690) were considered for further analysis. Based on the lncRNA-mRNA coexpression network and competing endogenous RNA (ceRNA) network analysis, they predicted three possible target genes (DCX [doublecortin], RAPGEF3 [RAS-associated protein guanine nucleotide exchange factor 3], and UBOX5 [U-box domain containing 5]) that were coexpressed with both the lncRNAs, and the expression of RAPGEF3 was concordant with the GO and Kyoto Encyclopedia of Genes and Genomes pathway analysis. RAPGEF3, also known as EPAC1, is a Rap1 (Ras-related protein 1) guanine nucleotide-exchange factor that is activated by cAMP, which is a prototypic second-messenger–mediating signaling pathway that is related to many diseases, including immunological and cardiac disorders and cancer. RAPGEF3-Rac1 signaling has previously been implicated in respiratory distress syndrome, and direct activation of EPAC attenuated the CS-induced IL-8 release from human airway smooth muscle cells conferring to the role of RAPGEF3 in COPD pathogenesis (31, 32).

lncRNAs in Animal Models of COPD

Investigators have also studied the roles of lncRNAs using animal models of COPD. A lung tissue RNA-seq analysis was recently performed to define the transcriptome profiles of a chronic CS-induced COPD mouse model (33). A total of 37,072 lncRNAs were detected, and 109 lncRNAs and 260 mRNAs were identified as being significantly differentially expressed in CS-exposed mice compared with control animals. Three of these significantly altered lncRNAs (NR_102714, fantom3_D330021G15, and fantom3_D830009E10) were found to be logically associated with likely COPD-relevant protein-coding gene targets (UCHL1 [ubiquitin C-terminal hydrolase L1], IL1RL1 [interleukin 1 receptor-like 1], and GGT5 [gamma-glutamyltransferase 5], respectively). Further qRT-PCR analyses of 16HBE cells treated with or without CSE (CS-extract), and PBMCs from six healthy individuals and seven patients with COPD confirmed an association between lncRNA NR_102714 and its putative target, UCHL1. Of note, UCHL1 is an established aging biomarker that is characteristically associated with COPD (34).

In a related mouse model study, Wang and colleagues conducted a microarray analysis to identify CS-induced lncRNAs and mRNAs, and identified 108 lncRNAs and 119 mRNAs that were differentially expressed in CS-exposed animals as compared with controls (35). By performing a coding-noncoding gene coexpression network analysis, this group identified lncRNAs AK076311 and uc007coi.2 as being negatively associated with CCR10 (chemokine receptor 10), and the lncRNA ENSMUST00000181247 as being positively coexpressed with CD177. Of note, CCR10 and CD177 are both inflammation-associated biomarkers that are commonly evaluated in lung disease (36, 37).

lncRNAs Modulated by Other Airborne/Environmental Toxicants

Li and colleagues performed a microarray analysis to identify abnormally expressed lncRNAs in human lung epithelial cells induced by different concentrations of wood smoke particulate matter less than or equal to 2.5 μm in aerodynamic diameter (WSPM_2.5) and arterial traffic ambient PM_2.5 (TAPM_2.5) (38). They found that compared with control cells, 94 lncRNAs were upregulated and 203 were downregulated in WSPM_2.5-treated cells, whereas 1,292 lncRNAs were upregulated and 1,362 were downregulated in TAPM_2.5-treated cells. They selected lncRNA MEG3 (maternally expressed gene 3) for further validation based on its differential expression and GO/Kyoto Encyclopedia of Genes and Genomes analysis, and demonstrated that lncRNA MEG3 mediates PM_2.5-induced cell apoptosis and autophagy by regulating the expression of p53.

In the respiratory tract, MUC5AC (Mucin-5 subtype AC) and MUC5B (Mucin-5 subtype B) act as the dominant gel-forming mucins, and in a pathological state are responsible for airway mucus obstruction resulting from mucin gene expression, mucin hypersecretion, and/or goblet cell hyperplasia (39–41). Dysregulation of any of these processes can contribute to a pathological decline in lung function and contribute to CS- and COPD-associated morbidity and mortality (39, 40, 42). Of note, high expression of MUC5B is observed in both COPD and lung adenocarcinomas. Interestingly, Yuan and colleagues found that lncRNA MUC5B-AS1 (MUC5B Antisense RNA 1) (ENST00000532061.2) is significantly upregulated in lung adenocarcinoma tissues (43). They also showed that lncRNA MUC5B-AS1 promotes cell migration and invasion by forming a protective RNA-RNA duplex with MUC5B, resulting in increased expression of MUC5B in these cancer cells. It has been reported that lncRNA HOTAIR (HOX [Homeobox gene] Transcript Antisense RNA) and SNHG16 (Small Nucleolar RNA Host Gene 16) have great oncogenic potential and modulate MUC5AC gene expression (44, 45). Both the chronic bronchitic and emphysematous COPD phenotypes exhibit pulmonary damage resulting from exaggerated and uncontrolled inflammation, which may be regulated by lncRNAs (46).

Chronic Inflammation, Aging, and COPD: Role of lncRNAs

Recent advances in sequencing have led to the discovery of novel lncRNAs associated with aging and inflammaging in various disease models as well as in patients with aging-associated diseases (recently reviewed in References 47 and 48). The aging-associated diseases are broadly immunologic, metabolic, and neurologic in origin and include cardiovascular diseases, diabetes, arthritis, and Alzheimer’s disease, whose incidence increases exponentially with age (49). Aging-associated lncRNAs include TERRA (telomeric repeat-containing RNA) and TERC (telomerase RNA component), which control telomere function; Xist (X-inactive specific transcript), H19, MANTIS (a lncRNA n342419), ANRASSF1, and PINT (P53 induced Transcript lincRNA), which are involved in epigenetic changes; MEG3, 7SL, GAS5 (growth arrest specific 5 lncRNA), LincRNA-p21, and HOTAIR, which are associated with loss of proteostasis; ANRIL, MALAT1 (metastasis-associated lung adenocarcinoma transcript 1), SRA (steroid receptor activator), HEIH (highly expressed in hepatocellular carcinoma lncRNA), HULC (hepatocellular carcinoma up-regulated lncRNA), UCA1 (urothelial cancer associated 1 lncRNA), and Gadd7 (growth-arrested DNA damage-inducible gene 7), which are involved in the regulation of immunosenescence; 17A, Lnc-IL7R, THRIL (TNF and HnRNPL [heterogenous nuclear ribonucleoprotein L] - Related Immunoregulatory lncRNA), and Lethe, which are associated with inflammaging; AK028326, AK141205, ES-1,-2,-3, and lnc-ROR, which are involved in the regulation of stem cell function; and TUC339 (the ultra-conserved lncRNA) and Tie-1AS (Tyrosine kinase with Ig- and EGF-like domains 1 Antisense RNA), which are implicated in intercellular communication.

Aging lungs undergo structural changes, such as an increase in the size of the alveolar space without any inflammation or alveolar wall destruction (senile emphysema), a reduction in elastic recoil, diminished respiratory muscle strength, and a less effective mucociliary clearance apparatus (50). The hallmarks of aging can be broadly divided into processes that affect transcription (genomic instability, telomere attrition, and epigenetic alterations), metabolism (loss of proteostasis, deregulated nutrient sensing, and mitochondrial dysfunction), and cellular processes (cellular senescence, stem cell exhaustion, and altered intracellular communication). The key molecules/pathways associated with aging and COPD are γ-H2A.X; Ku70 and Ku80, which are involved in genomic instability; TPP1 (tripeptidyl peptidase 1), which regulates telomere shortening; HDAC, SIRT1, and SIRT6, which governs epigenetic signatures; UCHL1, p62, and autophagy regulators that balance proteostasis; mTOR and IGF1 for deregulated nutrient sensing; reactive oxygen species (ROS), oxidized DNA, Nrf2, and mitophagy (mitochondrial dysfunction); SA-β-gal, p16, and p21 (senescence); MMP/TIMP dysregulation, ELN, FBLN5, and LTBP2 (extracellular matrix dysregulation); and Notch pathway and WNT pathways (stem cell exhaustion). Studies on the aging pathophysiologies associated with COPD pathology have been aptly summarized by Brandsma and colleagues (51). Inflammaging is one of the hallmarks defined as chronic low-grade inflammation mostly associated with immunosenescence. John-Schuster and colleagues showed that aged mice exposed to CS were more susceptible to emphysema than younger mice (52). In this model, they observed a significantly increased iBALT volume/basal membrane and expression of CXCL13, MCP1, MIP1α, IL-6, TNF-α, and p16, and decreased expression of SIRT-1 in CS-exposed old mice compared with young mice.

Aging-related genomic instability due to the accumulation of damaged DNA along with dysregulated repair mechanisms also contributes to cellular and tissue senescence and COPD pathogenesis (53). A recent lung tissue transcriptomic analysis of patients with severe COPD found that alterations in DNA repair responses are associated with COPD disease severity (54), and another study suggested that some CS-exposure–altered lncRNAs could also play an important role in DNA repair mechanisms (55). However, the association of specific lncRNAs and DNA repair mechanisms in the context of COPD pathogenesis has yet to be established.

Cellular Senescence and Mitochondrial Dysfunction in COPD: Role of lncRNAs

Cellular senescence is a dynamic cellular process that results in cell-cycle arrest that is instigated by a stress response to either exogenous stimuli (e.g., oxidative stress due to smoke exposure) or endogenous stimuli via replication stress (e.g., telomere erosion). Airway epithelial cell senescence contributes strongly to COPD pathogenesis, as discussed comprehensively in a review by Nyunoya and colleagues (42). Mitochondrial dysfunction and the unfolded protein response can also promote cellular senescence. p53/p21 and p16/RB are two of the major pathways responsible for driving cellular senescence and the senescence-associated secretory phenotype, which is characterized by the secretion of proteases, inflammatory factors (e.g., IL-6, IL-8, and IL-1α), and growth factors such as vascular endothelial growth factor and granulocyte-macrophage colony-stimulating factor (56). In addition, cells undergoing senescence show increased metabolic and chromatin remodeling that results in the alteration of cell-death and survival pathways, including autophagy and mitophagy. Mitochondria modulate cellular oxidative homeostasis and undergo dynamic remodeling whereby they continually fuse and divide to control their quality, distribution, size, and motility. Several proteins tether to mitochondria via an interaction with mitochondria-associated proteins, and control organelle morphology. In addition to morphological changes, mitochondrial homeostasis is maintained by biogenesis and autophagic degradation or mitophagy (57–60). It has been shown that protein-associated Damage-associated molecular patterns (DAMPs) are increased in patients with COPD, as well as during exacerbations (61, 62). However, it remains unknown whether lung epithelial cells harbor increased mitochondrial DAMPs (e.g., mitochondrial DNA) due to impaired mitophagy, which alters innate inflammatory and mucus responses to CS exposure during COPD exacerbations. In addition, several mitochondria-associated lncRNAs (e.g., LIPCAR [LIncRNA Predicting CArdiac Remodeling], Sox2ot [SRY (sex determining region Y)-box-2 Overlapping Transcript], and MALAT-1) have been associated with mitochondrial biosynthesis, bioenergetics, and cell-death pathways in cancer cells (63–66), although their role in normal cell physiology and COPD pathophysiology remains to be elucidated.

The mechanisms by which lncRNAs could influence aging and senescence-associated pathologies are starting to be unraveled (67, 68), but further research is needed to gain a deeper understanding of lncRNA functions and exploit that knowledge to develop novel and effective therapeutic strategies. Such analyses may be hampered by the lack of sequence conservation and multiple transcript variants that may have different functions based on their subcellular localization. More importantly, some of the currently annotated lncRNAs could also encode for yet-to-be-found short peptides that may have been overlooked.

Recently, Liu and colleagues used a broad and systematic approach to develop a genome-wide screening platform that identifies functional lncRNAs (69). Using a CRISPRi (clustered regularly interspaced short palindromic repeats-based interference) gene-editing technology, the authors identified lncRNAs that modified robust cell growth in seven different human cell lines. The results were significant because they substantially increased the number of known functional lncRNA loci. They further demonstrated that lncRNAs function in a highly cell-type–specific manner and directly regulate important cellular processes.

Limited sample sizes and a lack of information regarding confounding factors such as sex, age, smoking, and medication history represent recurring major limitations in the majority of studies discussed here. In addition, investigators have analyzed an array of tissues (e.g., lung tissues, immune cells, isolated PBMCs, and whole-blood cells) to examine the role of lncRNAs in COPD pathogenesis. Of particular concern is the fact that no studies attempting to characterize lncRNAs that may be relevant to COPD (regardless of the tissue analyzed) have been able to independently generate data that suggest a likely role for any specific lncRNA in COPD. However, only a handful of such studies have been reported to date, and lncRNA profiles can differ dramatically among various pathophysiologies, tissues, and cell types. In addition, the limited number of patients with COPD examined to date have undoubtedly undergone different pharmacotherapy regimens, which may also contribute to lncRNA variability. Nevertheless, lncRNAs that are differentially expressed in COPD are readily identifiable via RNA-seq analyses, and there exists a considerable body of available RNA-seq data relevant to COPD that can likely be used to identify single lncRNAs and/or lncRNAs that act cooperatively to regulate specific pathways driving COPD. There are very limited data on the direct impact of CS exposure on the long noncoding transcriptome of lungs and other tissues, and more studies are needed to address this issue. In any event, rigorous experiments, including (but not limited to) targeted gain- or loss-of-function characterizations in appropriate COPD models (e.g., lung tissue organoids and lung-on-a-chip models), will ultimately be required to establish functional roles for individual lncRNAs in COPD, and to differentiate between lncRNAs that drive COPD pathology and those that are misexpressed in response to disease.

Conclusions

The literature reviewed here provides evidence that lncRNAs may be important contributors to cellular responses to CS exposure and COPD pathogenesis. The evidence gathered so far positively supports the role of specific lncRNAs in cellular and molecular pathways in both basal and diseased conditions. These findings will have to be placed in context with other pathogenic contributions from other regulatory mechanisms, such as DNA methylation, miRNAs, and specific genetic polymorphisms that are not discussed here. Specifically, the interplay between the effects of CS exposure on the long noncoding transcriptome and associated functions could lead to COPD pathobiology, as summarized in Figure 1.

Figure 1.

Tobacco smoke exposure alters the expression and functions of long noncoding RNAs, which could lead to augmented chronic obstructive pulmonary disease pathologies, including those associated with mitochondrial dysfunction, chronic inflammation, mucus dysregulation, epigenetic alterations, and cellular senescence. ANRIL = antisense noncoding RNA in the INK 4 locus; COPD = chronic obstructive pulmonary disease; HOTAIR = HOX (Homeobox gene) transcript antisense RNA; LIPCAR = long intergenic noncoding RNA predicting cardiac remodeling; lncRNAs = long noncoding RNAs; M-AS1 = Alpha-2-macroglobulin protein antisense RNA 1 (A2M-AS1); MALAT1 = metastasis-associated lung adenocarcinoma transcript 1; MANTIS = a nuclear lncRNA n342419 that facilitates endothelial angiogenesis; MEG3 = maternally expressed gene 3; MUC5B-AS1 = mucin-5 subtype B antisense RNA 1; PINT = P53 induced Transcript lincRNA; SAL-RNA = senescence-associated long noncoding RNA; SNHG-16 = a small nucleolar RNA host gene 16; Sox2ot = SRY (sex determining region Y)-box-2 overlapping transcript; Xist = X-inactive specific transcript.

It is evident that there is still a large gap with regard to the identification and functional characterization of these molecules, and we have very limited knowledge about the specific lncRNAs that regulate the pathogenic mechanisms of COPD. Nonetheless, the establishment of novel ncRNA libraries and new biological markers is helping investigators find novel lncRNAs and assess their respective functions. The lncRNA-associated databases summarized in Table 2 provide important resources for such studies. Substantial research efforts are still needed to identify novel lncRNAs and assess their specific roles in both physiological and diseased settings. Such studies will help us understand other genetic or transcriptomic factors that contribute to the development of debilitating chronic airway diseases and associated consequences (e.g., steroid resistance, susceptibility to infections, and comorbid conditions) that are notably responsible for the extensive healthcare burden and worsening health in the ever-increasing aging population.

Table 2.

List of Current Databases That Are Useful for Long Noncoding RNA Research

Database	Description	Reference
NRED	Contains expression information for thousands of human and mouse lncRNAs	70
lncRNAdb	Reference database that provides comprehensive annotations of eukaryotic functional lncRNAs	71, 72
NONCODEv5.0	An integrated database dedicated to noncoding RNA from 17 species, including human, mouse, and rat, providing the location, strand, exon number, length, sequence, expression profile, exosome expression profile, predicted function, and disease relation	73, 74
lncRNome	A comprehensive searchable database for human lncRNAs	75
MONOCLdb	The MOuse NOnCode Lung database provides the annotations and expression profiles of mouse lncRNAs involved in influenza and SARS-CoV infections	76
LncRNAWiki	A comprehensive wiki-based, editable and open-content platform that harnesses collective intelligence to collect, edit, and annotate information about human lncRNAs and their users and publishers	77
MiTranscriptome	Catalog of RNA-Seq data for computational analysis of lncRNAs from over 6,500 samples spanning diverse cancer and tissue types	78
C-It-Loci	A tool for exploring and comparing the expression profiles of conserved loci for in silico screening of tissue-enriched lncRNAs	79
slncky Evolution Browser	An lncRNA discovery tool with high-quality RNA-sequencing data for analyzing evolutionary constraints	80
deepBase v2.0	An updated platform to decode the evolution, expression patterns, and functions of diverse ncRNAs across 19 species; provides lncSeeker to identify 176,680 high-confidence lncRNAs from 14 species	81
LncBook	A curated knowledge base of human lncRNAs that features a comprehensive collection of human lncRNAs and systematic curation of lncRNAs by multi-omics data integration, functional annotation, and disease association	82
LNCipedia5	A comprehensive compendium of human lncRNAs with updated literature curation of 2,482 lncRNA articles and official gene symbols	83

Definition of abbreviations: ncRNA = noncoding RNA; SARS-CoV = severe acute respiratory syndrome–associated coronavirus.