PARP1 Inhibition and Effect on Burn Injury-Induced Inflammatory Response and Cardiac Function

Abstract

BACKGROUND:

Burn injury induces multiple signaling pathways leading to a significant inflammatory storm that adversely affects multiple organs, including the heart. Poly (ADP-ribose) polymerase inhibitor 1 (PARP1) inhibition, with specific agents such as N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide (PJ34), is effective in reducing oxidative stress and cytokine expression in the heart. We hypothesized that PARP1 inhibition would reduce inflammatory signaling and protect against burn injury–induced cardiac dysfunction.

STUDY DESIGN:

Male Sprague-Dawley rats (8 weeks old, 300 to 350 g) were randomly assigned to sham injury (Sham), 60% total body surface area burn (24 hours post burn), or 60% total body surface area burn with intraperitoneal administration of PJ34 (20 mg/kg, 24 hours post burn + PJ34) and sacrificed 24 hours after injury. Cardiac function was determined using Vevo 2100 echocardiography. Genetic expression of 84 specific toll-like receptor–mediated signal transduction and innate immunity genes were examined using microarray to evaluate cardiac tissue. Qiagen GeneGlobe Data Analysis Center was used to analyze expression, and genetic clustering was performed using TreeView V2.0.8 software. Real-time quantitative polymerase chain reaction was used to validate identified differentially expressed genes.

RESULTS:

Burn injury significantly altered multiple genes in the toll-like receptor signaling, interleukin-17 signaling, tumor necrosis factor signaling, and nuclear factor-κB signaling pathways and led to significant cardiac dysfunction. PARP1 inhibition with PJ34 normalized these signaling pathways to sham levels as well as improved cardiac function to sham levels.

CONCLUSIONS:

PARP1 inhibition normalizes multiple inflammatory pathways that are altered after burn injury and improves cardiac dysfunction. PARP1 pathway inhibition may provide a novel methodology to normalize multiple burn injury–induced inflammatory pathways in the heart.

Burn injuries represent a significant cause of trauma globally. In the US alone, approximately 40,000 individuals are hospitalized for burns each year.¹ After the initial burn injury, numerous inflammatory mediators are released into circulation leading to altered function in multiple organ systems. Cardiac dysfunction has been found to occur as early as 8 hours after burn injury and to last up to 2 years.² This persistent dysfunction is characterized by decreased contractility, slowed isovolumic relaxation, and decreased diastolic compliance.^2–4

After burn injury, there is a decrease in plasma circulating volume secondary to increased capillary permeability leading to decreased cardiac output and a compensatory increase in heart rate and systemic vascular resistance. However, the continued decline in cardiac output after burn injury is not always recovered after repletion of plasma volume.^5–7 Other mechanisms such as direct myocardial damage,^8,9 myocardial ischemia,⁶ persistent β-adrenergic stimulation,¹⁰ accumulation of reactive oxygen species,^11,12 and circulating myocardial depressants^13,14 have been investigated. Recent studies and our previous work has found that mitochondrial oxidative stress and damage contributes to burn-induced myocardial dysfunction.^15–17

Our laboratory has found that the AMPK-SIRT1-PGC1α-NRF2-ARE pathway is an important mechanism in the defense against mitochondrial oxidative stress and mitochondrial biogenesis.¹⁸ PGC1α is a transcription regulator involved in energy metabolism modulated by the proteins AMPK and SIRT1. NRF2 is a transcription factor bound to ARE that performs an adaptive response against oxidative stress to prevent uncontrolled inflammation. Both SIRT1 and PGC1α activate NRF2, linking the two pathways. Burn injury results in the decreased expression of AMPK, SIRT1, and NRF2-ARE.¹⁸ This results in increased oxidative stress, decreased mitochondrial respiration, and ultimately mitochondrial DNA damage. SIRT1 deficiency causes acetylation of cardiac proteins which, in turn, leads to cardiac dysfunction.¹⁹

Poly (ADP-ribose) polymerase inhibitor 1 (PARP1) is a protein primarily involved in repairing DNA single-stranded breaks caused by reactive oxygen species. This occurs by catalyzing the transfer of ADP-ribose from nicotinamide adenine dinucleotide to other proteins.²⁰ In states of severe DNA damage, PARP1 alternatively promotes inflammatory processes, oxidative stress, and apoptotic pathways.²⁰ Downstream, these effects decrease cardiac function by altering calcium handling, increasing inducible nitric oxide activation, reactive oxygen species production, and activating apoptotic pathways.²⁰ PARP1 inhibitors—such as N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide (PJ34)—have been found to be effective in reducing oxidative stress and cytokine expression in pathologic states of the heart by reducing inflammatory activity, nicotinamide adenine dinucleotide consumption, and abrogating nuclear factor κ-light-chain enhancer of activated B cells (NF-κB)–dependent cytokine expression after Toll-like receptor (TLR) activation.^20,21 Improvements in cardiac function have been found with PARP1 inhibitor therapy in congestive heart failure, reperfusion injury, myocardial infarction, and atherosclerosis.^20,22 The role of PARP1 inhibitors in cardiac dysfunction after burn injury has not yet been studied.

The purpose of this study is to examine the role of PARP1 in the heart after burn injury and identify novel pathway targets for intervention by examining changes the expression of inflammation-related genes in response to PARP1 inhibition. We hypothesize that PARP1 inhibition will improve cardiac function after burn injury.

METHODS

Ethics statement

The study was permitted by the Ethics Committee of the University of Texas Medical Branch at Galveston (Galveston, TX) and by our Institutional Animal Care and Use Committee (IACUC, Protocol number: 1509059).

Animals, burn model process, and sample collection

Male Sprague-Dawley rats (8 weeks old, 250 to 300 g) purchased from Harlan Laboratories (Indianapolis, IN). were adapted for 1 week before the procedure in our animal housing by maintaining a 12 hour:12 hour light–dark cycle. The temperature was maintained at 25°C in the animal housing and all procedure rooms. Rats were allowed to freely take food and water during the study. A well-established model was used to induce a 60% total body surface area full-thickness burn in the burn and treatment groups.^{17,18,23–25} Rats were first given buprenorphine (0.05 mg/kg) by subcutaneous injection then anesthetized with isoflurane (3% to 5%). The rats were then placed within a protective mold that exposed 30% of the dorsum and 30% of the abdomen for a 60% total body surface area burn. The dorsum was submerged in near-boiling water (95° to 98°C) for 10 seconds. The abdomen was submerged for 2 seconds. Resuscitation with lactated Ringer’s solution (40 mL/kg) and oxygen at room temperature was performed immediately afterward. Rats in the treatment arm received N-(5,6-dihydro-6-oxo-2-phenanthridinyl)-2-acetamide hydrochloride (PJ34 HCl, 15 mg/kg weight) by intraperitoneal injection. Buprenorphine (0.05 mg/kg) was given at 6 hours post burn by subcutaneous injection. Rats were sacrificed at 24 hours post burn. Blood and tissue were collected into microtubes, flash frozen in liquid nitrogen, and stored at −80°C for later processing.

Echocardiography

Rats were sedated with 1.5% isoflurane/100% O₂ and placed supine on an electrical heating pad at 37°C. Heart rate and respiratory physiology were continuously monitored by echocardiography. After shaving the chest, warmed ultrasound gel was applied, and transthoracic echocardiography was performed using the Vevo 2100 ultrasound system (VisualSonics, Toronto, Canada) equipped with a high-frequency linear array transducer (MS250 13 to 24 MHz).²⁶ All measurements were obtained in triplicate. Data were analyzed using the Vevo 2100 standard measurement package.

RNA isolation

Total RNA was isolated from 10 g of heart tissue using RNeasy Mini Kit (Qiagen, Gaithersburg, MD, Catalog No. 74104) according to manufacturer instructions. The genomic DNA was digested with DNase I, RNase-free (Westlake, LA, Catalog No. M0303S). To judge the integrity and overall quality of isolated RNAs, 2 μg RNAs were run on 1% native agarose gels. A spectrophotometer (The DU 700 UV/Visible, Beckman Coulter, Pasadena, CA), NanoDrop (NanoDrop ND-1000 Spectrophotometer), and Agilent 2100 Bioanalyzer were applied to measure RNA quality by determination of absorbance at 260 and 280 nm (OD260/280 ratio 2, 1 OD260 Unit = 40 g/mL RNA), and the ratio of 28 S to 18 S rRNA (≥1.7), separately.

First-strand complementary DNA synthesis

Two micrograms of RNA was used to make complementary DNA (cDNA) using a SuperScript III Reverse Transcriptase (Invitrogen, Waltham, MA) according to manufacturer’s instructions. The cDNA synthesis mixture (13 μL) containing 2 μg of total RNA, 1 μL of 500 ng oligo(dT)18, 1 μL of 10 mmol/L dNTP mix, and sterile RNase-free water was heated to 65°C for 5 minutes to denature the RNA, then incubated on ice for 1 minute. A total of 4 μL of 5 first-strand buffer, 1 μL of 0.1 mol/L dithiothreitol, 1 μL of RNase inhibitor, and 1 μL of SuperScript III reverse transcriptase were added to the reaction tube after centrifugation. This was then mixed by pipetting gently and incubating at 42°C for 60 minutes. The reaction was inactivated by heating at 95°C for 15 minutes and diluted to a final volume of 102 μL with double-distilled water to be stored at −80°C. The cDNA was used as a template to run real-time PCR array and qPCR using iCycler Thermal Cycler (Bio-Rad, Hercules, CA) with SYBR-Green Supermix (SYBR Green Real-Time PCR Master Mixes, ThermoFisher, Sugar Land, TX, Catalog No. 4309155).

Real-time RT2 Profiler PCR Array (QIAGEN, Catalog No. PARN-018Z)

The high-quality cDNAs were used on the real-time RT2 Profiler PCR Array in combination with RT2 SYBR Green qPCR Mastermix (Catalog No. 330529). The PCR array was used to identify and pool the differentially expressed gene (DEG) profiles that were related to TLR pathways and profile gene expression, according to the instructions from the manufacturer. RNA from 2 randomly selected rats in each group was converted to cDNA and then amplified by PCR. The means were used for data analysis. The PCR array was performed in triplicate. The delta threshold cycle (Ct) was calculated by subtracting Ct for the average of 6 housekeeping genes, including β-actin (H1), β-2-macroglobulin (H2), GAPDH (H3), hypoxanthine phosphoribosyl transferase 1 (H4), ribosomal protein lateral stalk subunit P0 (H5), and (R)-2-hydroxyglutaryl-CoA dehydratase–activating ATPase (H6) from Ct for genes of interest. The rat profile PCR array profiles the expression of 84 key genes associated with the TLR-signaling pathway as well as related pathway-activity signature genes. In brief, a reaction mixture was prepared by successively adding 550 μL of 2RT2 SYBR Green qPCR Mastermix, 102 μL of the diluted first-strand cDNA synthesis reaction, and 448 μL of double-distilled water. The mixture was then administrated to the PCR array and PCR performed on a Bio-Rad iCycler at 95°C for 10 minutes, 40 cycles of 95°C for 15 seconds, and 60°C for 1 minute.

Analysis of real-time RT2 Profiler PCR array

Online Analysis Software (https://geneglobe.qiagen.com/us/analyze [accessed on June 6, 2021]) was used to analyze the PCR array data. All plates had 3 positive PCR controls and 3 reverse-transcription controls. Calculations of contamination with rat genomic DNA, according to the manufacturer’s instructions, showed the presence of genomic DNA in an acceptable range that would not influence experiment performance. Values of the cycle threshold (Ct) obtained in quantification were used for calculations of fold changes in messenger RNA (mRNA) abundance, according to the 2^ΔΔCt method.²⁷ The average was chosen from the group of 6 housekeeping genes as the varying reference gene.²⁷ Changes in mRNA level for evaluated genes were assessed in all groups in relation to the control group of animals, with mRNA abundance set up arbitrarily as 1.

String-DB analysis of differentially expressed genes

All derived protein names were entered into the String-DB website (https://string-db.org/ [accessed on June 6, 2021]) in Multiple Proteins by Names/Identifiers under SEARCH. Data were obtained by following the website’s instructions. The results from String-DB included more than 10 parameters, such as Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, function, compartments, component, InterPro, clusters, etc. These are summarized in Supplemental Digital Content 1 to 3 (http://links.lww.com/JACS/A195).

Signaling pathway mapping for differentially expressed genes

The KEGG Mapper (https://www.genome.jp/dbget-bin/www_bget?rno [accessed on March 3, 2021]) was used to analyze DEGs. The ratio mean of gene expression was calculated in the control and burn groups, which was visualized in a color code (reds for upregulated DEGs and blues for downregulated DEGs) in the top 4 pathways, including TLR-signaling pathway (KEGG rno04620), interleukin (IL)-17 signaling pathway (KEGG rno04657), RIG-I-like receptor signaling pathway (KEGG rno04622), and NF-κB signaling pathway (KEGG rno04064).

Validation of identified genes by quantitative polymerase chain reaction

Total RNA isolation and cDNA production were similar to the methods described in the RNA Isolation and First Strand cDNA Synthesis sections. A total of 25 L of reaction mixture was prepared by successively adding 12.5 μL of 2 RT2 SYBR Green qPCR Mastermix, 2 μL of the diluted first-strand cDNA synthesis reaction, 0.6 μL of primer mixture, and 9.9 μL of double-distilled water. The mixture was then administrated to the PCR tube and PCR was performed on a Bio-Rad iCycler at 95°C for 10 minutes, 40 cycles at 95°C for 15 seconds, and at 60°C for 1 minute. The results were analyzed using the ΔΔCt method. The primers we designed were based on identified DEGs using National Center for Biotechnology Information’s Primer-BLAST. The primers for qPCR tests used in the study are shown in Table 1. The Ct values of target genes were normalized to the housekeeping gene GAPDH and calculated as fold change.

Table 1.

Oligonucleotides Used for Validation of the Identified PJ34 Recovered Burn-Induced Toll-Like Receptor-Related Differentially Expressed Genes

Gene	5’-Forward-3’	5’-Reverse-3’	Amplicon size, bp	Accession no.
Casp8	AATAAAGACAACCCGAGGAACA	ATGCCACAGCCCATCTTCACACTA	101	NM_022277.1
Ccl2	GTCTCAGCCAGATGCAGTTAAT	CTGCTGGTGATTCTCTTGTAGTT	105	NM_031530.1
Cd14	GTCCAGCGCTTTTCTATGCG	TTCCGTGTCCACACGCTTTA	94	AF087943.1
Cd86	CCGAGTGAGCTCGTAGTATTT	GGTACTTGGCATTCACGTTATC	100	NM_020081.1
Cd180	CACCTTCGTGAGCTTGACCT	GGCCCAGGGGTTCATTGTAG	121	NM_001106405.2
Cebpb	CTTGATGCAATCCGGATCAAAC	CCCGCAGGAACATCTTTAAGT	113	NM_024125.5
Clec4e	GGTTTCGGGACTTGATGCAAT	CAGTTACACGTGTGTTGCGT	81	NM_001301720.1
Csf2	CCCCGGAAACTGACTGTGAA	CCTCATTTCTGGACCGGCTT	115	NM_053852.1
Csf3	AGGCCAGGAACACTGAACTG	TCAGGGAAGCCTTGGGGATA	111	NM_017104.2
Cxcl10	GGTCCCATCAGCTGCTACTC	GGGAAGATTCGGACCAGCTC	115	NM_139089.2
Fos	GAGCAGTCAGAGAAGGCAGG	TAGGTCTACGGGAACCCCTC	96	NM_022197.2
Gapdh	ACTCCCATTCTTCCACCTTTG	CCCTGTTGCTGTAGCCATATT	105	NM_017008.4
Hmgb1	TCGGCCTTCTTCTTGTTCTG	GTTGTTCCACATCTCTCCTAGTT	108	NM_012963.2
Hspa1a	CGTTTGACACTCTGTTGCTTTC	ACGGCCAGGCAAGATTATAC	89	NM_031971.2
Ifna1	AGAGAGAGAGAGAGAGAGAGAGA	GTGTGATTCCACATTTGCAGTAG	110	XM_001076062.4
Ifnb1	ACTTGGGTGACATCCACGAC	GGACCACCATCCAGGCATAG	106	NM_019127.2
IL10	AGTGGAGCAGGTGAAGAATG	GAGTGTCACGTAGGCTTCTATG	109	NM_012854.2
IL1a	GAGGCCATAGCCCATGATTTA	CAGATGGGCTAACTAAGGGATAAG	105	NM_017019.2
IL1b	GAGGCCATAGCCCATGATTTA	CTCCTGCTTGACGATCCTTATC	105	NM_017019.2
IL1r1	AATCGAGGCAGTGAACTGGG	GACCCCCAAAGTGTCCTCAG	121	NM_001412594.1
IL2	GCAGGCCACAGAATTGAAAC	CCAGCGTCTTCCAAGTGAA	108	NM_053836.1
IL6	GAAGTTAGAGTCACAGAAGGAGTG	GTTTGCCGAGTAGACCTCATAG	105	M26744.1
Kcnh8	GAAGACAGAGCCAAAGGAAGA	CAGGTGTCCTGAGATGTGATAAA	96	NM_145095.2
Lta	GCGTCAGTTACCACAGAACA	GCCTCGGGCTTTCTTCTAAA	100	NM_080769.2
Ly98	CCGAAGCGCAAGGAAATTG	TGTGATGGCCCTTAGGAAATAG	133	NM_001024279.1
Map2k3	CTGGGGCAGCACTTACCTAC	TTGGGACAGCTAGTTGCGAG	104	NM_001100674.1
Map3k7	AATCCACTGTCTCGCTGCAT	GCACGTACAGCTACGTTCAC	98	NM_001107920.2
Mapk8	ATGTCTGCAATACCCAGGCAC	AAGGGGCTTCTTCCTACAACC	112	NM_053829.2
Nfkb1	GGTTACGGGAGATGTGAAGATG	GTGGATGATGGCTAAGTGTAGG	100	NM_001276711.1
Nr2c2	CTGATAGCCACTCCCACATTT	GAACTGTACCATCCTCACGTATC	118	NM_017323.2
Peli1	TACCTACGTTCAACCGGTGC	CACTGGTCAGGAGCCTTGAG	131	NM_001100565.1
Pglyrp1	GCACGTCCAGCATGTTGTTT	CACGTAGCGGACTGGTTTCT	146	NM_053373.1
Ptgs2	GGCCATGGAGTGGACTTAAA	GTCTTTGACTGTGGGAGGATAC	132	NM_017232.3
Ripk2	CCTTTGCCTCCTGTCTTTCT	CCTCTACCCAACCAGCATATT	100	NM_001191865.1
Tbk1	GAAGAAGCTGAAGGAGGAGATG	CAGACTCCCGAAGAAGCTAAAG	137	NM_001106786.1
Ticam2	CTGCAAACCATCAATGCCTTAG	CACTTCGTGTCTCCTTCCATAC	118	NM_001108890.1
Tlr1	AGGTACAGTGATGCGGGAAG	GACCTTGGGAGGTAAGCACC	111	NM_001172120.2
Tlr2	CCAAGAGGAAGCCCAAGAAA	CATGAGGTTCTCCACCCAATAG	98	NM_198769.2
Tlr5	CCATCCAGGGAAGCCACATT	GGTCCAGTTGTAGCACCGAA	138	NM_001145828.1
Tlr6	CCTGGATGTGGAGCCTTCAG	ATCGGAATGGGTTGTTCCCC	133	NM_207604.1
Tlr7	AACCTTTCCCAGAGCATACAG	AGCCTCTGATGGGACAAATAAA	110	NM_001097582.1
Tnf	ACCTTATCTACTCCCAGGTTCT	GCAGATGGGCTGTACCTTATC	108	X66539.1
Tnfrsf1a	GGTGCTGACCCTCTCCTCTA	ATACAGCATCGCAGGGTCTG	126	NM_013091.2
Tollip	CCCTCCTCTGATGTTGTATGTG	GGTAAGCAAGACAGGAGGTTAG	108	NM_001109668.1
Tradd	CTGGACCTTCTGAAACCTAGATG	CTTAGGTCCCACGATGAGAATG	106	NM_001100480.1
Ube2n	GCCCAGGCATCTTCATCTTT	CTAGCCTCTGCTCATGCTTTAC	89	NM_053928.2
Ube2v1	GAAGAAGGCCAGAAAGGAGTAG	TTCGTGGAGGCCCAATTATC	109	NM_001110345.2

PJ34, N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide.

Protein–protein interaction networks

Protein–protein interaction networks of differentially expressed protein were constructed using STRING 9.1 (http://string-db.org/ [accessed on June 6, 2021]). The name of each individual protein was given as a query to the STRING database, and the corresponding protein–protein interaction information was retrieved by enabling different prediction methods. The networks were made with a confidence cutoff of 0.45.

Big data analysis of high-mobility box group

Free public resources, including tools and databases, were enlisted to obtain data and draw figures. To guarantee the objectivity of the results, all analyses throughout the study were employed according to the default settings of the database or tool, unless otherwise stated. When conducting analysis using default parameters, the operating tools or databases displayed needed data instantly for free download. Figures were made using GraphPad Prism 9.4.0 unless otherwise specified.

The online tool DISEASES²⁸ was used to analyze the associations of high-mobility box group protein 1 (HMGB1) with diseases. Only the human gene HMGB1 was selected, and the Z-score data were downloaded by selecting “text mining.” All genes of competitive endogenous RNA (ceRNA) of HMGB1 were obtained from ENCORI for functional analysis.²⁹ For ceRNA, the parameters are as follows: miRNA number ≥2, p value, and false discovery rate both ≤0.01. GTEx data in the GEPIA2 database were employed to extract 200 coexpressed (coRNA) genes of HMGB1 in the heart by conducting the Pearson’s correlation coefficient analysis.²⁹ Gene function comments including gene ontology (GO) terms for biological processes, cellular component, molecular functions, and KEGG pathways for coRNA, and ceRNA datasets were achieved using well-known tool DAVID³⁰ and illustrated using ImageGP. eVenn³¹ was used to find intersection genes. Functional information about all intersecting genes was downloaded from InnatedDB.³² Only human genes were selected. GO terms were extracted directly. KEGG analysis for HMGB1 and 6 intersection genes between coRNA and ceRNA genes was conducted using KOBAS.³³ The statistical method used was “hypergeometric test/Fisher’s exact test,” and the false discovery rate correction method was “Benjamini and Hochberg.” Tissue and cell profiling data of intersecting genes were directly downloaded from HPA, an online database set that includes massive amounts of human RNA and protein expression information.²⁸

The functional processes of intersecting genes were downloaded from National Center for Biotechnology Information and drawn using ImageGP. Gene-chemical interactions were queried using gene symbols from the CTD database,³⁴ a repository embodying the effects of environmental chemicals on human gene expression. All compounds that hit the targets—HMGB1, COX5A, NDUFB8, SOD1, and MDH2—were downloaded. The network of gene-chemical interplay was visualized using the software Cytoscape.³⁵ The targeted network is automatically visualized after the text files with genes and chemicals were directly input into the software.

A drug-matching analysis was performed using DrugBank to dissect potential effects for all chemicals. Queried chemicals in Drugbank were downloaded, and disease classification analysis was performed manually one by one.³⁶

Statistical analysis

Data are presented as the mean + SD or a number (%). Statistical analysis was performed using GraphPad 9.4 (GraphPad Holdings, LLC, San Diego, CA). A two-tailed Student’s t-test was applied for this analysis. Before the Student’s t-test, the Fisher–Snedecor test was used to verify equality of variances. Unequal variances were not found. Therefore, a t-test with Welch correction was not used in the project. Significance is expressed with * (24 hours post burn [hpb] rats vs sham rats) (*p < 0.05, **p < 0.01, ***p < 0.001) and # (24 hpb rats vs treatment with PJ34 rats) (#p < 0.05, ##p < 0.01, ###p < 0.001). For PCR array data, the data analysis was conducted using the ΔΔCt module on the Qiagen GeneGlobe Data Analysis Center portal. The efficiency of all the primers used in the kits was shown to be greater than 90%. The real-time PCR arrays contained control wells/samples for the determination and/or verification of rat genomic contamination, reverse transcription control, and positive PCR controls. At least 6 reference genes were used for data normalization. In these analyses, genes with a greater than 1.5-fold change in expression at a value of p < 0.05 between burn- and sham-group rats were defined as differentially expressed and selected for inclusion in comparative analyses. Gene Ontology (GO) analysis was conducted using String-DB resources (http://string-db.org [accessed on June 6, 2021]).

RESULTS

Effect of PJ34 on burn-induced cardiac heart dysfunction

Echocardiography demonstrated that left ventricular (LV) systolic function was significantly decreased after burn injury. Ejection fraction (Fig. 1A), fraction shortening (Fig. 1B), and cardiac output (Fig. 1C), LV mass (Fig. 1D), and interventricular septal size (Fig. 1E) decreased 21%, 21%, 25%, and 10%, respectively, after burn injury. Heart rate (Fig. 1E) and LV mass (Fig. 1F) increased after burn injury by 33% and 28%, respectively. PJ34 administration preserved ejection fraction, fraction shortening, cardiac output, interventricular septum, and heart rate after burn injury. LV mass was not preserved after PJ34 administration. The data in Figure 1 suggest that heart dysfunction after burn injury is attenuated through PARP1-NF-κB inhibition.

Figure 1.

Echocardiography data of sham vs burn vs treatment (PJ34) groups. Decreased left ventricular function after burn injury evidenced by a significant decrease in ejection fraction (C), fractional shortening (D), and cardiac output (E). Heart rate (A) increased significantly after burn injury. Interventricular septal size (B) did not significantly decrease after burn injury. PJ34 recovered all measures except interventricular septal size after burn injury. *Significant change between sham and burn group (p < 0.05). #Significant change between burn and treatment (PJ34) group (p < 0.05). BPM, beats per minute; hpb, hours post burn; PJ34, N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide.

Polymerase chain reaction array in cardiomyocytes

High-purity RNA

Highly qualified/purified RNA is necessary for the transcriptome/microarray/real-time PCR array profile. We used different methods to test our RNA quality, including agarose electrophoresis (data not shown), a spectrophotometer, NanoDrop, and an Agilent 2100 Bioanalyzer. Supplemental Digital Content 1 at http://links.lww.com/JACS/A195 shows the qualities and purities of isolated sham RNA (Supplemental Digital Content 1A), 24 hours postburn RNA (Supplemental Digital Content 1B), and 24 hours postburn with PJ34 administration (Supplemental Digital Content 1C). Quality/purity were satisfactory to move forward, as evidenced by ratios of 28S:18S RNA in sham, control, and treatment groups of approximately 2 (Supplemental Digital Content 1D).

Polymerase chain reaction array

The TLR RT2 Profiler PCR Array had 84 TLR-signaling pathway-related genes (Supplemental Digital Content 3A, http://links.lww.com/JACS/A195, wells A1 to G12) and 6 housekeeping genes (Supplemental Digital Content 3A H1 to 6), 3 positive genes (Supplemental Digital Content 3A H7 to 9) and 3 negative genes (Supplemental Digital Content 3A H10 to 12). The Qiagen analysis center provided the heatmap of the real-time RT2 Profiler PCR Array and the fold changes in burn injury with/without PJ34 treatment compared with the sham control in each of the 96 wells and in burn injury with PJ34 compared with burn injury without PJ34 (Fig. 2A). The fold-change calculations were conducted using the online software of the Qiagen Analysis Center, with pairwise comparison of 3 groups.

Figure 2.

Expression of toll-like receptor (TLR)-related genes in burn-injured rats with/without PJ34. Analysis performed with Rat TLR RT2 Profiler PCR Array. (A) The heatmap displays hierarchical clustering of the entire dataset of expressed genes, displaying coregulated genes across sham control and burn experimental groups with/without PJ34. Clustering analysis showed that genes upregulated in burn-injury rats belong to apoptosis modulation, IL-1 signaling, TLR signaling, RIG-I-like receptor signaling, and IL-17 signaling pathways (n = 4 rats analyzed for each experimental condition).

(B) The Venn diagram and list of TLR-related genes modulated after burn injury rats are presented. Genes with fold upregulation values greater than 1.5 and less than 1.5 compared with controls are shown. hpb, hours post burn; PJ34, N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide.

Cardiomyocytes in 24 hpb rats demonstrated 39 upregulated (39 genes, fold change 2.0) and 19 downregulated (fold change 2.0)²⁷ TLR pathway–related genes (Fig. 2B). PJ34 treatment significantly restored all burn injury–induced upregulated genes, and 10 of 19 downregulated genes (Fig. 2B). PJ34 significantly recovered all burn-induced upregulated DEGs (Fig. 2B, S3; Table 2) compared with the 24-hour postburn group. Ten of 19 burn-induced downregulated DEGs were restored to normal sham expression levels (Fig. 2B, S3; Table 2).

Table 2.

Gene Ontology for PJ34 Recovered Identified Burn Injury-Induced Differentially Expressed Genes

Term ID	Term description	Observed gene count	Matching proteins in network
GO:0070498	Interleukin-1-mediated signaling pathway	4	Traf6,Il1r1,Irak2,Tollip
GO:0019221	Cytokine-mediated signaling pathway	5	Traf6,Il1r1,Irak2,Tnfrsf1a,Tollip
GO:0071310	Cellular response to organic substance	10	Traf6,Il12a,Nr2c2,Il1r1,Cd14,Irak2,Ppara,Tnfrsf1a,Actb,Tollip
GO:0071396	Cellular response to lipid	7	Traf6,Il12a,Nr2c2,Cd14,Irak2,Ppara,Tnfrsf1a
GO:0001819	Positive regulation of cytokine production	5	Traf6,Il12a,Il1r1,Cd14,Tnfrsf1a
GO:0002711	Positive regulation of T cell-mediated immunity	3	Traf6,Il12a,Il1r1
GO:0006950	Response to stress	10	Traf6,Il12a,Nr2c2,Il1r1,Cd14,Irak2,Ppara,Tnfrsf1a,Actb,Tollip
GO:0032729	Positive regulation of interferon-γ production	3	Il12a,Il1r1,Cd14
GO:0006954	Inflammatory response	5	Il1r1,Cd14,Irak2,Tnfrsf1a,Tollip
GO:0032496	Response to lipopolysaccharide	5	Traf6,Il12a,Cd14,Irak2,Tnfrsf1a
GO:0034097	Response to cytokine	6	Traf6,Il1r1,Cd14,Irak2,Tnfrsf1a,Tollip
GO:1901701	Cellular response to oxygen-containing compound	7	Traf6,Il12a,Il1r1,Cd14,Irak2,Tnfrsf1a,Actb
GO:0071222	Cellular response to lipopolysaccharide	4	Traf6,Il12a,Cd14,Irak2
GO:0006952	Defense response	6	Il12a,Il1r1,Cd14,Irak2,Tnfrsf1a,Tollip
GO:0007166	Cell surface receptor signaling pathway	7	Traf6,Il12a,Il1r1,Cd14,Irak2,Tnfrsf1a,Tollip
GO:1901700	Response to oxygen-containing compound	8	Traf6,Il12a,Il1r1,Cd14,Irak2,Ppara,Tnfrsf1a,Actb
GO:0009605	Response to external stimulus	8	Traf6,Il12a,Cd14,Irak2,Ppara,Tnfrsf1a,Actb,Tollip
GO:0051707	Response to other organism	6	Traf6,Il12a,Cd14,Irak2,Tnfrsf1a,Tollip
GO:0010628	Positive regulation of gene expression	7	Traf6,Il12a,Nr2c2,Il1r1,Cd14,Ppara,Tnfrsf1a
GO:0007165	Signal transduction	9	Traf6,Il12a,Nr2c2,Il1r1,Cd14,Irak2,Ppara,Tnfrsf1a,Tollip
GO:0060255	Regulation of macromolecule metabolic process	10	Traf6,Il12a,Nr2c2,Il1r1,Cd14,Irak2,Ppara,Tnfrsf1a,Actb,Tollip
GO:0002726	Positive regulation of T-cell cytokine production	2	Traf6,Il1r1
GO:0042088	T-helper 1 type immune response	2	Traf6,Il12a
GO:0032101	Regulation of response to external stimulus	5	Traf6,Il12a,Il1r1,Ppara,Tnfrsf1a
GO:0010604	Positive regulation of macromolecule metabolic process	8	Traf6,Il12a,Nr2c2,Il1r1,Cd14,Ppara,Tnfrsf1a,Tollip

GO, gene ontology; PJ34, N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide.

Analysis of interaction and function network of differentially expressed genes

The burn injury–induced upregulated and PJ34 recovered DEGs were used for analysis. In detail, in Figure 3B, the colored nodes indicate first shell interactions; the edges represent protein–protein associations; blue indicates that the information is from curated databases and pink indicates that it is experimentally determined; green is gene neighborhood; dark blue represents gene co-occurrence; yellow represents information gathered from text mining; black represents co-expression; and light blue represents protein homology. Proteins with no interactions are not included in the figure for clarity. The functional interaction networks based on the STRING database and kmeans clustering method are shown in different colors. The red nodes correspond to proteins involved in the chemokine-related pathway, forming a multimeric SMAD3/SMAD4/JUN/FOS complex. These play a role in signal transduction; cell proliferation and differentiation; the generation of purine nucleotides through the purine salvage pathway; and regulation of the immune response. The brown nodes correspond to proteins involved in mediating the innate immune response, and recognizing damage-associated molecular patterns and pathogen-associated molecular patterns; these are multifunctional redox-sensitive proteins with various roles in different cellular compartments, in the protein quality-control system, in ensuring the correct folding of proteins and the refolding of misfolded proteins, in controlling the targeting of proteins for subsequent degradation, and in the clearance of cytotoxic polyglutamine (polyQ) protein aggregates. The dark gold rod node corresponds to proteins involved in CD180 binding. The green-yellow nodes correspond to proteins involved in innate immunity and function in the intracellular killing of bacteria. The 2 green nodes correspond to proteins acting as a growth factor for activated T and NK cells, enhancing the lytic activity of NK/lymphokine-activated killer cells, and stimulating the production of interferon (IFN)-γ by resting peripheral blood mononuclear cells. The green nodes correspond to proteins involved in regulating the expression of genes involved in immune and inflammatory responses, playing a significant role in adipogenesis, as well as in the gluconeogenic pathway. The blue nodes correspond to proteins involved in the important repression of nuclear receptor signaling pathways. The light sky-blue nodes correspond to proteins involved in: upstream protease of the activation cascade of caspases, which trigger intracellular signaling cascades, leading to transcriptional upregulation and mRNA stabilization; various processes such as neuronal proliferation, differentiation, migration, and programmed cell death; mediating the ubiquitination of free/unanchored polyubiquitin chains that leads to MAP3K7 activation and activation of NF-κB and JUN; dendritic cell maturation and/or activation; and the error-free DNA repair pathway, contributing to the survival of cells after DNA damage are shown (Figs. 3B and 4B and Supplemental Digital Content 2, http://links.lww.com/JACS/A195).

Figure 3.

Quantitative polymerase chain reaction validation of upregulated genes after burn injury. (A) Shows a heatmap of the upregulated genes in sham, 24 hours after burn, and treatment (PJ34) groups. (B) Represents the functional interaction network of the genes upregulated after burn injury. The changes in expression of the upregulated genes after burn injury and PJ34 treatment are graphed in C. Changes in expression corresponds to the relative changes seen in the heatmap (A). hpb, hours post burn; PJ34, N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide.

Figure 4.

Validation of burn-induced downregulated genes by quantitative polymerase chain reaction. Heatmap (A) of the downregulated toll-like receptor (TLR)-related differentially expressed genes in sham, 24 hours post burn, and treatment (PJ34) groups. (B) Represents the functional interaction network of the genes downregulated after burn injury. The changes in expression of the downregulated differentially expressed genes after burn injury and PJ34 treatment are graphed in C. Changes in expression correspond to the relative changes seen in the heatmap (A). hpb, hours post burn; PJ34, N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide.

Protein–protein interaction network and function network of identified PJ34 recovered upregulated genes after burn injury

Figure 3B and 3C represent the networks of the PJ34 recovered burn injury–upregulated DEGs (fold change >2.0). By analyzing the functional interaction network, the top 6 clusters were obtained (Fig. 3). The lime green nodes correspond to proteins involved in cell apoptosis/death regulation/activation or that have a role in the proinflammatory cytokines. They may also participate in T-cell effector function, and perhaps in T-cell development, signal transduction, cell proliferation and differentiation, regulation of the immune response, or lymphocyte and monocyte differentiation. The green nodes correspond to proteins acting as a pattern-recognition receptor of the innate immune system, recognizing damage-associated molecular patterns and pathogen-associated molecular patterns, and with various roles in different cellular compartments, replication, transcription, chromatin remodeling, V(D)J recombination, DNA repair, and genome stability. They also play a pivotal role in the protein quality-control system, ensuring the correct folding of proteins and the refolding of misfolded proteins, and controlling the targeting of proteins for subsequent degradation. The red nodes correspond to proteins involved in mediating the transcriptional activation of target genes; these play a role in the control of progress through the cell cycle; in differentiation; and in the error-free DNA repair pathway, with contributions to the survival of cells after DNA damage or protein ubiquitination. The yellow nodes correspond to proteins involved in potent proinflammatory cytokines, promoting Th17 differentiation of T cells, and synergizing with IL-12 to induce IFN-γ synthesis from T-helper 1 cells. These include various processes such as neuronal proliferation, differentiation, migration, and programmed cell death, stimulating the stress-activated protein kinase/c-Jun N-terminal kinase (SAP/JNK) signaling pathway. The blue nodes correspond to proteins involved in the potassium voltage-gated channel. The purple nodes correspond to proteins involved in immunity to Gram-positive bacteria by interfering with peptidoglycan biosynthesis.

PJ34 recovered identified downregulated genes after burn injury

Similar to the previous section, PJ34 recovered burn injury–induced downregulated DEGs were analyzed in the protein–protein interaction network (Fig. 4). A functional cluster analysis resulted in 3 pools. The red nodes correspond to the proteins involved in mediating the innate immune response; the recruitment of adapter molecules, triggering intracellular signaling cascades and leading to transcriptional upregulation and mRNA stabilization; connecting the ubiquitin pathway to autophagy by functioning as a ubiquitin-ATG8 family adapter and, thus, mediating autophagic clearance of ubiquitin conjugates; the clearance of cytotoxic polyQ protein aggregates; and mediating the ubiquitination of the free/unanchored polyubiquitin chain that leads to MAP3K7 activation and the activation of NF-κB and JUN. The light green nodes correspond to the proteins that act as a repressor or activator of transcription, an important repressor of nuclear receptor signaling pathways. The light blue nodes correspond to the proteins that are key regulators of the expression of inflammatory mediators such as IL-6, tumor necrosis factor (TNF)-α, and IFN-β; act as a coreceptor (along with the Toll-like receptor TLR 4 and MD-2) for the detection of bacterial lipopolysaccharide; and are involved in the survival and prognosis of B-cell chronic lymphocytic leukemia.

Analysis of enriched gene ontology for burn injury-induced differentially expressed genes recovered by PJ34

Enriched GO is presented as GO molecular functions, GO cellular components and the GO biological processes. Table 2 reveals the top 10 GO molecular functions, top 10 GO cellular components, and top 10 GO biologic processes based on the strengths.

Signaling pathway mapping for PJ34 recovered burn injury–induced differentially expressed genes

In this section, we explored the specific pathways involved in PJ34 recovery after burn injury. KEGG signaling network mapping (KEGG ruo) was used to compare the expression levels of the sham controls and burn injury groups with/without PJ34 (Fig. 5, red colors represented PJ34 recovered burn injury–induced upregulated DEGs and blue colors represented PJ34 recovered burn injury–induced downregulated DEGs). The signaling pathway mapping of PJ34 recovered burn injury–induced DEGs showed that the top 4 pathways included the TLR signaling pathway (Fig. 5B and Supplemental Digital Content 3A at http://links.lww.com/JACS/A195, KEGG rno04620) involved in the 25 DEGs (Tnf, Cd80, Tlr6, Cd86, Cxcl10, Tlr5, Tlr7, Traf6, Il1b, Map3k7, Ifnb1, Map2k3, Tbk1, Fos, Il12a, Tlr2, Il6, Casp8, Cd14, NFκB1, Tlr3, Tollip, Ifna1, Mapk8, Tlr1); the IL-17 signaling pathway (Fig. 5B and Supplemental Digital Content 3B, KEGG rno04657), involved in 18 DEGs (Tnf, Ptgs2, Cxcl10, Traf6, Il1b, Map3k7, Tbk1, Ccl2, Ifng, Fos, Csf3, Il6, Casp8, Tradd, NFκB1, Csf2, Mapk8, Cebpb); RIG-I-like receptor signaling pathway (Fig. 5B and Supplemental Digital Content 3C, KEGG rno04622), involved in 12 DEGs (Tnf, Cxcl10, Traf6, Map3k7, Ifnb1, Tbk1, Il12a, Casp8, Tradd, NFκB1, Ifna1, Mapk8); and TNF signaling pathway (Fig. 5B and Supplemental Digital Content 3D, KEGG rno04668), involved in the 18 DEGs (Tnf, Lta, Ptgs2, Cxcl10, Il1b, Map3k7, Ifnb1, Map2k3, Ccl2, Fos, Il6, Casp8, Tradd, NFκB1, Csf2, Tnfrsf1a, Mapk8, Cebpb) based on the strength value. The top 5 molecular functions included type I interferon signaling, JUN kinase activity, IL-1 receptor binding, CXC chemokine binding, and TNF receptor binding (Fig. 5A).

Figure 5.

Bar graphs (A) and (B) represent the molecular function and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of upregulated differentially expressed genes recovered by PJ34. Bar graphs (C) and (D) represent the molecular function and KEGG pathway analysis of downregulated differentially expressed genes recovered by PJ34. PJ34, N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide.

Validation of burn injury–induced Toll-like receptor–related differentially expressed genes using quantitative polymerase chain reaction

One disadvantage of this method is that many identified DEGs/proteins may be false positives and require confirmation using qPCR for DEGs or Western blotting/ELISA/IHC for differentially expressed proteins. Therefore, we selected real-time qPCR to validate the PJ34 recovered identified burn injury–induced DEGs. The primers for the qPCR tests are listed in Table 1. Our qPCR data show that all selected identified burn injury–induced DEGs, whether upregulated (Figure 3C) or downregulated (Fig. 4C), had the same trend as our PCR array results.

Characterizing protein–protein interactions via STRING analysis

String (version 11.5) was employed to postulate the protein–protein interactions and networks. Eighteen upregulated DEGs (Fig. 6A) and 5 downregulated DEGs were analyzed (Fig. 6B). The predicted protein–protein interactions and networks for genes upregulated after burn injury recovered by PJ34 with a more than 30-fold change are listed in Table 3. The predicted pathways/functions are additionally listed. Common protein processes affected included cardiac remodeling, cytokine signaling, and apoptosis. Table 4 lists the predicted protein–protein interactions and networks for genes downregulated after burn injury recovered by PJ34 with a more than 30-fold change. Common predicted processes included cytokine signaling and cardiac remodeling.

Figure 6.

STRING analysis of up-regulated differentially expressed genes (DEG) after burn injury. STRING analysis revealed protein interaction networks for 18 of the upregulated DEGs (A) and 5 downregulated differentially expressed genes (B) (fold change ≥5). Proteins are presented as nodes. Lines represent predicted interactions. Line thickness represents the confidence level (0.7 to 0.9). Red indicates the presence of fusion evidence. Green indicates neighborhood evidence. Blue indicates co-occurrence evidence. Purple indicates experimental evidence. Yellow indicates text mining evidence. Light blue indicates database evidence. Black indicates coexpression evidence.

Table 3.

Predicted Protein-Protein Interactions and Functions/Pathways of Upregulated Genes with More Than 30-Fold Change after PJ34 Administration

Protein	Predicted protein-protein interaction	Predicted function/pathway
Ccl2	CCr1, Ccr3, Ccr4, and Ccr5	Cardiac remodeling
Cd86	Cdc42, Cd25 and Cd84 proteins	Cardiomyocyte proliferation and immune cell activation
Cebpb	Smad4, Crebbp, Ep300, Myb, Esr1	Transforming growth factor-β signaling, apoptosis, and cardiac hypertrophy
Hmgb1	Ddx5, Becn1, Thbp	Multiple membrane trafficking pathways, cytokine signaling, apoptosis
IL10	CD4, Ifnγ, IL10, IL10a, IL10b, IL4	Anti-inflammatory pathways and regeneration
IL1a	IL6, Myd88, Rela, TNF	Cytokine signaling, apoptosis, altered calcium handling, and increased iNOS activity
IL6	IL6r, IL6st, IL1b	Cytokine signaling
Ticam2	Ptpn4, TLR9	Interactions at junctions between the membrane and cytoskeleton
Tlr2	Hspa4, ENSRN, Clec7a, Ly96, Tollip, Cd14, Ggn	Cytokine signaling and cardiac remodeling
Tlr7	Ctsk, Ctsl, Ctsb, Lgmn	Cardiac remodeling and clearance of apoptotic cells
LTB	LTA, Tnfrs14, Tnfrs1a, Tnfrs1b, Ltbr	Cytokine signaling, alteration in calcium handling, and increased iNOS activity

Common predicted processes included cardiac remodeling, cytokine signaling, and apoptosis.

iNOS, inducible nitric oxide; PJ34, N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide.

Table 4.

Predicted Protein-Protein Interactions and Functions/Pathways of Downregulated Genes with More than 30-Fold Change after PJ34 Administration.

Protein	Predicted protein-protein interaction	Predicted function/pathway
IL1r1	Traf6, Irak1, IL1b, IL1a, IL1rn, IL1rap	Downstream activation of NF-κB, proinflammatory cytokine production
Nr2c2	Jazf1	Cardiomyocyte development, cell homeostasis
Ripk2	Birc3, Ngfr, Nod1 and Nod2	Cardiomyocyte calcium handling, cardiac hypertrophy
Tbk1	Ikbke, Ikekg, Irf3, Mavs, Optn, Tank, Tbk1, Tmem173, Traf3	Mitochondrial damage sensing, inflammatory signaling via NF-κB
Tollip	Irak1, Tlr2, Tom1 proteins	Inflammatory signaling

Common predicted processes included cytokine signaling and cardiac remodeling.

NF-κB, nuclear factor κ-light-chain enhancer of activated B cells; PJ34, N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide.

DISCUSSION

In this study, we demonstrated that PARP1 inhibition with PJ34 normalized cardiac function after burn injury in a rat model. Heart rate, ejection fraction, fractional shortening, and cardiac output returned to or exceeded preburn state function levels after administration of PJ34 (Fig. 1). Previous studies on the effects of PARP1 inhibitors in cardiac pathologies involving acute inflammation (reperfusion injury, myocardial infarction) have yielded similar results.^37,38 We additionally identified 47 upregulated and 19 downregulated DEGs after burn injury (Fig. 2). PARP1 inhibition after burn injury reversed 36 upregulated and 10 downregulated DEGs involved in cardiac function, inflammation, and fibrosis. Twelve of the 36 upregulated DEGs returned to sham expression levels (CD80, CEBPB, CXCL10, IL2, HSPA1A, IFNA, IL6, LTA, MAPK8, PGLYRP1, SARM1, and TLR5; Fig. 3). The predicted protein–protein interactions involving upregulated DEGs recovered by PJ34 involved cardiac remodeling, cytokine signaling, and apoptosis. The predicted protein–protein interactions for downregulated DEGs recovered by PJ34 involved NF-κB signaling and cardiac homeostatic processes. Taken together, these results imply that PARP1 inhibition improves cardiac function after burn injury and modulates genes involved in inflammatory signaling.

The systemic inflammatory response after burn injury induces cytokine production and TLR expression in the heart.^39,40 We have previously identified that IL-1, IL-17, TLR, and apoptosis signaling increased significantly after burn injury in heart tissue.²⁷ During KEGG pathway analysis for upregulated DEGs, four of the top pathways modulated by PJ34 administration involved TLR, IL-17, RIG-I–like receptor, and NF-κB signaling (Figure 5B). Interestingly, the top 4 pathways affected by PJ34 administration in downregulated DEGs were involved in TLR, IL-17, RIG-I–like receptor, and NF-κB signaling as well. The significance of these inflammatory pathways in the mechanism of burn-induced cardiac dysfunction is further solidified by these results. PARP1 inhibition with PJ34 has demonstrated modulation of genes involved in these signaling pathways, indicating its role in modulating burn-induced cardiac dysfunction.

TLRs are an important class of pattern recognition receptors activated by damage-associated molecular patterns and pathogen-associated molecular patterns leading to activation of innate immune cells and release of cytokines and adhesion molecules.^41,42 Fibrosis and inflammation can be limited by altering TLR signaling. Circulating levels of decorin, a ligand for TLR2 and TLR4, were found to be a predictor for of hypertrophic scar formation at 2 weeks after burn injury.⁴³ TLRs have been used as potential targets in hypertrophic scarring but with increased risk of infection and inhibited healing.⁴⁴ HMGB1-TLR signaling contributes to tissue damage in ischemia-reperfusion injury outside the heart.⁴⁵ During burn wound progression, necrotic periburn tissue releases HMGB1, a chomatin bound protein, first into the cytosol then extracellularly.⁴⁶ HMGB1 interacts with TLR4 leading to cytokine production and inflammation.⁴⁷ In the heart, knockout of TLR4 has been found to protect against lipopolysaccharide--induced cardiac dysfunction⁴⁸ and decrease myocardial specific secretion of TNF-α, IL-1β, IL-6, and IL-10 after burn injury.⁴⁹ TLR signaling plays a role in cardiac damage and dysfunction after burn injury and warrants further study as a potential therapeutic target.

RIG-I–like receptors (RLRs) compose a class of pattern recognition receptors important in the detection of viral RNA and processed self RNA.^50,51 Once activated, RLRs trigger innate immunity and inflammation to control infection. There are a limited number of studies on the role of RLRs in sterile inflammation. Mutations in RNA posttranscriptional modification have been implicated in autoimmune disorders, promoting chronic inflammation through type I interferon pathways.^52,53 Embryonic kidney and colon stem cell cultures treated with radiation have shown increased numbers of small nuclear RNA.⁵⁴ Small nuclear RNA translocate into the cytosol and interact with RLRs to promote inflammation and cell death.⁵⁴ These studies indicate that RLRs play a role in sterile inflammation in response to cellular damage. Both our current study and previous work have found RLR signaling to be modulated in cardiomyocytes after burn.²⁷ Further study of the role of RLRs in burn injury and burn-induced cardiac dysfunction is warranted.

NF-κB is a family of transcription factors and central regulators of innate and adaptive immunity that has made it a central focus in postburn cardiac dysfunction.⁵⁵ The canonical NF-κB pathway is activated by a number of diverse stimuli through pattern recognition receptors, cytokine receptors, and TNF receptors.⁵⁵ Activation of NF-κB results in the expression of proinflammatory genes (ie IL-1β, IL-6, and TNFα), T cells, the inflammasome and apoptotic pathways.^55,56 Specifically in the heart, persistent NF-κB signaling has been implicated in cardiac remodeling,^57,58 proapoptotic signaling,⁵⁶ and promotion of local inflammation.⁵⁶

Finally, IL-17 is a proinflammatory cytokine that promotes granulocyte development, neutrophil recruitment, IL-6 release, and matrix metalloproteinase release.⁵⁹ These functions lend toward IL-17’s association with the atherosclerotic progression⁶⁰ and left ventricular remodeling.⁶¹ After burn injury, IL-17 has been shown to have local and distant organ effects. IL-17 concentration has been found to be increased in the burn wound and small intestines, possibly contributing to bacterial translocation and endotoxemia.^62–64 IL-17 activity in the small intestine after burn injury alludes to its possible activity at other tissue sites, including the heart. Given IL-17’s effects at distant sites and promotion of cardiac remodeling and dysfunction, this cytokine may be a potential therapeutic target in burn-induced cardiac dysfunction.

PARP1 functions downstream of most of the previously described pathways. Its primary function is repair of single-stranded DNA breaks caused by reactive oxygen species.⁶⁵ This is done by catalyzing the transfer of ADP-ribose from nicotinamide adenine dinucleotide to other proteins.⁶⁵ The alternative functions of PARP1 (illustrated in Fig. 7) involve the promotion of cytokine signaling and inflammatory gene transcription. First, PARP1 has been found to enhance NF-κB–dependent gene expression by promoting the transcription of inflammatory mediators.⁶⁵ Second, in response to DNA alkylating damage, PARP1 regulates the translocation of a chromatin bound protein—high mobility group box protein 1 (HMGB1)—from the nucleus to cytosol.⁶⁶ HMGB1 may then be released from the cell after necrosis. HMGB1 acts as an inflammatory signaling protein by acting at TLRs or promoting chemotaxis of immune cells.⁶⁷ Finally, PARP1 promotes the assembly of the signalosome complex (IKKy, PIASy, and ATM) through poly(ADP-ribose) synthesis. This signalosome complex in turn activates NF-κB.⁶⁸ The extent of cell injury and DNA damage drives the expression of PARP1. More severe levels of DNA damage result in increased PARP1 expression and a shift toward proinflammatory functions.⁶⁸ Improvements in cardiac function have been found with PARP1 inhibitor therapy in congestive heart failure, reperfusion injury, myocardial infarction, and atherosclerosis.^20,22 PARP1 inhibition may attenuate this shift to proinflammatory pathways in states of severe DNA damage such as after severe burn injury.

Figure 7.

Alternate functions of PARP1. In response to DNA damage, PARP1 promotes the formation of the signalosome (IKKy, PIASy, and ATM) which activates NF-κB downstream. PARP1 interacts with NF-κB to promote transcription of proinflammatory mediators. Finally, PARP1 releases HMGB1 bound to chromatin. HMGB1 undergoes translocation into the cytosol where it may be released extracellularly during cell damage. HMGB1 acts as an inflammatory mediator at extracellular receptor sites, promoting the inflammatory signaling cascade. Image created with BioRender. HMGB1, high-mobility group box protein 1; NF-κB, nuclear factor κ-light-chain enhancer of activated B cells; PARP1, poly (ADP-ribose) polymerase inhibitor 1.

The attenuation In proinflammatory gene expression may be seen in the PJ34 recovered upregulated DEGs molecular functions that included type I interferon receptor signaling (such as RLRs), JUN kinase activity (proapoptotic), IL-1 receptor binding (Fig. 5, A and B), CXC chemokine binding, and TNF receptor binding. IL-1–mediated signaling and TNF-mediated signaling functions were recovered by PJ34 in the downregulated DEGs (Fig. 5, C and D).

The pathway and protein interaction analysis of this study is based on database comparisons and predictions. Our findings only demonstrate the statistical probabilities of protein–protein interactions. Further in vitro and in vivo experiments will be needed for verification.

CONCLUSIONS

PARP1 inhibition is a promising anti-inflammatory–based approach to burn-induced cardiac dysfunction. We have shown that PARP1 inhibitor PJ34 improves cardiac function after burn injury and normalized multiple inflammatory pathways. TLR, RLR, NF-κB, and IL-17 signaling were identified as the top signaling pathways for upregulated and DEGs. The role of these pathways in burn-induced cardiac dysfunction has not been fully studied. Further study may elucidate the mechanism of PARP1 inhibition in burn-induced cardiac dysfunction and identify additional novel therapeutic targets.

CONTINUING MEDICAL EDUCATION CREDIT INFORMATION.

Accreditation:

The American College of Surgeons is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians.

AMA PRA Category 1 Credits^™:

The American College of Surgeons designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit^™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

Of the AMA PRA Category 1 Credits^™ listed above, a maximum of 1 credits meet the requirement for Self-Assessment.

Abbreviations and Acronyms

cDNA

Complementary DNA

ceRNA

Competitive endogenous RNA

coRNA

Coexpressed RNA

Ct

Threshold cycle

DEG

Differentially expressed gene

ELISA

enzyme-linked immunoassay

GAPDH

Glyceraldehyde 3-phosphate dehydrogenase

GO

Gene ontology

HMGB1

High mobility box group protein 1

IL-17

Interleukin 17

IHC

Immunohistochemistry

KEGG

Kyoto Encyclopedia of Genes and Genomes

mRNA

Messenger RNA

NF-κB

Nuclear factor κ-light-chain enhancer of activated B cells

PARP1

Poly (ADP-ribose) polymerase inhibitor 1

PCR

Polymerase chain reaction

PJ34 HCl

N-(5,6-Dihydro-6-oxo-2-phenanthridinyl)-2-acetamide hydrochloride

PolyQ

Polyglutamine

qPCR

Quantitative polymerase chain reaction

RLR

RIG-I–like receptors

RIG-I

Retinoic acid-inducible gene I

TLR

Toll-like receptor

Biographies

Discussion

DR DAVID HERNDON (Augusta, GA): I am going to allow Dr Efron to dissect the genetic assumptions and the genetic contributions of this, because that is more his area of expertise. I want to bring the clinical perspective that myocardial depression is a real burn problem. The use of PARP1 inhibition has been successful in some inflammatory conditions. Burn is a huge and terrible inflammatory condition. Hypertrophic scars, prolonged cardiac morbidity, cachexia, are probably caused by prolonged inflammation.

My question to Dr Radhakrishnan and his excellent group is that infection is the major cause of death in burns. Burn sepsis, burn wound infection, pneumonia. For PARP1 inhibition to be clinically employed and, indeed, many of the strategies of reducing inflammation that you also bring to mind in your excellent discussion and from your titillating results might be targets, may have downsides. What happens to infection in burns if you inhibit inflammation? A host response gone awry clearly occurs, but the timing of intervention and the nature of intervention is extremely complex.

I would like you to answer this one question. What is worse than burns? Infection or inflammation? Or how do you tame one and not allow the individual to die of the other?

What an excellent example of the best of science has been presented here, where you take a clinical problem, analyze it in the laboratory, and come up with so many potential targets for further investigation.

DR PHILIP A EFRON (Gainesville, FL): There actually will not be many genetic questions from me, because Dr Herndon’s question is already spot on regarding what is beneficial vs detrimental to treat such an inflammatory chronic condition.

The authors have identified a significant and clinically relevant issue to investigate, which is the acute and chronic burn-induced cardiac dysfunction in adult burn patients. This pathology has been associated with worse outcomes in the burn cohort and, in modifying this response, burn patient morbidity and mortality could be improved.

As already mentioned by Dr Herndon, the authors have specifically investigated PARP1, otherwise known as the polyadenosine diphosphate-ribose polymerase 1, which is a molecule that can be produced in response to host cellular DNA damage. The authors have targeted PARP1 with an inhibitor in a clinically relevant model of burn injury and demonstrated that this intervention can improve cardiac function acutely as well as induce a unique transcriptome in cardiac tissue that reveals effects on the pathways of cardiac function, inflammation, and fibrosis.

Again, as we examine these post-ICU syndromes that are becoming more common, this novel approach could allow the resolution of organ-specific pathology in burn injury, as well as represent a therapeutic for burn, which, like sepsis, has seen little progress during the past several decades to improve outcomes.

Although investigators often have to do the extremes of pathology in clinically relevant animal models to demonstrate a difference, 60% burn is not necessarily the average size in a clinical practice. Do you believe this therapeutic, or potential therapeutic, will work just as well in a burn >20%, after which you start to see that significant inflammatory response associated with poor outcomes?

A hallmark of burn injury is how long it takes the patients to return to some sort of baseline or homeostasis after the event. Some of Dr Herndon’s pioneering work shows that it can take more than a year for a burn patient to return to base line or what we call homeostasis,