Cathepsin A inhibition attenuates myocardial infarction-induced heart failure on the functional and proteomic levels

Six to seven week old male C57BL/6 mice weighing 19−25 g were purchased from Charles River Laboratories. Twelve weeks old Wistar rats weighing 250–300 g were purchased from Harlan. Animals were housed in an air-conditioned room with a 12 h dark/light cycle and received standard mouse chow with free access to tap water. They were allowed 7 days to adjust to the new environment before starting the experiments. All animal studies conformed to the German law for the protection of animal guidelines and the guide for the care and use of laboratory animals published by the US National Institutes of Health (NIH Publications No 85–23, revised 1996) as well as to Sanofi-Aventis Ethical Committee guidelines.

SAR1 was synthesized in the chemical department of Sanofi. SAR1 ((S)-3-{[1-(2-Fluoro-phenyl)-5-methoxy-1H-pyrazole-3-carbonyl]-amino}-3-o-tolyl-propionic acid) is a new orally active cathepsin A inhibitor previously described as compound 2a by Ruf et al. [8].

Study animals were randomized to MI versus sham operation. Mice were anaesthetized with 0.1 ml/10 g body weight i.p. of a mixture of ketamine (100 mg/ml), xylazine (20 mg/ml) and atropine (0.5 mg/ml). They were placed on a heated operating table and ventilated at a tidal volume of 7 µl/g body weight. The body temperature was controlled by a rectal probe and maintained at 37 °C. A left thoracotomy was performed via the third intercostal space. MI was induced by placing a 7/0 silk suture around the left anterior descending coronary artery near the atrial auricle. For animals undergoing a sham operation only the chest cavity was opened without further surgical procedures. The lungs were inflated by increasing positive end-expiratory pressure and the thoracotomy site was closed. Animals were allowed to recover and treated with 0.1 mg/10 g body weight buprenorphine (0.3 mg/ml) by subcutaneous injection for post-operative analgesia. Mice were then randomized to placebo or SAR1 (100 mg/kg in chow). Treatment started 1 day after LAD ligation and lasted for 4 weeks. In the SAR1 group fourteen animals died during the 1st week in contrast to eight in the placebo group. No further deaths occurred between weeks two and four in both groups. The difference in mortality under SAR1 treatment did not reach significance (p value >0.08, Mantel-Cox log-rank test). The final sample size at the end of the study for each group was the following: sham (n = 10), MI-placebo (n = 15), MI-SAR1 (n = 10). For determination of infarct size hearts were cut open at the end of the study, photographed and digitized. Images were analyzed for total and the more lucid infarcted area with the Explora Nova image analysis software Morpho Expert (La Rochelle, France) using a Leica- macrosetup.

Surgery in rats was performed in a similar way as described before [14]. They were randomized to placebo or SAR1 (3, 10 and 30 mg/kg, p.o.). Treatment started 1 day after LAD ligation and lasted for 7 days. Sample size for each group was the following: sham (n = 10), MI-placebo (n = 13), MI-SAR1 3 mg/kg (n = 11), MI-SAR1 10 mg/kg (n = 11), MI-SAR1 30 mg/kg (n = 11).

For the assessment of pressure–volume relationships, mice were anesthetized with 2 % Isoflurane. Respiration was controlled through a tracheotomy cannula, and animals were mechanically ventilated (Minivent, Hugo Sachs) at 140 breaths/min and a tidal volume of 7 µl room air/g BW. Physiological temperature was maintained with a heated operating table with rectal probe. The left ventricle was catheterized retrogradely via the right carotid artery using a 1.4 F impedance-micromanometer tip catheter (Millar Instruments, Houston, TX, USA). Pressure–volume signals were recorded at steady state and during transient preload reduction achieved by vena cava inferior occlusion. Artificial ventilation was stopped during sampling of approximately ten cardiac cycles for analysis of pressure–volume measurement. Data were digitalized with a sampling rate of 1000 Hz and recorded using specialized software (HEM 3.5, Notocord, France). For subsequent analysis of pressure–volume loops, Notocord and Excel software was used.

In rats hemodynamic parameters were measured in anaesthetized animals using Millar Tip catheters (Millar Instruments Inc, Houston, USA) introduced from the right arteria carotis and advanced into the left ventricle.

Plasma BNP32 was determined with commercially available ELISAs (RayBiotech, Phoenix Pharmaceuticals Inc.)

For proteome comparison, left ventricles from sham operated mice (sham) and mice which underwent left coronary artery ligation treated with placebo or SAR1 were prepared as described above. Preparation of mass spectrometry samples was performed as described previously, including stable isotope labeling with either formaldehyde light [d(0)¹² C], medium [d(2)¹² C] or heavy [d(2)¹³ C] for quantitative comparison and pre-fractionation via strong cation exchange chromatography [15]. LC–MS/MS analysis is described in the corresponding section. MS files were analyzed by MaxQuant version 1.3.0.5 with the Uniprot mouse database downloaded on October 2011, counting 44819 entries. MaxQuant analysis included an initial search with a precursor mass tolerance of 20 ppm for mass recalibration. In the main Andromeda search precursor mass and fragment mass were searched with initial mass tolerance of 6 ppm and 0.5 Da respectively. The search included variable modifications of methionine oxidation and N-terminal acetylation, and fixed modification of carbamidomethyl cysteine. Minimal peptide length was set to seven amino acids and a zero missed cleavages were allowed. The false discovery rate (FDR) was set to 0.05 for peptide and protein identifications; however we only considered proteins that were independently identified in at least three replicates. For comparison between samples we used a labeling scheme based on multiplicity three: dimethLys0/dimethNter0 (light label); dimethLys4/dimethNter4 (medium label); dimethLys8/dimethNter8 (heavy label). A minimum of two ratio counts was used to determine the normalized protein intensity. Protein table were filtered to eliminate the identifications from the reverse database, and common contaminants.

Analysis was performed on an Orbitrap XL (Thermo scientific) mass spectrometer that was coupled to an Ultimate 3000 micro pump (Thermo scientific). Buffer A was 0.5 % acetic acid, buffer B 0.5 % acetic acid in 80 % acetonitrile (HPLC grade). Liquid phases were applied at a flow rate of 300 nl/min with an increasing gradient of organic solvent for peptide separation. Reprosil-Pur 120 ODS-3 (Dr. Maisch, Ammerbuch, Germany) was used to pack column tips of 75 µm inner diameter and 11 cm length. The MS was operated in data dependent mode and each MS scan was followed by a maximum of five MS/MS scans.

30 μg of heart tissue lysate were loaded on to 12 % SDS–polyacrylamide gels. GAPDH served as an internal loading control. After electrophoretic separation, proteins were transferred on polyvinylidene fluoride membranes using a semidry blot system (Bio-Rad, Munich, Germany). After blocking, the membranes were exposed to the primary antibodies (GAPDH, 1:1000; carbonic anhydrase 3, 1:200; periostin, 1:500; troponin T, 1:1000; cytoglobin, 1:100) overnight at 4 °C. After washing, the membranes were incubated for 2 h with the secondary antibody. The membranes were washed and developed with the West Pico Chemiluminescent substrate (Pierce). Peroxidase activity was detected with a LumiImager device (Roche Applied Science, Mannheim, Germany). The primary antibodies were purchased from Abcam (Cambridge, MA) (GAPDH: Cat.No. Ab9484), Santa Cruz Biotechnology (Santa Cruz, CA) (carbonic anhydrase 3, Cat.No. Sc-50714; cytoglobin, Cat.No. Sc-66855), R&D Systems (Minneapolis, MN) (periostin: Cat. No. AF2955), Sigma-Aldrich (St. Louis, MO) (troponin T: Cat.No. SAB2502131). Western blots were quantified using the Fusioncapt advance software (Vilber Lourmat, Eberhardzell, Germany).

H9C2 rat cardiac myoblasts were purchased from ATCC (Cat.No. CRL-1446) and cultured in Dulbecco’s modified eagle medium (DMEM; PAN) supplemented with 10 % fetal calf serum (FCS; PAN), 1 % non-essential amino acids, 1 % MEM vitamins, 1 % penicillin/streptomycin at 37 °C in humidified air containing 5 % CO₂. To mimic ischemia, cell were grown to 70 % confluency, washed three times with PBS and incubated in DMEM serum free supplied with SAR1 or DMSO for 24 h in a special incubation bag containing Anaerocult^® A mini (Merck, Darmstadt, Germany), a chemical mixture that completely binds oxygen, creating an oxygen-free environment. The bag was sealed and severe hypoxia (<0.5 % O₂) was assessed by color change of a test strip (Anaerotest, Merck) from blue to white within the first 90 min of incubation [16, 17]. Human umbilical vein endothelial cells (HUVECs) were kindly provided by Dr. Jennifer Esser. Cells were cultured in enhanced endothelial cell growth medium (PELOBiotech GmbH, Martinsried, Germany) and used for angiogenesis assays.

Caspase-3 activity was determined using a caspase-3 activity assay kit (Enzo Life Sciences, Lörrach, Germany, Cat.No. ALX-260-031). Following the manufacturer’s protocol, cell lysates were incubated with reaction buffer and caspase-3 substrate Ac-DEVD-AMC for 10 min at 37 °C. Fluorescence was measured using a TECAN Infinite M200 microplate reader with excitation and emission wavelengths of 360 and 440 nm, respectively. FACS analysis using PI and YO-PRO^®-1 (Invitrogen, Cat.No. V13243) was performed on cells plated in 6-well plates at 80,000 cells/well according to manufacturer’s instructions. Fluorescence intensity was measured by flow cytometry (FACScalibur, Fortessa) and evaluated with FlowJo (Tree Star, Inc.). The data were recorded for a total of 15,000 events per sample.

To determine the effect of SAR1 on chronic heart failure, we used a mouse model of MI induced by permanent LAD ligation. Three groups were investigated: (1) a sham control group (receiving only a left thoracotomy) (n = 10); (2) a placebo-treated MI group (n = 15); (3) MI group given the cathepsin A inhibitor SAR1 daily starting one day after operation (100 mg/kg) (n = 10) (Fig. 1). Data for the left ventricular (LV) and body weights were determined 4 weeks after operation on the day of hemodynamic measurements. The placebo group showed a significant increase in LV mass/body mass ratio compared to the sham group (Fig. 2), as it is commonly observed in the permanent LAD ligation model of MI [18]. The increase in LV mass/body mass ratio was not affected by SAR1 treatment (Fig. 2). Infarct sizes ranged between 25 and 30 % among mice and were not significantly different between placebo and SAR1-treated animals.

Next, we assessed how cathepsin A inhibition affects cardiac functionality compared to placebo treatment in the murine MI model. As expected, placebo treated animals developed severely impaired cardiac function after 4 weeks of permanent LAD ligation (Fig. 2). These animals displayed hallmark features of post-MI heart failure, including increased left ventricular end systolic volume (LVESV) together with increased left ventricular end diastolic volume (LVEDV) and—pressure (LVEDP). Consequently, cardiac output (CO) and ejection fraction (EF) were severely decreased. Mean arterial pressure (MAP) and heart rate (HR) were not affected by placebo-treated permanent LAD ligation. In line with limited cardiac functionality, dP/dt values were severely impaired; this is also reflected by an increased Tau-weiss value.

Cathepsin A inhibition dampened the increase of LVESV and LVEDP as compared to placebo treatment (Fig. 2), although not meeting statistical significance (One-way Anova test). Its effect on LVEDV was less pronounced. Similarly, cathepsin A inhibition (as compared to placebo treatment) displayed a trend towards a protective effect concerning cardiac output (CO), ejection fraction (EF) and stroke volume (SV). However, the difference between placebo-treated and SAR1-treated animals did not reach statistical significance (One-way anova test). The same picture emerges for dP/dt and Tau-weiss values. In an independent experimental setup, we demonstrated dose-dependent improvement of cardiac functionality by SAR1 in a rat model of permanent LAD ligation (Additional file 1).

Plasma brain natriuretic peptide (BNP) is considered a gold standard biomarker in determining the severity and prognosis of heart failure [19, 20]. We probed BNP plasma levels to monitor the progression of LAD ligation-induced cardiac failure (Fig. 2). Placebo treatment yielded elevated BNP titers post—MI. SAR1 treatment significantly rescued this increase and yielded BNP levels that are within the range of sham-operated animals.

In summary, our data suggest that cathepsin A inhibition by SAR1 may contribute to improved cardiac performance in the rodent permanent LAD ligation model systems.

In order to further substantiate the promising results yielded by SAR1 treatment of permanent LAD ligation, we used mass-spectrometry-based quantitative proteomics in order to (1) identify proteome alterations associated with the murine permanent LAD ligation and (2) monitor the effect of cathepsin A inhibition on those proteins that are impacted by permanent LAD ligation in the mouse model.

Similar to the section on cardiac functionality, we compared three groups: (1) a sham control group (receiving only a left thoracotomy); (2) a placebo-treated MI group (MI-placebo); (3) MI group given the cathepsin A inhibitor SAR1 daily starting 1 day after operation (100 mg/kg) (MI-SAR1). Left ventricles were harvested 4 weeks after sham operation or LAD ligation. A triple labeling strategy was employed [21] with a total of four biological replicates (12 mice in total).

In each of the four replicate experiments we identified approximately 1300 proteins (Fig. 3a). 1074 proteins were identified in at least three replicate experiments and we focused on those proteins for further analysis (Additional file 2). As noticed earlier, partially incomplete overlap of proteome coverage is an intrinsic characteristic of mass—spectrometry based proteomics [22]. Protein ratios are expressed as fold-change (Fc) values (log₂ of light/heavy, light/medium and heavy/medium ratios). All replicates showed a near-normal distribution and an average Fc close to zero, indicating that the majority of the proteins are not altered (Fig. 3b).

We employed the “significance analysis of microarrays” (SAM) method [23] to identify proteins with significantly altered abundance in placebo-treated LAD ligation compared to sham operation. At a false discovery rate <1 % for significantly affected proteins, we identified 79 proteins as increased in placebo-treated LAD ligation and 25 proteins as decreased in placebo-treated LAD ligation, in comparison to sham operation alone. These proteins are summarized in the table included in Additional file 2. This “proteomic fingerprint” of permanent LAD ligation comprises known marker proteins of cardiac failure. For example, we found decreased levels of cardiac troponin T, heart fatty acid-binding protein and α-myosin heavy chain as well as increased levels of atrial natriuretic peptide, galectin-3 and several extracellular/matricellular proteins such as periostin, fibulin-1, versican, lumican and osteoglycin (mimican); thereby being in line with earlier reports [24‐26].

Noteworthy, cathepsin A was up-regulated approximately two-fold in placebo-treated permanent LAD ligation compared to sham-treated animals. This increase failed to meet the significance criteria of the SAM method. However, a similar increase of cathepsin A levels was observed in a rat model of chronic heart failure [6]. SAR1 treatment did not change cathepsin A levels in the murine hearts upon LAD ligation (Additional file 3).

Next, we assessed the effect of SAR1 treatment on the proteomic level. As shown by the scatter plot in Fig. 4a, there is a good correlation between proteome alterations observed upon placebo-treated LAD ligation (as compared to sham operation) and proteome alterations observed upon cathepsin A inhibition (as compared to sham operation). Simple linear regression yields a positive slope of 0.8, indicating that the overall proteome rearrangement upon LAD ligation is similar between placebo and SAR1 treatment but that, on average, the extent of quantitative increase or decrease is mildly dampened upon cathepsin A inhibition.

We then focused on those proteins that were initially found to be significantly affected by the placebo-treated LAD ligation (as compared to sham operation). The heatmap in Fig. 4b visualizes the effect of SAR1 on this subgroup. It becomes evident that SAR1 treatment partially restores the ligation-induced alterations for many proteins in this subgroup. For a more detailed view on proteins for which SAR1 rescues, at least in part, a ligation-induced alteration, we concentrated on those proteins that were (1) significantly affected by placebo-treated LAD ligation (as compared to sham operation) and (2) for which SAR1 treatment increased or decreased this effect by more than 50 % (based on averaged protein ratios from the replicate experiments; difference of log₂ ratios (MI-placebo/sham)– (MI-SAR1/sham) >0.58). These criteria were fulfilled by 21 proteins (Fig. 5a, b). In every case, SAR1 treatment dampened the alteration in protein abundance that was caused by LAD ligation (as compared to sham operation). This effect reached statistical significance (p < 0.05, two-sided t-test) for carbonic anhydrase 3, Igγ 2b, fatty acid synthase, and Rab-7a (attenuated increase in abundance, Fig. 5a) as well as cordon-bleu protein-like 1 and perilipin-4 (attenuated decrease in abundance, Fig. 5b).

Carbonic anhydrase 3 (CA3) is one of the most up-regulated proteins upon LAD ligation (~nine-fold increase compared to sham operation). Elevated plasma levels of CA3 in MI have been previously reported [27, 28]. Elevated plasma levels of CA3 are also found in Duchenne muscular dystrophy [29]. These findings suggest that increased levels of CA3 are associated with degenerating or stressed muscle tissue. SAR1 treatment rescued elevated CA3 levels caused by LAD ligation. This was further validated by western blot analysis (Fig. 5c).

FAS is a biosynthetic enzyme responsible for de novo fatty acid synthesis in mammals. FAS was found to be increased in two mechanistically distinct mouse models of heart failure and in the hearts of humans with end stage cardiomyopathy [30]. We find a more than six-fold increase of FAS abundance upon placebo-treated LAD ligation (as compared to sham operation). In line with an increased need for de novo fatty acid synthesis, we also noticed reduced levels of perilipin-4, which is required for intracellular storage of lipid droplets. In both cases, cathepsin A inhibition significantly dampened the altered abundance observed upon placebo-treated LAD ligation (as compared to sham operation, Fig. 5a, b).

Interestingly, we observed that cathepsin A inhibition also attenuates elevated levels of cardiac stress response proteins, such as periostin, troponin T and cytoglobin. It was recently found that cytoglobin expression is strongly up-regulated in the hypoxia-induced hypertrophic heart [31]. We corroborated the proteomic results observed for periostin, troponin T and cytoglobin by immunoblot analysis (Fig. 5c; Additional file 4).

In summary, our proteomic results indicate that SAR1 treatment reduces the extent of cardiac proteome rearrangement in permanent LAD ligation.

The rat cardiomyocyte cell line H9C2 has been previously used in cell models of heart ischemia and ischemia–reperfusion [32, 33]. In order to probe a putative cardio-protective effect of SAR1 in a cell-autonomous system, we subjected H9C2 cells to culture conditions of simulated ischemia, which consist of serum deprivation and severe hypoxia. After 24 h, we assessed caspase-3 activity as a marker of apoptosis. SAR1 treatment significantly reduced the induction of caspase-3 activity elicited by simulated ischemia (Fig. 6a). Flow-cytometry of cells stained for the apoptotic marker YO-PRO-1 and the necrosis marker PI showed that SAR1 treatment during simulated ischemia significantly reduced the percentage of cells undergoing apoptosis and, to a lesser extent, necrosis (Fig. 6b). Our findings suggest that cathepsin A inhibition attenuates ischemia-induced apoptosis in H9C2 cells. Additionally, we excluded a hypothetical effect of SAR1 on angiogenesis by a matrigel capillary-like sprouting assay using HUVECs treated with SAR1 for 16 h (Additional file 5).