Abstract

Metabolic remodeling plays an essential role in the pathophysiology of heart failure (HF). Many studies have shown that the disruption of phosphoinositide-dependent protein kinase-1 (PDK1) caused severe and lethal HF; however, the metabolic pattern of PDK1 deletion remains ambiguous. 1H nuclear magnetic resonance-based metabolomics was applied to explore the altered metabolic pattern in Pdk1-deficient mice. Principle component analysis showed significant separation as early as 4 weeks of age, and dysfunction of metabolism precedes a morphological change in Pdk1-deficient mice. A time trajectory plot indicated that disturbed metabolic patterns were related to the pathological process of the HF in Pdk1-deficient mice, rather than the age of mice. Metabolic profiles demonstrated significantly increased levels of acetate, glutamate, glutamine, and O-phosphocholine in Pdk1 deletion mice. Levels of lactate, alanine, glycine, taurine, choline, fumarate, IMP, AMP, and ATP were significantly decreased compared with controls. Furthermore, PDK1 knockdown decreased the oxygen consumption rate in H9C2 cells as determined using a Seahorse XF96 Analyzer. These findings imply that the disruption of metabolism and impaired mitochondrial activity might be involved in the pathogenesis of HF with PDK1 deletion.

Introduction

Cardiovascular diseases continue to be the leading cause of morbidity and mortality worldwide [1]. Heart failure (HF) is a progressive, complex, end-stage of various forms of cardiovascular diseases [2]. Ventricular remodeling, abnormal myocyte calcium cycling, inflammation, accumulation of oxidative stress, apoptosis, and mutations have been shown to contribute to the progress of HF [3]. Metabolic dysfunction has also been associated with the onset and progression of HF, such as alterations in fatty acid oxidation, glucose, ketone body, and amino acid metabolism during HF [47]. Thus, we believe that exploring potential metabolic mechanisms is an important part for further understanding of HF pathogenesis. Metabolomics, a dynamic portrait of the metabolic status of living systems, offers potential biomarkers and underlying pathophysiology of diseases. Nuclear magnetic resonance (NMR) spectroscopy is an attractive method for studying metabolomics due to fast analysis, simple sample preparation and high reproducibility [8]. It has been used as a promising tool in research field of heart disease [9,10].

3-Phosphoinositide-dependent protein kinase-1 (PDK1) is a key member of the AGC serine/threonine kinase family. Many studies have shown that the PDK1/AGC kinase signaling pathway regulates physiological processes relevant to proliferation, growth, autophagy, and apoptosis [11]. Moreover, PDK1 is involved in HF and pathologic heart remodeling. Mice with Pdk1 deletion in cardiac muscle displayed multiple abnormalities that included thinner ventricular walls, smaller cardiomyocytes, vascular remodeling, and with severe and lethal HF [12,13]. Budas et al. [14] reported that PDK1 regulated mitochondrial through PKB-GSK-3β in the heart. Feng et al. [11] revealed the role of PDK1 in cardiovascular development through the activation of Akt and Snail. Liu et al. [15] suggested that differentially lncRNA expression profiles may be involved in the pathological processes of HF in Pdk1 deletion mice. However, the mechanisms of HF in Pdk1 knockout mice are still elusive, and the underlying metabolism profiles of Pdk1 deficiency remain to be resolved.

To understand the function of PDK1 in metabolism characteristics, time-related metabolic alterations in cardiac tissues were explored by NMR in comparison with controls. The results have revealed the essential role of PDK1 in metabolism profiles. Furthermore, the mitochondrial function was assessed as well. The present study could help to elucidate new underlying pathophysiological mechanisms of HF.

Experimental procedures

Animals

Mice used in this study were on C57BL/6 genetic background and were maintained in a specific pathogen-free colony of the Laboratory Animal Center of Wenzhou Medical University (Wenzhou, China) on a 12-h light/dark cycle and given free access to food and water. Animal experiments were approved by the Institutional Animal Care and Use Committee of Wenzhou Medical University. Pdk1 floxed mice were kindly provided by Dr Zhongzhou Yang (Ministry of Education Key Laboratory of Model Animal for Disease Study, Model Animal Research Center, Nanjing). Mice with cardiomyocyte-specific deletion of Pdk1 (Pdk1F/F; αMHC-Cre) were used in this study, and Pdk1F/F littermates without the αMHC-Cre transgene used as control.

Cardiac tissue preparation and extraction

Animals were humanely sacrificed by 4, 5, 6, and 7 weeks of age. Specimens of the heart were dissected immediately, frozen in liquid nitrogen, and stored at −80°C until use. Cardiac tissues were homogenized with methanol (2 ml/g), chloroform (4 ml/g), and distilled water (6 ml/g) at 4°C. Then, the mixture was kept on ice for 15 min and centrifuged at 10 000 g for 15 min at 4°C. The supernatant was lyophilized for 48 h and resuspended in 500 µl of D2O containing 0.05% sodium trimethylsilyl propionate-d4 (TSP) for NMR spectroscopy.

NMR-based metabolomic analysis

The 1H NMR experiments were performed with a Bruker AVANCE III 600 MHz NMR spectrometer. Spectra were acquired using a one-dimensional Carr–Purcell–Meiboom–Gill pulse sequence with water suppression. 1H NMR acquisition parameters were set as follows: temperature, 298 K; data points, 32 K; spectral width, 12 000 Hz; relaxation delay, 4 s; acquisition time, 2.65 s/scan. Spectra were zero-filled at 64 K, and 0.3 Hz exponential line-broadening function was used to the free induction decay prior to the Fourier transformation.

The spectra were corrected for phase and baseline and referenced to the methyl peaks of TSP (CH3, δ1.314) at 0 ppm using Topspin 2.1 software (Bruker BioSpin, Rheinstetten, Germany). The spectral was sub-divided and integrals of segments for each NMR spectrum were normalized to the total sum of the peak integrals as our previous study [16]. Metabolic pathways were produced based on the KEGG database (www.genome.jp/kegg/) and HMDB (http://www.hmdb.ca/metabolites/).

Cell culture and transfection assay

H9C2 cells were maintained in DMEM and supplemented with 10% fetal bovine serum (Gemini Bio-Products). Cells were incubated at 37°C in 5% CO2 and 95% air. Small interfering RNAs (siRNAs) were obtained from GenePharma (Shanghai, China). The transfection experiment was performed with GenMute reagent (SignaGen), according to the manufacturer's instructions. The siRNA target sequence for PDK1 was 5′-GCCAACUCAUUUGUAGGAAtt-3′ (siRNA1), 5′-GGAUAAGCGAAAGGGUUUAtt-3′ (siRNA2).

RNA isolation and quantitative real-time PCR

The total RNA was extracted using TRIzol reagent (Invitrogen), and the PrimeScript RT reagent kit (TaKaRa) was used to obtain cDNA samples. Quantitative real-time PCR was performed with SYBR Ex Taq (TaKaRa). All genes examined were normalized to GAPDH. Primer sequences were as follows: HK2, CTCGCATATGATCGCCTGCT (F), GGTGTGTGGTAGCTCCTAGC (R); PDH, ATGGCTTCACCTTCACTCGG (F), GTACATGTGCATTGACCCGC (R); IDH1, GTGGGCGTCAAGTGTGCTA (F), CCACCCAGAATGTTTCGGATG (R); IDH2, GGAGAAGCCGGTAGTGGAGAT (F), GGTCTGGTCACGGTTTGGAA (R); SDH, GCTGCGTTCTTGCTGAGACA (F), ATCTCCTCCTTAGCTGTGGTT (R); FH, GAATGGCAAGCCAAAATTCCTT (F), TCTTACGGTCTGAGCACCATAA (R); MDH, GAACCAATCAGAGTCCTTGTGAC (F), GGCACAGTCTTGCAGTTCCA (R).

Measurement of mitochondrial respiration

The Seahorse XF96 Extracellular Flux Analyzer (Seahorse Bioscience, Chicopee, MA, U.S.A.) was used to assess the bioenergetic. Thirty-six hours post-transfection of siRNA, cells were seeded into 96-well XF Seahorse cell culture plates. XF assay medium supplemented with 5.5 mM/l pyruvate, 1.25 mM/l glucose, and 4 mM/l glutamine was used in the test. The subsequent assays were performed as previously described [17,18]. The oxygen consumption rate (OCR) is reported in pmol/minute, and the results were normalized to protein content.

Histopathological examination

Cardiac tissue was fixed in 10% formalin before embedded in paraffin. Serial sections of the embedded specimens were stained with hematoxylin and eosin as conventionally conducted. Histopathological changes in cardiac tissues were assessed randomly.

Statistical analysis

Multivariate statistical analysis was carried out using the SIMCA 13.0 software package (Umetrics, Umeå, Sweden). The time trajectory plot and principal component analysis (PCA) were performed to study the overall metabolic pattern between the Pdk1F/F; αMHC-Cre mice and controls. Moreover, corresponding loading plots were used to identify the metabolites contributed to the separation.

The statistical analyses were performed by SPSS software (version 16.0). Data were presented as mean ± standard deviation (SD). An independent sample t-test was applied to compare metabolite levels between two groups. A value of P < 0.05 was considered statistically significant.

Results

Morphological evaluation in cardiac tissue

To examine the deficiency of PDK1, we performed western blot. Significantly reduced PDK1 protein levels were observed in the heart of Pdk1F/F; αMHC-Cre (Figure 1A). To explore the abnormalities of myocardium, histological analysis on cardiac tissue derived from Pdk1F/F; αMHC-Cre and Pdk1F/F mice of 5, 6, and 7 weeks of age was performed. By 5 and 6 weeks of age, no significant changes of cardiomyocyte were observed between the two groups (Figure 1B,C). However, the histological changes of the tissue were observed at 7 weeks, and the related area of cardiomyocyte of 7-week-old Pdk1F/F; αMHC-Cre was significantly lower (Figure 1B,C). The result was consistent with previous reports that Pdk1 deletion mice with smaller cardiomyocytes [13].

Histological examination of cardiac tissues.

Figure 1.
Histological examination of cardiac tissues.

(A) Western blot analysis of Pdk1 expression in hearts from wild-type and Pdk1-deficient mice. Actin was tested as a loading control. (B) Representative graphs of HE staining of cardiac tissue in Pdk1F/F; αMHC-Cre and Pdk1F/F mice at 5, 6, and 7 weeks of age. (C) The relative cell area of cardiomyocyte was presented as mean ± SD.

Figure 1.
Histological examination of cardiac tissues.

(A) Western blot analysis of Pdk1 expression in hearts from wild-type and Pdk1-deficient mice. Actin was tested as a loading control. (B) Representative graphs of HE staining of cardiac tissue in Pdk1F/F; αMHC-Cre and Pdk1F/F mice at 5, 6, and 7 weeks of age. (C) The relative cell area of cardiomyocyte was presented as mean ± SD.

NMR-based metabonomics of cardiac tissue

Representative 1H NMR spectra in the heart tissue of mice are displayed in Figure 2A,B. 1H NMR spectra of cardiac tissue extracts contained many metabolites, including pantothenate, leucine, isoleucine, valine, thymol, 3-hydroxybutyrate, 5,6-dihydrothymine, lactate, alanine, acetate, glutamate, glutamine, succinate, glutathione, asparagine, creatine, choline, PC (O-phosphocholine), taurine, glycine, inosine, AMP, nicotinurate, fumarate, formate, IMP, ATP, and ADP.

NMR spectra of cardiac samples.

Figure 2.
NMR spectra of cardiac samples.

Typical 600 MHz 1H NMR spectra obtained from the cardiac extracts in Pdk1F/F; αMHC-Cre (A) and controls (B). Pan, pantothenate; Leu, leucine; Ile, isoleucine; Val, valine; 3-HB, 3-hydroxybutyrate; DHT, 5,6-dihydrothymine; Lac, lactate; Ala, alanine; Ace, acetate; Glu, glutamate; Gln, glutamine; Suc, succinate; GSH, glutathione; Asp, asparagine; Cre, creatine; Cho, choline; PC, O-phosphocholine; Tau, taurine; Gly, glycine; Ino, inosine; NIC, nicotinurate; Fum, fumarate; For, formate.

Figure 2.
NMR spectra of cardiac samples.

Typical 600 MHz 1H NMR spectra obtained from the cardiac extracts in Pdk1F/F; αMHC-Cre (A) and controls (B). Pan, pantothenate; Leu, leucine; Ile, isoleucine; Val, valine; 3-HB, 3-hydroxybutyrate; DHT, 5,6-dihydrothymine; Lac, lactate; Ala, alanine; Ace, acetate; Glu, glutamate; Gln, glutamine; Suc, succinate; GSH, glutathione; Asp, asparagine; Cre, creatine; Cho, choline; PC, O-phosphocholine; Tau, taurine; Gly, glycine; Ino, inosine; NIC, nicotinurate; Fum, fumarate; For, formate.

To investigate the metabolic patterns, cardiac tissues collected from Pdk1 deletion mice and age-matched controls were detected by NMR-based metabonomic. The PCA data illustrated that the clear discrimination metabolic patterns of cardiac tissues between the two groups (Figure 3, left). More importantly, the results suggested that Pdk1F/F; αMHC-Cre mice exhibited metabolic pattern dysfunction as early as the 4 weeks of age (Figure 3A). To identify metabolites contributed to separation between the groups, the corresponding loading plots were applied. The plots showed that lactate, glutamine, glutamate, taurine, creatine, glycine, and alanine are among the major contributors to the separation (Figure 3, right).

Pattern recognition analysis of cardiac samples at different time points.

Figure 3.
Pattern recognition analysis of cardiac samples at different time points.

PCA (left) and its corresponding plots (right) for the models discriminating the Pdk1F/F; αMHC-Cre (red dots) and time-matched controls (black squares) at 4 weeks (Pdk1F/F; αMHC-Cre: n = 12, Pdk1F/F: n = 9) (A), 5 weeks (Pdk1F/F; αMHC-Cre: n = 8, Pdk1F/F: n = 7) (B), 6 weeks (Pdk1F/F; αMHC-Cre: n = 11, Pdk1F/F: n = 10) (C), and 7 weeks (Pdk1F/F; αMHC-Cre: n = 11, Pdk1F/F: n = 7) (D).

Figure 3.
Pattern recognition analysis of cardiac samples at different time points.

PCA (left) and its corresponding plots (right) for the models discriminating the Pdk1F/F; αMHC-Cre (red dots) and time-matched controls (black squares) at 4 weeks (Pdk1F/F; αMHC-Cre: n = 12, Pdk1F/F: n = 9) (A), 5 weeks (Pdk1F/F; αMHC-Cre: n = 8, Pdk1F/F: n = 7) (B), 6 weeks (Pdk1F/F; αMHC-Cre: n = 11, Pdk1F/F: n = 10) (C), and 7 weeks (Pdk1F/F; αMHC-Cre: n = 11, Pdk1F/F: n = 7) (D).

To further evaluate the time-dependent interventional effect of Pdk1 deletion, the time trajectory plot was conducted. The trajectory plot showed distinguishable of metabolic pattern between the two groups (Figure 4A). Results from the Pdk1F/F; αMHC-Cre mice displayed obvious trajectory space from 4 to 7 weeks, while the controls revealed a huddled trajectory plot over time, indicating that the metabolic profile changes were related to the pathological process of the HF, rather than the age of mice. The age-dependent metabolic profiles indicated continuous metabolic alteration overtime, which were manifestations of the pathological processes. Figure 4B showed that lactate, creatine, glutamate, taurine, AMP, ADP, and Ino (inosine) are among the major contributors to the separation between groups at PC1. Figure 4C showed that lactate, creatinine, taurine, AMP, ADP, and Ino are responsible for the discrimination overtime at PC2.

Time trajectory analysis of cardiac extracts.

Figure 4.
Time trajectory analysis of cardiac extracts.

(A) Time trajectory plot based on the mean 1H NMR spectra of cardiac tissues collected from the Pdk1F/F; αMHC-Cre (red dots) and age-matched controls (black squares) at various time points. A loading plot revealed the metabolites with major contributors to the separation at PC1 (B) and PC2 (C).

Figure 4.
Time trajectory analysis of cardiac extracts.

(A) Time trajectory plot based on the mean 1H NMR spectra of cardiac tissues collected from the Pdk1F/F; αMHC-Cre (red dots) and age-matched controls (black squares) at various time points. A loading plot revealed the metabolites with major contributors to the separation at PC1 (B) and PC2 (C).

Metabolic changes in cardiac tissue of Pdk1F/F; αMHC-Cre mice

The normalized integral values of metabolites are shown in Table 1. Figure 5 demonstrates the holistically metabolic pathways of cardiac tissue between Pdk1F/F; αMHC-Cre and Pdk1F/F at four time points. We found that the levels of acetate, glutamate, glutamine, and O-phosphocholine were significantly increased upon Pdk1 deletion. Besides, Pdk1F/F; αMHC-Cre mice had significantly decreased levels of lactate, alanine, glycine, taurine, choline, fumarate, IMP, AMP, and ATP compared with control mice. Levels of formate and inosine exhibited a relatively slight change in Pdk1 deletion mice. Together, Pdk1 deficiency resulted in the reduced TCA cycle, decreased glycolysis, and other metabolic disorders. We next conducted RT-PCR analysis on key enzymes of TCA cycle and observed the decreased expression of HK2, FH, IDH1, and MDH, confirming the aforementioned effects of PDK1 on heart tissues (Figure 6).

Disturbed metabolic pathways in cardiac at four time points.

Figure 5.
Disturbed metabolic pathways in cardiac at four time points.

The pathways referenced to the KEGG database and HMDB reveal the interrelationship of the identified metabolic pathways involved in Pdk1F/F; αMHC-Cre mice. Metabolites with colors represent significant changes in levels compared with control.

Figure 5.
Disturbed metabolic pathways in cardiac at four time points.

The pathways referenced to the KEGG database and HMDB reveal the interrelationship of the identified metabolic pathways involved in Pdk1F/F; αMHC-Cre mice. Metabolites with colors represent significant changes in levels compared with control.

Analysis of TCA cycle enzymes of cardiac samples.

Figure 6.
Analysis of TCA cycle enzymes of cardiac samples.

Analysis of TCA cycle key gene expression in Pdk1 wild-type or knockout mice by qRT-PCR.

Figure 6.
Analysis of TCA cycle enzymes of cardiac samples.

Analysis of TCA cycle key gene expression in Pdk1 wild-type or knockout mice by qRT-PCR.

Table 1
Summary of change trends of metabolites compared with age-matched controls in relative integral levels
No. Age 4 weeks 5 weeks 6 weeks 7 weeks 
Metabolites Pdk1 KO (n = 12) versus WT (n = 9) Pdk1 KO (n = 8) versus WT (n = 7) Pdk1 KO (n = 11) versus WT (n = 10) Pdk1 KO (n = 11) versus WT (n = 7) 
P-value Rate of variation P-value Rate of variation P-value Rate of variation P-value Rate of variation 
Pantothenate 0.022 0.184 0.010 0.193 0.099 −0.079 0.808 −0.014 
Leucine 0.557 0.041 0.518 0.045 0.322 −0.059 0.342 0.096 
Isoleucine 0.306 0.078 0.889 0.013 0.503 −0.045 0.776 0.025 
Valine 0.411 0.060 0.776 0.023 0.663 −0.021 0.451 0.068 
Thymol 0.360 0.073 0.707 0.051 0.289 −0.092 0.598 −0.066 
3-Hydroxybutyrate 0.148 −0.170 0.387 −0.129 0.031 −0.307 0.646 −0.050 
5,6-Dihydrothymine 0.695 −0.024 0.943 −0.010 0.022 −0.190 0.178 −0.102 
Lactate 0.000 −0.273 0.000 −0.312 0.003 −0.150 0.001 −0.195 
Alanine 0.735 −0.025 0.078 0.080 0.858 0.009 0.528 0.054 
10 Acetate 0.011 0.158 0.012 0.195 0.162 0.088 0.187 0.143 
11 Glutamate 0.528 −0.026 0.012 0.159 0.038 0.136 0.200 0.077 
12 Glutamine 0.184 0.044 0.056 0.041 0.008 0.116 0.100 0.106 
13 Succinate 0.136 −0.350 0.129 −0.319 0.772 0.106 0.165 −0.315 
14 Glutathione 0.047 −0.122 0.605 0.033 0.864 0.011 0.206 0.083 
15 Asparagine 0.091 0.379 0.011 0.535 0.096 0.331 0.042 1.477 
16 Creatine 0.952 0.001 0.016 0.069 0.008 −0.088 0.000 −0.235 
17 Choline 0.096 −0.285 0.037 −0.389 0.235 −0.160 0.499 −0.094 
18 PC 0.004 0.439 0.012 0.304 0.006 0.283 0.027 0.421 
19 Taurine 0.000 −0.070 0.003 −0.046 0.000 −0.128 0.000 −0.136 
20 Glycine 0.368 −0.064 0.479 −0.077 0.849 0.113 0.026 −0.151 
21 Inosine 0.012 0.304 0.685 0.068 0.093 0.304 0.026 0.527 
22 AMP 0.047 −0.448 0.767 −0.085 0.002 −0.605 0.009 −0.499 
23 Nicotinurate 0.110 −0.046 0.037 −0.058 0.000 −0.204 0.000 −0.211 
24 Fumarate 0.000 −0.702 0.000 −0.681 0.000 −0.656 0.000 −0.632 
25 Formate 0.979 −0.003 0.212 0.161 0.431 0.070 0.497 0.125 
26 IMP 0.078 −0.148 0.114 −0.150 0.000 −0.342 0.000 −0.485 
27 ATP 0.011 −0.340 0.447 −0.210 0.061 −0.227 0.083 −0.334 
28 ADP 0.004 −0.272 0.028 −0.237 0.000 −0.345 0.001 −0.392 
No. Age 4 weeks 5 weeks 6 weeks 7 weeks 
Metabolites Pdk1 KO (n = 12) versus WT (n = 9) Pdk1 KO (n = 8) versus WT (n = 7) Pdk1 KO (n = 11) versus WT (n = 10) Pdk1 KO (n = 11) versus WT (n = 7) 
P-value Rate of variation P-value Rate of variation P-value Rate of variation P-value Rate of variation 
Pantothenate 0.022 0.184 0.010 0.193 0.099 −0.079 0.808 −0.014 
Leucine 0.557 0.041 0.518 0.045 0.322 −0.059 0.342 0.096 
Isoleucine 0.306 0.078 0.889 0.013 0.503 −0.045 0.776 0.025 
Valine 0.411 0.060 0.776 0.023 0.663 −0.021 0.451 0.068 
Thymol 0.360 0.073 0.707 0.051 0.289 −0.092 0.598 −0.066 
3-Hydroxybutyrate 0.148 −0.170 0.387 −0.129 0.031 −0.307 0.646 −0.050 
5,6-Dihydrothymine 0.695 −0.024 0.943 −0.010 0.022 −0.190 0.178 −0.102 
Lactate 0.000 −0.273 0.000 −0.312 0.003 −0.150 0.001 −0.195 
Alanine 0.735 −0.025 0.078 0.080 0.858 0.009 0.528 0.054 
10 Acetate 0.011 0.158 0.012 0.195 0.162 0.088 0.187 0.143 
11 Glutamate 0.528 −0.026 0.012 0.159 0.038 0.136 0.200 0.077 
12 Glutamine 0.184 0.044 0.056 0.041 0.008 0.116 0.100 0.106 
13 Succinate 0.136 −0.350 0.129 −0.319 0.772 0.106 0.165 −0.315 
14 Glutathione 0.047 −0.122 0.605 0.033 0.864 0.011 0.206 0.083 
15 Asparagine 0.091 0.379 0.011 0.535 0.096 0.331 0.042 1.477 
16 Creatine 0.952 0.001 0.016 0.069 0.008 −0.088 0.000 −0.235 
17 Choline 0.096 −0.285 0.037 −0.389 0.235 −0.160 0.499 −0.094 
18 PC 0.004 0.439 0.012 0.304 0.006 0.283 0.027 0.421 
19 Taurine 0.000 −0.070 0.003 −0.046 0.000 −0.128 0.000 −0.136 
20 Glycine 0.368 −0.064 0.479 −0.077 0.849 0.113 0.026 −0.151 
21 Inosine 0.012 0.304 0.685 0.068 0.093 0.304 0.026 0.527 
22 AMP 0.047 −0.448 0.767 −0.085 0.002 −0.605 0.009 −0.499 
23 Nicotinurate 0.110 −0.046 0.037 −0.058 0.000 −0.204 0.000 −0.211 
24 Fumarate 0.000 −0.702 0.000 −0.681 0.000 −0.656 0.000 −0.632 
25 Formate 0.979 −0.003 0.212 0.161 0.431 0.070 0.497 0.125 
26 IMP 0.078 −0.148 0.114 −0.150 0.000 −0.342 0.000 −0.485 
27 ATP 0.011 −0.340 0.447 −0.210 0.061 −0.227 0.083 −0.334 
28 ADP 0.004 −0.272 0.028 −0.237 0.000 −0.345 0.001 −0.392 

Knockdown of PDK1 reduces mitochondrial spare respiratory capacity

Figure 5 shows lower levels of ATP, ADP, and AMP of cardiac tissues in Pdk1F/F; αMHC-Cre mice. Moreover, Pdk1 deficiency mice revealed decreased levels of overall energy charge [19] relative to control (Figure 7). To explore the causality of mitochondrial respiration, siRNA was used to silence PDK1 expression followed by the Seahorse extracellular flux analyzer. The efficiency of RNA interference was detected by real-time PCR and western blot (Figure 8A). The silence of PDK1 displayed a decreased OCR, an indicator of mitochondrial oxidative respiration (Figure 8B). Reduced ATP compounds and decreased energy charge in Pdk1 deficiency tissues were consistent with a reduced OCR and impaired spare respiratory capacity in vitro. These data revealed that Pdk1 deficiency had significantly diminished energy status.

Overall energy charge assay of cardiac samples.

Figure 7.
Overall energy charge assay of cardiac samples.

Quantitation of energy charge in Pdk1F/F; αMHC-Cre heart tissues compared with age-matched controls at 4 weeks (Pdk1F/F; αMHC-Cre: n = 13, Pdk1F/F: n = 9) (P = 0.014), 5 weeks (Pdk1F/F; αMHC-Cre: n = 8, Pdk1F/F: n = 7) (P = 0.439), 6 weeks (Pdk1F/F; αMHC-Cre: n = 11, Pdk1F/F: n = 11) (P =0.064), and 7 weeks (Pdk1F/F; αMHC-Cre: n = 9, Pdk1F/F: n = 14) (P =0.06). Energy charge = ([ATP] + 0.5[ADP])/([ATP] + [ADP] + [AMP]).

Figure 7.
Overall energy charge assay of cardiac samples.

Quantitation of energy charge in Pdk1F/F; αMHC-Cre heart tissues compared with age-matched controls at 4 weeks (Pdk1F/F; αMHC-Cre: n = 13, Pdk1F/F: n = 9) (P = 0.014), 5 weeks (Pdk1F/F; αMHC-Cre: n = 8, Pdk1F/F: n = 7) (P = 0.439), 6 weeks (Pdk1F/F; αMHC-Cre: n = 11, Pdk1F/F: n = 11) (P =0.064), and 7 weeks (Pdk1F/F; αMHC-Cre: n = 9, Pdk1F/F: n = 14) (P =0.06). Energy charge = ([ATP] + 0.5[ADP])/([ATP] + [ADP] + [AMP]).

Knockdown of PDK1 reduces the OCR.

Figure 8.
Knockdown of PDK1 reduces the OCR.

(A) Expression of PDK1 was performed by real-time PCR and western blot in cells transfected with siRNA. (B) Effects of PDK1 deficiency on the OCR were examined. OCR, oxygen consumption rate. Significant levels: *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 8.
Knockdown of PDK1 reduces the OCR.

(A) Expression of PDK1 was performed by real-time PCR and western blot in cells transfected with siRNA. (B) Effects of PDK1 deficiency on the OCR were examined. OCR, oxygen consumption rate. Significant levels: *P < 0.05, **P < 0.01, ***P < 0.001.

Discussion

Many researches have shown that deletion of Pdk1 causes severe and lethal HF [1113,20]. However, the underlying mechanisms of abnormalities are not clear. Recent researches have suggested that metabolic reprogramming participate in the onset and progression of HF, including alterations in fatty acid, glucose, ketone body, and amino acid metabolism [5,21,22]. The aim of this study was to explore the cardiac metabolome in Pdk1 deletion mice by 1H NMR-based metabolomics.

In the present study, metabolism profiles were detected at four time points between Pdk1 deletion mice and matched controls. Statistical analysis of 1H NMR spectral data showed the abnormal cardiac metabolome in Pdk1F/F; αMHC-Cre mice. Moreover, the disorders were observed as early as 4 weeks of age, and dysfunction of metabolism precedes a morphological change in Pdk1 deletion mice. The data suggested that impaired metabolism is an early event in HF pathogenesis. Besides, our study indicated that Pdk1 deficiency resulted in the reduced TCA cycle, decreased glycolysis, and other metabolic disorders.

Glucose metabolism is critical for sustaining cardiac function. In the present study, significantly reduced concentration of lactate, succinate, and fumarate was found in Pdk1 deletion mice. The decreased levels of lactate and TCA intermediates indicated that anaerobic glycolysis and TCA cycle were down-regulated compared with controls. Diakos et al. [23] reported that lactate, succinate, and fumarate levels were reduced in patients with severe HF. The findings suggest that the perturbation of glucose metabolism may contribute to the pathology of the heart in Pdk1 deletion mice.

Amino acids play important roles in energy production, protein synthesis, and signaling intermediates. Glycine has been proved to exhibit a precursor for creatine synthesis and against reoxygenation damage in mitochondria of cardiomyocytes [24]. Studies suggested that taurine associated with calcium homeostasis, stabilization of membrane and oxidation resistance, and played a critical role in cardiovascular function [25]. Researches showed that HF patients displayed impaired amino acid metabolism [26]. It should be noted that levels of alanine, glycine, and taurine decreased occurred in Pdk1 knockout mice relative to controls. Besides, glutamate and glutamine served as anaplerotic substrates for TCA cycle, which abnormalities may be contributory to the impaired energy synthesis and influences cellular activities in Pdk1F/F; αMHC-Cre mice. The accumulation of glutamate and glutamine was consistent with the notion that amino acid catabolism was down-regulated in failing heart [27]. The abnormality of amino acid metabolism in Pdk1F/F; αMHC-Cre mice may be a potential pathogenic mechanism underlying HF.

Beside the imbalance in fuel availability, levels of choline pathway metabolites were also found disordered in Pdk1 deficiency mice. Choline is the principal component of cell membranes, and recent studies indicated that plasma choline levels could be a prognosis marker in HF patients [28]. We speculate that the alteration in choline metabolism could be responsible for the morphological changes of heart tissue in Pdk1F/F; αMHC-Cre mice.

Furthermore, we also found that the energy biomolecules including IMP, AMP, ATP, and ADP levels and energy charge were decreased in the Pdk1F/F; αMHC-Cre cardiac tissues. Doenst et al. [29] indicated that the ATP level was reduced in failing human hearts compared with controls. Pointon et al. [30] proposed that inadequate ATP generation was responsible for HF. Since Pdk1 deficiency with disordered energy metabolic and decreased ATP compounds, we postulated that mitochondrial function was impaired. In accordance with this, a reduced oxygen consumption rate was detected by the Seahorse Analyzer in PDK1 knockdown cells. The data consistent with previous study reported impaired respiratory function in HF [17]. However, we have not determined the expression of key genes involved in metabolism and the activity of respiratory chain complexes.

In the present study, we have demonstrated first the holistically metabolism patterns of cardiac in Pdk1 deletion mice. The data suggest that abnormalities in Pdk1 may be implicated in various metabolic disorders and a reduced oxygen consumption rate. Here, we propose a hypothesis, deletion of Pdk1 in cardiomyocyte leading to metabolism reprogramming, which can contribute to irregular heart function and HF. However, further studies are required to elucidate the underling mechanisms of metabolism remodeling in Pdk1 deficiency mice of the HF.

Abbreviations

     
  • Cre

    creatine

  •  
  • HF

    heart failure

  •  
  • Ino

    inosine

  •  
  • NMR

    nuclear magnetic resonance

  •  
  • OCR

    oxygen consumption rate

  •  
  • PC

    O-phosphocholine

  •  
  • PCA

    principal component analysis

  •  
  • PDK1

    phosphoinositide-dependent protein kinase-1

  •  
  • SD

    standard deviation

  •  
  • siRNA

    small interfering RNA

  •  
  • Tau

    taurine

  •  
  • TSP

    trimethylsilyl propionate

Author Contribution

C.L. and Y.N. contributed to experimental studies, data analysis, and manuscript writing. H.Z., L.Z., and C.Y. contributed to the data analysis. C.S., Q.C., and Z.Y. played a major role in organizing animal experiments. H.G. contributed to financial support and gave final approval for the publication of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China [nos. 81771386, 21575105, and 81770830], the Zhejiang Provincial Natural Science Foundation [nos. LQ18H160027 and LY17H160049].

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Benjamin
,
E.J.
,
Blaha
,
M.J.
,
Chiuve
,
S.E.
,
Cushman
,
M.
,
Das
,
S.R.
,
Deo
,
R.
et al.  (
2017
)
Heart disease and stroke statistics-2017 update: a report from the American heart association
.
Circulation
135
,
e146
e603
2
Harjola
,
V.P.
,
Mebazaa
,
A.
,
Čelutkienė
,
J.
,
Bettex
,
D.
,
Bueno
,
H.
,
Chioncel
,
O.
et al.  (
2016
)
Contemporary management of acute right ventricular failure: a statement from the Heart Failure Association and the Working Group on Pulmonary Circulation and Right Ventricular Function of the European Society of Cardiology
.
Eur. J Heart Fail.
18
,
226
241
3
Tanai
,
E.
and
Frantz
,
S.
(
2015
)
Pathophysiology of heart failure
.
Compr. Physiol.
6
,
187
214
4
van Bilsen
,
M.
(
2004
)
‘Energenetics’ of heart failure
.
Ann. N. Y. Acad. Sci.
1015
,
238
249
5
Wende
,
A.R.
,
Brahma
,
M.K.
,
McGinnis
,
G.R.
and
Young
,
M.E.
(
2017
)
Metabolic origins of heart failure
.
JACC Basic Transl. Sci.
2
,
297
310
6
Ho
,
J.E.
,
Lyass
,
A.
,
Courchesne
,
P.
,
Chen
,
G.
,
Liu
,
C.
,
Yin
,
X.
et al.  (
2018
)
Protein biomarkers of cardiovascular disease and mortality in the community
.
J. Am. Heart Assoc.
7
,
e008108
7
Melenovsky
,
V.
,
Benes
,
J.
,
Franekova
,
J.
,
Kovar
,
J.
,
Borlaug
,
B.A.
,
Segetova
,
M.
et al.  (
2017
)
Glucose homeostasis, pancreatic endocrine function, and outcomes in advanced heart failure
.
J. Am. Heart Assoc.
6
,
e005290
8
Ni
,
Q.Z.
,
Yang
,
F.
,
Can
,
T.V.
,
Sergeyev
,
I.V.
,
D'Addio
,
S.M.
,
Jawla
,
S.K.
et al.  (
2017
)
In situ characterization of pharmaceutical formulations by dynamic nuclear polarization enhanced MAS NMR
.
J. Phys. Chem. B
121
,
8132
8141
9
Hunter
,
W.G.
,
Kelly
,
J.P.
,
McGarrah
, III,
R.W.
,
Kraus
,
W.E.
and
Shah
,
S.H.
(
2016
)
Metabolic dysfunction in heart failure: diagnostic, prognostic, and pathophysiologic insights from metabolomic profiling
.
Curr. Heart Fail. Rep.
13
,
119
131
10
Heather
,
L.C.
,
Wang
,
X.
,
West
,
J.A.
and
Griffin
,
J.L.
(
2013
)
A practical guide to metabolomic profiling as a discovery tool for human heart disease
.
J. Mol. Cell. Cardiol.
55
,
2
11
11
Feng
,
Q.
,
Di
,
R.
,
Tao
,
F.
,
Chang
,
Z.
,
Lu
,
S.
,
Fan
,
W.
et al.  (
2010
)
PDK1 regulates vascular remodeling and promotes epithelial-mesenchymal transition in cardiac development
.
Mol. Cell. Biol.
30
,
3711
3721
12
Di
,
R.M.
,
Feng
,
Q.T.
,
Chang
,
Z.
,
Luan
,
Q.
,
Zhang
,
Y.Y.
,
Huang
,
J.
et al.  (
2010
)
PDK1 plays a critical role in regulating cardiac function in mice and human
.
Chin. Med. J.
123
,
2358
2363
PMID:
[PubMed]
13
Mora
,
A.
,
Davies
,
A.M.
,
Bertrand
,
L.
,
Sharif
,
I.
,
Budas
,
G.R.
,
Jovanovic
,
S.
et al.  (
2003
)
Deficiency of PDK1 in cardiac muscle results in heart failure and increased sensitivity to hypoxia
.
EMBO J.
22
,
4666
4676
14
Budas
,
G.R.
,
Sukhodub
,
A.
,
Alessi
,
D.R.
and
Jovanović
,
A.
(
2006
)
3′Phosphoinositide-dependent kinase-1 is essential for ischemic preconditioning of the myocardium
.
FASEB J.
20
,
2556
2558
15
Liu
,
H.
,
Song
,
G.
,
Zhou
,
L.
,
Hu
,
X.
,
Liu
,
M.
,
Nie
,
J.
et al.  (
2013
)
Compared analysis of LncRNA expression profiling in pdk1 gene knockout mice at two time points
.
Cell. Physiol. Biochem.
32
,
1497
1508
16
Gao
,
H.C.
,
Zhu
,
H.
,
Song
,
C.Y.
,
Lin
,
L.
,
Xiang
,
Y.
,
Yan
,
Z.H.
et al.  (
2013
)
Metabolic changes detected by ex vivo high resolution 1H NMR spectroscopy in the striatum of 6-OHDA-induced Parkinson's rat
.
Mol. Neurobiol.
47
,
123
130
17
Hoes
,
M.F.
,
Grote Beverborg
,
N.
,
Kijlstra
,
J.D.
,
Kuipers
,
J.
,
Swinkels
,
D.W.
,
Giepmans
,
B.N.G.
et al.  (
2018
)
Iron deficiency impairs contractility of human cardiomyocytes through decreased mitochondrial function
.
Eur. J. Heart Fail.
20
,
910
919
18
Nguyen
,
H.M.
,
Mejia
,
E.M.
,
Chang
,
W.
,
Wang
,
Y.
,
Watson
,
E.
,
On
,
N.
et al.  (
2016
)
Reduction in cardiolipin decreases mitochondrial spare respiratory capacity and increases glucose transport into and across human brain cerebral microvascular endothelial cells
.
J. Neurochem.
139
,
68
80
19
Peoples
,
J.N.R.
,
Maxmillian
,
T.
,
Le
,
Q.
,
Nadtochiy
,
S.M.
,
Brookes
,
P.S.
,
Porter
, Jr,
G.A.
et al.  (
2018
)
Metabolomics reveals critical adrenergic regulatory checkpoints in glycolysis and pentose-phosphate pathways in embryonic heart
.
J. Biol. Chem.
293
,
6925
6941
20
Ito
,
K.
,
Akazawa
,
H.
,
Tamagawa
,
M.
,
Furukawa
,
K.
,
Ogawa
,
W.
,
Yasuda
,
N.
et al.  (
2009
)
PDK1 coordinates survival pathways and β-adrenergic response in the heart
.
Proc. Natl Acad. Sci. U.S.A.
106
,
8689
8694
21
Ikegami
,
R.
,
Shimizu
,
I.
,
Yoshida
,
Y.
and
Minamino
,
T.
(
2017
)
Metabolomic analysis in heart failure
.
Circ. J.
82
,
10
16
22
McGarrah
,
R.W.
,
Crown
,
S.B.
,
Zhang
,
G.F.
,
Shah
,
S.H.
and
Newgard
,
C.B.
(
2018
)
Cardiovascular metabolomics
.
Circ. Res.
122
,
1238
1258
23
Diakos
,
N.A.
,
Navankasattusas
,
S.
,
Abel
,
E.D.
,
Rutter
,
J.
,
McCreath
,
L.
,
Ferrin
,
P.
et al.  (
2016
)
Evidence of glycolysis up-regulation and pyruvate mitochondrial oxidation mismatch during mechanical unloading of the failing human heart: implications for cardiac reloading and conditioning
.
JACC Basic Transl. Sci.
1
,
432
444
24
Ruiz-Meana
,
M.
,
Pina
,
P.
,
Garcia-Dorado
,
D.
,
Rodriguez-Sinovas
,
A.
,
Barba
,
I.
,
Miró-Casas
,
E.
et al.  (
2004
)
Glycine protects cardiomyocytes against lethal reoxygenation injury by inhibiting mitochondrial permeability transition
.
J. Physiol.
558
,
873
882
25
Froger
,
N.
,
Moutsimilli
,
L.
,
Cadetti
,
L.
,
Jammoul
,
F.
,
Wang
,
Q.P.
,
Fan
,
Y.
et al.  (
2014
)
Taurine: the comeback of a neutraceutical in the prevention of retinal degenerations
.
Prog. Retin. Eye Res.
41
,
44
63
26
Weitzel
,
L.B.
,
Ambardekar
,
A.V.
,
Brieke
,
A.
,
Cleveland
,
J.C.
,
Serkova
,
N.J.
,
Wischmeyer
,
P.E.
et al.  (
2013
)
Left ventricular assist device effects on metabolic substrates in the failing heart
.
PLoS ONE
8
,
e60292
27
Lai
,
L.
,
Leone
,
T.C.
,
Keller
,
M.P.
,
Martin
,
O.J.
,
Broman
,
A.T.
,
Nigro
,
J.
et al.  (
2014
)
Energy metabolic reprogramming in the hypertrophied and early stage failing heart: a multisystems approach
.
Circ. Heart Fail.
7
,
1022
1031
28
Guasch-Ferré
,
M.
,
Hu
,
F.B.
,
Ruiz-Canela
,
M.
,
Bulló
,
M.
,
Toledo
,
E.
,
Wang
,
D.D.
et al.  (
2017
)
Plasma metabolites from choline pathway and risk of cardiovascular disease in the PREDIMED (Prevention With Mediterranean diet) study
.
J. Am. Heart Assoc.
6
,
e006524
29
Doenst
,
T.
,
Nguyen
,
T.D.
and
Abel
,
E.D.
(
2013
)
Cardiac metabolism in heart failure: implications beyond ATP production
.
Circ. Res.
113
,
709
724
30
Pointon
,
A.V.
,
Walker
,
T.M.
,
Phillips
,
K.M.
,
Luo
,
J.
,
Riley
,
J.
,
Zhang
,
S.D.
et al.  (
2010
)
Doxorubicin in vivo rapidly alters expression and translation of myocardial electron transport chain genes, leads to ATP loss and caspase 3 activation
.
PLoS ONE
5
,
e12733

Author notes

*

These authors contributed equally to this work.