OPF/GO hydrogel preparation and characterization

OPF and OPF/GO hydrogels were fabricated as described previously ^{9, 10}. Briefly, 0.16 g of OPF 10 K and 0.08 g of PEGDA were dissolved in 380 µL of phosphate buffered saline (PBS) or the GO dispersion (1.0 mg/mL) and incubated at 23 ℃ for 45 min to eliminate air bubbles. Thermal radical initiators, 74.9 µL of 300 mM N, N, N', N'-tetramethylethylenediamine (TEMED) solution (Sigma) and 74.9 µL of 300 mM ammonium persulfate (APS) solution (Sigma), were then added and incubated at 37 ℃ for 8 min for crosslinking. The electrical resistance of the hydrogel was detected by an electrochemistry workstation (CHI660D) (n=3/group).

Degradability was monitored as follows: OPF and OPF/GO hydrogels were prepared as described above ^{9, 10}, except the polymer solutions were injected in Teflon molds (6 mm diameter by 0.5 mm thick) after mixing, and placed at 37 °C for 20 min afterwards. The resulting hydrogels were subject to degradation in PBS in 12-well tissue culture dishes at 37 °C on a shaker table shaking at 80 rpm. The PBS (2.5 mL/well) was completely changed every other day for the first week and the mass of the hydrogels was monitored every week. The mass remaining ratio of hydrogels (n=3) at each time point were represented as w_d/w_i × 100%, where w_i and w_d are the weight of the dried composite after fabrication prior to swelling, and the weight of the dried composite after swelling at each time point, respectively (n=3/group).

Ex vivo measurement of skeletomuscular signal propagation through OPF/GO

After euthanizing adult rats, rat leg skeletal muscle segments were explanted from the hind limbs. The muscles were cut in strips with a width of 10 mm and placed into a cell dish leaving a gap of 5 mm between two muscle pieces. This gap was filled with OPF or OPF/GO hydrogel, which acted as a bridge structure to connect two pieces of muscle tissue together. Voltage was then applied between two gel-connected muscle strips until contraction of both pieces could be observed to determine the voltage threshold (n=3).

Biocompatibility of OPF/GO

Rat cardiac fibroblasts cells were isolated from 1-day-old neonatal Sprague-Dawley rats and seeded onto uncoated polystyrene dishes or those coated with OPF or OPF/GO at a concentration of 250 cells/mm² to assess cell attachment ^{13, 30, 37}. After 48 h, cells were stained with live & dead kit to analyze the cytocompatibility of OPF/GO.

To do immunocytochemistry, cultured cardiac fibroblasts were fixed in 4% paraformaldehyde for 30 min at room temperature. After fixation, cells were treated with 0.3% Triton X-100 for 30 min. Cells were then stained with vimentin primary antibody (Abcam) at 1: 200 dilution in blocking buffer for 24 h at 4 ℃. Next, the sample was treated with secondary antibody (1: 200 dilution) in goat serum for 40 min. After washing with PBS, the sample was stained with DAPI (1: 200) diluted in PBS for 40 min at room temperature. The stained samples were then imaged with an inverted laser scanning confocal microscope (Leica SP5X MP, Germany). To assess the viability of cardiac fibroblasts on different materials, an MTT assay was performed according to the manufacturer's instructions. Briefly, a total of 1×10⁴ cardiac fibroblasts were seeded on polystyrene, OPF or OPF/GO-coated 96 well surface, and grown for 48 h. Cardiac fibroblasts were incubated with MTT for 3 h at 37 °C, and cell viability was determined by measuring a test wavelength of OD 570 nm and a reference wavelength of 630 nm.

In vivo evaluation in rat infarct model

MI was induced in 27 male Sprague Dawley rats using an established model ¹⁹. Briefly, the rats were anesthetized by injecting with 2% pentobarbital sodium into cavum abdominis. A left-sided open thoracotomy was performed, the heart was exposed, and the proximal left anterior descending artery (LAD) was ligated with 6-0 polypropylene. The rats were randomized into 3 groups and received separate peri-infarct intramyocardial injections of PBS (100 µL; n=7), OPF hydrogel (100 µL; n=10) or OPF/GO hydrogel (100 µL; n=10). Then, the thoracotomy was closed in multiple layers.

Echocardiography

Echocardiography (ECHO) was performed 4 weeks after hydrogel or PBS injection. Rats were anesthetized with 2% pentobarbital sodium. Standard transthoracic echocardiography was performed using the echocardiography system equipped with a 13 MHz linear ultrasonic transducer (15L8; Acuson Corporation, Mountain View, CA) in a phased array format. All analyses were performed by a single investigator who was blinded to the treatment groups (n=6/group).

Histological analysis and immunohistochemistry

After 24 h, 1, 2, 3, and 4 weeks of injections, rats in all surgical groups were anesthetized, and the heart was exposed and arrested by apical injection of 1 mL 15% KCl. The hearts were explanted and fixed in 2% paraformaldehyde overnight and then were paraffin-embedded. The embedded tissues were serially sectioned at 3 µm in the LV transverse direction. Masson's trichrome staining was performed to measure infarct size. Hematoxylin and eosin staining was performed to measure the maintenance and distribution of OPF or OPF/GO hydrogel in the infarct region. Immunohistochemical staining was performed with antibodies against CD68 (Abcam), F-actin (Abcam), N-cadherin (Abcam), plakoglobin (Abcam), desmoplakin, alpha-smooth muscle actin (α-SMA, Abcam), connexin 43 (Cx43, Abcam) and α-actinin (Abcam). Moreover, the number of CD68⁺ cells was counted from 6 randomly selected fields at the infarcted zone under a magnification of 60×. Nuclei were stained with 4', 6-diamidino-2-phenyindole (DAPI, Sigma). Morphometric analysis was performed to assess infarct and myocardial wall thickness. Infarct thickness was computed at each time point from the average spoke length. Infarct size was calculated as percentage of length of the infarcted endocardial circumference as previously described ²¹. Capillaries were recognized as tubular structures positively stained for α-SMA as described in the literature ²².

Reverse transcription and quantitative polymerase chain reaction

4 weeks after injections, total RNA of the infarcted region was extracted using Trizol (Invitrogen). Total RNA was treated with DNase I (Invitrogen, CA, USA) to eliminate the genomic DNA carry over. DNase-treated RNA was reverse transcribed using high capacity cDNA kit (Toyobo). Real-time quantitative PCR (qPCR) was performed with Fast SYBR Green Master Mix (Toyobo). The corresponding primers were presented in Table S1.

Western blot analysis

Western blot analysis was carried out using lysates from infarct heart tissues 4 weeks after MI. The infarcted tissue and protein extract (TIANDZ, Beijing, China) were homogenized on ice. The supernatant was collected after centrifugation at 12000×g for 10-15 min. The extracted proteins were determined by using BCA Protein Assay Kit (Thermo Scientific, USA). Equal amounts (50 µg) of denatured proteins were separated on 12% sodium dodecyl sulfate-polyacrylamide gels and transferred to nitrocellulose membrane. The membranes were blocked with 5% defatted milk for 1 h and were incubated with these primary antibodies overnight at 4 ℃. After being washed with Tris-base buffer, they were further incubated using appropriate secondary antibodies (goat anti-mouse or anti-rabbit IgG (Boster, Wuhan, China)) for 1 h at room temperature. The labeled proteins were visualized by enhanced chemiluminescent reagent (Applygen, Beijing, China). GAPDH was used as an internal control to correct the variations of different samples.

Isolation of adult CMs

Adult CMs were isolated as described in literature ²³, with minor modifications. Briefly, 4 weeks after injections, rats in all surgical groups were anesthetized. Hearts were removed and perfused via aortic cannulation with a constant flow of 3 mL/min in a Langendorff apparatus. Hearts were firstly perfused at 37 °C for 5 min with Tyrode's Solution containing 137 mM NaCl, 5.4 mM KCl, 0.5 mM MgCl₂, 5.0 mM D-glucose, 11.6 mM HEPES, and 1.8 mM CaCl₂ (pH 7.4). Then, the hearts were perfused for 3 min with the Tyrode's solution containing 130 mM NaCl, 5.4 mM KCl, 1.2 mM KH₂PO₄, 6.0 mM HEPES, and 10.0 mM D-glucose (pH 7.4), followed by digestion solution (0.8 mg/mL colleganse II) for 16 min. Hearts were then removed from the Langendorff apparatus while the tissues were gray and loosely connected. Desired areas (infarct zone) were then dissected in KB solution containing 50 mM L-glutamic acid, 30 mM KCl, 80 mM KOH, 30 mM KH₂PO₄, 20 mM taurine, 10 mM HEPES, 10 mM D-glucose, 3 mM MgSO₄, 0.5 mM EGTA (pH 7.4), mechanically dissociated, triturated, and resuspended in Tyrode's solution, and calcium was gradually reintroduced through a series of washes. For calcium transient measurements, cells were used on the same day as isolation.

Calcium-transient assessment

Calcium transients were assessed as previously described ²³. Briefly, isolated cardiomyocytes were loaded with Fluo-4 for 30 min at room temperature before measurements. The loading solution contained a 1:10 mixture of 5 mM Fluo-4 AM in dry DMSO, which was diluted 100-fold into extracellular Tyrode's solution containing suspended myocytes. After 30 min, the loading solution was extracted and added to the fresh Tyrode's solution. Then Fluo-4 fluorescence transients were recorded via live cell station.

Biocompatibility of OPF/GO Hydrogel in vitro

We investigated the biocompatibility of OPF/GO on cardiac fibroblasts, an important cell type in the heart after MI (Figure S2A). To demonstrate cell morphology and biomaterial attachment, rat fibroblasts were seeded on the surface of OPF- or OPF/GO hydrogel-coated wells and grown for 48 h (Figure ​Figure22A and Figure S2B). The effects of OPF/GO hydrogel on cardiac fibroblasts were investigated via live & dead staining. Compared with the cardiac fibroblasts seeded on OPF-coated surfaces, cells seeded on OPF/GO-coated surfaces maintained excellent viability and few dead cells (red) could be observed on all OPF/GO hydrogels, indicating that the introduction of GO did not induce cytotoxicity even at GO concentrations up to 1.0 mg/mL (Figure ​Figure22B). In addition, the introduction of GO nanoparticles improved the poor cell adhesion on the OPF surface, which correlated with increasing GO concentration (Figure ​Figure22C). A similar effect was observed for the MTT assay (Figure ​Figure22D): the introduction of GO in the OPF hydrogel greatly enhanced cell adhesion, and the adhesion effects were correlated with GO concentration. No significant differences were observed in terms of cytotoxicity between OPF with 0.6 mg/mL GO and 1.0 mg/mL GO. On the other hand, the results in Figure ​Figure11D suggested that the electrical conductivity of OPF/GO hydrogel with the composite ratio of 1.0 mg/mL was significantly higher than the rest. Therefore, OPF/GO with the composite ratio of 1.0 mg/mL was used in implantation research after MI.

Open in a separate window

Figure 2

OPF/GO hydrogel supports cell growth without toxicity. (A) Representative morphological images of cardiac fibroblast cells cultured for 48 h on OPF- or OPF/GO-coated substrates. (B) Rat cardiac fibroblast cells were plated on the surfaces of wells coated with OPF or OPF/GO, grown for 48 h, and then stained with live & dead kit. (C) Immunofluorescence images of cardiac fibroblasts seeded on OPF- or OPF/GO-coated substrates (vimentin, red). (D) An MTT assay of cardiac fibroblasts cultured on the surface of pure OPF and different composite OPF/GO (0.3 mg/mL, 0.6 mg/mL, 1.0 mg/mL) (n=3, *P<0.05). MTT indicates 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazoliumbromide.

In vivo biocompatibility evaluation

MI was induced in a rat model by occlusion of the left anterior descending coronary artery according to established laboratory procedures. Subsequently, 100 μL of injectable hybrid hydrogel was injected at three different sites in the peri-infarct regions of the heart. The rat models were divided into three groups and received PBS, OPF and OPF/GO injections, respectively. The rats were sacrificed at different time points (24 h, 1, 2, and 4 weeks) after injection to observe the histological responses to scaffolds using light microscopy.

As shown in Figure S3A, OPF/GO was still visible in H&E-stained heart sections of OPF/GO-treated hearts. The implant experiments suggested that the OPF/GO hydrogels could provide mechanical support to the infarcted region and interact directly with the infarcted tissue. 24 h after injection post MI, OPF or OPF/GO hydrogels were observed in cross sections of left ventricular tissue, which indicates direct interaction with the infarcted tissue at the early stage of MI. Along with partial degradation of OPF hydrogel with time after implantation, the biomaterial was gradually replaced by connective tissue and myofibroblasts. At 2 weeks and 4 weeks, only rare, weak pink-stained isolated islands of biomaterial were detected in the infarct area after OPF injection. However, in the OPF/GO-injected group, OPF/GO hydrogel was still visible and distributed continuously along the infarcted region at 4 weeks post MI (Figure S3B).

The inflammatory responses of injectable OPF/GO hydrogel in infarcted hearts

Considering the role of OPF and OPF/GO hydrogels in promoting macrophage recruitment and activation, we analyzed the infiltration of macrophages in the infarcted hearts. CD68⁺ cells were evaluated by immunofluorescence staining in the infarcted area of myocardial sections at 1 week after MI. No significant differences were observed in macrophage infiltration between PBS, OPF and OPF/GO-injected infarcts (Figure ​Figure33A). Compared to the PBS group, the ratio of CD68⁺ cells in the OPF- and OPF/GO-injected hearts was slightly increased (Figure ​Figure33B). Moreover, the recruitment of CD68⁺ cells was not significantly enhanced in or around the OPF/GO hydrogels in the infarcted region (Figure ​Figure33C).

Open in a separate window

Figure 3

Effects of OPF and OPF/GO hydrogels on macrophage infiltration in infarcted hearts. (A) Representative fluorescence images show CD68-positive cells (left column, merged channel; middle column, CD68 (red); right column, DAPI (blue)) in PBS-, OPF-, and OPF/GO-injected hearts at 1 week post MI. (B) Quantification of CD68-positive cells of the corresponding images of (A). (C) CD68-positive signal is co-localized with the presence of OPF/GO hydrogel (CD68 (green), DAPI (blue)).

The injectable OPF/GO hydrogel enhanced cytoskeletal structure and intercalated disc assembly

As shown in Figure ​Figure44A, in the OPF/GO- and OPF-treated group, a cross-striated pattern can be obviously observed in the F-actin-positive area at 4 weeks post MI. Whereas in the PBS injected group, the pattern of the F-actin-positive area was weak and ambiguous (Figure ​Figure44A-B). In order to confirm if the enhanced generation of cytoskeletal structures was induced by the mechanical support of OPF and OPF/GO, we further explored the mechanism of the enhanced cytoskeletal structures and examined intercalated disc (ID) formation.

Open in a separate window

Figure 4

OPF/GO enhanced cytoskeletal structure 4 weeks after MI. (A) F-actin (green) in the infarcted areas of PBS-, OPF- and OPF/GO-treated groups at 4 weeks post MI (DAPI (blue)). (B) α-Tubulin (green) in the border zone cardiomyocytes (marked by actinin staining in red, DAPI (blue)).

ID is important for cardiac integrity and function ²⁴. To determine whether the injection of OPF/GO had any influence on the assembly and formation of ID, ID-related proteins in the infarcted regions were characterized by immunofluorescence staining. Marker proteins including N-cadherin (NC) for adherens junctions, and plakoglobin (PG) and desmoplakin (DP) for desmosomes were examined 4 weeks post injection. As shown in Figure ​Figure55A, increases in localized structures were seen in the cell contact region, including the plaque pattern of N-C protein, stair-like distribution of PG protein (Figure ​Figure55B), and strand-like appearance of DP (Figure ​Figure55C). Meanwhile, NC, PG and DP showed larger linear regions in the OPF/GO groups compared to those in the OPF and control groups, indicating the formation of better developed mechanical and gap junctions. To further quantify the ID-associated proteins, we quantified the ID-associated proteins by western blotting. The results suggested that ID-related protein expression in the OPF/GO group had significantly higher levels relative to those in the groups injected with OPF and PBS after 4 weeks (Figure ​Figure55D).

Open in a separate window

Figure 5

Immunofluorescence staining and quantification of ID-related proteins in the hearts of PBS-, OPF-, OPF/GO-injected groups at 4 weeks post MI. Marker proteins including N-cadherin (NC) for adherens junctions (A), and plakoglobin (PG) (B) and desmoplakin (DP) (C) for desmosomes. (D) Western blot revealed the OPF/GO-injected group had significantly higher expression levels of NC, PG, and DP compared to the PBS- and OPF-injected groups.

The injectable OPF/GO hydrogel promotes the expression of gap connexin proteins

Immunohistochemistry and histology of tissue sections revealed higher gap junction remodeling in the infarcted region of the OPF/GO-treated group than those in the OPF and PBS groups. Cx43 red fluorescence could be observed at the gap connexin region of the tissue by immunohistochemistry staining. As shown in Figure S4A, no significant differences were observed in the formation of gap junction-associated proteins between PBS-, OPF- and OPF/GO-injected infarcts 2 weeks after MI. The higher gap junction remodeling was observed obviously in the infarcted region of the OPF/GO hydrogel-treated group compared to those of the OPF- and PBS-treated groups at 3 weeks and 4 weeks (Figure S4B and Figure ​Figure66A). The expression levels of gap junction-associated markers Cx43, Cx40, Cx37 and Cx45 were analyzed by qPCR (Figure ​Figure66B). The results were consistent with those of the immunohistochemistry analysis, where the mRNA expression level of gap connexin-associated genes in OPF/GO hydrogel was higher than that of OPF and PBS injection at 4 weeks post injection.

Open in a separate window

Figure 6

The formation of gap junction-associated proteins in the infarcted region. (A) Immunofluorescence staining revealed higher gap junction remodeling in the infarcted region of the OPF/GO hydrogel-treated group compared to the OPF- and PBS-treated groups. The highlighted areas are presented in the bottom images. (B) qPCR analysis of the expression levels of gap junction-associated markers Cx43, Cx40, Cx37 and Cx45. (C) Western blot revealed the downstream signaling molecules of canonical Wnt signaling pathway. (D) A schematic of the possible mechanisms.

To further clarify the possible mechanism of the enhanced generation of connexin-associated proteins, we quantified the downstream signaling molecules of canonical Wnt signaling pathway by western blot (Figure ​Figure66C). Our data suggests that the introduction of GO improved the expression of total Akt. Although there were no significant differences in the expression of total GSK-3β among all experimental groups, the introduction of GO could improve the phosphorylation levels of GSK-3β (pGSK-3β) at 14 days post MI. Similarly, pGSK-3β could then induce β-catenin signaling and subsequently cause an increase in Cx43 expression (Figure ​Figure66D).

Intramyocardial injection of the OPF/GO hydrogel improves cardiac function recovery after MI

Masson's Trichrome assay showed that there was extensive fibrous tissue formation in the border area of the PBS group, whereas little fibrosis was observed in the OPF/GO group. The control group receiving PBS injection had collagen zones occupying a large portion of the LV wall of the infarct site (Figure ​Figure77A), indicating the presence of widely spread scar tissue. In contrast, the animals injected with OPF/GO had visually more myocardial tissue at the infarct site. Consistent with this observation, LV wall thickness was significantly preserved (from 0.37±0.096 mm to 0.77±0.079 mm) in the OPF/GO injected group (Figure ​Figure77B); the infarct size was drastically decreased (from 50.7% to 31.7%; Figure ​Figure77C). These results suggest that the structural basis of the infarcted heart was effectively restored after OPF/GO injection.

Open in a separate window

Figure 7

Masson's trichrome staining and echocardiographic measurements represent the effects of OPF/GO hydrogel on the morphology and myocardial functional recovery of infarcted hearts. (A) Masson trichrome staining of PBS-, OPF- or OPF/GO-injected heart 4 weeks post injection. Four weeks after injection, hearts were excised and left vetricle wall thickness (B) and infarct size (C) were measured. (D) Systolic ventricular function was improved after OPF/GO hydrogel injection into the border zone. Echocardiography (ECHO) was performed at 4 weeks post injection. Quantification of the parameters reflecting blood pumping function, including ejection fraction (EF) (E) and fractional shortening (FS) (F) 6 weeks after MI. Quantification of the parameters reflecting ventricular filling function, including left ventricular end-diastolic dimension (LVEDD) (G) and left ventricular end-systolic dimension (LVESD) (H). Data is presented as mean ± standard deviation; *p < 0.05, or **p < 0.01 (n=6/group).

The MI-damaged left ventricle undergoes progressive remodeling, leading to the deterioration of cardiac function ²⁴. Cardiac function was evaluated using echocardiography (ECHO) at 4 weeks post MI (Figure ​Figure77D). The results of echocardiography showed that LV ejection fraction (EF; Figure ​Figure77E) and fractional shortening (FS; Figure ​Figure77F), two important indicators for blood pumping function, were improved in both OPF and OPF/GO groups compared to PBS-injected rats. In particular, the hearts treated by OPF/GO show a significantly higher recovery of cardiac function than the OPF-treated group. Consistently, the OPF/GO group had a decrease in left ventricular end-diastolic dimension (LVEDD; Figure ​Figure77G) and left ventricular end-diastolic volume (LVESD; Figure ​Figure77H), indicating a functional improvement in ventricular filling. Together, these data proved that OPF/GO injection effectively promotes myocardial functional recovery after MI.

The injectable OPF/GO hydrogel promotes neovascularization

Angiogenesis restores blood supply to myocardium, facilitating cardiac function recovery ²⁰. We next investigated the effects of OPF/GO injection on neovascularization after MI. The OPF/GO injection significantly increased the mRNA level of α-SMA (Figure S5A), vWF (Figure S5B) and VEGF (Figure S5C) at 4 weeks after LAD ligation, suggesting that OPF/GO injection may have an angiogenic effect. 4 weeks post injection, rats that received OPF/GO injections had a higher density of α-SMA-positive capillaries in the infarcted myocardium compared to both PBS- and OPF-treated rats (Figure S5D). The undegraded OPF/GO hydrogel appeared to be distributed in the anterior wall and cell ingrowth was found inside or around the hydrogel area, with some regions of this tissue structure staining positively for α-SMA in higher magnification images (Figure S5E). These data suggested angiogenesis can be an outcome of gap junction enhancement.

The authors thank Antonios G. Mikos (Rice University) for providing OPF raw materials. This work was supported by Projects of International Cooperation and Exchanges NSFC (No. 31320103914), National Natural Science Funds for Outstanding Young Scholar (No. 81622027), the Key Program of the National Key Research and Development Program of China (No. 2016YFY1101303, No. 2017YFA0106100), National Natural Science Foundation of China (No. 31370987, 31470940, 81371696), Joint Funds for National Natural Science Foundation of China (No. U1601221), Beijing NOVA Program of China (No. 2016B615).