Animals
All animal experiments were conducted in accordance with guidelines from the Directive 2010/63/EU of the European Parliament on animal experimentation in compliance with European guidelines and International Laws and Policies (EC Council Directive 86/609, OJL 34, 12 December 1987) and were approved by the ICGEB Animal Welfare Board, the Ethical Committee and the Italian Ministry of Health (authorization number 1303/2015-PR) and by Institutional Animal Care Committee of Medical Faculty, University of Zagreb and Ministry of Agriculture, Republic of Croatia (authorization number 525-10/0255-15-6). Animals were housed in ventilated cages and subjected to a light dark cycle. Temperature was kept at 20–22 °C and relative humidity at 45–65%. Adult Sprague-Dawley rats, C57BL6 and αSMA-RFP/COLL-EGFP mice53 were used for the in vivo procedures. Neonatal Sprague-Dawley rats and αSMA-RFP/COLL-EGFP53 mice were used to obtain primary fibroblasts and cardiomyocytes for in vitro studies.
BMP1.3 monoclonal antibody
Purified BMP1.3 peptide (amino acids 972-986; RYTSTKFQDTLHSRK) was used as immunogen for the production of the monoclonal antibody (mAb), using the standard hybridoma technique25. The specific monoclonal antibody cell line was produced by Promab Biotechnologies, Richmond, CA.
For the neutralization assay, recombinant dentin matrix protein 1 (DMP-1, 1 µg) was incubated with rBMP1.3 (125 ng) in reaction buffer (50 mM Tris-HCl, pH 7.5; 150 mM NaCl; 5 mM CaCl2) with protease inhibitors (Sigma, P-2714) and incubated for 20 h at 37 °C. The reaction was stopped by adding 4X Lithium Dodecyl Sulfate (LDS) buffer containing β-mercaptoethanol and heating samples at 96 °C for 7 min. The BMP1.3 antibody (5 µg) was added to rBMP1.3 in reaction buffer and incubated for 3 h at 37 °C and then mixed with DMP-1, as before. All samples were resolved by SDS-PAGE and stained with Coomassie brilliant blue stain.
BMP1.3 ELISA
Human plasma samples were obtained from eight patients of both genders who had a recent MI and eight aged/gender matched healthy donors without comorbidities. Among AMI patients, 6 had hyperlipidemia and others did not have significant comorbidities. All participants were older than 40 years. Human samples were collected in compliance with national ethical regulations, upon approval by the Ethics Committee of the School of Medicine University of Zagreb (EC KBC 5367001) and informed consent by all patients. Participants did not receive any compensation.
Plasma levels of BMP1.3 were measured using the ELISA Human BMP1 Matched Antibody Pair Kit (Cloud-Clone PSA653Hu01) and additional reagents supplied in Antibody Pairs Support Pack (Cloud-Clone IS077). Both capture and detection antibodies were rabbit polyclonal antibodies raised to an immunogen sequence Glu610-Ser843, specific for the BMP1.3 isoform. Microtiter 96-well plates (Immulon-HB) were coated with 100 µl/well capture antibody (final concentration 4 µg/mL) and incubated overnight at 4 °C. Microtiter plates were then blocked with provided blocking buffer for 90 min at 37 °C. Recombinant hBMP1.3 (Cloud-Clone) was used as a standard for the calibration curve. Plasma samples were incubated for 1 h at 37 °C, followed by the addition of biotinylated detection antibody (final concentration 1 µg/mL) for 1 additional hour. HRP-streptavidin (Cloud-Clone) was diluted 1:100 and incubated 30 min at 37 °C. The reaction was visualized by the addition of the chromogenic substrate tetramethylbenzidine and stopped with 1 M H2SO4. Plates were washed with provided wash buffer after each step. All samples were analyzed in duplicate. Absorbance was measured with a microplate reader (Biotek, EL808B) at 450 nm.
Plasma levels of mouse BMP1.3 were measured using the Mouse BMP-1 ELISA kit (Novus Biologicals) which detects specifically BMP1.3 isoform. Plasma samples were obtained from mice 24 h after MI and from sham-operated animals.
BMP1.3 antibody pharmacokinetics
Single-dose pharmacokinetics of the monoclonal mouse anti-BMP1.3 antibody was performed after either intravenous (i.v.) or intraperitoneal (i.p.) injection in 2 months old female Sprague-Dawley rats weighing 300 g (n = 3 per administration route) at a dose of 50 µg/kg (injection volume 0.25 mL i.v., 0.5 mL i.p.). Blood samples were taken from tail vein immediately before and at 5, 15, 60, 180, and 360 min and then at 24, 48, 72, 144, 168, 192, 240, 312, and 384 h post-injection. Serum anti-BMP1.3 antibody concentrations were measured by indirect ELISA method with the injected mouse monoclonal antibody used as a standard for the calibration curves (concentration range 10 pg/mL to 10 µg/mL, diluted in 1% BSA in PBS and in rat serum pool). Microtiter plates were coated overnight at 4 °C with 150 ng/mL rhBMP1.3 diluted in carbonate buffer (15 mM Na2CO3, 35 mM NaHCO3 buffer prepared without azide, pH 9.8) and then blocked with 5% sucrose and 1% BSA in PBS for 1 h at room temperature. Undiluted serum samples were incubated for 2 h at room temperature followed by the addition of biotinylated goat anti-rat IgG (R&D Systems), diluted 1:5000 as recommended by the manufacturer, for additional 2 h. Streptavidin-HRP (R&D Systems, 890803) was diluted 1:200 in 1% BSA in PBS and incubated for 20 min at room temperature. The reaction was visualized by the addition of chromogenic substrate (tetramethylbenzidine, TMB) and stopped with 1 M H2SO4. Plates were washed three times with PBS containing 0.05% Tween 20 (PBS-T) (pH 7.4) after each step. Absorbance was read at 450 nm using Absorbance Microplate Reader ELx 808TM. All serum samples were analyzed in duplicate.
Rat and mouse models of cardiac fibrosis
Cardiac injury was induced by systemic administration of isoproterenol in rats and by ligation of the LAD coronary artery in mice.
For rat experiments, 2 months old Sprague-Dawley animals, weighting ~280 g were used. Control and treated groups were injected subcutaneously with isoproterenol hydrochloride (Sigma Aldrich, Saint Louis, MO) at 85 mg/kg/day diluted in 2 mL of saline on two consecutive days with an interval of 24 h. Three experimental groups were established as follows: (1) sham—2 mL of saline on two consecutive days subcutaneously; (2) control—isoproterenol i.p.; (3) isoproterenol i.p. with anti-BMP1.3 antibody at 150 µg/kg, according to the protocol shown in Supplementary Fig. 2a. In a separate set of animals, the LAD was ligated in rats (n = 20) and the animals were treated with 150 µg/kg anti-BMP1.3 antibody according to the standard protocol. After 4 weeks rats were euthanized, and scars were dissected under microscope (Olympus BX53) and loupe (Olympus SZX10) and frozen at −80 °C for analysis of collagen crosslinking.
Mouse MI was induced by ligation of the LAD. Two months old, either CD-1 or αSMA-RFP/COLL-EGFP mice, weighing ~27 g, were used in these experiments. Mice were anesthetized with an i.p. injection of Xylazine 0.6 ml/kg and Ketamine (0.8 ml/kg). After left side thoracotomy, ligation of the LAD was performed with a 6–0 ethilon suture, 1–2 mm below the tip of the left auricle. The chest cavity and the skin were closed with 4–0 ethilon suture. Mice were intubated and ventilated during the procedure. Mice were randomized to four experimental groups: (1) sham—thoracotomy without LAD ligation; (2) control—MI without treatment; (3) MI treated with anti-BMP1.3 antibody (150 µg/kg); (4) MI treated with anti-BMP1.3 antibody (500 µg/kg). Therapy was administered as shown in Supplementary Fig. 2a. To evaluate the cardioprotective effect of anti-BMP1.3 antibody, mice were sacrificed at 2 days after LAD. To compare the efficacy of the anti-BMP1.3 antibody with alternative anti-fibrotic treatments, a new set of MI was performed in 16 mice, which were randomized to receive (1) anti-BMP1.3 antibody (150 µg/kg); (2) anti-TGFβ1 antibody (BioXCell, BE0057, 2 mg/kg)54; (3) SB-431542 (Sigma Aldrich, Saint Louis, MO, 10 mg/kg)55, a chemical inhibitor specifically blocking the TGFβ receptors Alk4, Alk5, and Alk7, and (4) IgG1 antibody (150 µg/kg) as isotype control.
Troponin T plasma levels
Myocardial damage and necrosis were evaluated by measuring plasmatic levels of Troponin T on days 0, 1, 2, 3, 7 and 10, using a commercially available kit (Roche). Blood samples were drawn from the orbital plexus and collected in heparinized tubes. Samples were promptly centrifuged at 2000 × g for 15 min prior to analysis.
Primary culture of cardiomyocytes and cardiac fibroblasts
Primary cardiomyocytes and fibroblasts were isolated as previously described39. Primary cell cultures were exposed to reduced oxygen levels (2%) for 48 h in an hypoxic chamber (Ruskinn IN VIVO2 200) and treated with either 5 µg/ml anti-BMP1.3 antibody or 0.1 µg/ml rBMP1.3. Alternatively, supernatants enriched with either BMP2 and/or BMP5 were produced as previously described39, using plasmids encoding for mouse Bmp2 and Bmp5. Enriched supernatants were added to the culture media (1:1 ratio with fresh medium) together, with anti-BMP1.3 antibody or rBMP1.3. Silencing of Bmp1.3, Bmp2 and Bmp5 in cardiomyocytes and fibroblasts was performed the day of plating using Lipofectamine RNAiMax (Thermofisher), following manufacturer instruction. A pool of siRNAs for Bmp2 (siGenome Rat Bmp2 siRNA smartpool, Dharmacon, gene ref. 29373) and Bmp5 (siGenome Rat Bmp2 siRNA smartpool, Dharmacon, gene ref. 315824) were used, while a specific siRNA for Bmp1.3 (antisense strand 5′ UGU ACC GCA GGU GAA AGC CUU 3′) was custom designed and synthetized by Eurofin genomics.
Cells were fixed with 4% paraformaldehyde (PFA) in PBS at the end of the experiment and stained.
Isolation of different cell types from ischemic mouse hearts
Different cell types were purified from the heart 2 days after LAD ligation. Cardiomyocytes were isolated as described56. For endothelial cells, inflammatory cells and fibroblasts, hearts were digested to single cell suspension using the Skeletal Muscle dissociation kit (Miltenyi Biotech). Endothelial cells were positively selected using anti-CD31 coated magnetic beads (Miltenyi Biotech), following manufacturer instructions. CD31-depleted cells were subsequently incubated with anti-CD45 magnetic beads (Miltenyi Biotech) to separate inflammatory cells from cardiac fibroblasts. Purified cells were pelleted at 4 °C at 300 × g for 5 min for RNA extraction.
Echocardiography
Echocardiography was performed on sham, control and anti-BMP1.3 antibody-treated rats and mice at days 7, 14, 30, and 60 using a VisualSonics Vevo 2100 ultrasound device with 55 MHz MicroScan transducer (in mice) and at day 30 using a 25 MHz MicroScan transducer (in rats). Mice were anesthetized with 1% isoflurane during the imaging. Parasternal short-axis and long-axis B-mode and M-mode views were recorded, as recommended by recent guidelines57.
Left ventricular systolic function was assessed by ejection fraction, according to the formula: EF (%) = [(EDv − ESv)/EDv] × 100. Echocardiography was performed by two independent educated observers and mean value was taken.
Gene expression analysis
Total RNA from cultured cells and hearts was isolated using TRIzol (Invitrogen), following manufacturer instructions. Tissues were digested with tissue lysis buffer (20 mM Tris-HCl (pH 8.0), 1 mM CaCl2, 0.1% sodium dodecyl sulfate, and 200 μg/ml proteinase K) and incubated at 55 °C for 12 h with shaking. cDNA was synthesized and amplified from 1 μg of total RNA using Super Script III First-Strand Synthesis System (Invitrogen). Gene expression of interest was measured using a LightCycler FastStart DNA Master SYBR Green kit in a LightCycler instrument (Roche Diagnostics), as described58. Values for specific genes were normalized to Gapdh, 18S or β actin housekeeping controls. Results are represented as a fold change of the comparative expression level. The list of primers used for gene expression analysis is shown in Supplementary Table 3.
Histology and immunostaining
Heart samples were fixed in formalin solution and embedded in paraffin. For Sirius Red staining, 4 μm sections were stained with hematoxylin-eosin and observed under light microscopy with camera. Sirius Red staining was quantified using S-form software.
Alternatively, hearts were cut on their sagittal plane and freshly frozen in liquid nitrogen-cooled isopentane. Frozen tissue was immersed in cryoprotective embedding medium (Killik, Bio Optica) and cut in 4 μm sections using a Leica cryostat.
Immunohistochemistry was performed after deparaffinization in xylene and rehydration in descending concentrations of ethanol with final incubation in PBS. Heat-induced epitope retrieval was performed in sodium citrate buffer pH 6.0 in a microwave oven for 15 min, followed by cooling for 30 min at room temperature. Either Mouse or Rabbit Specific HRP/AEC IHC Detection Kit—Micro-polymer (Abcam) was used in all procedures. To eliminate endogenous peroxidase activity and non-specific staining, sections were pretreated with Hydrogen Peroxide Block and Protein Block reagents from the same kit. Slides were incubated with mouse anti-BMP2 (1:200, Abcam ab6295), rabbit anti-BMP5 antibody (1:1000, Thermo Fisher Scientific PA5-78878), rabbit anti-TGFβ1 antibody (1:200, Abcam ab92486), mouse anti-Lox (1:100, R&D Systems MAB2639) and rabbit anti phosphorylated SMAD2 (1:100, kind gift of Peter ten Dijke) overnight at 4 °C in a moist chamber. Goat secondary antibody coupled with horseradish peroxidase (HRP) (Mouse and Rabbit Specific HRP/AEC IHC Detection Kit—Micro-polymer; Abcam: ab236467) was used and detected using 3-amino-9-ethylcarbazole (AEC) chromogen or diaminobenzidine (DAB, ImPACT DAB Substrate kit, Vector Laboratories). Slides were counterstained in hematoxylin and mounted using ImmunoHistoMount™ (Sigma-Aldrich, I1161); images were taken using the Olympus BX53 microscope. For immunofluorescence, cell death was evaluated using in Situ Cell Death Detection Kit, TMR red (Roche) on 5 µm thick section as previously described39. Slides were subsequently blocked in 0.1% Triton X-100 (Sigma), 10% horse serum (Gibco) solution in PBS and stained with anti-actinin antibody (1:3000, Abcam ab9465). After three washes in PBS, cells were incubated with an Alexa Fluor 488 anti-mouse secondary antibody diluted 1:500 in blocking solution for 1 h at room temperature. Nuclei were stained with DAPI. Alternatively, apoptosis was evaluated using antibodies against cleaved Caspase 3 (1:100, Cell Signaling 9661 s).
Primary cells were fixed with 4% PFA, permeabilized 10 min at room temperature with a 0.1% Triton X-100 (Sigma) solution in PBS, blocked with 2% BSA (Sigma) 0.2% Tween20 (Sigma) solution in PBS and stained with anti-α-actinin (1:200) in blocking solution. After three washes in PBS, cells were incubated with Alexa Fluor 488 anti-mouse secondary antibody (1:500, Invitrogen A-21202) in blocking solution for 1 h at room temperature. Nuclei were stained with DAPI.
Collagen cross-link analysis
Scar areas of control (n = 6) and treated rats (n = 6) were quantitatively analyzed following mincing and demineralization with EDTA (6 h), reduced with NaB3H4, hydrolyzed with 6 N HCl and subjected to amino acid and cross-link analysis59. Reducible cross-links (dehydro-dihydroxylysinonorleucine (deH-DHLNL)/its ketoamine) and a major aldehyde (hydroxylysine-aldehyde) were analyzed as their reduced forms; i.e. DHLNL, HLNL and DHNL, respectively. The levels of non-reducible cross-links, pyridinoline (Pyr) and deoxypyridinoline (d-Pyr) and reducible cross-links/aldehyde were expressed as moles of cross-link/mole of collagen. Total aldehyde involved in cross-linking was calculated by (DHNL + DHLNL + HLNL + 2xPyr + 2xd-Pyr). Maturation of cross-links was assessed by Pyr + d-Pyr/DHLNL + HLNL.
Western blot
To assess the level of Smad2/3 phosphorylation, fibroblasts were lysed using RIPA buffer (1% NP40, 10 mM Tris-HCl pH 7.4, 5 mM EDTA, 150 mM NaCl, 0.1% SDS, 0.1% sodium deoxycholate, 2 mM sodium orthovanadate, 10 mM sodium fluoride) supplemented with protease and phosphatase inhibitors (Sigma). Samples were denatured by boiling for 5 min at 95° degrees in Laemmli buffer and separated by SDS-Page using a 10% bis-tris acrylamide gel. Proteins were blotted on a nitrocellulose membrane using the semi-dry Power Blotter system (Invitrogen). Membranes were blocked in 5% BSA (for pSMAD2/3) or 5% skimmed milk (for total SMAD2/3) in TBS 0.1% Tween20 (TBST) for 30 min at room temperature, then incubated overnight at 4 °C with rabbit anti-Smad2/3 (Cell Signaling 8685) or rabbit anti-phospho-Smad2 (Ser465/467)/Smad3 (Ser423/425) (Cell Signaling 8828), diluted 1:1000 in their blocking solution. After five washes with TBST, membranes were incubated with goat anti-rabbit HRP-conjugated antibody (DAKO, p0448), diluted in their blocking solution 1:2000 for 1 h at room temperature. After five washes with TBST, membranes were incubated with SuperSignal West Pico PLUS Chemiluminescent Substrate (Thermoscientific) and chemiluminescence was detected using Chemidoc Touch imaging system (Promega).
Luciferase assay
HEK293T cells stably expressing the TGFβ-sensitive (CAGA)12-luciferase reporter (pGL3 CAGA12 LC + PuroR-P2A-Renilla-T2A-EGFP)60 were cultured in serum-free DMEM. After 6 h starvation, 2.5 pg/ml of human recombinant TGFβ1 (Peprotech) was added to the medium, either alone or in combination with 5 µg/ml anti-BMP1.3 antibody or 0.1 µg/ml rBMP1.3. After additional 17 h, cells were lysed with Glo-lysis buffer (Promega) and assayed using the Dual-Glo luciferase kit (Promega) on the Wallac Envision plate reader (Perkin Elmer).
Statistical analysis
Data were collected using Microsoft Excel 2009 and analyzed with SPSS software (iBM SPSS Statistics version 2.2.0., New York, USA) or Graphpad Prism 6.0 (Graphpad software). Values are reported as mean ± S.E.M. Gaussian distribution was assessed and the appropriate Student′s t-test of Mann–Whitney U test was used to compare two groups, with one-way ANOVA followed by the appropriate post hoc test to compare more than two groups. Two-way ANOVA for repeated measures was used for comparisons between groups over time. Sample size was described in each figure legend. P values ≤ 0.05 were considered statistically significant.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
