Radiation-induced pulmonary fibrosis: roles of therapy-induced senescence and microRNAs

Abstract

Purpose

Progressive, irreversible radiation-induced pulmonary fibrosis (RIPF) is a clinically significant intermediate- to a late-occurring side effect of radiotherapy. Known mechanisms of RIPF include oxidative stress-induced activation of TGF-β with activation of SMAD signaling, TNF-α elaboration, and activation of the Angiotensin Converting Enzyme (ACE) mediated production of angiotensin II with resulting activation of profibrotic cytokine signaling and vasoconstriction. The pioneering work of John Moulder, to whom this paper is dedicated, and several of his colleagues demonstrated that inhibiting the conversion of ACE with drugs such as Captopril, Enalapril, and Losartan can ameliorate radiation fibrosis in various tissues. While this work led several groups to probe mechanism-based pharmacological mitigation of RIPF, in this article, we explore and discuss the roles of microRNAs (miRNA) and therapy-induced senescence (TIS) in the pathogenesis of and potential biomarkers for RIPF.

Conclusion

Our analysis of the published literature in the last decade on RIPF, miRNA, and TIS identifies TIS as a mechanism in the onset and progression of RIPF, which is regulated through several miRNAs. This work may lead to the discovery and development of the next generation of miRNA therapeutics and/or the repurposing of approved pharmaceutical agents and the development of early biomarker panels to predict RIPF.

Keywords:

• Radiation
• radiotherapy
• fibrosis
• senescence
• microRNA

Introduction

Progressive, irreversible radiation-induced pulmonary fibrosis (RIPF) is a clinically significant intermediate- to late-occurring side effect of the radiotherapy (Stone et al. 2003; Prasanna et al. 2012). RIPF may initially manifest as pneumonitis and then can gradually develop into chronic debilitating fibrosis, characterized by loss of alveolar structure, disorganized thickening of septa, the collapse of alveolar space, and replacement of normal parenchyma with fibrotic tissue. Radiation injury from oxidative stress can lead to the deposition of extracellular matrix (ECM), apoptosis, senescence of pneumocytes, and loss of barrier function can lead to dysregulated inflammatory response, leukocyte infiltration, cytokine, chemokine, and growth factor production and secretion, resulting in chronic hypoxia (Wang et al. 2015). Acute inflammatory response, although essential for wound repair and tissue remodeling, can also lead to chronic inflammation, resulting in additional injury from oxidative stress.

Chronic oxidative stress activates transforming growth factor (TGF-β), a ubiquitously expressed immunomodulatory cytokine that acts as a biological lynchpin, triggering Suppressor of Mothers Against Decapentaplegic (SMAD) signaling pathways and stimulating fibroblast proliferation and ECM deposition (Barcellos-Hoff 2022). Together with TGF-β, TNF-α, an inflammatory cytokine produced by macrophages/monocytes during acute inflammation, is also a signaling axis in modulating radiation fibrosis (Dancea et al. 2009).

The Renin Angiotensin System (RAS) is a well-documented pathogenic factor implicated in fibrosis of various tissues, including lungs (Ghosh, Zhang et al. 2009), kidneys (Cohen et al. 2002; Moulder et al. 2011) and liver (Uhal et al. 2012). RAS plays an essential role in regulating blood pressure via Angiotensin Converting Enzyme (ACE). Angiotensin II (Ang II), a vasoconstricting hormone, is a central effector of the RAS through binding to its receptors AT1 and AT2 (Inagami 1994; Ghosh, Zhang et al. 2009). In addition, Ang II stimulates the over-expression of TGF-β and a connective tissue growth factor (CTGF) to drive lung fibroblast/myofibroblast proliferation and ECM deposition (Finckenberg et al. 2003). Increased ACE concentrations have been reported in the broncho-alveolar fluid in fibrotic lungs (Specks et al. 1990), and the gene angiotensin is overexpressed in pulmonary fibrosis patients (Selman et al. 2006), supporting the role of this pathway in fibrotic progression.

RAS and ACE are implicated in fibrosis, mediated via TGF-β overexpression by ANGII (Konigshoff et al. 2007) and ECM deposition (Rajasekaran et al. 2015). TGF-β and ECM deposition are both regulated by the microRNAs (Liu et al. 2010; Kang 2017). John Moulder and several of his colleagues demonstrated that inhibiting ACE or conversion of Ang I into Ang II with drugs such as Captopril (Cohen et al. 2008), Enalapril (Gao, Fish et al. 2013; Gao, Narayanan et al. 2013; Cohen et al. 2016) and Losartan ameliorates radiation fibrosis in various tissues (Ghosh, Wu et al. 2009; Kma et al. 2012). His work led to further probing into the mechanisms of radiation-induced fibrosis in various tissues, including RIPF, by several groups with the goal of finding a pharmacological approach to mitigate fibrosis from radiotherapy and accidental radiation exposures.

MicroRNAs (miRNAs) are small non-coding RNAs that regulate the post-transcriptional gene expression (Yates et al. 2013). Primarily they act through competitive binding to the target mRNA, leading to mRNA degradation, and inhibition of the translation (Bushati and Cohen 2007). While new miRNAs are continued to be discovered, their tissue-specific heterogeneity in regulatory functions depends on their subcellular localization, relative abundance, and their target (Bushati and Cohen 2007). The role of several miRNAs (let-7 family, miR-29, miR-30, miR-155 and miR-21) in Idiopathic Pulmonary Fibrosis (IPF), a chronic, progressive and lethal fibrotic lung disease, has been previously described (Pandit et al. 2011). A recurrent global theme that miRNAs in IPF are both regulated by TGF-β1 and regulate TGF–β1 signaling pathway by their target genes (Pandit et al. 2011).

Exposure of non-cancerous cells to radiation and chemotherapy can result in therapy-induced senescence (TIS) due to DNA damage and oxidative stress (Ewald et al. 2010; Prasanna et al. 2021), which is different from replicative senescence resulting from repeated cell division (Hayflick and Moorhead 1961). TIS is an unfortunate consequence of cancer therapy that may lead to late toxicities and cancer recurrence. Activation of stress signals, including DNA damage response (DDR), growth arrest by CDK inhibitors, and heterochromatin changes without apparent loss of telomere function characterize TIS, (Kuilman et al. 2010). Two pathways, p53/p21 and p16/RB, appear to be predominant in senescence stress activation, including chronic oxidative stress (Loaiza and Demaria 2016).

Here in this article, we explore and discuss the roles of miRNAs in TIS and the pathogenesis of RIPF. For this minireview, for the query of the published literature between 2012–2021, we used the U.S. Library of National Medicine’s PubMed database using iCite (https://icite.od.nih.gov/), a web application to access data on published bibliometric information for publications, developed by the National Institute of Health’s (NIH) Office of Portfolio Analysis. We used the MeSH terms ‘Radiation induced pulmonary fibrosis,’ ‘Therapy induced senescence,’ ‘Therapy induced senescence and microRNA,’ and ‘Radiation induced pulmonary fibrosis and miRNA’ as search terms. Figure 1(A) shows the relative number of publications in the above fields, and Figure 1(B) shows the yearly breakdown of the number of articles published in ‘RIPF,’ ‘TIS,’ ‘RIPF and miRNA,’ and ‘TIS and miRNA,’ respectively. This analysis shows the roles of senescence in the pathogenesis of pulmonary fibrosis and miRNAs in TIS. However, miRNAs in RIPF are less described. We focus our discussions on miRNAs’ roles in TIS and RIPF. Understanding the TIS as a mechanism of RIPF regulated through miRNA may help develop and translate next-generation miRNA therapeutics and repurpose approved pharmaceutical agents.

Figure 1. Analysis of published literature on roles of therapy-induced senescence and microRNAs in radiation-induced pulmonary fibrosis. We used the U.S. Library of National Medicine’s PubMed database using iCite (https://icite.od.nih.gov/), a web application to access data on published bibliometric information for publications developed by the National Institute of Health’s (NIH) Office of Portfolio Analysis. The database was queried for the published literature for ten years between 2012 and 2021, using the MeSH terms ‘Radiation induced pulmonary fibrosis,’ ‘Therapy induced senescence,’ ‘Therapy induced senescence and microRNA,’ and ‘Radiation induced pulmonary fibrosis and miRNA’ as search terms. (A) shows the relative number of publications in the above fields, and (B) shows the yearly breakdown of the number of articles.

Roles of miRNAs in TIS

After radiation exposure, senescence of type II alveolar cells (AECII) was observed in a dose- (5-Gy vs. 17.5 Gy and 5 × 5 Gy or 5 × 6 Gy) and time-dependent (up to 30 wks) manner that correlated with the progression of the fibrosis in mice (Citrin et al. 2013). Because AECII can also serve as alveolar stem cells, replenishing depleted AECII and AECI cell types after injury, senescence of AECII depletes the stem cell reserve of the lung and can lead to parenchymal depletion and fibrotic progression (Citrin et al. 2013). Further, senescent cells (SnCs) secrete the ‘senescence-associated secretory phenotype (SASP),’ a proteomic consequence of senescence that comprises several proinflammatory cytokines implicated in RIPF. These include IL-6, IL-1β, EGF, VEGF, MMPs, TIMPs, ICAM-1, and EGFR (Coppe et al. 2010). A robust elevated expression of p16, SASP chemokines (CCL2, CXCL10, and CCL17), and SASP matrix metalloproteinases (MMP2, MMP9, and MMP12) was also observed in the macrophages of the irradiated lung, which is implicated in the development of fibrotic phenotype in mouse pulmonary fibroblasts (Su et al. 2021). Further, eliminating senescent AECII cells with compounds capable of preventing senescence or compounds capable of selectively killing senescent cells (senolytics) after irradiation mitigates RIPF in mouse models (Citrin et al. 2013; Chung et al. 2016; Citrin et al. 2017; He et al. 2019). It appears that senescent AECs controlled by the PTEN/NF‐κB pathway facilitate collagen accumulation in fibroblasts, resulting in lung fibrosis in a bleomycin-induced pulmonary fibrosis mouse model (Tian et al. 2019).

While the mechanisms of RIPF and the role of senescence in pulmonary fibrosis are well studied, studies on TIS are also emerging; but how miRNAs regulate TIS is far from understood (Figure 1). It appears that several miRNAs influence senescence by modulating the abundance of crucial senescence regulatory proteins either by lowering the stability or effecting the translation of mRNAs that encode such factors (Munk et al. 2017). The expression of p53, p21, MDM2, SIRT1, MYC, TERT, and MCD1 are involved in senescence, and several miRNAs regulate their expression (Munk et al. 2017). For example, miR-125b, miR-504, miR-25, and miR-30d regulate the expression of p53 (Kumar et al. 2011), miR-181a, miR-181b, miR138 regulate SIRT1 (Zhou et al. 2017) in TIS pathways. (Figure 2). Similarly, the levels of miR-27a-3p were up-regulated in TGF-β1-treated human lung fibroblasts in a Smad2/3-dependent manner and fibroblasts isolated from lungs of mice with experimental pulmonary fibrosis. Its overexpression inhibited, whereas knockdown enhanced, the differentiation of lung fibroblasts into myofibroblasts (Cui et al. 2016). Thus, miR-27a-3p functions via a negative-feedback mechanism in inhibiting lung fibrosis, and targeting miR-27a-3p may be a novel therapeutic approach to treat lung fibrosis (Cui et al. 2016).

Figure 2. Illustration of miRNAs that regulate important molecular pathways of Radiation-Induced Pulmonary Fibrosis. For comprehension, this figure was created from the references that appear in this manuscript (given below) for each molecular pathway. The references were extracted from the manuscript to the Microsoft® Excel (v16.66.1) to create the ‘sunburst chart. While the ‘branches’ illustrate the molecular pathways involved in RIPF, the ‘leaves’ illustrate the miRNAs involved in these pathways. In general, the miRNAs identified above act through competitive binding to the target mRNA in specifically identified pathways, leading to mRNA degradation, and inhibition of the translation. References: AKT3, miR207 (Tan et al. 2014), ATM, miR-214-3p (Lei et al. 2021), ATR, miR-185 (Wang et al. 2013), ECM, miR-29 (Cushing et al. 2011) miR-377, (Wang et al. 2008), miR-449 (Wang et al. 2008), EMT, miR-18a-5p (Zhang et al. 2017), miR-155-5p (Wang et al. 2021), miR-483-5p (Li et al. 2022), miR-486-3p (Yan et al. 2022), MYC, miR-34a family (Rogers et al. 2020), NF-kβ, miR-155-p (Wang et al. 2021, Zoico et al. 2021), p16, miR-9 family (O'Loghlen et al. 2015), p21, miR-29c (Shang et al. 2016), miR-155a (Zoico et al. 2021), p53, miR-25 (Kumar et al. 2011), miR-29-c-3p (Shang et al. 2016), miR-30d (Kumar et al. 2011), miR-125 (Kumar et al. 2011), miR-486-3p (Yan et al. 2022), SASP, miR-146a/b (Bhaumik et al. 2009), SOD, miR-222 (Liu et al. 2019), TGF-β, miR-9p (Fierro-Fernandez et al. 2015), miR-21 (Liu et al. 2010), miR-27a-3p (Cui et al. 2016), miR-29 (Cushing et al. 2011), miR-34b-3p family (Rogers et al. 2020), miR-424 (Xiao et al. 2015), miR-483-5p (Li et al. 2022), TNF-α, miR-21 (Fabbri et al. 2012, Munk et al. 2017), miR-146a/b (Bhaumik et al. 2009), SIRT1, miR-155-5p (Zoico et al. 2021).

Several miRNAs have also been implicated in the p16-mediated senescence (Munk et al. 2017). Senescence, along with carcinogenesis and cellular differentiation, is also epigenetically regulated by the polycomb repressive complexes, PRC1 and PRC2, due to their ability to repress INK4/ARF locus and CBX7 (an orthologue of Drosophila polycomb and part of PRC1). The miR-9 family of miRNAs downregulates the expression of CBX7. In turn, CBX7 represses miR-9-1 and miR-9-2 as a part of the regulatory negative feedback loop. Thus miR-9/CBX7 feedback loop is a regulatory module contributing to the induction of the cyclin-dependent kinase inhibitor (CDKI), p16(INK4a), during senescence in cells (O'Loghlen et al. 2015).

MiRNAs also appear to regulate the SASP, particularly interleukins, MMPs, and TNF-α (Munk et al. 2017). For example, senescence is associated with increased production and secretion of ILs, specifically IL6 and IL8, which are part of SASP. Screening of quiescent and senescent human fibroblasts for differentially expressed miRNAs showed that miR-146a/b was significantly elevated during senescence, suggesting delayed miR-146a/b induction might be a compensatory response to restrain inflammation (Bhaumik et al. 2009). Elevated expression of miR-146a/b in primary human fibroblasts suppressed IL-6 and IL-8 secretion. MiR-146a/b also downregulated IRAK1, a crucial component of the IL-1 receptor signal transduction pathway. This elevated expression in response to rising inflammatory cytokine levels is a part of the negative feedback loop that restrains excessive SASP activity (Bhaumik et al. 2009).

Similarly, numerous MMPs, which are responsible for the degradation of ECM (including collagens, elastins, proteoglycans, etc.), cleavage of cell surface receptors, and ligands, promote tissue damage through enhancement of chronic inflammation and ECM remodeling (Verma and Hansch 2007). MiRNAs regulate MMP levels. Quantitative RT-PCR analysis showed that MMP-2 but not MMP-9 was induced by TGF-β1 treatment, which was not affected by overexpression of miR-133a in mouse fibroblast cultures (Wei et al. 2019). Another component of the SASP, the cytokine TNF-α is also regulated by miRNAs. Circulating miR-21 can activate TLR-8 to stimulate the secretion of TNF-α and IL-6 secretion acting as a paracrine agonist of TLRs in mice (Fabbri et al. 2012).

Senescence is also promoted through the overexpression of p53/p21 and p16/RB pathways by miR-29c-3p in MSCs, which enhances the SASP production (Shang et al. 2016; Munk et al. 2017). A direct reduction in levels of TERT (which maintains telomere length to prevent replicative senescence) and SIRT1 with miR195 in old MSCs induces senescence, and its inhibition enhances the expression of TERT and SIRT1, reducing the p53 levels and delays the onset of senescence in a mouse model (Okada et al. 2016).

Studies also indicate that the EMT of AECs contributes to RIPF by down-regulating miR-155-5p using the GSK-3β/NF-kβ pathway in mice (Wang et al. 2021). In a mouse model, miRNA array analysis two weeks after a single dose of 20-Gy thoracic irradiation, the upregulated miRNAs included miR-21, whose expression increased in parallel with EMT progression in lung epithelial cells. Further, its downregulation by transfection of its inhibitor prevented radiation-induced EMT (Liu et al. 2019). EMT of AECs also seems to be mediated through a decrease in miR-486-3p level, activation of the transcription factor, BCL6, and expression of Snail, a family of transcription factors that promote the repression of adhesion molecule, E-cadherin to regulate EMT (Yan et al. 2022). Similarly, the role of miR-483-5p was also investigated in TGF-β1-induced EMT. It was found that miR-483-5p was upregulated both in fibrotic lung tissue and A549 cells treated with TGF-β1, and inhibition of this miRNA inhibited TGF-β1-medicated EMT (Li et al. 2022). Since exosomal miR-466f-3p derived from MSCs may possess antifibrotic properties, it may help prevent radiation-induced EMT through inhibition of AKT/GSK3β via c-MET; thus, providing a promising therapeutic approach for RIPF (Li et al. 2022).

Roles of microRNAs in RIPF

In general, tissue-specific roles of miRNAs in the onset and progression of fibrosis in various tissues are well-described (Jiang et al. 2010; Lorenzen et al. 2011). After radiation exposure, analysis of RNA expression has shown significant alterations in mRNAs and non-coding RNAs (ncRNA), including miRNA, circRNA, and lncRNA, implying their accentuating role in cellular stress responses (May et al. 2021). MiRNAs can act as post-transcriptional regulators of gene expression, including inflammation, regulated cell death such as apoptosis, cellular proliferation, differentiation, metastasis, and disease pathogenesis and progression, enabling the deposition of ECM and collagen in the pulmonary fibrosis (Vettori et al. 2012). In the context of pulmonary fibrosis, specific patterns of dysregulation appear to be involved in TGF-β (Rajasekaran et al. 2015), the platelet-derived growth factor (PDGF) (Svegliati et al. 2005), and the Wnt-β catenin (Bergmann et al. 2011) pathways.

In RIPF, miRNA regulation of TGF-β and CTGF signaling appears necessary. Upregulation of miR-21 was proportional to the severity of the lung fibrosis and the expression of TGF-β in human idiopathic fibrosis (IPF) and murine Bleomycin lung fibrosis model. Further, this relationship between miR-21 and TGF-β1 appears to be causal because the administration of miR-21 antisense oligonucleotides in vivo showed attenuation of fibrosis in mice (Liu et al. 2010).

Several miRNAs target the components of the TGF-signaling pathways and can play profibrotic or antifibrotic roles in TGF-β mediated pathways of the fibrosis (Kang 2017). MiR-21, which targets Smad7 (Liu et al. 2010), and miR-424, which targets Smurf2 (Xiao et al. 2015) are the two critical profibrotic miRNAs that regulate fibrosis by amplifying the TGF-β the pathway. The overexpression of miR-21 promotes the development of bleomycin-induced pulmonary fibrosis in mice, whereas its inhibition reduces the severity of fibrosis and myofibroblast differentiation. In addition, the level of collagen deposition is also under the control of miR-21 (Liu et al. 2010). Similarly, in a mouse model of stereotactic radiotherapy, miR-21 expression was found to increase at the radiation injury site, concurrent with collagen deposition, but the inhibition of miR-21 by its specific inhibitor anti-miR-21 only marginally affected EMT in lung endothelial cells; however, this inhibition significantly reduced collagen synthesis in lung fibroblasts (Kwon et al. 2016).

The profibrotic functions of TGF-β signaling are suppressed by several antifibrotic miRNAs. For example, while TGF-β1 stimulates the production of NADPH oxidase 4 (NOX4)-dependent ROS, promoting lung fibrosis, miR-9-5p seems to exert an opposite inhibitory effect on TGF-β receptor type II (TGF-βR2) and NOX4 expression. Increased expression of miR-9-5p abrogates TGF-β1-dependent myofibroblast phenotypic transformation in a bleomycin lung fibrosis mouse model (Fierro-Fernandez et al. 2015). Similarly, using cultured pleural mesothelial cells (PMC) and a pulmonary fibrosis animal model, down-regulation of miR-18a-5p in PMCs with bleomycin resulted in the EMT of PMCs. miR-18a-5p binds to the 3′ UTR region of TGF-βR2 allowing the suppression of the TGF-β Smad2/3 signaling (Zhang et al. 2017). However, the characterization of radiation-induced changes in the circulating miRNA profile within the first 30 days after thoracic irradiation of mice indicated a strain difference in their ability to predict survival outcomes. In C57Bl/6 mice, a set of miR-34b-3p, −96-5p, and −802-5p associated with TGF-β/SMAD signaling predicted survival outcomes. In C3H mice, a different set of miR-34a-5p, −100-5p, and −150-5p were associated with pro-inflammatory NF-κβ-mediated signaling pathways (Rogers et al. 2020).

ECM deposition is a critical process in the pathogenesis of RIPF. In a large-scale screening study of miRNAs in the bleomycin-induced lung fibrosis mouse model, expression of the miR-29 family was found to be significantly reduced in fibrotic lungs, and their levels inversely correlated with the expression levels of profibrotic target genes, and the severity of lung fibrosis (Cushing et al. 2011). Further, miR-29 was suppressed by TGF-β and many fibrosis-associated genes upregulated by TGF-β are derepressed by miR-29 knockdown and de-repression of reported miR-29 targets, including several collagens and upregulation of ECM-associated and remodeling genes (Cushing et al. 2011). Besides miR-29, miR-377 and miR-449 have also been known to regulate ECM synthesis, which appears to be mediated via the upregulation of fibronectin. Targeting inhibitors of fibronectin with p21-activated kinase and superoxide dismutase (SOD) seems to be a promising approach to mitigate the progression of the fibrosis (Wang et al. 2008).

While emerging profibrotic and antifibrotic roles are becoming apparent for several miRNAs, therapeutic intervention using anti-fibrotic miRNAs may be one promising approach for mitigating RIPF. However, identifying novel relevant miRNAs and understanding their regulatory roles in the pathogenesis of RIPF will be essential to develop successful therapeutic strategies. It was reported that extracellular vesicle (E) derived from MSCs could ameliorate RIPF in a mouse model by attenuating radiation-induced lung vascular damage, inflammation, and fibrosis via a downregulating ATM/P53/P21 signaling, which was under the control of miR-214-3p. In addition, the MSC-EVs also inhibited the development of SASP and attenuated the radiation-induced injury of endothelial cells (Lei et al. 2021).

Biomarkers of radiation exposure as predictors of lung injury

Studies on RNA biomarkers for cellular responses to radiation have become an increasingly important field of research for accidental and therapeutic radiation exposure. Several studies have explored the potential of miRNAs and long non-coding RNAs (lncRNA) in predicting normal tissue injury sustained during accidental and therapeutic radiation exposures using different model systems (Aryankalayil, Chopra, Makinde et al. 2018; May et al. 2021; May et al. 2022). The utility of miRNA biomarkers in predicting lung injury in non-human primate (NHP) serum after whole-body irradiation has previously been reported (May et al. 2022). MiRNA biomarkers could predict NHP survival at up to 71% accuracy by nine days post-irradiation and predict pleural effusion with 100% accuracy by 21 days. Pleural effusion could lead to fibrosis, suggesting that miRNA markers could be explored for predicting pulmonary fibrosis sustained due to extensive partial- or total-body irradiation. Thus, not only are miRNA biomarkers helpful in predicting survival outcomes, but also these markers can accurately predict lung injury after ionizing radiation (May et al. 2022).

Even though the role of several miRNAs in predicting radiation injury is very significant, the differential expression of any single miRNA can depend on dose and post-irradiation time. Therefore, using multiple miRNA biomarkers in a panel to differentiate doses at different time points following exposure is essential. In a whole-body irradiated mouse model, significant alterations in the expression patterns of lncRNA, miRNAs, and target mRNAs were observed at various time points after different doses of radiation (Aryankalayil, Chopra, Levin et al. 2018; Aryankalayil, Chopra, Makinde et al. 2018). Importantly to triage victims of a radiological incident, a single biomarker, for example, let-7e-3p, may have limited utility within a set of biomarkers because of their varying upregulation and downregulation across radiation doses and timepoints (Coleman et al. 2020). While this may also be true in radiotherapy-induced pulmonary fibrosis, the blood-based RNA is currently the primary molecule of interest. In contrast, other circulating markers, such as exosome-derived RNA, including circular RNAs, might also help predict tissue-specific injury after radiation treatment (Coleman et al. 2020).

Conclusion and future directions

RIPF is an adverse effect of thoracic irradiation. Its pathogenesis involves lung parenchyma, vasculature, inter-alveolar septa, pleura, and excessive deposition of ECM (Stone et al. 2003; Prasanna et al. 2012; Citrin et al. 2017). As a response to radiation, excessive production and release of pro-inflammatory, profibrotic factors such as IL-13 and TGF-β, TNF-α, promote EMT and stimulate interstitial fibroblasts and myofibroblast proliferation, ultimately leading to lung fibrosis (Wilson and Wynn 2009; Wang et al. 2015). Identification of RAS as a critical factor in the pathogenesis is an essential milestone in our understanding of RIPF, which lead to the successful evaluation of several inhibitors of ACE as interventional treatment approaches (Moulder and Cohen 2007; Ghosh, Zhang et al. 2009; Uhal et al. 2012).

Rapid progress in the field of RIPF has led to the identification of several targetable molecular pathways for pharmacological interventions of RIPF. These include but are not limited to RAS, TGF-β, SMAD signaling, senescence, SASP, EMT, p53, p21, MDM2, SIRT1, MYC, TERT, etc. Several of these pathways are also governed by post-transcriptional regulation of RIPF by miRNAs.

This article attempts to identify the roles of several miRNAs and TIS in the onset and progression of fibrosis beyond the traditionally studied molecular pathways to generate interest in research opportunities for developing next-generation mitigators against RIPF. Figure 2 provides an overview of molecular pathways that are regulated by miRNAs. While several approaches of miRNA-based interventional strategies can be created, three examples are discussed below.

First, given that several miRNAs are also involved in regulating radiation-induced cell death and govern cellular responses to DNA damage and replication stressors, these can serve as putative targets for early intervention (Li et al. 2013; Wang et al. 2013). For example, miR-185 is a negative regulator of ATR expression at the post-transcriptional level, inhibits cellular proliferation by repressing the ATR pathway, thereby enhancing radiation-induced apoptosis, and thus can be used to sensitize tumor cells to reduce normal tissue injury (Wang et al. 2013). Similarly, miR-207 enhances radiation-induced apoptosis by directly targeting AKT3, and anti-miR-207 might protect normal tissue from radiation injury (Tan et al. 2014).

Second, numerous miRNAs are implicated in TIS, and SnCs are an essential target. Eliminating SnCs in a ‘one-two punch’ cancer therapy paradigm with senolytics can inhibit proinflammatory factors, including SASP, reduce EMT and potentially prevent fibrosis (Prasanna et al. 2021). For example, inhibition of Bcl-2/XL with ABT-263 selectively eliminated senescent AECII and reversed persistent RIPF (Pan et al. 2017). Typical senescence-associated features, including increased beta-galactosidase activity (SA-ß-gal) and p21 due to activation of ROS and increased expression of pro-inflammatory cytokines, could be reduced with a treatment with quercetin, which decreased miR-155-5p expression possibly through the modulation of NF-κB and SIRT-1 (Zoico et al. 2021). The role of cellular senescence in RIPF and the development of senolytics as a therapeutic strategy is a rapidly emerging field research area (He et al. 2019).

Third, miRNAs have been found to regulate several fibrogenic pathways (e.g. TGF-β, SMAD, etc.). Targeting profibrotic pathways using small molecule inhibitors can prevent mRNA degradation associated with fibrosis. In an IPF model, miR-21 upregulation in mouse lungs was associated with profibrogenic activity, which could be mitigated by administering miR-21 antisense probes (Liu et al. 2010). Antifibrogenic miRNAs such as miR-9-p, miR-29, and miR-153 possess endogenous protective activity, which can be mimicked and intensified, enabling the downregulation of profibrotic mRNAs (Chan et al. 2012; Liang et al. 2015). Similarly, miR-155-5p was significantly down-regulated in radiation-induced EMT and RIPF, and ectopic miR-155-5p expression inhibited radiation-induced-EMT and ectopic miR-155-5p expression alleviated RIPF in mice via the GSK-3β/NF-κB pathway. Thus, radiation downregulates miR-155-5p in AECs that induce EMT, which contributes to RIPF using the GSK-3β/NF-κB pathway, a functional target of miR-155-5p (Wang et al. 2021).

Thus, miRNA-based therapeutics represent a new category of small molecule pharmaceuticals and may add to the arsenal of RIPF therapeutics, regulating the expression of crucial signaling fibrogenesis pathways. However, only a small number of miRNA-based therapeutics have moved to the clinical studies (Rupaimoole et al. 2016). One major challenge in miRNA-based therapeutics is to identify the best candidates or therapeutic targets for RIPF due to the heterogeneity of miRNA expression. Other challenges include the design of miRNA delivery vehicles, drug stability, tissue-specific targeting, and off-target toxicities.

Predictive biomarkers for late-occurring RIPF are desperately needed. Several reports indicate a causal relationship between RIPF, TIS, miRNAs, and RIPF (Bhaumik et al. 2009; Munk et al. 2017). However, given the causal relationship and the heterogeneity in the response of miRNAs to dose- and time-, which are tissue-specific further understanding of molecular pathways that govern RIPF at the intersection of TIS and RIPF for using miRNAs as predictive biomarkers is crucial. Further, a single miRNA may not be sufficient to predict RIPF; it is more likely that a miRNA profile may be necessary to predict TIS and RIPF. Such an approach may eventually lead to the early prediction of RIPF and the personalization of treatment following therapeutic or accidental radiation exposures. In summary, our analysis of published literature indicates that TIS is a mechanism by which radiation induces fibrosis and contributes to its progression, and our understanding of its regulation through miRNAs is starting to emerge.

Epilog

That progress has been made in understanding and potentially mitigating RIPF is due in good measure to the efforts of John Moulder. We coauthors from the NCI Radiation Research Program and Radiation Oncology Branch have been colleagues and friends of his, some of us, for decades. His research has been insightful and impeccably well done, with his interpretations honest, carefully crafted, and always open for discussion of the next steps. Being of a similar vintage to the senior author of this mini-review (CNC), a curiosity about how he started in radiation brought us to one of his earliest papers, ‘The steepness of the dose-response curve in radiation therapy. Theoretical considerations and experimental results’), coauthored with Jim J Fischer (Fischer and Moulder 1975), a colleague of CNC from Yale who was one of the brightest leaders in our field. John had strong roots in innovation and basic science, and through his tenacity and quality investigation, his research and clinical colleagues were world leaders in normal tissue injury. When the specter of 9/11/2001 unfolded, John quickly joined in with the radiation scientific community answering the societal request for drugs that would pharmacologically mitigate the adverse health effects of a potential terrorist improvised nuclear device (Coleman et al. 2004). He reminded us to live an active life through his actions, including remarkable long-distance cycling. His wealth of colleagues and friends gravitated to him for his humor and genuine appreciation of the efforts and work of others. John’s is a legacy of character and enthusiasm whose impact will remain within the foundation for understanding normal tissue radiation injury and as an example of the power of honesty and straightforward home-spun decency.

Author contributions

PGSP designed, acquired, analyzed, interpreted the data, and drafted the initial version of the manuscript. MA contributed to the design, acquisition, analysis, and interpretation of the data on miRNAs and biomarkers. DEC contributed to the design and validated the data on therapy-induced senescence and miRNAs, and critically reviewed the manuscript. CNC contributed to the design, research, and interpretation of the data, reviewing the draft, and drafting the epilog. All authors approved the final version of the manuscript.

Abbreviations
ACE	=	Angiotensin-converting enzyme
AEC I and II	=	Alveolar epithelial cell types I and II
Ang I and II	=	Angiotensin I and II
AT1 and ATII	=	Angiotensin Type 1 and II receptors
ATM	=	Ataxia telangiectasia mutated
BCL	=	B-cell lymphoma
CBX	=	Chromobox homolog
CDK	=	Cyclin D kinase
CCL	=	Chemokine ligand
CXCL	=	Chemokine (C-X-C motif) ligand
CTGF	=	Connective tissue growth factor
DDR	=	DNA damage response
ECM	=	Extracellular matrix
EGF	=	Epidermal growth factor
EGFR	=	Epidermal growth factor receptor
EMT	=	Epithelial-mesenchymal transition
GSK	=	Glycogen synthase kinase
ICAM	=	intercellular adhesion molecule
IL	=	Interleukin
IPF	=	Idiopathic pulmonary fibrosis
IRAK	=	Interleukin receptor-associated kinase
mRNA	=	messenger RNA
miRNA(s)	=	microRNA(s)
MCD	=	Mast cell deregulating peptide
MDM2	=	Murine double minute
MMP	=	Matrix metalloproteinases
MSC	=	Mesenchymal stem cells
ncRNA	=	noncoding RNA
NF-κB	=	Nuclear factor kappa B
NHP	=	non-human primate
PDGF	=	Platelet-derived growth factor
PMC	=	Pleural mesothelial cells
PRC1 and 2	=	Polycomb repressive complexes
PTEN	=	Phosphatase and tensin homolog
RAS	=	Renin-angiotensin system
RB	=	Retinoblastoma
RIPF	=	Radiation-induced pulmonary fibrosis
ROS	=	Reactive oxygen species
SASP	=	Senescence associated secretory phenotype
SIRT	=	Sirtuin
SMAD	=	Suppressor of mothers against decapentaplegic
SnCs	=	Senescent cells
SOD	=	Super oxide dismutase
TERT	=	Telomerase reverse transcriptase
TGF-β	=	transforming growth factor - β
TIMP	=	Tissue inhibitors of metalloproteinases
TIS	=	Therapy-induced senescence
TLR	=	Toll-like receptor
TNF	=	Tumor necrosis factor
VEGF	=	Vascular endothelial growth factor

Disclosure statement

The views and opinions expressed in this article are those of the authors and do not reflect the views, opinions, or policies of the NCI, NIH, or HHS. Authors have no competing interests to declare.

Additional information

Funding

This work was supported by the Radiation Research Program, Division of Cancer Treatment and Diagnosis, and the Radiation Oncology Branch, Center for Cancer Research, National Cancer Institute, NIH, Bethesda, MD, USA.

Notes on contributors

Pataje G. S. Prasanna

Pataje G. S. Prasanna, PhD, is a Program Director in the Radiation Research Program, Division of Cancer Treatment and Diagnosis, at the National Cancer Institute, Bethesda, Maryland.

Molykutty Aryankalayil

Molykutty Aryankalayil, PhD, is a Research Scientist in the Experimental Therapeutics Section, Radiation Oncology Branch at the National Cancer Institute, Bethesda, Maryland.

Deborah E. Citrin

Deborah E. Citrin, MD, is a Senior Investigator in the Radiation Oncology Branch of the Center for Cancer Research, National Cancer Institute, in Bethesda, Maryland.

C. Norman Coleman

C. Norman Coleman, MD, is Associate Director of the Radiation Research Program, Division of Cancer Treatment and Diagnosis, Senior Investigator, Radiation Oncology Branch, Center for Cancer Research, National Cancer Institute, Bethesda, MD, and Senior Medical Advisor, Administration for Strategic Preparedness and Response, Department of Health and Human Services, Washington DC.