Systolic and diastolic dysfunction is exacerbated by age and spinal cord injury in male and female mice with central nervous system serotonin deficiency

Abstract

The present study was designed to explore whether the depletion of serotonin (5-HT) in the central nervous system (CNS^5-HT) leads to systolic and diastolic dysfunction and whether this dysfunction is exacerbated by sex, age and spinal cord injury. Echocardiographic assessment of systolic and diastolic function was completed in young and old male and female tryptophan hydroxylase 2 knockout (TPH2^-/-) and wild-type (TPH2^+/+) mice with intact spinal cords, as well as in C₂ spinal cord hemisected young TPH2^-/- and TPH2^+/+ mice. In addition, lumbar sympathetic nervous system activity was recorded in elderly male and female intact TPH2^-/- and TPH2^+/+ mice. Systolic and diastolic dysfunction was evident in young TPH2^-/- mice, including a higher left ventricular mass (P < 0.001), left ventricular outflow parameters (e.g. peak velocity) and E/A (P < 0.001). Reductions in ejection fraction and fractional shortening were also evident (P < 0.001), although stroke volume and cardiac output were maintained. The assessed dysfunction was exacerbated by age and spinal cord injury, resulting in reductions in cardiac output (P ≤ 0.01). The dysfunction was accompanied by increases in sympathetic burst height (P = 0.038) and incidence (P = 0.001). Reductions in CNS^5-HT are coupled to systolic and diastolic dysfunction, which is exacerbated by age and spinal cord injury. This dysfunction is coupled to increases in sympathetic nervous system activity in elderly mice. Our findings are an initial step toward determining whether reductions in CNS^5-HT are a unifying mechanism that links central sleep apnoea, sympathoexcitation and heart failure in intact and spinal cord injured individuals. KEY POINTS: Reductions in central nervous system serotonin (CNS^5-HT) may contribute to systolic and diastolic dysfunction. This dysfunction may be linked to increases in sympathetic nervous system activity and exacerbated by sex, age and spinal cord injury. Echocardiographic assessment of systolic and diastolic function was completed in young and old male and female intact TPH2^+/+ and TPH2^-/- mice, as well as in C₂ spinal cord hemisected young mice. Lumbar sympathetic nervous system activity was also recorded in elderly male and female intact TPH2^+/+ and TPH2^-/- mice. Systolic and diastolic dysfunction was evident in young TPH2^-/- mice. This dysfunction was exacerbated by age and spinal cord injury. The cardiac dysfunction was accompanied by increases in lumbar sympathetic nervous system activity. Our findings are an initial step toward determining whether reductions in CNS^5-HT is a unifying mechanism that links central sleep apnoea, sympathoexcitation and heart failure in intact and spinal cord injured individuals.

Keywords: burst height; burst incidence; echocardiogram; heart failure; lumbar sympathetic nerve activity; mice; serotonin.

Figure 1. Left ventricular mass is increased whereas ejection fraction and fractional shortening are decreased in TPH2^−/− mice

Box plots, coupled with scatterplots of individual measures showing (A) left ventricular mass, (C) ejection fraction and (E) fractional shortening measured from the three main groups, sex (males vs. females), genotype (TPH2^+/+ vs. TPH2^−/−) and age (young vs. old). In addition, similar plots showing (B) left ventricular mass, (D) ejection fraction and (F) fractional shortening for each sex × genotype × age interaction [young and old male and female TPH2^+/+ (white boxes and circles) and TPH2^−/− (grey boxes and circles) mice]. A, left ventricular mass was higher in males compared to females, TPH2^−/− compared to TPH2^+/+ mice and in old compared to young mice. After standardizing for weight (not shown) the observed differences between the TPH2^−/− and TPH2^+/+ mice remained. C and E, ejection fraction and fractional shortening were similar in males and females. By contrast, ejection fraction and fractional shortening were lower in TPH2^−/− compared to TPH2^+/+ mice and in old compared to young mice. Three‐way analysis of variance was used to determine statistical differences. Young TPH2^+/+ male mice (n = 22), old TPH2^+/+ male mice (n = 10), young TPH2^+/+ female mice (n = 17), old TPH2^+/+ female mice (n = 9), young TPH2^−/− male mice (n = 12), old TPH2^−/− male mice (n = 9), young TPH2^−/− female mice (n = 12) and old TPH2^−/− female mice (n = 10).

Figure 2. Left ventricular outflow velocity, peak velocity and E/A are increased in TPH2^−/− mice

Box plots, coupled with scatterplots of individual measures showing (A) left ventricular outflow velocity time integral (LVOT VTI), (C) left ventricular peak velocity and (E) E/A [the ratio of peak velocity blood flow from left ventricular relaxation in early diastole (E wave) to peak velocity flow in late diastole caused by atrial contraction (A wave)] measured from the three main groups, sex (males vs. females), genotype (TPH2^+/+ vs. TPH2^−/−) and age (young vs. old). In addition, similar plots showing (B) left ventricular outflow velocity time integral, (D) left ventricular peak velocity and (F) E/A for each sex × genotype × age interaction [young and old male and female TPH2^+/+ (white boxes and circles) and TPH2^−/− (grey boxes and circles) mice]. A and B, LVOT VTI was lower in old compared to young mice. LVOT VTI was lower in TPH2^+/+ male mice compared to TPH2^+/+ female mice and was higher in TPH2^−/− male mice compared to TPH2^+/+ male mice. C, left ventricular peak velocity was higher in TPH2^−/− mice compared to TPH2^+/+ mice and was lower in old compared to young mice. E, the E/A ratio was similar in males and females. The E/A was higher in TPH2^−/− compared to TPH2^+/+ mice, whereas the E/A ratio was similar in young and old mice. Three‐way analysis of variance was used to determine statistical differences. LVOT VTI and LVOT peak velocity: young TPH2^+/+ male mice (n = 22), old TPH2^+/+ male mice (n = 10), young TPH2^+/+ female mice (n = 17), old TPH2^+/+ female mice (n = 9), young TPH2^−/− male mice (n = 12), old TPH2^−/− male mice (n = 9), young TPH2^−/− female mice (n = 12) and old TPH2^−/− female mice (n = 10). E/A: young TPH2^+/+ male mice (n = 22), old TPH2^+/+ male mice (n = 9), young TPH2^+/+ female mice (n = 15), old TPH2^+/+ female mice (n = 8), young TPH2^−/− male mice (n = 12), old TPH2^−/− male mice (n = 6), young TPH2^−/− female mice (n = 12) and old TPH2^−/− female mice (n = 8).

Figure 3. Heart rate, stroke volume and cardiac output are reduced in elderly mice

Box plots, coupled with scatterplots of individual measures showing (A) heart rate, (C) stroke volume and (E) cardiac output measured from the three main groups, sex (males vs. females), genotype (TPH2^+/+ vs. TPH2^−/−) and age (youngvs. old). In addition, similar plots showing (B) heart rate (D) stroke volume and (F) cardiac output for each sex × genotype × age interaction [young and old male and female TPH2^+/+ (white boxes and circles) and TPH2^−/− (grey boxes and circles) mice]. A, heart rate was lower in old compared to young mice and (C) stroke volume was lower in old compared to young mice, although statistical significance was not achieved (P = 0.053). E, cardiac output was lower in old compared to young mice. Three‐way analysis of variance was used to determine statistical differences. Young TPH2^+/+ male mice (n = 22), old TPH2^+/+ male mice (n = 10), young TPH2^+/+ female mice (n = 17), old TPH2^+/+ female mice (n = 9), young TPH2^−/− male mice (n = 12), old TPH2^−/− male mice (n = 9), young TPH2^−/− female mice (n = 12) and old TPH2^−/− female mice (n = 10).

Figure 4. Burst height, frequency and incidence is increased in TPH2^−/− mice

A, a raw record showing lumbar sympathetic nerve activity in one elderly TPH2^+/+ mouse (left) and one elderly TPH2^−/− mouse (right). Box plots, coupled with scatterplots, showing individual measures of (B) lumbar sympathetic nerve burst height, (C) burst frequency (D) burst incidence in TPH2^+/+ (white boxes and circles) and TPH2^−/− (grey boxes and circles) mice. Burst height was higher in TPH2^−/− male mice compared to TPH2^+/+ male mice and TPH2^−/− female mice. Burst frequency and incidence were increased in TPH2^−/− male and TPH2^−/− female mice compared to their TPH2^+/+ counterparts. Two‐way analysis of variance was used to determine statistical differences. TPH2^+/+ mice (n = 13; 7 male and 6 female); TPH2^−/− mice (n = 12; 6 male and 6 female).

Figure 5. TNFα is higher in TPH2^−/− mice

Box plots, coupled with scatterplots of individual measures, showing the expression of TNFα in the left ventricle of TPH2^+/+ and TPH2^−/− mice. The densitometry data was normalized to the TPH2^+/+ group. Inset: examples of western blotting were completed to assess the relative abundance of TNFα using β‐actin as the loading control. The expression of TNFα is increased in TPH2^−/− compared to TPH2^+/+ mice. The following modifications were made to the western blot for the purpose of presentation. One additional column in the western blot was deleted because this column was unrelated to the groups shown in the scatterplot. In addition, in the original western blot, the position of the TPH2^+/+ column was to the right of the TPH2^−/− column. The position of the columns was reversed to conform with the presentation of the groups shown in the scatterplot and other plots throughout the manuscript (i.e. TPH2^+/+ to the left of TPH2^−/−). An unpaired t test was used to determine statistical differences. TPH2^+/+ mice (n = 10; 6 male and 4 female); TPH2^−/− mice (n = 6; 2 male and 4 female). [Colour figure can be viewed at wileyonlinelibrary.com]

Figure 6. Left ventricular mass is highest whereas ejection fraction and fractional shortening is lowest in spinal cord injured TPH2^−/− mice

Box plots, coupled with scatterplots of individual measures, showing (A) left ventricular mass, (B) ejection fraction and (C) fractional shortening in spinal cord injured or sham injured mice TPH2^+/+ (white boxes and circles) and TPH2^−/− (grey boxes and circles) mice. A, left ventricular mass was greater in spinal cord injured TPH2^−/− mice compared to TPH2^+/+ mice. This relationship remained unaltered after left ventricular mass was standardized to mouse weight (not shown). The left ventricular mass was lower in spinal cord injured TPH2^+/+ mice compared to sham. However, this difference was not evident once left ventricular mass was standardized to weight. The increase in left ventricular mass of spinal cord injured TPH2^−/− compared to sham did not achieve statistical significance. Nonetheless, a statistically significant increase was evident after left ventricular mass was standardized to mouse weight (not shown). B and C, the ejection fraction and fractional shortening were lower in the spinal cord injured TPH2^−/− mice compared to TPH2^+/+ mice. Likewise, the ejection fraction and fractional shortening were lower in the TPH2^−/− and TPH2^+/+ spinal cord injured animals compared to sham. Two‐way analysis of variance was used to determine statistical differences between the injured and sham injured mice, and between the injured and young and old uninjured mice. Spinal cord injured TPH2^+/+ mice (n = 12), spinal cord injured TPH2^−/− mice (n = 10), sham spinal cord injured TPH2^+/+ mice (n = 8) and sham spinal cord injured TPH2^−/− mice (n = 2).

Figure 7. E/A is higher in TPH2^−/− mice following spinal cord injury

Box plots, coupled with scatterplots of individual measures, showing (A) left ventricular outflow velocity time integral (LVOT VTI), (B) left ventricular peak velocity and (C) E/A [the ratio of peak velocity blood flow from left ventricular relaxation in early diastole (E wave) to peak velocity flow in late diastole caused by atrial contraction (A wave)] in spinal cord injured or sham injured mice TPH2^+/+ (white boxes and circles) and TPH2^−/− (grey boxes and circles) mice. A and B, the left ventricular outflow velocity time integral and peak velocity were similar in the spinal cord injured TPH2^−/− mice compared to the TPH2^+/+ mice. Reductions in the velocity time integral and peak velocity were evident in the spinal cord injured TPH2^−/− and TPH2^+/+ mice compared to sham. C, the E/A was higher in the spinal cord injured TPH2^−/− mice compared to TPH2^+/+ mice. Likewise, the E/A was higher in the spinal cord injured TPH2^−/− and TPH2^+/+ mice compared to sham. Two‐way analysis of variance was used to determine statistical differences between the injured and sham injured mice, and between the injured and young and old uninjured mice. Spinal cord injured TPH2^+/+ mice (n = 12), spinal cord injured TPH2^−/− mice (n = 10), sham spinal cord injured TPH2^+/+ mice (n = 8) and sham spinal cord injured TPH2^−/− mice (n = 2).

Figure 8. Stroke volume and cardiac output are lowest in spinal cord injured TPH2^−/− mice

Box plots, coupled with scatterplots of individual measures, showing (A) heart rate, (B) stroke volume and (C) cardiac output in spinal cord injured or sham injured mice TPH2^+/+ (white boxes and circles) and TPH2^−/− (grey boxes and circles) mice. A, heart rate was similar in the spinal cord injured TPH2^−/− mice compared to TPH2^+/+ mice and was similar in the spinal cord injured animals compared to sham. B and C, stroke volume and cardiac output were lower in the spinal cord injured TPH2^−/− mice compared to TPH2^+/+ mice. Stroke volume and cardiac output was lower in the spinal cord injured TPH2^+/+ mice compared to sham, while cardiac output was lower in the TPH2^−/− spinal cord injured animals compared to sham. Two‐way analysis of variance was used to determine statistical differences between the injured and sham injured mice, and between the injured and young and old uninjured mice. Spinal cord injured TPH2^+/+ mice (n = 12), spinal cord injured TPH2^−/− mice (n = 10), sham spinal cord injured TPH2^+/+ mice (n = 8) and sham spinal cord injured TPH2^−/− mice (n = 2).