|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||
Background Hyperphenylalaninemia is a common inherited metabolic disease that is due to phenylalanine hydroxylase deficiency, and at least half the affected patients have mild clinical phenotypes. Treatment with a low-phenylalanine diet represents a substantial psychosocial burden, but alternative treatments have not been effective.
Methods To explore the therapeutic efficacy of tetrahydrobiopterin, we performed a combined phenylalanine–tetrahydrobiopterin loading test and analyzed the in vivo rates of [13C]phenylalanine oxidation in 38 children with phenylalanine hydroxylase deficiency (age range, 1 day to 17 years). We assessed whether responsiveness to tetrahydrobiopterin was associated with specific genotypes, and we mapped mutations using a structural model of the phenylalanine hydroxylase monomer.
Results In 27 (87 percent) of 31 patients with mild hyperphenylalaninemia (10 patients) or mild phenylketonuria (21 patients), tetrahydrobiopterin significantly lowered blood phenylalanine levels. Phenylalanine oxidation was significantly enhanced in 23 of these 31 patients (74 percent). Conversely, none of the seven patients with classic phenylketonuria had a response to tetrahydrobiopterin as defined in this study. Long-term treatment with tetrahydrobiopterin in five children increased daily phenylalanine tolerance, allowing them to discontinue their restricted diets. Seven mutations (P314S, Y417H, V177M, V245A, A300S, E390G, and IVS4–5C
Conclusions Tetrahydrobiopterin responsiveness is common in patients with mild hyperphenylalaninemia phenotypes. Responsiveness cannot consistently be predicted on the basis of genotype, particularly in compound heterozygotes.
G) were classified as probably associated with responsiveness to tetrahydrobiopterin, and six mutations (A403V, F39L, D415N, S310Y, R158Q, and I65T) were classified as potentially associated. Four mutations (Y414C, L48S, R261Q, and I65V) were inconsistently associated with this phenotype. Mutations connected to tetrahydrobiopterin responsiveness were predominantly in the catalytic domain of the protein and were not directly involved in cofactor binding.
In approximately 50 genetic diseases of humans involving enzyme deficiencies, treatment with high doses of a cofactor can increase enzyme activity.5 Tetrahydrobiopterin is a natural cofactor of aromatic amino acid hydroxylases and nitric oxide synthase. Supplementation with this compound is an established treatment for the rare patients with hyperphenylalaninemia that is due to defects in the biosynthesis of tetrahydrobiopterin.6,7 However, more than 98 percent of patients with hyperphenylalaninemia have mutations in the phenylalanine hydroxylase gene, and they have elevated rather than decreased plasma concentrations of biopterin owing to the action of guanosine triphosphate cyclohydrolase I feedback regulatory protein.8 The therapeutic use of tetrahydrobiopterin in patients with phenylalanine hydroxylase deficiency had therefore not been considered.
Recently, however, individual patients with mutations in the phenylalanine hydroxylase gene have been shown to have a decrease in blood phenylalanine concentrations after tetrahydrobiopterin loading.9,10,11,12,13 Phenylalanine concentrations in the peripheral circulation, however, are governed by various genetic loci and modifying factors,1,14 and there is no evidence that the beneficial effect of tetrahydrobiopterin occurs at the level of phenylalanine hydroxylation. Therefore, we prospectively studied 38 children with phenylalanine hydroxylase deficiency in an effort to determine the frequency of sensitivity to tetrahydrobiopterin in these patients, whether tetrahydrobiopterin restores their oxidative capacity for phenylalanine, whether responsiveness to tetrahydrobiopterin is related to specific genotypes, and whether associated mutations map to distinct regions of the protein.
Methods
Patients
The study was conducted from December 2000 through December 2001 and was approved by the medical board of the Children's Hospital Research Center. We obtained written informed consent from the families of 38 children with various classes of hyperphenylalaninemia stratified according to the plasma phenylalanine concentration before treatment (normal, 30 to 120 µmol per liter): 10 patients had mild hyperphenylalaninemia (phenylalanine, less than 600 µmol per liter; age, 15 days to 10 years), 21 had mild phenylketonuria (phenylalanine, 600 to 1200 µmol per liter; age, 8 days to 17 years), and 7 had classic phenylketonuria (phenylalanine, more than 1200 µmol per liter; age, 1 day to 9 years). A defect in the synthesis or recycling of tetrahydrobiopterin was excluded by analysis of urinary pterins and dihydropteridine reductase activity in erythrocytes. We analyzed 7 patients during the newborn period and 31 at older ages. Five affected siblings from four families were included, because nongenetic factors are known to influence phenylalanine homeostasis. Mean daily phenylalanine tolerance was determined by calculating the dietary phenylalanine intake according to nutritional protocols.
Combined Phenylalanine and Tetrahydrobiopterin Loading Test
Phenylalanine loading was accomplished by having patients consume a meal containing 100 mg of phenylalanine per kilogram of body weight. One hour after the end of the meal the patients ingested 20 mg of tetrahydrobiopterin per kilogram (Schircks Laboratories). Blood phenylalanine concentrations were determined by electrospray ionization–tandem mass spectrometry before phenylalanine loading and before and 4, 8, and 15 hours after the tetrahydrobiopterin challenge. During the test period newborns were breast-fed, while older children received a standardized protein intake (10 mg of phenylalanine per kilogram) between six and eight hours after the challenge with tetrahydrobiopterin.
In Vivo Analysis of Phenylalanine Oxidation
The rate of phenylalanine oxidation was determined twice (on two different days) in each child — once without treatment and once during treatment with tetrahydrobiopterin (10 mg per kilogram over a 24-hour period). The tests were performed after a four-hour fast in infants and an overnight fast in older children. A total of 6 mg per kilogram of L-[1-13C]phenylalanine (Euriostop), dissolved in a 25 percent dextrose solution (2 mg per milliliter), was given orally. Breath samples were subsequently collected over a period of 180 minutes and stored in evacuated glass tubes until analysis by isotope-ratio mass spectrometry (deltaS, Thermoquest). The recovery of carbon-13 in breath samples was calculated as described by Treacy et al.,15 assuming a total carbon dioxide production of 300 mmol per hour per square meter of body-surface area.16,17 The amount of labeled carbon dioxide formed was expressed as the cumulative percentage of the dose administered as a function of time. The validity of results in newborns might be influenced by the diet or by the fact that breath sampling is more challenging than in older subjects. The base-line percentage of carbon-13 measured at time 0, however, did not differ significantly between newborns and older children. Values were considered to be below the limit of detection when the signal intensity of the atom excess (expressed as a percentage at time t, obtained by subtraction of the mean base-line value) did not allow sufficient distinction from atmospheric carbon-13 dioxide. On average, fewer than 1 of 27 consecutive measurements of carbon dioxide obtained during the 180 minutes of individual testing was uninterpretable in older children and fewer than 2 of 27 were uninterpretable in newborns, and these variations had a negligible influence on the final calculation. For the comparison among patients, we normalized the data by expressing individual results as a percentage of the mean value for the control group, which consisted of 12 healthy children (age, 2 days to 13 years).
Mutational Analysis
DNA was extracted from leukocytes according to standard protocols. Thirteen genomic fragments covering the entire coding sequence and the exon-flanking intronic sequences of the phenylalanine hydroxylase gene were amplified by the polymerase chain reaction followed by direct sequencing.18
Mapping of Phenylalanine Hydroxylase Gene Mutations
We constructed a model of the full-length, tetrahydrobiopterin-bound phenylalanine hydroxylase monomer from the crystal structures of several truncated forms19,20,21,22 by superimposing the catalytic domains using the tools provided by SWISS-MODEL/Swiss-Pdb Viewer.23
Results
Effects of Tetrahydrobiopterin on Blood Phenylalanine Levels and Rates of Phenylalanine Oxidation
Patients were classified as responsive to tetrahydrobiopterin when blood phenylalanine levels 15 hours after tetrahydrobiopterin challenge had decreased by more than 30 percent from the value obtained before the administration of tetrahydrobiopterin. An improvement in the rate of phenylalanine oxidation was considered to be significant when supplementation with tetrahydrobiopterin increased the individual normalized value by at least 15 percent. Tetrahydrobiopterin sensitivity was observed during the loading test in all 10 patients with mild hyperphenylalaninemia and in 17 of 21 patients with mild phenylketonuria (27 of 31, or 87 percent). Only four patients with mild phenylketonuria and all seven patients with classic phenylketonuria did not fulfill the criterion of responsiveness to tetrahydrobiopterin (Figure 1). Some patients had a rapid decrease in phenylalanine resembling that seen in patients with defects in the synthesis of tetrahydrobiopterin, whereas others had a slow response, which reached a maximum 15 hours after the administration of the cofactor (data not shown).
|
|
|
Values for individual patients and examples of notable imbalances in the effect of tetrahydrobiopterin are shown in Figure 4. In four patients (Patients 2, 4, 7, and 26) who were responsive to tetrahydrobiopterin, the normalized increase in phenylalanine oxidation was in the range of 8 to 14 percent (data not shown) and therefore below our predefined level of significance. One patient with classic phenylketonuria (Patient 35) had a slight decrease in the blood phenylalanine concentration that did not fulfill the criterion of responsiveness, whereas the increase in the oxidation rate (24 percent) was significant. Notably, 7 of the 11 patients who did not meet the criterion of responsiveness in the loading test had a slight increase in the rate of phenylalanine oxidation (range, 2 to 14 percent) with short-term treatment with tetrahydrobiopterin.
|
The parents of five children with mild phenylketonuria (age, 4 to 14 years) provided written informed consent for their children to participate in a therapeutic trial replacing dietary phenylalanine restriction with the oral administration of tetrahydrobiopterin. Cofactor treatment at daily doses of 7.1 to 10.7 mg per kilogram for a mean of 207.0±51.3 days (range, 166 to 263) led to an increase in the mean daily phenylalanine tolerance, from 18.7±8.6 mg per kilogram (range, 8.5 to 30.0) before treatment to 61.4±27.9 mg per kilogram (range, 17.9 to 90.0) during treatment (P=0.04), with little effect on the mean blood concentrations of phenylalanine (366±120 µmol per liter during dietary treatment and 378±173 µmol per liter during cofactor treatment).
Identification and Mapping of Phenylalanine Hydroxylase Gene Mutations
In 37 of 38 patients, two mutant alleles were identified (Table 1). We classified seven mutations (P314S, Y417H, V177M, V245A, A300S, E390G, and IVS4–5CG) as probably responsible for responsiveness to tetrahydrobiopterin, because they were present in either the homozygous or a functional hemizygous state. Six additional mutations were potentially connected to tetrahydrobiopterin responsiveness because of considerable residual in vitro enzyme activity (A403V, F39L, D415N, R158Q, and I65T), as previously proposed,24 or (in the case of S310Y) because of a known severe mutation on the second allele. Four mutations (Y414C, L48S, R261Q, and I65V) were inconsistently associated with tetrahydrobiopterin responsiveness. Eight of 12 missense mutations connected to tetrahydrobiopterin responsiveness mapped to the catalytic domain, whereas 2 mapped to the regulatory domain and 2 to the tetramerization domain. None of them affected residues at the active site or amino acids that interacted directly with the cofactor (Figure 5).
|
|
We present two lines of evidence that the metabolic phenotype of phenylalanine hydroxylase deficiency can be modified by pharmacologic doses of tetrahydrobiopterin. First, tetrahydrobiopterin loading led to normal or nearly normal blood phenylalanine concentrations in most patients with residual phenylalanine hydroxylase activity, suggesting that responsiveness to tetrahydrobiopterin is a common feature of mild hyperphenylalaninemia phenotypes. Second, tetrahydrobiopterin enhanced residual phenylalanine oxidative capacity in these patient groups.
Our findings suggest that the in vivo phenylalanine oxidation test can discriminate among classes of hyperphenylalaninemia of different severity. This observation is in accordance with data on the ability of the method to measure the dose effects of the phenylalanine hydroxylase gene.15 However, because of the multifactorial nature of hyperphenylalaninemia,25,26 the whole-body rate of phenylalanine oxidation is not a simple equivalent of phenylalanine hydroxylase activity. The decrease in blood phenylalanine concentrations was accompanied by a significant increase in phenylalanine oxidative capacity in the majority of patients who were identified as responsive to tetrahydrobiopterin. Taken together, these observations are consistent with the hypothesis that impaired phenylalanine hydroxylation is corrected by tetrahydrobiopterin therapy.
The extent of the fractional change in the disposal of phenylalanine did not always correspond to the change in phenylalanine oxidation — a finding not unexpected with respect to genetically determined enzyme deficiencies in general27 and phenylalanine hydroxylase deficiency in particular.26 We observed slow and rapid responses as well as differences in the time course and relative extent of formation of labeled carbon dioxide, suggesting that tetrahydrobiopterin may exert its effects through various mechanisms and with different degrees of efficacy. In addition to the proposal that high-dose tetrahydrobiopterin treatment may compensate for the decreased affinity of the mutant phenylalanine hydroxylase for tetrahydrobiopterin,28 other mechanisms need to be considered. Tetrahydrobiopterin treatment may up-regulate the expression of the phenylalanine hydroxylase gene,24 stabilize phenylalanine hydroxylase messenger RNA,29 facilitate the formation of functional phenylalanine hydroxylase tetramers, or protect a misfolded enzyme protein from proteolytic cleavage.30,31
The use of genotyping to predict the phenotype may present difficulties in the case of complex traits such as hyperphenylalaninemia,32 particularly in compound heterozygotes. We identified predominantly "mild" genotypes in the group of patients with a response to tetrahydrobiopterin, whereas most of the patients without a response had "severe" genotypes.2 The weight of the evidence of the association of distinct mutations with responsiveness to tetrahydrobiopterin varied. The Y414C mutation occurs in more than one clinical phenotype.33,34 We and others12 have identified this mutation in a functional hemizygous state in two patients with identical genotypes but discordant responses to tetrahydrobiopterin. This observation may be explained by the influences of modifying loci in hyperphenylalaninemia, since this trait is polygenic.1 In a homozygous state, and thus one in which homopolymeric tetramers are formed, the Y414C and the L48S mutations were reported to confer responsiveness to tetrahydrobiopterin.24,35 However, we detected these mutations in a functional hemizygous state in patients with classic phenylketonuria who had no response to tetrahydrobiopterin. Under these conditions, heteropolymerization may impede the formation of functional tetramers.
Our data confirm that most missense mutations associated with sensitivity to tetrahydrobiopterin are in the catalytic domain of the protein, but they do not map to residues at the active site and are not directly involved in cofactor binding.28 These mutations may affect interactions between domains in a monomer or influence residues in the dimer or tetramer interfaces,36 resulting in the misfolding of the protein and reduced enzyme activity. Tetrahydrobiopterin may act as a chemical chaperone and thus prevent misfolding.
In vitro expression analysis has been used to predict the functional effect in vivo of mutations in the phenylalanine hydroxylase gene.37,38,39 This type of analysis may result in the overestimation of phenylalanine hydroxylase activity in vitro,38 perhaps because such analyses have been carried out almost exclusively in the presence of high concentrations of natural or synthetic cofactors,2 thereby contributing to genotype–phenotype inconsistencies.33 Revised experimental protocols to assess the intrinsic severity of mutations should include a range of tetrahydrobiopterin concentrations.40
Since responsiveness to tetrahydrobiopterin cannot be predicted on the basis of pretreatment phenylalanine concentrations, we would suggest a new clinical classification: tetrahydrobiopterin-unresponsive hyperphenylalaninemia and tetrahydrobiopterin-responsive hyperphenylalaninemia, which includes tetrahydrobiopterin-responsive phenylalanine hydroxylase deficiency and defects in the synthesis of tetrahydrobiopterin. A phenylalanine–tetrahydrobiopterin loading test with an extended observation period (at least 15 hours) can reliably and safely discriminate between patients with a response and patients without a response and ought to be performed in all patients with hyperphenylalaninemia to identify those who may benefit from tetrahydrobiopterin treatment. Our short-term study design does not exclude the possibility of subtle effects, which may become evident only after prolonged treatment even in some patients with classic phenylketonuria.
Our data suggest that long-term therapy with tetrahydrobiopterin could lead to an increase in phenylalanine tolerance. Cofactor treatment instead of a phenylalanine-restricted diet might be possible in many patients and would be expected to improve their quality of life substantially. Tetrahydrobiopterin treatment may also be helpful in cases of maternal phenylketonuria, since metabolic control, which is key to the prevention of serious adverse effects in the offspring,41 is difficult to maintain during pregnancy. However, the safety of tetrahydrobiopterin therapy during pregnancy has not been established. Worldwide, more than 350 patients with tetrahydrobiopterin deficiency have been treated with the cofactor.42 Some dose-dependent adverse reactions, including sleep disorders, polyuria, and loose stools, were reported in a safety evaluation.43
Several obstacles must be overcome before tetrahydrobiopterin treatment can be used routinely. First, tetrahydrobiopterin has not yet been approved for therapeutic use in most countries. Second, this compound is expensive. Third, dose-finding studies and clinical trials are needed to determine the bioavailability and long-term effects of tetrahydrobiopterin therapy in patients with phenylalanine hydroxylase deficiency.
In conclusion, we found that pharmacologic doses of tetrahydrobiopterin corrected impaired phenylalanine oxidation in the majority of patients with mild hyperphenylalaninemia phenotypes. Our findings have implications for the diagnostic workup and clinical classification of this defect as well as for therapeutic interventions. In the near future, in a large number of patients with hyperphenylalaninemia, cofactor treatment may obviate the need for the most burdensome dietary restrictions.
Supported by intramural funds from the Children's Research Center, University of Munich, Dr. von Hauner Children's Hospital.
We are indebted to the patients for their participation; to Florian Lagler, Esther Maier, Stephan Hiedl, Kathrin Vosberg, and Wolfgang Sperl for clinical care; to Katharina Dokoupil for nutritional advice; to Markus Phirrmann for assistance with the statistical analysis; to Berthold Koletzko and Dietrich Reinhardt for continuous support; to Nenad Blau for the measurement of pterins and the dihydropteridine reductase enzyme analysis; to Ralph Fingerhut and Bernhard Olgemöller for the measurement of phenylalanine; to Claudia Nevinny-Stickel and Ekkehard Albert for genotyping; to Barbara Paton, Debbie Stern, and Tricia LeVan for careful review of the manuscript; and to Alf Heger for his expert in silico skills. This article is part of an M.D. thesis to be submitted by Mr. Habich.
Source Information
From the Research Center, Department of Biochemical Genetics and Molecular Biology (A.C.M., W.R., M.H., A.A.R.), and the Department of Metabolic Disorders and Nutrition (H.D., B.H.), Dr. von Hauner Children's Hospital; and the Department of Clinical Chemistry and Clinical Biochemistry, Surgical Clinic (C.P.S.) — all at Ludwig Maximilians University, Munich, Germany.
Address reprint requests to Dr. Roscher at Dr. von Hauner Children's Hospital Research Center, Ludwig Maximilians University, Lindwurmstrasse 2a, D-80337 Munich, Germany, or at adelbert.roscher{at}kk-i.med.uni-muenchen.de.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Tetrahydrobiopterin and Mild Phenylketonuria
Ponzone A., Peduto A., Spada M., Hanley W. B., D'agostino R. D., Roscher A. A., Muntau A. C., Röschinger W.
Extract |
Full Text |
PDF
N Engl J Med 2003;
348:1722-1724, Apr 24, 2003.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
