Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jan 16.
Published in final edited form as: Respir Physiol Neurobiol. 2023 Apr 7;313:104060. doi: 10.1016/j.resp.2023.104060

Effects of DNA methylase inhibitors in a murine model of severe BPD

Kathryn M Heyob a, Zahra Khuhro a, Aiman Q Khan a, Dorian Brown a, Trent E Tipple c, Lynette K Rogers a,b,*
PMCID: PMC11736813  NIHMSID: NIHMS2043703  PMID: 37031925

Abstract

DNA methylation is necessary for developmental gene regulation, but adverse environments result in aberrant methylation and gene silencing. The current pilot study tested the hypothesis that treatment with DNA methylation inhibitors (decitabine; RG108) would improve alveolarization in a newborn murine model of severe bronchopulmonary dysplasia. Newborn mice exposed to maternal inflammation (LPS) and neonatal hyperoxia (85% O2) were treated with decitabine (p3, 0.1 mg/kg; p2, 4, 6, 0.1 mg/kg; or p2, 4, 6, 0.15 mg/kg) or RG108 (p3, 0.0013 mg/kg) delivered intranasally. Modest improvements in alveolarization were observed with decitabine, but no differences were observed with RG108. Attenuated phospho-SMAD2/3 levels and greater surfactant protein C protein levels compared to vehicle were observed with some tested doses. No detrimental side effects were observed with the doses used in this study. In summary, our pilot investigations identified a safe dose for intranasal administration of both methylation inhibitors and provides a foundation for further studies into methylation inhibitors in the context of neonatal lung injury.

Keywords: Alveolarization, DNMT, TGFβ1, RG108, decitabine

1. Introduction

Growing evidence indicates that early life environment significantly impacts development and life-long health. Epigenetic modifications cause changes in the expression of genes and thus protein levels and activity. Altered epigenetic signatures often result in abnormal development or disease progression including body weight, glucose metabolism, energy production, and neurodevelopment (reviewed by Scarpato (Scarpato et al., 2020; Diaz et al., 2017; Hillman et al., 2015; Tilley et al., 2018). One significant epigenetic modification is DNA methylation which involves the addition of a methyl group to cytosine residues in CpG dinucleotides. DNA methylation and demethylation are essential events that regulate normal development. In healthy mammals approximately 70% of CpG-enriched DNA is methylated at any given time. During normal development or under homeostatic conditions, cytosine methylation is tightly controlled by the activities of DNA methyltransferases (DNMTs) (Cuna, 2015). Hypo- or hypermethylated regions have been linked to maternal health during pregnancy, specifically obesity and extreme preterm birth (Tilley et al., 2018). Further, maternal exposure to either internal or external stressors has been associated with altered placental methylation patterns and development of chronic lung disease in preterm infants (reviewed by Jackson 2022(Jackson et al., 2022)). Irregular or abnormal methylation patterns have been identified in several pulmonary pathologies such as asthma (Bae et al., 2020), acute respiratory distress syndrome (Huang et al., 2016), and developmental morbidities such as bronchopulmonary dysplasia (BPD) (Bik-Multanowski et al., 2018; Everson et al., 2020; Zhao et al., 2018). BPD is a serious pulmonary morbidity afflicting preterm infants and is associated with life-long pulmonary insufficiency (Day and Ryan, 2017). Several studies both in human and animal models have identified epigenetic alterations in CpG methylation and microRNA expression in the context of BPD pathology (Durrani-Kolarik et al., 2017; Everson et al., 2020; Olave et al., 2016; Tong et al., 2021). Developmental changes to DNA patterns have been analyzed in preterm vs term lung tissues and 20 genes and 3 pathways have been identified as having differential methylation patterns (Cuna et al., 2015). Further, Cuna, et al. found 32 genes with differential methylation in lung tissues from infants born preterm vs preterm with BPD (Cuna et al., 2015). We and others have found global DNA hypermethylation in autopsy lung tissues of infants diagnosed with BPD vs infants without BPD and in mouse models of BPD (Robbins et al., 2016; Rogers et al., 2015).

DNA methylation is regulated by three mammalian DNA methyltransferases (DNMT1, 3a, 3b) (Robertson et al., 1999). DNMT1 functions to maintain methylation of silenced genes while DNMT 3a and 3b function dynamically to facilitate responses to developmental and environmental stimuli. DNMT inhibitors have been studied for many years in the context of malignancies and used to target demethylation of tumor suppressor genes (Agrawal et al., 2018; Schneeberger et al., 2016; Yu et al., 2019). A small number of studies have tested the effectiveness of decitabine in acute lung injury models. In an adult model of cigarette smoke extract induced emphysema, treatment with decitabine improved lung function and preserved alveolar structure (He et al., 2016; He et al., 2017). Further, in an acute LPS-induced ARDS model, pretreatment with decitabine prevented increases in oxidative stress markers, and suppressed increases in phosphorylation of MAPK signaling pathways (Huang et al., 2016). More relevant to the current study, Zhao, et al. reported increased survival and modestly improved alveolarization in a rat model of newborn hyperoxia exposure (Zhao et al., 2018).

Currently, there are two primary classes of DNMT inhibitors, those that are nucleosidic, incorporate in DNA, and irreversibly bind DNMTs, such as decitabine (5-deoxy-azacytidine) and non-nucleosidic, that interfere with the transfer of the methyl group to the DNA base, such as RG108 (N-phthalyl-L-tryptophan) (Agrawal et al., 2018; Schneeberger et al., 2016). Methylation of tumor suppressor genes is an established mechanism in cancer progression, and decitabine is currently used to treat hematologic malignancies (Dan et al., 2019). Animal models of lung injury such as cigarette smoke induced emphysema (He et al., 2016, 2017), acute lung injury (Bik-Multanowski et al., 2018; Everson et al., 2020), and hyperoxic lung injury (Zhao et al., 2018) have reported improvement in lung function and alveolarization with decreases in fibrotic score and markers of oxidant stress after treatment with methylation inhibitors.

The goal of the present pilot study was to investigate the potential of methylation inhibitors, administered early in life, to improve lung growth and attenuate aberrant matrix remodeling in our murine model of severe BPD.

2. Methods

2.1. Animal studies

C3H/HeN mice (Envigo, Indianapolis, IN, USA) were paired for breeding, and the presence of a vaginal plug was designated as embryonic day 1 (E1). On E16, pregnant dams were injected intraperitoneally with 0.08 mg/kg lipopolysaccharide (LPS, serotype 0111:B4, no. 437627, Calbiochem), or an equal volume of saline. Within 12 h of birth, pups were redistributed between two dams and one group of pups was exposed to 85% oxygen for 14 days, while the other remained in room air. Dams were switched daily between the paired room air and oxygen litters to prevent oxygen toxicity (Durrani-Kolarik et al., 2017; Sugar et al., 2021; Velten et al., 2014, 2012, 2010). Pups were dosed intranasally with 0.1 mg/kg decitabine (MedChemExpress, Monmouth NJ, USA, HY-A0004) or saline on postnatal day 3 (p3), or dosed multiple days (p2, 4, 6) with 0.1 or 0.15 mg/kg decitabine (dissolved in sterile phosphate buffered saline and sonicated in a warm water bath for 5 min to increase solubility). Another set of pups were dosed p3 with 0.0013 mg/kg RG108 (Medchemexpress, Monmouth NJ, USA, HY-13642) (prepared by making a 0.1 mg/mL solution, heating to 70 oC in a heated shaker and sonicating for 10 sec in a warm water bath). Intranasal delivery was accomplished by holding the pup by the nape of the neck so that the head was vertical, placing 1–3 μl of drug directly on the nares of the mouse pup and then gently stroking the abdomen to induce a deep inhalation. Previously, this protocol of intranasal delivery was demonstrated in Durrani-Kolarik, et al. using GFP labeled treatment (Durrani-Kolarik et al., 2017), however we acknowledge that there may be differences between efficacy of a viral vector and diluted drug. Litters in oxygen were returned to room air on p14.

Due to the number of treatment groups, seventeen litters were analyzed over three complete experiments. Four to eleven mice per treatment group were analyzed for morphometrics with no more than two mice, one per each sex, included from any specific litter. Four to six mice per treatment group were included in the biochemical analyses with no more than two mice, one per each sex, from any specific litter. On p21, mice were euthanized, body weights and lung weights were recorded, and tissues were collected for analysis. No differences in body weights or lung weights were observed at p21 (data not shown). No sex differences in lung injury or lung developmental deficits during early life have been previously identified in this model and preliminary analyses revealed no differences were observed in these studies. Consequently, equal numbers of males and females were used for each analysis and the results are shown as mixed. While vehicle doses were tested in all groups and were not different, only the 0.1 mg/kg, day 3 dose in the saline/RA groups was added to the figures for simplicity.

All animal experiments were performed with approved Institutional Animal Care and Use protocols at Nationwide Children’s Hospital, Columbus, Ohio (IACUC #AR07–0028,) in accordance with National Institutes of Health and ARRIVE guidelines.

2.2. Morphometric analysis of lung tissue

Mice were euthanized by over-anesthesia with ketamine/xylazine at P21, and the left lung was inflation fixed with formalin at 25 cm H2O. Following paraffin embedding, tissue section (5 μm) were cut, and slides were stained with hematoxylin and eosin (H&E) for morphometric measurements as previously described (Durrani-Kolarik et al., 2017; Park et al., 2007; Sugar et al., 2021; Velten et al., 2012, 2010). Five images per animal were analyzed and averaged using digital image analysis software (Image-Pro® Plus 6.3; Media Cybernetics, Silver Spring, MD). Data are reported as number of alveoli per high power field and digitally calculated area of each alveolus.

2.3. PCR

Total RNA was isolated from frozen lung tissues using the RNeasy Mini kit (Qiagen, Valencia, CA). RNA was reverse-transcribed using Thermo Scientific Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Fisher Scientific, Waltham MA). Real-time PCR was performed using the Maxima SYBR Green/ROX qPCR Master Mix (Thermo Fisher Scientific) on the QuantStudio 6 Flex system (Thermo Fisher Scientific). PCR primers are listed in Table 1. Data are normalized to β-actin and expressed as fold change over the saline/RA exposure, vehicle (saline) treatment group.

2.4. Western blot analysis

Lung tissues were lysed via bead homogenization. Proteins were separated on 10% SDS-polyacrylamide gels and transferred to polyvinylidene fluoride membranes. After blocking, membranes were probed with primary antibodies to surfactant protein C (SPC) (ab211326, Abcam, Waltham, MA, USA), phospho-Smad 2 (#3108, Cell Signaling, Danvers, MA) Smad 2/3 (#3102, Cell Signaling, Danvers, MA), phospho-extracellular signal-regulated kinase (pERK)(# 20G11, Cell Signaling, Danvers, MA), ERK (# 137F5, Cell Signaling, Danvers, MA) and species specific horseradish peroxidase-labeled secondary antibodies (BioRad, Hercules, CA). Membranes were developed using enhanced chemiluminescence (ECL) with ChemiDoc imaging and quantified by densitometry using Image Lab software (v 5.2.1) (BioRad). Quantifications for all western blots were normalized by ratios of the phosphorylated to total protein (ERK and SMAD) or normalized to total protein using amido black staining.

2.5. Picro-Sirius red staining (PSR)

Lung tissues were embedded in paraffin and 6 μm sections were cut. Air-dried sections were deparaffinized using standard techniques. Staining was performed using 0.2% phosphomolybdic acid (26357–01, Electron Microscopy Sciences, Hatfield, PA) for 4 min, rinsed in deionized water and placed into 0.1% Sirius red solution in picric acid (Electron Microscopy Sciences, 26357–02) for 90 min. The tissue was washed in 0.01 N HCl for 2 min, then dehydrated in ascending ethanol concentrations (70%, 95%, and 100%) for 10 s each and set in xylene for 10 min. Sections were coverslipped with Permount mounting medium (Thermo Fisher Scientific, Waltham, MA). To quantify the Picro-Sirius red images, color brightfield and polarized Portable Network Graphics (png) images of Picro-Sirius red stained slides were captured on a Zeiss Axio Scope.A1 using an Axiocam 305 Color camera (426560–9030–000) (White Plains, NY). Irfanview was used to convert images to png format if necessary. Images were loaded into Python and split into their 3 primary colors using the Pillow module. Colors with relevant information were merged, converted to grayscale, and then into numpy arrays. Arrays were parsed and any value below a threshold was set to 0, while those above were set to 255 to create a black and white image. Thresholding was determined manually such that a single threshold would work for all color brightfield or all polarized images. A smoothing algorithm was used to remove lone black and lone white pixels. Each value of every pixel was recursively compared to its 8 neighbors and set to black or white if all 8 of its neighbors were also black or white respectively. The resulting images contained black pixels only in areas of interest against an all-white background. The number of polarized pixels was divided into the number of brightfield pixels to generate a ratio between collagen and tissue.

2.6. Immunohistochemistry

Paraffin embedded lung tissues were cut in 6 μm sections. Air-dried sections were deparaffinized using standard protocols. Antigen retrieval was performed in sodium citrate (10 mM pH 6.0) with 0.05% Tween-20 for one hour at 100 °C. Slides were blocked in 2.5% Normal Goat Serum Blocking Solution (S-1012–50, Vector Laboratories, Malvern, PA) for one hour, incubated overnight at 4 C with pro-surfactant protein C (SPC) primary antibody (ab211326, Abcam, Waltham, MA) at 1:500 in blocking buffer, followed by goat anti-rabbit IgG Alexa Fluor 594 (Invitrogen, Waltham, MA, USA), and nuclei were stained with DAPI (#2248, Thermo Fisher Scientific, Waltham, MA) in TBS at 1:200. Slides were coverslipped using Clearmount (Electron Microscopy Sciences, Hatfield, PA, US). Z-stack images were captured on a Zeiss 710 Confocal Microscope (White Plains, NY). All images were analyzed using a macro developed in ImageJ 1.53q, which quantified the number of nuclei and the number of cells positive for SPC. The macro selected one image within the Z-stack which showed the layer of maximum intensity for Texas Red and DAPI channels in grayscale, then split the channels into two separate images. Once speckles and noise from background were removed, objects of interest were counted using “Find Maxima” with specific prominence for each channel used for all images. The number of cells positive for SPC was divided into the number of nuclei to generate a ratio.

Statistics.

Data are expressed as means ± standard error of the mean (SEM). Statistical analyses were performed with GraphPad PRISM Windows version 8.0.0 (San Diego, CA) using a one-way ANOVA and Tukey’s post hoc.

3. Results

3.1. Morphometrics

Morphometric analysis of alveolarization was performed to determine the efficacy of methylation inhibitors to improve alveolarization. While LPS/O2 exposure caused significantly fewer alveoli and larger alveolar areas, post-hoc analysis revealed that intranasal dosing of decitabine (0.1 mg/kg) on p3 resulted in a significantly more alveoli (p = 0.008) compared to the LPS/O2 vehicle group (Fig. 1A). Further, post-hoc analysis revealed that smaller alveolar areas observed with a single dose (0.1 mg/kg) on p3 (p = 0.017) and multiple doses of decitabine (p2, 4, 6, 0.1 mg/kg) (p = 0.043) compared to the LPS/O2 vehicle group (Fig. 1B). The higher dose of decitabine (0.15 mg/kg) was not different than the LPS/O2 group dosed with vehicle. There was no improvement in alveolarization in the RG108-treated group. These data suggest that lower doses of decitabine modestly improved alveolarization.

Fig. 1.

Fig. 1.

Open in a new tab

Morphometric analysis. Lung tissue sections from p21 mice were stained with H&E. Photomicrographs were obtained from each section (five distinct fields) and alveolarization within each field was measured as described in Methods. A) number of alveoli per high power field; B) average area of individual alveoli using digital analysis. Data are expressed as means ± standard error of the mean (SEM) and analyzed by one way ANOVA with Tukey’s post hoc. n = 4–12 mice per group; scale bar = 100 μm.

3.2. PCR

Inflammation is a key indicator of inflammation-induced lung injury but may also be induced by treatment modalities, specifically those delivered directly to the lungs. Markers of inflammation and lung growth were assessed by RT-PCR. TNFα expression was different between exposures/treatments (p = 0.013, One way ANOVA) with the LPS/O2 exposed groups trending to be higher but post hoc analyses did not indicate significant individual differences (Fig. 2A). IL-1β expression in lung tissues was not different between exposures/treatment groups (Fig. 2B). TNFα and IL-1β measurements indicate that exposure associated inflammation was largely resolved at p21 and that dosing with methylation inhibitors did not cause lasting inflammation (Figs 2A and 2B). TGFβ1 expression is essential for lung growth and in matrix structure. TGFβ1 mRNA expression was higher in the LPS/O2 decitabine multiple dosed groups (p2, 4, 6; 0.1 mg/kg (p = 0.007) and p2, 4, 6; 0.15 mg/kg (p = 0.002)) than the LPS/O2 vehicle group, implicating improved growth or modest activation of matrix remodeling mechanisms. Additionally, the LPS/O2 RG108 group had greater expression of TGFβ1 than the saline/RA RG108 group, indicating a potentially synergistic increase due to dosing and exposure (Fig. 2C). VEGF expression is associated with angiogenesis which is essential to normal lung development. VEGF mRNA was higher in the decitabine-treated groups (p = 0.011, One way ANOVA), but no post hoc statistical differences were observed (Fig. 2D). DNMT1, 3a, and 3b expression are responsible for normal developmental methylation (DNMT1) as well as methylation associated with injury or disease (DNMT3a and 3b). DNMT expression levels were measured to determine if methylation inhibitors administered during early life exhibited persistent effects. All three DNMTs exhibited differences within individual groups by one way ANOVA (DNMT1, p = 0.043; DNMT3a, p = 0.014; DNMT3b, p = 0.030) but no clear pattern was observed (Fig. 3). Post hoc differences between both the vehicle dosed groups and the highest decitabine dose at multiple doses (decitabine p2, 4, 6; 0.15 mg/kg) were observed with DNMT1 but no other post hoc differences were indicated. Overall, methylation inhibitors caused no long-term inflammation, modestly improved parameters of growth (TGFβ1 and VEGF) and had no lasting effect on DNMT expression.

Fig. 2.

Fig. 2.

Open in a new tab

PCR analysis of inflammatory markers and growth factors. PCR was performed on RNA isolated from snap-frozen lung tissues. Data are presented as fold change compared to saline/RA vehicle. A) TNFα; B) IL-1β; C) TGFβ1; and D) VEGF. Data are expressed as means ± standard error of the mean (SEM) and analyzed by one way ANOVA with Tukey’s post hoc. n = 4–6 mice per group.

Fig. 3.

Fig. 3.

Open in a new tab

PCR analysis of DNMTs. PCR was performed on RNA isolated from snap-frozen lung tissues. Data are presented as fold change compared to saline/RA vehicle. A) DNMT1; B) DNMT3a; C) DNMT 3b. Data are expressed as means ± standard error of the mean (SEM) and analyzed by one way ANOVA with Tukey’s post hoc. n = 4–6 per group.

3.3. Western blots

To determine the effects of methylation inhibitors on expression of representative proteins, western blots were performed on extracellular related kinase (ERK), as a downstream marker of persistent inflammation, phosphorylated SMAD, as a marker of TGFβ1 function, activity, and matrix remodeling, and SPC, as a marker of lung epithelial integrity. Protein levels of phospho/total (p/t) ERK were not different between exposure/treatment groups (Fig. 4A). There was a trend toward higher levels of p/t SMAD2/SMAD2/3 in all the LPS/O2 exposed groups (p = 0.008, One way ANOVA). Lower protein levels were observed in decitabine (p2, 4, 6; 0.10 mg/kg) treatment (p = 0.008), and RG108 treatment (p = 0.043) compared to LPS/O2 vehicle. (Fig. 4B). This observation might imply a damping of aberrant matrix remodeling with decitabine treatment. Interestingly, protein levels of SPC were different between exposures (p < 0.001, One way ANOVA) and protein levels were higher in both LPS/O2 decitabine multiple dose groups compared to either of the vehicle dosed group (Fig. 4C). This observation would indicate that inhibition of methylation in early life may enhance SPC protein expression.

Fig. 4.

Fig. 4.

Open in a new tab

Western blot of key proteins. Lung tissues from p21 mice were homogenized and proteins separated by SDS-PAGE. Proteins were transferred to PVDF membranes and probed with the respective antibodies. A) p/t ERK; B) p/t SMAD 2/3; C) SPC. Data are expressed as means ± standard error of the mean (SEM) and analyzed by one way ANOVA with Tukey’s post hoc. n = 5–6 per group.

3.4. Picro-Sirius red

Tissues sections were stained with Picro-Sirius red to identify changes in collagen deposition due to methylation inhibitor treatment. We observed increased Picro-Sirius red staining (fluorescence) in all LPS/O2 exposed groups (p < 0.001) with a substantially greater fluorescence in the decitabine p2, 4, 6; 0.15 mg/kg, compared to all other groups (Fig. 5). This would indicate that methylation inhibitors did not attenuate the increases in collagen deposition observed in the LPS/O2 exposed mice and the even greater increase observed with the high dose decitabine on multiple days may have detrimental effects on increased collagen deposition.

Fig. 5.

Fig. 5.

Open in a new tab

Picro-Sirius red staining. Lung tissue section from p21 mice were stained with Picro-Sirius red as described in Methods. Photomicrographs were obtained from each section (five distinct fields) and fluorescence within each field was measured as described in Methods. Data are presented as fluorescent pixels per total tissue area. Data are expressed as means ± standard error of the mean (SEM) and analyzed by one way ANOVA with Tukey’s post hoc. n = 4–5 per group; scale bar = 100 μm.

3.5. SPC immunohistochemistry

Tissues sections were treated with SPC antibody and DAPI staining to identify SPC expression in situ. The number of cells fluorescent for SPC divided by the total number of DAPI positive nuclei (Fig. 6) was calculated for each slide to compensate for the decreased tissue content per high power field in the LPS/O2-treated mice. The results indicated differences between groups (p < 0.0001, One way ANOVA), with no observed differences between saline/RA vehicle and LPS/O2 vehicle treated groups. Lower levels of fluorescence were observed in the decitabine, (p3, 0.1 mg/kg) and the RG108 treated groups. This may indicate detrimental effects of methylation inhibitors on lung cell integrity.

Fig. 6.

Fig. 6.

Open in a new tab

SPC immunohistochemistry. Lung tissue sections from p21 mice were processed with SPC antibodies. Photomicrographs were obtained from each section (five distinct fields) and fluorescence within each field was measured as described in Methods. Total SPC positive cells were divided by the number of DAPI-positive nuclei. Data are expressed as means ± standard error of the mean (SEM) and analyzed by one way ANOVA with Tukey’s post hoc. n = 4 per group; scale bar = 200 μm.

4. Discussion

Using our well characterized murine model of severe BPD, we performed this pilot study to test the hypothesis that early intervention with a DNMT-blocking agent might attenuate the deficits in lung development previously observed in this model (Durrani-Kolarik et al., 2017; Robbins et al., 2016; Sugar et al., 2021; Velten et al., 2012, 2010). Our investigations began with 5-azacytidine, a precursor to the currently used decitabine. These experiments were halted due to extensive toxicity observed with intranasal dosing of newborn mice (data not shown). The more commonly used and less toxic derivative of 5-azacytidine, decitabine, became the focus of the current investigations. The dose used was based on the manufacturer’s suggested ip dose which we divided by one-tenth because of intranasal dosing in newborn pups. At the time these studies were initiated, there were no published studies in vivo using RG108. A dose was calculated using the manufacturer’s IC50 (115 nM). We readily acknowledge that the dose of RG108 may have been inadequate but was a starting point for these pilot studies. Earlier studies have indicated that acute inflammation due to hyperoxia exposure begins at approximately p3 (Rogers et al., 2009), thus we chose to begin dosing pups on p3. We have observed growth retardation previously in this model, with catch-up growth after 21 days. In the present study, we did not weigh the mice prior to 21 d, and consequently were unable to ascertain changes in growth due to the methylation inhibitors. We did observ modest improvements in alveolarization with a single dose of decitabine at p3 (Fig. 1), similar to that observed by Zhao et. al (Zhao et al., 2018). After identifying modest improvements in alveolarization, we chose to investigate multiple doses of decitabine at p2, 4, and 6 and a higher dose (0.15 mg/kg) to improve efficacy. Interestingly, no further improvements in alveolarization were observed (Fig. 1). RG108 has not been previously used to treat pulmonary injury and because of its alternative mechanism of methylation inhibition, we chose to test its efficacy in the same environment as decitabine (Dan et al., 2019; Schneeberger et al., 2016). RG108 was also administered on P3, using a calculated therapeutic dose. We observed no improvements in alveolarization with RG108, however alternative dosing regimens were not tested, and any lack of efficacy could be to poor drug distribution or that a therapeutic dose was not achieved. (Fig. 1). Further investigations are needed to determine if RG108 can be a viable option in our model.

Murine BPD models have identified early and sustained inflammation as key components to tissue injury and disruption of growth (Ryan et al., 2008). We measured two key inflammatory markers, TNFα and IL-1β to determine if demethylation treatment early in life changed the course of inflammatory resolution in these mice. We observed differences in TNFα between groups indicating a modest elevation in TNFα due to LPS/O2 treatment but no post hoc differences within groups (Fig. 2A). No differences in IL-1β were observed among all groups (Fig 2B). Alternatively, mitogen activated kinases (MAPK) are key regulators of inflammation and Huang, et al. reported modulation of MAPKs by decitabine administration in a model of LPS-induced acute lung injury (Huang et al., 2016). This led us to investigate the long-term effects of methylation inhibitors on the expression phosphorylation of extracellular signal-related kinase, ERK, a classical MAPK protein. We observed differences with the LPS/O2-exposed groups having higher levels of pERK but no post hoc differences were observed with treatments (Fig. 4A). This observation would indicate that most inflammation caused by maternal LPS or neonatal hyperoxia exposure is resolved by 21 days in our model and that dosing with the methylase inhibitors does not induce chronic inflammation.

Further, we examined the growth factors TGFβ1 and VEGF to identify any changes that occurred due to administration of the methylation inhibitors. One way ANOVA indicated differences between the saline/RA groups and the LPS/O2 groups with the latter having higher expression. Post-hoc analysis revealed greater TGFβ1 expression in pups dosed with both multiple doses of decitabine compared to the LPS/O2 vehicle and in pups dosed with RG108 compared to the saline/RA RG108 treated mice (Fig. 2C). TGFβ1 is essential for normal lung growth and development and is the isoform most closely associated with aberrant lung development in BPD. Ironically, TGFβ1 expression is essential for regulation of epithelial mesenchymal transition, adherens junction and tight junction proteins, branching morphogenesis, and type 1 cell differentiation in the embryonic and early stages of postnatal lung development but increases antenatally are associated with impaired lung branching, interstitial fibrosis and increased extra-cellular matrix deposition (Alejandre-Alcazar et al., 2007; Gauldie et al., 2003; Calthorpe et al., 2023). Dysregulation of TGFβ1 is associated with adult pulmonary fibrotic diseases such as chronic obstructive pulmonary disease and idiopathic pulmonary fibrosis reviewed by (Saito et al., 2018). Methylation of TGFβ1 pathway effectors has been reported in other rodent models of BPD and thus were of interest in these studies (Bik-Multanowski et al., 2018). To further characterize any differences in TGFβ1 expression with methylation inhibitor treatment, we investigated protein expression of the canonical pathway, phosphorylated SMAD2, a downstream effector of TGFβ1. As previously reported, p/tSMAD2/3 was elevated in the mice exposed to LPS/O2 compared to those exposed to saline/RA (Velten et al., 2014). Interestingly, mice treated with multiple doses of decitabine or a single dose of RG108 , demonstrated lower levels of p/t SMAD2/3 expression compared to the LPS/O2 vehicle treated group (Fig. 4B). Increases in TGFβ1 (Fig. 2C) but decreases in p/t SMAD 2/3 (Fig. 4B) as observed in this study, could be interpreted as enhancing alternative TGFβ1 signaling not directly related to the SMAD pathway but involving Rho GTPases or MAPK pathways.

VEGF is a key molecule in angiogenesis and vascularization of developing tissues (Abman, 2010). Suppression of VEGF expression has been observed in models of BPD, and methylation of VEGF was also observed (Bik-Multanowski et al., 2018). In the current study, there were differences in LPS/O2 decitabine -treated group trending toward higher VEGF expression than the saline/RA or LPS/O2 vehicle treated groups, but post hoc analysis did not indicate individual differences (Figs. 2C and 2D). This could indicate improved angiogenesis immediately after dosing and that multiple dosing would be required or that the present studies were insufficiently powered to overcome intra-animal variability.

Expression of DNMTs was also assessed at this later time point to identify any prolonged or chronic effects of early methylation inhibitor treatment. DNMT1 was increased in the multiple, higher dose decitabine (0.15 mg/kg) group (Fig. 3A). This is likely a rebound response to suppression in early life, but the consequences of this rebound are not known. Analysis of both DNMT3a and 3b indicated statistical differences between groups but post hoc analyses did not identify individual differences (Fig. 3B and C).

SPC is an essential lung protein and is produced by distal lung epithelial cells (Ruaro et al., 2021). Lung Type II cells line the distal airways and alveoli and are vulnerable to hyperoxia-induced lung injury (Ruaro et al., 2021). In our studies, we observed differences in SPC protein expression both by western blot and immunohistochemistry (Figs. 4C and 6). No individual differences in saline/RA and LPS/O2 vehicle treatments were observed at this later time point using western blot, but a substantial increased SPC protein expression in LPS/O2−-treated mice was observed with the multiple doses of decitabine (Fig. 4C). The implications of this increased protein expression are unclear but may be partially responsible for protecting the lung from injury or may be a lasting effect of methylation inhibition in these mice. Evaluation of SPC expression in lung tissue sections by immunohistochemical analyses and fluorescence detection indicated that SPC expression was not different between the saline/RA and LPS/O2 vehicle treated groups but that treatment with decitabine or RG108 at P3 caused a decrease in expression compared to vehicle controls (Fig. 6).

Our previous studies had identified abnormal collagen deposition in response to perinatal inflammation (Velten et al., 2012, 2010). Picro-sirius red staining is used to identify areas of abnormal collagen deposition, consequently, we investigated the effects of methylation inhibitors on Picro-Sirius red staining in lung tissue sections. While LPS/O2 treatment caused an increase in Picro-Sirius red fluorescence compared to saline/RA, we observed a further substantial increase in the groups treated with multiple doses of decitabine at the higher dose (p2, 4, 6; 0.15 mg/kg) (Fig. 5). This increase in collagen deposition correlates with the increased expression of DNMT1 (Fig. 3) and implies that the higher expression of the methylase may be partly responsible for aberrant collagen deposition in this group.

These studies provide new data into the effects of methylation inhibitors on a mouse model of severe BPD, and there are several strengths to our findings reported herein. First, we are one of few investigations into methylation inhibitors in the context of BPD. We were able to identify a tolerable dose for intranasal administration. Finally, we evaluated both nucleosidic and non-nucleosidic inhibitors. There are, however, several limitations. First, our studies evaluated long-term effects of methylation inhibitors at p21, when lung maturation is almost complete, and did not characterize early changes in expression levels. Further, the intricacy of intranasal administration of a drug to a very young mouse may cause variability in our evaluations. Because of the complexity of the double hit model, we were unable to perform extensive dose-response and time course evaluations and perhaps never reached an efficacious dose of RG108. There are also many newer methylation inhibitors that warrant further evaluation in developmental models with perhaps less toxicity.

In conclusion, some improvements in alveolarization were noted with intranasal administration of methylation inhibitors but further studies into dose-response relationships are needed. Interestingly, we observed increased production of SPC protein with multiple doses of decitabine and no lasting negative effects. These investigations begin to provide a foundation to develop a more targeted approach to DNA methylation observed in preterm infants with BPD.

Funding

This work was supported by the National Institutes of Health R01HD0880833 (L.K.R.).

Abbreviations:

DNMT

DNA methyltransferase

BPD

bronchopulmonary dysplasia

TNFα

tumor necrosis factor α

TGFβ

transforming growth factor β

IL1β

interleukin 1β

VEGF

vascular endothelial growth facator

References

  1. Abman SH, 2010. Impaired vascular endothelial growth factor signaling in the pathogenesis of neonatal pulmonary vascular disease. Adv. Exp. Med. Biol 661, 323–335. [DOI] [PubMed] [Google Scholar]
  2. Agrawal K, Das V, Vyas P, Hajduch M, 2018. Nucleosidic DNA demethylating epigenetic drugs - a comprehensive review from discovery to clinic. Pharmacol. Ther 188, 45–79. [DOI] [PubMed] [Google Scholar]
  3. Alejandre-Alcazar MA, Kwapiszewska G, Reiss I, Amarie OV, Marsh LM, Sevilla-Perez J, Wygrecka M, Eul B, Kobrich S, Hesse M, Schermuly RT, Seeger W, Eickelberg O, Morty RE, 2007. Hyperoxia modulates TGF-beta/BMP signaling in a mouse model of bronchopulmonary dysplasia. Am. J. Physiol. Lung Cell Mol. Physiol 292, L537–L549. [DOI] [PubMed] [Google Scholar]
  4. Bae DJ, Jun JA, Chang HS, Park JS, Park CS, 2020. Epigenetic changes in asthma: role of DNA CpG methylation. Tube Respir. Dis 83, 1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bik-Multanowski M, Revhaug C, Grabowska A, Dobosz A, Madetko-Talowska A, Zasada M, Saugstad OD, 2018. Hyperoxia induces epigenetic changes in newborn mice lungs. Free Radic. Biol. Med 121, 51–56. [DOI] [PubMed] [Google Scholar]
  6. Calthorpe R, Poulter D, Smyth AR, Sharkey D, Bhatt J, Jenkins G, Tatler AL, 2023. Complex roles of TGF-β signaling pathways in lung development and bronchopulmonary dysplasia. Am. J. Physiol. Lung Cell Mol. Physiol 324, L285–L296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cuna A, Halloran B, Faye-Petersen O, Kelly D, Crossman DK, Cui X, Pandit K, Kaminski N, Bhattacharya S, Ahmad A, Mariani TJ, Ambalavanan N, 2015. Alterations in gene expression and DNA methylation during murine and human lung alveolar septation. Am. J. Respir. Cell Mol. Biol 53, 60–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dan H, Zhang S, Zhou Y, Guan Q, 2019. DNA methyltransferase inhibitors: catalysts for antitumour immune responses. Onco Targets Ther. 12, 10903–10916. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Day CL, Ryan RM, 2017. Bronchopulmonary dysplasia: new becomes old again! Pediatr. Res. 81, 210–213. [DOI] [PubMed] [Google Scholar]
  10. Diaz M, Garcia C, Sebastiani G, de Zegher F, Lopez-Bermejo A, Ibanez L, 2017. Placental and cord blood methylation of genes involved in energy homeostasis: association with fetal growth and neonatal body composition. Diabetes 66, 779–784. [DOI] [PubMed] [Google Scholar]
  11. Durrani-Kolarik S, Pool CA, Gray A, Heyob KM, Cismowski MJ, Pryhuber G, Lee LJ, Yang Z, Tipple TE, Rogers LK, 2017. miR-29b supplementation decreases expression of matrix proteins and improves alveolarization in mice exposed to maternal inflammation and neonatal hyperoxia. Am. J. Physiol. Lung Cell Mol. Physiol 313, L339–L349. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Everson TM, O’Shea TM, Burt A, Hermetz K, Carter BS, Helderman J, Hofheimer JA, McGowan EC, Neal CR, Pastyrnak SL, Smith LM, Soliman A, DellaGrotta SA, Dansereau LM, Padbury JF, Lester BM, Marsit CJ, 2020. Serious neonatal morbidities are associated with differences in DNA methylation among very preterm infants. Clin. Epigenet 12, 151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gauldie J, Galt T, Bonniaud P, Robbins C, Kelly M, Warburton D, 2003. Transfer of the active form of transforming growth factor-beta 1 gene to newborn rat lung induces changes consistent with bronchopulmonary dysplasia. Am. J. Pathol 163, 2575–2584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. He ZH, Chen Y, Chen P, He SD, Ye JR, Liu D, 2016. Decitabine enhances stem cell antigen-1 expression in cigarette smoke extract-induced emphysema in animal model. Exp. Biol. Med 241, 131–139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. He ZH, Chen Y, Chen P, He SD, Zeng HH, Ye JR, Liu D, Cao J, 2017. 5-Aza-2’-deoxycytidine protects against emphysema in mice via suppressing p16(Ink4a) expression in lung tissue. Int. J. Chron. Obstruct. Pulmon. Dis 12, 3149–3158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hillman SL, Finer S, Smart MC, Mathews C, Lowe R, Rakyan VK, Hitman GA, Williams DJ, 2015. Novel DNA methylation profiles associated with key gene regulation and transcription pathways in blood and placenta of growth-restricted neonates. Epigenetics 10, 50–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Huang X, Kong G, Li Y, Zhu W, Xu H, Zhang X, Li J, Wang L, Zhang Z, Wu Y, Liu X, Wang X, 2016. Decitabine and 5-azacitidine both alleviate LPS induced ARDS through anti-inflammatory/antioxidant activity and protection of glycocalyx and inhibition of MAPK pathways in mice. Biomed. Pharmacother 84, 447–453. [DOI] [PubMed] [Google Scholar]
  18. Jackson WM, Santos HP Jr., Hartwell HJ, Gower WA, Chhabra D, Hagood JS, Laughon MM, Payton A, Smeester L, Roell K, O’Shea TM, Fry RC, 2022. Differential placental CpG methylation is associated with chronic lung disease of prematurity. Pediatr. Res 91, 1428–1435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Olave N, Lal CV, Halloran B, Pandit K, Cuna AC, Faye-Petersen OM, Kelly DR, Nicola T, Benos PV, Kaminski N, Ambalavanan N, 2016. Regulation of alveolar septation by microRNA-489. Am. J. Physiol. Lung Cell Mol. Physiol 310, L476–L487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Park MS, Rieger-Fackeldey E, Schanbacher BL, Cook AC, Bauer JA, Rogers LK, Hansen TN, Welty SE, Smith CV, 2007. Altered expressions of fibroblast growth factor receptors and alveolarization in neonatal mice exposed to 85% oxygen. Pediatr. Res 62, 652–657. [DOI] [PubMed] [Google Scholar]
  21. Robbins ME, Dakhlallah D, Marsh CB, Rogers LK, Tipple TE, 2016. Of mice and men: correlations between microRNA-17 approximately 92 cluster expression and promoter methylation in severe bronchopulmonary dysplasia. Am. J. Physiol. Lung Cell Mol. Physiol 311, L981–L984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Robertson KD, Uzvolgyi E, Liang G, Talmadge C, Sumegi J, Gonzales FA, Jones PA, 1999. The human DNA methyltransferases (DNMTs) 1, 3a and 3b: coordinate mRNA expression in normal tissues and overexpression in tumors. Nucleic Acids Res. 27, 2291–2298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Rogers LK, Tipple TE, Nelin LD, Welty SE, 2009. Differential responses in the lungs of newborn mouse pups exposed to 85% or >95% oxygen. Pediatr. Res 65, 33–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Rogers LK, Robbins M, Dakhlallah D, Yang Z, Lee LJ, Mikhail M, Nuovo G, Pryhuber GS, McGwin G, Marsh CB, Tipple TE, 2015. Attenuation of miR-17-92 cluster in bronchopulmonary dysplasia. Ann. Am. Thorac. Soc 12, 1506–1513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ruaro B, Salton F, Braga L, Wade B, Confalonieri P, Volpe MC, Baratella E, Maiocchi S, Confalonieri M, 2021. The history and mystery of alveolar epithelial type ii cells: focus on their physiologic and pathologic role in lung. Int. J. Mol. Sci 22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Ryan RM, Ahmed Q, Lakshminrusimha S, 2008. Inflammatory mediators in the immunobiology of bronchopulmonary dysplasia. Clin. Rev. Allergy Immunol 34, 174–190. [DOI] [PubMed] [Google Scholar]
  27. Saito A, Horie M, Nagase T, 2018. TFGβ singaling in lung health and disease. Int. J. Mol. Sci 19 (8), 2460. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Scarpato R, Testi S, Colosimo V, Garcia Crespo C, Micheli C, Azzara A, Tozzi MG, Ghirri P, 2020. Role of oxidative stress, genome damage and DNA methylation as determinants of pathological conditions in the newborn: an overview from conception to early neonatal stage. Mutat. Res. Rev. Mutat. Res 783, 108295. [DOI] [PubMed] [Google Scholar]
  29. Schneeberger Y, Stenzig J, Hubner F, Schaefer A, Reichenspurner H, Eschenhagen T, 2016. Pharmacokinetics of the experimental non-nucleosidic DNA methyl transferase inhibitor N-Phthalyl-L-Tryptophan (RG 108) in rats. Basic Clin. Pharmacol. Toxicol 118, 327–332. [DOI] [PubMed] [Google Scholar]
  30. Sugar SS, Heyob KM, Cheng X, Lee RJ, Rogers LK, 2021. Perinatal inflammation alters histone 3 and histone 4 methylation patterns: effects of MiR-29b supplementation. Redox Biol. 38, 101783. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Tilley SK, Martin EM, Smeester L, Joseph RM, Kuban KCK, Heeren TC, Dammann OU, O’Shea TM, Fry RC, 2018. Placental CpG methylation of infants born extremely preterm predicts cognitive impairment later in life. PLoS One 13, e0193271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Tong Y, Zhang S, Riddle S, Zhang L, Song R, Yue D, 2021. Intrauterine hypoxia and epigenetic programming in lung development and disease. Biomedicines 9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Velten M, Heyob KM, Rogers LK, Welty SE, 2010. Deficits in lung alveolarization and function after systemic maternal inflammation and neonatal hyperoxia exposure. J. Appl. Physiol 108, 1347–1356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Velten M, Britt RD Jr., Heyob KM, Tipple TE, Rogers LK, 2014. Maternal dietary docosahexaenoic acid supplementation attenuates fetal growth restriction and enhances pulmonary function in a newborn mouse model of perinatal inflammation. J. Nutr 144, 258–266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Velten M, Britt RD Jr., Heyob KM, Welty SE, Eiberger B, Tipple TE, Rogers LK , 2012. Prenatal inflammation exacerbates hyperoxia-induced functional and structural changes in adult mice. Am. J. Physiol. Regul. Integr. Comp. Physiol 303, R279–R290. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Yu J, Xie T, Wang Z, Wang X, Zeng S, Kang Y, Hou T, 2019. DNA methyltransferases: emerging targets for the discovery of inhibitors as potent anticancer drugs. Drug Discov. Today 24, 2323–2331. [DOI] [PubMed] [Google Scholar]
  37. Zhao SM, Wu HM, Cao ML, Han D, 2018. 5-aza-2’-deoxycytidine, a DNA methylation inhibitor, attenuates hyperoxia-induced lung fibrosis via re-expression of P16 in neonatal rats. Pediatr. Res 83, 723–730. [DOI] [PubMed] [Google Scholar]

RESOURCES