In vivo administration of urolithin A and B prevents the occurrence of cardiac dysfunction in streptozotocin-induced diabetic rats

Abstract

Background: Emerging evidence suggests that specific (poly)phenols may constitute new preventative strategies to counteract cell oxidative stress and myocardial tissue inflammation, which have a key role in the patho-physiology of diabetic cardiomyopathy. In a rat model of early diabetes, we evaluated whether in vivo administration of urolithin A (UA) or urolithin B (UB), the main gut microbiota phenolic metabolites of ellagitannin-rich foods, can reduce diabetes-induced microenvironmental changes in myocardial tissue, preventing cardiac functional impairment.

Methods: Adult Wistar rats with streptozotocin-induced type-1 diabetes (n = 29) were studied in comparison with 10 control animals. Diabetic rats were either untreated (n = 9) or subjected to daily i.p. injection of UA (n = 10) or UB (n = 10). After 3 weeks of hyperglycaemia, hemodynamics, cardiomyocyte contractile properties and calcium transients were measured to assess cardiac performance. The myocardial expression of the pro-inflammatory cytokine fractalkine and proteins involved in calcium dynamics (sarcoplasmic reticulum calcium ATPase, phospholamban and phosphorylated phospholamban) were evaluated by immunoblotting. Plasma, urine and tissue distribution of UA, UB and their phase II metabolites were determined.

Results: In vivo urolithin treatment reduced by approximately 30% the myocardial expression of the pro-inflammatory cytokine fractalkine, preventing the early inflammatory response of cardiac cells to hyperglycaemia. The improvement in myocardial microenvironment had a functional counterpart, as documented by the increase in the maximal rate of ventricular pressure rise compared to diabetic group (+18% and +31% in UA and UB treated rats, respectively), and the parallel reduction in the isovolumic contraction time (-12%). In line with hemodynamic data, both urolithins induced a recovery of cardiomyocyte contractility and calcium dynamics, leading to a higher re-lengthening rate (+21%, on average), lower re-lengthening times (-56%), and a more efficient cytosolic calcium clearing (-32% in tau values). UB treatment also increased the velocity of shortening (+27%). Urolithin metabolites accumulated in the myocardium, with a higher concentration of UB and UB-sulphate, potentially explaining the slightly higher efficacy of UB administration.

Conclusions: In vivo urolithin administration may be able to prevent the initial inflammatory response of myocardial tissue to hyperglycaemia and the negative impact of the altered diabetic milieu on cardiac performance.

Keywords: Cardiac performance; Cardiomyocyte mechanics; Diabetes; Ellagitannins; Urolithins.

Fig. 1

Accumulation of urolithin metabolites in tissues of rats treated with UA or UB (n = 5 for each group). Control samples (CTRL and D3 groups) did not contain detectable levels of these metabolites. The fraction under each column accounts for the number of samples where the compound was detected. For Urolithin A and Urolithin B, the y axes on the right refer to the concentration of the two metabolites in the heart tissue. ^§ p < 0.05 significant differences between D3-UA and D3-UB (Mann–Whitney U test). °Outlier values

Fig. 2

Hemodynamic measurements. Mean values ± SEM of left ventricular systolic pressure (LVSP; a), left ventricular end diastolic pressure (LVEDP; b), maximal rate of ventricular pressure rise (+dP/dt_max; c), and decline (−dP/dt_max; d), isovolumic contraction time (IVCT; e), and total cycle duration (Tcycle; f), measured in control rats (CTRL, n = 10), untreated diabetic rats (D3, n = 9), and urolithin A-treated (D3-UA, n = 10) or urolithin B-treated (D3-UB, n = 10) diabetic animals. *p < 0.05 significant differences vs CTRL; ^# p < 0.05 significant differences vs D3 (two-way ANOVA)

Fig. 3

Cell mechanics and calcium transients. Representative examples of sarcomere shortening (a) and corresponding calcium transients (b; normalized traces: fold increase) recorded from CTRL, D3, D3-UA and D3-UB ventricular myocytes. In c–h bar graphs means values ± SEM of sarcomere fraction of shortening (FS; c), maximal rate of shortening (−dL/dt_max; d), maximal rate of re-lengthening (+dL/dt_max; e), time to 10, 50 and 90% of total cycle length (TBL10%, TBL50%, and TBL90%; f), calcium transient amplitude expressed as peak fluorescence normalized to baseline fluorescence (f/f0; g), and time constant of the intracellular calcium decay (tau; h), measured in CTRL (66 and 33 cells, for mechanics and calcium transients respectively), D3 (91 and 57 cells), D3-UA (100 and 63 cells), and D3-UB (102 and 42 cells). *p < 0.05 significant differences vs CTRL; ^# p < 0.05 significant differences vs D3; ^§ p < 0.05 significant differences vs D3-UA (2-factor Nested ANOVA)

Fig. 4

Expression levels of functional proteins and fractalkine by immunoblot assay. Average values ± SEM of SERCA2 (a), PLB (b), SERCA2/PLB ratio (c), PLB-P/PLB ratio (d), and fractalkine (e, CX3CL1) expression levels in left ventricular myocardium of CTRL (n = 2), D3 (n = 3), D3-UA (n = 4), and D3-UB (n = 4), in technical triplicate. Data are expressed in densitometric units. Sets of bands related to protein expression in 4 animals representative of the average behaviour observed in each group are reported in f. In columns 1–4 CTRL, D3, DR-UA, and D3-UB animals. *p < 0.05 significant differences vs CTRL; ^# p < 0.05 significant differences vs D3 (non-parametric statistical test Kruskal–Wallis and U Mann–Whitney test)