The attenuated hepatic clearance of propionate increases cardiac oxidative stress in propionic acidemia

Abstract

Propionic acidemia (PA), arising from PCCA or PCCB variants, manifests as life-threatening cardiomyopathy and arrhythmias, with unclear pathophysiology. In this work, propionyl-CoA metabolism in rodent hearts and human pluripotent stem cell-derived cardiomyocytes was investigated with stable isotope tracing analysis. Surprisingly, gut microbiome-derived propionate rather than the propiogenic amino acids (valine, isoleucine, threonine, and methionine) or odd-chain fatty acids was found to be the primary cardiac propionyl-CoA source. In a Pcca^-/-(A138T) mouse model and PA patients, accumulated propionyl-CoA and diminished acyl-CoA synthetase short-chain family member 3 impede hepatic propionate disposal, elevating circulating propionate. Prolonged propionate exposure induced significant oxidative stress in PCCA knockdown HL-1 cells and the hearts of Pcca^-/-(A138T) mice. Additionally, Pcca^-/-(A138T) mice exhibited mild diastolic dysfunction after the propionate challenge. These findings suggest that elevated circulating propionate may cause oxidative damage and functional impairment in the hearts of patients with PA.

Keywords: Cardiac complication; Human induced pluripotent stem cell-derived cardiomyocytes; Microbiome; Propionate; Propionic acidemia; Stable isotope-based metabolic flux.

Fig. 1

Propionyl-CoA turnover in the perfused rat heart. A is a scheme of propionyl-CoA turnover traced by [¹³C₃]propionate. C3 AC: propionylcarnitine, C3-CoA: propionyl-CoA, MM-CoA: methylmalonyl-CoA, Suc-CoA: succinyl-CoA, Mito: mitochondria. B is the protocol of 7 heart perfusion groups with[¹³C₃]propionate (M3 Prop) and various substrates. Ile: isoleucine, Thr: threonine, Met: methionine, Val: valine, BSA: bovine serum albumin; AAs: mixed 20 amino acids. C is propionyl-CoA labeling in perfused hearts from 7-group perfusions with [¹³C₃]propionate and various substrates. D is the calculated contributions to propionyl-CoA production from various substrates in heart. E is the metabolite level detected in perfused hearts from 7 groups. N=5–6/group. * denotes significant differences between indicated group with p<0.05.

Fig. 2

The metabolism of odd-chain fatty acid and propiogenic amino acids in the perfused hearts. A-E are schemes of metabolism of deuterium labeled C17 fatty acid (D33 C17 fatty acid) and ¹³C labeled amino acids (valine, isoleucine, threonine, and methionine) leading to the formation of propionylcarnitine (C3 AC). F-J are stable isotope labeling of metabolites in rat hears perfused with either 80 μM D33 C17 fatty acid, 230 μM M5 valine ([¹³C₅]valine), 71 μM M6 isoleucine ([¹³C₆]isoleucine), 170 μM M4 threonine ([¹³C₄]threonine), or 23 μM M5 methionine ([¹³C₅]methionine). C17-CoA: heptadecanoyl-CoA; C17 AC, C13 AC, C11 AC, C9 AC, C7 AC, C5 AC, and C3 AC are acylcarnitines with 17, 13, 11, 9, 7, 5, and 3 carbons, respectively; Thre: threonine; 2HB: 2-hydroxybutyrate; Ile: isoleucine; Tigyly AC: tigylycarnitine; Met: methionine. n=3–4/group.

Fig. 3

The consumption of propionate and propiogenic amino acids by the h-iPSC-CMs. A is the experiment design. Following the differentiation of h-iPSCs into h-iPSC-CMs, 1 million cells were cultured in RPMI 1640/B27 containing 100 μM propionate (0.1-Prop), 1 mM propionate (1-Prop), 382 μM isoleucine, 101 μM methionine, 168 μM threonine, and 171 μM valine. After 4 days, media samples were collected and analyzed using mass spectrometry (MS). B is the levels of 100 μM propionate (0.1-Prop), 1 mM propionate (1-Prop), isoleucine, valine, threonine, and methionine in both pre- and post-cultured media. The levels of substrates in the post-cultured media were normalized to the levels in the pre-cultured media. C is the consumption of substrates after 4 days of culturing. The data represent N=4. **, ***, and **** denote significant differences between indicated groups with p<0.01, p<0.005, and p<0.001, respectively.

Fig. 4

Propionate and short-chain acylcarnitines in control and germ-free (GF) mice. A is a simplified circulating system, with syringe markers indicating locations for blood sample withdrawal. B is propionate levels in plasma obtained from the hepatic vein (HV), portal vein (PV), and renal vein (RV) in both control and GF mice. C is C3 AC (propionylcarnitine) levels in the portal vein for both control and GF mice. D is acylcarnitine levels in the portal vein for both control and GF mice. The data include n=5–6 per group. **, ***, and **** denote significant differences between indicated groups with p<0.01, p<0.005, and p<0.001, respectively.

Fig. 5

Propionate disposal by liver in Pcca^−/−(A138T) mice was attenuated. A-B are propionate and propionylcarnitine levels in the plasma of WT and Pcca^−/−(A138T) mice (n=5–6/group). C-D are propionate and propionylcarnitine levels in human plasma from control subjects (n=9) and PA patients (n=6). E-F are M3 and M0 propionate concentrations in plasma at 0, 10, and 20 minutes after the intraperitoneal injection of 500 mg/kg [¹³C₃]propionate in control and Pcca^−/−(A138T) mice. G is a schematic of glucose synthesis from [¹³C₃]propionate blocked by the PCC mutation, indicating MM-CoA (methylmalonyl-CoA), Suc-CoA (succinyl-CoA), OAA (oxaloacetate), and PEP (phosphoenolpyruvate). H is plasma glucose labeling at 10 minutes after the intraperitoneal injection of 500 mg/kg [¹³C₃]propionate in WT and Pcca^−/−(A138T) mice (n=5). I is ACSS3 activity in hearts from WT and Pcca^−/−(A138T) mice (n=10 in triplicate assays). *** and **** denote significant differences between indicated groups with p<0.005, and p<0.001, respectively.

Fig. 6

Propionate supplement induces mild diastolic dysfunction in Pcca^−/−(A138T) mice. Cardiac function in mice was assessed using echocardiography before and after a 2-week and 16-week propionate supplementation (250 mM) via drinking water. n=5–6 per group for both control and Pcca^−/−(A138T) mice. A-H are E, A, E/A, Decel Slope, DT, IVRT, E’, and E/E’ values recorded at pre-( 0), week 2 (2), and week 16 (16) after both control and Pcca^−/−(A138T) mice started propionate administration via drinking water. E: early diastolic mitral inflow velocity; A: late diastolic mitral inflow velocity; E/A: Ratio of E/A; Decel slope: early diastolic mitral inflow velocity slope of deceleration; DT: early diastolic mitral inflow velocity deceleration time; IVRT: isovolumic relaxation time; E’: early diastolic mitral valve velocity; E/E’: Ratio of E/E′. * and ** denote significant differences between indicated groups with p<0.05 and p<0.01, respectively.

Fig. 7

Propionate treatment increases oxidative stress to PCCA KD HL-1 cells. A is PCCA protein level in Con (control), NC (negative control), PCCAKD HL-1 (PCCA knockdown HL-1) cells. N=3. B is fluorescence image of ROS production in NC and PCCA knockdown cells exposed to varying doses of propionate (0 to 5 mM). C is the fluorescence intensity of NC and PCCA knockdown cells exposed to varying doses of propionate (0–5 mM). N=77. *, **, and **** denote significant differences between indicated groups with p<0.05, p<0.01, and p<0.001, respectively.

Fig. 8

Propionate exposure elevates oxidative stress both in vitro and in vivo. A-B show western blot quantification of the fold changes of fold changes of total HNE-modified proteins (fold changes) in negative control (NC) and PCCA knockdown HL-1 (PCCAKD HL-1) cells exposed to varying doses of propionate (Prop, 0 to 5 mM), n=3. C-D display GSH levels, GSSG levels, GSH/GSSG ratio, and methionine sulfoxide levels in the hearts of wild-type (WT) and Pcca^−/−(A138T) mice after 16 weeks of propionate administration (250 mM) in drinking water. N=5. *, **, ***, and **** denote significant differences between indicated groups with p<0.05, p<0.01, p<0.001, and p<0.0001, respectively.