Cardioprotection and lifespan extension by the natural polyamine spermidine

Abstract

Aging is associated with an increased risk of cardiovascular disease and death. Here we show that oral supplementation of the natural polyamine spermidine extends the lifespan of mice and exerts cardioprotective effects, reducing cardiac hypertrophy and preserving diastolic function in old mice. Spermidine feeding enhanced cardiac autophagy, mitophagy and mitochondrial respiration, and it also improved the mechano-elastical properties of cardiomyocytes in vivo, coinciding with increased titin phosphorylation and suppressed subclinical inflammation. Spermidine feeding failed to provide cardioprotection in mice that lack the autophagy-related protein Atg5 in cardiomyocytes. In Dahl salt-sensitive rats that were fed a high-salt diet, a model for hypertension-induced congestive heart failure, spermidine feeding reduced systemic blood pressure, increased titin phosphorylation and prevented cardiac hypertrophy and a decline in diastolic function, thus delaying the progression to heart failure. In humans, high levels of dietary spermidine, as assessed from food questionnaires, correlated with reduced blood pressure and a lower incidence of cardiovascular disease. Our results suggest a new and feasible strategy for protection against cardiovascular disease.

Figure 1. Spermidine extends lifespan and improves cardiac diastolic function in mice.

(a) Schematic overview of spermidine administration to aging wild-type C57BL/6J mice. Spermidine was supplemented to drinking water starting from the age of 4 months (life-long) or 18 months (late-in-life) and cardiovascular parameters (Fig. 1f-l) or molecular phenotypes (Figs. 2 and 3) were analyzed at the indicated time points (M, months). (b-d) Kaplan-Meier survival analyses. In the life-long supplementation experiment using C57BL/6J female mice (b, c), the same control group was used. The late-in-life supplementation experiment (d) used C57BL/6J male and female mice. Dashed lines depict median lifespans. N=40/41 (b, control/spermidine), N=40/20/17 (c, control/putrescine/spermine), N=91/86 (d, control/spermidine) mice (see Supplementary Tables 1 and 2 for more details). P-values, calculated using Breslow test, represent pairwise comparisons of survival curves between the groups Spermidine vs. Control (b), Spermine vs. Control (c), and Spermidine vs. Control (d). (e-k) Effects of late-in-life supplementation of spermidine in C57BL6/J male mice, analyzed at the indicated ages (M, months). Shown are whole blood polyamine levels (e), left ventricular mass-to-tibia length ratio (LVmass/TL), indicative of cardiac hypertrophy (f), representative hemodynamic pressure-volume loops (g), left ventricular end-diastolic pressure (EDP) (h), myocardial stiffness constant (end-diastolic pressure-volume relationship [EDPVR] β obtained from exponential fits presented in Supplementary Fig. 4a) (i), ejection fraction, as determined by echocardiography (j) and ventricular-vascular coupling (VVC) (k). N=15/18 (e, 24M/24M+S), N=10/14/20/20 (f, j, 4M/18M/23M/23M+S), N=10/8/10/10 (h, i, k, 4M/18M/24M/24M+S) mice. (l) Systolic and diastolic arterial blood pressures in mice analyzed in (e). N=8/11 (l, 24M/24M+S) mice. ***p<0.001, **p<0.01, *p<0.05 and #p<0.06 (ANOVA with post-hoc Tukey, ANCOVA (panel i) or Welch’s t-test (panel e), see Methods). For box-and-whisker plots, whiskers show minima and maxima within 1.5 interquartile range.

Figure 2. Spermidine improves cardiomyocyte composition and mitochondrial function in mice.

C57BL/6J male mice were supplemented with spermidine (+S) late-in-life (see Fig. 1a for the feeding scheme) and hearts were subjected to molecular and biochemical analyses at the indicated ages (M, months). (a, b) Representative transmission electron micrographs (a) and quantification of left ventricular cardiomyocyte composition using design-based stereology (b). Relative volumes of mitochondria Vv(mi/myo), myofibrils Vv(mf/myo) and mitochondria- and myofibril-free sarcoplasm Vv(sp/myo) per cardiomyocyte are shown. mi, mitochondria; mf, myofibrils; myo, cardiomyocyte; nu, nucleus; sp, sarcoplasm. Scale bar represents 2 µm. N=10/15/14 (4M/24M/24M+S) mice. (c) Oxygen consumption (complex I-mediated respiration) of isolated cardiac mitochondria using high-resolution respirometry. N=8 (left), N=5 (right) mice/group. (d) Plasma TNFα levels after late-in-life spermidine supplementation of C57BL6/J male mice. Mice that had a possibly acute inflammatory condition were excluded (see Supplementary Fig. 8 and Methods). N=12/12/13/9/10 (5M/21M/21M+S/23M/23M+S) mice. (e) Titin isoform composition and phosphorylation. Representative Western Blots probed with a pan-phospho-serine/threonine-antibody for detection of total N2B phosphorylation (P-N2B, quantified in g), Coomassie-stained PVDF membrane for detection of total N2B (Total-N2B), and a Coomassie Blue-stained gel for detection of N2BA and N2B isoforms quantified in f (see Methods for details on isoform identification). (f) The N2BA/(N2BA+N2B) ratio was quantified by densitometry using normalization standards for inter-gel comparisons (see Methods). (N=12 mice/group). (g) Quantification of total N2B phosphorylation (left) and serine 4080 (S4080)-specific N2B phosphorylation (right, from densitometry of western blots representatively shown in Supplementary Fig. 5c). (N=12 mice/group). Panels b,c, f,g: ***p<0.001, **p<0.01 and *p<0.05 (ANOVA with post-hoc Tukey, Welch’s test with post-hoc Games-Howell or t-test (paired in panel c) as appropriate, see Methods). Panel d: P-value represent factor (T, treatment; A, age) comparisons by two-way ANOVA including 21M and 23M groups followed by simple main effects (**p<0.01, *p<0.05 vs. age-matched control); ⁺⁺⁺p<0.001 (ANOVA post-hoc Tukey comparing controls). For box-and-whisker plots, whiskers show minima and maxima within 1.5 interquartile range.

Figure 3. Spermidine ameliorates cardiac function through induction of autophagy.

(a, b) Cardiac autophagic flux assessed by the LC3-II/GAPDH ratio, determined using western blot analysis (Supplementary Fig. 9e) (a) and cardiac tissue levels of spermidine (b) 50 min after intraperitoneal injection of leupeptin or vehicle in 13-month-old C57BL/6J male mice with or without 4 weeks of spermidine supplementation to the drinking water. N=7/13 (Vehicle/Leu) mice/group. Co, Control; Spd, Spermidine; Leu, Leupeptin. (c, d) Young (3-month-old) transgenic mice harboring cardiac-specific expressed tandem-fluorescence mRFP-GFP-LC3 (tf-LC3) were subjected to spermidine treatment for 2 weeks and hearts were analyzed by confocal microscopy 4 hours after intraperitoneal injection of chloroquine (CQ) or vehicle to assess autophagic flux. Representative RFP/GFP/Hoechst overlays (c) and quantification of autophagosomes (orange puncta in c, arrows) and autolysosomes (red puncta in c, arrowheads) (d) are shown. Scale bars represent 50 µm. N=3 mice/group. (e) Mitophagy was assessed in young (6-month-old) or aged (18-month-old) C57BL/6J wild-type mice treated with or without 2 weeks of spermidine supplementation and injected with AAV9-Mito-Keima. The positive ratiometric area (see Supplementary Fig. 9b-d) of Mito-Keima fluorescence (561 nm/457 nm excitation), indicative of mitophagy, was quantified. N=3 mice/group. (f-k) Cardiomyocyte-specific Atg5-deficient mice (Atg5^-/-) were analyzed at 16 weeks of age and compared to age-matched Atg5^+/+ littermates, with or without 12 weeks of spermidine supplementation (see Supplementary Fig. 10). Shown are tibia length-normalized left ventricular mass (LVmass/TL) (f), ejection fraction (EF) (g), representative pressure-volume loops (h), myocardial stiffness constant (end-diastolic pressure-volume relationships [EDPVR] β obtained from exponential fits, Supplementary Fig. 10d) (i), end-systolic elastance (Ees, slope of the end-systolic pressure-volume relationship, Supplementary Fig. 10e) (j) and ventricular-vascular coupling (VVC) (k). N=16/15 and N=12/14 (f, g, Co/Spd) mice, Atg5^+/+ and Atg5^-/-, respectively. N=10/10 and N=9/11 (i-k, Co/Spd) mice, Atg5^+/+ and Atg5^-/-, respectively. P-values represent factor (T, treatment; G, genotype) comparisons by two-way ANOVA (ANCOVA in panel i) followed by simple main effects (***p<0.001, **p<0.01, *p<0.05 and #p<0.06). Dot-plots show means ± s.e.m. For box-and-whisker plots, whiskers show minima and maxima within 1.5 interquartile range.

Figure 4. Spermidine ameliorates salt-induced hypertension and heart failure in Dahl salt-sensitive rats.

(a) Schematic overview of spermidine administration (starting from the age of 7 weeks) to Dahl salt-sensitive rats fed a high-salt diet (8% NaCl) ad libitum. Cardiovascular parameters (b-k) and renal tissue characteristics (l, m) were recorded at the indicated ages. (b) Mean arterial blood pressure (MBP) by the non-invasive tail-cuff method. (N=10 rats/group). Red symbols denote the first time point at which there is a non-significant difference in MBP compared to the peak MBP of the group. (c) Plasma spermidine and ornithine content and global arginine bioavailability ratio (GABR). N=12/12/12, 11/12/12, 11/12/10 (7wk/9wk/9wk+S) rats, left, center and right sub-panels, respectively. (d, e) Echocardiographic assessment of tibia length-normalized LV mass (LVmass/TL) (d) and the ratio of peak early Doppler transmitral flow velocity (E) to the corresponding myocardial tissue Doppler velocity (E') (e). N=10 rats/group. (f-k) Representative pressure-volume loops (f), LV end-diastolic pressure (EDP) (g), myocardial stiffness constant for indexed volumes (end-diastolic pressure-volume relationship [EDPVR] β_i) (h), tibia length (TL)-normalized lung and liver weights (i), ejection fraction (EF) (j), and ventricular-vascular coupling (VVC) (k). N=10/9/9/10/10 (7wk/14wk/14wk+S/19wk/19wk+S) rats. (l) Representative micrographs of 9 rats/group analyzed for renal fibrosis in 19-week-old rats, as assessed by picrosirius red collagen staining. (m) Urinary lipocalin-2 (Lcn-2) levels. N=12/10/10 (7wk/19wk/19wk+S) rats. ***p<0.001, **p<0.01 and *p<0.05 (ANOVA with post-hoc Tukey in panel c, m, Kruskal-Wallis with corrected multiple-comparisons in panel d). Panels b, e, g-k: p-values represent factor (T, treatment; A, age) comparisons by two-way ANOVA (mixed-design in b and e) including 14wk and 19wk groups followed by simple main effects (***p<0.001, **p<0.01, *p<0.05 vs. age-matched control); ⁺⁺⁺p<0.001, ⁺⁺p<0.01, ⁺p<0.05 (ANOVA with post-hoc Tukey comparing controls). Dot- and line-plots show means ± s.e.m. For box-and-whisker plots, whiskers show minima and maxima within 1.5 interquartile range.

Figure 5. Dietary spermidine intake inversely correlates with human cardiovascular disease.

(a-c) Associations of polyamine intake (spermidine or putrescine) in human subjects with death due to heart failure (a), clinically overt heart failure (b) and incident cardiovascular disease (CVD, a composite of acute coronary artery disease, stroke and death to due vascular disease) (c). Hazard ratios (a and c, time-to-event analysis) and odds ratios (b, cross-sectional analysis) are indicated for one standard deviation higher intake of the given polyamine. Models were unadjusted (U) or had multivariable adjustment (M) for age, sex, total caloric intake, current smoking, diabetes, alcohol consumption, and diastolic blood pressure. *Death due to heart failure was defined according to the International Statistical Classification of Diseases and Related Health Problems, 10th revision (ICD-10) codes I50.x, I13.0, I13.2, I11.00, I11.01, or I97.1, and the results represent sub-distribution hazard ratios based on the Fine and Gray model and account for the competing risk of death due to causes unrelated to heart failure. **Diagnosis of clinically overt heart failure (ascertained in 2010) relied on gold standard Framingham criteria. ***See methods for incident CVD criteria. (d) Association of polyamine (spermidine or putrescine) intake with systolic (BPsys) and diastolic (BPdia) blood pressures repeatedly assessed in 829 participants of the Bruneck Study (1995-2010). The effects shown represent the average difference in blood pressure (mmHg) associated with one standard deviation higher intake of the given polyamine (M1) or between the first and third tertile groups (M2) under adjustment for age, sex, and total caloric intake.

Figure 6. Mechanistic model of spermidine-mediated cardioprotection in aging and hypertensive heart failure.

Oral (dietary) supplementation of spermidine improves cardiac function by (i) promoting protective autophagy and mitophagy in cardiomyocytes; (ii) reducing subclinical, chronic inflammation (circulating TNFα levels) that impinges on cardiomyocyte function; (iii) improving systemic arginine bioavailability that may favor the production of the vasodilator nitric oxide (NO), and thus decrease systemic blood pressure; and (iv) inhibiting kidney damage through induction of autophagy. Improved renal function by spermidine treatment may additionally contribute to reduced arterial blood pressure and cardioprotection in the setting of salt-induced hypertension. In conjunction with spermidine’s anti-inflammatory action, its autophagy-dependent effects on cardiomyocytes lead to enhanced mitochondrial volume and function, increased titin phosphorylation and reduced hypertrophy, which in turn result in improved mechano-elastical properties of cardiomyocytes.