Primary hyperoxaluria and kidney stones

Although ascorbate (vitamin C) can apparently give rise to oxalate, the capacity for ascorbate metabolism is very limited, and at levels of intake above about 100 mg/day it is excreted unchanged in the urine. There is some evidence that the formation of oxalate from ascorbate is an artefact, and occurs during storage of urine rather than in the body.

These studies suggest most strongly that the problem in hyperoxaluria is in the metabolism of glycine, with excess formation of oxalate leading to the increased urinary excretion, and hence the problem of crystallisation and formation of renal stones.

Glyoxylate is formed from glycine either by the action of glycine oxidase (forming ammonium and hydrogen peroxide) or by a variety of transaminases, the most important of which is alanine-glyoxalate transaminase. In common with most other enzymes that produce hydrogen peroxide, glycine oxidase is located in the peroxisomes.

At this stage it is not possible to say whether the defect in patients with primary hyperoxaluria is increased formation of glyoxylate or failure of transamination of glyoxylate back to glycine; however, most inborn errors of metabolism result in loss of enzyme activity (although there are examples of conditions associated with excessive activity of an enzyme).

Both the oxidation of glyoxylate to oxalate and the reduction of glyoxylate to glycollate are catalysed by lactate dehydrogenase.

The graphs below show the excretion of oxalate, glycollate and glyoxylate in patients with type I primary hyperoxaluria and in red the range found in normal control subjects:

[From data reported by Hockaday TDR, Clayton JE, Frederick EW & Smith LH (1964) Primary hyperoxaluria. Medicine 43: 315]

These results confirm that the primary problem is excessive formation of glyoxylate from glycine; excretion of glyoxylate, glycollate and glycine are all very high in the patients compared with controls.

Frederick et al. investigated the metabolism of a test dose of [14C]glyoxylate in patients with hyperoxaluria, their parents and control subjects:

[Frederick EW, Rabkin MT, Richie RH & Smith LH (1963) Studies on primary hyperoxaluria. New England Journal of Medicine 269: 821-9.]

These results show increased formation of oxalate and glycollate from glyoxylate in hyperoxaluric patients compared with control subjects, suggesting that it is most likely that the problem is indeed one of failure to transaminate glyoxylate back to glycine. The parents of the affected children (who show no clinical signs) show increased excretion of oxalate (but not glycollate); this is what would be expected of heterozygotes for a recessive condition, and is consistent with partial failure to remove glyoxylate by transamination, so that it accumulates and is oxidised to oxalate.

A number of attempts were made to identify the enzyme defect in patients with primary hyperoxaluria; early studies on small biopsy samples showed no lack of any enzyme.

The breakthrough came when Watts and coworkers treated a patient by combined liver and kidney transplantation. This corrected the patient's metabolic lesion in vivo (and replaced the damaged kidneys). It also provided a whole liver from an affected person for biochemical investigation, as opposed to the small amounts of biopsy tissue or post mortem tissue (from patients who had died of uraemia, so that the liver was grossly abnormal) that had been available previously.

The key finding was that there was a nearly complete deficiency of alanine:glyoxylate transaminase in the peroxisomes from this patient, although there was more or less normal activity in whole liver homogenate and in mitochondria.

[Danpure CJ & Jennings PR (1986) Peroxisomal alanine:glyoxylate transaminase deficiency in primary hyperoxaluria type I. FEBS Letters 201: 20-4].

The defect seems to be mistargeting of the enzyme, so that while it is present in the cell, there is little or none in the peroxisomes; rather it has been targeted to mitochondria. Glycine oxidase, which catalyses the formation of glyoxylate is also a peroxisomal enzyme. This means that while glyoxylate can be synthesised in the peroxisomes, it cannot be transaminated back to glycine in these patients.

The occurrence of hyperoxaluria when this transaminase is mistargeted and absent from peroxisomes shows the importance of glycine oxidase in nitrogen metabolism. The glycine-glyoxylate couple has a major role in the transdeamination of a wide variety of amino acids.

Over the years it has been observed that experimental vitamin B6 deficiency leads to increased excretion of oxalate; indeed, measurement of urinary oxalate excretion has been suggested as an index of vitamin B6 nutritional status (however, it is too variable from person to person and from day to day to be useful in nutritional studies).

Takada et al. studied the effect of vitamin B6 deficiency in rats, measuring the excretion of oxalate (the graph on the left) and the activity of alanine:glyoxylate transaminase in liver peroxisomes (the graph on the right).

[From data reported by Takada Y, Mori T & Noguchi T (1984) The effect of vitamin B6 deficiency on alanine:glyoxylate transaminase isoenzymes in rat liver. Archives of Biochemistry and Biophysics 229: 1-6.]

These results with vitamin B6 deficient rats confirm that alanine-glyoxylate transaminase (like all other mammalian transaminases) is pyridoxal phosphate dependent, and that loss of activity of this enzyme due to loss of its cofactor results in increased formation and excretion of oxalate. This provides further confirmation that impairment of the transaminase is the defect in primary hyperoxaluria.

There are three main variants of type I primary hyperoxaluria, due to different abnormalities of alanine-glyoxalate transaminase:

For patients in group B, the enzyme can be activated in vitro by adding pyridoxal phosphate, and in vivo by providing very high intakes of vitamin B6 (of the order of 50 - 100 mg /day, compared with a reference intake of < 2 mg /day).

There seem to be three different types of primary hyperoxaluria associated with defective alanine-glyoxylate transaminase:

Type (A) is a classical inborn error of metabolism; there is no enzyme activity and no enzyme protein. This may be a deletion or nonsense mutation leading to early termination of translation, or a mutation in the promoter region of the gene so that is is not transcribed.

Type (B) is an example of a vitamin dependency disease. There is enzyme protein (detected immunologically), but it has no catalytic activity at normal concentrations of its coenzyme, pyridoxal phosphate. Feeding extremely high intakes of vitamin B6 leads to significant activation of the enzyme suggesting that the mutation affects the affinity of the coenzyme binding site and a much higher than normal concentration of pyridoxal phosphate is required for binding. This

Type (C) is an extremely interesting group of patients who show both immunoreactive protein and enzyme activity in the whole liver homogenate, but have classical type I primary hyperoxaluria. In liver biopsy samples from nine patients in this group, it was possible to perform assays for the enzyme in different subcellular compartments; all of them showed no activity in peroxisomes, but significant activity in the mitochondria.

[Takada Y, Kaneko N, Esumi H, Purdue PE & Danpure CJ (1990). Human peroxisomal L-alanine:glyoxylate transaminase. Biochemical Journal 268: 517-20. ]

The defect seems to be in the targeting of the enzyme, so that after synthesis on the ribosomes it is targeted to the mitochondria rather than to the peroxisomes.

The subcellular distribution of alanine:glyoxylate transaminase differs in different species:

Investigation of the mitochondrial and peroxisomal forms of AGT in the rat shows that the mature proteins have the same amino acid sequence. However, there are two different forms of mRNA for AGT; they differ only in that the one mRNA has a 5' extension which codes for a 22 amino acid amino-terminal sequence.

Purdue et al. deduced the amino terminal amino acid sequences of the products of translation of these two mRNA species,

type A: Met-Phe-Arg-Met-Leu-Ala-Lys-Ala-Ser-Val-Thr-Leu-Gly-Ser-Arg-Ala-Ala-Ser-Trp-Val-Arg-Asn-Met-Gly-Ser---

type B: Met-Gly-Ser---

The amino terminal extension of type A mRNA is a typical mitochondrial-targeting signal peptide; four basic amino acids in an alpha-helical region, arranged in such a way that the basic amino acids and the terminal methionine lie on the same side of the helix.

[Purdue PE, Lumb MJ & Danpure CJ (1992) Molecular evolution of alanine/glyoxylate transaminase I intracellular targeting: analysis of the marmoset and rabbit genes. European Journal of Biochemistry 207: 757-66]

In various peroxisomal proteins, two tripeptide sequences have been shown to act as peroxisomal targeting sequences: -Ser-Lys-Leu- in some proteins and -Gly-Arg-Leu- in others. In each case the targeting sequence is near, but not at, the carboxyl terminal of the protein.

The difference between the rat mitochondrial and peroxisomal forms of AGT is the 22 amino acid amino terminal extension of the mitochondrial isoenzyme, which is a mitochondrial targeting signal peptide. As with other signal peptides, this serves to target the protein from the ribosome to its appropriate site, but is then removed during post-translational processing of the protein. Remember that proteins are synthesised from the amino terminal (the 5'-end of the mRNA), so that it is most likely that the signal peptide will enter the mitochondria before translation is complete. Therefore a protein synthesised with the mitochondrial sequence will go into the mitochondria even if it has a peroxisomal targeting sequence near the carboxyl terminal.

The cDNA sequences show that rat and marmoset have two initiator sequences (ATG). Translation from the 5'-most would give the longer protein, with a mitochondrial targeting sequence; translation from the second would give the enzyme without a mitochondrial targeting sequence, which will therefore respond to its peroxisomal targeting sequence.

Purdue et al. compared the 5'-region nucleotide sequence of cDNA for alanine glyoxylate transaminase, and the amino acid sequence deduced from the nucleotides, for the rat, marmoset, human and rabbit enzymes. For the 5'-region nucleotide sequences the ATG initiator codons are shown in red. For the predicted amino acid sequences those shown in blue are translation of the codons that would be expected if there was an initiator codon at the 5' end of this sequence

Rat
ATG-TTC-CGG-ATG-TTG-GCC-AAG-GCC-AGT-GTG-ACG-CTG-GGC-TCC-AGG-GCA-GCA-AGT-TGG-GTA-CGG-AAC-ATG-GGC-TCG--
Met-Phe-Arg-Met-Leu-Ala-Lys-Ala-Ser-Val-Thr-Leu-Gly-Ser-Arg-Ala-Ala-Ser-Trp-Val-Arg-Asn-Met-Gly-Ser--

Marmoset
ATG-TTC-CAG-GCT-CTG-GCC-AAG-GCC-AGT-GCA-GCC-CTG-GGT-CCC-AGA-GCA-GCA-GGT-TGG-GTG-AGG-ACT-ATG-GCG-TCG--
Met-Phe-Gln-Ala-Leu-Ala-Lys-Ala-Ser-Ala-Ala-Leu-Gly-Pro-Arg-Ala-Ala-Gly-Trp-Val-Arg-Thr-Met-Ala-Ser--

Human
ATA-TTC-CAG-GCT-TTG-GCC-AAG-GCC-AGT-GCA-GCC-CCA-GGT-TCC-CGA-GCG-GCA-GGT-TGG-GTG-CGG-ACC-ATG-GCC-TCT--
Ile-Phe-Gln-Ala-Leu-Ala-Lys-Ala-Ser-Ala-Ala-Pro-Gly-Ser-Arg-Ala-Ala-Gly-Trp-Val-Arg-Thr-Met-Ala-Ser--

Rabbit
ACA-CCC-CAG-TCA-CCA-GCC-AAG-GCC-AGT-GTG-GCC-CTG-GGG-CCC-CGA-CGC-GCA-GGT-CGT-GTG-CAG-ACC-ATG-GCC-TCC--
Thr-Pro-Gln-Ser-Pro-Ala-Lys-Ala-Ser-Val-Ala-Leu-Gly-Pro-Arg-Ala-Ala-Gly-Arg-Val-Gln-Thr-Met-Ala-Ser--

[Purdue PE, Lumb MJ & Danpure CJ (1992) Molecular evolution of alanine/glyoxylate transaminase I intracellular targeting: analysis of the marmoset and rabbit genes. European Journal of Biochemistry 207: 757-66]

Comparison of the human 5' sequence with that of rat and marmoset suggests that at some stage in evolution the first ATG initiator codon has been mutated to ATA, so that the mitochondrial targeting sequence is not translated. A mutation of this ATA back to ATG would give an intact mitochondrial targeting sequence (compare the untranslated region of the human gene, shown in blue above, with the rat and marmoset sequences).

Remember that proteins are synthesised from the amino terminal (the 5'-end of the mRNA), so that it is most likely that the signal peptide will enter the mitochondria before translation is complete. Therefore a protein synthesised with the mitochondrial sequence will go into the mitochondria even if it has a peroxisomal targeting sequence near the carboxyl terminal.

Key points from this exercise:

• Type I primary hyperoxaluria is a recessive genetic condition, with excessive excretion of oxalate, which crystallises in the kidneys,
• It is due to increased synthesis of oxalate from glycine, and the defective enzyme of alanine-glyoxylate transaminase
• Three variants of type I primary hyperoxaluria can be distinguished:
• absence of immunoreactive protein and enzyme activity - a classical inborn error of metabolism due to a deletion or nonsense mutation
• absence of enzyme activity but with immunologically cross-reactive protein that can be activated by high concentrations of its cofactor, pyridoxal phosphate. Here the mutation is in the coenzyme binding site, leading to a low affinity for pyridoxal phosphate
• presence of both immunologically cross-reactive protein and enzyme activity, but incorrectly targeted to mitochondria instead of peroxisomes
• In species in which alanine-glyoxylate transaminase is located in both the mitochondria and peroxisomes there are two initiation sites; the mRNA for the mitochondrial enzyme codes for a 22 amino acid extension at the amino terminal end of the protein which codes for a mitochondrial targeting sequence.
• The human alanine-glyoxylate transaminase gene has this mitochondrial targeting sequence, but the initiator codon has mutated from ATG to ATA, so that it is not transcribed, and the enzyme is targeted to peroxisomes because it also has a peroxisomal targeting sequence near the carboxyl terminal end.
• In patients with primary hyperoxaluria in whom the enzyme is found in mitochondria rather than peroxisomes there has bee a reverse mutation to ATG, so that the mitochondrial targeting sequence is expressed.
• Alanine-glyoxylate transaminase and glycine oxidase provide a pathway for the transdeamination of most amino acids, and therefore have a central role in amino acid and N metabolism.