Adenine Phosphoribosyltransferase Deficiency:A Previously Undescribed Genetic Defect in Man

WILLIAM N. KELLEY, ROBERTI. LEvy,' FREDEucKM. ROSENBLOOM,J. FANKHENDERSON,and J. EDWINSEEGMILLER

From the Section on HumanBiochemical Genetics, National Institute ofArthritis and Metabolic Diseases and Laboratory of Molecular Diseases,National Heart Institute,* National Institutes of Health,Bethesda, Maryland 20014

A B ST RA CT A deficiency of adenine phospho-ribosyltransferase (A-PRTase) is described infour members in three generations of one family.A-PRTase is coded by an autosome and themutants described in this report are heterozygotesfor this enzyme defect. The level of enzyme activ-ity in these heterozygotes was inappropriately low,ranging from 21 to 37% of normal rather thanthe expected 50%o of normal. Examination ofvarious physical and chemical properties of theA-PRTase obtained from the mutant heterozy-gotes failed to reveal differences from the normalenzyme. These patients have no discernable ab-normality in uric acid production despite thefinding that patients with a deficiency of a closelyrelated enzyme, hypoxanthine-guanine phospho-ribosyltransferase, invariably produce excessivequantities of uric acid. A relationship of theA-PRTase deficiency to the disturbance in lipo-protein metabolism observed in the propositus hasnot been firmly established. Possible manifesta-tions of the homozygous form of this enzymedeficiency will require identification of such indi-viduals in the future.

This study was presented in part at the 60th AnnualMeeting of the American Society for Clinical Investiga-tion, Atlantic City, N. J., 6 May 1968. (J. Clin. Invest.47: 53 a. 1968.)

Address reprint requests to Dr. Kelley. Present address:Dept. of Medicine, Duke University Medical Center,Durham, N. C.

Received for publication 19 April 1968.

INTRODUCTION

A partial deficiency of an enzyme of purine me-tabolism, hypoxanthine-guanine phosphoribosyl-transferase (HG-PRTase) (Fig. 1), has recentlybeen described in five patients with gout whoproduce excessive quantities of uric acid (1).Subjects with a complete deficiency of this enzymealso produce uric acid in excess but in additionhave a bizarre neurological and behavioral syn-drome characterized by self-mutilation, choreo-athetosis, spasticity, and mental deficiency (2).These findings have suggested that the enzymeHG-PRTase is somehow concerned with the nor-mal regulation of purine biosynthesis in man.Genetic data based on biochemical (3) as well asclinical studies (4-7) have established that thegene for this enzyme is located on the X-chromo-some.

In the present study a previously undescribeddeficiency of a closely related enzyme, adeninephosphoribosyltransferase (A-PRTase) (Fig. 1),is reported in four members in three generationsof one family. Despite apparently close functionalsimilarity of this enzyme to HG-PRTase, its par-tial deficiency in these subjects is not associatedwith excessive uric acid excretion. The propositus,however, does have Type II hyperbetalipopro-teinemia (8). From the limited genetic data avail-able in this kindred, the gene for A-PRTaseappears to be located on an autosome rather thanbeing X-linked and affected individuals may beheterozygous for the defect, even though their

The Journal of Clinical Investigation Volume 47 1968 2281

ADENINE PHOSPHORIBOSYLTRANSFERASE

NH2

+ PRPP arCNt H + PP

HC.~ C

R-P

Adenylic Acid

HYPOXANTHINE-GUANINEPHOSPHORIBOSYLTRANSFERASE

OH

NC

H

Hypoxantkhin.

OH

H24CHH

+ PRPP

OH

II| CHHN /C N/

R-PInosinic Acid

+ PRPP is

Guanine

OH

I \CHN vC

R-P

Guanylic Acid

FIGURE 1 Reactions catalyzed by the two enzymes, adenine phosphoribosyltransferase andhypoxanthine-guanine phosphoribosyltransferase.

enzyme activity is substantially less than 50%o ofnormal.

Case report. The propositus, E. H., a 45 yr oldscientist (Ph.D.) and administrator, who has con-

tinued to enjoy excellent health, first developedarcus juvenilis at age 35. Serum cholesterol at thattime was said to be elevated. In 1962 he beganto develop nodules on both external pinnae. In1967 a colleague felt that the lesions on his ears

might represent tophaceous deposits and a serum

urate was reported to be 7.2 mg/100 ml. Becauseof this finding, he was referred to our service atthe Clinical Center for further evaluation. Thepatient had no history suggestive of gouty arthritisor nephrolithiasis, but a disorder of lipoproteinmetabolism was suspected on the initial examina-tion because of the presence of arcus and the his-tory of a high cholesterol in the past. Erythrocyteswere obtained for determination of HG-PRTaseand A-PRTase as part of a control series beingobtained in our out-patient department. His eryth-

rocyte A-PRTase activity was quite low despitenormal HG-PRTase activity, and he was admittedto the Clinical Center for further study.

Clinical history revealed an asymptomatic pro-

teinuria until age 18 and an episode of orchitis in1961 at the age of 39. There was no history ofxanthomata, diabetes, or early death in the family.His mother had a history a recurrent mild mono-

articular arthritis associated with mild hyper-uricemia which had been diagnosed as gout. Thepatient has three daughters (Fig. 2) two of whom(III-4 and III-5). are exceedingly bright, alert,and healthy young ladies. A third daughter(III-6), equally cooperative, has a mild cerebellardisorder attributed to an episode of encephalitiswhich occurred at the age of 8 months.

Physical examination including detailed neuro-

logical and ophthalmological evaluation was nega-

tive except for the presence of bilateral arcus

juvenilis. There were no xanthelasmata nor tendonor tuberous xanthomata. Hemogram, including

2282 Kelley, Levy, Rosenbloom, Henderson, and Seegmiller

l H2

H

Adenine

+ PP

+ PP

I

l 2 , 3 4 5 6

2 3 4 5 6 7 8 9

* * Examined, Affected / Propositus8l 0 Examined, NormalO 0 Not Examined

FIGURE 2 "H" Family Pedigree, showing inheritance of adenine phosphoribo-syltransferase deficiency.

hemoglobin, hematocrit, white blood cell count anddifferential, erythrocyte sedimentation rate, redblood cell indexes, platelet count, and reticulo-cytes, was normal. Renal function assessed byurinalysis, timed phenolsulfonphthalein excretion,blood urea nitrogen, and creatinine clearance wasnormal. There was no proteinuria. Protein-boundiodine, 24 hr 131I uptake, thyroid scan, oral glucosetolerance test, basal excretion of 17-hydroxycorti-coids and 17-ketosteroids, and serum calcium andphosphorus concentrations revealed no evidence ofendocrine dysfunction. Liver function tests includ-ing alkaline phosphatase, serum glutamic oxala-cetic transaminase, serum glutamic pyruvic trans-aminase, lactate dehydrogenase, thymol turbidity,and bilirubin were normal. There was, however,10% retention of Bromsulfalein after 45 min.Additional studies including protein electrophore-sis, electrolytes (Na+, K+, Cl-, HCO3- and Mg++),acid phosphatase, lipase, amylase, ceruloplasmin,copper, total serum iron, and VDRLwere normal,as were ECG and EEG. Audiometry revealedbilateral high tone hearing loss of the sensori-neural type. X-rays of the chest, axial skeleton,and large joints were unrevealing. The lipoproteinvalues obtained in this patient are indicated inTable I. The values remained essentially un-changed on repeated determinations over a spanof 5 months, including a period in the hospitalwhen his dietary intake of cholesterol was con-trolled.

METHODSA-PRTase and HG-PRTase activities were assayed indialysed lysates of washed erythrocytes by a radiochem-ical method described previously (1). The samples forenzyme assay obtained from the propositus, E. H., aswell as his wife and three daughters, were frozen im-mediately without loss of enzyme activity. The sampleson the remainder of the family were airmailed in icefrom California, with 2 to 3 days intervening before thesamples could be frozen. This delay could have resultedin a loss of not more than 20% of the HG-PRTase andA-PRTase activities. Orotate phosphoribosyltransferase(orotidylic pyrophosphorylase) was assayed in a simi-lar manner using 0.6 mMorotic-6-'4C acid hydrate (3.5mc/mmole) (New England Nuclear Corp. Boston,Mass.); 1 mm5-phosphoribosyl-1-pyrophosphate (PRPP)(Pabst Research Laboratories Milwaukee, Wis.); 55 mMTris buffer, pH 7.4; 5 mmMgCl2; and 3-6 mg of he-molysate protein in a final volume of 100 pl. After in-cubation for 3 hr at 38°C, the reactions were terminatedby the addition of 2 ,umoles of ethylenediaminetetraace-tate (EDTA) and immediately frozen in a dry iceacetone bath. 20 pi of the reaction mixture was placedon 3MMWhatman paper with 0.06 ,umole of orotidylic(a gift of Dr. Herbert Windmueller) and uridylic acid,and the reaction products were separated from the sub-strate by high voltage electrophoresis in 0.05 M boratebuffer, pH 9.0, containing 0.001 M EDTA at 4000 v for30 min. The areas of the paper containing orotidylic anduridylic acid were located by inspection of the paperunder ultraviolet light or by radioautography and cut outand counted in a liquid scintillation counting system at60%o efficiency. The total counts present in orotidylic anduridylic acid were taken as a measure of phosphoribosyl-transferase activity. The values so obtained compareclosely to the values reported in erythrocytes by Smith,Huguley, and Bain using a different assay system (9).

Cholesterol (10) and triglyceride (11) concentrations

Adenine Phosphoribosyltransferase Deficiency 2283

TABLE IErythrocyte Adenine Phosphoribosyltransferase Activity and Plasma Lipids in

Members of the "H" Family

Adenine phos-phoribosyl-transferase Total VrLDL* Beta Alpha

Patient Relationship Age activity cholesterol Triglyceride cholesterol cholesterol cholesterol

mpumoles/mgprotein per hr mg/100 ml mg/100 ml mg/100 ml mg/100 ml mg/1OO ml

E. H. Propositus 11-3 45 6.6t 318§ 160 36 248§ 34P. H. Daughter III-4 12 9.0 140 54 8 90 42D. H. Daughter II1-5 10 9.6 162 28 4 113 45V. H. G. Mother 1-2 6.0 298 164 58 204 36Er. H. Daughter I11-6 8 21.7 170 30 6 110 54J. H. Wife I1-4 40 26.0 142 25 2 80 60G. H. Brother 11-6 44 28.8 180 63 10 120 50L. H. Niece III-7 16 27.1 178 100 20 116 42Ev. H. Niece 111-8 18 27.1 168 122 28 103 35Je. H. Niece 111-9 14 27.0 146 80 16 79 510. H. Brother II-1 47 22.8 260 262§ 52 180 28Ba. H. Niece III-1 17 25.8 160 24Ga. H. Nephew III-2 11 29.2 188 40S. H. Niece III-3 14 27.5 164 53G. P. Cousin 20 37.6 146 64 14 93 39

* Very low density lipoprotein (d > 1.006 g/ml).Normal value - 31 i 6 (mean :1 1 SD in 32 normal control subjects).

§ Abnormal value (8).

were determined by methods previously described. Themethodology of lipoprotein quantification as well as ageand sex-corrected normal limits for these values hasbeen described elsewhere (8). The blood samples in eachcase were drawn with the patient in the fasting state.

We are indebted to Mrs. Mary McGinniss and Mrs.Mary Ann Campbell of the Clinical Center Blood BankDepartment for determination of the red cell phenotypesin members of the "H" family.

RESULTS

The specific activity of the A-PRTase assayed indialysed erythrocyte hemolysates obtained from 15members of the "H" family is indicated in TableI. The A-PRTase activity was distinctly reducedin four members of this family, the propositus(E. H. II-3), his mother (V. H. G. I-2) and twoof his three daughters (P. H. III-4, D. H. III-5).The values obtained in these four subjects (6.0-9.6 mumoles/mg protein per hr) ranged from 21to 31% of normal when compared with those ofother family members. A comparison of A-PRTaseand HG-PRTase activity in 13 of the familymembers is illustrated in Fig. 3. For each memberof the family tested, including those with reducedA-PRTase activity, the HG-PRTase activity,

_ 400wz

o 35

cr

3CE

u- 25

cr

.mC 20

IOn

0o1CLa-

zzUl 50ox

IF-

.0000

0

- 8D.H.P.H.

0 EH._ vG.

w 160

cO 140X

a. 2

2E5-

12E

(nl00zL-

0

z 600to

0He 40

< 20

co

0

>

oV.G.

o D.H.

0

PH.i E.H.2

*S.___ ____.

.c 160

CL

mM 1400cra-

E 120n

I-aFE

E 100

.2

cn 80

C 60

0CL

n 40

=

f 20(D

I0

008 E.H.0

- D.H.8 PH.

,--.

FIGURE 3 Specific activity of adenine phosphoribosyl-transferase and hypoxanthine-guanine phosphoribosyl-transferase in members of the "H" family. The solidand broken lines indicate the mean ± 2 SD in 32 normalcontrol subjects. The subjects with reduced A-PRTaseactivity are indicated by the open circles.

2284 Kelley, Levy, Rosenbloom, Henderson, and Seegmiller

t---------

TABLE I IAdenine Phosphoribosyltransferase Activity and Plasma Lipids in Patients with

Familial Hyperbetalipoproteinemia

Adenine phos-phoribosyltrans- Total VLDL* Beta Alpha

Patient Age ferase activity cholesterol Triglyceride cholesterol cholesterol cholesterol

mpsmoles/mgprotein per hr mg/100 ml mg/100 ml mg/100 ml mg/100 ml mg/100 ml

J. N. 48 20.5T 312§ 128 10 264§ 48E. B. 43 21.2 314§ 224§ 46 235§ 33J. B. 30 22.8 338§ 214§ 58 261§ 27H. J. 34 33.8 382§ 179§ 44 306§ 32M. B. 24 32.2 418§ 64 6 370§ 42M. H. 43 32.2 462§ 70 22 388§ 52

* Very low density lipoproteins (d > 1.006 g/ml).t Normal value - 31 1 6 (mean :1 1 SD in 32 normal control subjects).§ Abnormal values (8).

when assayed with either hypoxanthine or guanine and pH optimum have also failed to reveal prop-as substrate, was within two standard deviations erties unique to this variant. Therefore, no evi-of the mean. Mixtures of the hemolysate obtained dence to suggest a structural alteration in thefrom the propositus, E. H., with a hemolysate deficient enzyme has yet been obtained.known to have normal A-PRTase activity, gave In addition to a partial deficiency of A-PRTase,the value expected from such a combination, indi- the propositus, E. H., also had an abnormal lipo-cating that the decreased A-PRTase activity ob- protein pattern consistent with type II hyperlipo-served in patient E. H. was not due to the pres- proteinemia (8) characterized by an increasedence of a reversible inhibitor. concentration of cholesterol and betalipoprotein in

The electrophoretic migration of the deficient his plasma. Investigation of 14 other members ofenzyme in starch gel, as well as its heat stability, the family, including the two daughters andhas been normal on repeated testing. In addition, mother who had a reduction in A-PRTase activitypreliminary studies involving gel filtration through m wSephadex, electrophoretic migration in Geon- failed to reveal a similar lipoprotein abnormalityPevikon, determination of its Michaelis constants,' (Table I). His brother, 0. H. (II-1), had an

abnormal lipoprotein pattern quite unlike that1 Henderson, J. F., W. N. Kelley, F. M. Rosenbloom,

and J. M. Seegmiller. The kinetics of human adenine found in the propositus, a pattern characterizedphosphoribosyltransferase. Manuscript in preparation. by endogenous hypertriglyceridemia (Type IV

TABLE IIIRed Cell Antigenic Phenotype in Members of the "H" Family

Rh factors MNS P Kell Duffy Kidd Lewis

Patient ABO C C- D Du E c e M N S s U Pi K k Kpa Kpb Fya Fyb Jk- Jkb Lea Leb Xg&

E.H.II-3 0 + - + - + + + + + + + + + + - + + + + + - + -

P. H. III-4 0 + - + - - + + - + + + + + + - + - + - + - + +D. H. III-5 0 + - + - - + + + + + + + + + - + + + - + + -

Er. H. III-6 0 + - + - - + - + - + + + - + - + + + + + + +J.H.II-4 A + - + - - + + + - + + + - + - + + + - + + +G. H. II-6 0- - - + + + + + + + + + + - + - + + + + - -

L. H. III-7 A + - + - + + + + ++ ++- + + - + + + + - + -Ev. H. III-8 0 + - + - + + + + + + + + _ + _ + + _ + +Je. H. III-9 0 + - + - + + + + + + + + _ + + _ + _ +O. H. II-l 0 - ±- ++ - + ++- - -

Ba. H. III-t 0 + + - + + + - + + + + - + - + - + + + +Ga. H. III-2 0 - - + + + - + - + + - + - + - + + - +S. H. III-3 0 - - + + + - + - + + - + - + - + + - - + +

Adenine Phosphoribosyltransferase Deficiency 2285

hyperlipoproteinemia). A repeat sample of blood2 months later showed no abnormality.

A-PRTase activity was assessed in six patientspreviously demonstrated to have familial type IIhyperlipoproteinemia (hyperbetalipoproteinemia)(8). In each of these subjects, the activity of thisenzyme was normal (Table II).

None of the affected members of the "H" familywere hyperuricemic. The propositus excreted 514mg of uric acid in his daily urine after 4 days ofdietary purine restriction (normal < 600 mg/day). Serum urate at that time was 5.2 mg/100ml.

The red cell antigenic phenotypes obtained oneach member of the "H" family are illustrated inTable III. The data obtained on each subject arecompatible with the pedigree proposed in Fig. 2.The small number of affected individuals preventsan accurate estimate of the linkage of this enzymewith any of these blood groups.

DISCUSSION

In the present study a previously unreported defi-ciency of an enzyme of purine metabolism, adeninephosphoribosyltransferase (A-PRTase), has beendescribed. The reduced A-PRTase activity in thepropositus, E. H., has been constant with repeatedassay over a period of 8 months, indicating thatthis defect was not due to some unknown transienteffect on his circulating erythrocytes. The observa-tion of reduced activity in his mother and in twoof his three children also provides evidence thatthe deficiency of this enzyme represents a geneticand not an acquired defect.

The genetic data available suggest that the pro-positus, his mother, and his two affected daughtersare probably heterozygotes for this enzyme defectand that the defective allele is most likely locatedon an autosome. If the propositus were a mutanthomozygote carrying a double dose of the defectivegene, all three of his children would be hetero-zygotes and comparable values for A-PRTasewould have been predicted in all three progeny.However, one had normal A-PRTase activity,whereas the other two had distinctly reducedactivity. Similarly, if the defective gene werelocated on the X-chromosome and the proposituswere a nmtutant hemizygote, all three dauighterswould also have been obligate heterozygotes. Weare not able to exclude completely the possibility

of X-linkage of A-PRTase based on the study ofthis family, since only one cell type was examinedand male to male transmission of the enzymedefect was not observed. One could argue that theenzyme is X-linked and that patient Er. H. (III-6)is heterozygous for the enzyme defect despite nor-mal erythrocyte A-PRTase activity if it is assumedon the basis of the Lyon hypothesis that there wasearly inactivation of the abnormal X-chromosomein the hematopoietic system in this individual.However, recent studies in other families of athermostable A-PRTase isoenzyme which has nor-mal enzyme activity revealed that male to maletransmission of this allele did occur,2 providingstrong evidence that A-PRTase is not X-linkedand therefore must be coded for by autosomalDNA. The proposed mode of inheritance is alsobased on the assumption of full penetrance, theabsence of a "neutralizing" or "suppressor" factorin the apparently unaffected daughter, and theproposed parenthood which is in agreement withthe erythrocyte antigenic phenotypes.

The failure to detect evidence of a structuralabnormality in the A-PRTase obtained from thepropositus is compatible with the suggestion thathe is a heterozygote and that the enzyme presentis the product of only the normal allele. Thenormal allele would be expected to code for anormal A-PRTase, whereas the mutant allelemight code for a protein with no enzyme activityby our assay. It must be emphasized, however,that many structural gene mutations are possiblewhich might affect enzyme activity and not alterthe properties which we have examined and thatindisputable evidence for a structurally normalA-PRTase in this kindred would require knowl-edge of its full amino acid sequence.

In most autosomal disorders in which adequatestudies of enzyme activity in heterozygotes havebeen conducted, the enzyme activity in question isfound to be reduced to values close to 50% ofnormal, as would be expected. A striking excep-tion is found in orotic aciduria. Heterozygotes forthis disorder have approximately 20-35%o or nor-mal activity for the enzyme(s) orotate phospho-ribosyltransferase (orotidylic pyrophosphorylase)and orotidylic decarboxylase which are virtually

2Hendersoni, J. F., W. N. Kelley, F. M. Rosenbloom,and J. M. Seegmiller. Inheritance of purine phosphori-bosyltransferases in man. Manuscript in preparation.

2286 Kelley, Levy, Rosenbloom, Henderson, and Seegmiller

absent in the mutant homozygotes (9). Thisunexpected finding in the heterozygotes for oroticaciduria has been attributed to the presence of anabnormal regulator gene product produced at themutant locus (12). Preliminary data reported byKrooth, using cultured fibroblasts derived from amutant homozygote with this disorder, led himalso to conclude that the mutation may be at aregulator site, rather than on a structural gene(13, 14).

The A-PRTase activities observed in the foursubjects heterozygous for this enzyme defectranged from 21 to 37%o of normal. Because thesevalues are inappropriately low and similar inrange to the enzyme values reported in the hetero-zygotes for orotic aciduria, a regulator gene muta-tion could be proposed by analogy with the lattercase. The inappropriately low enzyme activity,however, could also result from a structural genemutation and this finding in itself does not neces-sarily require that a regulator mutation be postu-lated. One possible model for such a reduction inactivity due to a structural gene mutation wouldrequire the enzyme to be a dimer formed by ran-dom aggregation of identical subunits from eachallele. If this protein had enzymatic activity onlywhen it existed as a dimer composed of two nor-mal subunits, then a structural gene mutation in-volving one allele in the heterozygote would leadto a reduction in enzyme activity to values ap-proximately 25%o of normal, as is observed in bothA-PRTase deficiency and orotic aciduria. Thereis, in fact, some evidence which suggests a subunitcomposition for A-PRTase.3 It is apparent thatwhat seems by our present techniques to be astructurally normal A-PRTase in the mutant het-erozygotes is compatible not only with a mutationon a structural gene as discussed earlier, but alsowith an alteration at a regulator site. In conclusion,it is not possible at this time to determine whetherthe mutation in the "H" family involves a site ona structural or on a regulator gene.

In addition to the similarities noted in the het-erozygotes deficient for these enzymes, orotatephosphoribosyltransferase and A-PRTase alsohave quite similar enzymatic functions since theycatalyze the transfer of the ribose-5-phosphate

3 Gadd, R. E. A., and J. F. Henderson. Unpublisheddata.

moiety of 5'-phosphoribosyl-1-pyrophosphate to apyrimidine and purine base, respectively, to formthe appropriate 5'-mononucleotides. The possibilitythat the four subjects in this report were actuallyheterozygotes for orotic aciduria and that the re-duced A-PRTase activity observed reflected eithera deficiency of yet a third enzyme in orotic aciduriaor the reduced orotate phosphoribosyltransferaseactivity itself, has been excluded, since orotatephosphoribosyltransferase activty was normal inthe propositus.

The subjects in this report had no detectableabnormality of purine metabolism despite thefinding that the deficiency of a closely relatedenzyme, HG-PRTase, is always associated withexcessive purine synthesis.

The relationship of the A-PRTase deficiencyand the hyperbetalipoproteinemia observed in thepropositus is not known. This abnormal lipopro-tein pattern is often familial, being inherited as anautosomal dominant (8). The common causes ofacquired hyperbetalipoproteinemia including ex-aggerated dietary intake of cholesterol, hypothy-roidism, nephrosis, myeloma, macroglobulinemia,and obstructive liver disease were excluded in thispatient. The failure to find evidence of hyperbeta-lipoproteinemia in the remainder of the familystudied was somewhat surprising, but far frominconceivable on genetic grounds. However, sinceno other family members had hyperbetalipopro-teinemia, evidence of complete genetic segregationof these two biochemical disorders could not beobtained in this kinship. Evaluation of the fatherof the propositus could have provided definitiveevidence of such genetic segregation, but he diedin an accident in middle age without a clinicalhistory of cardiovascular disease.

There are many apparent interrelationships be-tween purine and lipoprotein metabolism whichare unanswered. An elevated serum urate, the ulti-mate end product of purine metabolism in man,frequently occurs in patients with hypertriglycer-idemia (Type III, IV, and V hyperlipoproteine-mia) (15, 16, 8) although this association is notusually noted in patients such as the propositus,with hyperbetalipoproteinemia (Type II) (17, 8).Beta lipoprotein release from the liver in rats canbe strongly inhibited by the administration oforotic acid, an effect which is completely reversedby exogenous adenine, suggesting that purines and

Adenine Phosphoribosyltransferase Deficiency 2287

pyrimidines may be involved in the regulation ofbetalipoprotein metabolism (18). Patients withfamilial Type II hyperlipoproteinemia have an in-creased quantity of a lipoprotein which is struc-turally normal (8), suggesting that the mutationresponsible for this abnormality acts by producinga defect in a normal control mechanism. Highnormal beta lipoprotein levels in the mother andnormal concentrations of beta lipoprotein in thetwo young daughters of the propositus, who alsohad reduced A-PRTase activity, suggest that theassociation of these two metabolic disorders mayhave been fortuitous in the propositus. It is possi-ble that the two girls with the enzyme defect willultimately manifest this disorder of lipoproteinmetabolism, but the failure of the mother of thepropositus to demonstrate this lipoprotein abnor-mality argues strongly against a causal relation-ship of these two disorders. Certainly the apparenthigh incidence of hyperbetalipoproteinemia in theAmerican population would not mitigate against achance association. The possibility still remainsthat hyperbetalipoproteinemia is expressed in asso-ciation with A-PRTase deficiency only in the male.A-PRTase activity, however, was normal in sixunrelated adult patients with familial Type IIhyperlipoproteinemia, suggesting that the defi-ciency of this enzyme, if causally related at all, isnot the basic defect in most patients with thisdisorder. Further studies are being conducted inthe propositus in an attempt to delineate this issue.

No attempt has been made to estimate frequencyof this mutation but the propositus was discoveredin the course of assay of blood samples from about150 subjects. The possible manifestations of thisenzyme defect in the homozygote are difficult topredict. However, it would seem unlikely to us atthis time that this would be a lethal mutation sincethe product of this reaction, adenylic acid, can beformed in most tissues by alternative biochemicalpathways. Furthermore, normal growth of mutantmammalian cells (AMK) completely deficient inthis enzyme, has been reported (19). On the otherhand, mature human erythrocytes which lack ade novo pathway for synthesis of purine ribo-nucleotides might be adversely affected by a verylow A-PRTase activity, particularly since thealternative pathways for the synthesis of adenylicacid from adenosine and inosinic acid are lackingin these cells (20). The possibility of finding a

more severe deficiency of A-PRTase among pa-tients with unexplained congenital hemolyticanemia remains to be explored.

ACKNOWLEDGMENTS

Wewish to thank the propositus, E. H., for his coopera-tion and active participation in obtaining the historicalinformation and appropriate blood samples on members ofhis family. We wish to thank Dr. Charles J. Epstein,previously of the National Institute of Arthritis andMetabolic Diseases, for his suggestions and specificallywish to acknowledge his role in proposing the modelby which a structural gene mutation could lead to a re-duction in enzyme activity to 25% of normal in aheterozygote.

REFERENCES

1. Kelley, W. N., F. M. Rosenbloom, J. F. Henderson,and J. E. Seegmiller. 1967. A specific enzyme defectin gout associated with overproduction of uric acid.Proc. Natl. Acad. Sci. U. S. 57: 1735.

2. Seegmiller, J. E., F. M. Rosenbloom, and W. N.Kelley. 1967. An enzyme defect associated with a sex-linked human neurological disorder and excessivepurine synthesis. Science. 155: 1682.

3. Rosenbloom, F. M., W. N. Kelley, J. F. Henderson,and J. E. Seegmiller. 1967. Lyon hypothesis andX-linked disease. Lancet. 2: 305.

4. Lesch, M., and W. L. Nyhan. 1964. A familial dis-order of uric acid metabolism and central nervoussystem function. Am. J. Med. 35: 561.

5. Hoefnagel, D., E. D. Andrew, N. G. Mireault, andW. 0. Berndt. 1965. Hereditary choreoathetosis,self-mutilation and hyperuricemia in young males.New Engl. J. Med. 273: 130.

6. Shapiro, S. L., G. L. Sheppard, Jr., F. E. Dreifuss,and D. S. Newcombe. 1966. X-linked recessive in-heritance of a syndrome of mental retardation withhyperuricemia. Proc. Soc. Exptl. Biol. Med. 122: 609.

7. Nyhan, W. L., J. Pesek, L. Sweetman, D. G. Car-penter, and C. H. Carter. 1967. Genetics of an

X-linked disorder of uric acid metabolism andcerebral function. Pediat. Res. 1: 5.

8. Fredrickson, D. S., R. I. Levy, and R. S. Lees. 1967.Fat transport in lipoproteins-an integrated ap-proach to mechanisms and disorders. New Engl. J.Med. 276: 34, 94, 148, 215, 273.

9. Smith, L. H., Jr., C. M. Huguley, Jr., and J. A. Bain.1966. Hereditary orotic aciduria. In Metabolic Basisof Inherited Disease. J. B. Stanbury, J. B. Wyngaar-den, and D. S. Fredrickson, editors. McGraw-HillBook Company, New York. 2nd edition. 739.

10. Total cholesterol procedure N-24. In AutoAnalyzerManual. 1964. Technicon Instrument Corporation.Chauncey, N. Y.

11. Kessler, G. and H. Lederer. 1965. Fluorometricmeasurement of triglycerides. In Automation in Ana-lytical Chemistry. p. 341. L. T. Skegg, Jr., editor.Mediad, Inc., 1966. New York.

2288 Kelley, Levy, Rosenbloom, Henderson, and Seegmiller

12. Fallon, H. J., M. Lotz, and L. H. Smith, Jr. 1962.Congenital orotic aciduria: demonstration of an en-

zyme defect in leukocytes and comparison with drug-induced orotic aciduria. Blood. 20: 700.

13. Krooth, R. S. 1964. Properties of diploid cell strainsdeveloped from patients with an inherited abnormalityof uridine biosynthesis. Cold Spring Harbor Symp.Quant. Biol. 29: 189.

14. Pinsky, L., and R. S. Krooth. 1967. Studies on thecontrol of pyrimidine biosynthesis in human diploidcell strains. I. Effect of 6-azauridine on cellularphenotype. Proc. Natl. Acad. Sci. U. S. 57: 925.

15. Feldman, E. B. and S. L. Wallace. 1964. Hypertri-glyceridemia in gout. Circulation. 29: 508.

16. Berkowitz, D. 1966. Gout, hyperlipidemia, and dia-betes interrelationships. J. Am. Med. Assoc. 197: 77.

17. Jensen, J., D. H. Blankenhorn, and V. Kornerup.1966. Blood-uric-acid levels in familial hypercho-lesterolaemia. Lancet. 1: 298.

18. Windmueller, H. G. and R. I. Levy. 1967. Totalinhibition of hepatic B-lipoprotein production in therat by orotic acid. J. Biol. Chem. 242: 2246.

19. Lieberman, I. and P. Ove. 1960. Enzyme studieswith mutant mammalian cells. J. Biol. Chem. 235:1765.

20. Lowy, B. A., M. K. Williams, and I. M. London.1962. Enzymatic deficiencies of purine nucleotidesynthesis in the human erythrocyte. J. Biol. Chem.237: 1622.

Adenine Phosphoribosyltransferase Deficiency 2289