Effects of alanine:glyoxylate aminotransferase variants and pyridoxine sensitivity on oxalate metabolism in a cell-based cytotoxicity assay

Abstract

The hereditary kidney stone disease primary hyperoxaluria type 1 (PH1) is caused by a functional deficiency of the liver-specific, peroxisomal, pyridoxal-phosphate-dependent enzyme, alanine:glyoxylate aminotransferase (AGT). One third of PH1 patients, particularly those expressing the p.[(Pro11Leu; Gly170Arg; Ile340Met)] mutant allele, respond clinically to pharmacological doses of pyridoxine. To gain further insight into the metabolic effects of AGT dysfunction in PH1 and the effect of pyridoxine, we established an "indirect" glycolate cytotoxicity assay using CHO cells expressing glycolate oxidase (GO) and various normal and mutant forms of AGT. In cells expressing GO the great majority of glycolate was converted to oxalate and glyoxylate, with the latter causing the greater decrease in cell survival. Co-expression of normal AGTs and some, but not all, mutant AGT variants partially counteracted this cytotoxicity and led to decreased synthesis of oxalate and glyoxylate. Increasing the extracellular pyridoxine up to 0.3μM led to an increased metabolic effectiveness of normal AGTs and the AGT-Gly170Arg variant. The increased survival seen with AGT-Gly170Arg was paralleled by a 40% decrease in oxalate and glyoxylate levels. These data support the suggestion that the effectiveness of pharmacological doses of pyridoxine results from an improved metabolic effectiveness of AGT; that is the increased rate of transamination of glyoxylate to glycine. The indirect glycolate toxicity assay used in the present study has potential to be used in cell-based drug screening protocols to identify chemotherapeutics that might enhance or decrease the activity and metabolic effectiveness of AGT and GO, respectively, and be useful in the treatment of PH1.

Keywords: Alanine:glyoxylate aminotransferase; Glyoxylate; Kidney stones; Oxalate; Primary hyperoxaluria; Pyridoxine.

Fig. 1

Compartmentalization of enzymes and the flux of metabolites in CHO cells stably transfected with GO and AGT. AGT: normal alanine:glyoxylate aminotransferase targeted to peroxisomes; AGT*: mutant AGT mistargeted to mitochondria; GO: glycolate oxidase; LDH: lactate dehydrogenase; ECF: extracellular fluid. Untransformed CHO cells express LDH but not AGT or GO. Solid arrows indicate metabolic pathways; dashed arrows indicate transport pathways.

Fig. 2

GO and AGT catalytic activities in stably transfected CHO cell lines. Units: GO nmol substrate/min/mg protein (black boxes); AGT μmol pyruvate/h/mg protein (gray histograms). The results of the AGT assay were published in part previously [29]. The reference ranges for GO and AGT activity in the human liver are 13.2–101.6, and 19.1–47.9, respectively [34,43]. Results (n = 5–9) are expressed as mean ± SEM. GO catalytic activities in different cell lines were compared to the GO activity in CHO GO AGT-MA cells (***p < 0.001).

Fig. 3

Influence of GO and AGT on cell survival after incubation with 2-carbon metabolites. CHO wt (white), CHO GO (black) and CHO GO AGT-MA (gray) cells were exposed to increasing concentrations (0–1 mM) of either glyoxylate (A), glycolate (B) or oxalate (C) for 1 day. Cell viability was normalized against untreated controls and expressed as mean% of the control (+SD; n ≥ 3). Significant differences are marked as follows: $: AGT-MA vs WT; #: AGT-MA vs GO; *: GO vs WT ($$$/###/***p < 0.001; #/*p < 0.05).

Fig. 4

Impact of AGT variants on cell survival after incubation with varying concentrations of glycolate. CHO wt cells (black △) and CHO cells stably expressing GO (orange ○) ± different AGT variants (AGT-MA: blue □, AGT-mi: green ◇, AGT-170: red ▲, AGT-152: purple ●, AGT-244: violet ■, AGT-41: yellow ◆) were incubated for 2 days with glycolate (0 to 1500 μM). Cell survival was normalized against untreated controls. Data expressed as mean ± SEM (n ≥ 2). Cell survival curves were fitted with Graphpad Prism 6.0 software.

Fig. 5

Synthesis of oxalate and glyoxylate in CHO cells expressing GO and AGT after incubation with varying concentrations of glycolate. CHO wt and CHO cells expressing GO ± AGT-Ma, AGT-mi or AGT-170 were incubated with different glycolate concentrations (a: 0, b: 100, c: 250 μM) for 24 h. The concentrations of oxalate (white bars), glycolate (black bars) and glyoxylate (gray bars) released in the media were measured by IC, IC/MS and HPLC, respectively and plotted as stacked columns. Results (n ≥ 3) are expressed as mean ± SD.

Fig. 6

Effect of pyridoxine on the survival of CHO GO AGT cells after glycolate incubation. CHO cells stably expressing GO ± different AGT variants (AGT-MA, AGT-mi, AGT-170) grown in different concentrations of pyridoxine (0: light gray, < 0.3: dark gray, 0.3: black, 50 μM: hatched) were incubated for 2 days with glycolate (0, 500, 750 μM). Cell survival was normalized against untreated controls grown in the same pyridoxine concentration and expressed as a ratio to the survival at matched (500 and 750 μM). Glycolate in 0.3 μM pyridoxine. Results are expressed as mean and range (n ≥ 4) of both glycolate concentrations. Different superscript letters denote significant differences between columns (p < 0.01). The estimated range of normal PLP levels in human plasma is 10–100 nmol/l.

Fig. 7

Reactive oxygen species (ROS) production. Intracellular ROS production was measured 4 h after incubation with glycolate (0–250 μM) in CHO wt (dashed line) and CHO GO (full line) CHO GO + AGT-MA (dotted line) after loading with DCF-DA. Results are expressed as mean ± SEM relative increases in specific fluorescence intensities (n = 2–4), statistical difference between CHO wt and CHO GO ± AGT-MA is indicated (**p < 0.01; ***p < 0.001).