Long-term oral atazanavir attenuates myocardial infarction-induced cardiac fibrosis

Abstract

Atazanavir is an antiretroviral medication used to treat and prevent HIV/AIDS, but its effects on cardiac fibrosis are unknown. The aim of this study was to determine the effects of atazanavir on myocardial infarction (MI)- induced cardiac fibrosis in rats and used a TLR 9 antagonist,
hydroxychloroquine (HCQ) to elucidate the po- tential mechanism in vitro. The results indicated that atazanavir significantly attenuated CoCl2-induced neonatal rat cardiac fibroblast (rCFs) proliferation in a concentration-dependent manner. Treatment of rCFs with ataza- navir 1–10 µM blocked CoCl2-induced nuclear factor kappaB phosphorylation (p-NF-κB), p-IκBα and high-mo- bility group box 1 (HMGB1) expression. Treatment of rCFs with atazanavir 3 µM blocked HMGB1 downstream, p-NF-κB by blocking HMGB1 binds to toll-like receptor 9 (TLR 9). Intragastric administration of atazanavir sulfate 30 mg/k ameliorated changes in the left ventricular systolic pressure (LVSP), + dp/dtmax, and − dp/ dtmax after 4 weeks. This was associated with attenuation of α-SMA, HMGB1, p-NF-κB, TLR 9, collagen I, collagen III expression and hydroxyproline (Hyp) content in ischemic myocardial tissue. Additionally, con-
tinuous intragastric administration of atazanavir for 28 days attenuated cardiac remodeling. These data sug- gested that the protective effect of atazanavir is likely due to blocking of myocardial inflammatory cascades through an HMGB1/TLR 9 signaling pathway.

1. Introduction

Myocardial infarction (MI) is the main pathogenic factor underlying heart failure (HF) (Samuel et al., 2008). In the post-MI phase, the heart undergoes extensive myocardial remodeling in response to the ischemic injury, leading to thickening or stiffening of regions of the heart, with progressive deterioration of cardiac function, which can progress to HF (Kirk and Cingolani, 2016). Cardiac fibrosis plays a major role in car- diac remodeling after MI and is a predisposing factor for HF. The pa- thological features of cardiac fibrosis include phenotypic changes in cardiac fibroblasts, excessive proliferation, and deposition of extra- cellular matrix (ECM) proteins such as collagen types I and III (Sutra et al., 2008).

Fibroblasts undergo dynamic phenotypic alterations and direct re- parative response following MI, especially fibroblasts undergo myofi- broblast transdifferentiation forming stress fibers and expressing con- tractile proteins (such as α-smooth muscle actin) in the infarct area at the proliferative phase of healing during hypoxia (Chen and Frangogiannis, 2013). In addition, previous studies have indicated that nuclear factor kappaB (NF-κB) plays a key role in inflammatory response during MI injury and subsequent cardiac hypertrophy and cardiac fibrosis. Suppression of NF-κB activation diminishes cardiac hypertrophy and cardiac fibrosis (Hamid et al., 2011; Zelarayan et al., 2009). High mobility group box-1 protein (HMGB1) is a key cytokine to play an extracellular role involving cellular activation and proinflammatory responses (Lotze and Tracey, 2005; Yang et al., 2005). HMGB1 binds to toll-like receptor 9 (TLR 9), interacts and preassociates with TLR9 to regulate inflammatory responses (Ivanov et al., 2007). Knockout TLR9 showed a reduction of fibrosis (Watanabe et al., 2007; Gäbele et al., 2008). HMGB1 plays a critical role to amplify fibrosis and involves the activation of NF-κB. Inhibiting HMGB1 blocks in- flammatory response and may be a therapeutic target during cardiac fibrosis (Jiang et al., 2012; Su et al., 2014).

Atazanavir is an antiretroviral protease inhibitor class of drug that plays a key role in the treatment of HIV infection. In exploring novel pharmacological targets for cardiac fibrosis, we found that atazanavir attenuated hypoxia induced cardiac fibroblasts proliferation. This mo- tivated us to experimentally examine the pharmacological activity of atazanavir in a cardiac fibrosis model. Therefore, this study utilized both in vitro and in vivo cardiac fibrosis models to characterize the role of atazanavir in signaling modulation that contributes potentially to the observed cardiac attenuation.

2. Materials and methods

2.1. Cell culture and expressional analysis

Rat cardiac fibroblasts (rCFs) from newborn (1- to 2-day-old) Sprague-Dawley rats were isolated according to previous method (Villarreal et al., 1993). Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) with 10% fetal bovine serum, 100 kU/L peni- cillin and 100 mg/L streptomycin at 37 °C with 5% CO2 in a humidified incubator. The cells were cultured to approximately 70% confluency and starved in serum-free DMEM overnight prior to the treatment. The cells were then treated with 3 μM atazanavir sulfate (purity > 99.0%; CAS No.: 229975-97-7; Hanxiang Biomedical Company, Shanghai, China) with or without cobalt chloride (CoCl2; 100 μM) for 72 h and thereafter proteins were extracted.

2.2. rCFs proliferation assay and expressional assessment

To assess cellular proliferation, rCFs were maintained as described above. Cells were exposed to CoCl2 at 100 μM to mimic hypoxia and treated with varying concentrations of atazanavir (0, 1, 3, 10 μM) with or without 3 μM hydroxychloroquine (HCQ), a TLR 9 antagonist, for 72 h. Cellular proliferation levels were determined via cell counting.

To examine changes in expression, the cells were seeded into 6-well flat bottom plates and maintained as described above, with one well per plate maintained as an untreated control. Cells were treated with 3 μM atazanavir sulfate with or without CoCl2 (100 μM) for 72 h and there-after the supernatants were collected and the proteins were extracted. Collagen I and collagen III were examined by ELISA kits. The expression levels of TLR 9, HMGB1, p-NF-κB, p-IκBα and total NF-κB were examined by Western blot and normalized and displayed as described above. To investigate the possible mechanism of reduction in rCF pro- liferation, cells were treated with 3 μM atazanavir sulfate with or without 3 μM HCQ for 72 h, TLR 9 and expression levels of HMGB1 and p-NF-κB were examined using Western blot as described above.

2.3. Ethics approval and consent to participate

All animal experimental procedures in this study were performed in accordance with the Institutional Animal Care and the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Maryland, USA). The protocol was approved by the Committee on the Ethics of Animal Experiments of Binzhou Medical University (Permit Number: SCXK 20140013).

2.4. Induction of myocardial infarction (MI) model and experimental assessment

An MI model were induced according to a previously procedure (Jiang et al., 2010). Briefly, Rats were anesthetized with ketamine 100 mg/kg (i.m.) and xylazine 10 mg/kg (i.m.) and ventilated with room air using a rodent respirator. The chest was opened by middle thoracotomy and the left coronary artery was ligated at 2–3 mm from its origin between the left atrium and pulmonary artery conus using a 6- 0 prolene suture. A successful operation was confirmed by the occur- rence of ST-segment elevation in an electrocardiogram. This operation was performed by an experimenter who was blinded to the group as- signments of the animals to avoid subjective bias of the experimenter on the outcome. The sham-operated group underwent thoracotomy and cardiac exposure without coronary ligation. Thirty rats were divided into three groups including (I) non-MI rats; (II) MI rats received saline alone; (III) MI rats received intragastric administration of atazanavir sulfate (30 mg/kg) plus ritonavir (10 mg/kg). Atazanavir is a low oral bioavailability compound and, clinically, is generally coadministrated with Ritonavir, which boosts the oral bioavailability of atazanavir by inhibiting cytochrome P450 (CYP) 3A4, and P-glycoprotein via the same metabolic pathway (Le Tiec et al., 2005, 021567s026lbl). The rats were administered daily via intragastric administration of corre- sponding drug for continuous 28 days after MI 24 h. Treatment was orally administered on a daily basis for atazanavir-treated animals, while animals in the vehicle-treated and sham groups were given an equal volume of saline. At day 29, determine hemodynamics and ana- lyze histopathological change.

2.5. Determination of hemodynamics

Twenty eight days after MI, the rats were anesthetized with sodium pentobarbital (40 mg/kg) through intraperitoneal injection and a Millar vessel was inserted into the left ventricular cavity via the right common carotid artery. The pressure was transduced and amplified by a pressure transducer. Left ventricular systolic pressure (LVSP) and ± dp/dtmax were recorded and programmed by using a biotic signal collection and processing system (BIOPIC, American).

2.6. Histological examination

After fixation, three cross-sections through the ventricles were ob- tained and embedded. Paraffin sections were stained with Masson’s trichrome and aniline blue for determination of collagen volume frac- tion, stained with Masson’s trichrome for measurement of infarct size, the infarct size was expressed as previously described (Fishbein et al., 1978). The sections of HE staining measured myocyte size. For the measurement of cardiomyocyte cross-sectional area and diameter in the noninfarcted LV, a total of 30 myocytes sectioned transversely for area and longitudinally for diameter at the level of the nucleus were ran- domly chosen from each section at ×400 magnifications, and traced. To measure collagen volume fraction, 16 fields in the border and re- mote myocardium of the noninfarcted LV walls per section were scanned at a magnification of ×200. The interstitial collagen volume fraction was measured while omitting fibrosis of the perivascular, epi-, and endocardial areas from the study. The collagen volume fraction was obtained by calculating the mean ratio of connective tissue to the total tissue area of all the measurements of the section. The collagen-positive areas from all sections were determined by a single investigator who was unaware of the experimental groups.

2.7. Measurement of hydroxyproline (Hyp)

Frozen heart tissue samples were washed with saline and hydro- lyzed with 6 mol/L hydrochloric acid at 100 °C for 5 h. The Hyp content was determined by adding p-dimethylaminobenzaldehyde and was quantified on a spectrophotometer at 560 nm and expressed as milli- grams per gram of the wet heart tissue.

2.8. Western blots analysis of myocardial tissue

The heart samples (area at risk) were suspended in a buffer that contained 10 mM Tris (pH7.5),1.5 mM MgCl2, 10 mM KCl, and 0.1% Triton X-100 and lysed by homogenization. Nuclei were recovered by microcentrifugation at 6288 g for 5 min. The supernatant was collected and stored at − 80 °C for Western blot analysis. Nuclear proteins were extracted at 4 °C by gently resuspending the nuclei pellet in buffer that contained 20 mM Tris (pH7.5), 20% glycerol, 1.5 mM MgCl2, 420 mM NaCl, 0.2 mM EDTA, and 0.1% Triton X-100, followed by 1 h incubation with occasional vortexing at 4 °C. After microcentrifugation at 18,894 g for 15 min, the supernatant was collected. Protein concentrations of the extracts were measured by BCA assay. Equal amounts of cell protein (50 μg) were separated by SDS-PAGE and analyzed by western blot using specific antibodies to collagen I, collagen III, TLR 9, p-NF-κB, HMGB1 and GAPDH. Optical densities of the bands were scanned and quantified with a Gel Doc 2000. Results were expressed as fold increase over sham group.

2.9. Statistical analysis

Histopathological scores between vehicle-treated group and ataza- navir group were compared using sum of ranks test. Quantitative ex- perimental data are expressed as mean ± S.D. Every one of the sta- tistical tests was performed with a two-sided test, and the level of significance was established as P ≤ 0.05. If ANOVA revealed a sig- nificance (P ≤ 0.05), Dunnett’s test was conducted for multiple com- parisons. Statistical analysis was done with GraphPad Prism software (version 5), and a probability level of < 0.05 was determined to be statistically significant.

3. Results

3.1. Effects of atazanavir on rCFs proliferation, collagen production and proteins expression

The rCFs were examined in the absence or the presence of CoCl2 to mimic a pro-fibrotic environment during hypoxic conditions. Following CoCl2 induced hypoxia, rCFs proliferation increased compared with the normal group (P < 0.01), as shown in Table 1, but was significantly inhibited in a concentration-dependent manner following atazanavir sulfate treatment at concentrations between 1 and 10 μM compared with the CoCl2 group (P < 0.05).

To further characterize this inhibitory effect, atazanavir sulfate treatment was combined with HCQ, a TLR 9 antagonist. However, it found no further decline in rCFs pro- liferation compared with the HCQ group (P > 0.05), as shown in Table 1. In addition, the content of collagen I and collagen III was measured in CoCl2 stimulated rCFs. The results showed the contents of collagen I and collagen III were increased compared with the normal group (P < 0.01). However, collagen I and collagen III levels were significantly reduced in a concentration-dependent manner following atazanavir sulfate treatment at concentrations between 1 and 10 μM compared with the CoCl2 group (P < 0.05)), as shown in Table 1. It found no further decline in collagen I and collagen III in atazanavir 3 μM plus HCQ 3 μM compared with the HCQ group (P > 0.05), as shown in Table 1.

To further examine the mechanism of atazanavir sulfate on reducing rCFs proliferation during hypoxia, we investigated the expression of

HMGB1, p-NF-κB, p-IκBα and total NF-κB with or without atazanavir sulfate. Following CoCl2 induced hypoxia, HMGB1, p-NF-κB, p-IκBα and TLR 9 expression were increased compared with the normal group (P < 0.01), as shown in Fig. 1A and B, but HMGB1, p-IκBα and p-NF- κB expression were significantly inhibited following atazanavir sulfate treatment at 1–10 μM compared with the CoCl2 group (P < 0.05 or P < 0.01). Compared with the CoCl2 group, atazanavir 3 μM group has no change in total NF-κB expression, and no decline in TLR 9 expression (P > 0.05), as shown in Fig. 1A and B. HCQ treatment reduced HMGB1, p-NF-κB and TLR 9 expression (P < 0.05 or P < 0.01). Ata- zanavir treatment was combined with HCQ has no further decline in HMGB1, TLR 9 and p-NF-κB expression (P > 0.05) compared with the HCQ group (P > 0.05), as shown in Fig. 1C and D. These findings
suggest that atazanavir attenuates hypoxia induced rCFs proliferation by modulating the HMGB1/TLR 9 pathway.

3.2. Effects of atazanavir sulfate on myocardial function

We evaluated the effect of atazanavir on LVSP and ± dp/dtmax of the left ventricle after MI 28 days. Compared with vehicle-treated an- imals, rats treated with atazanavir had significantly improved LVSP,
+ dp/dtmax and − dp/dtmax 28 days after MI as shown in Table 2. In addition, we found no further change in SP, DP and HR compared with the HCQ group (P > 0.05). It is clear that continuous atazanavir treatment 28 days provided long-term benefits for the myocardial function recovery after MI.

3.3. Effects of atazanavir on cardiac collagen volume and myocytes hypertrophy after MI 28 days

To clarify the mechanism of long-term improved cardiac perfor- mance caused by atazanavir, we examined the effects of atazanavir treatment on mural hypertrophy and collagen volume in the non-in- farcted region and infarct size. There was no difference in infarct size between the vehicle-treated group and atazanavir 30 mg/kg group (38.11 ± 4.15% and 38.80 ± 4.62%, respectively). The cross-sec- tional area and diameter of myocytes in the non-infarcted LV and hy- pertrophy of the myocytes significantly increased in vehicle-treated rats compared with Sham rats, while inhibited by atazanavir, as shown in Fig. 2A, C and D. Atazanavir significantly attenuated an increase in morphometrical collagen volume fraction in the border left ventricle, as shown in Fig. 2B and E. In agreement with the above results, the heart index (heart-weight to body-weight ratio) which was increased in the vehicle-treated rats compared with sham rats, was significantly (p < 0.05) lowered by continuous atazanavir treatment, as shown in Fig. 2F.

3.4. Effects of atazanavir on the expressions of α-SMA, HMGB1, p-NF-κB, TLR 9, collagen I, collagen III and the content of Hyp in vivo

Changes in the expressions of α-SMA, HMGB1, TLR 9, p-NF-κB, collagen I and collagen III were also examined by Western blot analysis, as shown in Figs. 3–5. In vehicle-treated rats, all of the examined pro- tein expressional levels and the content of Hyp increased relative to the sham animals (P < 0.01), while those protein expressional levels and the content of Hyp decreased following atazanavir treatment compared with the vehicle-treated rats (P < 0.01). The results of in vitro and in vivo investigations suggest that atazanavir can reduce fibroblast pro- liferation and collagen deposition by modulating the HMGB1/TLR 9 pathway.

4. Discussion

In this study, we confirmed that the presence of CoCl2 to mimic a pro-fibrotic environment during hypoxic condition significantly in- creased the rCFs proliferation, which were evidenced via cell counting in the culture medium. However, pretreatment with different con- centrations (1, 3 and 10 μM) of atazanavir greatly decreased rCFs pro- liferation in a concentration-dependent way, and decreased the content of collagen I and collagen III in supernatant of CoCl2 stimulated rCFs.

Fig. 1. Effects of atazanavir on HMGB1, TLR 9 and p-NF-κB expression in hypoxic rCFs. Fig. 1A and C: The expression of HMGB1 and p-NF-κB were analyzed by Western blotting in cobalt chloride (CoCl2) induced hypoxic rCFs with or without atazanavir. Fig. 1B and D: The expression of HMGB1, TLR 9 and p-NF-κB were analyzed by Western blotting in cobalt chloride (CoCl2) induced hypoxic rCFs with or without atazanavir or HCQ. Results were displayed as increase in percentage relative to the normal, with atazanavir (Ata) and CoCl2 treatments indicated. All data were displayed as a mean ± S.D. (n = 3). #P < 0.01 vs. the normal group; *P < 0.05, **P < 0.01 vs. the CoCl2 group, significance was determined by ANOVA, followed by a Dunnett's test.

We also examined the effects of atazanavir on LV remodeling induced by coronary artery ligation in rats, a clinically relevant model used to assess the pathology and mechanisms of MI, a major cause of HF. We found that continuous treatment with atazanavir showed marked pro- tection against MI induced LV dysfunction and fibrosis, potentially via downregulation of HMGB1, TLR 9 and p-NF-κB expression through the HMGB1/TLR 9 signaling pathway.

After MI, the damaged tissue is structurally replaced by develop- ment of a stable scar. This adaptive hypertrophy can help compensate for cardiac dysfunction caused by infarction (Kirk and Cingolani, 2016). However, the surviving myocardial tissue can also exhibit progressive injury, leading to hypertrophy, fibrosis, maladaptive remodeling, im- paired cardiac function, and ultimately myocardial failure (Lindsey et al., 2002; Sun et al., 2002). We observed a similar progressive loss of diastolic and systolic cardiac function in rats after MI, with decreases in LVSP, + dp/dtmax, and − dp/dtmax and increases in heart index. By contrast, continuous atazanavir administration significantly amelio- rated the changes in LVSP, + dp/dtmax, and − dp/dtmax, and there was a reduction in heart index. Thus, atazanavir has the potential to ame- liorate impairments in myocardial contractility and relaxation, re- sulting in improvement of MI-induced LV dysfunction.

Cardiac fibrosis is the main pathological feature of myocardial le- sions observed in late-phase post-MI and is caused by abnormal accu- mulation of ECM, which has marked effects on cardiac function (Siwik et al., 2001). Collagen I and collagen III are two main components of the ECM (Feldman et al., 1993), with collagen I contributing to myo- cardial stiffness and collagen III contributing to myocardial compliance ( Klappacher et al., 1995). Thus, the balance between collagen I and collagen III is important for maintaining the structural and functional integrity of the heart. In the present study, we evaluated cardiac fibrosis with Masson's trichrome staining. In addition, western data for collagen I and collagen III also indicated a marked increase in interstitial fibrosis in MI rats, which was reduced with continuous atazanavir treatment. These findings were validated by measuring: (i) Hyp content, an index used to measure collagen metabolism and assess fibrosis (Liu et al., 2016); (ii) α-SMA levels, a highly contractile protein expressed by myofibroblasts, of which persistent activation can contribute to cardiac fibrosis under pathological conditions; and (iii) infarct size. rCFs pro- liferation was significantly inhibited and collagen I and collagen III is reduced in a concentration-dependent manner following atazanavir sulfate treatment at concentrations between 1 and 10 μM in vitro. In addition, Hyp, α-SMA, collagen and myocyte diameter were all up- regulated in MI rats and were markedly reduced with atazanavir treatment. Thus, these results from in vitro and in vivo investigation suggest atazanavir can reduce fibrosis through inhibiting collagen de- position, leading to amelioration of ventricular function following myocardial infarction.

Fig. 2. Effects of atazanavir on pathological progress of myocardial tissue after MI 28 days. Fig. 2A1–A3: Representative light mi- croscopic appearance of cardiac hypertrophy in the border of noninfarcted left ventricle. (hematoxylin staining for sham (A1), vehicle- treated (A2) and atazanavir (A3). Fig. 2B1–B3: Representative light microscopic appearance of myocardial tissue (Masson trichrome staining for Sham (B1), vehicle-treated (B2) and ataza- navir (B3). Fig. 2C–D: Effects of atazanavir on cardiac hypertrophy in the border of non- infarcted left ventricle. Fig. 2E: Effects of ata- zanavir on collagen volume fraction in the border left ventricle. Fig. 2F: Effects of ataza- navir on the change of heart index. Data are expressed as mean ± S.D., n = 10 in each group. Ata indicated atazanavir. *P < 0.01 vs. vehicle-treated group; #P < 0.01 vs. sham group. Significance was determined by one- way analysis of ANOVA followed by Dunnett's test.

TLR 9-dependent activation by DNA-containing immune complexes is mediated by HMGB1 (Tian et al., 2007). TLR 9 and HMGB1 play important roles in myofibroblast proliferation in fibrotic diseases. In- hibition of HMGB1/TLR 9 can attenuate fibrosis (Li et al., 2014). Our study demonstrated that increased HMGB1 release was accompanied by the increase in NF-κB and the activation of HMGB1/TLR 9 signaling both in vitro and in vivo. Atazanavir showed marked protection against MI induced fibrosis, potentially via downregulation of HMGB1, TLR 9 and p-NF-κB expression through the HMGB1/TLR 9 signaling pathway.

5. Conclusions

In summary, this study demonstrated that atazanavir can inhibit rCFs proliferation and ameliorate myocardial fibrosis via HMGB1/TLR 9 signaling, highlighting a new drug that could be potentially useful in the treatment of myocardial fibrosis.

Fig. 3. Effects of atazanavir on α-SMA, HMGB1, p-NF- κB, TLR 9, after MI 28 days. Fig. 3A–D: At MI 28 days, equal amounts of cell protein (50 μg) were separated by SDS-PAGE and analyzed by western blot using specific antibodies to α-SMA, HMGB1, p-NF-κB and TLR 9. Ata indicated atazanavir. Data are expressed as mean ± S.D., n = 5 in each group. *P < 0.05, **P < 0.01 vs. vehicle-treated group; #P < 0.01 vs. sham group. Significance was determined by one-way analysis of ANOVA followed by Dunnett's test.

Fig. 4. Effects of atazanavir on collagen I and collagen III after MI 28 days. At MI 28 days, equal amounts of cell protein (50 μg) were separated by SDS- PAGE and analyzed by western blot using specific antibodies to collagen I and collagen III. Ata indicated atazanavir. Data are expressed as mean ± S.D., n = 5 in each group. *P < 0.01 vs. vehicle-treated group; #P < 0.01 vs. sham group. Significance was determined by one-way analysis of ANOVA followed by Dunnett's test.

Fig. 5. Effects of atazanavir on content of Hyp after MI 28 days. At MI 28 days, Hyp content was determined. Ata indicated atazanavir. Data are ex- pressed as mean ± S.D., n = 5 in each group. *P < 0.01 vs. Vehicle-treated group; #P < 0.01 vs. sham group. Significance was determined by one-way analysis of BMS-232632 ANOVA followed by Dunnett’s test.