The brush-border carbohydrate hydrolases are a family of small-intestinal enzymes that convert disaccharides and oligosaccharides into absorbable monosaccharides for absorption (1). The enzymes, maltase-glucoamylase (MGA), lactase-phlorizin hydrolase (LPH), and sucrase-isomaltase (SI), are encoded by the genes MGAM, LCT and SI that map to chromosomes 7, 2, and 3, respectively (2–4). Isolated genetically determined deficiencies of each of the enzymes are known (4–9).
A clinical disorder of starch oligomer digestion was described in 1994 (10). A reduction in mucosal hydrolysis of glucose oligomers, glucoamylase activity, was discovered in 2% of biopsies obtained from children suffering from chronic diarrhea. The biopsies in these subjects were morphologically normal (10). In this population of infants with chronic diarrhea, the incidence of congenital MGA deficiency was equivalent to that of congenital SI deficiency. Half of the infants with low glucoamylase activity had clinical carbohydrate intolerance with symptoms identical to those of patients with congenital lactase deficiency when receiving dietary loads of the offending carbohydrate. Two of the nine MGA-deficient patients had dual deficiencies with either low LPH or SI activities. In large studies of SI deficiency, LPH deficiency was also found in about 15% of congenital sucrase deficiency patients (9). In 16 patients with congenital lactase deficiency, concomitant SI deficiency was also detected in one patient; another patient had clinical intolerance to Nutramigen (11).
This report describes the first infant with congenital maltase-glucoamylase deficiency to be investigated at the molecular level. Biopsy lactase and sucrase-isomaltase activities were also reduced.
MATERIALS AND METHODS
The propositus was an African-American male admitted at 2.5 months of age to the Ben Taub General Hospital (a teaching unit of Baylor College of Medicine) in Houston, Texas, with chronic diarrhea since birth. The infant was being fed lactose-containing formula. Genetic history was incomplete but one informant suggested the possibility of consanguinity between the parents. At the time of admission, he weighed 3.2 kg, 300 g less than his birth weight. His stools were large, watery, were of low pH and contained glucose. He was severely acidotic. He was started on total intravenous nutrition, which resulted in a prompt resolution of diarrhea and acidosis. He began to gain weight. These presenting symptoms were consistent with congenital lactase deficiency (11). After two weeks of parenteral nutrition, he was changed to oral feedings of Nutramigen, which he tolerated. However, he failed to sustain weight gain despite the same level of energy intake. We suspected he might have MGA deficiency, and the formula was supplemented with 2% wt/vol glucose, the end product of MGA enzymatic activity. He responded with a prompt gain in weight and length. The response to glucose supplementation increased the suspicion of a coexisting congenital MGA deficiency. Endoscopic biopsies of the jejunum were obtained. The gross appearance of the distal duodenum was normal. Histologic examination revealed normal villous length, and no evidence of inflammation was seen. Laryngomalacia and stridor were a prominent feature from birth, and a feeding gastrostomy was placed at 4 months because of repeated aspirations of oral formula and pneumonitis. As of this report, at 36 months, he remains stunted with a length paralleling but well below the fifth percentile and weight/length was greater than the 95th percentile. There is significant psychomotor developmental delay. Because of recurrent aspiration pneumonias, he continues to be fed by gastrostomy.
The infant had excellent clinical tolerance of a formula containing 10% glucose oligomer and 2% glucose. The glucose supplement was discontinued at 6 months of age. At 3 and 6 months he was challenged for 24-hours by replacing half of the starch oligomers in the formula with 5% sucrose. This resulted in vomiting on both occasions. At 12 months, after a sucrose-loading test, the infant developed severe colic that lasted for 12 hours. The infant became progressively obese on the 10% oligomer formula, and Nutramigen was replaced with formula 3232A® (Mead-Johnson, Evansville, IN) with 5% starch oligomers at 6 months. He is physically inactive and is currently fed Vivonex® (Novartis Nutrition, St. Louis Park, MN) with 5% starch oligomers at 55 kcal/kg/d to prevent obesity.
Endoscopic biopsies were obtained from the distal duodenum. The tissue was oriented and submitted for routine sections and stains. The Texas Children's Hospital Pathology Department (Houston, Texas) consultants evaluated serial sections of biopsies obtained at 4.5 and 12 months of age.
Enzyme activity was determined in the supernatant of tissue or cell homogenates using 2% sucrose, lactose, maltose, starch oligomers (Polycose®, Ross, Columbus, OH), or soluble starch (Sigma Chemical Co., St. Louis, MO) as substrates (12–16). Enzyme-specific activity was expressed as U (mole min−1)/g protein (U/g protein). Normal activity cutoff values are from children biopsied for clinical dyspepsia that had normal histology (unpublished).
Starch Digestive Capacity
Studies of 13C-glucose absorption and its subsequent oxidation to 13CO2 in breath of children with congenital and acquired glucose malabsorption have been previously reported (17). We modified this method by using 13C-starch/starch oligomers as the loading substrate. Control studies used 13C-glucose/glucose or starch oligomers as the loading substrate (17). The starch was derived from tobacco plants grown in an environment enriched in 13CO2 (MSD, Montreal, Canada). The starch had a labeled 13C enrichment of 78% APE. A load of 2 mg/kg 13C-label was added to a feeding which provided 2 g/kg of unlabeled starch oligomer in Nutramigen formula. As an experimental control, 13C-glucose (78% APE, MSD) was administered as previously described (17). Breath specimens were collected at −2, −1, 1, 2, 3, 4, 5, 6, 7, and 8 hours after each test feeding. An age-matched infant with a gastric feeding tube served as a control subject. Both infants maintained their routine four-hour feeding schedules before and during each breath test. The breath samples were analyzed for 13CO2 by Europa 2020 Isotope Ratio MS (Crewe, UK) and for H2 by QuinTron MicroLizer (Milwaukee, WI). The breath 13CO2 enrichments were reported as delta over baseline (DOB).
Search for MGA Mutations
RNA from intestinal biopsy and DNA from blood were extracted and studied by RT/PCR or genomic PCR as previously described (12,13). Genomic PCR used exon specific primers (unpublished). The RT/PCR primer sets were those used in the original sequencing of MGA (2). The full cDNA for each subject was sequenced from intestinal biopsy RNA with eight primer sets that produced overlapping amplimers. The amplimers were gel purified and reamplified with the same primer sets to provide final DNA concentrations adequate for bi-directional direct sequencing. All sequencing was done on a Perkin-Elmer Model 373 version 3 instrument (Foster City, CA). The coding region was sequenced in the propositus and 9 control subjects with normal maltase or glucoamylase activities in mucosal biopsies with normal histology. These sequences were used to detect mutations in the propositus.
The MGA promoter region was sequenced from genomic DNA using the GenomeWalker kit (Clontech) to produce clones of the 5´ upstream region to ∼−2000 bp (GenBank accessions AF432182–432202). Gene primers were used to amplify the −580 through exon 1 region (+43) from genomic DNA from three subjects with normal glucoamylase activity and from the propositus, his mother and grandmother for sequencing.
Activity of Wildtype and Mutant cDNA Constructs Expressed in COS-1 Cells
To determine whether a C1673T mutation altered enzyme activity, this mutation was produced by the method of recombinant PCR (18–20) using the eukaryotic expression vector MGA-P1A2 as a template that encodes the N-terminal domain of MGA. Two of the obtained clones were expressed in COS-1 cells. The expressed mutant form and its ability to hydrolyze maltose, starch, and Polycose were compared with wildtype MGA-P1A2. Cells were labeled with 50 μCi of [35S]methionine (NEN, Boston, MA). The cells were removed from the dishes, pelleted by centrifugation in the presence of protease inhibitors. The homogenate was incubated with monoclonal antiserum coupled to protein A Sepharose beads. The expressed proteins were analyzed on 7.5% SDS-PAGE under reducing conditions. The enzymatic activity was measured directly in cell homogenates (20).
MGA, SI, and LPH Synthesis and Processing in Organ Culture
The tissue specimen was washed in methionine-free medium supplemented with streptomycin (100 μg/ml), penicillin (100 units/ml), and 10% dialyzed fetal calf serum (Sigma). The tissue was viewed under a dissecting microscope to orient the villous surface upwards on a stainless steel grid. The tissue was preincubated for two hours in mixed CO2+O2 (5:95, v/v) in a humidity box at 37°C. The tissue was then labeled by adding 100 μCi of 35S-methionine/cystine (NEN). After 8 hours, the tissue was rinsed with homogenization buffer and protease inhibitor mixture. The labeled specimens were homogenized and lysis buffer added. The homogenate was centrifuged and the supernatant was retained. The protein concentration of the supernatant was measured with the BCA Protein Assay (Pierce, Rockford, IL). The supernatant was sequentially extracted with monoclonal antibodies coupled in Protein-A Beads (Sigma). The antigen-antibody beads were run on 6% SDS-polyacrylamide gel electrophoresis and dried for autoradiography.
All clinical procedures were reviewed and approved by the Baylor College of Medicine Institutional Review Board for Human Investigation and the parent or legal guardian signed informed consent in each case.
Histologic sections of jejunal biopsies obtained at 4 and 12 months revealed normal villus structures with no increases of intraepithelial or lamina propria lymphocytes. The brush border appeared normal on trichrome stains.
Mucosal Enzyme Activities
Assays of the homogenates from the patient's biopsy revealed reduced substrate hydrolysis (Table 1) at 4 and 12 months of age. Because all measured activities were below the normal range at 4 months the results were interpreted as consistent with a congenital pandisaccharidase deficiency. Sucrase and maltase activity increased to the lower limit of normal at 12 months but lactose, starch, and Polycose hydrolysis remained very low. The second biopsy was processed in parallel with three control tissues (see legend to Fig. 2) to exclude systematic errors in tissue handling or analysis.
Starch Digestion Capacity
The propositus was challenged with 13C-starch loads added to a bolus feeding of starch oligomer containing formula on three occasions between 3.5 and 7.5 months of life. Breath was sampled at intervals and analyzed for 13CO2 and H2. The results were compared with a control infant of comparable age and formula composition. In both subjects, a gastric feeding tube was used to deliver the bolus feeding. The control subject had a peak 13CO2 (DOB) at 100 minutes that continued at about 50 DOB until 200 minutes and then declined rapidly to baseline. The propositus had peaks of 13CO2 at 200 to 300 minutes with maximum levels less than 40 DOB, which were still above baseline at 400 minutes. The breath 13CO2 responses were quantitated as % of 13C-dose excreted in breath during 3 hours. For the control, 60% of the dose was excreted. For the propositus, the 13CO2 excretion varied from 5 to 25% of the dose or 8 to 40% of the quantity of starch digested by the control subject. Replicate starch breath tests in the propositus had a coefficient of variation of <13%. 13C-glucose loads were given to the propositus and control to exclude glucose malabsorption (data not shown). The 13CO2 excretion after 13C-glucose was the same in the two subjects. The coefficient of variation for duplicate 13C-glucose loads given to the propositus was <8%. Breath H2 was measured for 6 hours after 3 separate bolus feedings of glucose oligomer formula feedings in the propositus and control subject. The results of the H2 breath tests were averaged. The propositus had a mean level of >15 ppm at all time points with peaks >20 ppm at about 180 minutes after each feeding. The control subject never exceeded 10 ppm and there were no overlapping points between the propositus and the control in the unaveraged H2 concentrations.
Coding Region Sequencing
The full cDNA sequence from the propositus was compared with the coding region of nine subjects with normal maltase or glucoamylase activities and the resequenced cDNA reported in our original publication (2). Single nucleotide differences (SNP) were detected in some of the samples (GenBank SS# 4318818–4318840). A homozygous single base change C1673T was detected in the propositus. This translates to a S542L mutation which is located 12 AA after the proton donor aspartic acid (D) in signature 1 site WIDMNE (2). This mutation is in a loop following a β-sheet 4 that normally contains 4 serines. The presence of the point mutation was confirmed with DNA from the propositus and his mother using an exon specific primer pair and restriction enzyme Bsa I (Fig. 1). As revealed in the same figure, his asymptomatic mother shared this mutation.
Expression of Mutant cDNA in COS Cells
The N-terminal coding region of MGA was used to introduce the C1673T mutation. The COS-1 cells expressing the C1673T mutant constructs had the same protein isoforms (not shown) and enzyme activity levels for maltose, starch, and starch oligomer substrates as the wildtype construct. A negative control experiment, without an insert in the vector, had negligible enzyme activities for the same substrates.
The 5´ upstream promoter region of the gene was sequenced to ∼−2000 by the GenomeWalker (Clontech) technique. For the purpose of this study, a primer set was used which amplified a 580-bp region that extended from the 3´ end of exon 1 to −577. Genomic DNA from three controls, which had normal glucoamylase activities, and also from the propositus, his mother and grandmother were amplified and directly sequenced. There were no differences between the promoter sequences of the propositus and that of controls.
Isoforms of Brush Border Glucohydrolases
Because of simultaneous reductions in the activities of lactase, sucrase, and glucoamylase in this propositus, a screen for mutations that truncate message or block processing was performed in all three coding regions. This consisted of biopsy culture in the presence of 35S-labeled sulfur amino acids for 8 hours followed by sequential immunoprecipitations from homogenates, normalized for protein concentration, with excess quantities of monoclonal antibodies for MGA, LPH, and SI as previously described (14–16,20). The isolates were separated by SDS-PAGE and analyzed by fluorography. The fluorogram is shown in Figure 2. In this experiment, biopsies from three subjects with normal histology served as controls. All of the enzyme activities of the controls were in the normal range. Two biopsies were studied from the propositus; the second was treated with endoglycosidase H (Endo H) after immunoisolation to identify the high mannose isoforms (15,16,21).
The LPH isoforms were of the expected size in each immunoisolate with the high mannose form at about 160 kDa, the mature complex glycosylated isoform at about 215, and a faint dimeric form at about 335 kDa. The LPH high mannose isoform was identified in the propositus biopsy by a reduction in size after treatment with Endo H (15). The SI isoforms were of the expected sizes in each immunoisolate, with the high maltose isoforms visualized at about 210 kDa and the complex glycosylated forms at about 245 kDa. The identity of the SI high mannose form was confirmed in the propositus biopsy by a reduced size after Endo H treatment (14). The isoforms of the MGA proteins were of the expected sizes of about 285 kDa and 335 kDa in the propositus and control 4 (16). The treatment of the propositus biopsy with Endo H resulted in the expected reduced size of the 285 isomer, the high mannose isoform. Controls 2 and 3 had additional MGA isoforms at about 210 and 265 kDa. This variant pattern occurs in controls with normal glucoamylase activity and is thought to be a variant of MGA dimerization (2). No qualitative variants of protein isoforms or processing to the mature brush-border peptides were detected by organ culture labeling of biopsies from the propositus. There were no correlations between protein band pattern or intensity and enzyme activities for the four specimens and three enzyme activities studied.
The clinical history of this child, normal villus histology and low activities of three enzyme complexes all suggest combined congenital lactase, sucrase and maltase-glucoamylase deficiencies. The activities of the mucosal homogenate at 12 months when the patient was well nourished were increased but still below normal for subjects with normal histology. It is unclear whether the lower values at 4.5 months were due to developmental or nutritional factors but in our experience the loss of brush border enzyme activity in marasmic infants is always associated with villus atrophy (12,13).
Reduced proximal digestion of starch was first suggested by his failure to gain on adequate intakes of a 10% glucose oligomer formula followed by rapid gains with 2% supplemented glucose. This response to glucose supplementation suggests that while the colonic salvage pathway of digestion was functional it was calorically inefficient. The literature suggests that in adults, carbohydrate digested through the colonic pathway has an energy density of only 2.0 cal/g (22). Our results in this infant suggest that the caloric density of glucose polymers digested in the colon might be closer to 3.0 cal/g because a 17% increase in available glucose calories compensated for the colonic oligomer digestion energy deficit and restored weight gain.
All previously reported patients with glucoamylase deficiencies were first investigated because of chronic diarrhea (10). Our patient had chronic diarrhea on lactose but not on glucose oligomer formulas. Carbohydrate intolerance requires a proximal enzyme deficiency and a colonic failure of the salvage pathway. The mechanism of colonic failure of carbohydrate digestion is unresolved; however, it is usually associated with diarrhea, reducing substances in the stool, low fecal pH, and failure to produce breath H2. These findings were present in this patient on lactose and sucrose but not on glucose oligomer formulas. An obvious mechanism might be the greater luminal osmotic force exerted by unabsorbed lactose or sucrose than the larger starch oligomers.
Cases of multiple intestinal glucohydrolase enzyme deficiencies have been previously reported (10,11,30–32). Of particular interest is a frequent association of MGA deficiency reported in patients with congenital SI deficiency (30–32). This has led to recognition of two phenotypes of CSID, patients with or without coexisting starch intolerance (9). There has been one prior case report of a patient with combined lactase, sucrase and glucoamylase deficiencies. This woman, studied at 21 years of age, had sucrose-induced diarrhea from childhood (32).
Search for Mutations
Although the observation of reduced activity of more than one disaccharidase do not point strongly to a causal mutation within the MGAM gene, the possible consanguinity in our patient's parents made the coincidental occurrence of two or more independent mutations in genes encoding the different disaccharidases possible. We therefore screened MGAM extensively for mutations in the MGA coding region and proximal promoter. None were found that could be associated with reduced enzyme activity. A unique single base change of C1673T was detected in the propositus that was absent in all control subjects. This translates to a S542L mutation located 12 AA after the acidophile aspartic acid (D) in catalytic site WIDMNE. No loss of enzyme activities could be detected after the expression of two clonal lines of the C1673T mutation in the N-terminal construct. Also the mother, as well as the propositus, was homozygous for this change. Further analysis showed the substitution to be present in other individuals of European, Indian, and African ancestry (unpublished).
We also tested the expression of the three enzymes in organ culture and could recognize no alterations of the pattern of synthesis or processing of any of these enzymes. Failure of the MGA protein patterns, on autoradiograms, to predict the glucoamylase enzyme values from the same subject have been found in a larger study (unpublished) using identical methods of investigation. The molecular observations support the hypothesis that the defect in the propositus is likely to reside in genetic or epigenetic factors up the regulatory cascade. Down stream translational regulation is unlikely because of normality of MGA, lactase, and sucrase protein isoforms. The composite data support the belief that shared regulatory suppression has reduced transcription of all three enzymes.
MGA deficiency was suspected in an infant who failed to thrive on a starch oligomer formula. Deficiency was confirmed by enzyme assays on two duodenal biopsies and by repeated starch substrate breath tests. The observation of reduced activity of more than one disaccharidase did not point strongly to a causal mutation within the MGAM gene, but the existence of possible consanguinity made the coincidental occurrence of two or more independent mutations in genes encoding the different disaccharidases possible. We therefore extensively screened MGAM for mutations. However, sequencing of the coding region and the promoter of MGAM revealed no causal mutation. The negative evidence of mutations or functional polymorphisms at coding and promoter region levels suggests that this case of pandisaccharidase deficiency is mediated by factors external to the MGAM gene and the association with lactase and sucrase deficiencies supports the hypothesis that the MGA deficiency of the propositus is due to shared upstream regulatory factors. The normality of protein isoforms of all three enzymes in organ culture clearly makes downstream transcriptional or peptide regulation unlikely. Likely genetic candidates for the pleiotropic (multiple) effects on expression observed in this infant would be transcription factors and cofactors.(23–29)
The authors thank the contribution of Polycose from Ross Laboratories, Columbus, OH. This work is a publication of the USDA/ARS Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine and Texas Children's Hospital, Houston, TX. The authors thank Carlos H. Lifschitz, M.D., and Gopala Krishna, M.D. for providing clinical care to the propositus. The NIH supported Child Health Research Center Sequencing Core, performed all sequencing. This project has been funded in part with federal funds from the U.S. Department of Agriculture, Agricultural Research Service under Cooperative Agreement number 58-6250-1-003 (BN) and from the Swiss National Foundation number 3200-052736.97 (ES). The contents of this publication do not necessarily reflect the views or policies of the U.S. Department of Agriculture, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. or Swiss Governments. The MGAM promoter sequence, revised cDNA sequence, and MGA coding region SNPs have been registered with GenBank. The sequences of the overlapping primers for direct sequencing of human MGA cDNA are available from the corresponding author.
1. Semenza G, Auricchio S (2000). Small-intestinal disaccharidases. In: Scriver, CR, Beaudet, AL, Valle, WS, Sly, D, eds. The Metabolic And Molecular Bases Of Inherited Disease. New York. McGraw-Hill: 1623–50.
2. Nichols BL, Eldering J, Avery S, et al. Human small intestinal maltase-glucoamylase
cDNA cloning: homology to sucrase-isomaltase
. J Biol Chem 1998: 273:3076–81.
3. Harvey CB, Fox MF, Jeggo PA, et al. Regional localization of the lactase-phlorizin hydrolase
gene, LCT, to chromosome 2q21. Ann Hum Genet 1993; 57:179–85.
4. West LF, Davis MB, Green FR, et al. Regional assignment of the gene coding for human sucrase-isomaltase
(SI) to chromosome 3q25–26. Ann Hum Genet 1988; 52:57–61.
5. Wang YX, Harvey CB, Pratt WS, et al. The lactase persistence/non-persistence polymorphism is controlled by a cis-acting element. Hum Molec Genet 1995; 4:657–662.
6. Boll W, Wagner P, Mantei N. Structure of the chromosomal gene and cDNAs coding for lactase-phlorizin hydrolase
in humans with adult-type hypolactasia or persistence of lactase. Am J Hum Genet 1991; 48:889–902.
7. Jarvela I, Sabri EN, Kokkonen J, et al. Assignment of the locus for congenital lactase deficiency to 2q21, in the vicinity of but separate from the lactase-phlorizin hydrolase
gene. Am J Hum Genet 1998; 63:1078–85.
8. Poggi V, Sebastino G. Molecular analysis of the lactase gene in the congenital lactase deficiency. Am J Hum Genet (Suppl.) 1991; 61:11–6.
9. Treem, WR. Congenital sucrase-isomaltase
deficiency. J Pediatr Gastroenterol Nutr 1995; 21:1–14.
10. Lebenthal E, U KM, Zheng BY, et al. Small intestinal glucoamylase deficiency and starch malabsorption: a newly recognized alpha-glucosidase deficiency in children. J Pediatr 1994; 124:541–6.
11. Savilahti E, Launiala K, Kuitunen P. Congenital lactase deficiency. A clinical study on 16 patients. Arch Dis Child 1983; 58:246–52.
12. Nichols BL, Dudley MA, Nichols VN, et al. Effects of malnutrition on expression and activity of lactase in children. Gasteroenterology 1997; 112:742–51.
13. Nichols BL, Nichols VN, Putman M, et al. Contribution of villous atrophy to reduced intestinal maltase in infants with malnutrition. J Pediatr Gastroenterol Nutr 2000; 30:494–503.
14. Naim HY, Roth J, Sterchi EE, et al. Sucrase-isomaltase
deficiency in humans; different mutations disrupt intracellular transport, processing, and function of an intestinal brush border enzyme. J Clin Invest 1988; 82:667–79.
15. Sterchi EE, Mills PR, Fransen JAM, et al. Biogenesis of intestinal lactase-phlorizin hydrolase
in adults with lactose intolerance. J Clin Invest 1990; 86:1329–37.
16. Naim HY, Sterchi EE, Lentze MJ. Structure, biosynthesis, and glycosylation of human small intestinal maltase-glucoamylase
. J Biol Chem 1988; 263:19709–17.
17. Lifschitz CH, Boutton TW, Carrazza F, et al. A carbon–13 breath test to characterize glucose absorption and utilization in children. J Pediatr Gastroenterol Nutr 1988; 7:842–7.
18. Higuchi R. Recombinant PCR. In: PCR Protocols a Guide to Methods and Applications. Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds.). San Diego, CA: Academic Press, Inc., 1990: 177–83.
19. Cullen BR. Use of eukaryotic expression technology in the functional analysis of cloned genes. Meth Enzymol 1987; 152:684–704.
20. Pischitzis, A, Hahn, D, Leuenberger, B, et al. N-Benzoyl-L-tyrosyl-p-aminobenzoic acid hydrolase (human meprin). A 13 amino-acid sequence is required for proteolytic processing and subsequent secretion. Eur J Biochem 1999; 261:421–9.
21. Hauri HP, Sterchi EE, Bienz D, et al. Expression and intracellular transport of microvillus membrane hydrolases in human intestinal epithelial cells. J Cell Biol 1985; 101:838–51.
22. Livesey G. A perspective on food energy standards for nutrition labelling. Br J Nutr 2001; 85:271–87
23. Asp NG, Berg NO, Dahlqvist A, et al. Intestinal disaccharidases in Greenland Eskimos. Scand J Gastroent 1975; 10:513–19.
24. Ament ME, Perera DR, Esther LJ. Sucrase-isomaltase
deficiency – a frequently misdiagnosed disease. J Pediatr 1973; 83:721–7.
25. Gardiner AJ, Tarlow MJ, Symonds J, et al. Failure of the hydrogen breath test to detect primary sugar malabsorption. Arch Dis Childhood 1981; 56:368–72.
26. Ford RPK, Barnes GL. Breath hydrogen test and sucrase isomaltase deficiency. Arch Dis Childhood 1983; 58:595–7.
27. Gudmand HE, Fenger HJ, Kern-Hansen P, et al. Sucrase deficiency in Greenland: incidence and genetic aspects. Scand J Gastroent 1987; 22:24–8.
28. McNair A, Gudmand-Hoyer E, Jarnum S, et al. Sucrose malabsorption in Greenland. Brit Med J 1972; 1:19–21.
29. Eggermont E, Hers HG. The sedimentation properties of the intestinal α-glucosidases of normal subjects and patients with sucrose intolerance. Eur J Biochem 1969; 9:488–96.
30. Schmitz J, Bresson JL, Triadou N, et al. Analyse en electrophorese sur gel de polyacrylamide des proteines de la membrane microvillositaire et d'une fraction cytoplasmique dans 8 cas d'intolerance congenital au saccharose. Gastroenterol Clin Biol 1980; 4:251–6.
31. Skovbjerg H, Krasilnikoff PA. Maltase-glucoamylase
and residual isomaltase in sucrose intolerant patients. J Pediatr Gastroenterol Nutr 1986; 5:365–71.
32. Mainguet P, Vanderhoeden R, Loeb H, et al. Congenital maltase-sucrase and maltase-isomaltase deficiency in an adult. Digestion 1973; 8:353–9.