liver galactose-l-phosphate uridyl transferase: activity...

Liver Galactose-l-Phosphate Uridyl Transferase:

Activity in Normal and Galactosemic Subjects

STANTONSEGAL, SHIRLEY ROGERS,and Puinp G. HOLTZAPPLE

From the Division of Biochemical Development and Molecular Diseases,Children's Hospital of Philadelphia, Philadelphia, Pennsylvania 19146, and theDepartment of Pediatrics, Medical School of the University of Pennsylvania,Philadelphia, Pennsylvania 19104

A B S T R A C T The kinetic characteristics of galactose-1-phosphate uridyl transferase have been determined inhomogenates of human liver biopsies obtained from con-trol subjects and in 50-fold purified enzyme preparationsfrom liver obtained at autopsy. A standardized assayprocedure employing linear kinetics was used to assessthe enzyme activity in homogenates of liver biopsy speci-mens from five control subjects and four patients withcongential galactosemia with demonstrated absence ofthe enzyme activity in red blood cells. Activity of con-trol specimens ranged from 11.8 to 17.2 mimoles ofUDPgalactose formed per min mg of protein. Liver oftwo galactosemic patients, both Caucasian, possessed nodetectable enzyme activity (less than 1-2% of normal).The tissue of two others, both Negro, who are knownto be capable of metabolizing intravenously administeredgalactose, contained easily detectable enzyme at approxi-mately 10% of the controls. No alternate enzymatic ac-tivity for formation of UDPgalactose was found in theliver of Negroes with galactosemia that was as active asthe residual galactose-l-phosphate uridyl transferase.The data suggest that the residual liver enzyme activityaccounts for the ability of Negroes with galactosemia tometabolize limited but significantly large quantities ofgalactose.

INTRODUCTIONThe disorder, congenital galactosemia, characterizedby cataracts, liver disease, and mental retardation, is dueto a deficiency of the enzyme galactose-l-phosphate uridyltransferase (UDPglucose:' a-n-galactose-l-P uridyltransferase [Enzyme Commission, 2.7.7.12]) (1). This

Received for publication 27 August 1970 and in revisedform 6 November 1970.

1Abbreviations used in this paper: UTP, uridine triphos-phate; UDPG, TDPG, CDPG, UDPGal, nucleotide sugars.

enzyme which catalyzes the reaction gal-l-P + UDP-glucose -4 glucose-l-P + UDPgalactose has been shownto be defective in a variety of tissues from galactosemicpatients, and has been studied from a kinetic point ofview in human erythrocytes (2), leucocytes (3, 4), cul-tured fibroblasts (4), and most recently, intestinal mu-cosa (5). Although the liver is the principle tissue fordisposal of galactose, and despite the fact that the defecthas been demonstrated in liver (6), there has been nodetailed kinetic characterization of human liver trans-ferase nor further exploration of the enzyme activity inliver of galactosemic patients.

Of particular interest has been the group of patientswith congenital galactosemia who despite absent red cellgal-i-P uridyl transferase, has been shown to metabolizesignificant amounts of galactose (7, 8). Since all of thesepatients have been Negro, the term "Negro variant" ofgalactosemia has been applied to this group (9). Ofseveral tissues from such a patient studied in vitro, onlyliver has been shown to oxidize galactose-l-"4C to 'CO2(10). In order to determine whether this oxidation isrelated to the presence of residual galactose-l-P uridyltransferase activity, we have assessed the level of thisenzyme in liver biopsies from galactosemic Negroes andCaucasians after first characterizing the kinetic proper-ties of the enzyme in liver of nongalactosemic subjects.Our results form the basis of this report.

METHODSSubjects. Pertinent clinical data on subjects from whom

liver biopsies were obtained are summarized in Table I. Sixpatients ranging in age from 4 months to 41 yr and withoutany history or stigmata of deranged galactose metabolismare considered control subj ects. Specimens were obtainedfrom two controls by punch needle biopsy (C with idio-pathic hypoglycemia and K who had cystinosis with aconsiderable degree of renal failure). Of the other four pa-tients from whom small wedge biopsies were surgically ob-

500 The Journal of Clinical Investigation Volume 50 1971

TABLE IPatient Clinical Characteristics and Activity of Liver Galactose-l-P- Uridyl Transferase

TransferaseSubject Sex Race Age Diagnosis Type of biopsy activity

Yr mpmoles/min/mgprotein

ControlsB F C 41 Cholecystitis Op 11.8S F C 37 Cholecystitis Op 17.2C F C 2 Hypoglycemia P 16.1K M C 9 Cystinosis P 14.7U F C 4 months Hypoglycemia Op 15.0Sm M N 42 Cerebral hemorrhage Autopsy 4.4

purified 327.0Ba M C 60 Perinephric abscess Autopsy 4.0

purified 225.0GalactosemicsW. W. M N 9 P 1.3M. W. M N 8 P 1.8C. W. F C 5 P nondetectableF. R. M C 2 P nondetectable

C, Caucasian; N, Negro.Op, operative; P, punch.Values are representative of at least three separate assays of liver homogenate activity.Methods are described in the text.

tained, two were undergoing gall bladder surgery, one,patient R, was being explored for recurrent Wilms tumorafter receiving actinomycin D and one, patient U, underwentpancreatic exploration for an islet-cell adenoma and sub-total pancreatectomy. None of these liver specimens washistologically abnormal.

Of the four galactosemic patients, two were Negro. Allhad absent red cell transferase consistent with the galacto-semic genotype. The Caucasian patients F. R. and C. W.had the usual clinical syndrome in the neonatal period. Anearlier clinical description of the Negro siblings M. W. andW. W. has been published (8). In more recent metabolicstudies, these brothers both demonstrated substantial abilityto metabolize small amounts of galactose-l-"C to "CO2 andbelong in the group termed the "Negro variant" of subjectswith galactosemia (9). Previously, liver tissue from patientW. W. had been shown in vitro to oxidize radioactive galac-tose to "CO2 (8). The liver tissue from all galactosemicpatients was examined histologically and found within nor-mal limits. All had normal liver function tests at the timeof biopsy.

The galactosemic patients had been on a galactose-freediet since the postnatal period except for patient W. W. whowas on a normal, milk-containing diet without symptomsfor the 1st yr of life when he was found to have absent redcell transferase. The control patients were on regular hos-pital diets and no attempt was made to assess the galactosecontent.

For enzyme purification, approximately 200 g of liver wasobtained at autopsy from two adult male patients within 6hr of death. These patients had no history of liver diseaseand livers exhibited no significant pathology. Sm, a Negromale, age 42, died from massive cerebral hemorrhage andBa, a 60 year old Caucasian male, died from a perinephricabscess and septicemia.

Handling of tissue and assay of Gal-i-P uridyl transferase.Punch biopsies weighing between 25 and 35 mg and wedgebiopsies weighing about 20 mg were kept at 4°C for 10-20min until homogenized in 20 vol of 20 mmKCl. After spin-ning at 30,000 g for 30 min the supernatant was dividedinto 0.1 ml aliquots which were frozen in glass vials andkept at - 44°C until used. The homogenate enzyme activitywas stable on storage at - 44°C for 6 months.

Liver obtained at autopsy was frozen and stored at - 44°Cuntil partially purified with minor modifications by themethod of Mayes and Hanson (11). The steps consisted ofhomogenization in 20 mmKCl, precipitation with ammoniumsulfate at 50% saturation, dialysis against distilled water,adsorption and elution from calcium phosphate gel, and chro-matography of the gel eluate on a DEAE-cellulose columnemploying elution with batches of 20 and 80 mMphosphatebuffer. The specific activity of the enzyme before and afterpurification is shown in Table I.

The determination of enzyme activity was performed bythe method of Bertoli and Segal (12) except that the alka-line phosphatase step 'was not used prior to chromatographyof the incubation mixture. The assay is based on the stoichio-metric formation of UDPgalactose-"C during the incubationof the enzyme with galactose-l-P-"'C and unlabeled UDPGaccording to the reaction Gal-l-P-"C + UDPG-4 UDPGal-'C + G-l-P. After incubation the labeled gal-l-P is sepa-rated from the radioactive product by chromatography onDEAE-cellulose ion exchange paper with 0.05 M LiCl assolvent. The product radioactivity was determined by count-ing the area of the paper containing the UDPgalactose-1'Cwith a liquid scintillation spectrometer employing Liquifluor(New England Nuclear Corp., Boston, Mass.) as the scin-

tillator. The amount of product formed can be calculatedusing the specific activity of the radioactive substrate.

Human Liver Galactose-1-Phosphate Uridyl Transferase 501

200

usE

0E

E 50

2000

1500

1000

500

- ~~~~~~~~~~~~~~~~~~~~~~A/

*~~~~~~Homogenate Normal

g / ~~~~~0Homogenate W.W.sow *~~~~~~~Hornogenate F.R.

--- &-- iR r-, v2 4 6 8 10 12 14 16

T IME (minutes )

FIGURE 1 The effect of incubation time on galactose-l-phosphate uridyl transferase activity of human liver. Theright ordinate refers to purified enzyme (10 jAg protein),the left to biopsy homogenates (50 /Ag protein).

The standard assay procedure was as follows: The incu-bation mixture contained a final concentration of 0.2 mmUDPglucose, 0.25 mmgalactose-l-P (0.0056 jiCi), 10 mmmercaptoethanol, 100 mmglycine buffer pH 8.2 in a totalvol of 200 ul and a final pH of 8.2. The reaction was startedby the addition of 20 1A1 of homogenate containing about 50Ag protein and carried out for 5 min. When partially puri-fied enzyme was studied 2 /Ad containing 10 fAg protein wasadded and incubated 2.5 min. The reaction was stopped byboiling for 90 sec and 20 pul of the reaction mixture wasspotted for chromatography. The specific activity of theenzyme is expressed as millimicromoles of UDPgalactoseformed per minute per milligram soluble protein.

Other assay procedures. UDPgalactose pyrophosphory-lase was assayed both by the technique of Abraham andHowell (13) and by the standard procedure reported herefor transferase except that 0.25 mmUTP was added as sub-strate in place of UDPG. Protein was estimated by themethod of Lowry, Rosebrough, Tarr, and Randall (14).

Materials. Alpha-D-galactose-l-phosphate, UDPG, UTP,UDPGal, TDPG, and CDPGwere products of Calbiochem,Los Angeles, Calif. Alpha-D-galactose-l-phosphate-'4C (uni-formly labeled, SA 13-20 mCi/mmole) was purchased fromInternational Chemical and Nuclear Corp., City of Industry,Calif. Whatman No. DE81 was the DEAE-cellulose ion ex-change paper employed.

RESULTS

Kinetic characteristics of normal enzyme. Initial ve-locity studies of the partially purified enzyme and wholehomogenate are shown in Figs. 1 and 2. The amount ofUDPgalactose formed by the purified preparation using10 sAg of protein was linear for about 3 min of incubationwhile whole homogenate was linear for 10 min when 50Ag of protein was used. On 2.5 min of incubation puri-fied enzyme produced UDPgalactose linearly up to 15 iLgof protein, and on 5 min of incubation whole homogenatesdemonstrated proportionality to protein, added up to 100i'g of protein. Therefore, the assay procedure for purifiedenzyme utilized, employed 2.5 min incubation and about10 ,g of protein, that for crude homogenate 5 min incuba-tion and about 50 iyg of protein.

Transferase activity measured in the pH range from8 to 9 revealed a broad optimum between 8.2 and 8.6 forthe purified preparation (Fig. 3). Essentially the sameresults were found with crude homogenate.

The relationship of substrate concentration to enzymeactivity is shown in Fig. 4 by a typical Lineweaver-Burkplot of the data obtained with both whole homogenate andpurified enzyme. Table II summarizes the Km and Vmaxvalues for UDPGwhen Gal-l-P was 0.25 mmand forGal-l-P when UDPGwas 0.20 mmfor the various en-zyme preparations studied. Km values were relativelyuniform in both homogenates and purified preparation.Vra. in the various homogenates was similar except forsubject Rwho had been given actinomycin D. The differ-ence in maximum velocity for purified and crude enzymereflects the approximately 50-fold purification obtained.

Various nucleotide sugars such as TDPG, CDPG, andUDPGal were tested at 0.20 mmas substrates in placeof UDPG for reaction with labeled Gal-l-P. In ourstandard incubation only UDPGal resulted in a la-beled compound migrating as a sugar nucleotide onDEAEpaper chromatography. The velocity of this reac-tion, Gal-l-P-w"C + UDPGal > UDPGal-14C + Gal-l-P,was concentration dependent as shown in Fig. 5. TheKmand Vmra of UDPGal when Gal-l-P was 0.25 mmareshown in Table II. These values are essentially the sameas those found for UDPG. No labeled nucleotide could beshown to be formed when UTP at 0.25 mmwas sub-stituted for UDPGas substrate with the 5 min incuba-tion time of the standard assay. On prolonged incuba-tions of crude homogenates for 20 min, however, smallamounts of labeled nucleotide sugar presumably UDP-galactose were formed at a rate of about 1.5 mumole/min per mgprotein.

5

c

E

0)E

E

4

2

* PurilA Homro Homr

I~~~~~~~~~~~~~~~

ified Enzymetogenate Normal -

togenote W.W.

0.5

0.4

0.3

0.2

0.1

20 40 60 80 100 120agrams Protein

FIGURE 2 The effect of protein concentration on enzymeactivity. The right ordinate refers to homogenate W. W.activity (5 min incubation) while that on the left refers topurified enzyme (2.5 min incubation) and normal homogenate(5 min incubation).

502 S. Segal, S. Rogers, and P. G. Holtzapple

I

150[

100[

Transferase activity in galactosemic patients. Thetransferase activity, measured under standard conditions,of liver biopsies obtained from five control patients andfour galactosemic subjects is shown in Table I. Levelsin controls ranged from 11.8 to 17.2 m/Amoles/min per mgprotein. In two galactosemic patients, both Caucasian,there was no detectable activity while in both Negroeswith galactosemia activity was observed at approxi-mately 10% of the control. The enzyme activity ob-served in the liver of the two Negro siblings was de-pendent on addition of UDPGto the homogenates; nodetectable labeled nucleotide sugar was formed withoutthis substrate, even on prolonged incubations of 20 min.The homogenates of liver, obtained at autopsy performedwithin 6 hr of death, have about a quarter of activity infresh biopsy specimens.

The formation of labeled product in homogenates fromthe two Negro galactosemic patients increased linearlywith time as shown in Fig. 1. 20-min incubation valuesnot shown on the graph fell on the line if extended. Ac-tivity was proportional to protein as seen in Fig. 2.Galactosemics C. W. and F. R. revealed no detectableliver enzyme activity on incubation up to 20 min, norwas activity seen on adding larger amounts of homoge-nate. Insufficient homogenate was available from galac-tosemic patients W. W. and M. W. to assay accuratelysubstrate concentration dependence. Liver from galac-tosemic patients M. W. and W. W. which had measur-able transferase activity under standardized conditionsalso revealed similar activity when UDPGal was sub-stituted for UDPGthus behaving qualitatively like thenormal.

A

161-

2 Fv

SI

4

B

8 16 24

[Gal-I- P1

32 40

201-

16/Yv

A

12[

8

8 16 124 32

[UDPG]

40

FIGURE 4 A. The concentration dependence of purified en-zyme (0) and biopsy homogenate (A) on galactose-1-phosphate (Gal-l-P). B. Concentration dependence of thesame enzyme preparations on UDPglucose (UDPG). v isexpressed as mltmoles/min per mgX 10' for the purifiedenzyme and m/Amoles/min per mgX 102 for the homogenate.

The possibility of the presence of alternate routes offormation of galactose containing sugar nucleotides was

* Purified Enzyme - 650 investigated in the liver biopsies of the galactosemicA Homogenate

children. An attempt was made with the method of600 Abraham and Howell (13) to assay UDPGal pyrophos-

phorylase which catalyzes Gal-1-P interaction with UTP18 - - 550 to form UDPGal plus pyrophosphate (15). The use of

E 50 Ag of liver protein in which transferase activity was.' 16 - - 500 easily detectable was insufficient to detect UDPGal pyro-E". / \ \phosphorylase either in normal liver or in that from pa-O 14 - - 450 tient M. W. Substitution of UTP at 0.25 mmfor UDPG0E in our assay for transferase was also employed as aE 12 - 400 UDPGal pyrophosphorylase assay. No radioactive

UDPGal was formed by any of the homogenates oflot 350 galactosemic or normal livers with 5 min of incubation.

7.0 7.5 8.0 8.5 9.0 9.5 After 20 min of incubation, radioactive sugar nucleotidepH formation was detected at an average rate of 1.5 m/Amole/

FIGURE 3 The pH dependence of liver galactose-1-phosphate min per mg in three control specimens while that fromuridyl transferase. 100 mmglycine buffer at the pH's shownwas employed in the standard assay. The right ordinate patient W. W., a Negro galactosemic patient, gave arefers to purified enzyme, the left to the homogenate. value of 0.5 mymole/min per mg.


TABLE I IKm and V.. Values of Human Liver Galactose-l-P- Uridyl

Transferase

Enzyme Substrate*

Gal-i-P UDPG UDPGal

Kn4 Vma§ K. VMs Km Vmn

HomogenateS 0.10 18 0.26 20 0.35 22B 0.11 15 0.16 22 1.0 50R 0.07 6 0.05 6C 0.11 26 0.08 31 0.20 40

Partially purifiedSm 0.30 1117 0.31 1111Ba 0.25 1000 0.20 1000

* In homogenate experiments when Gal-i-P was the variablesubstrate UDPGwas 0.20 mM. WhenUDPGor UDPGal werevaried Gal-i-P was 0.25 mM. Whenpurified enzyme was usedand Gal-i-P was variable UDPGconcentration was 0.25 mMand when sugar nucleotide was the varied substrate Gal-i-Pwas 0.35 mM.$ Km is expressed in mM.

§ V.. is expressed as mtzmoles/min per mg protein.

To examine the possibility that transferase in galac-tosemic liver may be altered thereby requiring a sugarnucleotide other than UDPG, we substituted thymidinediphosphoglucose for UDPGin the assay of galactosemicliver. No TDPGal was formed after a 5 min incubation.With a 20 min incubation, three normal livers formeddetectable nucleotide sugar at a velocity of 1.2 m/Amoles/min per mg, while galactosemic liver from patient F. R.formed none and that from patient W.W. formed 2mymoles/min per mg.

DISCUSSION

Much of our understanding of congenital galactosemiahas resulted from the study of galactose-l-phosphateuridyl transferase in peripheral blood cells wherein thedeficiency of the enzyme was first noted (1). Anderson,Kalckar, and Isselbacher have shown that transferaseactivity was also deficient in liver (6). Until recentlytheir data seemed sufficient for interpreting the meta-bolic derangement in galactosemic patients. The demon-stration that certain galactosemic patients with absentred cell transferase can metabolize administered galac-tose (7) has prompted a reinvestigation of liver trans-ferase activity in both normals and galactosemic patients.In the present work, we have analyzed in detail thekinetic parameters of transferase from the livers of non-galactosemic controls and have shown the similaritiesof partially purified enzyme to biopsy homogenates.The characteristics of transferase activity in normal

human liver are similar to those observed previously withrat liver (12).

Our values for liver transferase differ from those ofAnderson et al. whose data until now were the only dataavailable about the human enzyme (6). Although theradioactive assay used was similar in principle to thatdescribed here, no kinetic analysis of the conditions forthe assay was reported. The limit of sensitivity in theassay of Anderson et al. was said to be 0.3 Amoles/gmliver per hr while that of our assay is 0.25 mttmoles/mgliver per hr), (after converting from our velocity inmimoles/mg protein per min), a difference factor of1000. This difference is no doubt due to the nonlinearkinetics of their assay in which 7-15 mg of tissue and anincubation time of 30 min was employed. The differenceis also due to the addition of UTP to the earlier assaysystem since we have recently found UTP to be a potenttransferase inhibitor.2

Comparison of the transferase activity in liver of fourgalactosemic subjects with that of nongalactosemic con-trols has revealed that two galactosemic patients, Negrosiblings, possess easily detectable amounts of enzyme inliver amounting to 10% of the normal, while two Cau-casian with galactosemia had nondetectable levels (thatis, less than 1-2% of the normal). Of the two liver ho-mogenates assayed by Anderson et al. (6), one was anadult Negro male who had easily detectable enzymeactivity while the other was a Caucasian infant withoutdetectable enzyme.' The latter investigators felt that thefinding of detectable enzyme in the Negro adult was re-lated to his older age. Subsequent studies have indicatedthat being a Negro may be a factor in the ability ofgalactosemic patients to metabolize galactose (7, 9).Enzyme activity would not be expected to "mature"with age since developmental studies of rat liver trans-ferase have indicated that transferase has the highestspecific activity in the young liver and subsequently de-creases with age (12).

The Negro adult who had detectable enzyme estimatedat 5% of normal (6), was later extensively studied forhis metabolic capability and was found to be able toeasily metabolize galactose-l-"4C to 14CO2 (7, 9). Subse-quently, his liver biopsy specimen was found in vitro tooxidize galactose--1T4C to 14CO2 at a near normal rate andextent. Patient W. W. investigated here was shown tohave a galactose metabolic capability at age 3, and a

'Unpublished observations of the authors.'Although the race of the patients was not stated in the

paper by Anderson, Kalckar, and Isselbacher, the race ofthe adult is known by one of the authors, who was at Na-tional Institutes of Health at the time, and who subse-quently studied the same patient (7). The race of the infantwas made known to the present authors by Dr. George Don-nell of the Los Angeles Childrens Hospital, who sent thebiopsy to Dr. Kalckar.


punch liver biopsy at that time was capable in vitro ofgalactose oxidation (8). The ability of both Negro pa-tients W. W. and M. W. to oxidize galactose-l-"C wastested at the time of the liver biopsy for transferase ac-tivity and both were found able to rapidly and extensivelyconvert the radioactive substrate to 14CO2. The observa-tions that Negro galactosemic patients convert galactose-l-'4C to blood glucose (7), have inhibition of galactoseoxidation by ethanol (7, 16), and produce '4CO2 fromboth C-1 and C-2 labeled galactose in a normal pattern(17), are consistent with operation of the sugar nu-cleotide pathway of galactose metabolism. The findingsthat Negroes with galactosemia have residual transferasein the liver imply that it is this enzyme, though limitedin amount, which accounts for their ability to metabolizegalactose. This is consistent with our observations thatthe capacity of the Negro galactosemic to metabolizeintravenously administered galactose is limited, beingnearly normal for quantities up to only 2 g (7).

An alternate pathway involving a nucleotide sugar asan explanation of the galactosemic Negro's ability tometabolize galactose could be considered. The UDPGalpyrophosphorylase reaction in which Gal-l-P + UTPUDPGal + PP was postulated by Isselbacher to servethis function (15). Such a reaction would circumvent thetransferase step and satisfy the metabolic observationsenumerated earlier. Recently, Abraham and Howell stud-ied this enzyme activity in autopsy specimens of humanliver (13). Employing a standardized assay, they showedthat the specific activity of the enzyme was about onehundredth that of Gal-l-P uridyl transferase, there beingno increase with age as suggested by Isselbacher's ratexperiments (15). Indeed, the activity is so low as toraise the question of its functional significance. Knopand Hanson, who studied the substrate specificity ofhuman liver crystalline UDPglucose pyrophosphorylase,reported that what has been called UDPGal pyrophos-phorylase activity is in reality a small degree of reac-tivity of UDPGpyrophosphorylase with UDPGal (18).Ting and Hanson have suggested caution in attributingsignificance to the pyrophosphorylase route for metabo-lism of galactose (19). Nonetheless, we have looked forthis activity in our biopsies of galactosemic and normalsubjects. In short incubation times during which trans-ferase is very active in normal subjects and easily de-tectable in galactosemic Negroes, no UDPGal pyrophos-phorylase activity was seen. It is our feeling that UDPGalpyrophosphorylase activity is unlikely to explain therapid and quite proficient galactose metabolism seen inNegroes with galactosemia. Likewise, we have observedno other transferase activity employing another sugarnucleotide such as TDPGthat might adequately explainthe metabolic observations.

The delineation of a difference between Negro andCaucasian patients with galactosemia by the presence or

281

24[

IYV20[

16

8

4

0 8 16 24[

lUDPGal]

32 40

FIGURE 5 Effect of UDPgalactose (UDPGal) concentra-tion on enzyme activity. (0) purified enzyme with v ex-pressed as mumoles/min per mgX 10' and homogenate (A)with v expressed as m~umoles/min per mgX 10-2.

absence of liver transferase activity, appears to be paral-leled by similar finding with intestinal mucosa (5).About 10% of the normal intestinal Gal-l-P uridyl trans-ferase was found to exist in mucosal biopsies from threeNegroes with galactosemia, two of whomare subjects inthe present paper. This would appear to reflect the com-mon embryonic origin of the two tissues. The exact con-tribution of gastrointestinal tissue to the disposition ofgalactose, however, in the normal individual is notknown. Although the liver is thought to be the primeorgan for normal galactose disposal, it may well be thatthe liver and intestinal mucosa together provide the trans-ferase complement which enables the galactosemic Ne-gro to metabolize some galactose.

The exact explanation for possession of transferaseactivity by visceral tissue when other tissues of theNegro patients are deficient, remains to be explained.6 Negroes are in the group of 18 subjects with galac-tosemia that we have studied. All five of the six who havebeen tested have metabolized galactose to a significantdegree. In Hsia's group of 45 patients (20), two sib-lings are identified as Negroes, one of whom has beenwithout symptoms of galactose toxicity while ingestinggalactose (resembling our patient W. W.). Donnell'sclinic of 41 subjects with galactosemia includes twoNegroes (21). None of the patients in the latter twogroups has had a quantitation of galactose metabolismbut there is a clinical indication that the Negro in Hsia'sgroup has a galactose metabolic capability. Althoughthere appears to be a racial correlation with an ability


I I I

I2-

to metabolize galactose and the possession of some livertransferase, the hereditary basis for this difference fromother galactosemic patients has not been established.Kalckar' has suggested the possibility a provirus may beprogramming the formation of transferase in liver cellsof these individuals.

ACKNOWLEDGMENTSThe authors wish to express their appreciation to Dr. LesterBaker for permitting us to study his patients.

This work was supported by Research grant AM 10894,Training grant HD00215, and Clinical Research Centergrant FR240 from the National Institutes of Health. Dr.Holtzapple was a trainee in the Genetic and Metabolic Dis-ease Training Program.

4Kalckar, H. M. Personal communication.

REFERENCES1. Kalckar, H. M., E. P. Anderson, and K. J. Isselbacher.

1956. Galactosemia, a congenital defect in a nucleotidetransferase. Biochim. Biophys. Acta. 20: 262.

2. Beutler, E., and M. C. Baluda. 1966. Biochemical prop-erties of human red cell galactose-1-phosphate uridyl trans-ferase (UDP glucose: a-D-galactose-1-phosphate uridyltransferase E.C.2.7.7.12) from normal and mutant sub-jects. J. Lab. Clin. Med. 67: 947.

3. Inouye, T., H. L. Nadler. and D. Y. Y. Hsia. 1969. Somecharacteristics of Gal-1-P uridyl transferase of humanwhite blood cells. In Galactosemia. D. Y. Hsia, Editor,Charles C Thomas, Publisher, Springfield, Ill. 105.

4. Tedesco, T. A., and W. J. Mellman. 1969. The UDPGluconsumption assay for Gal-i-P uridyl transferase. InGalactosemia. D. Y. Hsia, Editor, Charles C Thomas,Publisher, Springfield, Ill. 66.

5. Rogers, S., P. G. Holtzapple, W. J. Mellman, and S.Segal. 1970. Some characteristics of galactose-l-phosphateuridyl transferase in intestinal mucosa of normal andgalactosemic humans. Metabolism. 19: 701.

6. Anderson, E. P., H. M. Kalckar, and K. J. Isselbacher.1957. Defect in uptake of galactose-1-phosphate intoliver nucleotides in congenital galactosemia. Science(Washington). 125: 113.

7. Segal, S., A. Blair, and H. Roth. 1965. The metabolismof galactose by patients with congenital galactosemia.Amer. J. Med. 38: 62.

8. Baker, L., W. J. Mellman, T. A. Tedesco, and S. Segal.1966. Galactosemia: symptomatic and asymptomatic ho-mozygotes in one Negro sibship. J. Pediat. 68: 551.

9. Segal, S. 1969. The Negro variant of congenital galac-tosemia. In Galactosemia. D. Y. Hsia, Editor, Charles CThomas, Publisher, Springfield, Ill. 176.

10. Topper, Y. J., L. Laster, and S. Segal. 1962. Galactosemetabolism: phenotypic differences among tissues of apatient with congenital galactosaemia. Nature (London).196: 1006.

11. Mayes, J. S., and R. G. Hanson. 1967. Galactose-1-phos-phate uridyl transferase. In Methods in Enzymology. 9:708.

12. Bertoli, D., and S. Segal. 1966. Developmental aspectsand some characteristics of mammalian galactose 1-phos-phate uridyl-transferase. J. Biol. Chem. 241: 4023.

13. Abraham, H. D., and R. R. Howell. 1969. Humanhepatic uridine diphosphate galactose pyrophosphorylase.J. Riol. Chem. 244: 545.

14. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J.Randall. 1951. Protein measurement with the folinphenol reagent. J. Biol. Chem. 193: 265.

15. Isselbacher, K. J. 1957. Evidence for an accessory path-way of galactose metabolism in mammalian liver. Science(Washington). 126: 652.

16. Segal, S., and A. Blair. 1961. Some observations on themetabolism of D-galactose in normal man. J. Clin. In-vest. 40: 2016.

17. Segal, S., and P. Cuatrecasas. 1968. Oxidation of 14Cgalactose by patients with congenital galactosemia: evi-dence for a direct oxidative pathway. Amer. J. Med. 44:340.

18. Knop, J. K., and R. G. Hansen. 1970. Uridine diphosphateglucose pyrophosphorylase. Crystallization and proper-ties of the enzyme from human liver. J. Biol. Chem. 245:2499.

19. Ting, W. K., and R. G. Hansen. 1968. Uridine diphos-phate galactose pyrophosphorylase from calf liver.Proc. Soc. Exp. Biol. Med. 127: 960.

20. Hsia, D. Y. 1967. Clinical variants of galactosemia.Metabolism 16: 419.

21. Donnell, G. N., R. Koch, and W. R. Bergren. 1969.Observations on results of management of galactosemicpatients. In Galactosemia. D. Y. Hsia, Editor. Charles CThomas, Publisher, Springfield, Ill. 247.


liver galactose-l-phosphate uridyl transferase: activity...

Documents