Share this article on:

Contribution of Villous Atrophy to Reduced Intestinal Maltase in Infants With Malnutrition

Nichols, Buford L.*; Nichols, Veda N.*; Putman, Margaret*; Avery, Stephen E.*; Fraley, J. Kennard*; Quaroni, Andrea; Shiner, Margot‡¶; Sterchi, Erwin E.§; Carrazza, Francisco R.

Journal of Pediatric Gastroenterology and Nutrition: May 2000 - Volume 30 - Issue 5 - p 494-502
Original Articles

Background It has been known for many years that small intestinal maltase activities are reduced in malnourished infants and in other patients with villous atrophy. The recent availability of human maltase-glucoamylase cDNA provides the opportunity to test the hypothesis that villous atrophy accounts for the reduced maltase enzyme activity in malnourished infants.

Methods Mucosal biopsy specimens obtained for clinical evaluation of malnourished infants with poor responses to refeeding were examined by quantitative methods for enzyme activity and mRNA levels.

Results Maltase activity and maltase-glucoamylase mRNA were reduced (approximately 45% of normal). When maltase-glucoamylase message was normalized to villin message, a structural protein expressed only in enterocytes, a preservation of maltase messages in surviving enterocytes was documented. The luminal glucose transporter–villin message was also preserved.

Conclusions The loss of maltase-glucoamylase message paralleled the reduction in villin message and degree of villous atrophy. The reduced maltase-glucoamylase message also paralleled sucrase-isomaltase message, previously found to be decreased in proportion to villous atrophy of malnourished infants. The data directly demonstrate, for the first time, that the terminal steps of starch 1-4 starch digestion and sucrase-isomaltase 1-6 starch digestion are decreased in malnourished infants, secondary to villous atrophy. These data in prior and present reports suggest that mechanisms underlying the chronic villous atrophy of malnutrition should be a priority for investigations in malnourished infants with slower than expected weight gain during refeeding.

*U.S. Department of Agriculture/Agriculture Research Service, Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine and Texas Children's Hospital, Houston, Texas; †Cornell University, Ithaca, New York; ‡Department of Pediatric Gastroenterology, Sackler Medical School, Assaf Harofeh Hospital, Tel Aviv University, Tel Aviv, Israel; §Institute for Biochemistry and Molecular Biology of the University of Bern, Bern, Switzerland; ∥Departamento de Pediatria, Instituto da Criança, Hospital das Clinicas da Faculdade de Medicina, Universidade de São Paulo, Sãao Paulo, Brazil

Received October 6, 1999;

revised February 2, 2000; accepted February 7, 2000.

¶Deceased July 31, 1998.

Address correspondence and reprint requests to Buford L. Nichols, MD, USDA Children's Nutrition Research Center, Baylor College of Medicine, 1100 Bates Street, Houston, TX 77030-2600.

Some severely malnourished infants have a slow response to dietary rehabilitation. These infants are usually fed with mixed milk-, sugar-, and cereal-based diets. We recently presented evidence that the severe suppression of lactase-phlorizin hydrolase (LPH) expression and activity in malnourished infants may be transcriptional and suggested that the accompanying reduction of sucrase-isomaltase (SI) expression and activity is an alteration secondary to the villous atrophy common in malnourished infants. This deduction was based on lower levels of lactase than sucrase message (1). Our recent cloning of the human maltase-glucoamylase (MGA) cDNA (2) presented the opportunity to study the message expression and activity of this key enzyme in starch digestion in the same malnourished subjects as in our previous study (1). Our hypothesis was that MGA would be reduced by the loss of enterocytes due to villous atrophy. To test our hypothesis of the structural basis for maltase reductions, we predicted that MGA mRNA levels would decrease in proportion to a reduction in the mRNA for villin, a structural protein, and sodium-activated glucose-galactose luminal transporter 1 (SGLT 1), a functional protein. These two genes, expressed only by enterocytes within small intestinal mucosa, were selected to serve as a representation of the population of enterocytes within the total biopsy.

Starches, the most common carbohydrates in the postweaning human diet, are the main carbohydrate source in cereal-based programs of nutritional rehabilitation of malnourished infants. They are also used to fortify or complement milk-based refeeding diets (3). Starches (recently reviewed in this journal [4]) are poorly soluble storage granules from food plants that are composed of two structurally different polysaccharides: amylose, linear 4-O-α-d-glycopyranosyl-d-glucose polymers, and amylopectin, with additional 6-O-α-d-glycopyranosyl-d-glucose links that result in a branched configuration. The dietary starches used in refeeding malnourished infants are a mixture of approximately 25% amylose with amylopectin. All starch granules require cooking to improve enzyme digestibility. α-Amylase (EC, an endoenzyme found in normal mature human salivary and pancreatic secretions, acts to produce oligosaccharide chains by hydrolysis of internal α-1-4 linkages. It produces soluble linear maltose and branched isomaltose dextrins of 2 to approximately 40 glucose units (4,5). α-Amylase, from various sources, is used to malt starches and produce the soluble oligosaccharides (dextrins) widely used in premature infant formulas and the fortification of energy levels for older infants. There has been a recent trend to use dextrins in the refeeding of infants with diarrhea (6) and malnutrition (7). These dextrins are not metabolized without further processing by a β-amylase (EC, which hydrolyzes the internal, nonreducing ends at 1-4 and 1-6 linkages to free glucose. In mammals, this penultimate act is carried out by two brush border enzymes, SI and MGA (EC and 3) (5). The 1-4 activity of SI is specific for maltose, but the activity of MGA encompasses all 1-4 maltosides from 2 to beyond 7 glucose units and is specific for starch (8,9). The activity of SI is specific for all 1-6 isomaltosides from 2 to beyond 7 glucose units (8). The final act of carbohydrate assimilation is transmembrane transport by SGLT 1, which requires sodium to move glucose or galactose through the brush border of the enterocyte (4,5). It has been hypothesized that the brush border enzymes responsible for sugar and starch hydrolysis to glucose are functionally linked to SGLT 1 transport of glucose (10).

Salivary and pancreatic amylase secretion and activity are undetectable in healthy humans less than 4 to 6 months of age (11,12). A reversible suppression of pancreatic amylase secretion and activity also occurs in severely malnourished 1-to 3-year-old children (13,14). All brush border carbohydrate hydrolases are sensitive to experimental developmental and nutritional factors (5). In the rat and rabbit, LPH enzyme activity is high at birth and declines when the animal is weaned to the adult diet. The activity of SI is virtually absent until weaning occurs. MGA activity is present from birth, but increases with consumption of the adult diet. In adult animals, SI and MGA activities increase when a high-carbohydrate diet is fed, but LPH activity appears to be independent of dietary levels of lactose (5). In the pig and human a modified pattern of enzyme activity is observed in which LPH is more sustained after weaning, and SI and MGA activities are present before birth. In the mature pig SI and MGA activities increase on high-carbohydrate diets (5). In mice, rats, and sheep SGLT 1 activity is upregulated at weaning (15,16). All brush border carbohydrate hydrolase activities increase in severely malnourished rats and pigs (17,18) but activities of LPH, SI, and MGA are decreased in malnourished children (reference 1, and the current study). One report suggested the possibility of a genetic deficiency of intestinal glucoamylase occurring in young children with chronic diarrhea and responding to a starch-free diet. These patients had normal mucosal morphology and low mucosal starch hydrolyzing activity (19). SGLT 1 has been found to undergo a variety of genetic mutations in infants with clinical glucose and galactose malabsorption (20). The mucosal histology is normal in these infants (20). An acquired form of this phenotype has been reported in young, malnourished infants with severe villous atrophy that responds to nutritional rehabilitation (21). In these acquired cases, the degree of glucose malabsorption was proportional to the loss of small intestinal surface area, thus suggesting that the glucose transport defect was due to villous atrophy, rather than genetic mutations of the transporter (21–24).

Recent advances in the molecular biology of these enzymes and the transporter have allowed preliminary insights into the regulation of SI and LPH in human malnutrition and in genetic disorders (1,20). LPH is believed to be regulated at a transcriptional level, whereas SI is thought to be normally expressed in enterocytes, although total enzyme activity and message are reduced because of villous atrophy (1). We wanted to perform studies of MGA message regulation in malnourished infants, which subsequently became possible when MGA cDNA sequence data became available (2). We report in this article the first studies of MGA message regulation. The messages for villin, a brush border structural protein, and SGLT 1, a functional brush border transport protein within the same pathway of carbohydrate assimilation, were included in the investigation to directly test our hypothesis that changes in MGA message expression are unchanged in surviving enterocytes.

Back to Top | Article Outline


Experimental Design

Two groups of infants were enrolled in the study (Tables 1–4). These are the same subjects reported in our previous investigation but differ slightly in number because only the subjects with maltase assays are reported in the current study (1). Twenty-four Brazilian infants with malnutrition, who had z scores for weights and heights for age less than −2.0, comprised the M group. Portions of jejunal biopsy specimens, obtained as part of the routine clinical management of poor response to dietary rehabilitation, were used for this study. All M subjects were admitted to the hospital more than 2 weeks before the biopsy and had previously received a home-prepared cow's milk–based formula (whole cow's milk, diluted by two thirds with the addition of a 5% mixture of sucrose and starch) and a variety of table foods before hospital admission. The infants received commercially available formulas and mixed cereal and fruit diets after admission.









The control group was composed of nine infants with normal weight/height z scores and normal mucosal histology, and were thus viewed as normal subjects (N) within our experimental design. The jejunal tissue for this group was taken from nine Brazilian patients admitted to the hospital for the Kasai surgical procedure for biliary atresia. The specimens were obtained during the routine operative procedure. The N subjects were younger and of different sex ratios than the malnourished infants, but they were matched for height and weight, and the z score and histologic criteria for their admission to the study were insensitive to age and sex ratio differences. As a confirmation, we tested older control subjects (8–10 years of age) without biliary atresia who were selected by these same criteria, and could detect no effect of age, biliary status, or biopsy procedure on the control mucosal histology and maltase activity presented in this report (data not shown). Jejunal tissue from a 4-year-old organ donor (donor IV) was used for making monoclonal antibody and for internal standards throughout the assays.

Back to Top | Article Outline

Ethical Considerations

The 24 M and 9 N subjects (Table 1) were admitted to the Instituto da Crianças, Hospital das Clinicas da Faculdade de Medicina da Universidade de São Paulo (ICR:HC FMUSP), Brazil. The organ donor tissue was collected at one of the Baylor College of Medicine–affiliated hospitals, The Methodist Hospital of Houston, Texas, U.S.A. The investigation had been approved by the Human Investigation Review Boards of the Baylor College of Medicine and affiliated hospitals and of the Department of Pediatrics ICR:HC FMUSP, and signed parental consent was obtained for each subject. Clinical examinations and data were recorded using the procedures and data management system reported previously (1,24).

Back to Top | Article Outline

Biopsy Procedures

Biopsy tissues were obtained from the M children after a 2-to 3-week hospital stay. These are the same biopsies reported in our previous publication (1). The jejunal tissues were obtained using a multiport pediatric Crosby–Kugler capsule, as previously described (1). Each specimen was divided as follows: a portion in Zenker's solution, embedded in paraffin, sectioned, and stained for light microscopy and diagnostic histology; and a portion embedded in an optimal cutting temperature embedding compound (OCT, Tissue-Tek, Elkhart, IN, U.S.A.), frozen in liquid nitrogen, and used for immunofluorescence, enzyme histochemistry, and RNA isolation. The remainder was frozen at −70°C until used for biochemical assay, at which time the tissue was thawed and homogenized in Dulbecco's phosphate-buffered saline containing protease inhibitors (1). In the N children undergoing the Kasai procedure for biliary atresia, 1 to 2 cm of the jejunum immediately distal to the ligament of Treitz was obtained by surgical resection during the construction of the Roux-en-Y anastomosis. Tissue was treated as described.

Back to Top | Article Outline

Biochemical Analysis

Disaccharide Activity

The results of the maltase assays were not reported in our previous publication (1). Briefly, maltase activity was determined in biopsy homogenates using 0.112 M maltose as the substrate (1). Glucose production was measured on a an autoanalyzer (COBAS Fara II; Roche Diagnostic Systems, Montclair, NJ, U.S.A.), using a glucose oxidase kit (Glucose Trinder 100; Sigma Chemical Co., St. Louis, MO, U.S.A.), and Monitrol (Roche Diagnostic Systems) as the glucose control (1). Protein was measured using a bicinchoninic acid protein assay kit (Pierce Co., Rockford, IL, U.S.A.). Activity was expressed as units (U = μM/min)/g protein (U/g protein). Normal values were those from our laboratory, and hypomaltasia was defined as less than 93 U/g protein (16).

Back to Top | Article Outline



Villous atrophy was judged from the hematoxylin and eosin–stained, paraffin-embedded or frozen sections and graded 1 to 4+ by a single histologist who graded the slides in blinded fashion (1). The mean histology grades were the same as in our previous report (1). Absence of atrophy (1+) was estimated by comparison with tissue from an organ donor, and 4+ was judged in reference to children with active celiac disease (1). For purposes of comparison, in the context of other M values, villous atrophy was recalculated as a percentage of N villous length by taking the reciprocal of the grade of atrophy (in this case, normal = grade 4 in Table 4).

Back to Top | Article Outline

Molecular Procedures

Fresh assays of MGA, villin, and β-actin mRNA were performed in a subset of 10 M and 9 N subjects from Table 1 for this investigation. The subset comprising Table 3 was defined by adequacy and quality of biopsy RNA for the assay of a housekeeping gene message, β-actin. The assays for SGLT 1 were performed at the time of our previous investigation but were not reported in our previous article, which reported only LPH and SI messages (1). The present investigation used the same small intestinal RNA as our previous publication (1). Fresh reverse transcription–polymerase chain reaction (RT-PCR) analyses of mRNA were performed according to our published procedures (1). Briefly, five frozen tissue sections (10 μm) were cut from the OCT block at −20°C and attached to glass slides. RNA was isolated from scraped sections (RNeasy kit; Qiagen, Chatsworth, CA, U.S.A.). The total RNA was quantitated by optical density (OD) measurement at 260 nm. Fresh reverse transcription was performed using 100 ng of RNA at 37°C for 2 hours. The 50-μl reaction volume was composed of 10× buffer, 4 mM deoxyribonucleoside triphosphate (dNTP), 4 U RNase inhibitor, 30 ng random nonamer primers, 20 U M-MLV reverse transcriptase, and diethyl pyrocarbonate (DEPC)-treated water (RT-PCR Kit; Stratagene, La Jolla, CA, U.S.A.). The RT reaction product was amplified by PCR (Stratagene). The amplification primers for human MGA, SI, and SGLT 1 were synthesized (Genosys, The Woodlands, TX, U.S.A.). Maltase primers 5´ (1669–1690) and 3´ (2049–2028) were based on our cloning primers (2). SGLT 1 primers were those described by Wang et al. (25) Villin primers were those described by Fajardo et al., (26) and primers for human β-actin were purchased (#5211.h; Continental Laboratory Products, San Diego, CA, U.S.A.). The PCR reactions had volumes of 100 μl including 10× buffer, 0.8 mM dNTP, the specific primers (0.4 μM MGA, SGLT-1, or villin, and 0.5 μM total β-actin), 2.5 U Taq polymerase, and 3 μl of the RT reaction product. After a hot start at 94°C for 5 minutes, the reaction was stepped through 30 cycles of 52° for 40 seconds, 70° for 60 seconds and 94°C for 30 seconds in a thermal cycler (model 480; Perkin-Elmer, Foster City, CA, U.S.A.). SGLT 1 reactions were taken to 35 cycles because of low copy numbers. A 5-μl sample of each PCR product was visualized on a 1.5% agarose gel run in TAE buffer containing 0.5 μg/ml ethidium bromide. To quantitate the amplimers, 10 μl of the RT-PCR reaction mixture was assayed as previously described by liquid chromatography (LC) with the DEAE-NPR column (TSK kit; Perkin-Elmer) (1). RNA from organ donor IV was run in parallel with the experimental subjects as an internal standard for the assays. The LC results were expressed as nanograms amplimer/100 ng total tissue RNA. The amplifications from diluted donor IV total intestinal RNA were linear (r = 0.995–0.999) in the calibration range of 10 to 180 ng. Within-run LC coefficients of variation were less than 10%. The specificity of all PCR amplimers was confirmed by sequencing with an automated sequencer (model 373A; Applied Biosystems, Foster City, CA, U.S.A.). The quality of mRNA was confirmed by parallel amplifications from the same cDNA with the β-actin primers.

Back to Top | Article Outline

Statistical Procedures

Subject data were maintained using the SIR Database Management System (SIR, Inc., Evanston, IL, U.S.A.) (1). Statistical analyses were performed using the Minitab programs (Minitab, Inc., State College, PA). Categorical variables were analyzed with χ2 or Fisher's exact test. All t-tests were computed using Bonferroni's correction. Numeric variables were analyzed by one-way analysis of variance (ANOVA). Missing data were excluded from consideration and account for the variations of sample size (Tables 1–4). Only complete data sets were used for regression analyses.

Back to Top | Article Outline


Clinical Characteristics

Evaluation of the nutritional status, as indicated by age-independent measures of growth and body composition, revealed severe malnutrition characterized by growth failure and tissue wasting in the M group. The nutritional status (weight/height z scores) of the infants in the M group differed significantly (P = 0.003) from that of the N subjects (Table 1). The M group had significant (P = 0.006) morphologic alterations of the jejunum with partial villous atrophy (Table 1). A correlation between villous atrophy and weight/age z score was present (r = 0.65, n = 22).

Back to Top | Article Outline

Enzyme Activity

Reductions in the enzyme activities of maltase were present in the M subjects. Mean enzyme activity (units per gram of biopsy sample protein) for the nutritional groups is shown in Table 2. Maltase activities were reduced in the M specimens (to 37% of N) but statistical significance was marginal (P ≈ 0.15). Maltase activity was below normal (<93 U/g protein) in 13 of 25 M subjects. Maltase activities were within the normal range in all the N subjects. A decline in activity as weight declined from standard for age (z score) was not found for maltase activity.

Back to Top | Article Outline


The mRNA for maltase, SGLT 1, villin, and β-actin was assayed by quantitative LC amplimer measurement. Except for SGLT 1, all were amplified under identical conditions from freshly transcribed total RNA. SGLT 1 message quantitation was performed at the time of our previous study of lactase expression in the same subjects (1). Inclusion of M and N subjects in the subgroups reported in this and our previous publication (1) was based on adequacy of β-actin amplification from the biopsy specimen RNA. The present results are summarized in Table 3.

Mean maltase activity was 83 ± 52 and 224 ± 78 U/g protein in the M and the N subgroups, respectively (P = 0.001). The mean quantity of MGA mRNA amplimers was 218 ± 167 ng/100 ng RNA in the M and 485 ± 207 ng/100 ng in the N subgroups, respectively (P = 0.016). (The N values differ slightly from those previously reported [1] because one subject was discovered to have adult-type hypolactasia. The RT-PCR for this subject were not included in the present investigation.) SGLT 1 amplimers were 5585 ± 1936 and 8389 ± 3193 ng/100 ng RNA (P = 0.057) in the M and N groups. Villin amplimers were measured at 1179 ± 508 ng in M and 2245 ± 607 ng in N (P = 0.003). The mean mRNA amplimers for β-actin were 4180 ± 758 ng and 4794 ± 957 ng in the M and N subgroups, respectively, and were not significantly different (P = 0.185). The quantity of MGA amplimer correlated with maltase activity (r = 0.32, n = 15). There was a correlation between MGA and SGLT 1 mRNA (P = 0.67, n = 15).

Villin mRNA concentration correlated with histologic grade of villous atrophy (r = 0.69), and villin/β-actin ratios correlated with villous atrophy (r = 0.76). SGLT 1 and SGLT 1/β-actin ratios also correlated with grade of atrophy (r = 0.47 and 0.54, respectively). Maltase mRNA correlated with villous atrophy (r = 0.73), but β-actin did not (r = 0.01, n = 15).

MGA and SGLT 1 messages correlated with villin message (Fig. 1). The regression equations are shown in the legend to Figure 1 (r ≥ 0.74). The intercepts ± SD were near zero for MGA and above zero for SGLT 1. These intercept values are consistent with the M values normalized to villin message in Table 3.

FIG. 1.

FIG. 1.

Back to Top | Article Outline


Enzyme Activity

Levels of maltase enzyme activity were reduced to approximately half of normal in our malnourished Brazilian infants. The loss of enzyme activity was similar to that reported in malnourished infants in other countries (27–47). The degree of villous atrophy was comparable to that reported in malnourished adults as well as children from other countries (48–67). Maltase activity levels also have been reported to be reduced in children and adults with other atrophic mucosal disorders such as intractable diarrhea of infancy and celiac syndrome (21,22,69–76). Although the association of villous atrophy has been assumed, there has been no direct proof of this association. In the present investigation, the mechanism for the reduction of mucosal maltase was examined in the context of villous atrophy, other brush border carbohydrate hydrolases, and the glucose transporter.

In all the intestinal mucosal disorders associated with villous atrophy, maltase activity has always correlated with sucrase activity (21,22,69–76). It has been reported that 80% of the maltase activity in human small intestine is contributed by SI, with the remaining by MGA (5). This correlation of maltase activity with sucrase activity was confirmed in the present study (r = 0.95, n = 32). This correlation has been attributed to the overlap of the two enzyme specificities for the maltose substrate used in the assays (21,69–76). In the present study we demonstrate, for the first time, that the correlations between the two enzyme activities exist at the mRNA levels (r = 0.73, n = 16) and cannot be explained by the overlap of enzyme specificity for maltose substrate (5).

Back to Top | Article Outline

Enzyme Messages

We previously suspected that sucrase activity in malnourished infants was due to the loss of enterocytes secondary to villous atrophy and deduced that lactase message was downregulated (1). We also demonstrated that our malnourished infants with hypolactasia did not have the heterozygotic allelic polymorphism that is characteristic of all reported hypolactasic adults (1). To test the hypothesis that maltase was reduced by loss of enterocytes, we normalized MGA messages to additional messages specific for the enterocyte. The first was villin, a structural protein required for formation of microvilli on the enterocyte luminal surface (77). The second was SGLT 1, the luminal glucose transporter located on the microvillus membrane (20). The villin message and SGLT 1 correlated (r = 0.69 and r = 0.53, respectively;n = 15) with villous atrophy. Normalizing villin to β-actin mRNA improved its correlation to atrophy (r = 0.76), but normalizing SGLT 1/β-actin did not. From this analysis, it became clear that expression of the structural protein villin message level is a function of villous atrophy and that the functional protein SGLT 1 message was regulated more independently.

The functional messages for the carbohydrate hydrolases and the glucose transporter message concentrations (nanograms amplimer/nanograms RNA template) were normalized to the structural message for villin (Table 3). Values recalculated from our prior publication are indicated by (1) :

  • • MGA/villin message ratio was unchanged by malnutrition (P = 0.248).
  • • SI (1) /villin message ratio was significantly increased by malnutrition (P = 0.026).
  • • The mean SGLT 1/villin message ratio was conserved in malnutrition (P = 0.154).
  • • The previously deduced reduction in lactase mRNA (1) /enterocyte was directly confirmed by the reduced LPH/villin message in malnutrition (P = 0.043) (1).
  • • The increases in the SI/villin ratios led to a discovery that some messages may be enhanced in the surviving enterocytes of malnourished infants.
  • • The conserved or enhanced MGA, SI, and SGLT 1 message levels are in striking contrast to the 40% reduction in the enterocyte LPH/villin message level, and clearly support the hypothesis that sucrase and maltase activities are reduced in malnourished infants by loss of enterocytes. The unexpected discovery of upregulation of sucrase/villin message suggests that compensatory diet-sensitive transcriptional responses may occur in the enterocyte that regulate the proteins responsible for the hydrolysis and absorption of sugar, starch, and glucose. This is consistent with a hypothesis proposed by Pappenheimer (10), which linked the expression of brush border hydrolases to the microvillus membrane transporters.
Back to Top | Article Outline

Dietary Implications

The malnourished infants were studied while they were fed a diet of formulas fortified with sucrose and starch and complemented with cereals and fruit. There was no obvious relationship between the nature of the diets and maltase activities and message levels. All infants were experiencing increasing weight but at an unsatisfactory rate of gain. If the reduced maltase activity depicted in Table 4 is representative of the whole small intestine, it appears that the average message expression and enzyme activity of MGA, crucial for terminal starch digestion, was reduced to approximately 40% of normal. The SGLT 1 message, the final phase of starch assimilation, averaged 66% of normal. These calculations should be tested directly in this population. It has been intuitively believed, but is now demonstrated by experimental evidence, that the activities that determine essential pathways for starch α-exo-hydrolysis and glucose transport are sensitive to losses of enterocytes.

The conserved or enhanced enterocyte glucose hydrolases and transporter also suggest the possibility of a potential bottleneck at the level of enterocyte transport of the summed glucose produced by the carbohydrolases. This leads to the speculation that severe villous atrophy may lead to an excess of glucose production from a mixed diet, by combined lactase (1), sucrase (1), and maltase hydrolytic activities, and a relative inadequacy of glucose transporters. The excess glucose produced could be backflushed into the lumen and malabsorbed by the small intestine, or because of excessive glucose concentrations in the unstirred layer of the brush border, could be forced through paracellular pathways (10). This potential excess paracellular osmolar load, because of hydrolase–transporter mismatch, may contribute to ongoing basolateral ballooning and shedding of enterocytes (56–65) and to the persistence of villous atrophy. This atrophic process could prolong the reductions of maltase and sucrase, contribute to limited small intestinal starch and sugar digestion and assimilation and, because of normal inefficiency of carbohydrate assimilation and oxidation in the colon, to reduced weight gains with the usual dietary energy intakes.

There has been a recent trend to use either whole starch or malted starch in oral hydration solutions for infants with acute diarrhea, and malted starch for the refeeding of malnourished children (4,5). In the younger infant with α-amylase developmental delay and the malnourished infant with acquired α-amylase deficiency, it is assumed by the proponents that the activity of MGA will be able to compensate for missing α-amylase activities in the digestion of starch. It is necessary to caution that the feeding of malted starch only bypasses amylase deficiency. In this investigation, we have shown that MGA is reduced by an average of 40% in malnourished infants secondary to villous atrophy. It is important that reduced brush border hydrolytic capacities, varying in severity, be considered in the prescription of mixtures of dietary starch, sugar, and lactose for refeeding of malnourished infants.

Back to Top | Article Outline


The results of quantitative measurements made of maltase activity and message levels in this study are summarized in Table 4, where the observations in the M infants are expressed as percentages of of N mean values. There is a correlation between the morphologic degree of villous atrophy and the level of villin message that justifies the use of this structural message as a more quantitative measure of the number of enterocytes. Mean maltase activity and message were reduced proportional to number of enterocytes (villin message level, Figure 1). The glucose transporter message was increased in ratio to villin message in most subjects. These data advance our hypothesis that maltase (and sucrase [1]) messages are reduced because of villous atrophy and suggest that MGA sucrase and glucose transport messages, although reduced in aggregate, may be conserved in surviving enterocytes. The fundamental mucosal lesion, villous atrophy, requires additional investigations because of its direct effect on starch and starch oligomer digestion by the small intestine of malnourished infants.

Back to Top | Article Outline


Dr. Andy Feste of Biochroma (Spring, TX, U.S.A.) set up the liquid chromatography system for quantitation of PCR amplimers and Ms. Jestina Mason of the Child Health Research Core Laboratory at Baylor College of Medicine performed amplimer sequencing. Mr. Scott Perkinson performed the enzyme activity assays. The authors thank Ms. Conceicao Maria L. Godoy, nutritionist; Maraci Rodrigues, M.D., and Ary Lopes Cardoso, M.D., for their excellent clinical care of the infants in this study at ICR HCFMUSP; Joao Gilberto Maksoud, M.D., of the Surgical Department of ICR HCFMUSP, for provided the tissue from infants undergoing the Kasai procedure; Dr. John Waterlow, Dr. Magdalena Araya, and Dr. Laszlo G. Kömüves for reading the manuscript and providing critical reviews; Ms. Jane Schoppe for secretarial support; and Children's Nutrition Research Center editor Leslie Loddeke for editorial support.

This work was funded in part by a research grant from Bristol-Myers. It is a publication of the U.S. Department of Agriculture/Agriculture Research Service Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine and Texas Children's Hospital, Houston, Texas and was funded in part with federal funds from the U. S. Department of Agriculture, Agricultural Research Service under Cooperative Agreement number 58-6250-6-001. 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. Government.

Back to Top | Article Outline


1. Nichols BL, Dudley MA, Nichols VN, et al. Effects of malnutrition on expression and activity of lactase in children. Gastroenterology 1997; 112:742–51.
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. Brewster DR, Manary MJ, Menzies IS, et al. Comparison of milk and maize based diets in kwashiorkor. Arch Dis Child 1997; 76:242–8.
4. Christian M, Edwards C, Weaver LT. Starch digestion in infancy. J Pediatr Gastroenterol Nutr 1999; 29:116–24.
5. Semenza G, Auricchio S. Small-intestinal disaccharidases. Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic Basis of Inherited Disease. New York: McGraw-Hill, 1989:2975–97.
6. Thillainayagam AV, Hunt JB, Farthing MJ. Enhancing clinical efficacy of oral rehydration therapy: Is low osmolality the key? Gastroenterology 1998; 114:197–210.
7. Brown KH. Complementary feeding in developing countries: Factors affecting energy intake. Proc Nutr Soc 1997; 56:139–48.
8. Heymann H, Breitmeyer D, Gunther S. Human small intestinal sucrase-isomaltase: Different binding patterns for malto-and isomalto-oligosaccharides. Biol Chem Hoppe-Seyler 1995; 376:249–53.
9. Heymann H, Gunther S. Calculation of subsite affinities of human small intestinal glucoamylase. Biol Chem Hoppe-Seyler 1994; 375:451–5.
10. Papenheimer JR. On the coupling of membrane digestion with the intestinal absorption of sugars and amino acids. Am J Physiol 1993; 265:G409–17.
11. Sevenhuysen GP, Holodinsky C, Dawes C. Development of salivary-amylase in infants from birth to 5 months. Am J Clin Nutr 1984; 39:584–8.
12. Lebenthal E, Lee PC. Development of functional response in human exocrine pancreas. Pediatrics 1980; 66:556–60.
13. Barbezat GO, Hansen JDL. The exocrine pancreas and protein-calorie malnutrition. Pediatrics 1968; 42:77–92.
14. Danus O, Urbina AM, Valenzuela I, Solimano G. The effect of refeeding on pancreatic exocrine function in marasmic infants. J Pediatr 1970; 77:334–7.
15. Shirazi-Beechey SP. Intestinal sodium-dependent D-glucose transporter: Dietary regulation. Proc Nutr Soc 1996; 55:167–78.
16. Ferraris RP, Diamond JM. Specific regulation of intestinal nutrient transporters by their dietary substrates. Ann Rev Physiol 1989; 51:125–41.
17. Galluser M, Belkhou R, Freund J-N, et al. Adaption of intestinal hydrolases to starvation in rats: Effect of thyroid function. J Comp Physiol Series B Biol Sci 1991; 161:357–61.
18. Pond WG, Ellis KJ, Mersmann HJ, et al. Severe protein deficiency and repletion alter body and brain and organ weights in infant pigs. J Nutr 1996; 126:290–302.
19. 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.
20. Martin MG, Lostoa MP, Turk E, et al. Compound missense mutations in sodium/D-glucose cotransporter (SGLT 1) results in trafficking defects. Gastroenterology 1997; 112:1206–12.
21. Klish WJ, Udall JN, Rodriguez JT, et al. Intestinal surface area in infants with acquired monosaccharide intolerance. J Pediatr 1978; 92:566–71.
22. Shiner M, Putman M, Nichols VN, Nichols BL. Pathogenesis of small-intestinal mucosal lesions in chronic diarrhea of infancy: I. A light microscopic study. J Pediatr Gastroenterol Nutr 1990; 11:455–63.
23. Calvin RT, Klish WJ, Nichols BL. Disaccharidase activities, jejunal morphology, and carbohydrate tolerance in children with chronic diarrhea. J Pediatr Gastroenterol Nutr 1985; 4:949–53.
24. Nichols BL, Carrazza F, Nichols VN, et al. Mosaic expression of brush border enzymes in infants with chronic diarrhea and malnutrition. J Pediatr Gastroenterol Nutr 1992; 14:371–9.
25. Wang Y, Harvey C, Rousset M, et al. Expression of human intestinal mRNA transcripts during development: Analysis by a semiquantitative RNA polymerase chain reaction method. Pediatr Res 1994; 36:514–21.
26. Fajardo O, Naim HY, Lacey SW. The polymorphic expression of lactase in adults is regulated at the messenger RNA level. Gastroenterology 1994; 106:1233–41.
27. James WPT. Effects of protein-calorie malnutrition on intestinal absorption. Ann NY Acad Sci 1971; 176:244–61.
28. James WPT. Sugar absorption and intestinal motility in children when malnourished and after treatment. Clin Sci 1970; 39:305–18.
29. James WPT. Intestinal absorption in protein-calorie malnutrition. Lancet 1968; 2:333–5.
30. James WPT. Jejunal disaccharidase activities in children with marasmus and with kwashiorkor: Response to treatment. Arch Dis Child 1971; 46:218–20.
31. James WPT. Comparison of three methods used in assessment of carbohydrate absorption in malnourished children. Arch Dis Child 1972; 47:531–6.
32. Stanfield PJ. The small intestine in protein energy malnutrition. In: Creamer B, ed. The Small Intestine. Chicago: William Heinemann Medical Books Publication, 1974: 247–59.
33. Cevini G, Giovanni M, Careddu P. Alterazioni della digestione e dell'assorbimento intestinale dei glucidi nei disturbi acuti e cronici della nutrizione del lattante. Minerva Pediatr 1962; 14:831–5.
34. Bowie MD, Brinkman GL, Hansen JDL. Diarrhoea in protein-calorie malnutrition. Lancet 1963; 2:550–1.
35. Bowie MD, Brinkman GL, Hansen JDL. Acquired disaccharide intolerance in malnutrition. Pediatrics 1965; 66:1083–91.
36. Bowie MD, Barbezat GO, Hansen JDL. Carbohydrate absorption in malnourished children. Am J Clin Nutr 1967;89–97.
37. Kerpel-Fronius E, Jani L, Fekete M. Disaccharide malabsorption in different types of malnutrition. Ann Pediatr 1966; 206:245–57.
38. Wharton B, Howells G, Phillip I. Diarrhoea in kwashiorkor. BMJ 1968; 4:608–11.
39. Prinsloo JG, Wittmann W, Pretorius PJ, et al. Effect of different sugars on diarrhoea of acute kwashiorkor. Arch Dis Child 1969; 44:593–9.
40. Brasseur D, Hennart P, Vis HL. Malabsorption of intact lactose. Lancet 1985; 2:100–01.
41. Brasseur D, Vis HL. L'activite lactaseique de labordure en brosse de l'enterocyte. Acta Gastroenterol Belg 1984; 46:115–22.
42. Nordio S, Lamedica GM, Berio A, et al. Disaccharidase activities of duodenal mucosa in children. Ann Paediatr 1966; 206:287–312.
43. Cook GC, Lee FD. The jejunum after kwashiorkor. Lancet 1966; 2:1263–7.
44. DeLarrechea I, Ibarra R, Sampayo RR, et al. Malabsorption in malnutrition [abstract]. Gastroenterology 1967; 52:1103.
45. Prinsloo JG, Wittmann W, Kruger H, et al. Lactose absorption and mucosal disaccharidases in convalescent pellagra and kwashiorkor children. Arch Dis Child 1971; 46:474–8.
46. Romer H, Urbach R, Gomez MA, et al. Moderate and severe protein energy malnutrition in childhood: Effects on jejunal mucosal morphology and disaccharidase activities. J Pediatr Gastroenterol Nutr 1983; 2:459–64.
47. Sokucu S. Enterokinase and disaccharidase activities of the small intestinal mucosa in infants with protein energy malnutrition and in normal infants. Turk J Pediatr 1987; 29:15–24.
48. Burman D. The jejunal mucosa in kwashiorkor. Arch Dis Child 1965; 40:526–31.
49. Stanfield JP, Hutt MSR, Tunnicliffe R. Intestinal biopsy in kwashiorkor. 1965;2:519–23.
50. Cook GC, Lee FD. The jejunum after kwashiorkor. Lancet 1966; 2:1263–7.
51. Brunser O, Reid A, Monckeberg F, et al. Jejunal biopsies in infant malnutrition: With special reference to mitotic index. Pediatrics 1966; 38:605–12.
52. Barbezat GO. Gastrointestinal studies in human protein-calorie malnutrition [abstract]. S Afr Med J 1967; 41:1211.
53. Barbezat GO, Bowie MD, Kaschula ROC, et al. Studies on the small intestinal mucosa of children with protein-calorie malnutrition. S Afr Med J 1967; 41:1031–6.
54. Nunez-Montiel O, Bauza CA, Brunser O, et al. Ultrastructural variations of the jejunum in the malabsorption syndrome. Lab Invest 1967; 12:16–24.
55. Tandon BN, Magotra ML, Saraya AK, et al. Small intestine in protein malnutrition. Am J Clin Nutr 1968; 21:813–9.
56. Brunser O, Reid A, Monckeberg F, et al. Jejunal mucosa in infant malnutrition. Am J Clin Nutr 1968; 21:976–83.
57. Berkel I, Kiran O, Say B. Jejunal mucosa in infantile malnutrition. Acta Paediatr Scand 1970; 59:58–64.
58. Theron JJ, Wittmann W, Prinsloo JG. The fine structure of the jejunum in kwashiorkor. Exp Mol Pathol 1971; 58:184–99.
59. Schneider RE, Viteri FE. Morphological aspects of the duodenojejunal mucosa in protein-calorie malnourished children and during recovery. Am J Clin Nutr 1972; 25:1092–102.
60. Mayoral LG, Tripathy K, Bolanos O, et al. Intestinal, functional, and morphologic abnormal in severely protein-malnourished adults. Am J Clin Nutr 1972; 25:1084–91.
61. Shiner M, Redmond AOB, Hansen JDL. The jejunal mucosa in protein-energy malnutrition: A clinical, histological, and ultrastructural study. Exp Mol Pathol 1973; 19:61–78.
62. Duque E, Bolanos O, Lotero H, Mayoral LG. Enteropathy in adult protein malnutrition: Light microscopic findings. Am J Clin Nutr 1975; 28:901–13.
63. Duque E, Lotero H, Bolanos O, Mayoral LG. Enteropathy in adult protein malnutrition: Ultrastructural findings. Am J Clin Nutr 1975; 28:914–24.
64. Brunser O, Castillo C, Araya M. Fine structure of the small intestinal mucosa in infantile marasmic malnutrition. Gastroenterology 1976; 70:495–507.
65. Neto UF, Wehba J, Patricio FRS, et al. Morphological and functional study of the small intestine in marasmic patients. Arq Gastroenterol 1977; 14:241–8.
66. Kaschula ROC, Fajjar PD, Mann M, et al. Infantile jejunal mucosa in infection and malnutrition. Isr J Med Sci 1979; 15:356–61.
67. Nassar AM, El Tantawy SA, Khalifa S, et al. Ultrastructural changes in the mucosa of the small intestine due to protein-calorie malnutrition. J Trop Pediatr 1980; 26:62–72.
68. Zoppi G, Hadorn B, Gitzelmann R, Kistler H, Prader A. Intestinal-galactosidase activities in malabsorption syndromes. Gastroenterology 1966; 50:557–61.
69. Desjeux JF, Sassier P, Tichet J, et al. Sugar absorption by flat jejunal mucosa. Acta Pediatr Scand 1973; 62:531–7.
70. Berg NO, Dahlqvist A, Lindberg T, et al. Correlation between morphological alterations and enzyme activities in the mucosa of the small intestine. Scand J Gastroenterol 1973; 8:703–12.
71. Lebenthal E, Antonowicz I, Schwachman H. The interrelationship of enterokinase and trypsin activities in intractable diarrhea of infancy, celiac disease, and intravenous alimentation. Pediatrics 1975; 56:585–91.
72. Welch JD, Poley JR, Bhatia M, et al. Intestinal disaccharidase activities in relation to age, race, and mucosal damage. Gastroenterology 1978; 75:847–55.
73. Harrison M, Walker-Smith JA. Reinvestigation of lactose intolerant children: Lack of correlation between continuing lactose intolerance and small intestinal morphology, disaccharidase activity, and lactose tolerance tests. Gut 1977; 18:48–52.
74. Lebenthal E, Lee PC. Glucoamylase and disaccharidase activities in normal subjects and in patients with mucosal injury of the small intestine. J Pediatr 1980; 97:389–93.
75. Rossi TM, Lebenthal E, Nord KS, et al. Extent and duration of small intestinal mucosal injury in intractable diarrhea of infancy. Pediatrics 1980; 66:730–5.
76. Shulman RJ, Langston C, Lifschitz CH. Histologic findings are not correlated with disaccharidase activities in infants with protracted diarrhea. J Pediatr Gastroenterol Nutr 1991; 12:70–5.
77. Louvard D. The function of the major cytoskeletal components of the brush border. Curr Opin Cell Biol 1989; 1:51–7.

Glucose transporter mismatch; Maltase-glucoamylase; Marasmic malnutrition; Starch hydrolysis; Villous atrophy

© 2000 Lippincott Williams & Wilkins, Inc.