- Email: [email protected]
Molecular and Biochemical Basis of Galactosemia
MOLECULAR GENETICS AND METABOLISM ARTICLE NO.
63, 263–269 (1998)
GM982678
Molecular and Biochemical Basis of Galactosemia Boris B. T. Wang,*,1 Yan-Kang Xu,* Won G. Ng,* and Lee-Jun C. Wong*,†,‡,2 *Division of Medical Genetics, Department of Pediatrics, and †Molecular Genetics Laboratory, Department of Pathology, Children’s Hospital, Los Angeles, California 90027; ‡Institute for Molecular and Human Genetics, Georgetown University, Washington, DC 20007 Received December 1, 1997, and in revised form December 22, 1997
of S135L compound heterozygotes expressed variable amounts of GALT activity. We speculate that heterodimeric subunit interaction plays an important role in determining the overall enzymatic activity. Various genotypes thus result in biochemical and clinical heterogeneity among the patients.
Galactosemia is a clinically heterogeneous autosomal recessive inborn error of metabolism caused by deficiency of galactose-1-phosphate uridylyltransferase (GALT). Despite the numerous point mutations identified in the GALT gene, the prevalence of these mutations in different ethnic groups has not been studied. Reports on genotype/ phenotype correlation are not consistent due to the small sample sizes studied and the lack of a sensitive enzyme assay. We applied multiplex PCR/ASO dot blot analysis to screen 293 galactosemic patients for 17 known point mutations in exons 5, 6, and 10. Our data demonstrate that only 7 of these mutations were detected in our patients, accounting for 65% of the GALT mutant alleles. Although Q188R is the most common mutation in Caucasian and Hispanic patients, the S135L mutation is most common in African-Americans. Another mutation, F171S, was observed only among African-American patients. An improved, sensitive, and accurate method was used to measure GALT activity in patient’s red blood cells. The results indicated that patients homozygous for Q188R have no enzyme activity while those homozygous for S135L had residual enzyme activity. Interestingly, both Q188R/S135L and S135L/ F171S compound heterozygotes demonstrated zero enzyme activity. Overall, 85% of Q188R compound heterozygotes also did not have any enzyme activity, whereas the remaining Q188R and the majority
q 1998 Academic Press
Key Words: galactosemia; molecular screening; GALT mutations; phenotype/genotype.
Galactosemia is a clinically heterogeneous inborn error of metabolism. It is an autosomal recessive disorder which occurs with a frequency of about 1 in 62,000 (1). The affected infant usually appears normal at birth with symptoms developing after milk feeding (2). Clinical manifestations include feeding difficulties, vomiting, lethargy, diarrhea, jaundice, hepatomegaly, failure to thrive, increased intracranial pressure, and cataracts (2). Abnormal laboratory findings include hyperbilirubinemia, proteinuria, amino aciduria, elevated levels of galactose-1-phosphate in blood, and high levels of galactose and galactitol in urine. Treatment is with dietary galactose restriction. Despite dietary restriction, long-term complications may involve growth retardation, speech defects, ovarian failure, learning disabilities, and other neurological and psychological abnormalities (2). The biochemical defect in galactosemia is the impairment of galactose-1-phosphate uridylyltransferase (GALT), which catalyzes the second step in the Leloir pathway of galactose metabolism. Traditional laboratory confirmation of the disease is established by measurement of the enzyme activity in red blood cells (3). Recently, a sensitive radiochemical method has been developed which can detect GALT activity as low as 0.1% of normal in both
1 Present address: Genzyme Genetics, 2000 Vivigen Way, Santa Fe, NM 87505. 2 To whom correspondence and reprint requests should be addressed at Director, Molecular Diagnostics Laboratory, Institute for Molecular and Human Genetics, Georgetown University Medical Center, 3800 Reservoir Road NW, Suite 4000, Washington, DC 20007. Fax: (202) 784-1770. E-mail: [email protected].
263 1096-7192/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MGM 2678
/
af11$$$$81
04-29-98 08:12:52
bmmal
AP: MGM
264
WANG ET AL.
erythrocytes and leukocytes (4). This method is extremely valuable in the determination of minute amounts of GALT activity in a biochemically heterogeneous galactosemia population. The human GALT gene has been mapped to chromosome 9p13 (5). The gene, containing 11 exons and spanning 4 kb, has been cloned and sequenced (6). More than 50 point mutations have been identified by SSCP (single-strand conformation polymorphism) analysis and sequencing (7–11). Many of the mutant cDNAs have been expressed in COS cells (8) and yeast (12) to determine the functional consequences of the mutations. Galactose-1-phosphate uridylyltransferase functions as a dimer which is structurally stabilized by a pair of Zn/Fe (13). In the Escherichia coli enzyme, two UMP molecules are covalently bound per uridylyl enzyme intermediate (14,15). Although the overall amino acid sequence identity of the E. coli, yeast, and human enzymes is only 35% (16), the structurally and functionally conserved domains have been described. Studies have shown that mutations in the structurally and functionally conserved regions result in low or absent enzyme activity (17,18). It has been hypothesized that different mutations lead to biochemical and clinical heterogeneity depending on the nature and position of the altered amino acid. The most common mutation in GALT is Q188R, which has a prevalence of 60–70% in galactosemic Caucasians (6,19). Patients homozygous for Q188R have essentially no detectable enzyme activity in erythrocytes and lymphocytes. This subgroup of galactosemic patients has a more severe clinical phenotype, including reduced intellect and ovarian failure. In contrast, patients with the African-American-specific mutation, S135L, generally have had a mild clinical course. Since the number of patients studied in previous reports was small (10), the incidence of many mutations, such as the S135L allele among Caucasians and the Q188R allele among African Americans, as well as the enzyme activities in patients with various genotypes, remain unclear. In order to unravel these questions, 293 galactosemic individuals were screened for 17 known point mutations in exons 5, 6, and 10, and the obtained genotypes were correlated with biochemical phenotypes. Here we report the allele frequency of each of these point mutations in Caucasian, Hispanic, and African-American patients. The biochemical implications of various genotypes are discussed. MATERIALS AND METHODS This study was carried out according to the protocols, CCI 86-773 and CCI 90-117, approved by the
AID
MGM 2678
/
af11$$$$81
04-29-98 08:12:52
Committee on Clinical Investigation at Children’s Hospital, Los Angeles. Over the years, we have identified about 1000 galactosemia patients. Blood samples were available from approximately half of them. A total of 293 patients, including 210 Caucasians, 28 Hispanics, and 55 African-Americans, were studied. The GALT activity in the red blood cells of these patients was measured using a sensitive radiochemical assay (4). Patient DNA was extracted from either the blood spots or whole blood (20,21). A simple, rapid, and cost-effective multiplex PCR/ASO (allele-specific oligonucleotide) hybridization method (22) was employed to screen 17 known point mutations in exons 5, 6, and 10 of the GALT gene. These exons were chosen because they are apparently the mutation hot spots and they contain the two most common mutations, Q188R and S135L. The forward and reverse primers for PCR amplification of exons 5 and 6 were 5*CTGCCCGTAGCACAGCCA3* and 5*GCTGGCTCAGACTCAGCC3*, respectively. The primers for exon 10 are 5*GGGTTTGGGAGTAGGTGCT3* and 5*GGGCAACAGAAGTATCAGGT3*. PCR conditions and ASO dot blot hybridization conditions were similar to the previously published procedures (22). Briefly, 96-well plates were used for multiplex PCR amplification using a Perkin–Elmer 9600 thermal cycler. Each 100-ml PCR mixture contained 11 Promega PCR buffer, 1.5 mM MgCl2 , 0.2 mM concentrations of each dNTP, 0.5 mM concentrations of each of the two pairs of primers stated above, 1 unit of Taq DNA polymerase, and 100 ng of genomic DNA. The reaction mixture was denatured at 947C for 2 min, followed by 32 cycles of 30 s of denaturation at 947C, 30 s of reannealing at 547C, and 40 s of extension at 727C. The PCR was completed by a final extension cycle at 727C for 4 min. Two microliters of PCR product was spotted onto a Biodyne B/ membrane. Dot blots were prepared for each of the normal and mutant probes listed in Table 1. The ASO probes were end-labeled with [g-32P]ATP using T4 polynucleotide kinase. RESULTS Figure 1 shows representative results of PCR/ ASO dot blot analysis for the Q188R and F171S mutations. Only 7 (Q188R, S135L, F171S, N314D, R148W, R333G, and R333W) of 17 mutations screened were detected in our patient population. Table 2 summarizes the mutant allele frequencies in a total of 293 galactosemia patients. The overall detection rate was 65%. The detection rate in each
bmmal
AP: MGM
265
MOLECULAR MECHANISM OF GALACTOSEMIA
TABLE 1 Nucleotide Sequence of the ASO Probes Probe
Sequence (5* to 3*)
Sense (S) or antisense (A)
S135 S135L T138, L139, M142 T138M L139P M142K R148, V151 R148W R148Q V151A F171 F171S Q188 Q188R E308 E308K N314 N314D H319 H319Q A320 A320T Y323 Y323D R333, K334 R333G R333W K334R
CCCCTGGTCGGATGTAAC CCCCTGGTTGGATGTAAC TGTAACGCTGCCACTCATGTCG GGATGTAATGCTGCCACT ATGTAACGCCGCCACTCA CCACTCAAGTCGGTCCC GATCCGGGCTGTTGTTGATG TGAGATCTGGGCTGTTGT TGAGATCCAGGCTGTTGT GGCTGTTGCTGATGCATG GATAGATCTTTGAAAACAAAGG GATAGATCTCTGAAAACAAAGG ACCCTTACCTGGCAGTGG ACCCTTACCCGGCAGTGG CCCAGCCTCTGATCCTGT CCCAGCCTTTGATCCTGT CCAATGGTTCCAGTTGGC CCAATGGTCCCAGTTGGC CAGCTGCACGCTCATTAC CAGCTGCAAGCTCATTAC CAGCTGCACGCTCATTACT CAGCTGCACACTCATTACT CGGAGGGTAGTAATGAGC CGGAGGGTCGTAATGAGC ATGAATTTCCGGACAGTGG ATGAATTTCCCGACAGTGG ATGAATTTCCAGACAGTGG ATGAATCTCCGGACAGTGG
S S S S S S S S S S S S A A A A A A S S S S A A A A A A
ethnic group was 67% in Caucasians, 48% in Hispanics, and 68% in African-Americans. As shown in the table, the most prevalent mutant allele, Q188R, accounts for 63% of the Caucasian GALT mutant alleles in our population. Forty-two percent of the Caucasian patients are homozygous and 42% are compound heterozygous for the Q188R mutation. The S135L mutation, the most common mutant allele in African-American GALT patients, is present in a small number of Caucasian patients, accounting for 1.4% (6/420) of the mutant alleles. In African-American patients, the most common mutation is S135L, accounting for 45% of the mutant alleles, followed by Q188R (20%) and F171S (3.6%). One homozygote and two compound heterozygotes for the F171S mutation were found in 55 AfricanAmerican patients. The F171S mutation has not been detected among any Caucasian or Hispanic patients in this study. The only mutation identified among the Hispanic patients was Q188R, for which 28% patients were homozygous and 46.4% were compound heterozygous for this mutation. Among all
AID
MGM 2678
/
af11$$$$82
04-29-98 08:12:52
three populations, the unidentified mutations were prevalent, including 33% of Caucasian, 32% of African-American, and 52% of Hispanic mutant alleles. There have been questions regarding the predictability of the biochemical and clinical phenotypes associated with particular genotypes. Conflicting results for the enzymatic activity of the S135L mutant allele have been reported (7,12). In this study, we measured the red blood cell transferase activity in 258/293 (88%) of our patients using a more sensitive assay. Table 3 lists the genotypes and the red blood cell GALT activities in the galactosemia patients. Our data indicate that all of the Q188R homozygotes and most (85%) of the compound heterozygotes have essentially no GALT activity. Residual enzyme activity was observed in two patients with Q188R/ R333G but not in the single patient with the Q188R/ R333W genotype. The arginine (R) to tryptophan (W) change results in the drastic replacement of a highly basic, positively charged amino acid with a very hydrophobic aromatic amino acid. A complete loss of enzyme activity may be predicted.
bmmal
AP: MGM
266
WANG ET AL.
FIG. 1. ASO dot blot analysis of point mutations in the GALT gene. (A and B) Hybridization with Q188R normal and mutant probes, respectively. A1, D5, and C5 are, respectively, examples of a normal, heterozygote, and homozygote for the Q188R mutation. A12, C4, and H6 are no-DNA controls. (C and D) Hybridization with F171S normal and mutant probes, respectively. G6 is a heterozygote for the F171S mutation. A1 and H7 contain no DNA template in the PCR.
Consistent with its milder clinical outcome, patients homozygous for the S135L mutant allele demonstrate residual red blood cell GALT activity (10). However, when the S135L allele is found in compound heterozygotes with Q188R, no GALT activity was detected. Variable amounts of residual enzyme activity are present in patients with an S135L/unknown genotype. Patients homozygous for F171S or compound heterozygous for F171S/Q188R have no
enzyme activity, suggesting that F171S is a severe mutation, consistent with previous data (18). DISCUSSION PCR/ASO dot blot analysis is a simple, accurate, sensitive, and cost-effective method for the screening of multiple, recurrent, known mutations. In this report, we multiplex-amplified exons 5/6 and 10, where
TABLE 2 Mutant Allele Frequencies in GALT Gene Caucasian
Hispanic
African-American
Total
Mutation
No. alleles
%
No. alleles
%
No. alleles
%
No. alleles
%
Q188R S135L F171S N314D R333G R333W R148W Total identified Unidentified
266 6 0 4 2 1 1 280 140
63 1.4 0 1 0.5 0.2 0.2 67 33
27 0 0 0 0 0 0 27 29
48 0 0 0 0 0 0 48 52
22 49 4 0 0 0 0 75 35
20 45 3.6 0 0 0 0 68 32
315 55 4 4 2 1 1 382 204
53.75 9.39 0.68 0.68 0.34 0.17 0.17 65.18 34.81
Total
AID
420
MGM 2678
/
af11$$$$82
56
04-29-98 08:12:52
110
bmmal
AP: MGM
586
267
MOLECULAR MECHANISM OF GALACTOSEMIA
TABLE 3 Correlation of Genotype with Red Blood Cell Transferase Activity in Galactosemia Patients Number of patients Mutation (allele 1/allele 2)
Caucasian
Q188R/Q188R Q188R/S135L Q188R/R333G Q188R/R333W Q188R/R148W Q188R/unknown S135L/S135L S135L/F171S S135L/unknown F171S/F171S Unknown/unknown
AfricanAmerican
89 3 2 1 1 81
3 30
4 10
7
4 10 2 17 1 7
13
most common mutations occur. Seventeen known point mutations in this region were screened in 293 galactosemia patients. Only 7 different mutations, Q188R, S135L, N314D, F171S, R333W, R333G, and R148W, were detected. This suggests that most or all of the other 10 mutations are private mutations with almost no recurrence rate. The overall detection rate for the 7 point mutations is 67–68% for Caucasians and African-Americans, but only 48% for Hispanics (Table 2). The lower detection rate among Hispanics is similar to that currently noted for other diseases, such as cystic fibrosis, phenylketonuria, and glycogen storage disease type 1a, and is likely secondary to either a greater genetic heterogeneity or that mutations to be screened are typically identified by centers whose patient population is predominately Caucasian and African-American. For the molecular diagnosis of galactosemia, we recommend that the first step should be PCR/ASO dot blot analysis for the 7 recurrent mutations. If zero or only one mutant allele is detected, then SSCP or other mutation detection methods should be used to screen for unknown mutations in all 11 exons, followed by direct DNA sequencing to confirm and identify the mutation. We have recently identified several novel mutations using the TTGE (temporal temperature gradient gel electrophoresis) method (to be published elsewhere). The small percentage (1.4%) of S135L alleles found in the Caucasian galactosemia patients could be due either to racial admixture or to an independent mutation which arose in the Caucasian population. On the other hand, Q188R, which accounts for 20% of the mutant GALT alleles in the African-
AID
MGM 2678
/
af11$$$$82
Hispanic
04-29-98 08:12:52
8
GALT activity (% of normal in RBC) 0 0 0.3 0 0 0–22 0.2–1.7 0 0–17 0 0–29
American galactosemia population, is probably due predominately to racial admixture considering the high frequency of Caucasian genes among AfricanAmericans (23). Of particular importance is to correlate the molecular genotypes with their corresponding GALT activities. Because of the large number of patients studied, we are able to examine the enzyme activity in various homozygous and compound heterozygous combinations. Our results suggest that subunit interaction plays an important role in determining the enzyme activity. Although S135L homozygotes show a residual enzyme activity of 0.2–1.7% of normal activity, the activity of the S135L compound heterozygote varies from 0 to 17% of normal controls (Table 3). The compound heterozygotes S135L/Q188R and S135L/F171S show no detectable enzyme activity (Table 3). Similarly, the enzymatic activity of the Q188R compound heterozygote varies from 0 to 20%. These data suggest that there is possibly a dominant negative effect of one mutation over the other by heterodimeric interaction. Assuming that the S135L and the Q188R polypeptides are produced in equal amounts, homodimers and heterodimers are formed without preference, and there is no subunit predominant effect, one would expect that the S135L/Q188R heterozygote would have an enzyme activity of 50% that of a S135L homozygote, which is not the case. If the Q188R mutation inactivated the S135L-containing subunit in a Q188R/S135L heterodimer and there is no preference towards homo- or heterodimer formation, then one would expect the enzyme activity to be 25% that of a S135L homozygote, since only the S135L homodimer is functional. A possible ex-
bmmal
AP: MGM
268
WANG ET AL.
planation for the zero enzyme activity of the Q188R/ S135L compound heterozygote is that heterodimer formation is much more favorable than homodimer formation. Within the heterodimer, the Q188R mutant subunit either structurally destabilizes and facilitates degradation of the S135L subunit or functionally impairs the catalytic ability of the heterodimer. It is speculated that the heterodimer has higher affinity, which shifts the equilibrium toward the nonfunctional heterodimeric state by physically disabling the function of the other subunit. In addition, GALT enzyme activity varies from tissue to tissue. Thus tissue-specific cellular protein factors may also play important role in the regulation of GALT gene expression at the transcription, translation, or posttranslation level. Further studies of GALT activity should be extended from the red blood cells to include the nucleated cells, such as white blood cells or cultured skin fibroblasts. Not all Q188R heterodimers exhibit zero enzyme activity. About 15% of the Q188R compound heterozygous patients show residual activity. This observation is also consistent with our hypothesis that enzyme activity depends on the heterodimeric interaction. Which subunit is dominant over the other depends on the type of combination of the mutant protein subunits. Identification of the mutation in the second mutant allele will assist in the interpretation of the biochemical phenotypes in these patients. It is apparent that the enzyme activity depends on the combination of alleles in a given patient. This may partly explain the heterogeneous clinical outcome of patients who carry one common mutant allele. Studies to correlate the biochemical and molecular findings with patients’ clinical outcomes are currently being undertaken. ACKNOWLEDGMENTS The authors thank the clinicians who referred the patients to us.
REFERENCES 1. Levy HL, Hammersen G. Screening for galactosemia. In Inherited Disorders of Carbohydrate Metabolism (Burman D, Holton JB, Pennock CA, Eds.). Lancaster, England: MTP 1980, pp 133–139.
4. Xu YK, Kaufman FR, Donnell GN, Ng WG. Radiochemical assay of minute quantities of galactose-1-phosphate uridyltransferase activity in erythrocytes and leukocytes of galactosemia patients. Clin Chim Acta 235:125–136, 1995. 5. Sparkes RS, Sparkes MC, Funderburk SJ, Moedjono S. Expression of GALT in 9p chromosome alterations: assignment of GALT locus to 9cen r 9p22. Ann Hum Genet 43:343–347, 1980. 6. Leslie ND, Immerman EB, Flach JE, Florez M, FridovichKeil JL, Elsas LJ. The human galactose-1-phosphate uridyltransferase gene. Genomics 14:474–480, 1992. 7. Reichardt JKV, Woo SLC. Molecular basis of galactosemia: mutations and polymorphisms in the gene encoding human galactose-1-phosphate uridyl transferase. Proc Natl Acad Sci USA 88:2633–2637, 1991. 8. Reichardt J, Belmont J, Levy H, Woo S. Characterization of two missense mutations in human galactose-1-phosphate uridyl transferase: different molecular mechanisms for galactosemia. Genomics 12:596–600, 1992. 9. Elsas LJ, Langley S, Steele E, Evinger J, Fridovich-Keil JL, Brown A, Singh R, Fernhoff P, Hjelm LN, Dembure PP. Galactosemia: A strategy to identify new biochemical phenotypes and molecular genotypes. Am J Hum Genet 56:630– 639, 1995. 10. Lai K, Langley SD, Singh RH, Dembure PP, Hjelm LN, Elsas LJ. A prevalent mutation for galactosemia among black Americans. J Pediatr 128:89–95, 1996. 11. Maceratesi P, Sangiuolo F, Novelli G, Ninfali P, Magnani M, Reichardt JKV, Dallapiccola B. Three new mutations (P183T, V150L, 528insG) and eleven sequence polymorphisms in Italian patients with galactose-1-phosphate uridyltransferase (GALT) deficiency. Hum Mut 8:369–372, 1996. 12. Fridovich-Keil JL, Langley SD, Mazur LA, Lennon JC, Dembure PP, Elsas LJ II. Identification and functional analysis of three distinct mutations in the human galactose-1phosphate uridyltransferase gene associated with galactosemia in a single family. Am J Hum Genet 56:640–646, 1995. 13. Frey PA. The Leloir pathway: a mechanistic imperative for three enzymes to change the stereochemical configuration of a single carbon in galactose. FASEB J 10:461–470, 1996. 14. Wong LJ, Sheu KF, Lee SL, Frey PA. Galactose-1-phosphate uridylyltransferase: Isolation and properties of a uridylylenzyme intermediate. Biochemistry 16:1010–1016, 1977. 15. Frey PA, Wong LJ, Sheu KF, Yang SL. Galactose-1-phosphate uridylyltransferase: Detection, isolation and characterization of the uridylyl enzyme. Methods Enzymol 87:20– 36, 1982. 16. Flach JE, Reichardt JKV, Elsas LJ. Sequence of a cDNA encoding human galactose-1-phosphate uridylyltransferase. Mol Biol Med 7:365–369, 1990.
2. Segal S, Berry GT. Disorders of galactose metabolism. In The Metabolic and Molecular Bases of Inherited Disease (Scriver CR, Beaudet AL, Sly WS, Valle D, Eds.), 7th ed., Vol. 1. New York: McGraw-Hill, 1995, pp 967–1000.
17. Reichardt JKV, Packman S, Woo SLC. Molecular characterization of two galactosemia mutations: correlation of mutations with highly conserved domains in galactose-1-phosphate uridyltransferase. Am J Hum Genet 49:860–867, 1991.
3. Lee JES, Ng WG. Semi-micro techniques for the genotyping of galactokinase and galactose-1-phosphate uridyl transferase. Clin Chim Acta 124:351–356, 1982.
18. Reichardt JKV, Levy H, Woo S. Molecular characterization of two galactosemic mutations and one polymorphism: Implications for structure-function analysis of human galactose-1-
AID
MGM 2678
/
af11$$$$82
04-29-98 08:12:52
bmmal
AP: MGM
MOLECULAR MECHANISM OF GALACTOSEMIA phosphate uridyl transferase. Biochemistry 31:5430–5433, 1992. 19. Ng WG, Xu YK, Kaufman FR, Donnell GN, Wolff J, Allen RJ, Koritala S, Reichardt JKV. Biochemical and molecular studies of 132 patients with galactosemia. Hum Genet 94:359–363, 1994. 20. Gregersen N, Blakemore A, Winter V, Andersen BS, Kolvraa S, Bolund L, Curtis D, Engel PC. Specific diagnosis of medium-chain acyl-CoA dehydrogenase (MCAD) deficiency in dried blood spots by a polymerase chain reaction (PCR) assay
AID
MGM 2678
/
af11$$$$83
04-29-98 08:12:52
269
detecting a point-mutation (G985) in the MCAD gene. Clin Chim Acta 203:23–34, 1991. 21. Lahiri D, Nurnberger J Jr. A rapid non-enzymatic method for the preparation of HMW DNA from blood for RFLP studies. Nucleic Acids Res 19:5444, 1991. 22. Wong LJC, Senadheera D. Direct detection of multiple point mutations in mitochondrial DNA. Clin Chem 43:1857–1861, 1997. 23. Chakraborty R, Kamboh MI, Nwankwo M, Ferrell RE. Caucasian genes in American blacks: new data. Am J Hum Genet 50:145–155, 1992.
bmmal
AP: MGM
63, 263–269 (1998)
GM982678
Molecular and Biochemical Basis of Galactosemia Boris B. T. Wang,*,1 Yan-Kang Xu,* Won G. Ng,* and Lee-Jun C. Wong*,†,‡,2 *Division of Medical Genetics, Department of Pediatrics, and †Molecular Genetics Laboratory, Department of Pathology, Children’s Hospital, Los Angeles, California 90027; ‡Institute for Molecular and Human Genetics, Georgetown University, Washington, DC 20007 Received December 1, 1997, and in revised form December 22, 1997
of S135L compound heterozygotes expressed variable amounts of GALT activity. We speculate that heterodimeric subunit interaction plays an important role in determining the overall enzymatic activity. Various genotypes thus result in biochemical and clinical heterogeneity among the patients.
Galactosemia is a clinically heterogeneous autosomal recessive inborn error of metabolism caused by deficiency of galactose-1-phosphate uridylyltransferase (GALT). Despite the numerous point mutations identified in the GALT gene, the prevalence of these mutations in different ethnic groups has not been studied. Reports on genotype/ phenotype correlation are not consistent due to the small sample sizes studied and the lack of a sensitive enzyme assay. We applied multiplex PCR/ASO dot blot analysis to screen 293 galactosemic patients for 17 known point mutations in exons 5, 6, and 10. Our data demonstrate that only 7 of these mutations were detected in our patients, accounting for 65% of the GALT mutant alleles. Although Q188R is the most common mutation in Caucasian and Hispanic patients, the S135L mutation is most common in African-Americans. Another mutation, F171S, was observed only among African-American patients. An improved, sensitive, and accurate method was used to measure GALT activity in patient’s red blood cells. The results indicated that patients homozygous for Q188R have no enzyme activity while those homozygous for S135L had residual enzyme activity. Interestingly, both Q188R/S135L and S135L/ F171S compound heterozygotes demonstrated zero enzyme activity. Overall, 85% of Q188R compound heterozygotes also did not have any enzyme activity, whereas the remaining Q188R and the majority
q 1998 Academic Press
Key Words: galactosemia; molecular screening; GALT mutations; phenotype/genotype.
Galactosemia is a clinically heterogeneous inborn error of metabolism. It is an autosomal recessive disorder which occurs with a frequency of about 1 in 62,000 (1). The affected infant usually appears normal at birth with symptoms developing after milk feeding (2). Clinical manifestations include feeding difficulties, vomiting, lethargy, diarrhea, jaundice, hepatomegaly, failure to thrive, increased intracranial pressure, and cataracts (2). Abnormal laboratory findings include hyperbilirubinemia, proteinuria, amino aciduria, elevated levels of galactose-1-phosphate in blood, and high levels of galactose and galactitol in urine. Treatment is with dietary galactose restriction. Despite dietary restriction, long-term complications may involve growth retardation, speech defects, ovarian failure, learning disabilities, and other neurological and psychological abnormalities (2). The biochemical defect in galactosemia is the impairment of galactose-1-phosphate uridylyltransferase (GALT), which catalyzes the second step in the Leloir pathway of galactose metabolism. Traditional laboratory confirmation of the disease is established by measurement of the enzyme activity in red blood cells (3). Recently, a sensitive radiochemical method has been developed which can detect GALT activity as low as 0.1% of normal in both
1 Present address: Genzyme Genetics, 2000 Vivigen Way, Santa Fe, NM 87505. 2 To whom correspondence and reprint requests should be addressed at Director, Molecular Diagnostics Laboratory, Institute for Molecular and Human Genetics, Georgetown University Medical Center, 3800 Reservoir Road NW, Suite 4000, Washington, DC 20007. Fax: (202) 784-1770. E-mail: [email protected].
263 1096-7192/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
MGM 2678
/
af11$$$$81
04-29-98 08:12:52
bmmal
AP: MGM
264
WANG ET AL.
erythrocytes and leukocytes (4). This method is extremely valuable in the determination of minute amounts of GALT activity in a biochemically heterogeneous galactosemia population. The human GALT gene has been mapped to chromosome 9p13 (5). The gene, containing 11 exons and spanning 4 kb, has been cloned and sequenced (6). More than 50 point mutations have been identified by SSCP (single-strand conformation polymorphism) analysis and sequencing (7–11). Many of the mutant cDNAs have been expressed in COS cells (8) and yeast (12) to determine the functional consequences of the mutations. Galactose-1-phosphate uridylyltransferase functions as a dimer which is structurally stabilized by a pair of Zn/Fe (13). In the Escherichia coli enzyme, two UMP molecules are covalently bound per uridylyl enzyme intermediate (14,15). Although the overall amino acid sequence identity of the E. coli, yeast, and human enzymes is only 35% (16), the structurally and functionally conserved domains have been described. Studies have shown that mutations in the structurally and functionally conserved regions result in low or absent enzyme activity (17,18). It has been hypothesized that different mutations lead to biochemical and clinical heterogeneity depending on the nature and position of the altered amino acid. The most common mutation in GALT is Q188R, which has a prevalence of 60–70% in galactosemic Caucasians (6,19). Patients homozygous for Q188R have essentially no detectable enzyme activity in erythrocytes and lymphocytes. This subgroup of galactosemic patients has a more severe clinical phenotype, including reduced intellect and ovarian failure. In contrast, patients with the African-American-specific mutation, S135L, generally have had a mild clinical course. Since the number of patients studied in previous reports was small (10), the incidence of many mutations, such as the S135L allele among Caucasians and the Q188R allele among African Americans, as well as the enzyme activities in patients with various genotypes, remain unclear. In order to unravel these questions, 293 galactosemic individuals were screened for 17 known point mutations in exons 5, 6, and 10, and the obtained genotypes were correlated with biochemical phenotypes. Here we report the allele frequency of each of these point mutations in Caucasian, Hispanic, and African-American patients. The biochemical implications of various genotypes are discussed. MATERIALS AND METHODS This study was carried out according to the protocols, CCI 86-773 and CCI 90-117, approved by the
AID
MGM 2678
/
af11$$$$81
04-29-98 08:12:52
Committee on Clinical Investigation at Children’s Hospital, Los Angeles. Over the years, we have identified about 1000 galactosemia patients. Blood samples were available from approximately half of them. A total of 293 patients, including 210 Caucasians, 28 Hispanics, and 55 African-Americans, were studied. The GALT activity in the red blood cells of these patients was measured using a sensitive radiochemical assay (4). Patient DNA was extracted from either the blood spots or whole blood (20,21). A simple, rapid, and cost-effective multiplex PCR/ASO (allele-specific oligonucleotide) hybridization method (22) was employed to screen 17 known point mutations in exons 5, 6, and 10 of the GALT gene. These exons were chosen because they are apparently the mutation hot spots and they contain the two most common mutations, Q188R and S135L. The forward and reverse primers for PCR amplification of exons 5 and 6 were 5*CTGCCCGTAGCACAGCCA3* and 5*GCTGGCTCAGACTCAGCC3*, respectively. The primers for exon 10 are 5*GGGTTTGGGAGTAGGTGCT3* and 5*GGGCAACAGAAGTATCAGGT3*. PCR conditions and ASO dot blot hybridization conditions were similar to the previously published procedures (22). Briefly, 96-well plates were used for multiplex PCR amplification using a Perkin–Elmer 9600 thermal cycler. Each 100-ml PCR mixture contained 11 Promega PCR buffer, 1.5 mM MgCl2 , 0.2 mM concentrations of each dNTP, 0.5 mM concentrations of each of the two pairs of primers stated above, 1 unit of Taq DNA polymerase, and 100 ng of genomic DNA. The reaction mixture was denatured at 947C for 2 min, followed by 32 cycles of 30 s of denaturation at 947C, 30 s of reannealing at 547C, and 40 s of extension at 727C. The PCR was completed by a final extension cycle at 727C for 4 min. Two microliters of PCR product was spotted onto a Biodyne B/ membrane. Dot blots were prepared for each of the normal and mutant probes listed in Table 1. The ASO probes were end-labeled with [g-32P]ATP using T4 polynucleotide kinase. RESULTS Figure 1 shows representative results of PCR/ ASO dot blot analysis for the Q188R and F171S mutations. Only 7 (Q188R, S135L, F171S, N314D, R148W, R333G, and R333W) of 17 mutations screened were detected in our patient population. Table 2 summarizes the mutant allele frequencies in a total of 293 galactosemia patients. The overall detection rate was 65%. The detection rate in each
bmmal
AP: MGM
265
MOLECULAR MECHANISM OF GALACTOSEMIA
TABLE 1 Nucleotide Sequence of the ASO Probes Probe
Sequence (5* to 3*)
Sense (S) or antisense (A)
S135 S135L T138, L139, M142 T138M L139P M142K R148, V151 R148W R148Q V151A F171 F171S Q188 Q188R E308 E308K N314 N314D H319 H319Q A320 A320T Y323 Y323D R333, K334 R333G R333W K334R
CCCCTGGTCGGATGTAAC CCCCTGGTTGGATGTAAC TGTAACGCTGCCACTCATGTCG GGATGTAATGCTGCCACT ATGTAACGCCGCCACTCA CCACTCAAGTCGGTCCC GATCCGGGCTGTTGTTGATG TGAGATCTGGGCTGTTGT TGAGATCCAGGCTGTTGT GGCTGTTGCTGATGCATG GATAGATCTTTGAAAACAAAGG GATAGATCTCTGAAAACAAAGG ACCCTTACCTGGCAGTGG ACCCTTACCCGGCAGTGG CCCAGCCTCTGATCCTGT CCCAGCCTTTGATCCTGT CCAATGGTTCCAGTTGGC CCAATGGTCCCAGTTGGC CAGCTGCACGCTCATTAC CAGCTGCAAGCTCATTAC CAGCTGCACGCTCATTACT CAGCTGCACACTCATTACT CGGAGGGTAGTAATGAGC CGGAGGGTCGTAATGAGC ATGAATTTCCGGACAGTGG ATGAATTTCCCGACAGTGG ATGAATTTCCAGACAGTGG ATGAATCTCCGGACAGTGG
S S S S S S S S S S S S A A A A A A S S S S A A A A A A
ethnic group was 67% in Caucasians, 48% in Hispanics, and 68% in African-Americans. As shown in the table, the most prevalent mutant allele, Q188R, accounts for 63% of the Caucasian GALT mutant alleles in our population. Forty-two percent of the Caucasian patients are homozygous and 42% are compound heterozygous for the Q188R mutation. The S135L mutation, the most common mutant allele in African-American GALT patients, is present in a small number of Caucasian patients, accounting for 1.4% (6/420) of the mutant alleles. In African-American patients, the most common mutation is S135L, accounting for 45% of the mutant alleles, followed by Q188R (20%) and F171S (3.6%). One homozygote and two compound heterozygotes for the F171S mutation were found in 55 AfricanAmerican patients. The F171S mutation has not been detected among any Caucasian or Hispanic patients in this study. The only mutation identified among the Hispanic patients was Q188R, for which 28% patients were homozygous and 46.4% were compound heterozygous for this mutation. Among all
AID
MGM 2678
/
af11$$$$82
04-29-98 08:12:52
three populations, the unidentified mutations were prevalent, including 33% of Caucasian, 32% of African-American, and 52% of Hispanic mutant alleles. There have been questions regarding the predictability of the biochemical and clinical phenotypes associated with particular genotypes. Conflicting results for the enzymatic activity of the S135L mutant allele have been reported (7,12). In this study, we measured the red blood cell transferase activity in 258/293 (88%) of our patients using a more sensitive assay. Table 3 lists the genotypes and the red blood cell GALT activities in the galactosemia patients. Our data indicate that all of the Q188R homozygotes and most (85%) of the compound heterozygotes have essentially no GALT activity. Residual enzyme activity was observed in two patients with Q188R/ R333G but not in the single patient with the Q188R/ R333W genotype. The arginine (R) to tryptophan (W) change results in the drastic replacement of a highly basic, positively charged amino acid with a very hydrophobic aromatic amino acid. A complete loss of enzyme activity may be predicted.
bmmal
AP: MGM
266
WANG ET AL.
FIG. 1. ASO dot blot analysis of point mutations in the GALT gene. (A and B) Hybridization with Q188R normal and mutant probes, respectively. A1, D5, and C5 are, respectively, examples of a normal, heterozygote, and homozygote for the Q188R mutation. A12, C4, and H6 are no-DNA controls. (C and D) Hybridization with F171S normal and mutant probes, respectively. G6 is a heterozygote for the F171S mutation. A1 and H7 contain no DNA template in the PCR.
Consistent with its milder clinical outcome, patients homozygous for the S135L mutant allele demonstrate residual red blood cell GALT activity (10). However, when the S135L allele is found in compound heterozygotes with Q188R, no GALT activity was detected. Variable amounts of residual enzyme activity are present in patients with an S135L/unknown genotype. Patients homozygous for F171S or compound heterozygous for F171S/Q188R have no
enzyme activity, suggesting that F171S is a severe mutation, consistent with previous data (18). DISCUSSION PCR/ASO dot blot analysis is a simple, accurate, sensitive, and cost-effective method for the screening of multiple, recurrent, known mutations. In this report, we multiplex-amplified exons 5/6 and 10, where
TABLE 2 Mutant Allele Frequencies in GALT Gene Caucasian
Hispanic
African-American
Total
Mutation
No. alleles
%
No. alleles
%
No. alleles
%
No. alleles
%
Q188R S135L F171S N314D R333G R333W R148W Total identified Unidentified
266 6 0 4 2 1 1 280 140
63 1.4 0 1 0.5 0.2 0.2 67 33
27 0 0 0 0 0 0 27 29
48 0 0 0 0 0 0 48 52
22 49 4 0 0 0 0 75 35
20 45 3.6 0 0 0 0 68 32
315 55 4 4 2 1 1 382 204
53.75 9.39 0.68 0.68 0.34 0.17 0.17 65.18 34.81
Total
AID
420
MGM 2678
/
af11$$$$82
56
04-29-98 08:12:52
110
bmmal
AP: MGM
586
267
MOLECULAR MECHANISM OF GALACTOSEMIA
TABLE 3 Correlation of Genotype with Red Blood Cell Transferase Activity in Galactosemia Patients Number of patients Mutation (allele 1/allele 2)
Caucasian
Q188R/Q188R Q188R/S135L Q188R/R333G Q188R/R333W Q188R/R148W Q188R/unknown S135L/S135L S135L/F171S S135L/unknown F171S/F171S Unknown/unknown
AfricanAmerican
89 3 2 1 1 81
3 30
4 10
7
4 10 2 17 1 7
13
most common mutations occur. Seventeen known point mutations in this region were screened in 293 galactosemia patients. Only 7 different mutations, Q188R, S135L, N314D, F171S, R333W, R333G, and R148W, were detected. This suggests that most or all of the other 10 mutations are private mutations with almost no recurrence rate. The overall detection rate for the 7 point mutations is 67–68% for Caucasians and African-Americans, but only 48% for Hispanics (Table 2). The lower detection rate among Hispanics is similar to that currently noted for other diseases, such as cystic fibrosis, phenylketonuria, and glycogen storage disease type 1a, and is likely secondary to either a greater genetic heterogeneity or that mutations to be screened are typically identified by centers whose patient population is predominately Caucasian and African-American. For the molecular diagnosis of galactosemia, we recommend that the first step should be PCR/ASO dot blot analysis for the 7 recurrent mutations. If zero or only one mutant allele is detected, then SSCP or other mutation detection methods should be used to screen for unknown mutations in all 11 exons, followed by direct DNA sequencing to confirm and identify the mutation. We have recently identified several novel mutations using the TTGE (temporal temperature gradient gel electrophoresis) method (to be published elsewhere). The small percentage (1.4%) of S135L alleles found in the Caucasian galactosemia patients could be due either to racial admixture or to an independent mutation which arose in the Caucasian population. On the other hand, Q188R, which accounts for 20% of the mutant GALT alleles in the African-
AID
MGM 2678
/
af11$$$$82
Hispanic
04-29-98 08:12:52
8
GALT activity (% of normal in RBC) 0 0 0.3 0 0 0–22 0.2–1.7 0 0–17 0 0–29
American galactosemia population, is probably due predominately to racial admixture considering the high frequency of Caucasian genes among AfricanAmericans (23). Of particular importance is to correlate the molecular genotypes with their corresponding GALT activities. Because of the large number of patients studied, we are able to examine the enzyme activity in various homozygous and compound heterozygous combinations. Our results suggest that subunit interaction plays an important role in determining the enzyme activity. Although S135L homozygotes show a residual enzyme activity of 0.2–1.7% of normal activity, the activity of the S135L compound heterozygote varies from 0 to 17% of normal controls (Table 3). The compound heterozygotes S135L/Q188R and S135L/F171S show no detectable enzyme activity (Table 3). Similarly, the enzymatic activity of the Q188R compound heterozygote varies from 0 to 20%. These data suggest that there is possibly a dominant negative effect of one mutation over the other by heterodimeric interaction. Assuming that the S135L and the Q188R polypeptides are produced in equal amounts, homodimers and heterodimers are formed without preference, and there is no subunit predominant effect, one would expect that the S135L/Q188R heterozygote would have an enzyme activity of 50% that of a S135L homozygote, which is not the case. If the Q188R mutation inactivated the S135L-containing subunit in a Q188R/S135L heterodimer and there is no preference towards homo- or heterodimer formation, then one would expect the enzyme activity to be 25% that of a S135L homozygote, since only the S135L homodimer is functional. A possible ex-
bmmal
AP: MGM
268
WANG ET AL.
planation for the zero enzyme activity of the Q188R/ S135L compound heterozygote is that heterodimer formation is much more favorable than homodimer formation. Within the heterodimer, the Q188R mutant subunit either structurally destabilizes and facilitates degradation of the S135L subunit or functionally impairs the catalytic ability of the heterodimer. It is speculated that the heterodimer has higher affinity, which shifts the equilibrium toward the nonfunctional heterodimeric state by physically disabling the function of the other subunit. In addition, GALT enzyme activity varies from tissue to tissue. Thus tissue-specific cellular protein factors may also play important role in the regulation of GALT gene expression at the transcription, translation, or posttranslation level. Further studies of GALT activity should be extended from the red blood cells to include the nucleated cells, such as white blood cells or cultured skin fibroblasts. Not all Q188R heterodimers exhibit zero enzyme activity. About 15% of the Q188R compound heterozygous patients show residual activity. This observation is also consistent with our hypothesis that enzyme activity depends on the heterodimeric interaction. Which subunit is dominant over the other depends on the type of combination of the mutant protein subunits. Identification of the mutation in the second mutant allele will assist in the interpretation of the biochemical phenotypes in these patients. It is apparent that the enzyme activity depends on the combination of alleles in a given patient. This may partly explain the heterogeneous clinical outcome of patients who carry one common mutant allele. Studies to correlate the biochemical and molecular findings with patients’ clinical outcomes are currently being undertaken. ACKNOWLEDGMENTS The authors thank the clinicians who referred the patients to us.
REFERENCES 1. Levy HL, Hammersen G. Screening for galactosemia. In Inherited Disorders of Carbohydrate Metabolism (Burman D, Holton JB, Pennock CA, Eds.). Lancaster, England: MTP 1980, pp 133–139.
4. Xu YK, Kaufman FR, Donnell GN, Ng WG. Radiochemical assay of minute quantities of galactose-1-phosphate uridyltransferase activity in erythrocytes and leukocytes of galactosemia patients. Clin Chim Acta 235:125–136, 1995. 5. Sparkes RS, Sparkes MC, Funderburk SJ, Moedjono S. Expression of GALT in 9p chromosome alterations: assignment of GALT locus to 9cen r 9p22. Ann Hum Genet 43:343–347, 1980. 6. Leslie ND, Immerman EB, Flach JE, Florez M, FridovichKeil JL, Elsas LJ. The human galactose-1-phosphate uridyltransferase gene. Genomics 14:474–480, 1992. 7. Reichardt JKV, Woo SLC. Molecular basis of galactosemia: mutations and polymorphisms in the gene encoding human galactose-1-phosphate uridyl transferase. Proc Natl Acad Sci USA 88:2633–2637, 1991. 8. Reichardt J, Belmont J, Levy H, Woo S. Characterization of two missense mutations in human galactose-1-phosphate uridyl transferase: different molecular mechanisms for galactosemia. Genomics 12:596–600, 1992. 9. Elsas LJ, Langley S, Steele E, Evinger J, Fridovich-Keil JL, Brown A, Singh R, Fernhoff P, Hjelm LN, Dembure PP. Galactosemia: A strategy to identify new biochemical phenotypes and molecular genotypes. Am J Hum Genet 56:630– 639, 1995. 10. Lai K, Langley SD, Singh RH, Dembure PP, Hjelm LN, Elsas LJ. A prevalent mutation for galactosemia among black Americans. J Pediatr 128:89–95, 1996. 11. Maceratesi P, Sangiuolo F, Novelli G, Ninfali P, Magnani M, Reichardt JKV, Dallapiccola B. Three new mutations (P183T, V150L, 528insG) and eleven sequence polymorphisms in Italian patients with galactose-1-phosphate uridyltransferase (GALT) deficiency. Hum Mut 8:369–372, 1996. 12. Fridovich-Keil JL, Langley SD, Mazur LA, Lennon JC, Dembure PP, Elsas LJ II. Identification and functional analysis of three distinct mutations in the human galactose-1phosphate uridyltransferase gene associated with galactosemia in a single family. Am J Hum Genet 56:640–646, 1995. 13. Frey PA. The Leloir pathway: a mechanistic imperative for three enzymes to change the stereochemical configuration of a single carbon in galactose. FASEB J 10:461–470, 1996. 14. Wong LJ, Sheu KF, Lee SL, Frey PA. Galactose-1-phosphate uridylyltransferase: Isolation and properties of a uridylylenzyme intermediate. Biochemistry 16:1010–1016, 1977. 15. Frey PA, Wong LJ, Sheu KF, Yang SL. Galactose-1-phosphate uridylyltransferase: Detection, isolation and characterization of the uridylyl enzyme. Methods Enzymol 87:20– 36, 1982. 16. Flach JE, Reichardt JKV, Elsas LJ. Sequence of a cDNA encoding human galactose-1-phosphate uridylyltransferase. Mol Biol Med 7:365–369, 1990.
2. Segal S, Berry GT. Disorders of galactose metabolism. In The Metabolic and Molecular Bases of Inherited Disease (Scriver CR, Beaudet AL, Sly WS, Valle D, Eds.), 7th ed., Vol. 1. New York: McGraw-Hill, 1995, pp 967–1000.
17. Reichardt JKV, Packman S, Woo SLC. Molecular characterization of two galactosemia mutations: correlation of mutations with highly conserved domains in galactose-1-phosphate uridyltransferase. Am J Hum Genet 49:860–867, 1991.
3. Lee JES, Ng WG. Semi-micro techniques for the genotyping of galactokinase and galactose-1-phosphate uridyl transferase. Clin Chim Acta 124:351–356, 1982.
18. Reichardt JKV, Levy H, Woo S. Molecular characterization of two galactosemic mutations and one polymorphism: Implications for structure-function analysis of human galactose-1-
AID
MGM 2678
/
af11$$$$82
04-29-98 08:12:52
bmmal
AP: MGM
MOLECULAR MECHANISM OF GALACTOSEMIA phosphate uridyl transferase. Biochemistry 31:5430–5433, 1992. 19. Ng WG, Xu YK, Kaufman FR, Donnell GN, Wolff J, Allen RJ, Koritala S, Reichardt JKV. Biochemical and molecular studies of 132 patients with galactosemia. Hum Genet 94:359–363, 1994. 20. Gregersen N, Blakemore A, Winter V, Andersen BS, Kolvraa S, Bolund L, Curtis D, Engel PC. Specific diagnosis of medium-chain acyl-CoA dehydrogenase (MCAD) deficiency in dried blood spots by a polymerase chain reaction (PCR) assay
AID
MGM 2678
/
af11$$$$83
04-29-98 08:12:52
269
detecting a point-mutation (G985) in the MCAD gene. Clin Chim Acta 203:23–34, 1991. 21. Lahiri D, Nurnberger J Jr. A rapid non-enzymatic method for the preparation of HMW DNA from blood for RFLP studies. Nucleic Acids Res 19:5444, 1991. 22. Wong LJC, Senadheera D. Direct detection of multiple point mutations in mitochondrial DNA. Clin Chem 43:1857–1861, 1997. 23. Chakraborty R, Kamboh MI, Nwankwo M, Ferrell RE. Caucasian genes in American blacks: new data. Am J Hum Genet 50:145–155, 1992.
bmmal
AP: MGM