Protocatechuic acid prevents isoproterenol‐induced heart failure in mice by downregulating kynurenine‐3‐monooxygenase

Abstract Protocatechuic acid (3,4‐dihydroxybenzoic acid) prevents oxidative stress, inflammation and cardiac hypertrophy. This study aimed to investigate the therapeutic effects of protocatechuic acid in an isoproterenol‐induced heart failure mouse model and to identify the underlying mechanisms. To establish the heart failure model, C57BL/6NTac mice were given high‐dose isoproterenol (80 mg/kg body weight) for 14 days. Echocardiography revealed that protocatechuic acid reversed the isoproterenol‐induced downregulation of fractional shortening and ejection fraction. Protocatechuic acid attenuated cardiac hypertrophy as evidenced by the decreased heart‐weight‐to‐body‐weight ratio and the expression of Nppb. RNA sequencing analysis identified kynurenine‐3‐monooxygenase (Kmo) as a potential target of protocatechuic acid. Protocatechuic acid treatment or transfection with short‐interfering RNA against Kmo ameliorated transforming growth factor β1–induced upregulation of Kmo, Col1a1, Col1a2 and Fn1 in vivo or in neonatal rat cardiac fibroblasts. Kmo knockdown attenuated the isoproterenol‐induced increase in cardiomyocyte size, as well as Nppb and Col1a1 expression in H9c2 cells or primary neonatal rat cardiomyocytes. Moreover, protocatechuic acid attenuated Kmo overexpression–induced increases in Nppb mRNA levels. Protocatechuic acid or Kmo knockdown decreased isoproterenol‐induced ROS generation in vivo and in vitro. Thus, protocatechuic acid prevents heart failure by downregulating Kmo. Therefore, protocatechuic acid and Kmo constitute a potential novel therapeutic agent and target, respectively, against heart failure.


| INTRODUC TI ON
Heart failure is a clinical syndrome caused by various heart diseases, including myocardial infarction, hypertension and aortic stenosis, [1][2][3][4] which are associated with left ventricular hypertrophy and fibrosis. 5 Cardiac hypertrophy, which can be established using an osmotic minipump to infuse catecholamines (such as phenylephrine or isoproterenol), is defined as cardiac myocyte growth without an increased number of cells, resulting in an increased cardiac wall thickness and narrowed lumina of the cardiac chambers. The heart weight apparently increases in response to hypertrophic stimulation, whereas heart failure refers to a state wherein cardiac function deteriorates due to cardiac muscle weakening and luminal enlargement.
The successful establishment of an isoproterenol-induced heart failure model is dependent on the mouse species and the isoproterenol concentration. 6 Both transverse aortic constriction (TAC) and infusions of isoproterenol are used to establish animal models of heart failure characterized by a reduced ejection fraction, which occurs when the heart does not pump efficiently and causes lung oedema.
Cardiac fibrosis is characterized by collagen deposition in the myocardial interstitium and perivascular regions, 7 which leads to heart failure. Therefore, the suppression of cardiac fibrosis is a potential therapeutic strategy for preventing progression to heart failure.
Hydroxybenzoic acid isomers prevent inflammation, cancer, oxidation, hypertension, cardiac hypertrophy and heart failure. For example, gallic acid (3,4,5-trihydroxybenzoic acid), gentisic acid (2,5-dihydroxybenzoic acid) and protocatechuic acid (3,4-dihydroxybenzoic acid) prevent cardiac hypertrophy, fibrosis and hypertension, as reported recently. [8][9][10][11][12][13] Additionally, gallic acid and gentisic acid alleviate pressure overload-induced heart failure or angiotensin II-induced atrial fibrillation. [14][15][16][17] The structures of protocatechuic acid and gallic acid are similar; protocatechuic acid is present in vegetables, fruits and grains (purple rice bran) 18 and exerts many beneficial anticancer effects, 19 inflammation, fibrosis, 20 oxidation, 21 microbial activity and thrombosis. 22 In a recent study, protocatechuic acid was found to attenuate cardiac hypertrophy in isoproterenol-infused mice. 10 Protocatechuic acid-mediated prevention of fibrosis may provide a basis for improving the pathology of heart failure. 11,20 Fibrosis is a very critical factor in heart failure and is the final stage of most cardiovascular diseases. Heart failure is emerging as an important syndrome and is associated with a high hospitalization-related burden. 23 However, the effects of protocatechuic acid on heart failure and its underlying mechanisms have not been elucidated. In our study, we aimed to demonstrate that protocatechuic acid can be a new drug for heart failure treatment.
Kynurenine-3-monooxygenase (Kmo), which is a nicotinamide adenine dinucleotide phosphate-dependent flavin monooxygenase, catalyses the conversion of kynurenine to 3-hydroxykynurenine 24 and mediates the catabolic step of the kynurenine pathway, which is the crucial route of tryptophan catabolism. Cardiovascular conditions, such as heart failure, are positively correlated with serum levels of aberrant kynurenine catabolites, including kynurenine, 3-hydroxykynurenine and quinolinic acid. [25][26][27] Based on Kmo expression profile information determined using RNA sequencing data, we hypothesized that Kmo plays an important role in the pathogenesis of heart failure. In this study, using Kmo knockdown and Kmo overexpression in primary neonatal cardiomyocytes and fibroblasts, we found that Kmo regulated cardiac hypertrophy and fibrosis.
This study was conducted to investigate the therapeutic effects of protocatechuic acid in an isoproterenol-induced heart failure mouse model and to reveal the underlying pathogenetic mechanisms.

| Establishment of an animal model of heart failure
A heart failure model was established by administering high-dose isoproterenol to male C57BL/6NTac mice; with a modified version of a previously reported protocol. 6 All animal procedures were approved by the Animal Experimental Committee of Chonnam National University Medical School (CNUHIACUC-22007) and performed failure by downregulating Kmo. Therefore, protocatechuic acid and Kmo constitute a potential novel therapeutic agent and target, respectively, against heart failure.

K E Y W O R D S
cardiac hypertrophy, fibrosis, heart failure, kynurenine-3-monooxygenase, protocatechuic acid according to the Guide for the Care and Use of Laboratory Animals (US National Institutes of Health Publications, 8th edition, 2011).
To induce heart failure, 7-week-old male C57BL/6NTac mice were anaesthetised with ketamine (100 mg/kg body weight) and xylazine (5 mg/kg body weight). Next, an osmotic minipump was implanted in the mice to infuse isoproterenol (80 mg/kg body weight/ day) for 14 days. The animals were divided into the following four groups (n = 8 per group): sham + dimethyl sulfoxide (DMSO); sham + protocatechuic acid (100 mg/kg body weight/day); isoproterenol + DMSO; and isoproterenol + protocatechuic acid. Mice in the isoproterenol + protocatechuic acid group received daily intramuscular injections of protocatechuic acid from Day 6 of isoproterenol administration for 9 days. The dosage of protocatechuic acid (100 mg/ kg body weight/day) was chosen based on the tolerable and safe concentration identified in previous studies. 10,18,28
Echocardiography was performed using the Vivid S5 system model with a 13-MHz linear array transducer (GE Healthcare), as described previously. 10 M-mode images and parameters were acquired from the short-axis view of the left ventricle at the level of the papillary muscles. The cardiac parameters were measured on Days 5 and 14 of isoproterenol administration.

| Histological analysis and picrosirius red staining
Cardiac tissues were fixed with 4% paraformaldehyde, rinsed with tap water and embedded in paraffin. The paraffin-embedded tissues were sectioned to a thickness of 3 μm. For haematoxylin and eosin (H&E) staining, the sections were deparaffinized with xylene and hydrated in a descending ethanol series (100%-70%). The sections were washed with tap water and stained with haematoxylin for 4 min and eosin, followed by phloxine for 1 min. Next, the sections were dehydrated with ethanol, cleared with xylene and mounted for observation.
For picrosirius red staining, the paraffin-embedded sections were deparaffinized with xylene overnight, dehydrated and hydrated with distilled water. The sections were then incubated with a drop of picrosirius red for 60 min. Next, the sections were quickly rinsed twice with acetic acid solution, rinsed with ethanol, sequentially treated with ethanol and xylene and mounted for observation. The tissue images were captured using a microscope (Nikon Eclipse 80i microscope, Nikon).

| Quantitative real-time polymerase chain reaction (qRT-PCR)
Total RNA was extracted from the cardiac tissues using TRIzol reagent (Invitrogen), according to the manufacturer's instructions. To quantify the RNA concentration, the absorbance of the sample was measured at 260 nm using an ACTGene spectrophotometer (ASP-2680). The isolated RNA was reverse transcribed into complementary DNA (cDNA) using TOPscript RT DryMIX (Enzynomics). The SYBR green PCR kit (Enzynomics) and specific primers were used to conduct qRT-PCR analysis. The relative mRNA levels were calculated using the 2 −ΔΔCt method.
The primers used in the qRT-PCR analysis are shown in Table 1.
Immunoreactive signals were visualized using Immobilon Western blotting detection reagents (EMD Millipore), and the protein band intensities were quantified using ImageJ (National Institutes of Health).

| Primary rat cardiac fibroblast culture
Primary rat cardiac fibroblasts were isolated from the cardiac tissues of 1-to 2-day-old rat pups, as described previously. 15 The atrium was excised, and the ventricles were finely chopped using sterile scissors.
The samples were incubated in 1× ADS buffer (116 mM NaCl, 20 mM HEPES, 10 mM NaH 2 PO 4 , 5.5 mM glucose, 5 mM KCl and 0.8 mM MgSO 4 ) containing 0.1% collagenase type II on a shaker at 120 rpm for 2 h. To neutralize the collagenase activity, an equal amount of DMEM containing 20% FBS was added to the samples. The samples were then centrifuged, and the cell pellets were resuspended in DMEM containing 10% FBS. Cells that were passaged twice were used for the experiment. To investigate the effect of protocatechuic acid on fibrosis, rat cardiac fibroblasts were treated with protocatechuic acid 1 h before treatment with transforming growth factor (TGF)-β1 (10 ng/mL).

| Primary neonatal rat cardiomyocyte culture
Primary neonatal rat cardiomyocytes were isolated from the cardiac tissues of 1-to 2-day-old rat pups, as described previously. 8 The ventricles were minced and digested with 0.1% collagenase type II and pancreatin. To complete tissue dissociation, FBS was added to stop enzymatic digestion and centrifuged at 1200 rpm for 5 min.
Cell pellets were resuspended, and fibroblasts were allowed to adhere for 1 h in 100 mm dishes. Supernatant cells were centrifuged at 1200 rpm for 5 min, and the cells were subjected to Percoll gradients, with centrifugation at 3000 rpm for 30 min. The Percoll layer cells were washed using 1× ADS buffer and seeded in 8-well plates or 12well plates. Cells were used for cell size and Western blot analyses.

| Cell size and sarcomeric actin organization
Rat neonatal cardiomyocytes were plated on an 8-well chamber, and immunocytochemistry was performed as described previously. 8 The cells were transfected with si-control or si-Kmo (100 nM) for 48 h and serum-starved overnight. The cells were incubated with isoproterenol (10 μM) for 48 h, fixed with 2% paraformaldehyde for 30 min and permeabilized with 0.1% Triton-X 100. The cells were blocked with 3% goat serum for 1 h and incubated with sarcomeric α-actin (1:200) overnight at 4°C. Then, secondary antibody (Alexa Fluor 568 goat anti-mouse IgG) was added, and antifade reagent (including DAPI) was used to detect nuclei. The cell area was measured using NIS Elements Software (Nikon).

| Overexpression transfection
H9c2 cells or rat cardiac fibroblasts were seeded in 12-well plates and transfected with pCMV-SPORT6 or pCMV-SPORT6-mouse Kmo (1000 ng/well) using Lipofectamine 3000 transfection reagents according to the manufacturer's instructions. Two days later, transfected cells were treated with or without protocatechuic acid for 9 h.
The pCMV-SPORT6-mouse Kmo clone was purchased from the Korea Human Gene Bank, Medical Genomics Research Center, KRIBB. Media (Leica Biosystems) and frozen in liquid nitrogen. The heart tissue preparations (10 μm thickness) were incubated with 10 μM DHE for 20 min at 25°C in the dark. The heart tissues were washed three times with deionized water. DHE-stained tissues were analysed using ImageJ.

| Statistical analysis
All statistical analyses were performed using GraphPad Prism, version 8 (GraphPad Software). The means among three or more groups were compared using one-way analysis of variance, followed by the Bonferroni post hoc test.

| Protocatechuic acid alleviates isoproterenol-induced cardiac hypertrophy in mice
We evaluated the effects of three doses (25,40 or 80 mg/kg body weight/day) of isoproterenol infusion on C57BL/6NTac male mice.
The most efficient isoproterenol dose to establish this heart failure model was 80 mg/kg body weight/day (treatment for 2 weeks; Figure S1). Even at this high concentration, the mice did not die.
Isoproterenol treatment increased the end-diastolic interventricular septal thickness (IVSd) and the end-diastolic left ventricular posterior wall thickness (LVPWd) and it decreased the left ventricular diameter ( Figure S2). These findings indicate that cardiac remodelling is initiated on the fifth day of isoproterenol infusion.
For further investigation of whether protocatechuic acid prevents the transition from cardiac hypertrophy to heart failure, intramuscular protocatechuic acid was administered to C57BL/6NTac mice after they had received 5 days of isoproterenol infusion ( Figure 1A,B). Protocatechuic acid administration suppressed the isoproterenol-induced upregulation of the heart-weight-to-bodyweight ratio in C57BL/6NTac mice ( Figure 1C) such that this value was not markedly different between the sham + DMSO and sham + protocatechuic acid groups. Haematoxylin and eosin staining revealed that protocatechuic acid decreased cardiomyocyte size in isoproterenol-treated mice ( Figure 1D,E). To further evaluate whether isoproterenol administration induces heart failure, wet lungs were weighed. The wet lung-weight-to-body-weight ratio was increased in the isoproterenol group, and it was reduced in the protocatechuic acid-treated group ( Figure 1F). Additionally, protocatechuic acid decreased the isoproterenol-induced upregulation of Nppb, Lgals3 and Timp1 but did not alter Egr3 mRNA levels ( Figure 1G-J). Western blotting analysis revealed that the Nppb levels in the isoproterenol + protocatechuic acid group were downregulated compared with those in the isoproterenol + DMSO group ( Figure 1K,L).

| Protocatechuic acid alleviates pathological cardiac remodelling and improves cardiac function in this isoproterenol-induced heart failure model
Treatment with isoproterenol (80 mg/kg body weight/day) resulted in heart failure, which was characterized by an enlarged left ventricular luminal diameter and decreased fractional shortening (Figure 2A).
Heart rate was increased in the isoproterenol group compared with the sham group ( Figure 2B). The 14 days of isoproterenol infusion However, the fractional shortening and ejection fraction in the isoproterenol + protocatechuic acid group were significantly increased compared with those in the isoproterenol + DMSO group.

| Protocatechuic acid attenuates cardiac fibrosis in isoproterenol-treated mice
This animal model of heart failure exhibited cardiac fibrosis.
Picrosirius red staining revealed that cardiac collagen deposition (pink-stained areas) in the isoproterenol + DMSO group was greater than that in the sham + DMSO group ( Figure 3A). However, protocatechuic acid significantly alleviated isoproterenol-induced cardiac fibrosis ( Figure 3B). Additionally, protocatechuic acid suppressed the isoproterenol-induced upregulation of cardiac Col1a1 and Fn1 mRNA levels ( Figure 3C,D). The TGFβ-Smads signalling pathway is involved in fibrosis and cardiogenesis. 29 We examined whether Smad3 and Smad4 can change in this isoproterenol-induced heart failure model.
Analysis with qRT-PCR showed no differences in the Smad3 and Smad4 mRNA levels between the four groups ( Figure 3E,F). Fibrosis is affected by chronic inflammatory processes. 30 We investigated relevant inflammation-related gene expression. TNFα and IL-1β mRNA levels were not altered in heart tissues treated with protocatechuic acid or isoproterenol ( Figure 3G,H). Western blotting using

| Protocatechuic acid suppresses several signalling pathway-related genes in isoproterenol-treated mice
To investigate the genes that play crucial roles in protocatechuic acid-mediated suppression of heart failure, whole ventricular tissues from the sham + DMSO, isoproterenol + DMSO and isoproterenol + protocatechuic acid groups were subjected to RNA sequencing. By analysing differentially expressed genes (DEGs) obtained using gene expression values through mouse transcriptome sequence analysis, a heat map was used to visualize the degree of similarity in the expression patterns of the three groups ( Figure 4A).  organismal processes, multicellular organism development, system development, anatomical structure morphogenesis and developmental processes ( Figure 4D).
Based on the RNA sequencing results, genes that were upregulated in the isoproterenol group relative to the sham group, as well as genes that were downregulated in the isoproterenol + protocatechuic acid group, were selected as candidates. Protocatechuic acid de-

| Downregulation of Kmo alleviates fibrosis and suppresses hypertrophic marker genes in H9c2 cells
The

| Protocatechuic acid treatment or Kmo knockdown alleviates TGF-β1-induced collagen and fibronectin expression
The role of Kmo in fibrosis was examined. Primary neonatal rat cardiac fibroblasts were transfected with si-control and si-Kmo.
The endogenous Kmo mRNA levels were downregulated in the si-Kmo-transfected cells ( Figure 6A). Additionally, the Col1a1 and

| Kmo knockdown alleviates isoproterenol-induced cardiomyocyte hypertrophy in H9c2 cells and primary neonatal rat cardiomyocytes
To identify whether Kmo affects cardiomyocyte hypertrophy, the size of si-control-transfected or si-Kmo-transfected H9c2 cells that were treated with vehicle or isoproterenol was measured. Alexa Fluor 488 phalloidin staining revealed that Kmo knockdown suppressed the isoproterenol-induced increase in cell area ( Figure S4A

| Protocatechuic acid attenuates Kmo overexpression-induced hypertrophic marker gene expression in H9c2 cells
To identify whether protocatechuic acid affects Kmo overexpression-induced cardiac hypertrophy, qRT-PCR was performed. Kmo overexpression increased the endogenous mRNA levels of Kmo in H9c2 cells. The increase was attenuated by protocatechuic acid treatment ( Figure 7F). Protocatechuic acid downregulated Kmo overexpression-induced Nppb mRNA levels in H9c2 cells ( Figure 7G). Representative immunoblot images and quantification of the Fn1 and Col3 band intensities in cardiac tissues. Actb was used as a loading control and was the band detected on the same blot used in Figure 1K. ***p < 0.001; ### p < 0.001. Data are reported as the mean ± standard error of mean. Statistics: one-way analysis of variance, followed by Bonferroni post hoc tests. DMSO, dimethyl sulfoxide; ISO, isoproterenol; PCA, protocatechuic acid.

F I G U R E 5
Protocatechuic acid downregulates the expression levels of 10 candidate genes selected from RNA sequencing data in an isoproterenol-induced heart failure model. (A-J) Ten candidate genes were selected from the RNA sequencing analysis of different groups. The mRNA levels of Lefty1, Mmp12, Cysltr1, Kmo, Pamr1, Tnfrsf9, Cfap61, Spp1, Ubxn10 and Enkur were determined using qRT-PCR. ***p < 0.001; ### p < 0.001.

| Protocatechuic acid or Kmo knockdown attenuates ROS production in vitro and in vivo
To determine whether the association between Kmo and cardiac hypertrophy was attributable to oxidative stress, DHE staining was performed in H9c2 cells and heart tissues. Isoproterenol-treated cells had significantly higher levels of ROS generation than vehicletreated cells, and ROS generation was reduced by protocatechuic acid treatment ( Figure 8A,B). These observations were similar in isoproterenol-infused heart tissues with or without protocatechuic acid ( Figure 8C,D). To further evaluate whether protocatechuic acid-mediated ROS reduction occurs via Kmo, DHE staining was performed on isoproterenol-treated Kmo siRNA-transfected cells.

| DISCUSS ION
This study demonstrated that protocatechuic acid treatment alleviates cardiac hypertrophy, fibrosis, oxidative stress, and dysfunction in an isoproterenol-induced heart failure mouse model by downregulating Kmo ( Figure 8G). The most interesting observation in this study was the restoration of fibrosis-associated cardiac dysfunction by protocatechuic acid treatment. Protocatechuic acid suppressed the isoproterenol-induced downregulation of fractional shortening and ejection fraction by decreasing the left ventricular diameter in mice. These findings indicate that protocatechuic acid alleviates heart failure-induced pathological cardiac hypertrophy. Heart failure is accompanied by cardiac hypertrophy and lung oedema. In the present study, the isoproterenol-induced increase in wet lung mass was attenuated by protocatechuic acid treatment. The analyses of the heart-weight-to-body-weight ratio, cardiomyocyte cross-sectional area, and hypertrophic marker genes revealed that protocatechuic acid attenuated cardiac hypertrophy. The mRNA and protein levels of Nppb were downregulated in vivo upon protocatechuic acid treatment. A recent study showed that protocatechuic acid attenuated cardiac hypertrophy in isoproterenol (25 mg/kg body weight/day)treated Institute of Cancer Research (ICR) male mice, 10 and this finding is supported by the results of our study. Long-term isoproterenol infusion induces cardiac hypertrophy and fibrosis comparable to that achieved using the TAC model. 31 Isoproterenol, a non-selective β-adrenergic receptor agonist, contributes to cardiac hypertrophy and fibrosis through β1-adrenergic signalling activation. [32][33][34] The TGF-β1/Smad signalling pathway is implicated in cardiac fibrosis. 35 Indeed, isoproterenol treatment enhances ColI, Col III, and α-SMA expression. 8,36,37 However, in the present study, neither isoproterenol nor TGF-β1 treatment induced Smad3 and Smad4 expression in heart tissues or cardiac fibroblasts. In contrast with our animal model, the TAC model has been shown to increase Smad3 but not Smad4 protein expression. 38 Similarly, TGF-β1 has been shown to upregulate the phospho-form Smad3 protein but not Smad3 in adult rat cardiac fibroblasts. 39 Indeed, isoproterenol treatment increases phosphorylated Smad3 protein levels but not Smad3 protein levels in heart and kidney tissues. 8,40 Picrosirius red staining and analysis of fibrosis-related genes  of ROS generation in heart failure. 51 Indeed, we demonstrated that isoproterenol causes oxidative stress in heart tissue 52 and cardiomyocytes, 53 which was attenuated by protocatechuic acid treatment or Kmo downregulation. These findings indicate that Kmo or protocatechuic acid can be a novel target gene and therapeutic agent, respectively, in the inhibition of isoproterenol-induced cardiac hypertrophy and oxidative stress.

| Limitations
Kmo is one of the target genes that mediate the therapeutic effects of protocatechuic acid against heart failure. Therefore, this study considered Kmo as a candidate gene for alleviating cardiac fibrosis and hypertrophy. However, the functions of other target genes that were modulated by protocatechuic acid have not yet been elucidated.
Considering the therapeutic benefits of Kmo inhibitors in various diseases, the findings of this study will be strengthened if future studies demonstrate the alleviation of heart failure following treatment with Kmo inhibitors. Another limitation was the use of neonatal cardiac cells instead of adult cardiac cells to mimic cardiac hypertrophy in vivo. Neonatal cardiomyocyte isolation is much easier (and is associated with higher cell yields) than adult cardiomyocyte isolation. A major advantage of neonatal cardiomyocytes is their spontaneously beating phenotype that precludes the need to administer agents to induce contractions. 54 Even though isoproterenol infusion readily causes cardiac hypertrophy and fibrosis, the use of high doses of isoproterenol enhances the associated myocardial impairments.
Therefore, the use of animal models for left coronary ligation or TAC can overcome the limitations of the isoproterenol model.

| CON CLUS IONS
Protocatechuic acid alleviated cardiac dysfunction, fibrosis and oxidative stress by downregulating Kmo in vivo and in vitro in an isoproterenol-induced heart failure model. The therapeutic effects of protocatechuic acid against heart failure were mediated through Kmo. The findings of this study suggest that the protocatechuic acid-mediated downregulation of Kmo is a potential therapeutic strategy for preventing heart failure.

CO N FLI C T O F I NTER E S T S TATEM ENT
None of the authors have any conflicts of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this research are available from the corresponding author upon request.