GDF3 AS BIOMARKER AND BIOTARGET IN POST-ISCHEMIC CARDIAC REMODELING
20230375571 · 2023-11-23
Inventors
Cpc classification
G01N2800/56
PHYSICS
C07K16/22
CHEMISTRY; METALLURGY
International classification
Abstract
Markers of an intense scarring process in the early phase post-myocardial infarction (MI) are still undetermined, and the identification of patients at higher risk of developing large adverse fibrotic remodeling and heart failure remains challenging. Here, the inventors demonstrate the modulation in the paracrine behavior of resident PW1+ cells in scarring cardiac tissue post-MI and the differential abundance of 12 candidate markers in their secretome. Of these, growth differentiation factor 3 (GDF3), a member of transforming growth factor-β family, upregulates proliferation of cardiac fibroblasts, which are instrumental in fibrosis. GDF3 is upregulated in the scarred tissue and plasma of mice and humans post-MI, with the highest plasma levels predicting higher fibrotic cardiac remodeling and cardiac dilation. The inventors thus reveal the previously unidentified function of GDF3 in predicting adverse fibrotic cardiac remodeling post-MF Thus the present invention relates to the use of GDF3 as biomarker and biotarget in post-ischemic cardiac remodeling.
Claims
1. A method of determining whether a patient who experienced a myocardial infarction has or is at risk of having adverse post-ischemic cardiac remodeling and treating the patient, comprising determining the level of GDF3 in a sample obtained from the patient and administering a therapeutically effective amount of a GDF3 inhibitor to a subject identified as having high levels of GDF3.
2. The method of claim 1 wherein the sample is a blood sample, more particularly a serum sample.
3. The method of claim 1 wherein the level of GDF3 is determined 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 days after the myocardial infarction.
4. The method of claim 1 that includes contacting the sample with an agent that selectively binds to the GDF3 protein.
5. The method of claim 4 wherein the agent is an antibody.
6. The method of claim 1 wherein the level of GDF3 is determined by an enzyme linked immunosorbent assay.
7. The method of claim 1 wherein high levels of GDF3 indicate that the probability that the patient has or is at risk of having adverse post-ischemic cardiac remodeling is high and conversely low levels of GDF3 indicate that the probability that the patient has or is at risk of having adverse post-ischemic cardiac remodeling is low.
8. The method of claim 1 that comprises the steps of i) quantifying the level of GDF3 in the sample obtained from the patient ii) comparing the level quantified at step i) with a predetermined reference value and iii) concluding that the patient has or is at risk of having adverse post-ischemic cardiac remodeling when the level quantified at step i) is higher than the predetermined reference value or inversely concluding that the patient does not have or is not at risk of having adverse post-ischemic cardiac remodeling when the content quantified at step i) is lower than the predetermined reference value.
9. A method of treating adverse post-ischemic cardiac remodeling in a patient who experienced a myocardial infarction comprising administering to the subject a therapeutically effective amount of a GDF3 inhibitor.
10. The method of claim 9 wherein the GDF3 inhibitor is an anti-GDF3 neutralizing antibody.
11. The method of claim 10 wherein the anti-GDF3 neutralizing antibody binds to the mature domain of GDF3.
12. The method of claim 11 wherein the anti-GDF3 neutralizing antibody binds to the amino acid sequence that ranges from the amino acid residue at position 251 to the amino acid residue at position 364 in SEQ ID NO:1.
13. The method of claim 4, wherein the agent selectively binds to a mature domain of GDF3 protein.
Description
FIGURES
[0044]
EXAMPLE
Material & Methods
[0045] All procedures and animal care protocols were approved by our institutional research committee (CEEA34 and French ministry of research, N° 2019050221153452) and conformed the animal care guideline in Directive 2010/63/EU European Parliament. All animals received humane care in compliance with the “Principles of Laboratory Animal Care” formulated by the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH Publication No. 86-23, revised 1996).
Animal Study Design
[0046] Left anterior descending artery (LAD) surgery was performed on male 8- or 13-week-old male C57BL/6J and PW1-reporter (PW1.sup.nLacZ) mice, which were anesthetized in an induction chamber with 2% isoflurane mixed with 1.0 L/min 100% O.sub.2 and placed on a supine position on a heating pad to maintain body temperature. The mice were intubated with endotracheal tube and then connected to a rodent ventilator (180 breaths/min and a tidal volume of 200 μL). During surgical procedure, anesthesia was maintained at 1.5-2% isoflurane with O.sub.2. The chest was accessed from the left side through the intercostal space, and the pericardium incised. The LAD was exposed and encircled with an 8.0 prolene suture at the proximal position. The suture was briefly snared to confirm the ligation by blanching the arterial region. Mice were analyzed 7 days after LAD permanent ligation. Blood samples were collected in heparin-coated Eppendorf tubes and immediately centrifuged at 200×g for 15 min at 4° C. to separate the plasma, which was stored at −80° C. until analysis. Hearts were excised and immediately digested for FACS sorting or qPCR analysis.
Cell Isolation and Fluorescence-Activated Cell Sorting (FACS)
[0047] PW1.sup.+CD51.sup.+ cardiac cell sorting was performed as previously described.sup.29. Briefly, small cell suspensions were prepared from total heart upon atria removal from 8-week-old PW1.sup.nLacZ mice. The ventricles were enzymatically digested with collagenase II and dissociated. The following antibodies were used for cell sorting: BUV737-tagged anti-CD31 (1:100 dilution; BD Bioscience), BUV395-tagged anti-TER119 (1:50 dilution, BD Biosciences), phycoerythrin-cyanin7-tagged anti-CD45 (1:500 dilution; eBiosciences). To detect β-gal reporter activity, cells were incubated with the fluorescent substrate 5-dodecanoylaminofluorescein di-β-D-galactopyranoside (C.sub.12FDG) at 37° C. for 1 h. The different populations were gated, analyzed, and sorted on a FACS Aria II cytometer (BD Biosciences).
CyQUANT™ Cell Proliferation Assay
[0048] FACS-sorted PW1.sup.+ and PW1.sup.−(FDG.sup.−) cells were seeded in 24-well plates at a density of 15,000 cells/well and cultured under normal conditions in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS, Sigma) and 1% penicillin and streptomycin (Sigma) for 5 days. The medium was collected and used to incubate serum-starved MEFs (cultured for 24 h under normal conditions and then serum starved for 24 h) for 24 h. The proliferation of MEFs was evaluated using the CyQUANT cell proliferation assay as per the manufacturer's instructions. MEFs incubated with complete medium served as the control.
RNA-Sequencing and Bioinformatics Analysis
[0049] In total, 300 ng of total RNA extracted from freshly isolated cells with SureSelect Strand-Specific RNA kit (Agilent) was used to prepare a library, according to the manufacturer's instructions. The resulting library was quality checked and quantified by peak integration on Bioanalyzer High sensitivity DNA labchip (Agilent). A pool of equal quantity of 12 purified libraries was prepared, and each library was tagged with a different index. The mRNA pool libraries were finally sequenced on Illumina Hiseq 1500 instrument using a rapid flowcell. The pool was loaded on two lanes of the flowcell. A paired-end sequencing of 2×100 bp was performed.
[0050] After discarding reads that did not pass the Illumina filters and trimming low-quality sequenced bases (q<28) using the Cutadapt program.sup.30, we restricted our downstream analyses to reads with lengths greater than 90 bp. Selected reads were mapped to a murine reference transcriptome that was generated by the RSEM package.sup.31 from the full mouse reference genome and the gtf transcript annotations file from ENSEMBL.sup.32. Alignment and estimation of transcripts abundance in each of the 12 processed samples were performed using the RSEM program. Transcripts with abundance counts higher than 10 in more than two samples (N=36,948) were considered as expressed and retained for further analysis. Abundances of transcripts assigned to the same gene were combined together, leading to the profiling of 16,403 genes. Analyses were conducted under the R environment (version 3.2.2).
[0051] Galaxy 15.10 instance was locally installed on a server machine. WolfPsort, TMHMM, and SignalP were obtained from CBS prediction servers (https://services.healthtech.dtu.dk/, accessed Apr. 15, 2020). NLStradamus and PredictNLS were used in parallel to determine nuclear localization signals. Each dataset from RNA-seq, corresponding to a different population, was then processed through a pipeline designed to select sequences containing a signal peptide, no transmembrane segment, no nuclear localization signal, and structural features consistent with active secretion via classical or non-classical secretory pathways.
qPCR Analysis
[0052] RNA was extracted from cardiac cells isolated from MI and SHAM C57BL/6J mice after 7 days from surgery using the RNAqueous Micro Kit (AM1931, Invitrogen), as per the manufacturer's instructions. Then, 500 ng of extracted RNA was subjected to reverse transcription using the SuperScript IV VILO kit (11756050, Invitrogen) as per the manufacturer's instructions. The resulting cDNA was subjected to qPCR using SYBR Select Master Mix (4472908, Applied Biosystems) on Quant Studio 3 Real-Time PCR system (Thermo Fisher) as per the following condition: 95° C. for 10 min, 40 cycles of 95° C. for 15 s and 60° C. for 1 min, followed by 95° C. for 10 s and 60° C. for 1 min. The expression of target gene was analyzed using the −2.sup.ΔΔCT method following normalization to RPL13 expression.
HEK293 Cell Culture and Transfection
[0053] HEK293 cells were cultured at 37° C. in the presence of 5% CO.sub.2 in DMEM GlutaMax™-1 (Life Technologies) supplemented into 10% FBS and 1% penicillin and streptomycin. HEK293 cells were plated at 6×10.sup.5 cells/well in 6 well-plates in a medium without antibiotics. After 24 h, transfection of expression plasmids (Origene and Genscript) was performed with Lipofectamine® 2000 (Life Technologies) according to the manufacturer's protocol using 2 μg of plasmids and 6 μL of Lipofectamine® 2000 diluted in Opti-MEM (Life Technologies). The cells were cultured for 2 days and then serum starved for 8 h prior to the collection of conditioned media and centrifugation at 200×g for 10 min. Supernatants were stored at −80° C. until MEF treatment.
Isolation of Mouse Embryonic Fibroblasts
[0054] Primary MEFs were isolated from 13.5-days post coitus C57bL/6J mouse embryos. The pregnant females were euthanized by cervical dislocation and embryos were surgically excised and separated from maternal tissues and the yolk sac in ice-cold phosphate-buffered saline (PBS). Embryos were then decapitated and eviscerated (removal of the heart, spleen, liver and intestine). The bodies were washed in ice-cold PBS to remove blood before being finely minced in a Petri dish without PBS. Samples were incubated for 15 min at 37° C. in the digestion solution (0.05% trypsin-EDTA solution [Life Technologies], 0.1 mg/mL DNase 1 [Sigma]). The suspension was allowed to settle. The supernatant was drawn off, mixed with MEF culture medium (DMEM 4.5 g/L D-glucose [Life Technologies], 10% FBS, 1% penicillin-streptomycin, 1% non-essential amino acid [Life Technologies]) and centrifuged for 5 min at 200×g. After centrifugation, the pellet containing MEFs was resuspended in MEF culture medium. The pellet from the tissue digestion was resuspended in the digestion solution and incubated for 15 min at 37° C. The cells were allowed to settle, the supernatant was drawn off and processed as previously described. The cells from the first and second digestion steps were pooled and then plated in Petri dishes. Each Petri dish received a volume of cell suspension equivalent to 1.5 embryos.
[0055] After 12 h, the culture medium was changed to remove non-adherent cells and debris. The MEFs were passaged upon reaching 80% confluence. MEFs were harvested by trypsinization, centrifuged, and resuspended in a freezing medium (DMEM 4.5 g/L D-glucose, 1% penicillin-streptomycin, 10% dimethyl sulfoxide). Primary MEFs were cultivated between passage 0 and 4.
Western Blot Analysis
[0056] Proteins were extracted from frozen mouse heart tissues using a Dounce-Potter homogenizer into ice-cold radioimmunoprecipitation assay (RIPA) buffer (50 mM Tris pH 7.4, 150 mM sodium chloride, 1% IGEPAL CA-630, 50 mM deoxycholate, and 0.1% sodium dodecyl sulfate [SDS]) supplemented with 1% anti-proteases (Sigma-Aldrich), 1% anti-phosphatase inhibitors (Phosphatase Inhibitor Cocktail 2 and 3, Sigma-Aldrich), and 1 mM sodium orthovanadate. After 1 h incubation at 4° C., the homogenate was centrifuged for 15 min at 15,300×g and 4° C. and the supernatant containing proteins was collected. Cardiac PW1.sup.+ cells from 22 mice and PW1.sup.− cells from 16 mice were pooled, centrifuged at 500×g for 15 min at 4° C., and lysed in urea-thiourea buffer (5 M urea, 2 M thiourea, 50 mM dithiothreitol [DTT], and 0.1% SDS in PBS, pH 7.4). Proteins were extracted as described above. Protein concentrations for all samples were determined using a Bradford-based protein assay (Bio-Rad).
[0057] After isolation of cardiomyocytes and non-cardiomyocytes from the adult mouse hearts, as previously described.sup.7, the proteins were denatured for 10 min at 70° C. before loading on a NuPAGE™ Novex® 4-12% Bis-Tris gel (Life Technologies). After 3 h electrophoresis, proteins were transferred onto nitrocellulose membranes using the Trans-Blot Turbo Transfer System (Bio-Rad) and stained with 0.1% Ponceau S (w/v in 5% acetic acid) to assess transfer quality and homogeneous loading. Membranes were blocked for 1 h in Tris-buffered saline with 0.1% Tween-20 (TB S-Tween) containing 5% skim milk with constant shaking and then incubated for overnight at 4° C. with primary antibodies specific for GDF3 (1:1000 for tissue and 1:500 for plasma, Abcam) and FLAG (1:1000, Sigma-Aldrich) diluted in 5% skim milk/TB S-Tween. After washing, the membranes were incubated for 1 h at room temperature (23° C.) with horseradish peroxidase (HRP)-conjugated secondary antibodies diluted in 5% skim milk/TB S-Tween. Membranes were then washed and incubated for 5 min with SuperSignal™ West Pico PLUS Chemiluminescent Substrate (Life Technologies) before imaging with the Chemidoc® XRS+ camera (Bio-Rad) and analysis using the Image Lab™ software.
ELISA
[0058] GDF3 levels were measured by GDF3 sandwich ELISA assay (GenWay, GWB-KBBHW6) following the manufacturer's instructions. Briefly, standards and diluted samples (1:16 in standard diluent) were added to an anti-GDF3 microplate (pre-coated plate with an antibody specific for GDF3) and incubated for 1 h at 37° C. After removing standards and samples, a biotinylated GDF3 detector antibody was applied. The plate was incubated for 1 h at 37° C. Wells were washed and then incubated at 37° C. with an avidin-HRP conjugate for 30 min. Finally, after extensive washing, the wells were incubated with the 3,5,3′,5′-tetramethylbenzidine (TMB) substrate for 15 min in the dark at 37° C. The blue color product from the oxidation of TMB substrate changed into yellow after reaction termination with the addition of stop solution and incubation at 37° C. for 15 min. Absorbance at 450 nm was quantitatively proportional to the amount of GDF3 captured in well and measured using microplate reader.
PREGICA Cardiac MRI Sub-Study
[0059] We used banked plasma from 80 patients with a first STEMI and who were enrolled in the prospective PREGICA cardiac MRI sub-study (Predisposition Genetical in Cardiac Insufficiency, clinicaltrials.gov identifier NCT01113268. Details of the study have been described previously (Garcia R, Bouleti C, Sirol M et al. VEGF-A plasma levels are associated with microvascular obstruction in patients with ST-segment elevation myocardial infarction. Int J Cardiol 2019; 291:19-24). Briefly, the study involved 6 cardiology centers in France and enrolled patients between 18 and 80 years old referred for a first STEMI between 2010 and 2017 and seen within the first 24 h after symptoms onset. STEMI was defined by the presence of ST-segment elevation on the ECG, significant rise of troponin (≥3 fold higher than the upper limit reference) and the presence of at least 3 akinetic LV segments on the initial trans-thoracic echocardiography. Patients were not included if they had permanent atrial fibrillation, a diagnosis of previous MI or a history of cardiac disease. All patients had coronary angiography and primary PCI in the first 24 h. Cardiac MRI was performed using a 1.5-T unit at day 4±2 after hospital admission and at 6-month follow-up in a subset of patients defining the CMR substudy of the PREGICA cohort. A standardized MRI protocol was followed in all centers and images were centrally analyzed. Cine images were acquired using a breath-holf steady-state free-precession sequence in long-axis and short-axis views. A stack of short-axis slices covering from the atrioventricular ring to the apex was used to derive left ventricular (LV) volumes, and ejection fraction (EF). Ten minutes after intravenous injection of gadolinium-based contrast agent, late gadolinium enhancement (LGE) images were acquired using a breath-hold segmented T1-weighted inversion-recovery gradient-echo sequence in the same long-axis and short-axis views of cine images. LGE images were assessed for infarct size. Blood samples were drawn at the same time than cardiac MRI. GDF3 was quantified on available plasma drawn at day 4 (n=80) using ELISA assays. The study was approved by the institutional rewiew board, and all patients provided written informed consent.
Statistical Analysis
[0060] Mouse and in vitro studies. The number of samples (n) used in each experiment is recorded in the text and figure legends. All experiments were performed independently at least twice. The data are expressed as mean±standard deviation (SD). Quantitative data were analyzed using one-way analysis of variance (ANOVA, Kruskal-Wallis test) and pair-wise comparisons with Dunnett's post-hoc test for multiple comparisons. The Mann-Whitney U test was used for comparing continuous variables between two groups.
[0061] Analyses of PREGICA cardiac MRI sub-study. P-value were obtained from Chi-Square test statistics for binary variables and by using the Mann-Whitney U test for comparing continuous variables between two groups. The association between GDF3 levels and the likelihood of presenting adverse cardiac remodeling was assessed by linear regression models, with additional adjustment for age and sex.
[0062] All statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, Inc., San Diego, CA). A value of P<0.05 indicated statistical significance.
Results
[0063] PW1.sup.+ cells from ischemic hearts release factors that promote fibroblast proliferation. We subjected adult PW1.sup.nLacZ reporter mice to ischemic cardiac injury by left anterior descending coronary artery (LAD) ligation, as previously described.sup.6,7. Hearts were harvested from mice with MI as well as SHAM-operated mice at day 7 post injury, and PW1.sup.+ cells were isolated by fluorescence-activated cell sorting (FACS) (not shown). Following cultivation for 5 days, the conditioned media from these cells were collected and used to culture mouse embryonic fibroblasts (Miffs) for 24 h (not shown). The effect of the conditioned media on the proliferation of MEFs was evaluated using the CyQUANT™ Cell Proliferation Assay. The result of cell proliferation assay revealed the significant increase in the proliferation of MEFs incubated with the conditioned media from PW1.sup.+ cells isolated from ischemic hearts as compared with those treated with the conditioned media from control cells or PW1.sup.+ cells from SHAM-operated hearts (not shown). There was no significant increase in response to conditioned media from PW1.sup.− cells (not shown). These observations suggest that the activated PW1.sup.+ cells from the ischemic heart release pro-proliferative factors, which may induce the proliferation of resident fibroblasts.
RNA-Sequencing (RNA-Seq) and Bioinformatic Analyses Predict Candidate Biomarkers Involved in the Paracrine Action
[0064] The transcriptome of FACS-isolated PW1.sup.+ cells from SHAM and MI mice was characterized by RNA-seq to investigate the influence of the ischemic heart environment on the paracrine potential of PW1.sup.+ cells. We filtered, aligned, and quality-controlled RNA-seq output files to obtain a list of transcripts showing the greatest signal intensities (not shown). We performed a comparative analysis to understand the disease-induced alterations in the secretory behavior of cardiac PW1.sup.+ cells and shortlisted candidates with more than two-fold higher expression in ischemic conditions than in normal condition (not shown). We then examined the predicted amino acid sequences of the corresponding genes by a series of bioinformatic algorithms to identify secretory proteins. We considered proteins with a predicted N-terminal endoplasmic reticulum (ER)-targeting signal peptide but without predicted transmembrane domains and intracellular localization signals (i.e., no ER retention signal, mitochondrial targeting peptide, or nuclear export signal) (not shown). Progressive filtering rendered a total of 24 secreted proteins overexpressed by cardiac PW1.sup.+ cells under ischemic conditions (not shown). We next confirmed the significant increase in the expression of 12 of these 24 candidates in the ischemic heart (remote or infarct zones) as compared with normal heart by quantitative polymerase chain reaction (qPCR) (not shown).
[0065] In comparison with the secretome of control cells, the secretome of MI-activated PW1.sup.+ cells comprised several growth factors, cytokines, and enzymes as well as some poorly characterized factors (not shown). The secretion of the growth factor GDF3, cytokines such as NDP and CCL8, and enzymes such as CELA1 and PRTN3, were more than two-fold higher in MI hearts than in SHAM hearts (not shown).
Transfection Experiment Confirms the Candidates with Proliferative Effects on Fibroblasts
[0066] We then investigated the effects of candidates on the proliferation of cultured embryonic fibroblasts and adult cardiac fibroblasts. We selected six candidates (CCL8, CELA1, GDF3, NDP, PRNT3, PROK2) that were possibly associated with cell proliferation, as evident from their Gene Ontology biological functions. Reciprocally, we excluded the lipoproteins APOC2, APOC4, and SAA3, the coagulation factor F10, and the poorly characterized C1QTNF3 and DMKN. We separately cloned the cDNAs of these six proteins into mammalian expression plasmids, which were then used to transfect HEK-293 cells (not shown). FLAG epitope-tagged fibroblast growth factor 23 cDNA served as positive control, while empty vector was used as negative control. After 48 h from transfection, the conditioned media were collected, tested to confirm the overexpression of the secreted proteins (not shown), and then used to incubate serum-starved MEFs. Evaluation of cell proliferation rate after 24 h treatment revealed four factors, namely, growth differentiation factor-3 (GDF3), norrin cystine knot growth factor (NDP), prokineticin 2 (PROK2), and chymotrypsin-like elastase family member 1 (CELA1), that significantly induced the proliferation of MEFs as compared with control treatment (not shown). The proliferative effects of three (GDF3, NDP and PROK2) of the four candidates were further confirmed using freshly isolated adult cardiac fibroblasts (not shown). Thus, cell proliferation assays facilitated the selection of three candidates GDF3, NDP, and PROK2 from 12 overexpressed markers.
[0067] Of these three remaining candidates, GDF3 (also known as Vg-related gene 2) combined the largest over-expression in the ischemic hearts and one of the most important increase of fibroblast proliferation. GDF3 is a member of the TGF-(3 superfamily that is composed of 366 amino acid residues. Human and mouse GDF3 show 76.6% nucleotide homology and 69.3% peptide identity.sup.12. The predicted amino acid sequence comprises a signal sequence for secretion at the hydrophobic NH.sub.2-terminus, a prodomain that facilitates cysteine-mediated disulfide bond formation with another family member, and a putative proteolytic processing site at 114 amino acid residue (not shown). The cleavage of GDF3 at this residue generates a mature GDF3 protein, which is 114 residue long.sup.11. GDF3 plays an important role during early development in mice and humans.sup.13, but its expression is low in adult organs and particularly negligible in the adult heart.sup.10,14,15. While the functions and implications of GDF3 in the adult heart are yet unknown, GO biological functions suggest its involvement with the SMAD protein signal transduction pathway that is very relevant to the process of cardiac fibrosis. This is consistent with the highest proliferative effect of GDF3 among all candidates on MEFs (not shown). Our transfection experiment (not shown) confirms that GDF3 is a secreted protein, as indicated by the band of full-length protein on the western blot of supernatants (not shown). Together these observations suggest the potential role of GDF3 in the regulation of fibroblast proliferation in the scarring tissue and encouraged us to investigate the expression profile of GDF3 in MI hearts of mice and humans.
GDF3 Level Increases in the Plasma and Infarcted Area of the Mouse Heart After MI
[0068] Based on the transcriptome results, we sought to evaluate the expression of GDF3 and determine its cellular sources in the whole hearts by western blotting. We detected the mature form of GDF3 in both neonatal and adult normal hearts (not shown). In particular, GFD3 was expressed only in the non-cardiomyocyte fraction and not by cardiomyocytes in the adult heart (not shown). Further analysis of the non-cardiomyocyte fraction confirmed the specific expression of GDF3 in PW1.sup.+ cells but not in the PW1.sup.− cell population of normal hearts (not shown). To investigate any dysregulation in the expression of GDF3 in the mouse heart following MI, we generated a permanent LAD mouse model and excised hearts after 7 days. We separately analyzed GDF3 expression in the infarcted area corresponding to the scar tissue and the remote area. Western blotting confirmed the higher expression of GDF3 in the infarcted area of MI hearts than in the corresponding areas of SHAM hearts (not shown). Consistent with our previous observations and with the fibrogenic fate of cardiac PW1.sup.+ cells in response to MI, this result indicates that GDF3 is produced at the site of infarction and suggests an involvement in the scarring process following MI.
GDF3 is a Circulating Factor Secreted Post-MI
[0069] Considering the secretory nature of this protein, we investigated if free GDF3 could be detected in the circulation by analyzing plasma samples from MI and SHAM mice. Western blot analysis confirmed the presence as well as higher level of mature GDF3 in the plasma of MI mice than in the plasma of SHAM mice (not shown). We performed an enzyme-linked immunosorbent assay (ELISA) specific for GDF3 to investigate kinetic changes over time in the circulating levels of GDF3 in the mouse plasma by ELISA. The secreted protein level increased from day 0 to day 2 and decreased thereafter until day 7 post-MI.
[0070] Overall and in line with the fibrogenic fate of cardiac PW1.sup.+ cells, these results indicate that GDF3 may be a novel cardiokine secreted by these cells that may be related to adverse cardiac remodeling post-MI.
Circulating GDF3 Level Serves as a Marker of Post-MI Adverse Remodeling in Humans
[0071] To investigate the clinical relevance of our findings in the murine MI model, we first assessed GDF3 expression on left ventricular cardiac tissue samples from failing ischemic hearts and non-failing hearts of patients. Western blot analysis revealed the stronger expression of GDF3 in the failing hearts than in the non-failing hearts (
[0072] We then asked whether elevated circulating GDF3 levels can be linked to adverse cardiac remodeling post-MI. We analyzed circulating GDF3 levels in 80 patients with a first acute ST-elevation myocardial infarction (STEMI), seen <24 h after symptom onset and treated by primary percutaneous coronary intervention (PCI) (Predisposition Genetical in Cardiac Insufficiency [PREGICA] patient collection, NCT01113268). Patients had an initial clinical and biological evaluation at day 4 and serial cardiac magnetic resonance imaging (MRI) at 4 days and at 6 months after angioplasty. The details of the inclusion/exclusion criteria are mentioned at https://clinicaltrials.gov/ct2/show/NCT01113268. The baseline characteristics of these patients are shown in Table 1.
[0073] We first defined adverse cardiac remodeling as a >20% increase in left ventricular end-diastolic volume (LVEDV) indexed for body surface area (LVEDVi, ml/m.sup.2) at 6 months as compared to the initial evaluation on cMRI. Patients were accordingly classified as remodelers (n=24) and non-remodelers (n=56). GDF3 measured at day 4 post-PCI was detectable in the plasma of these patients and levels were significantly higher in remodelers than in non-remodelers (1364±521 versus 1090±532 pg/mL, p=0.033) (
[0074] We then classified patients according to this cut-off value (≥1375 pg/mL called high GDF3, n=25 and <1375 pg/mL called low GDF3, n=55). Table 2 reports the main characteristics and cardiac MRI findings at baseline and 6 months after MI in both groups. There was no significant imbalance in the main cardiovascular risk factors between the two groups. The delay between symptom onset and coronary disocclusion was significantly longer (by 1.2 h) in patients with high GDF3 levels (P<0.05). However, the peak of troponin, a surrogate marker of myocardial necrosis, was significantly lower in the high GDF3 group. In terms of cardiac remodeling, patients with high GDF3 showed a non-significant trend of higher cardiac dilation at day 4 post-MI than those with low GDF3. The values of LVEDVi were however significantly higher (P<0.005) and pathological (normal value<82 mL/m2) at 6 months post-MI in patients with high GDF3 (
Discussion
[0075] Acute myocardial infarction characterized by left ventricular remodeling may progress toward development of heart failure.sup.16. Markers reflective of myocardial damage may not predict long-term left ventricular remodeling (troponin and creatine kinase) or suffer from insufficient clinical data (galactin-3 and soluble interleukin-1 receptor-like 1).sup.17,18. For instance, galectin-3 was shown to be involved in fibrosis and inflammation and independently associated with incident peripheral artery disease in an observational study with Whites and Blacks only and without eliminating the effects of confounding factors. Thus, it is imperative to discover potential makers informative of preclinical HF to identify patients with an increased risk of HF and for timely disease management.
[0076] In an attempt to contribute to wound healing, the cardiac ECM undergoes constant remodeling upon injury.sup.19. Interestingly, the concept of ECM regulation through key molecules involved in intercellular communications has only recently emerged. Here, we focused on cardiac PW1.sup.+ cells, a cellular subpopulation that is suspected to orchestrate the reparative process in tissues, including the heart.sup.6,7. We investigated major differences in the secretome of these PW1.sup.+ stromal cells in ischemic mouse hearts. Of note, the pro-proliferative effect observed with conditioned media from cardiac PW1.sup.+ cells isolated from ischemic mouse hearts was not observed with cardiac PW1.sup.− cells nor with cardiac PW1.sup.+ cells isolated from normal mouse hearts. RNA-sequencing and bioinformatic analyses confirmed the upregulation in the expression of several factors in MI hearts, (12 secreted factors by MI-activated cardiac PW1.sup.+ cells), particularly GDF3, PROK2, NDP, which were confirmed to exhibit about seven-fold, three-fold, and three-fold expression upregulation following MI in qPCR validation experiment (not shown). Moreover, the 12 dysregulated candidate markers confirmed by qPCR validation are mostly enriched in GO biological processes such as angiogenesis, inflammation, chemotaxis, and proliferation, thus confirming the crucial response of PW1.sup.+ cell population to MI.
[0077] MI is characterized with an acute inflammatory response involved in myocardial repair.sup.20; however, uncontrolled chronic inflammation causes excessive damage and fibrosis, eventually leading to the loss of cardiac function.sup.21. Cardiac inflammation and endothelial dysregulation are related to the remodeling of the extracellular matrix (ECM).sup.22, and TGF-β pathways have been consistently highlighted as the key molecular mediators of cardiac fibrosis.sup.23,24.
[0078] As a member of TGF-β superfamily, GDF3 was initially shown to participate in early embryonic development, muscular development, adipose tissue homeostasis, and energy balance through its interaction with activin receptor-like kinase type I receptor B (ACVR1B, ALK4) and ACVR1C (ALK7) receptors.sup.25. Recent studies have highlighted the critical role of GDF3 in macrophage function and inflammation cascade. Wang et al. recently described the role of GDF3 in macrophage polarization and endotoxin/sepsis-induced cardiac injury.sup.26. In the present study, we identified the previously unrecognized function of GDF3 in cardiac fibrosis and demonstrate the dynamic changes in GDF3 levels in the blood and hearts of mice and humans following MI.
[0079] This is the first report to investigate the prognostic potential of GDF3 in a cohort of post-MI patients. Interest in GDF3 as a marker of adverse cardiac remodeling arose from our pre-clinical study in a murine model of MI. We observed a seven-fold increase in the mRNA expression of GDF3 and about a two-fold increase in the circulating level of GDF3 in the mouse heart within 7 days from MI. These results were replicated in clinical samples derived from patients with MI. Thus, our results confirm the transient increase in GDF3 levels in murine MI model and highlight the novel pivotal role of this marker as a paracrine factor secreted by cardiac PW1.sup.+ cells in the process of regulation of the properties of the scar tissue and cardiac fibroblasts following MI. Therefore, circulating GDF3 level may be considered while gauging the risk of adverse outcome in patients after MI.
[0080] In our previous report, we showed reduction in TGF-β activation in vitro and cardiac fibrosis post-MI in vivo following pharmacological blockade of αV-integrin on activated cardiac PW1+ cells.sup.7. This observation and the involvement of GDF3 in TGF-β signaling.sup.11 prompt the contribution of GDF3 to adverse cardiac remodeling post-MI. We speculate the participation of GDF3 in the inflammatory cascade during/post MI and support the concept of early intervention of GDF3 functions in the inflammatory cascade to prevent the myocardial damage. Risk stratification at an early stage after MI is challenging yet useful to tailor personalized treatment regimen in the future. Thus, our study paves a strong foundation for future studies directed to target GDF3 in the treatment of MI, supported by the association we found between circulating GDF3 levels, post-MI scarring processes, and cardiac functions.
[0081] Our clinical study is limited by the small sample size, as we specifically focused on post-MI patients with cardiac MRI evaluation for cardiac remodeling, an investigation that is not routinely performed in these patients. Our results indicate that patients with high GDF3 levels develop adverse cardiac remodeling based on imaging surrogates, but further studies would be required to validate its prediction on heart failure and cardiovascular outcomes. However, a large proportion of patients with high GDF3 levels had a significantly decreased LVEF (<50%) 6 months after MI. Further, while we only focused on the potential implications of GDF3, the role of other upregulated markers, particularly PROK2 and NDP, was not investigated and warrant future studies. Notably, we cannot exclude a potential synergy between these other secreted factors. Lastly, this study investigated the role of secreted factors on fibroblast proliferation as a primary mechanism underlying cardiac fibrosis. However, other mechanisms support the fibrotic transformation of the ischemic heart such as myofibroblast transformation.sup.27 and immune-inflammatory response.sup.28. The impact of GDF3 on these mechanisms merits further investigations.
Conclusion
[0082] We show here that the upregulation of GDF3, a secreted protein, is detected in the plasma of mice and humans following MI. We interpret that the levels of circulating GDF3 correlates with the local cardiac production in response to MI and that higher GDF3 circulating levels would indicate higher proliferation of fibroblasts and higher fibrogenesis. Concordantly, we show that high levels of plasma GDF3 levels in a cohort of post-MI patients 4 days following MI corresponded with adverse outcomes measured 6 months later including cardiac dilation, limited recovery of contractile function and higher number of akinetic segments. These data suggest that higher circulating GDF3 levels can be used to identify patients who will develop adverse cardiac remodeling.
[0083] In conclusion, PW1.sup.+ cells from the ischemic heart release pro-proliferative factors, which induce proliferation of resident fibroblasts. One such factor, GDF3 may serve as a novel marker of adverse fibrotic remodeling in the heart tissue following MI. Its applicability in the clinical setting may allow for the identification of patients that have an increased risk of severe myocardial fibrosis and HF as well as better and more specific disease management.
TABLES
[0084]
TABLE-US-00002 TABLE 1 Baseline characteristics of patients Characteristics Mean ± SD or N (%) Age (years) 57.6 ± 10.3 Male (%) 67 (83%) Body weight (kg) 76.4 ± 13.5 Height (cm) 171 ± 8 Body mass index (kg/m.sup.2) 25.9 ± 4.1 Body surface area (body formula) 1.92 ± 0.20 Pre-existing hypertension 27 (33.5%) T2DM 15 (18.7%) Dyslipidemia 47 (58.7%) Delay between symptoms and PCI (hours) 9.84 ± 3.36 LMCA as a culprit 52 (65%) T2DM, type 2 diabetes; LMCA, left main coronary artery; PCI, percutaneous coronary intervention; SD, standard deviation
TABLE-US-00003 TABLE 2 Main characteristics and cardiac RMI findings at baseline and 6 months after MI in both groups Patients Patients with with >1375 Characteristics GDF3 <1375 pg/mL pg/mL P value Age (years) 58.6 ± 8.6 55.5 ± 13.2 0.17 Males, n (%) 44 (80%) 23 (92%) 0.31 Body weight (kg) 74.5 ± 13.2 79.9 ± 13.6 0.08 Height (cm) 170 ± 7 173 ± 8 0.26 BMI 25.7 ± 4.2 26.4 ± 3.8 0.21 Pre-existing 21 (38.9%) 6 (24%) 0.06 hypertension T2DM 8 (14.5%) 7 (28%) 0.15 Dyslipidemia 36 (65%) 11 (44%) 0.07 Active smokers Delay between 9.6 ± 3.4 10.8 ± 3.4 0.049 symptoms and PCI (hours) LMCA as a culprit 35 (63.6%) 17 (68.0%) 0.13 Biology at day 4 Troponin peak 115 ± 117 66 ± 90 0.0052 C-reactive 26.8 ± 40.2 35.9 ± 64.9 0.55 protein (CRP) Cardiac parameters (MRI) at day 4 LVEDVi (mL/m.sup.2) 82.4 ± 16.6 91.8 ± 16.1 0.052 LVESVi (mL/m.sup.2) 43.9 ± 13.2 52.4 ± 12.9 0.0097 LVEF (%) 47.3 ± 8.4 43.3 ± 7.4 0.0296 Number of akinetic 4.66 ± 2.27 5.64 ± 2.15 0.035 segments Cardiac parameters (MRI) at Month 6 LVEDVi (mL/m.sup.2) 89.3 ± 20.2 103.0 ± 21.3 0.0065 LVESVi (mL/m.sup.2) 44.3 ± 17.1 56.5 ± 17.1 0.0032 LVEF (%) 51.3 ± 10.0 45.8 ± 7.1 0.0129 Infarct size (%) .sup. 26.3 ± 13.2% 31.2 ± 13.0 0.08 Number of akinetic 2.80 ± 2.51 4.52 ± 2.47 0.0037 segments Apical infarction 32 (58.1%) 19 (76%) 0.12 Cardiac parameters (Echo) at day 4 LVEDVi (mL/m.sup.2) 51.3 ± 11.8 55.6 ± 9.0 0.11 LVESVi (mL/m.sup.2) 26.7 ± 7.8 31.0 ± 7.5 0.031 LVEF (%) 48.1 ± 7.6 44.4 ± 7.4 0.0223 Cardiac parameters (Echo) at Month 6 LVEDVi (mL/m.sup.2) 58.7 ± 13.9 67.4 ± 14.2 0.0188 LVESVi (mL/m.sup.2) 28.8 ± 10.8 34.5 ± 10.4 0.0205 LVEF (%) 52.2 ± 8.1 49.3 ± 8.7 0.14 T2DM, type 2 diabetes; LMCA, left main coronary artery; PCI, percutaneous coronary intervention; SD, standard deviation; BMI, body mass index; LVEDV, left ventricular end-diastolic volume; LVESV, left ventricular end-systolic volume; LVEF, left ventricular ejection fraction.
[0085] Statistical analysis was performed with Mann-Whitney non-parametric i-test for continuous variables and Chi-square (Fisher if n<5) for binary variables. P<0.05
REFERENCES
[0086] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure. [0087] 1. Frangogiannis, N. G. Regulation of the inflammatory response in cardiac repair. Circ Res 110, 159-173 (2012). [0088] 2. Gulati, A., et al. Association of fibrosis with mortality and sudden cardiac death in patients with nonischemic dilated cardiomyopathy. JAMA 309, 896-908 (2013). [0089] 3. Thygesen, K., et al. Third universal definition of myocardial infarction. Circulation 126, 2020-2035 (2012). [0090] 4. Scardovi, A. B., et al. Brain natriuretic peptide plasma level is a reliable indicator of advanced diastolic dysfunction in patients with chronic heart failure. Eur J Echocardiogr 8, 30-36 (2007). [0091] 5. Hinderer, S. & Schenke-Layland, K. Cardiac fibrosis—A short review of causes and therapeutic strategies. Adv Drug Deliv Rev 146, 77-82 (2019). [0092] 6. Yaniz-Galende, E., et al. Fibrogenic Potential of PW1/Peg3 Expressing Cardiac Stem Cells. J Am Coll Cardiol 70, 728-741 (2017). [0093] 7. Bouvet, M., et al. Anti-integrin alphav therapy improves cardiac fibrosis after myocardial infarction by blunting cardiac PW1(+) stromal cells. Sci Rep 10, 11404 (2020). [0094] 8. Yokota, T., et al. Type V Collagen in Scar Tissue Regulates the Size of Scar after Heart Injury. Cell (2020). [0095] 9. Li, Q., et al. The conditioned medium from a stable human GDF3-expressing CHO cell line, induces the differentiation of PC12 cells. Mol Cell Biochem 359, 115-123 (2012). [0096] 10. The hHuman Protein Atlas. https://www.proteinatlas.org/ENSG00000184344-GDF3/tissue. Accessed July 16. [0097] 11. McPherron, A. C. & Lee, S. J. GDF-3 and GDF-9: two new members of the transforming growth factor-beta superfamily containing a novel pattern of cysteines. J Biol Chem 268, 3444-3449 (1993). [0098] 12. Clark, A. T., et al. Human STELLAR, NANOG, and GDF3 genes are expressed in pluripotent cells and map to chromosome 12p13, a hotspot for teratocarcinoma. Stem Cells 22, 169-179 (2004). [0099] 13. Levine, A. J. & Brivanlou, A. H. GDF3, a BMP inhibitor, regulates cell fate in stem cells and early embryos. Development 133, 209-216 (2006). [0100] 14. Mouse Genome Informatics. http://www.informatics.jax.org/gxd/phenogrid/MGI:95686. Accessed, J., 2020. [0101] 15. BGEE. Gene Expression Data in Animals. https://bgee.org/?page=gene&gene_id=ENSG00000184344. Accessed July 17. [0102] 16. Sutton, M. G. & Sharpe, N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation 101, 2981-2988 (2000). [0103] 17. Ponikowski, P., et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 18, 891-975 (2016). [0104] 18. Tyminska, A., et al. Association of Galectin-3 and Soluble ST2, and Their Changes, with Echocardiographic Parameters and Development of Heart Failure after ST-Segment Elevation Myocardial Infarction. Dis Markers 2019, 9529053 (2019). [0105] 19. Schellings, M. W., et al. Absence of SPARC results in increased cardiac rupture and dysfunction after acute myocardial infarction. J Exp Med 206, 113-123 (2009). [0106] 20. Frangogiannis, N. G. The inflammatory response in myocardial injury, repair, and remodelling. Nat Rev Cardiol 11, 255-265 (2014). [0107] 21. Fontes, J. A., Rose, N. R. & Cihakova, D. The varying faces of IL-6: From cardiac protection to cardiac failure. Cytokine 74, 62-68 (2015). [0108] 22. Franssen, C., et al. Myocardial Microvascular Inflammatory Endothelial Activation in Heart Failure With Preserved Ejection Fraction. JACC Heart Fail 4, 312-324 (2016). [0109] 23. Khalil, H., et al. Fibroblast-specific TGF-beta-Smad2/3 signaling underlies cardiac fibrosis. J Clin Invest 127, 3770-3783 (2017). [0110] 24. Tschope, C. & Diez, J. Myocardial fibrosis as a matter of cell differentiation: opportunities for new antifibrotic strategies. Eur Heart J 40, 979-981 (2019). [0111] 25. Andersson, O., Korach-Andre, M., Reissmann, E., Ibáñez, C. F., & Bertolino, P. Growth/differentiation factor 3 signals through ALK7 and regulates accumulation of adipose tissue and diet-induced obesity. Proc Natl Acad Sci USA. 105, 7252-7256 (2008). [0112] 26. Wang L., et al. GDF3 Protects Mice against Sepsis-Induced Cardiac Dysfunction and Mortality by Suppression of Macrophage Pro-Inflammatory Phenotype. Cells 9, 120 (2020). [0113] 27. Weber, K. T., Sun, Y., Bhattacharya, S. K., Ahokas, R. A., Gerling, I. C. Myofibroblast-mediated mechanisms of pathological remodelling of the heart. Nat Rev Cardiol 10, 15-26 (2013). [0114] 28. Kino, T., Khan, M., Mohsin, S. The Regulatory Role of T Cell Responses in Cardiac Remodeling Following Myocardial Infarction. Int J Mol Sci. 21, 5013 (2020). [0115] 29. ANTI-INTEGRIN av THERAPY IMPROVES CARDIAC FIBROSIS AFTER MYOCARDIAL INFARCTION BY BLUNTING CARDIAC PW1+ STROMAL CELLS. Scientific Reports (2020). [0116] 30. Chen, C., Khaleel, S. S., Huang, H. & Wu, C. H. Software for pre-processing Illumina next-generation sequencing short read sequences. Source Code Biol Med 9, 8 (2014). [0117] 31. Li, B. & Dewey, C. N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12, 323 (2011). [0118] 32. Yates, A., et al. Ensembl 2016. Nucleic Acids Res 44, D710-716 (2016).