Cardiomyocyte p38 MAPKα suppresses a heart-adipose tissue-neutrophil crosstalk in heart failure development

Abstract

Although p38 MAP Kinase α (p38 MAPKα) is generally accepted to play a central role in the cardiac stress response, to date its function in maladaptive cardiac hypertrophy is still not unambiguously defined. To induce a pathological type of cardiac hypertrophy we infused angiotensin II (AngII) for 2 days via osmotic mini pumps in control and tamoxifen-inducible, cardiomyocyte (CM)-specific p38 MAPKα KO mice (iCMp38αKO) and assessed cardiac function by echocardiography, complemented by transcriptomic, histological, and immune cell analysis. AngII treatment after inactivation of p38 MAPKα in CM results in left ventricular (LV) dilatation within 48 h (EDV: BL: 83.8 ± 22.5 µl, 48 h AngII: 109.7 ± 14.6 µl) and an ectopic lipid deposition in cardiomyocytes, reflecting a metabolic dysfunction in pressure overload (PO). This was accompanied by a concerted downregulation of transcripts for oxidative phosphorylation, TCA cycle, and fatty acid metabolism. Cardiac inflammation involving neutrophils, macrophages, B- and T-cells was significantly enhanced. Inhibition of adipose tissue lipolysis by the small molecule inhibitor of adipocytetriglyceride lipase (ATGL) Atglistatin reduced cardiac lipid accumulation by 70% and neutrophil infiltration by 30% and went along with an improved cardiac function. Direct targeting of neutrophils by means of anti Ly6G-antibody administration in vivo led to a reduced LV dilation in iCMp38αKO mice and an improved systolic function (EF: 39.27 ± 14%). Thus, adipose tissue lipolysis and CM lipid accumulation augmented cardiac inflammation in iCMp38αKO mice. Neutrophils, in particular, triggered the rapid left ventricular dilatation. We provide the first evidence that p38 MAPKα acts as an essential switch in cardiac adaptation to PO by mitigating metabolic dysfunction and inflammation. Moreover, we identified a heart-adipose tissue-immune cell crosstalk, which might serve as new therapeutic target in cardiac pathologies.

Keywords: Cardiac inflammation; Lipolysis; Metabolic dysfunction; Pressure overload; p38 MAPKα.

Fig. 1

Verification of knockout and cardiac function. A Protein expression of p38 MAPKα and p38 MAPKγ in cardiac tissue of control and iCMp38αKO animals. n = 4, data represent mean ± SD, unpaired, two-tailed t test, *p < 0.05. B Transcript expression of p38 MAPK isoforms in whole heart tissue normalized to Nudc. n = 4–5, data represent mean ± SD, unpaired, two-tailed t test, ***p < 0.001. C Phosphorylation of cardiac p38 MAPKα after 48 h AngII stimulation compared to unstimulated animals. Phospho values were normalized to total p38 MAPKα signal. n = 6–8, data represent mean ± SD, unpaired, two-tailed t test, **p = 0.006. D Mean arterial pressure measured in the first 30 h after AngII pump implantation in control mice. n = 5, data represent mean ± SD. E Phosphorylation of p38 MAPKα target MK2 at T334 in control and KO animals with and without AngII stimulation. AngII was applied for 48 h. n = 5–7, data represent mean ± SD, mixed-effects analysis. Bonferroni’s multiple comparisons test was used to compared control vs. KO at each timepoint, reported are p values below 0.05, ***p < 0.001. F Schematic of experimental outline for analysis of cardiac function. G B- and M-Mode images of parasternal long axes of control and iCMp38αKO hearts at baseline (left panel) and after 2d of AngII treatment (right panel). B-Mode images show wall displacement. H Ejection fraction (EF), end diastolic (EDV) and end systolic volume (ESV) of control and iCMp38αKO hearts at baseline and after 12, 24 and 48 h of AngII treatment. Control: grey square, KO: black triangle, n = 5–17, Data represent mean ± SD, two-way ANOVA followed by Bonferroni’s multiple comparisons test was used to compare control vs. KO at each timepoint, reported are p values below 0.05, **p < 0.01, ***p < 0.001

Fig.2

Transcriptomic analysis of gene in control and iCMp38αKO hearts after 2 days of AngII treatment, n = 3–4. A Hierarchical cluster analysis of differentially expressed genes (FC ≥ 1.5, p < 0.01). B Canonical pathways identified by ingenuity pathway analysis (IPA) of differentially expressed genes. The top 12 pathways are shown. Stacked bars represent percentages of regulated genes of all genes assembled under the specified term (red: upregulated; green: downregulated; gray unaltered). Blue: − log p value of regulation. C IPA analysis of upstream regulators. Upstream regulators with an absolute activation Z-score ≥ 4, − logP ≤ 4 are shown. D IPA analysis of altered “Diseases and Functions”. The top activated (red) and suppressed (green) functional terms (absolute activation Z-score ≥ 2, − logP ≤ 4) are listed. E Kinetic analysis of gene expression for selected genes as indicated. Data show the expression levels for control (gray squares) and iCMp38KO hearts (black triangles) over the indicated time course normalized to Nudc (n = 4 each data point). Data represent mean ± SD. Two-way ANOVA followed by Bonferroni’s multiple comparisons test was used to compare control vs. KO at each timepoint, reported are p values below 0.05, *p < 0.05, **p < 0.01, ***p < 0.001)

Fig. 3

Time course analysis of lipid accumulation. A Representative magnifications of hearts after sudan red 7B staining of control and iCMp38αKO hearts after 12, 24 and 48 h of AngII treatment. n = 5–7. B Anti-perilipin 2 immunofluorescence staining of control and iCMp38αKO hearts after 48 h of angiotensin II treatment. Red: perilipin 2, green: wheat germ agglutinin (WGA), blue: DAPI. Arrows point to lipid droplets. C Fatty acid profile of control and iCMp38αKO hearts at baseline and after 48 h of AngII treatment. n = 4–6. Kruskal–Wallis test with Dunn’s multiple comparisons test was used to compare each group to another. Reported are p values below 0.05. *p < 0.05, **p < 0.01, ***p < 0.001

Fig. 4

Time course analysis of inflammation. A Representative magnifications of hearts after anti-Ly6G immunofluorescence staining of control and iCMp38αKO hearts after 12, 24 and 48 h of AngII treatment. n = 5–7. B Colocalization of lipid accumulation and neutrophil infiltration in control and iCMp38αKO mice after 24 h AngII treatment. C Gene expression level of Il6, Il1β and Cxcl5 over time course in control and iCMp38αKO hearts. n = 4. Data represent mean ± SD. Statistical analysis was performed using two-way ANOVA followed by Bonferroni’s multiple comparisons test to compare control vs. KO at each timepoint. Reported are p values below 0.05. *p < 0.05, **p < 0.01, ***p < 0.001. E TUNEL positive nuclei (arrows) of control and iCMp38αKO hearts after 48 h of AngII treatment and corresponding quantification. n = 4–6. Data are presented as mean ± SD. Statistical analysis was performed using unpaired, two-tailed t test. **p < 0.01

Fig. 5

Atglistatin treatment: A schematic of the experimental outline. B Plasma glycerol level of control and iCMp38αKO mice (n = 6). C Representative images of sudan red 7B staining and quantification of lipid accumulation of these images of control and iCMp38αKO hearts (n = 6–8). D EF, fractional shortening (FS) and stroke volume (SV) of control and iCMp38αKO hearts. n = 10–13, unpaired, two-tailed t test was used to compare KO vs. KO Atglistatin. E Flow cytometric analysis of cardiac neutrophils (CD11b⁺Ly6G⁺). F Flow cytometric analysis of cardiac immune cells (total macrophages (CD11b⁺Ly6G⁻CD64⁺), CCR2⁻, CCR2⁺, Ly6C^lo and Ly6C^hi macrophages, total monocytes (CD11b⁺Ly6G⁻CD64⁻MHCII⁻), Ly6C^lo and Ly6C^hi monocytes, B-cells (CD45⁺CD19⁺CD3⁻) and CD4⁺- and CD8⁺ T-cells (CD45⁺CD3⁺CD19⁻)) (n = 4–7). G) Correlation between cardiac neutrophil number and EF. All data shown are from animals after 48 h of AngII treatment either with or without pretreatment with Atglistatin. Data represent mean ± SD, Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test to perform intergroup comparisons. The following groups were compared to each other: control vs. KO, control vs. control Atglistatin, control Atglistatin vs. KO Atglistatin, KO vs. KO Atglistatin. Correlation between cardiac neutrophil number and EF was tested by Pearson correlation. Reported are p values below 0.05. *p < 0.05, **p < 0.01, ***p < 0.001

Fig. 6

Granulocyte depletion: A schematic of experimental outline. B FACS analysis of neutrophils in blood of control and iCMp38αKO mice 60–64 h after treatment with anti-Ly6G antibody or isotype control (Rat IgG2A). Representative FACS plots are shown. C EF, cardiac output (CO), EDV, ESV and heart rate of control and iCMp38αKO mice after 48 h AngII treatment with either anti-Ly6G antibody or isotype control (IT). n = 6–8. D Gene expression level of Cxcl5 and Cxcr2 in control and iCMp38αKO hearts after 48 h AngII treatment with either anti-Ly6G antibody or isotype control (IT). n = 4–8, Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test to perform intergroup comparisons. The following groups were compared to each other: control IT vs. KO IT, control IT vs. control anti-Ly6G, control anti-Ly6G vs. KO anti-Ly6G, KO IT vs. KO anti-Ly6G. Reported are p values below 0.05 *p < 0.05, **p < 0.01, ***p < 0.001