Early activation of the cardiac CX3CL1/CX3CR1 axis delays β-adrenergic-induced heart failure

We recently highlighted a novel potential protective paracrine role of cardiac myeloid CD11b/c cells improving resistance of adult hypertrophied cardiomyocytes to oxidative stress and potentially delaying evolution towards heart failure (HF) in response to early β-adrenergic stimulation. Here we characterized macrophages (Mφ) in hearts early infused with isoproterenol as compared to control and failing hearts and evaluated the role of upregulated CX3CL1 in cardiac remodeling. Flow cytometry, immunohistology and Mφ-depletion experiments evidenced a transient increase in Mφ number in isoproterenol-infused hearts, proportional to early concentric hypertrophy (ECH) remodeling and limiting HF. Combining transcriptomic and secretomic approaches we characterized Mφ-enriched CD45+ cells from ECH hearts as CX3CL1- and TNFα-secreting cells. In-vivo experiments, using intramyocardial injection in ECH hearts of either Cx3cl1 or Cx3cr1 siRNA, or Cx3cr1−/− knockout mice, identified the CX3CL1/CX3CR1 axis as a protective pathway delaying transition to HF. In-vitro results showed that CX3CL1 not only enhanced ECH Mφ proliferation and expansion but also supported adult cardiomyocyte hypertrophy via a synergistic action with TNFα. Our data underscore the in-vivo transient protective role of the CX3CL1/CX3CR1 axis in ECH remodeling and suggest the participation of CX3CL1-secreting Mφ and their crosstalk with CX3CR1-expressing cardiomyocytes to delay HF.


In-vivo intramyocardial ultrasound-guided transthoracic siRNA delivery in mice
SMART pool On-target plus Scramble or Cx3cl1 or Cx3cr1 siRNA (Dharmacon, Cambridge, UK) were injected by ultrasound-guided transthoracic intramyocardial injection (see Figure S9) in mice (solution 2.5µM in lipofectamine/optimem, according to the manufacturer instructions, 80µl/mice), as described in Keck et al. 2 , at day 7 after iso pump implantation. Echocardiographic parameters were measured regularly as stated.

Measurement of cardiac parameters
Echocardiography was performed on lightly anesthetized animals under isoflurane (0.2-0.5 %) with a probe emitting ultrasounds from 9-to 14-MHz frequency (Vivid7 PRO apparatus; GE Medical System Co), as previously reported 2

Cardiomyocyte isolation and culture
Cardiomyocytes were isolated from adult mice using a simplified Langendorff-free method, introducing ex vivo optimized dissociation buffers to the heart by direct needle injection into the left ventricle, as previously reported 2 . After cutting of the descending aorta and inferior vena cava, the heart is immediately flushed with cold EDTA buffer (130 mM NaCl, 5 mM KCl, 0.5 mM NaH2PO4, 10 mM HEPES, 10 mM glucose, 10 mM 2,3-butanedione monoxime (BDM), 10 mM taurine, 5 mM EDTA pH 7.8), by injection in the right ventricle.
After clamping of the emerging aorta, the heart is transferred to 60mm dishes for subsequent injection and digestion steps. Deep myocardial perfusion via the coronary vasculature is induced by injection in the left ventricle, first with EDTA buffer for 6 min, then with a Perfusion buffer for 2 min (130 mM NaCl, 5 mM KCl, 0.5 mM NaH2PO4, 10 mM HEPES, 10 mM glucose, 10 mM BDM, 10 mM taurine, 1 mM MgCl2 pH 7.8). Digestion is performed in a collagenase buffer for 30 min at 37°C using an increasing flow rate from 0.5 to 2.5 ml/min (2.5 mg/mL collagenase 2 (Worthington) dissolved in perfusion buffer). A Stop buffer is made with perfusion buffer containing 5% sterile fetal bovine serum (FBS) to stop collagenase action.
Isolated cardiomyocytes are then purified by sequential gravity settling steps, with a gradual calcium reintroduction to produce calcium-tolerant cells. The protocol reproducibly yields 655000 ±156000 total cells/heart and 46 ± 3% viable, rod-shaped cells.

Measurement of cardiomyocyte hypertrophy
Cardiomyocytes were visualized using brightfield at x20 magnification and cell width, length and area were measured in at least 300 cells per condition per experiment. Results were the mean of at least three different experiments performed on two cell isolations (using at least 4 different Cmed from adherent CD45 + cells).

Isolation of cardiac immune cells for Conditioned media (Cmed) preparation and invitro proliferation assay
After perfusion with PBS, the mice heart was excised and digested in HBSS medium containing 2.5 mg/ml collagenase D (Roche, Meylan, France) for 30 min at 37°C, with stirring.
Erythrocytes were lysed by using red blood cell lysis buffer (MiltenyiBiotec, Paris, France).
Cardiac CD45 cells were isolated by centrifugation, enriched using an anti-CD45 antibody coupled to magnetic beads (MiltenyiBiotec, Paris, France) and all CD45 + cells isolated from each heart were systematically seeded in 2 wells from a 48 multiwell plate (final volume 500µl in a RPMI medium (Life Technologies, Courtaboeuf, France) supplemented with 10 mmol/LHepes). After 3h of adhesion, the medium was renewed (this allowed elimination of non-adherent cells such as lymphocytes or NKT cells) and adherent cells (containing an average of 64.2 ± 3.5 % CD64 positive cells) were incubated overnight before collection of their conditioned medium (Cmed). The Cmed was concentrated 3 times on Amicon 3kDa Ultra centrifugal filter (Millipore), kept at -80°C until in-vitro studies on cardiomyocyte hypertrophy and used at a 1/20 final dilution. RPMI medium treated in parallel and the same way as Cmed was used as internal experimental control. Cytokines and chemokines in the Cmed were quantified using a bio-plex immunoassay (Biorad, Marnes-La-Coquette, France). For in-vitro proliferation assays, Mφ were kept in culture for 48 hours in the presence or absence of CX3CL1 (50ng/ml) before fixation in paraformaldehyde and fluorescent staining.

Quantification of macrophage proliferation
For BrdU incorporation assays, mice were injected intraperitoneally with 1 mg BrdU (Santa Cruz, Heidelberg, Germany) in 100μl PBS 2 hours prior to sacrifice and organ harvest.

Isolation and Preparation of immune cells for flow cytometry analysis
After perfusion with PBS, the mice heart was excised and digested in HBSS medium containing 2.5 mg/ml collagenase D (Roche, Meylan, France) for 30 min at 37°C, with stirring.
Erythrocytes were lysed by using red blood cell lysis buffer (MiltenyiBiotec, Paris, France). Samples were blocked with Fc block (Ebioscience, Paris, France) prior to labeling with antibodies. Cytometry data were acquired on an LSR Fortessa cytometer. After gating on CD45 + cells, doublets were excluded and live cells were analyzed (PI exclusion). Cardiac cell numbers were quantified using polybeads (Polysciences, Le-Perray-en-Yvelines, France). Data were analyzed with FlowJo software (Tree Star).

Preparation of immune cells for fluorescence activated cell sorting
Cardiac immune cells were isolated by centrifugation, enriched by immunoselection using an anti-CD45 antibody coupled to magnetic beads (MiltenyiBiotec, Paris, France).
Samples were blocked with Fc block (Ebioscience, Paris, France) prior to labeling with antibodies. Cytometry data were acquired on a BD FACSAria II cell sorter (5 lasers). After gating on CD11b+ cells, doublets were excluded and live (PI exclusion) CD14 + /CD64 + Mφ were sorted directly into RLT lysis buffer (Qiagen) or water and kept at -80°C until RNAseq analysis.

RNA sequencing and statistical analysis:
Total RNA from FACS sorted CD64 + /CD14 + cells was isolated using the Nucleospin RNA XS kit (Macherey Nagel, Hoerdt, France), according to the manufacturer instructions.
cDNA libraries were generated using total RNA with SMART-Seq v4 Ultra Low Input RNA Kit (TAKARA) and constructed according to manufacturer protocols as previously reported 3 .
Paired end sequencing (2 x 750 bp) was performed by Nextseq 500 machine using High Output kit (150 cycles). Raw sequencing data was quality-controlled with the FastQC program.
Trimmomatic was used to remove adapter sequences, trim low quality reads, and discard reads shorter than 40 bp. Reads were aligned to the mouse reference genome (build mm10) with the TopHat2 tool. Mapping results were quality-checked using RNA-SeQC. Aligned reads were counted using the FeatureCounts and Express software, at the gene-level and transcript-level, respectively. Normalization and differential analysis were performed with the GLM EdgeR package. RNA-Seq data has been made publicly available through the NCBI Gene Expression Omnibus (GEO), GEO accession number GSE157035.

Quantitative RT-PCR
Total RNA from CD64 + sorted cells was isolated with the Nucleospin RNA XS kit (Macherey Nagel). RNA reverse transcriptase-PCR analysis was performed using Verso cDNA Synthesis kit (Thermo Scientific) on a Lightcycler 1536 DNA green master (Roche) and a Bravo system (Agilent Technologies, Massy, France).
Primer sequences are listed below. Transcript levels were normalized to the Rpl13 mRNA. Gene

Immuno-fluorescence
Immuno-fluorescence studies were performed on frozen sections or cells fixed in paraformaldehyde as described in 2 . Immune cells or Frozen cardiac sections (6µm) fixed in paraformaldehyde for 15 min at room temperature and were stained immuno-histochemically with the following antibodies: rabbit anti-mouse Ki67 antibodies revealed with goat anti-rabbit Tissue sections and cells were analyzed with a Zeiss Axio Observer Z1 microscope.
Image analysis was performed using ImageJ and Photoshop CS5 (Adobe, San Jose, CA, USA).
Results are expressed as the number of positive cells per field and were quantified from 4-10 mice per group and 20-32 images per animal.

Quantification of cardiomyocyte area and tissue fibrosis
Frozen sections fixed in paraformaldehyde were labeled with WGA-Alexa 647 (1/500 dilution, Thermo Scientific, Montigny-Le-Bretonneux, France). Tissue sections were analyzed with a Zeiss Axio Observer Z1 microscope using ImageJ software. A low vs. high threshold allowed quantification of cardiomyocyte area or tissue fibrosis, respectively, as previously reported 2 . Results were quantified from 6-7 mice/group (12-32 images/animal).

Western Blot
Isolated cardiomyocytes or tissue homogenates were lysed in 150 mM NaCl, 50 mM We displayed cropped gels and blots in the main paper to improve the clarity and conciseness of the presentation. However, full-length unedited material was provided in the supplementary information, as mentioned in the figure legends.        Upper part of membrane was incubated with anti-CX3CR1 Ab, and lower part with anti-GAPDH Ab, followed by HRP Ab. Detection of ECL signals was performed after reassembling of membrane parts and was recorded using a Camera LAS 4000.

Supplementary information
Figure S10

Supplementary information
Full unedited gels to Figure 6C: Blots of Figure 4C were obtained from gel 1 (A, right panel) and experiment reproduced from other samples in gel 2 (A, left panel). Nitrocellulose membranes were cut. Upper part of membrane was incubated with anti-CX3CL1 Ab, and lower part with anti-GAPDH Ab, followed by HRP Ab.
Detection of ECL signals was performed after reassembling of membrane parts and was recorded using a Camera LAS 4000. Detection of ECL signals was performed after reassembling of membrane parts and was recorded using a Camera LAS 4000.