Sustained release of a p38 inhibitor from non-inflammatory microspheres inhibits cardiac dysfunction

Abstract

Cardiac dysfunction following acute myocardial infarction is a major cause of death in the world and there is a compelling need for new therapeutic strategies. In this report we demonstrate that a direct cardiac injection of drug-loaded microparticles, formulated from the polymer poly(cyclohexane-1,4-diylacetone dimethylene ketal) (PCADK), improves cardiac function following myocardial infarction. Drug-delivery vehicles have great potential to improve the treatment of cardiac dysfunction by sustaining high concentrations of therapeutics within the damaged myocardium. PCADK is unique among currently used polymers in drug delivery in that its hydrolysis generates neutral degradation products. We show here that PCADK causes minimal tissue inflammatory response, thus enabling PCADK for the treatment of inflammatory diseases, such as cardiac dysfunction. PCADK holds great promise for treating myocardial infarction and other inflammatory diseases given its neutral, biocompatible degradation products and its ability to deliver a wide range of therapeutics.

Figure 1. Polyketal microparticles – non-inflammatory polymer chemistry for drug delivery

(a) Poly(cyclohexane-1,4-diyl acetone dimethylene ketone) (PCADK) degrades to neutral, non-toxic products through the acid catalyzed hydrolysis of the ketal linkage (dotted circle). Acetone is an endogenous metabolite, while 1,4-cyclohexanedimethanol is FDA-approved as an indirect food additive. (b) A single emulsion/solvent evaporation technique was used to produce large (~20 µm) microparticles loaded with SB239063. Particles were imaged using SEM (scale bar: 20 µm). (c) The left descending coronary artery was permanently ligated to create an infarcted zone. Microparticles were injected intramyocardially where they released the encapsulated inhibitor within the infarct zone.

Figure 2. Macrophages are activated by PLGA microspheres in vitro, while PK-p38_i treatment inhibits p38 activation

(a) RAW 264.7 macrophages were treated with empty PCADK and PLGA particles for 2, 4, and 6 hours. Densitometric evaluation (mean + SEM; n=3) of Western blots (representative blot for phospho-p38 shown) demonstrate that PCADK did not elevate p38-phosphorylation at any time point. PLGA increased p38-phosphorylation in a time dependent manner, resulting in a four-fold increase in activation at 4 and 6 hours (*p<0.05 vs. all groups; ANOVA followed by Tukey-Kramer post test). (b) Macrophages were treated with PK-p38_i particles for the indicated time and stimulated with tumor necrosis factor-alpha (TNF-α). A time-dependent inhibition of phosphorylation was observed with complete inhibition occurring at 4–6h, and no effect seen with empty PK treatment (mean ± SEM, n=4, *p<0.05 vs. control, ANOVA followed by Tukey-Kramer post test). (c) Inhibition curves demonstrating similar dose-response profiles of the encapsulated and free inhibitor. Macrophages were stimulated with TNF-α following incubation with increasing doses of free or encapsulated inhibitor. There was no difference between treatments at any dose, and both exhibited similar IC₅₀ values. (d) Superoxide production, a downstream effect of p38 activation, was measured using DHE-HPLC. HPLC traces were used to quantify the superoxide-specific oxidation product of dihydroethidium, 2-hydroxyethidium (sample trace from sham and infarcted animals shown, inset). No effect is seen with empty PK treatment, while PK-p38_i reduced superoxide levels following TNF-α stimulation (mean ± SEM, n=4, n.s.=not significant *p<0.05 vs. control; ANOVA followed by Tukey-Kramer post test; SFM = serum free medium).

Figure 3. PCADK microparticles demonstrate little inflammatory response following intramuscular injections

Rats were injected with a high dose (50 mg/mL) of empty polymer microparticles. Histological sections were made and stained with DAPI for nuclei (blue) and for CD-45 (green), an inflammatory cell-specific marker. Tissue injected with PLGA microspheres (a) show a large influx of inflammatory cells, while tissue injected with PCADK microspheres (b) have little CD-45-positive staining. In a separate study, cytokine levels were measured in muscle tissue following an injection of 1 mg of PCADK particles or phosphate buffered saline (PBS) using a multiplex assay. (c) Injection of 1 mg of PCADK particles into the leg of rats leads to a slight elevation of interleukin-1β (IL-1β) at day one. At days 8, 14, 21 following surgery, PCADK particles do not elevate IL-1 β above basal levels. (d) PCADK particles do not elevate IL-6 levels at all time points following injection of 1 mg of PCADK particles. (e) Interferon gamma (IFN-γ) levels are not affected by PCADK particles. IFN-γ was not detected by the Bioplex assay at 21 days (n.d.). (mean + SEM, *p<0.05 vs. PBS, Student’s t-test).

Figure 4. PK-p38_i particles inhibit p38 phosphorylation, superoxide production, and TNF-α production in vivo following infarction

(a) PK-p38_i treatment inhibited phosphorylation of p38 at three and seven days in the infarct zone while free inhibitor (p38_i) or empty particles had no effect on phosphorylation at either time point (mean ± SEM, n ≥ 4, *p<0.05 vs. other treatment groups, ANOVA followed by Tukey-Kramer post test; MI = myocardial infarction). (b) Infarct zone tissue at three days was analyzed for superoxide using DHE-HPLC. MI alone, free inhibitor, and empty particles had significantly greater superoxide levels compared to sham, while PK-p38_i decreased the amount of superoxide produced (mean ± SEM, n ≥ 4, *p<0.05 vs. other treatment groups, ANOVA followed by Tukey-Kramer post test). (c) The inflammatory cytokine TNF-α was measured in the infarcted zone by ELISA three days post-surgery. MI alone, free inhibitor, and empty particles had significantly greater amounts of TNF-α, while PK-p38_i treatment reduced TNF-α levels by nearly two-fold (mean ± SEM, n≥4; *p<0.05 vs. other treatment groups, ANOVA followed by Tukey-Kramer post test).

Figure 5. PK-p38_i therapy results in improved cardiac function and reduced fibrosis

(a) Dimensions of the left ventricle were taken at systole and diastole using MRI at day seven and day 21 post-occlusion. PK-p38_i showed a statistically significant difference between days 7 and 21, whereas all other treatments did not reach significance (mean + SEM, n>4, *p<0.05 repeated measures ANOVA). In addition, PK-p38_i fractional shortening was significantly higher on day 21 compared to MI alone and all other treatment groups (ANOVA followed by Tukey-Kramer post-test). (b) Fractional shortening was measured and expressed as an absolute percent difference between days 7 and 21. A positive value represents an improvement in cardiac function between days 7 and 21, while negative values represent progression of cardiac dysfunction. PK-p38_i treatment showed a significant 10% improvement (absolute value) in cardiac function while PLGA-p38_i treatment did not inhibit cardiac dysfunction (mean ± SEM, n ≥ 4, **p<0.001 vs. all groups). (c) The left-ventricular free wall was analyzed histologically in at least 3 serial sections for fibrosis using a collagen-specific Picrosirius red stain. MI alone, free inhibitor, and empty particle treatments had significant increases in fibrosis compared to sham operation. PK-p38_i treatment reduced fibrotic area by more than half, though not completely to sham levels (mean + SEM, n≥4, *p<0.05 vs. other treatment groups, ANOVA followed by Tukey-Kramer post test). (d–g) Representative Picrosirius red images of (d) MI, (e) MI + free inhibitor, (f) MI + PK, and (g) MI + PK-p38_i treatment are shown (scale bar: 200 µm).