Peroxisomes contribute to intracellular calcium dynamics in cardiomyocytes and non-excitable cells

(A) Quantification of D3cpV-px colocalisation with peroxisomes. Mander’s colocalisation coefficient was normalised to PMP70 and catalase colocalisation, which was set to 1. n = 5. (B) Förster resonance energy transfer ratio measured at different pH values for D3cpV (cyto) and D3cpV-px (pero). Cells were incubated in buffers with different pH values containing 10 µM nigericin. Cyto and pero show comparable results at physiological pH values of cytosol and peroxisomes. At the pH = 4 the Förster resonance energy transfer ratio decreases drastically because of the acceptor sensitivity. Cell numbers for cyto at pH 4 n = 51, 7.1 = 51, 7.35 = 51, 7.5 = 67, 7.65 = 48, 7.8 = 48; for pero at pH 4 n = 50, 7.1 = 50, 7.35 = 71, 7.5 = 71, 7.65 = 64, 7.8 = 64 from three independent experiments per condition. (C) Comparison of cytosolic and peroxisomal responses to ionomycin (Iono). In comparison to cytosol, peroxisomal signal increases gradually, n = 16 cells for D3cpV and n = 9 cells for D3cpV-px. (D) One-step experiment in HeLa cells with thapsigargin (Tg) addition in Ca²⁺-containing buffer. Cytosolic and peroxisomal Ca²⁺ increase upon Tg treatment. n = 31 (cyto), 30 (pero) from four (cyto) and five (pero) independent experiments. (E) Experimental paradigm of a two-step Ca²⁺ measurement in non-excitable cells. First peak after histamine (His) addition: ER-store depletion. Second peak, after addition of extracellular Ca²⁺: plasma membrane-based uptake. (F) Measurement of the effect of treatments from (E) on the acceptor. Neither cytosolic EYFP (cyto) nor peroxisomal Venus-ACOX3 (pero) undergo changes upon treatment, suggesting that the pH changes during the experiment will not affect the measurement with genetically encoded Ca²⁺ indicators. Data presented as mean, n = 44 (cyto), 44 (pero). (G) Measurement with D3cpV-px according to the paradigm in (E). Two Ca²⁺ peaks of the experimental paradigm are detectable with D3cpV-px, n ≥ 50 cells from three experiments. (H) In situ K_d calculation and the saturation curve of D3cpV and D3cpV-px. Curve fitting was performed with one-site model with Hill coefficient. (G, I) Absolute Ca²⁺ concentration dynamics calculated from the data in (G). (I, J) Basal and maximum (max) Ca²⁺ concentrations in peroxisomes based on (I). (K) Measurement with pericam-px according to the paradigm in (E), n = 27 cells from three experiments. (L) Simultaneous measurement of cytosolic (blue) and peroxisomal (green) Ca²⁺. No delay of signal increase after histamine addition, but a delayed drop of the signal in peroxisomes. Left y-axis: of D3cpV-px (peroxisomal sensor). Right y-axis: F_n/F₀ ratio of R-GECO1 (cytosolic sensor), n = 35 cells from three experiments. (L, M) Decline of F_n/F₀ ratio per minute (min) in the linear part of the curves in (L) (from second 65–115, t test). Kinetic delay in decrease in peroxisomal signal is seen. (A, B, C, G, H, I, K, M) Data presented as mean ± SEM. (I) Data presented as Tukey’s box plots. ***P < 0.001.

(A) Comparison of cytosolic, mitochondrial, and peroxisomal Ca²⁺ response measured following the two-step measurement described in Fig 1E. Characteristic two peaks present in all three compartments. (A, B) Basal levels of Ca²⁺ in peroxisomes are similar to mitochondria. Analysis performed based on the data from (A). (C) Peroxisomal Ca²⁺ increase upon ER-store depletion is smaller than that of cytosol or mitochondria. Analysis performed based on the data from (A). (A, D) Increase of Förster resonance energy transfer ratio per minute (min) in the linear part of the curves in (A) (from second 27–42). Slower increase in peroxisomal signal is seen. Data presented as mean ± SEM. (E) Peroxisomal Ca²⁺ increase upon plasma membrane-based cellular uptake of Ca²⁺ is comparable to mitochondria. Analysis performed based on the data from (A). (B, C, D, E) One-way ANOVA followed by Tukey’s post hoc test was used for the statistical analysis. ****P < 0.0001, Cyto: cytosolic, mito: mitochondrial, pero: peroxisomal. n = 83 (cyto), 116 (mito), 117 (pero) cells from six independent experiments. (B, C, E) Data presented as Tukey’s box plots.

(A) mRNA expression of MCU-silenced (siMCU) HeLa cells in comparison to control (siCtrl), quantified by RT-qPCR. (B) Mitochondrial Ca²⁺ response in siCtrl and siMCU measured after the histamine addition with Ca²⁺-containing buffer. MCU knockdown results in decreased Ca²⁺ uptake to mitochondria. (C) Mitochondrial Ca²⁺ increase is smaller in siMCU. (B) Analysis performed based on the data from (B). (D) Peroxisomal Ca²⁺ response in siCtrl and siMCU measured following the histamine addition with Ca²⁺-containing buffer. (E) MCU knockdown does not affect maximum Ca²⁺ uptake to peroxisomes. (D) Quantification of data from (D). Cell numbers n = 73 (siCtrl), 61 (siMCU). (C, E) Data presented as Tukey’s box plots. ****P < 0.0001.

(A) Experimental paradigm of Ca²⁺ measurement in excitable cells. The peak after thapsigargin (Tg) addition represents Ca²⁺ increase due to the sarcoplasmic/endoplasmic reticulum calcium ATPase inhibition and Ca²⁺ retention in the cytosol. (A, B) Cytosolic Ca²⁺ measurement in NRCMs following the experimental design in (A), n = 25 (Tg), 22 (control) from three experiments. Addition of Tg is compared with the addition of Tg-free buffer (control). (B, C) Basal levels are not different before the treatment in (B). (B, D) After Tg addition in (B) cytosolic Ca²⁺ increases. (A, E) Peroxisomal Ca²⁺ measurement in NRCMs following the experimental design in (A). Addition of Tg is compared with the addition of Tg-free buffer (control), n = 20 (Tg), 31 (control) from three experiments. (E, F) Basal levels of Ca²⁺ are not different before the treatment in (E). (E, G) Peroxisomal Ca²⁺ increases after Tg addition in (E). (H) Human-induced pluripotent stem cell (HiPSC)-CMs generation. Donor skin fibroblasts were reprogrammed to hiPSCs, which were then differentiated to CMs. (I) hiPSC-CMs were stained for cardiac troponin T (cTnT) and analysed by flow cytometry. Negative control without primary antibody. 94.8% of iPSC-CMs are cTnT-positive (cTNT⁺). (J) Immunofluorescence staining visualized α-actinin protein expression and regular sarcomeric organisation. Scale bar: 20 μm. (A, K) Cytosolic Ca²⁺ measurement in hiPSC-CMs with D3cpV following the experimental paradigm for excitable cells in (A). Addition of Tg is compared to the addition of Tg-free buffer (control) to avoid artefacts and false results of the mechanical effect on the cells because of the addition itself. n = 24 (Tg), 27 (control) from three experiments. (K, L) No difference is found between two groups before the treatment in (K). (K, M) Tg addition in (K) results in cytosolic Ca²⁺ increase. (A, N) Peroxisomal Ca²⁺ measurement in hiPSC-CMs with D3cpV-px following the experimental design for excitable cells depicted in (A). Addition of Tg is compared with the addition of Tg-free buffer (control). n = 26 (Tg), 33 (control) from three experiments. (N, O) Basal levels of Ca²⁺ are not different before the treatment in (N). (M, P) Peroxisomal Ca²⁺ increases after Tg addition in (M). (B, E, K, N) Data presented as means from three independent experiments. (C, D, F, G, L, M, O, P) Unpaired t test was used for the statistical analysis. ****P < 0.0001, Tukey’s box plots.

Human-induced pluripotent stem cell–CMs were either transfected with D3cpV-px (left panels) or stained with anti-Pex14 (right panels) as a peroxisomal marker. (A) Representative images from staining of L-type Ca²⁺ channel (LTCC) show occasional proximity of peroxisomes and LTCC. (B) Representative images from staining of ryanodine receptor (RyR2) show occasional yet more often contact of peroxisomes with the RyR2 than with LTCC. DAPI is shown in blue. Scale bar 5 μm.

(A) D3cpV transfected NRCMs are stimulated with 1 Hz. Images are taken every 50 ms. Oscillations of Förster resonance energy transfer ratio are seen, n = 3. (A, B) FFT from the data in (A). Signal increases are rhythmic and correspond to the pacing frequency. (C) Förster resonance energy transfer ratio oscillates in D3cpV-px transfected NRCMs stimulated with 1 Hz. Images are taken every 100 ms, n = 3. (C, D) FFT from the data in (C). Signal increases are rhythmic and correspond to the pacing frequency.