- Email: [email protected]
Rennes-like variant of galactosemia: Clinical and biochemical studies
50
July 1975 The Journal o f P E D I A T R I C S
Rennes-like variant of galactosemia: Clinical and biochemical studies The first recognized case of a Rennes-like variant form of galactosemia in a Caucasian individual is described. Galactose-l-phosphate uridyl transferase activity was approximately 10% of the normal in both erythrocytes and cultured skin fibroblasts. Electrophoretic mobility of the variant enzyme in erythrocytes was slower than that of normal individuals and identical to that of the two cases originally reported from Rennes, France. In normal cultured skin fibroblasts, four transferase bands were found. In this tissue, the patient again had a slower moving transferase. It is proposed that in transferase variants an altered subunit results in a specifically altered enzyme mobility analogous for each tissue.
Gerhard Hammersen, M.D., Sally Houghton, and Harvey L. Levy, M.D.,* Boston, Mass.
GALACTOSEMIA, an inborn error of galactose metabo-
lism, was possibly first described as early as 19081 and almost certainly described by Goeppert 2 in 1917. This entity, now known as "classical" galactosemia, has been frequently diagnosed and is characterized in the acute stage by galactosuria, hypoglycemia, neonatal jaundice, weight loss, vomiting, diarrhea, and even neonatal death and, in the chronic stage, by cataracts, cirrhosis, and mental retardation? The enzymatic defect, that of galactose-l-phosphate uridyl transferase (transferase), is at the second degradative step for galactose (Fig. 1). This enzyme, as well as other enzymes of galactose metabolism, occurs in many body tissues, including erythrocytes? Because of the accessibility of such cells, transferase has been extensively studied. These studies have led to the recognition that transferase deficiency is not a single disease but is associated with several genetiFrom the State .Laboratory Institute, Massachusetts Department o f Public Health, the Neurology Service, and the Joseph P. Kennedy, Jr., Laboratories, Massachusetts General Hospital, and the Department o f Neurology, Harvard Medical School. Supported in part by Deutsche Forschungsgemeinschaft Grant HA 855/1, USPHS H S M H A M C H Grant 0I-H-000111-04-0, N i H Grant N S 05096, and National Foundation Grant CRBS-260. *Reprint address: State Laboratory Institute, 305 South St., Boston, Mass. 02130.
Vol. 87, No. 1, pp. 50-57
cally distinct entities. Each of these entities appears to be characterized by a variant enzyme that differs from the normal enzyme and other variant enzymes in degree of activity and, in most instances, electrophoretic mobility. 4-8 The tendency for clinical complications to develop in these variants varies from apparent clinical normality in the relatively common Duarte type, 4 to perhaps mild symptoms in the "Negro" variant, 8 and to the severe galactosemia syndrome reported in the "classical, ''3 Indiana, 7 and Rennes variants? Abbreviation used UDPG: uridine diphosphoglucose The recognition that the symptoms of galactosemia can be prevented by the early institution of a galactose-free diet has led to grea t interest in the neonatal detection of these disorders? Such detection has been greatly aided by the development of relatively simple assays for either galactosel0. 11 or transferase 12 that can be applied to the filter paper blood specimen that is submitted for routine phenylketonuria testing. In the application of either or both methods of neonatal screening it is predictable that infants with the known variants plus infants with as yet undiscovered variants will be identified. For the purpose of therapy as well as for a better understanding of this important and complex group of disorders, it is important that infants with any form of transferase deficiency be studied as completely as possible.
Volume 87 Number 1
Rennes-like variant o f galactosemia
51
c ta a c t o s e - l - P h o @ s p h a t e Uridyl Transferase DG l u c o s e - l - P h o s p h a t e
Galaetose~lucose Monophosphate Shunt
UDP
Fig. 1. Pathway o f gatactose metabolism. The known h u m a n disorders in this pathway are indicated by the n u m b e r e d enzyme deficient in each disorder. T a b l e I. E n z y m e
t r a n s f e r a s e ) s p o t s c r e e n i n g test r e s u l t s a n d g a l a c t o s e c o n c e n t r a t i o n s i n p a t i e n t
Enzyme results* Specimen
Age (days)
Diet
3 hr
16 hr
Blood (umbilical cord) Blood Blood Blood Urine Blood Urine
Birth 3 21 24 24 28 28
None Lactose:~ Lactose Lactose Lactose Lactose-freew Lactose-free
----
+ + + + + + Not done
-
+ + Not done
Galactoset (mg/dO Not done 4 > 50 2 > 50 Negative Negative
*Results in this test are recorded in intensity of fluorescence varying from no fluorescence (-) to good fluorescence (+ + + +). Normally, fluorescence is good (+ + + +) after 3 hours incubation. "~Normally there is no detectable galactose in blood by this method. In our experience, urine galactose concentrations in the normal newborn infant vary from less than 2 mg/dl to 20 mg/dl, depending on the length of time of urine collection following food ingestion. :~Similac (Ross Laboratories). w (Mead-Johnson Laboratories). I n this r e p o r t
we p r e s e n t clinical a n d
biochemical
s t u d i e s o f a n i n f a n t w h o h a s a v a r i a n t t r a n s f e r a s e w h i c h in e l e c t r o p h o r e t i c m o b i l i t y a n d e n z y m e activity is i d e n t i c a l to t h e R e n n e s v a r i a n t ? T h e t r a n s f e r a s e a b n o r m a l i t y w a s d e m o n s t r a b l e b o t h in e r y t h r o c y t e s a n d in c u l t u r e d s k i n fibroblasts. CASE REPORT A Caucasian boy was born after a term pregnancy to a gravida 2, para 2, 25-year-old woman. The birth and neonatal period were normal. In particular, there was no jaundice or feeding problems. He was first seen by us at the Massachusetts General Hospital when he was 389 weeks old because of an abnormal result in the Beutler enzyme spot screening test performed routinely on filter paper specimens o f blood obtained from the umbilical cord and in the newborn nursery. The initial blood specimen and a repeat blood specimen obtained when he was 3 weeks old had revealed galactose elevations o f 4 m g / d l and > 50
mg/dl, respectively (Table I). On the first visit he was a healthy alert newborn infant with no jaundice or other signs of hepatic disease or s y m p t o m s of milk intolerance despite the ingestion since birth of a lactose-containing proprietary formula (Similac, Ross Laboratories, Columbus, Ohio). Laboratory determinations at this time revealed a blood galactose concentration of only 2 m g / d l but a urine galactose concentration of 60 to 80 mg/dl. Because of the galactose accumulations in this infant, the regular lactose-containing formula was discontinued and, at age 3~A weeks, he was fed a non-lactose-containing proprietary formula (Nutramigen, Mead Johnson Laboratories, Evansville, Ind.), which is still maintained. He is now a healthy 14-month-old infant who is normal developmentally and has no hepatomegaly or other signs of illness. Ophthalmologic examination revealed no evidence of cataracts or other abnormalities. The parents are nonrelated and of Irish/Irish (mother) and Irish/Italian (father) ancestry. There is no family history of mental retardation, liver disease, neonatal death, or early cataracts.
52
Hamrnersen, Houghton, and Levy
MATERIALS
AND METHODS
Screening for galactose-l-phosphate uridyl transferase deficiency was performed by the enzyme spot screening test fbr Beutler and BaludaJ 2 Ultraviolet fluorescence is an indication of the presence of enzyme activity. In a slight modification of the originally described method that we apply to blood specimens of newborn infants, incubation is continued at room temperature for 16 hours when no fluorescence is detected after three hours' incubation at 37 ~ C. By this procedure the specimen with no detectable activity can be differentiated from the more numerous specimens with very low activity, frequently encountered as a result of artifact during the summer months, '~ or as noted in true variant states. Measurement of galactose in blood and urine for the purposes of routine screening and for determinations in specific specimens was performed by the bacterial assay of Paigen and Pacholec. ~ This assay, as yet unpublished, utilizes discs from filter paper specimens identical to those used in the Guthrie tests .~ and measures galactose by the growth of a uridine diphosphogalactose-4-epimerase (epimerase)-deficient strain of Escherichia coli, which resists destruction by bacteriophage in the presence of galactose. This assay has been used in testing several hundred thousand specimens in the Massachusetts and Other screening programs for newborn infants ~ and is highly reliable and sensitive. Routine blood and urine screening for amino acid abnormalities was performed by the bacterial inhibition assays for phenylalanine, leucine, methionine, and tyrosine 9 and by unidimensional paper chromatography. TM Further amino acid study was performed by sequential two-day chromatography of urine '~ and by automated ion-exchange column chromatography of blood and urine. 16 To determine the tolerance to galactose in the patient, a proprietary,formula containing lactose (Enfamil, Mead Johnson Laboratories, Evansville, Ind.) was fed after an overnight fast. The patient ingested 96 m l of the formula, which contains 0.78 gm of galactose (in the form of lactose) per kilogram of body weight. Whole milk or proprietary formula containing equivalent amounts of lactose (galactose) were ingested by one infant with "classical" galactosemia, another infant with the mixed heterozygotic pattern for "classical" and the Duarte variant form of transferase deficiency, and infants with normal transferase values. Blood specimens were obtained in the fasting state and at intervals following ingestion. Galactose-l-phosphate uridyl transferase activity was measured in hemolysates by the uridine diphosphoglucose consumption method of Beutler and Baluda. '7
The Journal of Pediatrics July 1975
Hemolysates from newborn infants were preincubated with NADase (501~1 N A D a s e / m l hemolysate: NADase: 0.63 units/ml) to avoid artifacts due to epimerase activityl8
Electrophoretic mobility of transferase in hemolysates was determined by starch-gel electrophoresis as described bY Ng and associates. 19 In order to obtain visible activity bands, the hemolysate of the patient was not diluted with distilled water prior to freezing and thawing. Similar nondiluted hemolysate preparations were used for 2 of the 20 control subjects. Fibroblasts were grown from punch skin biopsies of the patient, his parents, two galactosemic patients, and normal control subjects. Cells were grown in Eagle's medium, supplemented with 15% fetal calf serum and "nonessential" amino acids. After ten days o[ subculture, the confluent monolayers were washed with phosphatebuffered saline and trypsinized. Cells were counted, centrifuged, washed twice with isotonic cold saline, and resuspended in distilled water at 2 to 6 x 107 cells/ml, and then frozen in a Dry-lce bath and thawed three times. The lysate was centrifuged 20 minutes at 700 x g and the supernates were used for the U D P G consumption assay and for starch-gel electrophoresis. In these fibroblast lysates transferase activity was determined by the U D P G consumption assay of Tedesco and Mellman. 2~To inactivate epimerase, 0.05 ml lysate was preincubated with 0.5M (0.05 ml) glycylglycine buffer, pH 8.7, and 0.1M (0.05 ml) dithiothreitol. ~1 RESULTS Erythrocyte transferase activity. The results of the enzyme spot screening test on filter paper specimens of blood are listed in Table I. Transferase activity was not detectable in any specimen after the usual incubation duration of three hours or even after incubation at 37 ~ C for five hours but was detectable in almost every specimen after incubation at room temperature for an additional 16 hours. Thus it was evident from the initial test results that the patient did have detectable erythrocyte transferase activity. This slight activity was confirmed by the U D P G consumption assay. As indicated in Table II, hemolysates from the patient when he was 3~ weeks old contained 0.65 units* of transferase activity and thereafter never more than 2.0 units of activity. This is in contrast to the normal control value of 22.0 _ 3.7 and to the virtually undetectable activity in "classical" galactosemic infants. Blood galactose concentrations. Blood galactose concentrations in the patient (Table I) varied from the only *One unit of transferase activityin hemolysatesis equal to 1 /~rnole UDPG consumedper hour per gram of hemoglobin.
Volume 87 Number 1
Rennes-like variant o f galactosemia
//
53
t!
50 E o 40 E uJ 3 0 (n
o p-
o 0
......... ,L............ : ........... 2 . ........... :- . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
s ............
TIME (HOURS) Fig, 2. Galactose tolerance curves following the ingestion of milk which contains 0.78 gm of galactose (in the form of lactose) per kilogram of body weight. A . . . A ... Normal control subjects; 9 9 Duarte/"classical" galactosemia mixed heterozygous patient; @ ~ _ ' " c l a s s i c a l " galactosemic patient (on dietary treatment);O~_O~patient with Rennes variant (on dietary treatment). Table lI. Results of U D P G consumption assays in hemolysates and in cultured skin fibroblasts
Subjects Normal control subjects (20) Patient Age 3~Awk Age 3 mo Age 589 mo Mother Father "Classical" galactosemic patients (2)
Hemolflsate (l~moles UDPG consumed~1 hr/gm hemoglobin)
Cultured skin fibroblasts (IzmolesUDPG consumed~1 hr/lO ~ cells)
22.0 _+ 3.7
38.1 _+ 11.1 (17,0-62.1)
0.6,0.7* 2.0* 1.2,2.0" 12.7 (12.4-12.9) 8.1 (7.3-9.6) 0-0.5
4.8 _+ 1.6 (2.3-7.6)~
24.1,24.1 13.8,15.1 0-1.5:~
*Values for the patient are those obtained on one or two assays on single blood specimens. tRepresents assays in duplicate performed on three separate growths of the cell line. ,:~Allassays on cultured skin fibroblastsfrom these patients, with the exceptionof a single reading, gave negativeresults. slightly increased 2 m g / d l to over 50 m g / d l during the first 31/z Weeks Of life when he was receiving a lactosecontaining formula. Again, this is in contrast to the constant marked increase in blood galactose concentrations seen in patients with "classical" galactosemia ingesting lactose-containing formulas. This variable increase in blood galactose concentration in the patient is explained by his tolerance to galactose, as depicted in Fig. 2. In "classical" galactosemia, the blood galactose concentrations increased to 50 m g / d l and continued to be elevated four hours following milk ingestion. By contrast, although the blood galactose concentration in the patient rose to > 5 0 mg/dl, there was no detectable blood galactose even three hours after milk ingestion. As noted in Fig. 2, there is no rise in the blood galactose concentration following milk ingestion among n o r m a l infagts and
only slight elevations in a newborn infant who is heterozygous for both "classical" galactosemia and the Duarte variant. Amino acids. Concentrations of blood amino acids were normal. Values of urine amino acids, as analyzed by paper chromatography' and measured by ion-exchange column chromatography, were also normal in all urine specimens obtained prior to the institution of dietary therapy. Starch-gel electrophoresis of hemolysates. These results are illustrated in Fig. 3. As noted, the normal transferase b a n d i n g pattern in hemolysates is that of one brightly fluorescing major b a n d and a slower moving faintly fluorescing minor band. In hemolysates of the patient, there was only a single faintly fluorescing b a n d with mobility slower than the mobility of the major b a n d seen
54
Hammersen, "Houghton, and Levy
The Journal of Pediatrics July 1975
|
em 10-
r-ZZZ
~ j ~ ~ ] Fluorescent ".:::::.:=:::.".'::::::-::::.:~ '.-Z-=~ Bands
~
~
5-
~
~
I Hemoglobin
| ORI61~
1
2
3
4
Fig. 3B. Diagram illustrating the precise banding pattern.
Fig. 3A. Starch-gel electrophoretic patterns of transferase in hemolysates of normal individual (1), father of patient (2), mother of patient (3), and patient with Rennes-like variant (4). In order to obtain visualization of the low-activity transferase bands in the patient, incubation with the reaction mixture was carried out for 120 to 150 minutes instead of the usual 30 to 90 minutes. Electrophoretic comparisons of patient and parents with normal subjects was always performed on a single gel. However, even with such prolonged incubation time, fluorescence of the lowactivity bands was not always intense enough for clear photographic representation. in hemolysates of normal individuals. This band in the patient could only be identified when undiluted hemolysate was applied to the starch gel. Applying undiluted hemolysates of normal individuals to the gel did not change the normal banding pattern. No activity bands could be demonstrated in either patient with "classical" galactosemia. Blood specimens of the two original patients described with the Rennes variant, 8 kindly sent to us by the father of these patients, were also examined by electrophoresis. Hemolysates from these specimens were prepared in a manner identical to that used for the blood of our patient. Mobility of transferase in the two R e n n e s siblings was identical to that of our patient. FIBROBLAST
STUDIES
The most striking evidence for an unusual transferase variant in the patient was found in cultured skin fibroblasts. As noted in Table II, fibroblast transferase activity was 4.8 __ 1.6 units.* This is approximately 10% of the corresponding activity in cultured skin fibroblasts from normal individuals (38.1 ___ 11.1) and is clearly higher than the essentially undetectable activity in cultured skin fibroblasts from "classical" galactosemic individuals. *One unit of transferase activity in cultured skin fibroblast equals 1 ~mole U D P G consumed per one hour per 10" cells.
The transferase banding pattern in lysates of cultured skin fibroblasts, as determined by starch-gel electophoresis, is illustrated in Fig. 4. In normal individuals there are four visible bands, each of which is slower moving than any of the hemolysate transferase bands. The fastest and slowest of these fibroblast bands (I and IV) were only faintly fluorescing, whereas the other two (II and III) manifested bright fluorescence. In lysates of cultured skin fibroblasts from the patient, no activity bands were demonstrable by the usual method of applying lysate to gel, as used for control subjects. However, when filter paper was immersed twice in the fibroblast lysate of the patient, with drying after the first immersion, and then applied to starch gel, three faint transferase bands were detectable. These three bands were slightly but distinctly slower moving than the first three bands seen in fibroblast lysates from normal individuals (Fig. 4). Fibroblast lysates from normal individuals were also studied by starch-gel electrophoresis using this "double-dip" method with no resulting change in the normal banding pattern. There were no visible bands in fibroblast lysates from either patient with "classical" galactosemia. STUDIES
IN PARENTS
As noted in Table II, the transferase activity in hemolysates from the mother and father were 12.7 and 8.1 units, respectively. In lysates of cultured skin flbroblasts, the transferase activity of the mother and father were 24.1 and approximately 14.5 units, respectively (Table II). Although the activities in both parents were approximately 35 to 60% of normal values in both tissues and thus within the usual range for carriers of the "classical" galactosemic gene, the mother consistently had greater transferase activity than the father, her values being in the 60% range and his in the 35 to 40% range. By starch-gel electrophoresis the slower moving transferase band, found in the patient, could not be demonstrated in either parent (Figs. 3 and 4). This failure to demonstrate the
Volume 87 Number 1
Rennes-like variant of galactosemia
55
cm
t0-
F
w
Fluorescent Bands
L Fig. 4A. Starch-gel electrophoretic patterns of transferase m cultured skin fibroblast lysates of normal individual (1), father (2) and mother of patient (3), and patient with Rennes variant (4). variant band in either parent is very likely an artifact in that the fluorescence of normal bands spreads out before t h e low-activity variant band becomes visible, thus obscuring the latter. However, in contrast to t h e n o r m a l relative intensities of the fluorescing bands, fibroblasts of the mother revealed equal intensities of the first three bands and a very faintly fluorescing fourth band, suggesting that she is a carrier for the slow-moving Rennes variant. DISCUSSION The transferase-deficient form of galactosemia represented by this patient is clearly different from "classical" galactosemia. This was evident even from the results of the first enzyme spot screening test. Based on our experience in testing over 490,000 blood specimens with this assay for galactosemia, using either umbilical cord blood or that of the neonate, the result in infants with "classical" galactosemia is that of no ultraviolet fluorescence and thus no evidence of transferase activity, even after prolonged incubation with the reaction mixture. However, in this patient there repeatedly was slight but definite ultraviolet fluorescence after 16 hours' incubation. This indicated that there was erythrocyte transferase activity, albeit much less than normal activity and even less than the activity seen in the relatively common carrier state for "classical" galactosemia. Further studies of transferase in hemolysates as well as in lysates of cuttured skin fibroblasts conclusively demonstrated that the patient had markedly reduced but still detectable transferase activity and an electrophoretically slow-moving enzyme.
, . . . . 2; ,z2-2-_'., C1722~
C')
if"3
5 "-,. ........ ", '-\---2" ~
Q2D ............... :
|
ORIGIN
-
1
2
3
4
Fig. 4B. As in Fig. 3, a diagram is provided to clearly demonstrate the precise banding patterns. Ancillary data in this patient are consistent with the presence of a nonclassical variant. First, although the infant clearly accumulated galactose in blood and urine when fed a lactose-containing diet, this accumulation was intermittent, unlike the constant galactosemia of the "classical" disease. Second, his tolerance to galactose was greater than that in patients with "classical" galactosemia, as judged by the blood galactose concentration curve after the ingestion of milk. Third, although he was on a normal lactose-containing infant diet during the first three weeks of life and accumulated galactose during this period of time, he remained completely free of any clinical signs such as jaundice, weight loss, or vomiting and of biochemical signs of renal tubular toxicity, such as increased urinary excretion of amino acids. The salient characteristics regarding the transferase variants that have been reported are listed in Table III. All except the high-activity Los Angeles variant and the moderately low-activity Duarte variant are associated with severe galactosemia and clinical symptoms. The Rennes variant, reported previously in two Congolese siblings living in France, ~ resulted in vomiting, hepatomegaly, cataracts, failure to thrive, and hyperaminoaciduria during the first few weeks of life prior to the institution of a low-galactose diet.* Our patient, who *For these clinicaldetails we are indebted to Dr. Charles Jezequel.
56
Hammersen, Houghton, and Levy
The Journal of Pediatrics July 1975
Table Ill. Characteristics regarding reported variants of galactose-l-phosphate uridyl transferase
Variant "Classical" Duarte "Negro" Indiana
Rennes Los Angeles
Erythrocyte transferase activity (% of normal)
]
Starch-gel electrophoretic mobility (related to normal)
0* 50%* 0*
Faster -
0-45%
Slower
7% 140%
Slower Faster
Other characteristics
10% activity in liver and intestine Unstable in heparinized blood and isotonic phosphate buffer
*This representsthe activityin individualshomozygousfor the variant. probably represents the third reported case of the Rennes variant and the first such case recognized in a Caucasian individual, remained free of clinical symptoms for the first 31/2 weeks of life while receiving a lactose-containing formula. However, his transferase both in activity and electrophoretic mobility was indistinguishable from that of the original patients with the Rennes variant, suggesting that he too would have developed clinical desease had galactose not been eliminated from his diet. On the other hand, infants with variants of galactosemia may also have a tendency to develop less dramatic clinical disease than those with "classical" galactosemia. The findings in lysates of cultured skin fibroblasts are of particular interest in that an electrophoretically variant transferase is demonstrable in this tissue as well as in hemolysates. Normally, the transferase pattern in cultured skin fibroblasts consists of four demonstrable bands, each of which is slower moving than either of the two demonstrable transferase bands in hemolysates (Fig. 4). In our patient studies in both tissues revealed that his enzyme bands were slower moving than the corresPonding normal bands. Analogous to this are our findings in the Duarte variant which we have found is faster moving than the normal transferase in both hemolysate and cultured skin fibroblasts. Thus it appears that variant transferase enzymes have similar variations in electrophoretic mobility in cultured skin fibroblasts as in hemolysates. This suggests that human galactose-l-phosphate uridyl transferase is composed of several subunits, combining variously in a manner specific for each tissue. Subunit structure for human transferase has been suggested for human fibroblasts~ and demonstrated in human erythrocytesY Thus through human genetic mutation an altered subunit may arise altering the normal transferase banding pattern in a precise manner analogous in each tissue. The exact genotypic pattern in the family was not clearly definable by transferase studies in the parents. However, values for transferase activity suggest that the
father is a carrier for the "classical" galactosemic gene while the mother, with consistently higher transferase activities, is a carrier for the variant gene. Furthermore, the relative fluorescent intensities of the transferase bands in fibroblast lysates of the mother is an additional suggestion that she is the Rennes-like variant carrier. If so, both of these presumably allelic genes would be present in the patient, rendering him a "mixed heterozygote" for both the "classical" variant and the Rennes-like variant. The alternate possibility that he is homozygous for the Rennes-like variant is much less likely not only on the basis of the results of enzyme studies in both parents but also on the basis of the presumed low gene frequency for this variant. We wish to gratefully acknowledge the aid of Dr. Ernest Beutler for performing the first assay of transferase activity in this patient and of Drs. Herman Kalckar and Vivian E. Shih for their suggestions regarding this manuscript. We also wish to acknowledge the aid of the father of the original Rennes siblings, who kindly sent us blood specimens from each of his children. REFERENCES
1. yon Reuss A: Zuckerausscheidung im Saeuglingsalter, Wien Med Wochenschr 58:799, 1908. 2. Goeppert F: Galaktosurie nach Milchzuckergabe bei angeborenem, familiaerem chronischem leberleiden, Berl Klin Wochenschr 54:473, 1917. 3. SegalS: Disorders of galactose metabolism, in Stanbury JB, Wyngaarden JB, and Fredrickson DS, editors: The metabolic basis of inherited disease, New York, 1972, McGrawHill Book Company, Inc, p 174. 4. Beutler E, Baluda MD, Sturgeon P, and Day RW: The genetics of galactose-l-phosphate uridyl transferase deficiency, J Lab Clin Med 68:646, 1966. 5. Mathai CK, and Beutler E: Electrophoretic variation of galactose-l-phosphate uridyl transferase, Science 154:1179, 1966. 6. Segal S, Blair A, and Topper YJ: Oxidation of carbon-14 labeled galactose by subjects with clinical galactosemia, Science 136:150, 1962. 7. Chacko CM, Christian JC, and Nadler HL: Unstable
Volume 87 Number 1
8.
9.
10.
11. 12. 13,
14.
15.
16.
galactose-1-phosphate uridyl transferase: a new variant of galactosemia, J PEDIATR 78:454, 1971. Schapira F, and Kaplan JC: Electrophoretic abnormality of galactose-l-phosphate uridyl transferase in galactosemia, Biochem Biophys Res Commun 35:451, 1969. Levy HL: Genetic screening, in Harris H, and Hirschhorn K, editors: Advances in human genetics, vol 4, New York, 1973, Plenum Press, p 1. Guthrie R: Screening for "inborn errors of metabolism" in the newborn i n f a n t - a multiple test program, Birth Defects: Original Article Series 4:92, 1968. Paigen K, and Pacholec F: A new method of screening neonates for galactosemia. In preparation. Beutler E, and Baluda MD: A simple spot screening test for galactosemia, J Lab Clin Med 68:137, 1966. Shih VE, Levy HL, Karolkewicz V, Houghton S, Efron ML, Isselbacher KJ, Beutler E, and MacCready RA: Galactosemia screening of newborns in Massachusetts, N Engl J Med 284:753, 1971. Efron ML, Young D, Moser HW, and MacCready RA: A simple chromatographic screening test for the detection of disorders of amino acid metabolism: a technique using whole blood or urine collected on filter paper, N Engl J Med 270:1378, 1964. Efron ML: High voltage paper electrophoresis, in Smith I, editor: Chromatographic and electrophoretic techniques, vol 2, Zone electrophoresis, New York, 1968, John Wiley & Sons, Inc, p 166. Effort ML: Quantitative estimation of amino acids in
Rennes-like variant o f galactosemia
17.
18.
19.
20.
21.
22.
23.
57
physiological fluids using a Technicon amino acid analyzer, in Skegg LT, editor: Automation in analytical chemistry, New York, 1966, Mediad, p 637. Beutler E, and Baluda MC: Improved method for measuring galactose-l-phosphate uridy! transferase activity of erythroeytes, Clin Chim Acta 13:369, 1966. Beutler E, and Mitchell M: UDP-GIu consumption methods, in Hsia DYY, editor: Galactosemia, Springfield, Ill., 1969, Charles C Thomas Publisher, p 72. Ng WG, Bergren WR, Fields M, and Donnell GN: An improved electrophoretic procedure for galacose-l-phosphate uridyl transferase: demonstration of multiple activity bands with the Duarte variant, Biochem Biophys Res Commun 37:354, 1969. Tedesco TA, and Mellman WJ: The UDP-Glu consumption assay for gal-l-P uridyl transferase, in Hsia DYY, editor: Galactosemia, Springfield, II1., 1969, Charles C Thomas Publisher, p 66. Tedesco TA, and Mellman WJ: Inhibition of mammalian uridinediphosphoglucose 4-epimerase by dithiothreitolstimulated formation on NADH, Biochim Biophys Acta 191:144, 1969. Nadler HL, Chacko CM, and Rachmeler M: Interallelic complementation in hybrid cells derived from human diploid strains deficient in galactose-l-phosphate uridyl transferase activity, Proc Natl Acad Sci 67:976, 1970. Tedesco TA: Human galactose-l-phosphate uridyl transferase, J Biol Chem 247:6631, 1972.
July 1975 The Journal o f P E D I A T R I C S
Rennes-like variant of galactosemia: Clinical and biochemical studies The first recognized case of a Rennes-like variant form of galactosemia in a Caucasian individual is described. Galactose-l-phosphate uridyl transferase activity was approximately 10% of the normal in both erythrocytes and cultured skin fibroblasts. Electrophoretic mobility of the variant enzyme in erythrocytes was slower than that of normal individuals and identical to that of the two cases originally reported from Rennes, France. In normal cultured skin fibroblasts, four transferase bands were found. In this tissue, the patient again had a slower moving transferase. It is proposed that in transferase variants an altered subunit results in a specifically altered enzyme mobility analogous for each tissue.
Gerhard Hammersen, M.D., Sally Houghton, and Harvey L. Levy, M.D.,* Boston, Mass.
GALACTOSEMIA, an inborn error of galactose metabo-
lism, was possibly first described as early as 19081 and almost certainly described by Goeppert 2 in 1917. This entity, now known as "classical" galactosemia, has been frequently diagnosed and is characterized in the acute stage by galactosuria, hypoglycemia, neonatal jaundice, weight loss, vomiting, diarrhea, and even neonatal death and, in the chronic stage, by cataracts, cirrhosis, and mental retardation? The enzymatic defect, that of galactose-l-phosphate uridyl transferase (transferase), is at the second degradative step for galactose (Fig. 1). This enzyme, as well as other enzymes of galactose metabolism, occurs in many body tissues, including erythrocytes? Because of the accessibility of such cells, transferase has been extensively studied. These studies have led to the recognition that transferase deficiency is not a single disease but is associated with several genetiFrom the State .Laboratory Institute, Massachusetts Department o f Public Health, the Neurology Service, and the Joseph P. Kennedy, Jr., Laboratories, Massachusetts General Hospital, and the Department o f Neurology, Harvard Medical School. Supported in part by Deutsche Forschungsgemeinschaft Grant HA 855/1, USPHS H S M H A M C H Grant 0I-H-000111-04-0, N i H Grant N S 05096, and National Foundation Grant CRBS-260. *Reprint address: State Laboratory Institute, 305 South St., Boston, Mass. 02130.
Vol. 87, No. 1, pp. 50-57
cally distinct entities. Each of these entities appears to be characterized by a variant enzyme that differs from the normal enzyme and other variant enzymes in degree of activity and, in most instances, electrophoretic mobility. 4-8 The tendency for clinical complications to develop in these variants varies from apparent clinical normality in the relatively common Duarte type, 4 to perhaps mild symptoms in the "Negro" variant, 8 and to the severe galactosemia syndrome reported in the "classical, ''3 Indiana, 7 and Rennes variants? Abbreviation used UDPG: uridine diphosphoglucose The recognition that the symptoms of galactosemia can be prevented by the early institution of a galactose-free diet has led to grea t interest in the neonatal detection of these disorders? Such detection has been greatly aided by the development of relatively simple assays for either galactosel0. 11 or transferase 12 that can be applied to the filter paper blood specimen that is submitted for routine phenylketonuria testing. In the application of either or both methods of neonatal screening it is predictable that infants with the known variants plus infants with as yet undiscovered variants will be identified. For the purpose of therapy as well as for a better understanding of this important and complex group of disorders, it is important that infants with any form of transferase deficiency be studied as completely as possible.
Volume 87 Number 1
Rennes-like variant o f galactosemia
51
c ta a c t o s e - l - P h o @ s p h a t e Uridyl Transferase DG l u c o s e - l - P h o s p h a t e
Galaetose~lucose Monophosphate Shunt
UDP
Fig. 1. Pathway o f gatactose metabolism. The known h u m a n disorders in this pathway are indicated by the n u m b e r e d enzyme deficient in each disorder. T a b l e I. E n z y m e
t r a n s f e r a s e ) s p o t s c r e e n i n g test r e s u l t s a n d g a l a c t o s e c o n c e n t r a t i o n s i n p a t i e n t
Enzyme results* Specimen
Age (days)
Diet
3 hr
16 hr
Blood (umbilical cord) Blood Blood Blood Urine Blood Urine
Birth 3 21 24 24 28 28
None Lactose:~ Lactose Lactose Lactose Lactose-freew Lactose-free
----
+ + + + + + Not done
-
+ + Not done
Galactoset (mg/dO Not done 4 > 50 2 > 50 Negative Negative
*Results in this test are recorded in intensity of fluorescence varying from no fluorescence (-) to good fluorescence (+ + + +). Normally, fluorescence is good (+ + + +) after 3 hours incubation. "~Normally there is no detectable galactose in blood by this method. In our experience, urine galactose concentrations in the normal newborn infant vary from less than 2 mg/dl to 20 mg/dl, depending on the length of time of urine collection following food ingestion. :~Similac (Ross Laboratories). w (Mead-Johnson Laboratories). I n this r e p o r t
we p r e s e n t clinical a n d
biochemical
s t u d i e s o f a n i n f a n t w h o h a s a v a r i a n t t r a n s f e r a s e w h i c h in e l e c t r o p h o r e t i c m o b i l i t y a n d e n z y m e activity is i d e n t i c a l to t h e R e n n e s v a r i a n t ? T h e t r a n s f e r a s e a b n o r m a l i t y w a s d e m o n s t r a b l e b o t h in e r y t h r o c y t e s a n d in c u l t u r e d s k i n fibroblasts. CASE REPORT A Caucasian boy was born after a term pregnancy to a gravida 2, para 2, 25-year-old woman. The birth and neonatal period were normal. In particular, there was no jaundice or feeding problems. He was first seen by us at the Massachusetts General Hospital when he was 389 weeks old because of an abnormal result in the Beutler enzyme spot screening test performed routinely on filter paper specimens o f blood obtained from the umbilical cord and in the newborn nursery. The initial blood specimen and a repeat blood specimen obtained when he was 3 weeks old had revealed galactose elevations o f 4 m g / d l and > 50
mg/dl, respectively (Table I). On the first visit he was a healthy alert newborn infant with no jaundice or other signs of hepatic disease or s y m p t o m s of milk intolerance despite the ingestion since birth of a lactose-containing proprietary formula (Similac, Ross Laboratories, Columbus, Ohio). Laboratory determinations at this time revealed a blood galactose concentration of only 2 m g / d l but a urine galactose concentration of 60 to 80 mg/dl. Because of the galactose accumulations in this infant, the regular lactose-containing formula was discontinued and, at age 3~A weeks, he was fed a non-lactose-containing proprietary formula (Nutramigen, Mead Johnson Laboratories, Evansville, Ind.), which is still maintained. He is now a healthy 14-month-old infant who is normal developmentally and has no hepatomegaly or other signs of illness. Ophthalmologic examination revealed no evidence of cataracts or other abnormalities. The parents are nonrelated and of Irish/Irish (mother) and Irish/Italian (father) ancestry. There is no family history of mental retardation, liver disease, neonatal death, or early cataracts.
52
Hamrnersen, Houghton, and Levy
MATERIALS
AND METHODS
Screening for galactose-l-phosphate uridyl transferase deficiency was performed by the enzyme spot screening test fbr Beutler and BaludaJ 2 Ultraviolet fluorescence is an indication of the presence of enzyme activity. In a slight modification of the originally described method that we apply to blood specimens of newborn infants, incubation is continued at room temperature for 16 hours when no fluorescence is detected after three hours' incubation at 37 ~ C. By this procedure the specimen with no detectable activity can be differentiated from the more numerous specimens with very low activity, frequently encountered as a result of artifact during the summer months, '~ or as noted in true variant states. Measurement of galactose in blood and urine for the purposes of routine screening and for determinations in specific specimens was performed by the bacterial assay of Paigen and Pacholec. ~ This assay, as yet unpublished, utilizes discs from filter paper specimens identical to those used in the Guthrie tests .~ and measures galactose by the growth of a uridine diphosphogalactose-4-epimerase (epimerase)-deficient strain of Escherichia coli, which resists destruction by bacteriophage in the presence of galactose. This assay has been used in testing several hundred thousand specimens in the Massachusetts and Other screening programs for newborn infants ~ and is highly reliable and sensitive. Routine blood and urine screening for amino acid abnormalities was performed by the bacterial inhibition assays for phenylalanine, leucine, methionine, and tyrosine 9 and by unidimensional paper chromatography. TM Further amino acid study was performed by sequential two-day chromatography of urine '~ and by automated ion-exchange column chromatography of blood and urine. 16 To determine the tolerance to galactose in the patient, a proprietary,formula containing lactose (Enfamil, Mead Johnson Laboratories, Evansville, Ind.) was fed after an overnight fast. The patient ingested 96 m l of the formula, which contains 0.78 gm of galactose (in the form of lactose) per kilogram of body weight. Whole milk or proprietary formula containing equivalent amounts of lactose (galactose) were ingested by one infant with "classical" galactosemia, another infant with the mixed heterozygotic pattern for "classical" and the Duarte variant form of transferase deficiency, and infants with normal transferase values. Blood specimens were obtained in the fasting state and at intervals following ingestion. Galactose-l-phosphate uridyl transferase activity was measured in hemolysates by the uridine diphosphoglucose consumption method of Beutler and Baluda. '7
The Journal of Pediatrics July 1975
Hemolysates from newborn infants were preincubated with NADase (501~1 N A D a s e / m l hemolysate: NADase: 0.63 units/ml) to avoid artifacts due to epimerase activityl8
Electrophoretic mobility of transferase in hemolysates was determined by starch-gel electrophoresis as described bY Ng and associates. 19 In order to obtain visible activity bands, the hemolysate of the patient was not diluted with distilled water prior to freezing and thawing. Similar nondiluted hemolysate preparations were used for 2 of the 20 control subjects. Fibroblasts were grown from punch skin biopsies of the patient, his parents, two galactosemic patients, and normal control subjects. Cells were grown in Eagle's medium, supplemented with 15% fetal calf serum and "nonessential" amino acids. After ten days o[ subculture, the confluent monolayers were washed with phosphatebuffered saline and trypsinized. Cells were counted, centrifuged, washed twice with isotonic cold saline, and resuspended in distilled water at 2 to 6 x 107 cells/ml, and then frozen in a Dry-lce bath and thawed three times. The lysate was centrifuged 20 minutes at 700 x g and the supernates were used for the U D P G consumption assay and for starch-gel electrophoresis. In these fibroblast lysates transferase activity was determined by the U D P G consumption assay of Tedesco and Mellman. 2~To inactivate epimerase, 0.05 ml lysate was preincubated with 0.5M (0.05 ml) glycylglycine buffer, pH 8.7, and 0.1M (0.05 ml) dithiothreitol. ~1 RESULTS Erythrocyte transferase activity. The results of the enzyme spot screening test on filter paper specimens of blood are listed in Table I. Transferase activity was not detectable in any specimen after the usual incubation duration of three hours or even after incubation at 37 ~ C for five hours but was detectable in almost every specimen after incubation at room temperature for an additional 16 hours. Thus it was evident from the initial test results that the patient did have detectable erythrocyte transferase activity. This slight activity was confirmed by the U D P G consumption assay. As indicated in Table II, hemolysates from the patient when he was 3~ weeks old contained 0.65 units* of transferase activity and thereafter never more than 2.0 units of activity. This is in contrast to the normal control value of 22.0 _ 3.7 and to the virtually undetectable activity in "classical" galactosemic infants. Blood galactose concentrations. Blood galactose concentrations in the patient (Table I) varied from the only *One unit of transferase activityin hemolysatesis equal to 1 /~rnole UDPG consumedper hour per gram of hemoglobin.
Volume 87 Number 1
Rennes-like variant o f galactosemia
//
53
t!
50 E o 40 E uJ 3 0 (n
o p-
o 0
......... ,L............ : ........... 2 . ........... :- . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
s ............
TIME (HOURS) Fig, 2. Galactose tolerance curves following the ingestion of milk which contains 0.78 gm of galactose (in the form of lactose) per kilogram of body weight. A . . . A ... Normal control subjects; 9 9 Duarte/"classical" galactosemia mixed heterozygous patient; @ ~ _ ' " c l a s s i c a l " galactosemic patient (on dietary treatment);O~_O~patient with Rennes variant (on dietary treatment). Table lI. Results of U D P G consumption assays in hemolysates and in cultured skin fibroblasts
Subjects Normal control subjects (20) Patient Age 3~Awk Age 3 mo Age 589 mo Mother Father "Classical" galactosemic patients (2)
Hemolflsate (l~moles UDPG consumed~1 hr/gm hemoglobin)
Cultured skin fibroblasts (IzmolesUDPG consumed~1 hr/lO ~ cells)
22.0 _+ 3.7
38.1 _+ 11.1 (17,0-62.1)
0.6,0.7* 2.0* 1.2,2.0" 12.7 (12.4-12.9) 8.1 (7.3-9.6) 0-0.5
4.8 _+ 1.6 (2.3-7.6)~
24.1,24.1 13.8,15.1 0-1.5:~
*Values for the patient are those obtained on one or two assays on single blood specimens. tRepresents assays in duplicate performed on three separate growths of the cell line. ,:~Allassays on cultured skin fibroblastsfrom these patients, with the exceptionof a single reading, gave negativeresults. slightly increased 2 m g / d l to over 50 m g / d l during the first 31/z Weeks Of life when he was receiving a lactosecontaining formula. Again, this is in contrast to the constant marked increase in blood galactose concentrations seen in patients with "classical" galactosemia ingesting lactose-containing formulas. This variable increase in blood galactose concentration in the patient is explained by his tolerance to galactose, as depicted in Fig. 2. In "classical" galactosemia, the blood galactose concentrations increased to 50 m g / d l and continued to be elevated four hours following milk ingestion. By contrast, although the blood galactose concentration in the patient rose to > 5 0 mg/dl, there was no detectable blood galactose even three hours after milk ingestion. As noted in Fig. 2, there is no rise in the blood galactose concentration following milk ingestion among n o r m a l infagts and
only slight elevations in a newborn infant who is heterozygous for both "classical" galactosemia and the Duarte variant. Amino acids. Concentrations of blood amino acids were normal. Values of urine amino acids, as analyzed by paper chromatography' and measured by ion-exchange column chromatography, were also normal in all urine specimens obtained prior to the institution of dietary therapy. Starch-gel electrophoresis of hemolysates. These results are illustrated in Fig. 3. As noted, the normal transferase b a n d i n g pattern in hemolysates is that of one brightly fluorescing major b a n d and a slower moving faintly fluorescing minor band. In hemolysates of the patient, there was only a single faintly fluorescing b a n d with mobility slower than the mobility of the major b a n d seen
54
Hammersen, "Houghton, and Levy
The Journal of Pediatrics July 1975
|
em 10-
r-ZZZ
~ j ~ ~ ] Fluorescent ".:::::.:=:::.".'::::::-::::.:~ '.-Z-=~ Bands
~
~
5-
~
~
I Hemoglobin
| ORI61~
1
2
3
4
Fig. 3B. Diagram illustrating the precise banding pattern.
Fig. 3A. Starch-gel electrophoretic patterns of transferase in hemolysates of normal individual (1), father of patient (2), mother of patient (3), and patient with Rennes-like variant (4). In order to obtain visualization of the low-activity transferase bands in the patient, incubation with the reaction mixture was carried out for 120 to 150 minutes instead of the usual 30 to 90 minutes. Electrophoretic comparisons of patient and parents with normal subjects was always performed on a single gel. However, even with such prolonged incubation time, fluorescence of the lowactivity bands was not always intense enough for clear photographic representation. in hemolysates of normal individuals. This band in the patient could only be identified when undiluted hemolysate was applied to the starch gel. Applying undiluted hemolysates of normal individuals to the gel did not change the normal banding pattern. No activity bands could be demonstrated in either patient with "classical" galactosemia. Blood specimens of the two original patients described with the Rennes variant, 8 kindly sent to us by the father of these patients, were also examined by electrophoresis. Hemolysates from these specimens were prepared in a manner identical to that used for the blood of our patient. Mobility of transferase in the two R e n n e s siblings was identical to that of our patient. FIBROBLAST
STUDIES
The most striking evidence for an unusual transferase variant in the patient was found in cultured skin fibroblasts. As noted in Table II, fibroblast transferase activity was 4.8 __ 1.6 units.* This is approximately 10% of the corresponding activity in cultured skin fibroblasts from normal individuals (38.1 ___ 11.1) and is clearly higher than the essentially undetectable activity in cultured skin fibroblasts from "classical" galactosemic individuals. *One unit of transferase activity in cultured skin fibroblast equals 1 ~mole U D P G consumed per one hour per 10" cells.
The transferase banding pattern in lysates of cultured skin fibroblasts, as determined by starch-gel electophoresis, is illustrated in Fig. 4. In normal individuals there are four visible bands, each of which is slower moving than any of the hemolysate transferase bands. The fastest and slowest of these fibroblast bands (I and IV) were only faintly fluorescing, whereas the other two (II and III) manifested bright fluorescence. In lysates of cultured skin fibroblasts from the patient, no activity bands were demonstrable by the usual method of applying lysate to gel, as used for control subjects. However, when filter paper was immersed twice in the fibroblast lysate of the patient, with drying after the first immersion, and then applied to starch gel, three faint transferase bands were detectable. These three bands were slightly but distinctly slower moving than the first three bands seen in fibroblast lysates from normal individuals (Fig. 4). Fibroblast lysates from normal individuals were also studied by starch-gel electrophoresis using this "double-dip" method with no resulting change in the normal banding pattern. There were no visible bands in fibroblast lysates from either patient with "classical" galactosemia. STUDIES
IN PARENTS
As noted in Table II, the transferase activity in hemolysates from the mother and father were 12.7 and 8.1 units, respectively. In lysates of cultured skin flbroblasts, the transferase activity of the mother and father were 24.1 and approximately 14.5 units, respectively (Table II). Although the activities in both parents were approximately 35 to 60% of normal values in both tissues and thus within the usual range for carriers of the "classical" galactosemic gene, the mother consistently had greater transferase activity than the father, her values being in the 60% range and his in the 35 to 40% range. By starch-gel electrophoresis the slower moving transferase band, found in the patient, could not be demonstrated in either parent (Figs. 3 and 4). This failure to demonstrate the
Volume 87 Number 1
Rennes-like variant of galactosemia
55
cm
t0-
F
w
Fluorescent Bands
L Fig. 4A. Starch-gel electrophoretic patterns of transferase m cultured skin fibroblast lysates of normal individual (1), father (2) and mother of patient (3), and patient with Rennes variant (4). variant band in either parent is very likely an artifact in that the fluorescence of normal bands spreads out before t h e low-activity variant band becomes visible, thus obscuring the latter. However, in contrast to t h e n o r m a l relative intensities of the fluorescing bands, fibroblasts of the mother revealed equal intensities of the first three bands and a very faintly fluorescing fourth band, suggesting that she is a carrier for the slow-moving Rennes variant. DISCUSSION The transferase-deficient form of galactosemia represented by this patient is clearly different from "classical" galactosemia. This was evident even from the results of the first enzyme spot screening test. Based on our experience in testing over 490,000 blood specimens with this assay for galactosemia, using either umbilical cord blood or that of the neonate, the result in infants with "classical" galactosemia is that of no ultraviolet fluorescence and thus no evidence of transferase activity, even after prolonged incubation with the reaction mixture. However, in this patient there repeatedly was slight but definite ultraviolet fluorescence after 16 hours' incubation. This indicated that there was erythrocyte transferase activity, albeit much less than normal activity and even less than the activity seen in the relatively common carrier state for "classical" galactosemia. Further studies of transferase in hemolysates as well as in lysates of cuttured skin fibroblasts conclusively demonstrated that the patient had markedly reduced but still detectable transferase activity and an electrophoretically slow-moving enzyme.
, . . . . 2; ,z2-2-_'., C1722~
C')
if"3
5 "-,. ........ ", '-\---2" ~
Q2D ............... :
|
ORIGIN
-
1
2
3
4
Fig. 4B. As in Fig. 3, a diagram is provided to clearly demonstrate the precise banding patterns. Ancillary data in this patient are consistent with the presence of a nonclassical variant. First, although the infant clearly accumulated galactose in blood and urine when fed a lactose-containing diet, this accumulation was intermittent, unlike the constant galactosemia of the "classical" disease. Second, his tolerance to galactose was greater than that in patients with "classical" galactosemia, as judged by the blood galactose concentration curve after the ingestion of milk. Third, although he was on a normal lactose-containing infant diet during the first three weeks of life and accumulated galactose during this period of time, he remained completely free of any clinical signs such as jaundice, weight loss, or vomiting and of biochemical signs of renal tubular toxicity, such as increased urinary excretion of amino acids. The salient characteristics regarding the transferase variants that have been reported are listed in Table III. All except the high-activity Los Angeles variant and the moderately low-activity Duarte variant are associated with severe galactosemia and clinical symptoms. The Rennes variant, reported previously in two Congolese siblings living in France, ~ resulted in vomiting, hepatomegaly, cataracts, failure to thrive, and hyperaminoaciduria during the first few weeks of life prior to the institution of a low-galactose diet.* Our patient, who *For these clinicaldetails we are indebted to Dr. Charles Jezequel.
56
Hammersen, Houghton, and Levy
The Journal of Pediatrics July 1975
Table Ill. Characteristics regarding reported variants of galactose-l-phosphate uridyl transferase
Variant "Classical" Duarte "Negro" Indiana
Rennes Los Angeles
Erythrocyte transferase activity (% of normal)
]
Starch-gel electrophoretic mobility (related to normal)
0* 50%* 0*
Faster -
0-45%
Slower
7% 140%
Slower Faster
Other characteristics
10% activity in liver and intestine Unstable in heparinized blood and isotonic phosphate buffer
*This representsthe activityin individualshomozygousfor the variant. probably represents the third reported case of the Rennes variant and the first such case recognized in a Caucasian individual, remained free of clinical symptoms for the first 31/2 weeks of life while receiving a lactose-containing formula. However, his transferase both in activity and electrophoretic mobility was indistinguishable from that of the original patients with the Rennes variant, suggesting that he too would have developed clinical desease had galactose not been eliminated from his diet. On the other hand, infants with variants of galactosemia may also have a tendency to develop less dramatic clinical disease than those with "classical" galactosemia. The findings in lysates of cultured skin fibroblasts are of particular interest in that an electrophoretically variant transferase is demonstrable in this tissue as well as in hemolysates. Normally, the transferase pattern in cultured skin fibroblasts consists of four demonstrable bands, each of which is slower moving than either of the two demonstrable transferase bands in hemolysates (Fig. 4). In our patient studies in both tissues revealed that his enzyme bands were slower moving than the corresPonding normal bands. Analogous to this are our findings in the Duarte variant which we have found is faster moving than the normal transferase in both hemolysate and cultured skin fibroblasts. Thus it appears that variant transferase enzymes have similar variations in electrophoretic mobility in cultured skin fibroblasts as in hemolysates. This suggests that human galactose-l-phosphate uridyl transferase is composed of several subunits, combining variously in a manner specific for each tissue. Subunit structure for human transferase has been suggested for human fibroblasts~ and demonstrated in human erythrocytesY Thus through human genetic mutation an altered subunit may arise altering the normal transferase banding pattern in a precise manner analogous in each tissue. The exact genotypic pattern in the family was not clearly definable by transferase studies in the parents. However, values for transferase activity suggest that the
father is a carrier for the "classical" galactosemic gene while the mother, with consistently higher transferase activities, is a carrier for the variant gene. Furthermore, the relative fluorescent intensities of the transferase bands in fibroblast lysates of the mother is an additional suggestion that she is the Rennes-like variant carrier. If so, both of these presumably allelic genes would be present in the patient, rendering him a "mixed heterozygote" for both the "classical" variant and the Rennes-like variant. The alternate possibility that he is homozygous for the Rennes-like variant is much less likely not only on the basis of the results of enzyme studies in both parents but also on the basis of the presumed low gene frequency for this variant. We wish to gratefully acknowledge the aid of Dr. Ernest Beutler for performing the first assay of transferase activity in this patient and of Drs. Herman Kalckar and Vivian E. Shih for their suggestions regarding this manuscript. We also wish to acknowledge the aid of the father of the original Rennes siblings, who kindly sent us blood specimens from each of his children. REFERENCES
1. yon Reuss A: Zuckerausscheidung im Saeuglingsalter, Wien Med Wochenschr 58:799, 1908. 2. Goeppert F: Galaktosurie nach Milchzuckergabe bei angeborenem, familiaerem chronischem leberleiden, Berl Klin Wochenschr 54:473, 1917. 3. SegalS: Disorders of galactose metabolism, in Stanbury JB, Wyngaarden JB, and Fredrickson DS, editors: The metabolic basis of inherited disease, New York, 1972, McGrawHill Book Company, Inc, p 174. 4. Beutler E, Baluda MD, Sturgeon P, and Day RW: The genetics of galactose-l-phosphate uridyl transferase deficiency, J Lab Clin Med 68:646, 1966. 5. Mathai CK, and Beutler E: Electrophoretic variation of galactose-l-phosphate uridyl transferase, Science 154:1179, 1966. 6. Segal S, Blair A, and Topper YJ: Oxidation of carbon-14 labeled galactose by subjects with clinical galactosemia, Science 136:150, 1962. 7. Chacko CM, Christian JC, and Nadler HL: Unstable
Volume 87 Number 1
8.
9.
10.
11. 12. 13,
14.
15.
16.
galactose-1-phosphate uridyl transferase: a new variant of galactosemia, J PEDIATR 78:454, 1971. Schapira F, and Kaplan JC: Electrophoretic abnormality of galactose-l-phosphate uridyl transferase in galactosemia, Biochem Biophys Res Commun 35:451, 1969. Levy HL: Genetic screening, in Harris H, and Hirschhorn K, editors: Advances in human genetics, vol 4, New York, 1973, Plenum Press, p 1. Guthrie R: Screening for "inborn errors of metabolism" in the newborn i n f a n t - a multiple test program, Birth Defects: Original Article Series 4:92, 1968. Paigen K, and Pacholec F: A new method of screening neonates for galactosemia. In preparation. Beutler E, and Baluda MD: A simple spot screening test for galactosemia, J Lab Clin Med 68:137, 1966. Shih VE, Levy HL, Karolkewicz V, Houghton S, Efron ML, Isselbacher KJ, Beutler E, and MacCready RA: Galactosemia screening of newborns in Massachusetts, N Engl J Med 284:753, 1971. Efron ML, Young D, Moser HW, and MacCready RA: A simple chromatographic screening test for the detection of disorders of amino acid metabolism: a technique using whole blood or urine collected on filter paper, N Engl J Med 270:1378, 1964. Efron ML: High voltage paper electrophoresis, in Smith I, editor: Chromatographic and electrophoretic techniques, vol 2, Zone electrophoresis, New York, 1968, John Wiley & Sons, Inc, p 166. Effort ML: Quantitative estimation of amino acids in
Rennes-like variant o f galactosemia
17.
18.
19.
20.
21.
22.
23.
57
physiological fluids using a Technicon amino acid analyzer, in Skegg LT, editor: Automation in analytical chemistry, New York, 1966, Mediad, p 637. Beutler E, and Baluda MC: Improved method for measuring galactose-l-phosphate uridy! transferase activity of erythroeytes, Clin Chim Acta 13:369, 1966. Beutler E, and Mitchell M: UDP-GIu consumption methods, in Hsia DYY, editor: Galactosemia, Springfield, Ill., 1969, Charles C Thomas Publisher, p 72. Ng WG, Bergren WR, Fields M, and Donnell GN: An improved electrophoretic procedure for galacose-l-phosphate uridyl transferase: demonstration of multiple activity bands with the Duarte variant, Biochem Biophys Res Commun 37:354, 1969. Tedesco TA, and Mellman WJ: The UDP-Glu consumption assay for gal-l-P uridyl transferase, in Hsia DYY, editor: Galactosemia, Springfield, II1., 1969, Charles C Thomas Publisher, p 66. Tedesco TA, and Mellman WJ: Inhibition of mammalian uridinediphosphoglucose 4-epimerase by dithiothreitolstimulated formation on NADH, Biochim Biophys Acta 191:144, 1969. Nadler HL, Chacko CM, and Rachmeler M: Interallelic complementation in hybrid cells derived from human diploid strains deficient in galactose-l-phosphate uridyl transferase activity, Proc Natl Acad Sci 67:976, 1970. Tedesco TA: Human galactose-l-phosphate uridyl transferase, J Biol Chem 247:6631, 1972.