Primary hyperoxaluria type 1 and peroxisome-to-mitochondrion mistargeting of alanine: glyoxylate aminotransferase

Biochimie (1993) 75, 309-315 © Soci6t6 frangaise de biochimie et biologie mol6culaire / Elsevier, Paris

309

Primary hyperoxaluria type I and peroxisome-to-mitochondrion mistargeting of alanine:glyoxylate aminotransferase CJ

Danpure

Biochemical Genetics Research Group, MRC Clinical Research Centre, Watford Road, Harrow, Middlesex HAl 3UJ, UK (Received 20 November 1992; accepted 3 December 1992)

Summary m Under the influence of dietary selection pressure, the intracellular compartmentalization of alanine:glyoxylate amino-

transferase (AGT) has changed on many occasions during the evolution of mammals. In some mammals, AGT is peroxisomal in others it is mainly mitochondrial while in yet others it is more-or-less equally divided between both organelles. Although in normal human liver AGT is usually found exclusively within the peroxisomes, in some individuals a small proportion (--5%) is found also in the mitochondria. This apparently trivial intracellular redistribution of AGT is caused by the presence of a PromlLeu polym.orphism which allows the N-terminus of AGT to fold into a conformation (ie a positively-charged amphiphilic o~-helix) which functions as a mitochondrial targeting sequence. In one third of patients with the autosomal recessive disease primary hyperoxaluria type I, there is a further redistribution of AGT so that the great majority (--90%) is located in the mitochondria and only a small minority (10%) in the peroxisomes. AGT cannot fulfil its proper metabolic role in human liver (ie glyoxylate detoxification) when located in the mitochondria. This erroneous compartmentalization is due to the presence of a GlylToArg mutation superimposed upon the Pro~Leu polymorphism. The GlytToArg mutation appears to have no direct effect on mitochondrial targeting and is predicted to enhance mitochondrial import of AGT by interfering with its peroxisomal targeting and/or import. The mitochondrial targeting sequence generated by the Prom~Leu polymorphism is not homologous to that found in the AGT of other mammals which localise AGT within the mitochondria normally. The identity of the peroxisomal targeting sequence in AGT is unknown, but the Glyn0Arg mutation is found in a highly conserved region of the protein which might be involved in some aspect of the peroxisomal import pathway for AGT.

primary hyperoxaluria type I / alanine:glyoxylate aminotransferase / peroxisomal diseases / peroxisomal targeting/import / mitochondrial targeting/import Primary hyperoxaluria type I and alanine:glyoxylate aminotransferase deficiency Clinical, biochemical and enzymological characteristics o f primary hyperoxaluria type 1 (PHI) PrimalT hyperoxaluria type 1 (PH1; McKusick 259900) is a rare autosomal recessive disease caused by a deficiency of the liver-specific peroxisomal enzyme alanine:glyoxylate aminotransferase (AGT, EC 2.6.1.44) [1, 2]. A G T is a homodimeric protein of

Abbreviations: AGT, alanine:glyoxylate aminotransferase; SPT, serine:pymvate aminotransferase (identical to AGT); MTS, mitochondrial targeting sequence; PTS, peroxisomal targeting sequence; PHI, primary hyperoxaluria type 1; CRM +, presence or absence of AGT immunologically cross-reacting material; ENZ +, presence or absence of AGT catalytic enzyme activity.

subunit molecular mass =40 kDa [3] which, in humans at least, catalyses the intraperoxisomal transamination of glyoxylate to glycine using alanine as the amino donor [4] (reaction a~ in fig 1). Failure of this detoxification reaction in PH1 allows glyoxylate to be oxidized to oxalate, catalysed by glycolate oxidase in the peroxisome (c 2 in fig 1) and/or lactate dehydrogenase in the cytosol (el), and reduced to glycolate in the cytosol catalysed by glyoxylate reductase (d~) and possibly lactate dehydrogenase (e2). Excessive synthesis of oxalate and glycolate leads to increased renal excretion, concomitant hyperoxaluria and hyperglycolic aciduria being characteristic: biochemical features of PH1 [5]. Calcium oxalate is very insoluble at physiological pHs and all the recognisable pathological sequelae of PH1 are caused by its inappropriate deposition, initially in the kidney as urolithiasis and/or nephrocalcinosis, but following renal

310

Peroxisome

MitochondrionGluco[ DG,~f neogenesi$ ~ d2

OHP

Ser Gtuco. OHP - - - ~ OHP < .eogenesis~,

X

~a Glycol

"',, ox . . . . .

Glycol

Cytosol Pro11+ HaN~ -22

* 1 11 170

~

oZ

/ I~1 *

~MTS =[] aved) L e u , ~ (PH1 patients) I- COOH ~392

Translationproduct ~ in marmoset& rat ~

I PTS = ?

~

Gly17o\

Pyr

Glyox<--Oxal

(cleaved} , ~

Enzymic heterogeneity of PH1

Ala

a k Na~. ~ ./ ~

f

Ser

~

I Pro/Leu,,+ GIy,ro

HiN.-~ * 1 11 1 7 0 / /

/

/-

failure, also throughout the body as systemic oxalosis [5]. Unlike the diseases of aberrant peroxisomal biogenesis, but like other isolated peroxisomal enzyme deficiency diseases, the peroxisomes in PHI liver are relatively normal in number and morphological appearance [6].

l- COOH 392

Translationproduct in human,rabbit, marmoset& rat

Fig 1. Intracellular targeting/mistargeting of AGT and its metabolic consequences. The longer of the two AGT translation products of 414 amino acids found in marmoset and rat possesses an N-terminal 22 amino acid MTS (-22 to-1). This peptide is targeted to the mitochondria where the MTS is cleaved. In these mammals, the principal role of AGT in the mitochondria is one of gluconeogenesis (reaction a4). The shorter AGT translation product of 392 amino acids found in human, rabbit, and also marmoset and rat does not possess a leader MTS and is targeted to the peroxisomes, where its main role is in the detoxification of glyoxylate (al). In marmoset, rat, rabbit and most normal humans, the residues at positions 11 and 170 are Pro and Gly respectively. In a minority of normal humans, Leu is found at position 11. Although this substitution generates a MTS, AGT in these individuals is still mostly targeted to the peroxisomes. In PHI patients with the peroxisome-to-mitochondrion mistargeting phenotype, the additional presence of Arg at position 170 inhihits peroxisomal targeting/import so that the MTS generated by the Leu at position 11 can function efficiently, targeting 90% of the product to the mitochondria, where the MTS is not cleaved. This mitochondrial AGT is unable to detoxify glyoxylate efficiently (a3). Bold line, glyoxylate detoxification reaction catalysed by peroxisomal AGT (at); hatched line, gluconeogenesis pathway initiated in the mitochondria by AGT catalysing the serine-

PHI is surprisingly heterogeneous at the enzymological level, with respect to both A G T catalytic activity and A G T immunoreactivity [7-9]. There are three classes of PH1 patients as determined by enzymological phenotype. About 50% of patients have undetectable levels of A G T immunoreactive protein ( C R M - ) and undetectable levels of A G T catalytic activity (ENZ-), while 20% are CRM+/ENZ - and 30% are CRM+/ENZ + [9]. AGT catalytic activity in the E N Z + patients can vary from =3% to nearly 50% of the mean normal level. The levels of A G T activity found in many of the latter patients are similar to those found in some asymptomatic obligate heterozygotes

Peroxisome-to-mitochondrion mistargeting of AGT in P H I Intracellular compartmentafization of AGT in mammals During the evolution of mammals, the subcellular distribution of A G T appears to have changed on numerous occasions. For example, in human, macaque, baboon and rabbit liver, A G T is 100% peroxisomai [3, 10-14]; in cat and dog liver it is =95% mitochondrial and =5% peroxisomal [3, 11, 15]; while in marmoset and rat liver it is approximately ..equally divided between peroxisomes and mitochondria [3, 11, 12,

pyruvate transamination (SPT) reaction (a4); dotted line, glyoxylate translocation/metabolic pathway required for its inefficient detoxification in the mitochondria (a3); I 1 leader MTS in marmoset and rat (amino acid residues -22 to -1); [] homologous region at N-termipus of the shorter translatior~ product which functions as a non-cleavable MTS in human AGT only after the Pro.Leu substitution (amino acid residues 1 to --16). Enzymes are as follows: at/a 3) AGT (also SPT a2/a4); b) D-amino acid oxidase; cm/c2) glycolate oxidase; dVd2) glyoxylate reductase/D-glycerate dehydrogenase; el/e 2) lactate dehydrogenase; f) multistep conversion to glucose 6-phosphate (gluconeogenesis). Metabolites are as follows: Glyox, glyoxylate; Glycol, glycolate; Gly, glycine; Oxal, oxalate; Ala, alanine; Pyr, pyruvate; Set', serine; OHP, hydroxypyruvate; DG, D-glycerate.

311 -22

~

1

392

-- H u m a n

(PH1):

H2N- i f q a 1 a k a s a a p g s r a a g w v . r tl~.SHI~LLVTPLIC&LLKPLS I P . . . . .

-- H u m a n

(norm):

H2N-ifqalakasaapgsraagwvrtMASHKLLVTPPKALLKPLS

IP ..... KKL-COOH

Marmoset:

H2 N - M F Q A L A K A S A A L G P R A A G W V R T M A S H Q L L V A P P K A L L K P L S

IP . . . . . K K L - C O O H

Rabbit:

H2N-tpqspakasva

Rat:

H 2N-MFRMLAKASVTLGSRAASWVRNMGSHQLLVPPPEALSKPLS

IgpraagrvqtMASRQLLVAPPEALRKPLCTP

I

I MTS (marmoset rat)

II

I (human

NNL-COOH

..... SQL-COOH

I P ..... NKL-COOH

_/

MTS PH1 p a t i e n t s )

LTJ PTS ?

Fig 2. N-terminal and C-terminal sequences of mammalian AGT. The species are ordered in their presumed phylogenetic relationship. Human (norm), most normal humans expressing the major allele; human (PH1), PH 1 patients with the peroxisometo-mitochondrion mistargeting phenotype and the minority of normal humans expressing the minor allele. Amino acid sequences: upper case, translated sequence; lower case, untranslated sequence; -22 to -1, strong MTS in marmoset and rat (double underlined) (arrow, site of post-import cleavage) but 5'UTR in human and rabbit; 1 to =16, non-cleaved weak MTS (double underlined) due to the Pro, mLeu substitution in PHI patients with the mistargeting phenotype and normal humans expressing the minor allele (the downstream extent of this MTS is unknown). The C-terminal tripeptides, although similar to the SKL PTS motif, are unlikely to act as PTSs in AGT (see text).

16-18]. Loss of the ability to target AGT to the mitochondria in human and rabbit has heen shown to be due to the evolutionary loss of the more 5' of two alternative translation initiation sites, so that the 22 amino acids which comprise the mitochondrial targeting sequence (MTS) in the marmoset [19] and rat [20] are excluded from the open reading frame [19, 21] (see fig 2). Evolutionary selection of such mutations is probably related to diet and associated with the putative dual metabolic role of AGT with respect to glyoxylate detoxification and gluconeogenesis [ 11]. Consequently, herbivorous diets (high in oxal~tte [22, 23] and glycolate [24]) would be expected to favour a glyoxylate detoxification role for AGT within the peroxisomes (an in fig 1), while carnivorous diets (low in oxalate and glycolate, but high in protein) would favour a gluconeogenic role for AGT within the mitochondria. AGT also has serine:pyruvate aminotransferase (SPT) activity [25, 26] and it is probably this SPT catalytic reaction (a t) which is principally involved with its gluconeogenic function.

Although the AGT in hepatocytes from most normal humans is exclusively localized within the peroxisomes [10, 13, 14, 27], a few percent of the normal population have low, but detectable, levels (--5%) in the mitochondria as well [28].

AGT mistargeting in PHI In all the CRM+/ENZ - PH 1 patients so far studied, the immunoreactive, but catalytically defunct, AGT is entirely peroxisomal [13, 29]. However, in all ENZ+/CRM+ PH 1 patients so far studied, the intracellular distribution of AGT is very different. In these patients, about 90% of the immunoreactive AGT is localized in the mitochondria and only 10% in the peroxisomes [30]. This apparent mislocalization is specific for AGT and is not found with other peroxisomal enzymes, such as catalase, glycolate oxidase, D-amino acid oxidase, acyl-CoA oxidase, thiolase or hydratase [30]. Although the mitochondrial AGT in these patients appears to be catalytically active in vitro, it

312 must be metabolically inactive (or at least very inefficient) in vivo. Presumably efficient transamination (detoxification) of glyoxylate can only occur within the peroxisome, which is its major site of synthesis from glycine, catalysed by D-amino acid oxidase (b in fig 1), and glycolate, catalysed by glycolate oxidase (c~) [31]. Glyoxylate detoxification within the mitochondria (a3) would require the glyoxylate to diffuse out of the peroxisome into the cytosol, where most would be either oxidised to oxalate, catalysed by lactate dehydrogenase (e~) or reduced to glycolate, catalysed by glyoxylate reductase (dr), before being taken up into the mitochondria (see fig 1). Such an organelle-to-organelle protein trafficking defect is without precedent in human genetic disease, and is all the more surprising as the targeting and import of proteins into peroxisomes and mitochondria is dependent on specific, yet different, targeting information contained within the protein itself (ie mitochondrial targeting sequences (MTSs) and peroxisomal targeting sequences (PTSs)). Molecular basis of AGT mistargeting in PHI The normal human AGT gene

AGT is encoded by a single gene in all mammals so far studied (ie human [32], marmoset, rabbit [19] and rat [331). The human gene, which has been cloned and sequenced at both the cDNA [21] and genomic levels 1321, is comprised of !1 exons spread over =!0 kb, and is located on chromosome 2q36-37 1321. A cDNA clone of 1619 nucleotides, isolated from a HepG2 library, contained a continuous open reading frame of 1179 nucleotides predicted to encode a peptide of 392 amino acids with a molecular mass of 43 026 Da 1211. Ttle deduced peptide sequence has 8{)--90% sequence identity with AGT characterized from other mammalian species I191. However, it is 22 residues

shorter than the full sequence deduced from marmoset and rat cDNAs, due to the exclusion of the N-terminal MTS from the open reading frame (see above and fig 2). Two different AGT alleles have heen identified in normal individuals. The less common (minor) allele, which has a frequency of =15% in Caucasian populations, varies from the major allele in at least three respects, two of which (CI54T and Aila2G point nucleotide substitutions) lead to amino acid differences (ProllLeU and Iso340Met) [28] (see table I). In addition, the minor allele contains a 74-bp duplication within intron I [34]. The

C154T(PronLeu) polymorphism generates a MTS

The functional significance of the intron 1 duplication is unknown and, as the 1SO34o residue is not evolutionarily conserved, its substitution by Met is unlikely to have any significant effect on the properties of the peptide. However, Pro~! is conserved and its replacement by Leu is predicted to have marked effects on the folding and functional characteristics of the N-terminus of human AGT [28, 35]. Using the criteria of Chou and Fasman [36] or Gamier et al [37], the Pro,~Leu substitution is predicted to increase the tendency of the N-terminus of AGT to fold into an o~-helix rather than a random coil [28]. Helical wheel analysis showed that the basic residues (His4, Lyss, Lysl2, Lys16 and the terminal Meh) in this region would align on one side of the helix, making the structure amphiphilic. Such conformations are typical of MTSs [38]. Empirical confirmation of these structural predictions is provided, at least in part, by circular dichroism studies, which have shown the increased tendency of a synthetic 20-mer peptide based on the Leu~-AGT N-terminal sequence to form an tx-helix compared to the equivalent Prol~-20-mer (unpublished observations). In addition, polyclonal antibodies raised against the Prol~-20-mer could only detect the Prot~-

Table !. Mutations/polymorphisms involved in the intracellular mistargeting/import of human AGT Nucleotide

Amino acid

Subceilular distribution of AGT

154

630

1142

!1

! 70

340

Perox (%)

Mito (%)

Major allele

C

G

A

Pro

Gly

lso

100

0

Minor allele

T

G

G

Leu

Gly

Met

95

5

PHla

T

A

G

Leu

Arg

Met

10

90

"Expressed allele(s) in almost all PHi patients with the peroxisome-to-mitochondrion mistargeting phenotype.

313 AGT encoded by the major allele in liver homogenates (ie not cross-react with LeuI~-AGT), whereas antibodies raised against the Leu~r20-mer could only detect Leu~ ~-AGT (unpublished observations). That the N-terminal s-helix of Leu~-AGT functions as a MTS has heen confirmed by in vitro mitochondrial import studies which showed that the Pro~Leu substitution is necessary and sufficient to target AGT to mitochondria, albeit with low efficiency [35]. The N-terminal 19 amino acids of Leu~.-AGT could also target DHFR constructs to mitochondria, whereas the equivalent sequence from Pro~-AGT could not. Unlike the situation found with most mitochondrial matrix proteins, L e u . r A G T was not cleaved after import, presumably due to the absence of a mitochondrial processing peptidase recognition site (see figs !, 2). The functional consequences of the Pro~-Leu polymorphism, as far as normal individuals are concerned, are shown by the fact that, although in individuals homozygous for the major allele (ie Pro~) AGT is 100% peroxisomal, in an individual homozygous for the minor allele (ie Leu~) some AGT (=5%) is targeted and imported into the mitochondria, while 95% remained peroxisomai [28] (table I). This would appear to reinforce the in vitro data that the MTS generated in AGT by the Pro,~Leu substitution is rather weak and inefficient, and in vivo cannot out compete the PTS Surprisingly, the MTS in human AGT generated by the Pro~Leu polymorphism is not homologous with the MTS found in the AGT of certain other mammals, such as the marmoset and rat, in which AGT is normally distributed equally between mitochondria and peroxisomes. In these animals, the MTS is encoded by the 22 codons immediately upstream of the region equivalent to the human MTS as a 22 amino acid cleavable leader sequence 119, 33, 391. The homologous region to this leader sequence is part of the 5' UTR in the human I32] (see fig 2).

Peroxisome-to-mitochondrion AGT mistargeting in PHI requires the presence of a G6.~oA(Glym~4rg) mutation on the allelic background of the CLs4T (PronLeu) polymorphism Analysis of the cDNA nucleotide sequence from one PHI patient with the peroxisome-to-mitochondrion AGT mistargeting phenotype demonstrated the presence of the C~54T and A~42G polymorphisms, previously associated with the minor allele [28]. In addition, a further G630Amutation was found which was not present in any normal control. This novel mutation is predicted to lead to a Gly!70-Arg substitution. Using a combination of genomlc PCR and either RFLP analysis or allele-specific oligonucleotide hybridization, it was shown that almost all PHi

patients with the AGT mistargeting phenotype (but none who did not) expressed the G630A mutation on a background of the minor allele (ie together with the CI54T and Alla2G polymorphisms). Although all such patients appear to be homozygous at the transcript (ie cDNA) level, only about 10-20% are homozygous at the genomic level, the remainder being heterozygous. In these latter patients, the apparently 'normal' allele appears not to be expressed, at least at the level of stable mRNA. As referred to previously, the minor AGT allele from normal individuals (ie without the G630A mutation) encodes a protein, only 5% of which is targeted and imported into the mitochondria in vivo. However, the superimposition of a G630A mutation on top of the C~54T polymorphism yields a protein, of which most (=90%) is imported into the mitochondria [28, 30] (table I). In vitro mitochondrial import studies showed that the Gly~70Arg substitution, resulting from the G63oA mutation, had no direct effect on mitochondrial import [35]. Therefore, the most likely explanation for the effect of this mutation is that it interferes with some part of the peroxisomal import pathway for AGT. In the presence of such a mutation, the rather weak MTS generated by the C~54T(PrOlI-Leu) polymorphism is able to target AGT to the mitochondria without competition from the peroxisomal targeting/import system.

Peroxisomal targeting sequence of AGT The PTSs in many peroxisomal proteins appear to be variants of the C-terminal SKL tripeptide motif first proposed by Gould et ai for firefly luciferase 1401. Although the C-terminuses of mammalian AGT are similar to this motif (KKL in human and marmoset, NKL in rat and SQL in rabbit), there are a number of reasons for suspecting that they might not act as PTSs in AGT. Firstly, in vitro mutagenesis of the C-terminus of firefly luciferase trom SKL to KKL abolished its peroxisomal targeting and import 11401. Secondly, although the consensus tripeptide motif allows some flexibility, a basic residue (ie K, R or H) as the penultimate residue is mandatory. On these counts, the C-terminuses of at least human, marmoset and rabbit AGT would not appear to contain the information necessary to target the protein to the peroxisomes. The GlytT0 in AGT, which is mutated to Arg in almost all PHI patients expressing the mistargeting phenotype, is itself evolutionarily conserved and is located in a highly-conserved internal run of =58 amino acids [ 19]. In addition, another mutation, which has also been implicated in AGT mistargeting, has also been localized to this area (unpublished observations). Therefore, it is possible that this internal region is important in the targeting and/or import of

314 AGT into the peroxisomes and might even contain the PTS.

Evolutionary postscript The recent elucidation of the molecular basis of the evolutionary loss of mitochondrial AGT targeting ability in human and rabbit and its reacquisition in a subset of PHI patients, has generated as many questions as answers, conceming not only the evolution of the AGT gene, but also the evolution of MTSs and PTSs in general. If the PTS of AGT is shown not to reside at the C-terminus, then the question arises as to why there should be a 2/3 match between the C-terminus of A G T and the SKL consensus PTS motif in all the mammals studied. This might be a coincidence or it might indicate that a C-terminal SKL-like motif was the PTS for AGT at some time in its evolutionary history, but that the acquisition of a new PTS in a different region of the protein removed the selection pressure to maintain the C-terminal motif. It is also not clear why a sequence (ie amino acid residues 1 to =16) that is close to being a MTS (ie such that a point mutation causes it to become one) should evolve immediately downstream of an already existing functional one (eg the marmoset/rat mitochondrial targeting presequence at -22 to - 1 ) (see fig 2). Perhaps at some point in the evolutionary history of AGT the MTS included not only the leader sequence found in the marmoset and rat, but also the next run of =16 amino acids (currently equivalent to the N-terminuses of human and rabbit AGT, the peroxisomal tbrm and the mature mitochondrial form of marmoset and rat AGT). Such extended sequences are likely to be more efficient MTSs than shorter ones [411. The ability to target AGT to the mitochondria has heen lost on at least six occasions during the evolution of mammals (unpublished observations). Whether the molecular mechanisms are all similar to those found for human and rabbit are not known. If they are, then this might indicate that the AGT gene contains a hot spot for such mutations, or it might simply demonstrate the strength of the dietary selection pressure on the evolutionary perpetuation of otherwise random mutational events. No doubt, further elucidation of the molecular bases of the evolutionary fluctuations of AGT intracellular compartmentalization will be required before these questions can be answered.

References Danpure CJ,' Jennings PR (1986) Peroxisomal alanine:glyoxylate aminotransferase deficiency in primary hyperoxaluria type 1. FEBS Lett 20 I, 20-24

2 Danpure CJ, Jennings PR (1987) Deficiency of peroxisomal alanine:glyoxylate aminotransferase in primary hyperoxaluria type 1. In: Peroxisomes in Biology and Medicine (Fahimi HD, Sies H, eds) Springer-Verlag, Berlin, 374-378 3 Takada Y, Noguchi T (1982) Subcellular distribution, and physical and immunological properties of hepatic alanine: glyoxylate aminotransferase isoenzymes in different mammalian species. Comp Biochem Physiol (B) 72, 597-604 4 Thompson JS, Richardson KE (1967) Isolation and characterization of an L-alanine:glyoxylate aminotransferase from human liver. J Biol Chem 242, 3614--3619 5 Williams HE, Smith LH (1983) Primary hyperoxaluria. In: The Metabolic Basis of Inherited Disease (Stanbury JB, Wyngaarden JB, Fredrickson DS, Goldstein JL, Brown MS, eds) McGraw-Hill, New York, pp 204-228 6 Iancu TC, Danpure CJ (1987) Primary hyperoxaluria type I" ultrastructural observations in liver biopsies. J Inherited Metab Dis 10, 330-338 7 Wise PJ, Danpure CJ, Jennings PR (1987) Immunological heterogeneity of hepatic alanine:glyoxylate aminotransferase in primary hyperoxaluria type 1. FEBS Lett 222, 17-20 8 Danpure CJ, Jennings PR (1988) Further studies on the activity and subcellular distribution of alanine:glyoxylate aminotransferase in the livers of patients with primary hyperoxaluria type 1. Ciin Sci 75, 315-322 9 Danpure CJ (1991) Molecular and clinical heterogeneity in primary hyperoxaluria type 1. Am J Kidney Dis 17, 366-369 10 Noguchi T, Takada Y (1979) Peroxisomal localization of alanine:glyoxylate aminotransferase in human liver. Arch Biochem Biophys 196, 645-647 11 Danpure CJ, Guttridge KM, Fryer P, Jennings PR, Allsop J, Purdue PE (1990) Subcellular distribution of hepatic alanine:glyoxylate aminotransferase in various mammalian species. J Cell Sci 97, 669--678 12 Takada Y, Noguchi T (1982) The evolution of peroxisomal and mitochondrial alanine:glyoxylate aminotransferase 1 in mammalian liver. Biochem Biophys Res Commun 108, 153-157 13 Cooper PJ, Danpure CJ, Wise PJ, Guttridge KM (1988) lmmunocytochemical localization of human hepatic alanine:glyoxylate aminotransferase in control subjects and patients with primary hyperoxaluria type 1. J Histochem Cytochem 36, 1285-1294 14 Yokota S, Oda T, lchiyama A (1987) Immunocytochemical localization of serine:pyruvate aminotransferase in peroxisomes of the human liver parenehymal cells. Histochemistry 87, 601--606 15 Okuno E, Minatogawa Y, Nakanishi J, Nakamura M, Kamoda N, Makino M, Kido R (1979) The subcellular distribution of alanine-glyoxylate aminotransferase and serine-pyruvate aminotransferase in dog liver. Biochem J 182, 877-879 16 Noguchi T, Minatogawa Y, Takada Y, Okuno E, Kido R (1978) Subcellular distribution of pyruvate (glyoxylate) aminotransferases in rat liver. Biochem J 170, 173-175 17 Noguchi 2", Takada Y, Oota Y (1979) Intraperoxisomal and intramitochondrial localization, and assay of pyruvate (glyoxylate) aminotransferase from rat liver. Hoppe Seyiers Z Physiol Chem 360, 919-927 18 Yokota S, Oda T (1984) Fine localization of serine:pyruvate aminotransferase in rat hepatocytes revealed by a post-embedding immunocytochemical technique. Histochemistry 80, 591-595

315 19

20

21

22 23 24 25 26

27 28

29

30

Purdue PE, Lumb M J, Danpure CJ (1992) Molecular evolution ot alanine:glyoxylate aminotransferase I intracellular targeting. Analysis of the marmoset and rabbit genes. Eur J Biochem 207, 757-766 Oda T, Miyajima H, Suzuki Y, Ichiyama A (1987) Nucleotide sequence of the eDNA encoding the precursor for mitochondrial serine:pyruvate aminotransferase of rat liver. Fur J Biochem 168, 537-542 Takada Y, Kaneko N, Esumi H, Purdue PE, Danpure CJ (1990) Human peroxisomal L-alanine: glyoxylate aminotransferase. Evolutionary loss of a mitochondrial targeting signal by point mutation of the initiation codon. Biochem J 268, 517-520 Zarembski PM, Hodgkinson A (1962) The oxalic acid content of English diets. Br J Nutr 16, 627-634 Zarembski PM, Hodgkinson A (1962) The determination of oxalic acid in food. Analyst 87, 698-702 Hands KS, Richardson KE (1980) Glycolate in the diet and its conversion to urinary oxalate in the rat. Invest Urol 18, 106-109 Noguchi T, Takada Y (1978) Purification and properties of peroxisomal pyruvate (glyoxylate) aminotransferase from rat liver. Biochem J 175,765-768 Noguchi T, Okuno E, Takada Y, Minatogawa Y, Okai K, Kido R (1978) Characteristics of hepatic alanine-glyoxylate aminotransferase in different mammalian species. Biochem J 169, 113-122 Noguchi T, Takada Y (1978) Peroxisomal localization of serine:pyruvate aminotransferase in human liver. J Biol Chem 253, 7598-7600 Purdue PE, Takada Y, Danpure CJ (1990) Identification of mutations associated with peroxisome-to-mitochondrion mistargeting of alanine/glyoxylate aminotransferase in primary hyperoxaluria type 1. J Cell Biol 111, 2341-2351 Cooper PJ, Danpure CJ, Wise PJ, Guttridge KM (1988) Immunoelectron-microscopic localisation of alanine:glyoxylate aminotransferase in normal human liver and type I hyperoxaluric livers. Biochem Soc Trans 16, 627-628 Danpure CJ, Cooper PJ, Wise PJ, Jennings PR (1989) An enzyme traff~'cking defect in two patients with primary hyperoxaluria type !: peroxisomal alanine/glyoxylate aml-

31 32

33

34

35

36 37

38 39

40 41

notransterase rerouted to mitochondria. J Cell Biol 108, 1345-1352 De Duve C, Baudhuin P (1966) Peroxisomes (microbodies and related particles). Physiol Rev 46, 323-357 Purdue PE, Lumb M J, Fox M, Griffo G, Hamon Benais C, Povey S, Danpure CJ (1991) Characterization and chromosomal mapping of a genomic clone encoding human alanine:glyoxylate aminotransferase. Genomics 10, 34-42 Oda T, Funai T, Ichiyama A (1990) Generation from a single gene of two mRNAs that encode the mitochondrial and peroxisomal serine:pyruvate aminotransferase of rat liver. J Biol Chem 265, 7513-7519 Purdue PE, Lumb MJ, Allsop J, Danpure CJ (1991) An intronic duplication in the alanine:glyoxylate aminotransferase gene facilitates identification ot mutations in compound heterozygote patients with primary hyperoxaluria type I. Hum Genet 87, 394-396 Purdue PE, Allsop J, Isaya G, Rosenberg LE, Danpure CJ (1991) Mistargeting of peroxisomal L-alanine:glyoxylate aminotransferase to mitochondria in primary hyperoxaluria patients depends upon activation of a cryptic mitochondrial targeting sequence by a point mutation. Proc Natl Acad Sci USA 88, 10900-10904 Chou PY, Fasman GD (1974) Prediction of protein structure. Biochemistry 13, 222-245 Gamier J, Ogusthorpe DJ, Robson B (1978) Analysis of the accuracy and implications of simple methods tor predicting the secondary structure of globular proteins. J Mol Biol 120, 97-120 yon Heijne C (1986) Mitochondrial targeting sequences may form amphiphilic helices. EMBO J 5, 1335-1342 Oda T, Ichiyama A, Miura S, Mori M (1984) Uptake and processing of serine:pyruvate aminotransferase precursor by rat liver mitochondria in vitro and in vivo. J Biochem (Tokyo) 95, 815-824 Gould SJ, Keller GA, Hosken N, Wilkinson J, Suhramani S (1989) A conserved tripeptide sorts proteins to peroxisomes. J Cell Biol 108, 1657-1664 Galanis M, Devenish RJ, Nagley P (1991) Duplication of leader sequence tbr protein targeting to mitochondria leads to increased import efficiency. FEBS Lett 282, 425-430