Accelerated Transcription of PRPS1 in X-linked Overactivity of Normal Human Phosphoribosylpyrophosphate Synthetase*

Phosphoribosylpyrophosphate (PRPP) synthetase (PRS) superactivity is an X-linked disorder characterized by gout with overproduction of purine nucleotides and uric acid. Study of the two X-linked PRS isoforms (PRS1 and PRS2) in cells from certain affected individuals has shown selectively increased concentrations of structurally normal PRS1 transcript and isoform, suggesting that this form of the disorder involves pretranslational dysregulation ofPRPS1 expression and might be more appropriately termed overactivity of normal PRS. We applied Southern and Northern blot analyses and slot blotting of nuclear runoffs to delineate the process underlying aberrant PRPS1 expression in fibroblasts and lymphoblasts from patients with overactivity of normal PRS. NeitherPRPS1 amplification nor altered stability or processing of PRS1 mRNA was identified, but PRPS1 transcription was increased relative to GAPDH (3- to 4-fold normal in fibroblasts; 1.9- to 2.4-fold in lymphoblasts) and PRPS2. Nearly coordinate relative increases in each process mediating transfer of genetic information from PRPS1 transcription to maximal PRS1 isoform expression in patient fibroblasts further supported the idea that accelerated PRPS1 transcription is the major aberration leading to PRS1 overexpression. In addition, modulated relative increases in PRS activities at suboptimal Piconcentration and in rates of PRPP and purine nucleotide synthesis in intact patient fibroblasts indicate that despite an intact allosteric mechanism of regulation of PRS activity, PRPS1transcription is a major determinant of PRPP and purine synthesis. The genetic basis of disordered PRPS1 transcription remains unresolved; normal- and patient-derived PRPS1s share nucleotide sequence identity at least 850 base pairs 5′ to the consensus transcription initiation site.

Phosphoribosylpyrophosphate (PRPP) 1 is a substrate in the synthesis of virtually all nucleotides (1) as well as an important regulator of rates of the de novo pathways of purine and pyrimidine nucleotide synthesis (2)(3)(4). PRPP synthesis from MgATP and ribose-5-phosphate is catalyzed in mammalian cells by a family of PRPP synthetase (PRS; EC2.7.6.1) isoforms in reactions requiring Mg 2ϩ and P i as activators and subject to inhibition by purine, pyrimidine, and pyridine nucleotides (5)(6)(7)(8). Of the three highly homologous human PRS isoforms identified to date, PRS1 and PRS2 are expressed in all tissues (9,10) and are encoded by genes (PRPS1 and PRPS2) that map, respectively, to the long and the short arms of the X chromosome (11,12). PRS3 expression is detectable only in the testes and is encoded autosomally (9,13).
Superactivity of PRS is an X chromosome-linked human disorder (14) characterized by PRPP, purine nucleotide, and uric acid overproduction (15)(16)(17), gout (15,18), and, in some affected families, neurodevelopmental impairment (18 -21). The kinetic mechanisms underlying inherited PRS superactivity are diverse and include defective allosteric regulation of PRS1 activity (regulatory defects) (15-19, 21, 22), increased apparent affinity of PRS for the substrate ribose-5-phosphate (23), and increased activity of the normal PRS1 isoform (formerly called catalytic superactivity) (24 -27). Study of the genetic and mechanistic bases of the heterogeneous kinetic alterations associated with PRS superactivity has shed light on the manner in which the synthesis of PRPP is regulated (3,16,21). In the case of regulatory defects, for example, patient-derived PRS1 cDNAs bear point mutations encoding recombinant mutant PRS1s with altered allosteric properties (resistance to noncompetitive purine nucleotide inhibition and increased sensitivity to P i activation) characteristic of those of PRS in cells from the respective affected individual (21). This finding provides evidence that allosteric control of PRS1 activity is important in regulating PRPP synthesis in human cells.
In contrast, overexpression of normal PRS1 transcript as well as PRS1 isoform has been demonstrated in cells from patients with overactivity of normal PRS (24). The association of increased PRS1 transcript level with increased PRS1 isoform content and enzyme activity suggests a pretranslational defect in the expression of PRPS1 in this type of inherited PRS superactivity (24). In studies aimed at further defining the process responsible for overexpression of normal PRS1 transcript and isoform, we have found selective acceleration of PRPS1 transcription in this disorder as well as evidence that in fibroblasts from affected individuals the rate of transcription of PRPS1 serves as a major determinant of PRPP and purine nucleotide production rates despite intact allosteric regulation of PRS activity.

EXPERIMENTAL PROCEDURES
Cell Lines-Fibroblast strains initiated from skin biopsies obtained from five normal individuals and three unrelated males with overexpression of normal PRS (25,26) were propagated in monolayer in Eagle's minimal essential medium containing 10% fetal bovine serum, 2 mM L-glutamine, nonessential amino acids, penicillin (100 units/ml) and streptomycin (100 g/ml). B lymphoblast lines, derived as described (28) from two normal individuals and two of the affected males, were propagated in RPMI 1640 medium containing 10% fetal bovine serum, 2 mM L-glutamine, and penicillin and steptomycin. Conditions for growth and study of cell cultures and procedures for preparing cell extracts for PRS activity determinations and PRS isoform analyses were as described (16,28). The extraction buffer was 8 mM sodium phosphate, 1 mM dithiothreitol, 1 mM EDTA (pH 7.5).
PRS Activity and Isoform Analyses-PRS activities were determined in crude cell extracts by a two-step procedure previously described in detail (24). In the first step, PRPP was generated at pH 7.5 in the presence of either 1.0 or 32.0 mM P i , saturating substrate concentrations (500 M MgATP; 350 M ribose-5-phosphate), and 5.0 mM MgCl 2 . Activities of PRS are expressed in units/mg of protein, where 1 unit is defined as 1 mol of PRPP formed/min at 37°C. Separation and quantitation of PRS1 and PRS2 isoforms in cell extracts were accomplished by a polyacrylamide-urea isoelectric focusing-immunoblotting procedure recently described in detail (24). The intensities of bands of identical mobility in samples of cell extracts and of purified recombinant PRS1 and PRS2 isoforms (8) were quantitated on a Molecular Dynamics (Sunnyvale, CA) computing densitometer. Cell extract band densities were related to those of the respective purified recombinant PRS isoform and to the amount of protein in the sample applied to the isoelectric focusing gel, permitting determination of the concentration of each isoform in the corresponding sample. Protein concentrations were determined by the method of Lowry et al. (29).
PRPP Generation and Purine Synthesis de Novo-Rates of PRPP generation were calculated as described (16,23) from simultaneously measured values for intracellular PRPP concentrations and rates of incorporation of [ 14 C]Ade into intracellular purine compounds. Incorporation of the purine precursor [ 14 C]formate into intracellular purines and purines excreted into the culture medium was measured to estimate rates of purine synthesis de novo in intact cells (16,28).
PRPS1 Genomic DNA Sequencing-Human genomic DNA clones of 8.5 and 13.6 kilobases were isolated and purified by repeated plaque hybridization (with a 450-bp oligo-[ 32 P]-labeled 5Ј PRS1 cDNA HindIII fragment) from a fetal fibroblast genomic DNA library in the vector FixI (Stratagene, Menasha, WI) (30). A 2.2-kilobase SacI fragment common to human DNA in both clones was subcloned into pGEM3Z, and the identity of the cloned DNA was confirmed by establishing sequence identity with that of exon 1 of PRPS1 and the adjacent 5Ј proximal genomic and 3Ј intron 1 DNA sequences published to date (31,32). The 2.2-kilobase PRPS1 genomic DNA was amplified in Escherichia coli and served as template in sequencing reactions utilizing SP6 and PRPS1-specific oligonucleotide primers (see Table I).
Genomic DNAs isolated from three normal fibroblast strains and three strains derived from affected individuals (1-2 ϫ 10 7 cells each) (30) served as templates for sequential polymerase chain reaction amplifications utilizing a nested primer approach to prepare PRPS1 genomic DNA segments inclusive of 850 bp of proximal 5Ј-untranscribed DNA and the 5Ј-transcribed but untranslated region of the gene (24). The amplification primers utilized are listed in Table I, with their respective locations relative to the PRS1 cDNA translation initiation site. Sequencing of the proximal untranscribed 5Ј genomic DNA and of the 144-bp 5Ј DNA segment encompassing the proposed human PRPS1 transcription initiation sites (31,32) to the translation initiation codon was carried out on both DNA strands from all strains, utilizing appropriately oriented primer sequences (Table I) from the PRPS1 promoter region and exon 1 of PRPS1. Sequence identity has previously been established for the 997-bp 3Ј-untranslated segment of the transcribed region of normal and patient cell-derived PRPS1 DNAs (24).
PRS1 and PRS2 Transcript Levels-Steady state levels of PRS1 and PRS2 transcripts were estimated by Northern blot analysis (30) after electrophoresis and transfer of cellular total RNA samples to nitrocellulose filters, prehybridization, and hybridization, all as described previously (24). Probes for hybridization were oligo[ 32 P]-labeled human PRS1 cDNA (2.3 kilobases), PRS2 cDNA (2.7 kilobases), and glyceraldehyde-3-phosphate (GAPDH) cDNA (1.8 kilobases). After washing at suitable stringency, radioactivities in the regions of the membrane corresponding to PRS, and control transcripts were quantitated on a PhosphorImager (Molecular Dynamics) for 12 h before exposure of the membrane to x-ray film for 24 -72 h at Ϫ70°C. Values for PRS1 and PRS2 transcript levels in a cultured cell total RNA sample are expressed relative to the GAPDH transcript level measured in that sample.
To estimate PRS transcript stability, relative PRS mRNA levels were determined in lymphoblasts incubated with the RNA polymerase II inhibitor, actinomycin D. Identical cultures of each lymphoblast line (20 ϫ 10 6 cells in 20 ml of growth medium) were incubated at 37°C for 0 to 24 h after the addition of actinomycin D (final concentration, 5 g/ml). At appropriate times, cultures were harvested by centrifugation, and cells were washed twice in ice-cold serum-free medium before extraction of RNA (30). Northern blot analysis was carried out as described above, except that the nitrocellulose filters were probed with a labeled cDNA probe (1.4 kilobases) for 18 S ribosomal RNA as well as labeled PRS cDNA probes. Values for PRS transcript levels are expressed relative to that of the 18 S ribosomal RNA level measured in the respective sample.
DNA Preparation and Filter Hybridization-Normal male and patient fibroblast genomic DNAs were digested with BamHI and HindIII. After electrophoresis of equal amounts of DNA in 0.8% agarose gels, DNAs were transferred to GeneScreen nylon membrane filters (NEN Research Products) under alkaline conditions (33). Blots were hybridized with oligo[ 32 P]-labeled PRS1 cDNA probes (12), including a fulllength (2.3 kilobase) PRS1 cDNA and a HindIII fragment containing the 5Ј 450 bp of PRS1 cDNA. Conditions of hybridization, washing, and exposure of filters to x-ray film were as described (12).
Rates of PRPS Gene Transcription-Transcription rates were estimated in fibroblasts and lymphoblasts by nuclear runoff analysis (34). Nascent RNA transcripts were [ 32 P]UTP-labeled in and isolated from fibroblast (7 to 9 ϫ 10 7 cells/assay) and lymphoblast (1.8 to 2.2 ϫ 10 8 cells/assay) nuclei prepared as described (30,34). RNAs in each nuclear preparation were hybridized to linearized plasmid pGEM3Z or to pGEM3Z-borne PRS1, PRS2, and GAPDH cDNAs immobilized by slot blotting onto a nitrocellulose filter (30). Binding of label to pGEM3Z without a cDNA insert served as background control, and a 28 S ribosomal RNA cDNA probe (35) was applied to the filter to bind labeled ribosomal RNA and thus reduce nonspecific binding elsewhere on the filter. After hybridization and washing at suitable stringencies, label corresponding to each slot was quantitated on a PhosphorImager, and the filter was exposed to x-ray film for 48 to 72 h at Ϫ70°C for densitometric quantitation. Rates of PRPS1 and PRPS2 transcription are expressed relative to transcription of GAPDH, determined on the same filter.
Ϫ a Except where indicated, oligodeoxynucleotide sequences have been reported (24). b Priming sites are numbered with respect to the A base of the translation initiation ATG codon and follow human PRS. c ϩ, sense strand; Ϫ, antisense strand.

PRPS1
Genomic DNA Sequencing-The 1161-bp sequence of normal human PRPS1 preceding the translation initiation ATG triplet is shown in Fig. 1. 2 Sequence identity between normal-and patient-derived PRPS1 genomic DNA was confirmed for the 5Ј-transcribed but -untranslated region of the gene from the 3 affected and 3 normal individuals, supporting the contention (24) that PRS1 transcript structure is normal in hemizygous males with overactivity of normal PRS1. In addition, the sequence of polymerase chain reaction-amplified PRPS1 DNA corresponding to the 850 bp immediately 5Ј to the consensus transcription initiation site (nucleotides Ϫ117 to Ϫ115, Fig. 1; Ref. 32) was identical regardless of whether normal-or patient-derived genomic DNA served as template.
Genomic DNA Hybridization-Southern blots of restriction enzyme-digested patient and normal male genomic DNAs showed identical patterns and intensities of hybridizing bands when filters were probed with either full-length normal PRS1 cDNA (Fig. 2) or a HindIII fragment containing only the 5Ј 450 bp of normal PRS1 cDNA (not shown). These findings exclude gene amplification as a likely basis for PRPS1 overexpression in this form of PRS superactivity.
PRS Transcript Levels and Stability-Northern blot analysis confirmed (24) increased concentrations of PRS1 but not PRS2 transcript in extracts of patient fibroblasts and, to a lesser degree, lymphoblasts when expressed relative to levels of either GAPDH transcript (Table II) 3 or 18 S ribosomal RNA (Fig.  3A) in the same extract. Nevertheless, rates of decrement of PRS1 mRNA (Fig. 3, A and B) and of PRS2 mRNA (Fig. 3, A and C) (relative to those of 18 S ribosomal RNA) from normal and patient lymphoblasts during incubation with actinomycin D were virtually indistinguishable. PRS1 transcript half-lives were 10.8 Ϯ 1.4 h and 11.1 Ϯ 0.9 h (mean ϮS.D. of 3 determinations in each cell line), respectively, in normal and patient cells. Corresponding mean half-lives for PRS2 transcripts were 13.1 and 12.2 h. In all instances, PRS1 and PRS2 transcripts were detectable only as single hybridizing bands at 2.3 and 2.7 kilobases respectively, and no additional hybridizing bands suggestive of immature or alternatively processed PRS transcripts were observed.
Rates of PRPS Gene Transcription-Slot-blot analyses of specific RNAs labeled in nuclei isolated from cultured normal and patient cells showed consistent differences with respect to relative rates of labeling of PRS1 mRNA ( Fig. 4; Table III). In fibroblasts from 3 affected males, rates of PRS1 transcript 2 The only sequence in the data base with significant relatedness to the sequence in Fig. 1 is the sequence gb/M31078 for rat PRPS1, exon 1, that shows 94% identity in the region corresponding to Ϫ149 to Ϫ94 in human PRPS1. 3 Normal and patient cell strains are designated by initials of donors in Figs. 3 and 4 and Tables II and III. Numbering is relative to the A of the ATG codon (ϩ1). Transcription initiation sites identified by primer extension and S1 nuclease protection assays (32) are indicated by arrowheads. The 5Ј end of the human PRS1 cDNA (Ϫ122) (cloned from normal human B lymphoblasts) is denoted by the horizontal arrow. With exception of a single base (C) addition at Ϫ410, the sequence from Ϫ437 to ϩ3 is as previously published (32). An identical sequence was found in the interval from Ϫ967 (vertical arrow) to ϩ3 when 3 normal and 3 patient fibroblast-derived PRPS1 DNAs were amplified and sequenced. 1 and 2) and a lymphoblast line derived from patient TB with overactivity of normal PRS (lanes 3 and 4). DNAs were digested with BamHI (lanes 1 and 3)and HindIII (lanes 2 and 4). Each lane contained 10 g of enzyme-restricted DNA, and after overnight electrophoresis in 0.8% agarose and transfer to a nitrocellulose filter, hybridization was carried out with a full-length oligo-labeled (2.3 kilobase) PRS1 cDNA. The filter was washed (33) and exposed to a PhosphorImager for 4 h. Shown on the right are the locations of bands corresponding to DNA standards of known size (kilobases). labeling relative to those of GAPDH transcript labeling were 3to 4-fold greater than in cells from 5 normal individuals (Table  II; Fig. 4A). In lymphoblasts, the corresponding increases in relative labeling of PRS1 transcript were 1.9-to 2.4-fold (Table  II; Fig. 4B). Relative rates of labeling of PRS2 mRNA were indistinguishable in normal and patient cells, a point more readily apparent in lymphoblasts, which have substantially higher relative rates of PRS2 transcription than fibroblasts (Table III). In conjunction with the preceding studies excluding PRPS1 gene amplification or altered PRS1 transcript structure or stability, these findings suggest that increased expression of PRPS1 in overactivity of normal PRS1 results at least in major part from selective acceleration of PRPS1 gene transcription.

FIG. 2. Southern blot analysis of restriction enzyme-digested human genomic DNA extracted from a normal male B lymphoblast line (lanes
PRPS1 Gene Transcription, PRS Expression, and Purine Nucleotide Synthesis-If accelerated transcription of PRPS1 is the major determinant of PRS overactivity in fibroblasts and lymphoblasts from affected patients, correspondence between rates of PRPS1 transcription and PRS1 mRNA levels, PRS1 isoform concentrations, PRS enzyme activities, and rates of PRPP and purine nucleotide synthesis should be demonstrable. Such correspondence would support the view that PRPS1 transcription rate can determine expression of PRS1 activity even when allosteric regulation of enzyme activity is intact. To assess these relationships, we compared relative rates of PRPS1 transcription, relative PRS1 mRNA concentrations, PRS isoform concentrations, and PRS activities in fibroblast strains and lymphoblast lines derived from normal individuals and affected patients. In addition, PRPP generation and rates of purine synthesis de novo were determined in these cells. Differences between normal and patient fibroblasts with respect to PRPS transcription rates, PRS mRNA and isoform levels, and maximal PRS enzyme activities are presented in Table II, where values for individual patient-derived strains are expressed relative to the respective mean values for the group of 5 normal fibroblast strains. For each patient-derived cell strain, nearly coordinate increases are apparent in all processes relating PRPS1 transcription to PRS activity measured at 32 mM P i , a concentration of P i at which allosteric inhibition of PRS activity by endogenous purine nucleotides is minimal (16). As is also shown in Table II, intact patientderived fibroblasts synthesize PRPP and purine nucleotides at increased relative rates, which are, however, more modest than the increased relative rates of PRPS1 gene transcription or the increased relative levels of PRS1 transcript or isoform, or maximal PRS activities. Relative increases in PRPP and purine nucleotide synthesis, in fact, more closely parallel relative in-creases in PRS activities measured at 1.0 mM P i , a concentration closer to that in intact fibroblasts (36) and at which allosteric inhibition by endogenous nucleotides is potent (16). The modulated relative increases of PRPP and purine nucleotide synthesis in patient cells thus appears to reflect both the operation of allosteric inhibition of PRS activity and the higher intracellular concentrations of purine nucleotide inhibitors in intact patient fibroblasts (16).
Similar relationships are detectable but less immediately apparent in lymphoblast lines (Table II) than in fibroblast strains, first, because of smaller differences in all measurements comparing patient and normal lymphoblasts, and, second, because the contribution of PRS2 to total PRS isoform content and PRS activity is substantially greater in lymphoblasts than fibroblasts (24). In fact, as previously noted (24,28), the small increases in maximal PRS activities measured in lymphoblast extracts from affected patients appear insufficient to drive excessive production of PRPP or purine nucleotides in intact cells. DISCUSSION The current studies provide evidence for a selectively increased rate of transcription of PRPS1 as the pretranslational aberration underlying increased expression of the normal PRS1 isoform in inherited PRS overactivity. In each of the fibroblast strains and lymphoblast lines cultured from affected individuals, transcription of PRPS1 was consistently greater relative to that of GAPDH than was the case in corresponding normal cells. Although the relative increases in PRPS1 transcription rates were greater in patient fibroblasts than lymphoblasts, study of the latter cell type permitted a more accurate demonstration of the selectivity of accelerated PRPS1 transcription. That is, relative rates of PRPS2 transcription in normal cells were substantially greater in lymphoblasts than in fibroblasts, but for both lymphoblasts and fibroblasts, relative rates of PRPS2 transcription in normal and patient cells were indistinguishable. Consistency was also found in the relative differences between normal and patient cells in each of the processes mediating flow of genetic information from PRPS1 transcription to maximal PRS1 isoform expression (enzyme activity at 32 mM P i ). This finding, in conjunction with the demonstration that patient and normal cells did not differ in PRS1 mRNA or isoform structure or in alternative pretranslational mechanisms that might otherwise explain PRS1 isoform overexpression, supports the contentions that PRS1 isoform concentrations are determined, at least in major part, at the TABLE II Relationships between accelerated PRPS1 transcription and PRS activity, PRPP generation, and rates of purine synthesis in human fibroblasts with overactivity of normal PRS All values are derived from the means of at least three separate determinations in each fibroblast strain and lymphoblast line. Means for the five normal fibroblast strains and the two normal lymphoblast lines were averaged and are shown. Each normal value was assigned a relative value of 1.0 to which the corresponding relative mean values measured in patient cells are compared. Relative values are in parentheses. level of transcription and that an inherited increase in PRPS1 transcription rate provides the basis for the increase in the concentration of the normal PRS1 isoform in cells of affected individuals.
The genetic basis of inherited acceleration of PRPS1 transcription remains to be determined. Sequence identity of patient and normal PRPS1 DNAs in the 850-bp region 5Ј to the consensus transcription initiation site (32) excludes mutation in the gene promoter and immediate 5Ј PRPS1 flanking sequence, for which examples of transcriptional dysregulation have been established, as in the thalassemias (37)(38)(39) and hereditary persistence of fetal hemoglobin (40,41). Among alternative possibilities to explain accelerated PRPS1 transcription are mutations in a more remote promoter element, either in contiguity with the immediate 5Ј-flanking sequence (42) or even substantially distant (43); in a cis-acting element within or adjacent to the PRPS1 gene, such as an intronic enhancer or suppressor (44) or a 3Ј-flanking DNA sequence (45); or in a trans-acting gene influencing the regulation of PRPS1 transcription. In any case, X chromosome-linked transmission of PRS catalytic superactivity (14) favors the view that the primary defect altering PRPS1 transcription is itself Xlinked. In addition to extended PRPS1 5Ј-flanking region sequencing, functional analysis of the PRPS1 promoter and adjacent 5Ј-flanking DNA, comparing PRPS1 promoter-plasmid construct expression in normal and patient cells, should prove helpful in distinguishing among these possibilities.
Prior studies (3,16) comparing mechanisms of purine nucleotide overproduction in fibroblasts from individuals with PRS superactivity (either catalytic or regulatory defects in PRS1) and severe deficiency of hypoxanthine-guanine phosphoribosyltransferase (EC 2.4.2.8) have confirmed that the rate of the pathway of purine synthesis de novo is controlled at the sequential PRS and amidophosphoribosyltransferase (EC lymphoblasts were incubated with 5 g/ml actinomycin D for the indicated time intervals. RNA was isolated, and 10 g was run in each lane of a denaturing gel, blotted onto a nitrocellulose membrane, and sequentially probed, first with full-length PRS1 and PRS2 cDNAs and then with an 18 S rRNA cDNA. B and C, PRS1 and PRS2 mRNAs and 18 S rRNA levels were quantitated using a Phosphor-Imager, and PRS1 and PRS2 mRNA expression was normalized for 18 S rRNA expression. q, normal cells; E, patient cells. 2.4.2.14) reactions, the latter the first committed step in the pathway. Within this regulatory domain, purine nucleotides inhibit both reactions (3,16,36), exerting more potent inhibition of amidophosphoribosyltransferase by antagonizing PRPP activation of this reaction (3,46). Although fibroblasts with either regulatory defects in PRS or overactivity of normal PRS share the biochemical hallmarks of PRS superactivity, increased rates of PRPP and purine synthesis and increased intracellular purine nucleotide concentrations, defect-specific differences in the intracellular control of PRPP and purine nucleotide synthesis are apparent (16).
In the case of PRS regulatory superactivity, where PRS activity in cell extracts or purified enzyme preparations are resistant to purine nucleotide inhibition, accelerated rates of intracellular PRPP and purine nucleotide synthesis are refractory to inhibition by exogenous purine base precursors of purine nucleotides or by endogenous increases in purine nucleotide concentrations (16). Thus, in cells bearing PRS1s with any of an array of point mutations (21), in vitro defects in allosteric regulation of PRS1 activities are paralleled by dysregulation of PRPP and purine synthesis. In contrast, allosteric regulation of PRS activity is normal in enzyme preparations from fibroblasts with overactivity of normal PRS, and suppression of PRPP and purine nucleotide synthesis in response to purine base addition is intact in the corresponding cells (16). Nevertheless, these cells express increased rates of PRPP and purine nucleotide synthesis.
The apparent paradox of increased rates of PRPP and purine nucleotide synthesis in fibroblasts with overactivity of normal PRS catalytic superactivity despite increased purine nucleotide inhibitor pools and normal allosteric regulation of PRS activity (16) is best resolved by the view that the increased concentration of the normal PRS1 isoform (24) results in a rate of PRPP synthesis sufficient to activate amidophosphoribosyltransferase despite coexisting increased levels of inhibitory purine nucleotides (16). Consistent with this formulation are the substantially more modest increases in rates of PRPP and purine nucleotide synthesis in fibroblasts with intact allosteric regulation than in cells in which mutations in PRPS1 impair this control mechanism (16). Thus, although intact allosteric inhibition apparently modulates expression of PRPP and purine overproduction in fibroblasts with excessive PRS1 isoform, this regulatory mechanism is insufficient to overcome the excessive expression of enzyme activity resulting from acceleration of PRPS1 transcription. The current studies provide, then, an example of a circumstance in which transcription of PRPS1 is a major determinant of PRPP and purine nucleotide synthetic rates.  a Values given are means of the ratios of densities in the respective slots, measured on a PhosphorImager screen during 24 h of exposure. For each cell strain or line, at least three determinations were made, with agreement among measurements within 20%, except for PRPS2 in fibroblasts, where agreement was within 40%.