co m m e nta r y
REFERENCES 1. Moosa MR, Kidd M. The dangers of rationing dialysis treatment: the dilemma facing a developing country. Kidney Int 2006; 70: 1107– 1114. 2. Dirks J, Remuzzi G, Horton S et al. Diseases of the kidney and urinary system. In: Disease Control Priorities in Developing Countries, 2nd edn. Oxford University Press: New York, 2006, pp 695–706. 3. Jha V, Chugh K. Economics of ESRD in developing countries: India. In: El Nahas M (ed). Kidney Diseases in the Developing World and
4.
5.
6.
Ethnic Minorities. Taylor and Francis: London, 2005, pp 39–58. Lacson E, Kuhlmann MK, Levin NW et al. Outcomes and economics of ESRF. In: El Nahas M (ed). Kidney Diseases in the Developing World and Ethnic Minorities. Taylor and Francis: London, 2005, pp 15–38. Eastwood JB, Conroy RE, Naicker S et al. Loss of health care professionals from subSaharan Africa: the pivotal role of the UK. Lancet 2005; 365: 1893–1900. Seguin B, Singer P, Daar A. Scientific diasporas. Science 2006; 312: 1602–1603.
see original article on page 1115
Is there a genotype–phenotype correlation in primary hyperoxaluria type 1? BB Beck1 and B Hoppe1 There is ongoing debate about a genotype-phenotype correlation in patients with primary hyperoxaluria type 1 and specific AGXT mutations. However, other determinants like environmental factors or modifer genes may play a pivotal role in the heterogeneity of the disease. The report of Lorenzo and co-workers highlights this situation, presenting data of a whole population with just one specific AGXT mutation. Kidney International (2006) 70, 984–986. doi:10.1038/sj.ki.5001797
The term ‘primary hyperoxaluria type 1’ (PH1) refers to a hepatic defect in glyoxylate metabolism, namely a decrease or even loss of alanine:glyoxylate aminotransferase (AGT) activity. The disease itself causes no liver dysfunction but manifests as the most severe known urolithiasis entity. This is true with respect to both stone burden and prognosis. PH1 is considered to be rare but surely remains underreported with an estimated prevalence ranging from one to three per million population in Europe and North America.1 Higher rates are reported from inbred populations. Un- or misdiagnosed PH1 patients can defini1Division of Pediatric Nephrology, University Children’s Hospital, Cologne, Germany Correspondence: B Hoppe, Division of Pediatric Nephrology, University Children’s Hospital, Kerpener Strasse 62, D-50937 Cologne, Germany. E-mail:
[email protected]
984
tively be found in many dialysis programs, as a substantial proportion (up to >30%) of patients is diagnosed late with end-stage renal failure (ESRF) or, even worse, more or less by chance after a biopsy specimen is obtained for suspected renal graft rejection. There are several reasons for this dilemma. One is unfamiliarity with the rare monogenic disorder presenting with urolithiasis and with the linkage between stones and metabolic disorders. The fact that about half the children with stones, surprisingly, do not complain about pain or do so very little may unfortunately support this attitude. Second, while ESRF due to PH1 fortunately is rare in childhood, a number of patients are seen only in adult nephrology departments, which might make it more acceptable or simply necessary to treat patients without established diagnosis. Third, with anuria present, the
most valuable screening instrument, the determination of urinary oxalate excretion, is no longer available. PH1 is a mendelian disorder of autosomal-recessive inheritance with absolute penetrance but extremely variable expressivity. The enzyme physiologically allocated to liver peroxisomes converts glyoxylate to glycine. In its absence or because of mistargeting to mitochondria, glyoxylate will be increasingly converted to oxalate and glycolate (Figure 1). Oxalate cannot be further metabolized in humans and therefore has to be cleared by the kidney. Unfortunately, urine becomes quickly supersaturated with marked endogenous oxalate production. Nephrocalcinosis, which is crystal deposition within the distal tubular and collecting duct lumen, and recurrent stone formation are the clinical hallmarks of PH1 (Figure 1).1 Progressive nephrocalcinosis and recurrent urolithiasis along with their secondary complications (chronic interstitial inflammation and fibrosis) result somehow in chronic renal failure, which usually progresses to ESRF over time. Much worse than that, once significant loss of renal function occurs, systemic deposition of calcium oxalate (systemic oxalosis) invariably takes place in various tissues, including the retina, myocardium, vessel walls, central nervous system, bones, skin, and many more. This superimposes additional morbidity — for example, blindness; cardiomyopathy; arrhythmias, including heart block; oxalate osteopathy; treatment-resistant anemia; and so on.1 No form of dialysis treatment will stop or reverse systemic oxalate deposition. Many of the complications take years to reverse after combined kidney–liver transplantation; sadly, some, such as loss of vision, are irreversible. The remarkable single-center study by Lorenzo and co-workers2 (this issue) describes impressively the devastating course of systemic oxalosis not in a single patient but for a whole population — no small thing considering the relative paucity of long-term data. They clearly outline several key points regarding the clinical course of PH1: (1) PH1 patients who stay on chronic (hemo)dialysis are likely to die from systemic oxalosis. Kidney International (2006) 70
co m m e nta r y
Cytosol
Hepatocyte
L- glycerate(PH 2) LDH
Peroxisome Pyruvate
Alanine
AGT X
Glycolate GO
Glycine
HPR D-glycerate X Hydroxypyruvate
Glyoxylate GO
DAO
Oxalate
GR
Glycolate (PH 1)
X Glyoxylate LDH
Oxalate (PH 1 / PH 2)
Figure 1 | Oxalate metabolism in patients with primary hyperoxaluria types 1 and 2 and clinical hallmarks: severe urolithiasis and either medullary or diffuse nephrocalcinosis. AGT, alanine: glyoxylate aminotransferase; GO: glycolate reductase; HPR: hydroxypyruvate reductase. (Scheme modified from Leumann and Hoppe, with permission.1)
(2) Time spent on (hemo)dialysis should be kept to the absolute minimum. (3) Patients deserve a high priority on the waiting list for combined liver–kidney transplantation. (4) Early diagnosis by liver biopsy and/or mutational analysis as well as timely medical therapy is of paramount importance. Since PH1 became apparent as an entity, physicians have wondered about the remarkable clinical heterogeneity of the disease. Symptoms can develop in a small subset of patients (10%) as early as the neonatal period, and all such patients Kidney International (2006) 70
develop rapid ESRF within a few weeks to months. The vast majority of patients will present in childhood and reach ESRF between the second and fourth decades of life. Individuals who manifest late in life, an even smaller group (<10%), represent the reverse side of the PH1 spectrum. Some of them have true late-onset cases of PH1, whereas others have revealed signs and symptoms before but remained undiagnosed. This latter group often presents with advanced renal failure. Starting with the ground-breaking studies performed by Danpure and colleagues, our understanding of the genetic
and molecular mechanisms involved in PH1 has greatly evolved over the past 20 years.1 Today more than 60 mutations in the alanine:glyoxylate aminotransferase (AGXT) gene located on chromosome 2q37.3, resulting in subsequent absolute or functional enzyme deficiency, have been published. There is no mutational hot spot, but various mutations have been found throughout the 11 exons of the AGXT gene, which makes mutational analysis a sometimes laborious investigation. Many more are likely to be reported in the future. The G508A missense mutation (20%–30% allelic frequency) is most commonly found in PH1 and results in AGT mistargeting from the peroxisomes to the mitochondria. Another frequently (6%–9%) encountered missense mutation, I244T, leads to AGT misfolding, which produces functionally inactive aggregates.3 Not surprisingly, PH1 proved to be equally heterogeneous on the molecular level. So it seemed quite tempting to explain clinical heterogeneity by genetics. Keeping in mind the grim consequences of renal failure and the risks involved with liver transplantation, predicting the natural outcome for each patient would be a powerful tool, allowing sound counseling and a tailored medical management. For all the noble efforts by different investigators, so far all attempts to establish a general genotype–phenotype correlation have failed.4–6 There are several reasons why a genotype–phenotype correlation is so difficult or, as some experts believe, even impossible to prove. Besides the magnitude of mutations found, most individuals from non-consanguineous families are compound heterozygous, which makes a large group necessary to avoid skewing of the data. Unfortunately, such numbers are not available at the moment. In the available studies one can easily spot controversial findings where a certain genotype is associated with a favorable outcome in one but an unfavorable outcome in another study. Patients from consanguineous families often show private mutations not discovered in other individuals. Moreover, in some families, extreme intrafamilial phenotype variation exists, ranging from infantile oxalosis to a seemingly asymptomatic relative despite identical 985
co m m e nta r y
genotype.7,8 It should be mentioned, however, that two remarkable retrospective studies independently described an association between the G508A mutation and pyridoxine response, with a greater reduction of urinary oxalate excretion to near normal levels in homozygous patients, but a still partial response in heterozygous individuals.4,6 Pyridoxine treatment, the essential cofactor used by all aminotransferase reactions, has been reported for PH1 since the 1960s, long before AGT deficiency was discovered. Despite systematically evaluating pyridoxine response in all our patients, we, however, are unable to confirm these associations and still await the first patient to show complete response. That is what makes the contribution of Lorenzo et al.2 so important, as it should offer some guidance in all this discussion. If there is a genotype–phenotype correlation, it is likely to be found in the population described. Unlike most patients who are compound heterozygous bearing two different mutations, interestingly, more than 90% of PH1 among Canary Islands inhabitants is caused by the I244T missense mutation, with almost all patients being homozygous.3 To date, I244T constitutes the only PH1 mutation associated with a founder effect. Looking at the data of Lorenzo et al.2 we feel unhappy again; no uniform phenotype appears even in this small population characterized by genetic isolation. So is PH1 no longer considered a monogenic disease? For all we know, this is still the case; at least no other gene responsible has been identified. How-
986
ever, it is increasingly recognized that mendelian disorders are influenced by modifier genes that transform the phenotype into more complex traits. In order to uncover environmental factors and modifier genes, promising results came from animal studies of gastrointestinal oxalate transport, where secretory pathways for extrarenal oxalate elimination have been recently identified.9 Oxalobacter formigenes is an obligate anaerobic bacterium naturally colonizing the intestinal tract, which uses oxalic acid as its sole source of energy. Absence of O. formigenes colonization is observed in recurrent calcium oxalate stone formers. Hatch et al. were able to demonstrate that intestinal degradation of oxalate by O. formigenes in hyperoxaluric rats with renal insufficiency can induce excretion of oxalate from the blood to the intestinal lumen.9 Following that, we were able to show that treatment with O. formigenes in PH1 leads to a considerable reduction of urinary oxalate excretion or plasma oxalate levels in ESRF patients.10 Until recently the molecules in epithelial oxalate transport were unknown, but a new study by Jiang et al. shows that mice with knockout of Slc26a6 (an anion exchanger expressed in kidney and intestinal epithelial tissue) develop a phenotype that resembles PH1.11 Concluding our remarks, we strongly feel that a specific genotype–phenotype correlation is more than difficult to obtain in patients with PH1. Heterogeneous as this disease is, environmental factors and modifier genes may play a more important role than currently anticipated. In addi-
tion, with a better awareness of the disease and hence higher prevalence numbers, a more profound basis would be available to answer whether a true genotype–phenotype correlation exists. REFERENCES 1.
Leumann E, Hoppe B. The primary hyperoxalurias. J Am Soc Nephrol 2001; 12: 1986–1993. 2. Lorenzo V, Alvarez A, Torres A et al. Presentation and role of transplantation in adult patients with type 1 primary hyperoxaluria and the I244T AGXT mutation: single-center experience. Kidney Int 2006; 70: 1115–1119. 3. Santana A, Salido E, Torres A, Shapiro LJ. Primary hyperoxaluria type 1 in the Canary Islands: a conformational disease due to I244T mutation in the P11L containing alanine:glyoxylate aminotransferase. Proc Natl Acad Sci USA 2003; 10: 7277–7282. 4. Monico CG, Rosetti S, Olson JB, Milliner DS. Pyridoxine effect in type 1 primary hyperoxaluria is associated with the commonest mutant allele. Kidney Int 2005; 67: 1704–1709. 5. Pirulli D, Marangella M, Amorosa A. Primary hyperoxaluria: genotype-phenotype correlation. J Nephrol 2003; 16: 1–18. 6. Van Woerden CS, Jaap W, Groothoff F et al. Clinical implications of mutation analysis in primary hyperoxaluria type 1. Kidney Int 2004; 66: 746–750. 7. Frishberg Y, Rinat C, Shalata A et al. Intra-familial clinical heterogeneity: absence of genotypephenotype correlation in primary hyperoxaluria type 1 in Israel. Am J Nephrol 2005; 25: 269–275. 8. Hoppe B, Danpure CJ, Rumsby G et al. A vertical (pseudodominant) pattern of inheritance in the autosomal recessive disease primary hyperoxaluria type 1: lack of relationship between genotype, enzymatic phenotype and disease severity. Am J Kidney Dis 1997; 29: 36–44. 9. Hatch M, Cornelius J, Allison M et al. Oxalobacter sp. reduces urinary oxalate excretion by promoting enteric oxalate secretion. Kidney Int 2006; 69: 691–698. 10. Hoppe B, Beck B, Gatter N et al. Oxalobacter formigenes: a potential tool for the treatment of primary hyperoxaluria type 1. Kidney Int advance online publication, July 19 2006, doi:10.1038/ sj.ki.5001707. 11. Jiang Z, Asplin JR, Evan AP et al. Calcium oxalate urolithiasis in mice lacking anion Slc26a6. Nat Genet 2006; 38: 474–478.
Kidney International (2006) 70