Inhibition of inflammation by minocycline improves heart failure and depression-like behaviour in rats after myocardial infarction

Abstract

Rationale: Patients with heart failure have an increased incidence of depression. Central and peripheral inflammation play a major role in the pathophysiology of both heart failure and depression.

Aim: Minocycline is an antibiotic that inhibits microglia activation and release of pro-inflammatory cytokines. We assessed effects of minocycline on extent of heart failure and depression at 2 and 8 weeks post myocardial infarction.

Methods/results: Male Wistar rats were randomly divided into 3 groups: (i) sham + vehicle; (ii) MI + vehicle; and (iii) MI + minocycline with n/group of 8, 9 and 9 at 2 weeks, and 10, 16, 8 at weeks, respectively. Oral minocycline (50 mg/kg/day) or vehicle started 2 days before surgery. Depression-like behaviour was assessed with sucrose preference and forced swim tests, and cardiac function with echo and hemodynamics. After myocardial infarction, microglia activation and plasma/brain pro-inflammatory cytokines increased, which were mostly prevented by minocycline. At 8 weeks, cardiac dysfunction was attenuated by minocycline: infarct size (MI + Vehicle 29±1, MI + Min 23±1%), ejection fraction (Sham 80±1, MI + Vehicle 48±2, MI + Min 58±2%) and end diastolic pressure (Sham 3.2±0.3, MI + Vehicle 18.2±1.1, MI + Min 8.5±0.9 mm Hg). Depression-like behaviour was significantly improved by minocycline in sucrose preference test (% Sucrose Intake: Sham 96±1, MI + Vehicle 78±2, MI + Min 87±2) and forced swim test (% Immobile: Sham 40±4, MI + Vehicle 61±3, MI + Min 37±6).

Conclusion: Rats post myocardial infarction develop systemic inflammation, heart failure and depression-like behaviour that are all attenuated by minocycline. Targeting (neuro) inflammation may represent new therapeutic strategy for patients with heart failure and depression.

Fig 1. Timeline showing the experimental design for the 8 weeks post MI treatment study and the sequence of the behaviour tests.

Male Wistar rats received orally daily minocycline or vehicle starting 2 days before MI surgery. Behavioural studies were carried out 6 to 8 weeks after surgery with first the sucrose preference test (SPT), followed by Morris Water Maze test (MWM), forced swim test (FST) and fear condition test (FCT). Except for the SPT, the behaviour tests were performed between 7:00 am and 2:00 pm in the dark cycle which started at 3:00 am.

Fig 2. Chronic oral treatment with minocycline improves cardiac function at 2 and 8 weeks post MI.

Values are means ± SE. For n/group, see bottom of bars. One-way ANOVA at 2 weeks: EDP F = 21.3, p<0.001; PSP F = 22.2, p<0.001; dP/dt _max F = 74.1, p<0.001 and EF F = 86.9, p<0.001. One-way ANOVA at 8 weeks: EDP F = 64.4, p<0.001; PSP F = 12.5, p<0.001; dP/dt _max F = 45.7, p<0.001 and EF F = 66.4, p<0.001. *p≤0.05 vs sham group, ^† p≤0.05 vs MI + vehicle.

Fig 3. Chronic oral treatment with minocycline improves parameters of cardiac function relative to MI scar size in rats treated for 8 weeks post MI.

Minocycline lowers LVEDP (A) and increases dP/dt _max (B) relative to MI scar size, but increases ejection fraction (C) in parallel with decrease in MI scar size. Data were analyzed by Analysis of Co-variance.

Fig 4. Changes in sucrose preference as percent of sham at 2, 4 and 8 weeks post MI (top panel), and improvement in the decrease in sucrose preference by minocycline at 8 weeks post MI (bottom panel).

Values are means ± SE. For n/group, see bottom of bars. One-way ANOVA for time-course: F = 13.6, p<0.001. *p<0.05 vs sham or 2 weeks post MI. One-way ANOVA for 2 weeks minocycline F = 1.4, NS, and for 8 weeks F = 25.3, p<0.001. *p<0.05 vs sham ^† p<0.05 vs MI + vehicle.

Fig 5. Chronic treatment with minocycline prevents the increase in immobility time in the forced swim test (FST) at 8 weeks post MI.

Values are means ± SE. For n/group, see bottom of bars. One-way ANOVA at 2 weeks F = 0.7, not significant. One-way ANOVA at 8 weeks F = 9.7, p<0.001. * p<0.05 vs sham ^† p<0.05 vs MI + vehicle.

Fig 6. Extent of freezing during the acquisition FC on day 1, contextual FC on day 2, and the cued FC on day 3.

Chronic treatment with minocycline prevents enhanced fear memory at 8 weeks post MI. Values are means ± SE. For n/group, see bottom of bars. Two-way repeated measures ANOVA for contextual FC: treatment effect F = 1.2, p = 0.33, time-effect F = 8.1, p = 0.006 and treatment × time F = 1.3, p = 0.28. § p <0.05 vs habituation day 1. Two-way repeated measures ANOVA for cued FC: treatment effect F = 4.7, p = 0.02, time-effect F = 17.6, p<0.001, and treatment × time F = 2.1, p = 0.06. *p<0.05 vs pre-tone, ^† p<0.05 vs other groups.

Fig 7. Spatial learning (top panel) and memory (bottom panel) in the MWM test are not significantly affected by HF at 7–8 weeks post MI.

Chronic treatment with minocycline increases latency to platform (spatial learning), but does not significantly affect spatial memory. The 25% line in bottom panel indicates level of random exploration. Values are means ± SE. For n/group, see bottom of bars. Two-way repeated measures ANOVA for latency to platform: time-effect F = 27.5, *p<0.001 vs day 1, and treatment groups F = 5.2, ^† p<0.05 vs other groups. One-way ANOVA probe 1 F = 2.3, p = 0.12, and probe 2 F = 2.2, p = 0.13.

Fig 8. Minocycline prevents the marked microglia activation in the PVN at 8 weeks post MI.

Upper panel presents representative images showing activated microglia with stronger staining, enlarged soma, and shorter, stubbier processes in the PVN of a MI + veh rat. Lower panel shows the summary data in bar graphs. Values are means ± SE (n = 4/group). One-way ANOVA for number of microglia F = 11.5, p<0.01 and for percent activated microglia F = 171.3, p<0.0001. * p<0.05 vs sham ^† p<0.05 vs MI + vehicle.