Maleic Acid--but Not Structurally Related Methylmalonic Acid--Interrupts Energy Metabolism by Impaired Calcium Homeostasis

Abstract

Maleic acid (MA) has been shown to induce Fanconi syndrome via disturbance of renal energy homeostasis, though the underlying pathomechanism is still under debate. Our study aimed to examine the pathomechanism underlying maleic acid-induced nephrotoxicity. Methylmalonic acid (MMA) is structurally similar to MA and accumulates in patients affected with methymalonic aciduria, a defect in the degradation of branched-chain amino acids, odd-chain fatty acids and cholesterol, which is associated with the development of tubulointerstitial nephritis resulting in chronic renal failure. We therefore used MMA application as a control experiment in our study and stressed hPTECs with MA and MMA to further validate the specificity of our findings. MMA did not show any toxic effects on proximal tubule cells, whereas maleic acid induced concentration-dependent and time-dependent cell death shown by increased lactate dehydrogenase release as well as ethidium homodimer and calcein acetoxymethyl ester staining. The toxic effect of MA was blocked by administration of single amino acids, in particular L-alanine and L-glutamate. MA application further resulted in severe impairment of cellular energy homeostasis on the level of glycolysis, respiratory chain, and citric acid cycle resulting in ATP depletion. As underlying mechanism we could identify disturbance of calcium homeostasis. MA toxicity was critically dependent on calcium levels in culture medium and blocked by the extra- and intracellular calcium chelators EGTA and BAPTA-AM respectively. Moreover, MA-induced cell death was associated with activation of calcium-dependent calpain proteases. In summary, our study shows a comprehensive pathomechanistic concept for MA-induced dysfunction and damage of human proximal tubule cells.

Fig 1. Induction and prevention of hPTEC damage by MA.

hPTECs were stressed with increasing amounts of MA (0, 1, 4, 8, 21 mM) for up to 24h. MA led to a concentration- and time- dependent LDH release. Data are presented as percent of untreated control of n = 20 independent experiments.

Fig 2. Induction and prevention of hPTEC damage by MMA.

hPTECs were stressed with increasing amounts of MMA (0, 1, 4, 8, 21 mM) for up to 24h. In contrast to MA, MMA only influenced cell vitality in higher concentrations. Data are presented as percent of untreated control of n = 4 independent experiments.

Fig 3. CAM and EHD stainings before and after MA treatment.

CAM and EHD stainings revealed that MA treatment decreased hPTEC vitality already after 5h. Data are presented as percent of untreated control of n = 4 independent experiments.

Fig 4. Activation of apoptosis pathways after MA treatment in hPTECs.

Increasing concentrations of MA led to more cell death and the activation of apoptosis pathways could be shown with annexin V–PI staining. Data are presented as percent of untreated control of n = 4 independent experiments.

Fig 5. Effects of diverse metabolites on MA cytotoxicity.

To prevent MA-toxicity hPTEC were cultivated for 24h with 21 mM MA +/- diverse amino acids (each 5 mM) and substrates for organic anion transporters (each 2mM). L-Alanine (Ala) and L-glutamate (Glu) prevented MA induced LDH release. There was however no significant difference between treatment with L-Alanine (Ala) and L-glutamate (Glu) (5). L-Glycine (Gly), D- and β-alanine diminished MA toxicity, whereas L-serine and L-proline (Pro), taurine (Tau), L-arginine (Arg), L-lysine (Lys) and L-phenylalanine (Phe) had no effect. Data are presented as percent of untreated control of n = 4 independent experiments.

Fig 6. Effects of organic anion transporter inhibitors on MA cytotoxicity.

The organic anion transporter inhibitor probenecid (P) blocked MA-induced LDH release, whereas succinic acid, p-aminohippuric acid (PAH) and taurocholic acid were ineffective. Data are presented as percent of untreated control of n = 4 independent experiments.

Fig 7. ATP levels in hPTECs before and after MA treatment.

hPTECs showed a decreased ATP content 6h after treatment that was not affected by rescuing amino acids. In line with cell vitality experiments, MMA did not change cellular ATP levels. Data are presented as percent of untreated control of n = 4 independent experiments.

Fig 8. Effects of MA treatment on the enzymes of glycolysis.

We analyzed activities of enzymes of glycolysis in hPTECs that were incubated for 24h with 21 mM MA. In treated cells, Hexokinase (HK) activity was increased, while phosphofructokinase (PFK), glyceraldehyde-3-phosphate dehydrogenase activity (GAPDH) and phosphoglycerate mutase (PGM) were reduced. Data are presented as percent of untreated control expressed of n = 3 independent experiments.

Fig 9. Effects of MA treatment on the citric acid cycle element activities.

We analyzed activities of enzymes of citric acid cycle in hPTECs that were incubated for 24h with 21 mM MA. Activities of triosephosphate isomerase (TPI), enolase (ENO), high and low affinity pyruvate kinase (PK HA/LA), and lactate dehydrogenase (LDH) were unchanged. Moreover, enzymatic activity of the citric acid cycle protein 2-oxoglutarate dehydrogenase complex was diminished by MA treatment, whereas citrate synthase andisocitrate dehydrogenase (IDH) activities were increased. Activities of fumarase (FUM) and malate dehydrogenase (MDH) were not affected significantly. Data are presented as percent of untreated control expressed of n = 3 independent experiments.

Fig 10. Effects of MA treatment on the respiratory chain complex activities.

We analyzed activities of enzymes of respiratory chain in hPTECs that were incubated for 24h with 21 mM MA. Activities of respiratory chain complexes I and II were reduced by MA treatment, whereas ATP synthase was mildly and complex III and IV remained unaffected. Data are presented as percent of untreated control expressed of n = 3 independent experiments.

Fig 11. Effect of MA treatment on the mitochondrial oxygen consumption.

As a measure of overall respiratory chain activity, we assessed mitochondrial oxygen consumption. hPTECs treated for 6h with MA did not reveal any NaCN-sensitive mitochondrial respiration. This figure depicts exemplary results from one experiment.

Fig 12. Effect of MA treatment on pyruvate and succinate oxidation.

To estimate overall mitochondrial function we investigate pyruvate and succinate oxidation. CO₂ production rates on both substrates were reduced after MA treatment (12). Data are presented as percent of untreated control expressed of n = 3 independent experiments.

Fig 13. Effect of calcium concentration on MA-induced LDH release.

Decreasing calcium concentrations in treatment buffer (0, 0.35, 0.7, 1.4 mM) reduced MA induced LDH release except for the highest applied MA concentration. Data are presented as percent of untreated control of n = 5 independent experiments.

Fig 14. BAPTA-AM reduces MA-induced LDH release.

The selective chelator of intracellular Ca²⁺ stores BAPTA-AM mimicked the previous approach (Fig 13). Data are presented as percent of untreated control of n = 5 independent experiments.

Fig 15. a-d. MA leads to vacuole formation in hPTECs.

Staining of actin revealed intensive vacuole formation in MA-loaded (21 mM) hPTECs that could be prevented by co-incubation with rescuing amino acids. Exemplarily results for amino acid L-glutamate are shown.

Fig 16. MA toxicity activates calpain pathways.

MA treatment activated calcium-dependent calpain proteases as indicated by the reduction of MA-induced LDH release by the inhibitor PD 150606 (50 µM). Data are presented as percent of MA treated control cells of n = 5 independent experiments.

Fig 17. Effect of chloride channel blocker NPPB on MA mediated LDH release.

The chloride channel blocker NPPB (10 µM) decreased MA induced LDH release to the corresponding control level. Data are presented as percent of untreated control of n = 6 independent experiments.

Fig 18. Effect of chloride ions on MA mediated LDH release.

Incubation of hPTEC in chloride-free KRB resulted in high LDH release rates. Strikingly, this effect was reduced by the addition of MA. Data are presented as percent of untreated control of n = 6 independent experiments.