- Email: [email protected]
Molecular and Clinical Heterogeneity in Primary Hyperoxaluria Type 1
Molecular and Clinical Heterogeneity in Primary Hyperoxaluria Type 1 Christopher J. Danpure, SSe, PhD • The autosomal recessive disease primary hyperoxaluria type 1 (PH1) is caused by a functional deficiency of the liver-specific peroxisomal enzyme alanine:glyoxylate aminotransferase (AGT). An analysis of liver samples from 59 PH1 patients showed considerable heterogeneity at the enzymic level. Approximately two thirds of patients had zero AGT catalytic activity, whereas the remaining one third had activities that ranged from 3% to 48% of the mean normal level. Two thirds of patients with zero AGT activity also had zero immunoreactive AGT protein, while the other one third, together with all the patients with detectable AGT catalytic activity, had levels of immunoreactive AGT protein that varied from normal to only a few percent of normal. All patients with AGT catalytic activity had their enzyme in the wrong intracellular compartment (ie, mitochondria). On the other hand, in all but one of the patients with immunoreactive AGT protein, but zero catalytic activity, the inactive AGT was correctly located within the peroxisomes. This enzymic heterogeneity was matched by considerable heterogeneity at the clinical level (eg, age at onset, rate of progression, age at renal failure, etc). No simple relationship was found between the level of hepatic AGT and the severity of the disease. It is suggested that a lack of AGT might be responsible for a broader pathological phenotype than classically associated with PH1. The possibility is advanced that some patients with idiopathic oxalate stone disease might owe their predisposition to stone formation to a functional deficiency of AGT. © 1991 by the National Kidney Foundation, Inc. INDEX WORDS: Primary hyperoxaluria type 1; alanine:glyoxylate aminotransferase; peroxisomes; mitochondria; oxalate.
AS CLASSICALLY DEFINED,! primary hyftperoxaluria type 1 (PHI) is a rare autosomal recessive inborn error of glyoxylate metabolism, which is characterized biochemically by increased synthesis and excretion of oxalate and glycolate, and clinically by urolithiasis, nephrocalcinosis, and, in severe cases, systemic oxalosis. Although a small minority of patients die in early infancy from an acute neonatal form of the disease, most patients suffer from a chronic progressive condition and die in the second or third decade from the cumulative effects of calcium oxalate deposition (ie, renal failure). PHI is caused by a deficiency of the liver-specific peroxisomal enzyme alanine: glyoxylate aminotransferase 1 (AGT).2 A reduction in the intraperoxisomal transamination of glyoxylate to glycine in PHI liver allows more glyoxylate to be oxidized to oxalate, catalyzed either within the peroxisome by L-a-hydroxy acid oxidase (glycolate oxidase) or in the cytosol by lactate dehydrogenase.2.3 Glyoxylate can also be reduced to glycolate in the cytosol, probably cataFrom the Biochemical Genetics Research Group, MRC Clinical Research Centre, Harrow, UK. Address reprint requests to Christopher 1. Danpure, BSc, PhD, Head, Biochemical Genetics Research Group, Division of Clinical Cell Biology, Clinical Research Centre, Watford Rd, Harrow, Middlesex HAl 3Ul, UK. © 1991 by the National Kidney Foundation, Inc. 0272-6386/9111704-0003$3.00/0
366
lyzed mainly by glyoxylate reductase. 4 All of the pathological sequelae of PH 1 appear to be due to the increased synthesis of oxalate and the low solubility of calcium oxalate. None of the clinical symptoms have been attributed to the overproduction of glycolate. Over the past 4 to 5 years, this laboratory has assayed AGT and other enzymes in 68 liver samples (mainly percutaneous needle biopsies, but also open biopsy, autopsy, and hepatectomy specimens), obtained from patients aged 3 weeks to 65 years with definite or suspected primary oxalosis. This material was provided by 41 centers in 12 countries. In nine of these patients, PHI was excluded on the basis of normal AGT and reevaluation of the clinical data. From an analysis of the remaining 59 AGT-deficient PHI liver samples, it has become clear that PHI is extremely heterogeneous, not only at the clinical level, but also at the molecular level. At the clinical level, this heterogeneity is manifested by, for example, enormous differences in the age at disease onset (6 months to 57 years in the present study), rate of progression, and age at death « 1 year to >65 years). Marked variations also exist in the relative contributions made by urolithiasis, nephrocalcinosis, and systemic oxalosis, as well as the degree of hyperoxaluria and hyperglycolic aciduria. For example, a quarter of patients in the present study did not have elevated glycolate excretion despite significant hy-
American Journal of Kidney Diseases, Vol XVII, No 4 (April), 1991: pp 366-369
367
PRIMARY HYPEROXALURIA TYPE I
•
50
ec: 8
•
20
•
•• • • ••
••
•••••• • 2
•
•
•
-------------o------~-~-----
eo
0
O.
0
I
0.5
2
O. 0
5
00
10
o~
20
00
50
Age (y) at biopsy
Fig 1. AGT in liver biopsies from 59 AGT-deficient PH1 patients as a function of age at time of sampling. AGT enzyme activity expressed as a percent of the mean control activity was determined as described previously and corrected for cross-over from glutamate:glyoxylate aminotransferase. 9 ,lo The bottom of the control range is equivalent to 62%. CRM +, e; CRM -, 0; determined by immunoblotting. 11
peroxaluria. Some patients (four in the present group) are responsive to pharmacological doses of pyridoxine (pyridoxal phosphate is a cofactor for AGT). The clinical heterogeneity of PH 1 is matched by its enzymic heterogeneity (Fig 1). Approximately 60 % of the patients studied did not have any hepatic AGT enzyme activity (ENZ - ) (Figs 1 and 2). The remaining 40% of PHI patients (ENZ + ) had activities that ranged from 3 % to 48 % of the mean control level (Fig 1). Many of these ENZ + patients had levels of AGT activity that were similar to those found in asymptomatic obligate heterozygotes, and therefore would not have been expected to have had disease. Most of the ENZ patients also had undetectable levels of crossreacting materiallimmunoreactive AGT protein (CRM -). The others (CRM + IENZ -) had levels that varied from normal to only a few percent of normal. As expected, all the ENZ + patients were also CRM + (Figs 1 and 2). Of the 35 CRM + patients, 11 were ENZ - and 24 ENZ + (Fig 2). In CRM + patients it is possible to identify the subcellular distribution of the immunoreactive AGT protein irrespective of whether it is enzymically active or not, using protein A-gold immunoelectron microscopy.5 All except one of the six CRM + IENZ - patients studied had a normal peroxisomal subcellular distribution of immunoreactive AGT protein. On the other hand, in
all 14 of the 24 CRM + IENZ + patients studied, the majority (> 90 %) of the hepatic AGT was localized within the wrong intracellular compartment (ie, mitochondria).6 In these 14 ENZ + patients, and probably the other 10 ENZ + patients as well, the disease appeared to be due, at least in part, to a unique type of trafficking defect, in which AGT was erroneously targeted to the mitochondria instead of the peroxisomes. Because the peroxisome is probably the main site of glyoxylate synthesis in the liver, AGT is unable to properly perform its metabolic role (glyoxylate detoxification) when located in the mitochondria. Sequence analysis of the human AGT gene has shown it to have the potential to encode both carboxy-terminal peroxisomal targeting sequences and amino-terminal mitochondrial targeting sequences. 7 The molecular mechanisms behind the variable functional expression of these signals have yet to be fully elucidated. However, preliminary evidence (Purdue PE and Danpure CJ, 1990, unpublished observations) indicates that the mutations in the AGT gene responsible for the AGT targeting defect are the same in each patient. Attempts have been made to relate the level of hepatic AGT in PHI patients to various clinical manifestations of the disease. For example, all four pyridoxine-responsive patients in the present study were ENZ + , while all five patients with the acute neonatal form of the disease were ENZ - . Although early results showed that there appeared to be a relationship between the level of AGT deficiency and various measures of disease severity, 8 more extensive analysis of the much greater numTotal PHI patients-
CRM -
ENZ-
Subcell -
14M
1M
5P
Fig 2. Summary of AGT analysis of 59 AGT-deficient PH1 livers. ENZ + and ENZ - refer to presence (> 2%) or absence (:::; 2%) of significant AGT enzyme activity.9,10 CRM + and CRM - refer to presence or absence of detectable immunoreactive AGT protein. 11 P and M, principal subcellular localization of AGT: P = >99% peroxisomal, M = >90% mitochondrial. 5 .6
368
ber of patients presented here has failed to find any clinically useful relationship between AGT activity and disease severity. Because of the progressive nature of PHI and the likelihood that most patients are not subjected to liver biopsy without due cause, it is conceivable that the age of a patient at the time of liver sampling bears some relationship to disease severity. It can be seen from Figure I that, although all patients below the age of 5 years at the time of sampling were ENZ - , there were also much older patients (up to 65 years) who were also ENZ -. There are a number of possible reasons for this lack of correlation between AGT activity and disease severity. First, it is difficult to agree on independent quantifiable markers of intrinsic disease severity in PHI. Many of the measurable parameters characteristic of PHI (eg, systemic oxalosis, hyperoxaluria, hyperoxalemia) are themselves influenced secondarily by other consequences of the disease process, such as renal failure. Because of the progressive nature of PH I, most of the parameters would need to be age-related. But the progression is unlikely to be linear and it is almost impossible to determine the relative weighting that should be attributed to each one. Second, as far as the experience of this laboratory over the past 4 to 5 years is concerned, it is rare for all of the above potential markers of severity to have been measured or to be known. With respect to the patients in the present study, urinary oxalate was nearly always measured, but in more than a quarter of patients urinary glycolate was not determined. Plasma oxalate and glycolate were rarely measured. In addition, significant numbers of patients in this study had presented in renal failure, making oxalate determinations almost meaningless. The third reason is of a more fundamental nature. Although the underlying cause of PHI is clearly monogenic (ie, AGT deficiency), the quantitative and qualitative manifestations of its clinical effect are equally clearly multifactorial. The other factors involved are likely to be both genetic and environmental. In the former category, the levels of expression of a number of enzymes, such as Damino acid oxidase, L-a-hydroxy acid oxidase, glutamate:glyoxylate aminotransferase, lactate dehydrogenase, and glyoxylate reductase, would be expected to have effects on the endogenous synthetic rate of oxalate (Fig 3). Also in this category
CHRISTOPHER J. DANPURE
Other metabolic routes
Glycine
~( )~ Glyoxylate
LDH
LHO LDH
GR
Glycolate
~l~ AGT
Other metabolic routes
>
Oxalate
I
OIl(
Exogenous oxalate (diet)
Solubility factors (+&-)
t
Insoluble calcium oxalate Fig 3. Genetic and environmental factors likely to influence the effect of AGT deficiency in PH1. Peroxisomal enzymes: alanine:glyoxylate aminotransferase/serine:pyruvate aminotransferase (AGT) [deficient in PH1], o-amino acid oxidase/glycine oxidase (DAO), L-a-hydroxy acid oxidase/glycolate oxidase (LHO). Cytosolic enzymes: glutamate:glyoxylate aminotransferase/alanine:2-oxoglutarate aminotransferase (GGT), glyoxylate reductase/o-glycerate dehydrogenase (GR) [deficient in PH2], lactate dehydrogenase (LDH).
would be a host of genetic factors that, rather than influencing oxalate synthesis, would modify its effective solubility in the urine or elsewhere. In the environmental category could be placed a whole variety of influences from diet, infection (enteric and renal), and appropriateness of clinical management, which might be expected to affect oxalate synthesis, absorption, or solubility. The multifactorial nature of idiopathic oxalate stone disease is generally accepted. The heterogeneous and equally multifactorial nature of the clinical manifestations of PHI (as defined by AGT deficiency) cause the dividing line between this disease and idiopathic oxalosis to be ill-defined. The wide range of age (6 months to 57 years) at which the first symptoms of PHI appear and the presence or absence of hyperglycolic aciduria are good examples of heterogeneity that can not be explained purely on the level of AGT deficiency. Nevertheless without the clear demonstration of functional AGT deficiency (either complete deficiency of enzyme activity, or the presence of enzyme activity but in the wrong subcellular or-
369
PRIMARY HYPEROXALURIA TYPE I
ganelle), a number of patients in the present study might not have been diagnosed as PH 1 using classical criteria. I From these findings it is tempting to speculate that a proportion of idiopathic oxalate stone formers may owe their apparent predisposition to oxalosis to an underlying functional deficiency of AGT. The liver-specific expression of AGT in humans has hampered attempts to address such ideas. However, the cloning and sequencing of the human AGT gene 7 has opened up the possibility for the first time of screening such idiopathic
oxalate stone-forming populations by molecular genetic analysis, using material as readily accessible as peripheral blood leukocytes. ACKNOWLEDGMENT I wish to thank the following colleagues who have made invaluable contributions to this work over the past 5 years: Patricia Jennings (enzyme assays). Pauline Wise and Jenny Allsop (immunoblotting), Keith Guttridge, Penelope Cooper, and Patricia Fryer (immunoelectron microscopy) and Richard Watts who aroused my interest in PHI 5 years ago and provided me with the first clinical specimens.
REFERENCES 1. Williams HE, Smith LH: Primary hyperoxaluria, in Stanbury lB, Wyngaarden JB, Fredrickson DS (eds): The Metabolic Basis of Inherited Disease (ed 5). New York, NY, McGraw-Hili, 1983, pp 204-228 2. Danpure CJ, Jennings PR: Peroxisomal alanine:glyoxylate aminotransferase deficiency in primary hyperoxaluria type 1. FEBS Lett 201 :20-24, 1986 3. Danpure CJ: Recent advances in the understanding, diagnosis and treatment of primary hyperoxaluria type 1. J 1nher Metab Dis 12:210-224, 1989 4. Danpure CJ, Jennings PR, Mistry J, et al: Enzymological characterization of a feline analogue of primary hyperoxaluria type 2: A model for the human disease. J Inherited Metab Dis 12:403-414, 1989 5. Cooper Pl, Danpure Cl, Wise Pl, et al: Immunocytochemical localization of human hepatic alanine:glyoxylate aminotransferase in control subjects and patients with primary hyperoxaluria type 1. J Histochem Cytochem 36: 1285-1294, 1988 " 6" Danpure CJ, Cooper PJ, Wise Pl, et al: An enzyme trafficking defect in two patients with primary hyperoxaluria type
1: Peroxisomal alanine/glyoxylate aminotransferase rerouted to mitochondria. J Cell BioI 108:1345-1352, 1989 7. Takada Y, Kaneko N, Esurni H, etal: Human peroxisomal L-alanine:glyoxylate aminotransferase. Evolutionary loss of a mitochondrial targeting signal by point mutation of the initiation codon. Biochem 1 268:517-520, 1990 8. Danpure CJ, Jennings PR, Watts RWE: Enzymological diagnosis of primary hyperoxaluria type 1 by measurement of hepatic alanine:glyoxylate aminotransferase activity. Lancet 1:289-291, 1987 9. Danpure Cl, Jennings PR: Further studies on the activity and subcellular distribution of alanine:glyoxylate aminotransferase in the livers of patients with primary hyperoxaluria type 1. Clin Sci 75:315-322, 1988 10. Allsop J, lennings PR, Danpure Cl: A new micro-assay for human liver alanine:glyoxylate aminotransferase. Clin Chim Acta 170:187-194, 1987 11. Wise PJ, Danpure CJ, Jennings PR: Immunological heterogeneity of hepatic alanine:glyoxylate aminotransferase in primary hyperoxaluria type 1. FEBS Lett 222:17-20, 1987
AS CLASSICALLY DEFINED,! primary hyftperoxaluria type 1 (PHI) is a rare autosomal recessive inborn error of glyoxylate metabolism, which is characterized biochemically by increased synthesis and excretion of oxalate and glycolate, and clinically by urolithiasis, nephrocalcinosis, and, in severe cases, systemic oxalosis. Although a small minority of patients die in early infancy from an acute neonatal form of the disease, most patients suffer from a chronic progressive condition and die in the second or third decade from the cumulative effects of calcium oxalate deposition (ie, renal failure). PHI is caused by a deficiency of the liver-specific peroxisomal enzyme alanine: glyoxylate aminotransferase 1 (AGT).2 A reduction in the intraperoxisomal transamination of glyoxylate to glycine in PHI liver allows more glyoxylate to be oxidized to oxalate, catalyzed either within the peroxisome by L-a-hydroxy acid oxidase (glycolate oxidase) or in the cytosol by lactate dehydrogenase.2.3 Glyoxylate can also be reduced to glycolate in the cytosol, probably cataFrom the Biochemical Genetics Research Group, MRC Clinical Research Centre, Harrow, UK. Address reprint requests to Christopher 1. Danpure, BSc, PhD, Head, Biochemical Genetics Research Group, Division of Clinical Cell Biology, Clinical Research Centre, Watford Rd, Harrow, Middlesex HAl 3Ul, UK. © 1991 by the National Kidney Foundation, Inc. 0272-6386/9111704-0003$3.00/0
366
lyzed mainly by glyoxylate reductase. 4 All of the pathological sequelae of PH 1 appear to be due to the increased synthesis of oxalate and the low solubility of calcium oxalate. None of the clinical symptoms have been attributed to the overproduction of glycolate. Over the past 4 to 5 years, this laboratory has assayed AGT and other enzymes in 68 liver samples (mainly percutaneous needle biopsies, but also open biopsy, autopsy, and hepatectomy specimens), obtained from patients aged 3 weeks to 65 years with definite or suspected primary oxalosis. This material was provided by 41 centers in 12 countries. In nine of these patients, PHI was excluded on the basis of normal AGT and reevaluation of the clinical data. From an analysis of the remaining 59 AGT-deficient PHI liver samples, it has become clear that PHI is extremely heterogeneous, not only at the clinical level, but also at the molecular level. At the clinical level, this heterogeneity is manifested by, for example, enormous differences in the age at disease onset (6 months to 57 years in the present study), rate of progression, and age at death « 1 year to >65 years). Marked variations also exist in the relative contributions made by urolithiasis, nephrocalcinosis, and systemic oxalosis, as well as the degree of hyperoxaluria and hyperglycolic aciduria. For example, a quarter of patients in the present study did not have elevated glycolate excretion despite significant hy-
American Journal of Kidney Diseases, Vol XVII, No 4 (April), 1991: pp 366-369
367
PRIMARY HYPEROXALURIA TYPE I
•
50
ec: 8
•
20
•
•• • • ••
••
•••••• • 2
•
•
•
-------------o------~-~-----
eo
0
O.
0
I
0.5
2
O. 0
5
00
10
o~
20
00
50
Age (y) at biopsy
Fig 1. AGT in liver biopsies from 59 AGT-deficient PH1 patients as a function of age at time of sampling. AGT enzyme activity expressed as a percent of the mean control activity was determined as described previously and corrected for cross-over from glutamate:glyoxylate aminotransferase. 9 ,lo The bottom of the control range is equivalent to 62%. CRM +, e; CRM -, 0; determined by immunoblotting. 11
peroxaluria. Some patients (four in the present group) are responsive to pharmacological doses of pyridoxine (pyridoxal phosphate is a cofactor for AGT). The clinical heterogeneity of PH 1 is matched by its enzymic heterogeneity (Fig 1). Approximately 60 % of the patients studied did not have any hepatic AGT enzyme activity (ENZ - ) (Figs 1 and 2). The remaining 40% of PHI patients (ENZ + ) had activities that ranged from 3 % to 48 % of the mean control level (Fig 1). Many of these ENZ + patients had levels of AGT activity that were similar to those found in asymptomatic obligate heterozygotes, and therefore would not have been expected to have had disease. Most of the ENZ patients also had undetectable levels of crossreacting materiallimmunoreactive AGT protein (CRM -). The others (CRM + IENZ -) had levels that varied from normal to only a few percent of normal. As expected, all the ENZ + patients were also CRM + (Figs 1 and 2). Of the 35 CRM + patients, 11 were ENZ - and 24 ENZ + (Fig 2). In CRM + patients it is possible to identify the subcellular distribution of the immunoreactive AGT protein irrespective of whether it is enzymically active or not, using protein A-gold immunoelectron microscopy.5 All except one of the six CRM + IENZ - patients studied had a normal peroxisomal subcellular distribution of immunoreactive AGT protein. On the other hand, in
all 14 of the 24 CRM + IENZ + patients studied, the majority (> 90 %) of the hepatic AGT was localized within the wrong intracellular compartment (ie, mitochondria).6 In these 14 ENZ + patients, and probably the other 10 ENZ + patients as well, the disease appeared to be due, at least in part, to a unique type of trafficking defect, in which AGT was erroneously targeted to the mitochondria instead of the peroxisomes. Because the peroxisome is probably the main site of glyoxylate synthesis in the liver, AGT is unable to properly perform its metabolic role (glyoxylate detoxification) when located in the mitochondria. Sequence analysis of the human AGT gene has shown it to have the potential to encode both carboxy-terminal peroxisomal targeting sequences and amino-terminal mitochondrial targeting sequences. 7 The molecular mechanisms behind the variable functional expression of these signals have yet to be fully elucidated. However, preliminary evidence (Purdue PE and Danpure CJ, 1990, unpublished observations) indicates that the mutations in the AGT gene responsible for the AGT targeting defect are the same in each patient. Attempts have been made to relate the level of hepatic AGT in PHI patients to various clinical manifestations of the disease. For example, all four pyridoxine-responsive patients in the present study were ENZ + , while all five patients with the acute neonatal form of the disease were ENZ - . Although early results showed that there appeared to be a relationship between the level of AGT deficiency and various measures of disease severity, 8 more extensive analysis of the much greater numTotal PHI patients-
CRM -
ENZ-
Subcell -
14M
1M
5P
Fig 2. Summary of AGT analysis of 59 AGT-deficient PH1 livers. ENZ + and ENZ - refer to presence (> 2%) or absence (:::; 2%) of significant AGT enzyme activity.9,10 CRM + and CRM - refer to presence or absence of detectable immunoreactive AGT protein. 11 P and M, principal subcellular localization of AGT: P = >99% peroxisomal, M = >90% mitochondrial. 5 .6
368
ber of patients presented here has failed to find any clinically useful relationship between AGT activity and disease severity. Because of the progressive nature of PHI and the likelihood that most patients are not subjected to liver biopsy without due cause, it is conceivable that the age of a patient at the time of liver sampling bears some relationship to disease severity. It can be seen from Figure I that, although all patients below the age of 5 years at the time of sampling were ENZ - , there were also much older patients (up to 65 years) who were also ENZ -. There are a number of possible reasons for this lack of correlation between AGT activity and disease severity. First, it is difficult to agree on independent quantifiable markers of intrinsic disease severity in PHI. Many of the measurable parameters characteristic of PHI (eg, systemic oxalosis, hyperoxaluria, hyperoxalemia) are themselves influenced secondarily by other consequences of the disease process, such as renal failure. Because of the progressive nature of PH I, most of the parameters would need to be age-related. But the progression is unlikely to be linear and it is almost impossible to determine the relative weighting that should be attributed to each one. Second, as far as the experience of this laboratory over the past 4 to 5 years is concerned, it is rare for all of the above potential markers of severity to have been measured or to be known. With respect to the patients in the present study, urinary oxalate was nearly always measured, but in more than a quarter of patients urinary glycolate was not determined. Plasma oxalate and glycolate were rarely measured. In addition, significant numbers of patients in this study had presented in renal failure, making oxalate determinations almost meaningless. The third reason is of a more fundamental nature. Although the underlying cause of PHI is clearly monogenic (ie, AGT deficiency), the quantitative and qualitative manifestations of its clinical effect are equally clearly multifactorial. The other factors involved are likely to be both genetic and environmental. In the former category, the levels of expression of a number of enzymes, such as Damino acid oxidase, L-a-hydroxy acid oxidase, glutamate:glyoxylate aminotransferase, lactate dehydrogenase, and glyoxylate reductase, would be expected to have effects on the endogenous synthetic rate of oxalate (Fig 3). Also in this category
CHRISTOPHER J. DANPURE
Other metabolic routes
Glycine
~( )~ Glyoxylate
LDH
LHO LDH
GR
Glycolate
~l~ AGT
Other metabolic routes
>
Oxalate
I
OIl(
Exogenous oxalate (diet)
Solubility factors (+&-)
t
Insoluble calcium oxalate Fig 3. Genetic and environmental factors likely to influence the effect of AGT deficiency in PH1. Peroxisomal enzymes: alanine:glyoxylate aminotransferase/serine:pyruvate aminotransferase (AGT) [deficient in PH1], o-amino acid oxidase/glycine oxidase (DAO), L-a-hydroxy acid oxidase/glycolate oxidase (LHO). Cytosolic enzymes: glutamate:glyoxylate aminotransferase/alanine:2-oxoglutarate aminotransferase (GGT), glyoxylate reductase/o-glycerate dehydrogenase (GR) [deficient in PH2], lactate dehydrogenase (LDH).
would be a host of genetic factors that, rather than influencing oxalate synthesis, would modify its effective solubility in the urine or elsewhere. In the environmental category could be placed a whole variety of influences from diet, infection (enteric and renal), and appropriateness of clinical management, which might be expected to affect oxalate synthesis, absorption, or solubility. The multifactorial nature of idiopathic oxalate stone disease is generally accepted. The heterogeneous and equally multifactorial nature of the clinical manifestations of PHI (as defined by AGT deficiency) cause the dividing line between this disease and idiopathic oxalosis to be ill-defined. The wide range of age (6 months to 57 years) at which the first symptoms of PHI appear and the presence or absence of hyperglycolic aciduria are good examples of heterogeneity that can not be explained purely on the level of AGT deficiency. Nevertheless without the clear demonstration of functional AGT deficiency (either complete deficiency of enzyme activity, or the presence of enzyme activity but in the wrong subcellular or-
369
PRIMARY HYPEROXALURIA TYPE I
ganelle), a number of patients in the present study might not have been diagnosed as PH 1 using classical criteria. I From these findings it is tempting to speculate that a proportion of idiopathic oxalate stone formers may owe their apparent predisposition to oxalosis to an underlying functional deficiency of AGT. The liver-specific expression of AGT in humans has hampered attempts to address such ideas. However, the cloning and sequencing of the human AGT gene 7 has opened up the possibility for the first time of screening such idiopathic
oxalate stone-forming populations by molecular genetic analysis, using material as readily accessible as peripheral blood leukocytes. ACKNOWLEDGMENT I wish to thank the following colleagues who have made invaluable contributions to this work over the past 5 years: Patricia Jennings (enzyme assays). Pauline Wise and Jenny Allsop (immunoblotting), Keith Guttridge, Penelope Cooper, and Patricia Fryer (immunoelectron microscopy) and Richard Watts who aroused my interest in PHI 5 years ago and provided me with the first clinical specimens.
REFERENCES 1. Williams HE, Smith LH: Primary hyperoxaluria, in Stanbury lB, Wyngaarden JB, Fredrickson DS (eds): The Metabolic Basis of Inherited Disease (ed 5). New York, NY, McGraw-Hili, 1983, pp 204-228 2. Danpure CJ, Jennings PR: Peroxisomal alanine:glyoxylate aminotransferase deficiency in primary hyperoxaluria type 1. FEBS Lett 201 :20-24, 1986 3. Danpure CJ: Recent advances in the understanding, diagnosis and treatment of primary hyperoxaluria type 1. J 1nher Metab Dis 12:210-224, 1989 4. Danpure CJ, Jennings PR, Mistry J, et al: Enzymological characterization of a feline analogue of primary hyperoxaluria type 2: A model for the human disease. J Inherited Metab Dis 12:403-414, 1989 5. Cooper Pl, Danpure Cl, Wise Pl, et al: Immunocytochemical localization of human hepatic alanine:glyoxylate aminotransferase in control subjects and patients with primary hyperoxaluria type 1. J Histochem Cytochem 36: 1285-1294, 1988 " 6" Danpure CJ, Cooper PJ, Wise Pl, et al: An enzyme trafficking defect in two patients with primary hyperoxaluria type
1: Peroxisomal alanine/glyoxylate aminotransferase rerouted to mitochondria. J Cell BioI 108:1345-1352, 1989 7. Takada Y, Kaneko N, Esurni H, etal: Human peroxisomal L-alanine:glyoxylate aminotransferase. Evolutionary loss of a mitochondrial targeting signal by point mutation of the initiation codon. Biochem 1 268:517-520, 1990 8. Danpure CJ, Jennings PR, Watts RWE: Enzymological diagnosis of primary hyperoxaluria type 1 by measurement of hepatic alanine:glyoxylate aminotransferase activity. Lancet 1:289-291, 1987 9. Danpure Cl, Jennings PR: Further studies on the activity and subcellular distribution of alanine:glyoxylate aminotransferase in the livers of patients with primary hyperoxaluria type 1. Clin Sci 75:315-322, 1988 10. Allsop J, lennings PR, Danpure Cl: A new micro-assay for human liver alanine:glyoxylate aminotransferase. Clin Chim Acta 170:187-194, 1987 11. Wise PJ, Danpure CJ, Jennings PR: Immunological heterogeneity of hepatic alanine:glyoxylate aminotransferase in primary hyperoxaluria type 1. FEBS Lett 222:17-20, 1987