Overactivity of phosphoribosylpyrophosphate synthetase (PRS) is an X chromosome–linked disorder of purine metabolism that is characterized by gout with uric acid overproduction and, in some families, neurodevelopmental impairment. We present the case of a 24-year-old Spanish woman with renal colic and hyperuricemia, which first manifested at age 11 years. Results of enzymatic and genetic studies supported the view that accelerated purine nucleotide and uric acid production in this woman resulted from defective allosteric regulation of PRS activity, which is, in turn, a consequence of a mutation in one of the patient's PRPS1 genes: an A-to-T substitution at nucleotide 578, encoding leucine for histidine at amino acid residue 192 of the mature PRS1 isoform. A previous example of disordered regulation of PRS1 activity in a family with a different substitution at the same amino acid residue strengthens this proposed mechanism. This is the first reported instance of PRS overactivity in which the propositus and sole affected family member is a woman.
Phosphoribosylpyrophosphate (PRPP) is an important regulatory substrate in the synthesis of purine, pyrimidine, and pyridine nucleotides (1). PRPP synthesis from Mg-ATP and ribose-5-phosphate is catalyzed in mammalian cells by a family of PRPP synthetase (PRS; EC 22.214.171.124) isoforms in reactions that require Mg2+ and inorganic phosphate (Pi) as activators and are subject to inhibition by purine nucleotides (1). PRPS genes encoding the 3 highly homologous mammalian PRS isoforms (PRS1, PRS2, and PRS3, respectively) identified to date are differentially expressed: PRPS1 and PRPS2 are expressed in all tissues (2) and map to the long and short arms of the X chromosome, respectively (3). Expression of the autosomal PRPS3 gene is restricted to the testes (2).
Superactivity of PRS is an X chromosome–linked inborn error of metabolism (4) characterized by increased rates of PRPP, purine nucleotide, and uric acid production, in association with gout and uric acid urolithiasis. In some families, however, neurodevelopmental impairment, particularly sensorineural deafness (5), accompanies the metabolic abnormalities of the disorder. The differing clinical expressions of inherited PRS superactivity are typically paralleled by different kinetic alterations in the enzyme. The major kinetic alterations in PRS superactivity include regulatory defects, in which allosteric control of PRS activity by purine nucleotides and Pi is impaired (5, 6) and “catalytic defects,” characterized by increased maximum reaction velocity but normal substrate, inhibitor, and activator responsiveness (7). Families with defects in the allosteric regulation of PRS activity usually show infantile or early childhood disease onset in affected males, clinical expression in heterozygous female carriers, and more severe biochemical and clinical phenotypes (5).
Point mutations in the translated region of the PRPS1 gene have been found in patients with regulatory defects (6). In contrast, PRS “catalytic overactivity” results from an increased concentration of the normal PRS1 isoform (7) as a result of selective acceleration of PRPS1 transcription (8). Among the nearly two dozen affected families identified to date, 6 different single-base substitutions in the translated region of PRPS1 have been identified in 6 unrelated male patients with allosterically altered PRS1 (6). We present herein the case of a female patient in whom PRS overactivity resulted from a mutation in PRPS1 that led to an amino acid substitution in 1 of her 2 PRPS1 gene products. In this heterozygous woman, the clinical consequences of the genetic defect arose from dominant expression of allosteric dysregulation of PRPP production.
The patient, a 24-year-old woman, was in good health until age 11 years, when she experienced right renal colic and passed sandy red–tinged urine. Her blood pressure was normal, but her serum urate level was 10 mg/dl, with a 24-hour urinary uric acid level of 1,300 mg. Results of intravenous pyelography and bone radiography were normal. Recurrence of symptoms 5 months later prompted hospital admission. Her serum creatinine level at that time was 2.6 mg/dl, and her blood urea nitrogen level was 108 mg/dl. She was treated with intravenous hydration and bicarbonate, and her serum creatinine level returned to normal (0.6 mg/dl). Nevertheless, the serum urate and urinary uric acid levels remained in the range of 10 mg/dl and 1,300–1,400 mg/day (30 mg/kg/day), respectively. Upon treatment with allopurinol (200 mg/day), the serum urate level decreased to 4.9 mg/dl and the 24-hour urinary uric acid level to 352 mg.
The patient was referred (at age 11 years) to La Paz University Hospital for further study. She denied having articular symptoms, and on examination, there were no joint deformities, inflammation, or tophi. Results of neurologic examination, including hearing tests, were normal. Routine hematologic and biochemical tests also yielded normal results. The urinary sediment was sandy red and contained abundant uric acid crystals. The urinary calcium level was low (47 mg/24 hours), but serum calcium and parathyroid hormone levels were normal. Intelligence quotient estimations ranged between 85 and 90. Findings of audiometry, electroencephalography, and computed tomography of the head were normal. Renal echography showed normal kidneys, with no evidence of lithiasis or impaired urinary flow.
Serum urate and 24-hour urinary uric acid determinations in her parents, 2 brothers, and 1 sister yielded normal results. Activities of hypoxanthine guanine phosphoribosyltransferase (HPRT) and adenine phosphoribosyltransferase in erythrocyte lysates from the patient, her mother, and a brother were normal. The dynamics of the adenine nucleotide pool were examined in these 3 individuals, as described below.
Although suspected, PRS overactivity could not be confirmed by enzymatic and purine metabolic analyses at the time of the initial clinical presentation. Over the ensuing 13 years, however, the patient has been in good health, and her serum uric acid levels have remained normal during treatment with allopurinol. At the time of the most recent investigation, she was receiving 100 mg of allopurinol daily. After the patient expressed her wish to have children, genetic and additional enzymatic studies were performed to confirm the diagnosis of PRS overactivity.
MATERIALS AND METHODS
All studies were approved by the Institutional Research and Ethics Review Committees of La Paz University Hospital. As appropriate to the patient's age, informed consent was obtained from the patient and/or her parents and from her 23-year-old brother.
The patient, her mother, and a brother were admitted to the Clinical Research Unit of La Paz University Hospital. Purine metabolism was studied after all medications had been discontinued for 2 weeks. The subjects were started on a weight-maintenance, isocaloric, purine-restricted diet (<75 mg/day of purines, with a protein content of 10–15%) for 7 days before the studies. Purine metabolism was examined during a 5-day period by measuring urinary radioactivity (following the infusion of 25 μCi of 8-14C-adenine to radiolabel the adenine nucleotide pool) and by determining plasma and 24-hour urinary creatinine, hypoxanthine, xanthine, and uric acid concentrations (9). An increased rate of urinary excretion of radioactivity after 8-14C-adenine infusion strongly supports the presence of enhanced purine nucleotide degradation and, in the steady state, purine nucleotide overproduction (9). On at least 3 occasions, venous blood was obtained from the subjects (fasting for 10 hours before blood draw) at the end of 24-hour urine collections. Urine samples were collected daily (with 3 ml of toluene as preservative) for determination of creatinine and purine concentrations and radioactivity. For each subject, data presented relevant to purines are the means of at least 3 plasma and 5 urine determinations.
Aliquots of venous blood samples from the patient and her mother were processed for isolation of leukocyte genomic DNA. Washed red blood cell lysates were also prepared and stored at −20°C until PRS activity determinations, as previously described (10). Fibroblasts from the patient and her mother were cultured from skin biopsy specimens, as described elsewhere (11). RNA was extracted from cultured fibroblasts with a commercial kit (RNeasy Minikit; Qiagen, Hilden, Germany). Rates of de novo purine synthesis and inhibition of this process in response to the addition of the purine bases adenine, hypoxanthine, and guanine were measured in intact fibroblasts, as described elsewhere (11). PRS activity was measured in activated charcoal–treated erythrocyte lysates by a nonisotopic high-performance liquid chromatography method (10) and in dialyzed fibroblast extracts by a 2-step isotopic method (11). PRS1 and PRS2 complementary DNA (cDNA) were prepared by reverse transcription–polymerase chain reaction (RT-PCR) amplification of PRS1 and PRS2 transcripts using the oligonucleotides and conditions previously described (6).
PRPS1 genomic DNA, encompassing exon 5, was amplified using primers F-EXON5PRS1 (5′-AAGTGCTGGGATTACAGGCGTG-3′) and B9-EXON5PRS1 (5′-CTAACTACCAGCCCCATCAATCC-3′). Approximately 100 ng of genomic DNA was used for PCR. PCR conditions were 30 cycles at 94°C for 1 minute, 64°C for 1 minute, and 72°C for 1 minute, with a final extension at 72°C for 7 minutes. Amplified PRS1 and PRS2 cDNA and PRPS1 exon 5 were sequenced using an automated DNA sequencer (ABI Prism model 377; Applied Biosystems, Foster City, CA).
Renal function was assessed by measurement of serum creatinine and creatinine clearance, and the findings were normal in the patient, her mother, and her brother (Table 1). All parameters of purine metabolism in the patient's mother and brother were normal compared with those of premenopausal women and male subjects, respectively. In contrast, marked purine nucleotide and uric acid overproduction were evident in the propositus (Table 1). The patient's plasma urate levels and daily uric acid excretion were markedly increased, and the ratio of uric acid to creatinine in urine was nearly 1. Daily urinary excretion of hypoxanthine was also increased. The patient's excretion of urinary radioactivity in the 5 days after administration of 14C-labeled adenine was ∼1.7 times the mean rate determined in healthy control subjects (12).
Table 1. Purine metabolism in the patient, her mother, and a brother, as compared with that in controls*
Data for the controls (mean ± SD) were obtained from references9 and12. HPRT = hypoxanthine guanine phosphoribosyltransferase; Hgb = hemoglobin; APRT = adenine phosphoribosyltransferase.
HPRT activity, nmoles/hour/mg of Hgb
87.0 ± 16.5
APRT activity, nmoles/hour/mg of Hgb
34.8 ± 3.2
0.9 ± 0.2
2.3 ± 1.1
0.7 ± 0.2
4.9 ± 1.0
Creatinine, mg/24 hours/1.73 m2
1,372 ± 352
Uric acid, mg/24 hours/1.73 m2
442 ± 86
Uric acid:creatinine ratio, mg/mg
0.34 ± 0.11
Hypoxanthine, μmoles/gm of creatinine
35 ± 13
Xanthine, μmoles/gm of creatinine
2,439 ± 583
Total purines, μmoles/gm of creatinine
32 ± 13
Creatinine clearance, ml/minute/m2
99 ± 12
14C excretion, % of administered dose excreted in 5 days
5.0 ± 1.3
Mean maximum PRS activities (measured at 32 mM Pi) in dialyzed fibroblast extracts and in charcoal-treated hemolysates derived from the patient were each only 23% of the mean maximum values established for the corresponding cell-free preparations derived from normal individuals. Allosteric regulation of the PRS activity in the patient was next evaluated by measurement of the responsiveness of the enzyme to Pi activation and purine nucleoside diphosphate (ADP and GDP) inhibition. When activation of fibroblast or hemolysate PRS activities by Pi was examined, PRS activities at the lowest Pi concentrations were substantially greater in cell extracts derived from the patient than those derived from normal subjects. That is, activation of the patient's PRS by Pi occurred at Pi concentrations substantially lower than those required for activation of normal PRS (Figures 1A and B).
In contrast, resistance of the patient's PRS activity to inhibition by ADP or GDP could not be demonstrated by direct enzyme assay in either charcoal-treated hemolysates or Sephadex G-50–chromatographed fibroblast extracts (data not shown). This dissociation in the previously established criteria for dysregulation of allosteric function of PRS (6, 11) was, however, reconciled by the demonstration that rates of de novo purine synthesis in intact fibroblasts from the patient were resistant to the addition of the purine bases adenine (Figure 1C), as well as hypoxanthine and guanine (data not shown) (11). PRS activities and allosteric regulation of PRS in extracts of cells derived from the patient's mother were entirely normal (data not shown).
The complete sequence of the patient's PRS1 cDNA was studied on multiple occasions and on both sense and antisense cDNA strands. Uniquely, and in all instances, dual peaks were observed in the area corresponding to nucleotide 578 of the translated sequence. Peaks of both adenine and thymine were observed at this position in each sequencing run and in both DNA orientations. For all other nucleotide residues of the patient's PRS1 and PRS2 cDNA and her mother's PRS1 and PRS2 cDNA, single peaks corresponding to those confirming the respective normal PRS cDNA sequences (13, 14) were observed.
In order to confirm these findings, genomic DNAs were isolated from the patient and her mother, and these served as templates for PCR amplification of the respective PRPS1 exons 5. Sequencing of the resulting DNAs confirmed dual peaks (adenine and thymine) at nucleotide 578 in the patient's PRS1 exon 5 sequence and a single peak at all residues in her mother's PRS1 exon 5 sequence (Figure 2). Substitution of adenine by thymine at PRS1 cDNA nucleotide 578 predicts a corresponding substitution of leucine for histidine at PRS1 amino acid residue 192.
Based on the results of these studies we propose that accelerated purine nucleotide synthesis and uric acid overproduction in our patient resulted from PRS superactivity due to a point mutation in PRPS1, a T-for-A substitution at nucleotide 578 in the PRPS1 coding region. This mutation has 2 distinct functional effects on the activity of the PRS1 isoform thus encoded. First, as is the case in most of the point mutations encountered to date in affected male patients (6), the mutant PRS1 in our patient appears to be more labile than its normal counterpart, resulting in reduced concentrations and activities of the mutant isoform in the patient's cells and accounting for the subnormal maximum PRS activities measured in her cell extracts at maximally activating Pi concentrations. Second, the mutation in the patient's PRS1 imparts altered allosteric regulatory properties on the activity of the isoform. These altered properties include increased responsiveness of enzyme activity to activation by Pi (directly demonstrated in cell extracts at the lowest Pi concentrations) and impaired responsiveness to purine nucleotide inhibition (indirectly shown by blunted inhibition of rates of de novo purine synthesis in intact cells incubated with purine base precursors of nucleotide inhibitors). These kinetic defects are the hallmarks of regulatory defects that result in PRS superactivity (6, 11).
On first consideration, our failure to detect resistance of the patient's PRS activity to ADP or GDP by direct assay in cell extracts seems paradoxical, especially since, as predicted in our previous studies of human PRS1 mutations (6) and confirmed by structural analysis of Bacillus subtilis PRS (15), Pi and purine nucleotide inhibitors bind to an identical regulatory site on PRS and function antagonistically at this site. We believe that the explanation for the apparently normal nucleotide inhibition response of the patient's PRS resides in the combination of enhanced lability of the patient's mutant PRS1 and the considerable residual PRS1 encoded by the normal PRPS1 allele in extracts of cells from this patient, a heterozygous female with respect to the PRPS1 gene.
Mutations in PRPS1 that affect amino acid residues spanning nearly half of the PRS1 polypeptide (from residue 51 to residue 192) give rise to disordered allosteric regulation of PRS1 (6). Our group has previously predicted (6), on this basis, that the PRPS1 mutations described to date involve residues conveying allosteric signals to the regulatory site of PRS1 rather than residues comprising the allosteric site itself. This prediction is supported by the structural analysis of B subtilis PRS (15), which shares with human PRSs nearly complete identity at the residues comprising the proposed allosteric regulatory site. The base substitution at nucleotide 578 in the patient's PRPS1 predicts a mutation with similar effects on allosteric control of PRS1, because nucleotides 578 and 579 are substituents of the same codon specifying amino acid 192 of PRS1 and because a G-for-C transversion at nucleotide 579, predicting a substitution of glutamine for histidine at amino acid 192 (6), was shown to result in allosteric dysregulation of PRS1 in a second affected Spanish family (5).
Our patient was informed that she is an affected heterozygous carrier for PRS overactivity and that the phenotype of a male patient with a mutation also affecting histidine 192 (6) includes tophaceous gout, purine nucleotide and uric acid overproduction, mental retardation, and sensorineural deafness (5). Prenatal diagnosis has been offered to this patient in the event that she becomes pregnant.
We believe that the possibility of PRS overactivity should be considered and confirmed or ruled out in any young patient whose gout and/or purine nucleotide and uric acid overproduction is not otherwise explained, as, for example, by HPRT deficiency. Identification of hemizygous affected males or heterozygous female carriers for PRS overactivity should provide the basis for more accurate genetic counseling.