Secondary Logo

Journal Logo

BASIC SCIENCE

Inhibitory effect of HIV-1 Tat protein on the sodium-D-glucose symporter of human intestinal epithelial cells

Canani, Roberto Bernia; De Marco, Giulioa; Passariello, Annalisaa; Buccigrossi, Vittoriaa; Ruotolo, Serenaa; Bracale, Ileanaa; Porcaro, Francescoa; Bifulco, Giuseppeb; Guarino, Alfredoa

Author Information
doi: 10.1097/01.aids.0000198088.85572.68

Abstract

Introduction

Intestinal diseases are a hallmark of HIV-1 infection [1]. During the progression of the disease, chronic diarrhoea, dehydration, and malabsorption lead to progressive weight loss, and so contribute to the morbidity and mortality of HIV-1-positive subjects [2]. Functional and structural changes of gut mucosa may be detected before the onset of opportunistic infections and will eventually be responsible for intestinal dysfunction [1,3]. Carbohydrate and lipid malabsorption, and increased small bowel transepithelial permeability are common in patients who are not on HAART [3]. HAART rapidly improves intestinal sugar absorption [4–6]. The aetiology of HIV-1 associated intestinal dysfunction is largely unknown, and has been variously attributed to opportunistic infections, cytokine secretion in response to chronic inflammation, and a direct role of HIV-1 virus itself [1,3,7,8]. The latter concept is supported by the finding that recovery of intestinal digestive-absorptive function paralleled the decrease of viral load in children started on HAART [1,4]. Primary HIV-1 induced enteropathy is also consistent with the detection of viral proteins and/or nucleic acids in the intestinal mucosa of AIDS patients [9]. However, HIV-1 is not invariably found in the intestinal epithelium of AIDS patients. In addition, diarrhoea and nutrient malabsorption do not correlate with the presence of HIV-1 in gut mucosa [10].

Some of the effects induced by HIV-1 are not mediated by lytic propagation of viral particles but are induced by viral factors [11,12]. In addition to structural and enzymatic proteins, HIV-1 encodes a group of regulatory proteins including Tat, a transactivating peptide essential for HIV-1 replication [11,13–15]. Despite its nuclear localization, Tat is secreted from HIV-1-infected cells and taken up by uninfected neighbouring cells. Tat can occur in the sera of AIDS patients in the absence of massive lysis of infected cells, and is involved in many processes that contribute to immune and non-immune changes associated with HIV-1 infection [11,13–15]. Several effects induced by Tat require activation L-type Ca2+ channels and/or the mobilization of intracellular Ca2+ stores [11,13–15]. We previously reported that the addition of Tat to human enterocytes, and to human colonic mucosa, induces electrolyte secretion similar to that caused by classical bacterial enterotoxins, which suggests that Tat is directly involved in AIDS-related diarrhoea [16]. The finding that Tat induced a potent anti-proliferative effect in human enterocytes, links it to the pathogenesis of HIV-1-related intestinal mucosal atrophy [16].

Sugar malabsorption is the most frequent and severe feature of AIDS-related intestinal dysfunction, and it contributes to AIDS-associated malnutrition [17]. In the human intestine, and in the human intestinal cell line, Caco-2, glucose absorption is coupled with Na+ absorption through the Na+-D-glucose symporter 1 (SGLT-1) located on the enterocyte apical membrane. The transporter GLUT-2, which is located on the basolateral membrane, then carries intracellular glucose to the bloodstream [18–21]. The aim of this study was to test the hypothesis that, by inhibiting SGLT-1 activity in the intestinal epithelium, Tat is involved in the pathogenesis of glucose malabsorption in AIDS patients.

Materials and methods

Cell growth

Caco-2 cells were obtained from the American Type Culture Collection (Rockville, Maryland, USA). Cells were grown in Dulbecco's modified Eagle medium with high glucose concentration (4.5 g/l) supplemented with 10% foetal calf serum, 1% non-essential amino acids, penicillin (50 mU/ml), and streptomycin (50 mg/ml) and were incubated in 5% CO2/95% air. The medium was changed daily.

Glucose uptake studies

Caco-2 cells were grown on 24-well plates. After 15 days post-confluence cells were incubated for 30 min with the non-metabolizable radiolabelled glucose analogue [14C]-α-Methyl-L-D-glucopyranoside (AMG, 0.1 mM). The cells were lysed in 0.1 N NaOH. An aliquot was assayed for protein content (Bradford method, Bio-Rad Laboratories, Munich, Germany) and another for [14C]AMG content using a Packard scintillation spectrometer. To verify the presence of SGLT-1 activity in the cell line, the same experiment was performed in the presence of the selective competitive inhibitor of SGLT-1 phlorizin (100 μM) or in Na+-free buffer for 1 h (using choline chloride and K2HPO4 in place of NaCl and Na2HPO4 adjusted to pH 7.4 with KOH). Tat was added at increasing concentrations (from 0.01 to 1.0 nM) for 1 h, in the presence or absence of anti-Tat polyclonal antibodies (10: 1, w/w) or the specific L-type Ca2+ channels agonist, Bay K8644 (1 μM) as reported previously [16]. All data was expressed as c.p.m./mg protein.

Western blot analysis

The phenotypic expression of SGLT-1 was analysed in whole Caco-2 cell and in preparations of brush border membrane (BBM) vesicles. Briefly, cell blots and BBM vesicle preparations, obtained by magnesium precipitation method as described previously [22], were incubated with a rabbit polyclonal antibody, raised against the synthetic peptide corresponding to amino acids 564–575 of rabbit intestinal SGLT-1 sequence. Purified BBM vesicles from Caco-2 cells were pre-incubated with Tat and lysed with a buffer (150 mM NaCl, 10% glycerol, 10 mM EDTA, 10 mM Na4P2O7, 1 mM Na3VO4, 10 μg/ml aprotinine, 10 μg/ml leupeptine, 100 mM NaF, 1 mM phenilmethilsulphonyl fluoride, 1% Triton X-100 in 50 mM HEPES buffer, pH 7.5). In parallel experiments, the antibody was preadsorbed with the corresponding antigenic peptide, to confirm hybridization specificity. BBM vesicles were solubilized in Laemmli buffer (23 mmol/l Tris–HCl, pH 6.8, 2% SDS, 10% glycerol, 0.001% Bromophenol blue, and 5% 2-mercaptoethanol) and resolved by 8% SDS–PAGE. Proteins were electrotransferred onto nitrocellulose membranes using a transblot apparatus (Bio-Rad, Hercules, California, USA). Non-specific binding sites were blocked with phosphate-buffered saline (PBS) containing 0.05% Tween 20 and 3% bovine serum albumin (BSA). Blots were incubated with the primary antibody at a 1: 5000 dilution for 16 h at 4°C. In control experiments, nitrocellulose membranes were incubated with the same antibody previously pre-adsorbed with the antigenic peptide. Anti-SGLT-1 antibody was detected by enhanced chemiluminescence (Amersham International, Buckinghamshire, UK) using a peroxidase-conjugated anti-rabbit IgG (Sigma Chemical Co., St. Louis, Missouri, USA) as secondary antibody (1: 3000 dilution). After detection, hybridization bands were quantified by scanning densitometry.

Immunofluorescence of α-tubulin

Caco-2 cells were seeded in 24-well plates on glass coverslips. At 15 days post-confluence, cells were exposed to increasing Tat concentrations ranging 0.01 to 1.0 nM for 1 h. At the end of incubation, cells were fixed with 4% paraformaldehyde (w/v), then treated with NH4Cl (50 mM) in PBS for 10 min, and permeabilized with ice-cold methanol. Cells were blocked with 3% BSA (w/v) in PBS for 30 min. For α-tubulin staining, coverslips were incubated in a humid atmosphere with the specific primary antibody (1: 300) for 1 h at room temperature in blocking medium. Fluorescein-conjugated secondary antibody (1: 100) was added in 3% BSA in PBS for 1 h at room temperature. After washing, samples were fixed in 90% glycerol (v/v), 0.2% N-propylgallate (w/v) in PBS and observed with a fluorescence microscope (Nikon Eclipse E600, Tokyo, Japan).

Reagents

Non-metabolizable radiolabelled glucose analogous AMG was from Amersham Pharmacia Biotech (Milan, Italy). Phlorizin was from Sigma Chemical Co. Chemically synthesized, high-performance liquid chromatography 96% pure HIV-1 Tat, as well as rabbit polyclonal antibody anti-Tat were from Tecnogen (Piana di Monteverna, Italy). Mouse monoclonal antibody against α-tubulin and fluorescein-conjugated secondary antibody were from Sigma Chemical Co. Bay K8644 was purchased from Calbiochem (La Jolla, California, USA).

Statistics

Data are expressed as mean ± SE, and significance was evaluated by the nonparametric, two-tailed Mann–Whitney U test. A P value < 0.05 was considered significant. The SPSS software package for Windows (release 12.0.1; SPSS Inc., Chicago, Illinois, USA) was used for the statistical analysis.

Results

Effects of Tat on intestinal glucose uptake

Firstly, we determined the basal activity of SGLT-1 in Caco-2 cells by measuring [14C]-AMG absorption, in Na+-free medium or in the presence of phlorizin. As shown in Fig. 1, glucose uptake was significantly inhibited in Na+-free conditions and in the presence of phlorizin, which demonstrated normal functioning of the Na+-dependent glucose absorption pathway in Caco-2 cells. We then performed experiments to test the effects of increasing doses of Tat. Glucose uptake was significantly inhibited by incubation for 1 h with Tat. The effect was dose-dependent and saturable with a maximal effective concentration of 0.1 nM (Fig. 2). The magnitude of the maximal inhibitory effect induced by Tat was comparable to that observed with the maximal effective dose of phlorizin (70% vs. control cells). To investigate the specificity of Tat effects on glucose uptake, neutralization experiments were performed in the presence of specific anti-Tat antibodies. In this condition, the inhibitory effect of Tat on glucose uptake was almost totally abolished (Fig. 3). As Tat activates L-type Ca2+ channels, we also tested the effect of the specific L-type Ca2+ channels agonist BayK8644 on glucose uptake. BayK8644 alone did not significantly affect Caco-2 cells glucose uptake. However, it significantly inhibited the effect induced by Tat on glucose absorption (Fig. 3).

Fig. 1
Fig. 1:
Characterization of Ne+-D-glucose symporter activity in differentiated Caco-2 cells. The uptake of 0.1 μM of non-metabolizable radiolabelled glucose analogous AMG was measured at a single time point (1 h) in standard conditions (white bar) or in Na+-free condition (lined bar) or in the presence of phlorizin (pointed bar). Values are means ± SE of three independent measurements. *P < 0.01 versus standard condition.
Fig. 2
Fig. 2:
Effect of increasing concentrations of Tat on non-metabolizable radiolabelled glucose analogous AMG uptake in Caco-2 cells. Data are means of six different observations for each data point.
Fig. 3
Fig. 3:
Modifications of non-metabolizable radiolabelled glucose analogous AMG absorption in human enterocytes induced by Tat in various experimental conditions. Experiments were performed to investigate the involvement of Ca2+ (in the presence of Bay K8644), and the specificity (in the presence of anti-Tat antibody) of the effect of Tat in glucose transport. Values are expressed as the maximal percent of glucose transport inhibition in cells exposed for one hour to Tat versus control cells. Data are means ± SE. *P < 0.01 versus Tat in standard conditions.

Western blotting analysis

To test whether the inhibitory Tat effect on glucose uptake was dependent on SGLT-1 expression, we studied the symporter expression either in whole Caco-2 cells and in BBM vesicle preparations. Total SGLT-1 protein intracellular expression was not affected by incubation with Tat concentrations up to 1.0 nM. On the contrary, incubation of Caco-2 cells with Tat resulted in a dose-dependent inhibition of SGLT-1 expression in BBM (Fig. 4). Maximum inhibition was obtained with 0.1 nM (corresponding to the concentration that induced the maximal effect on glucose uptake). These experiments suggested that Tat induces a missorting of the symporter to apical membrane.

Fig. 4
Fig. 4:
Na+-D-glucose symporter 1 (SGLT-1) protein expression in Caco-2 cells after 1 h of incubation with increasing concentrations of Tat. (a) The symporter expression is revealed by the appearance of a 75-kDa band that corresponds to human SGLT-1. SGLT-1 protein expression was sought in whole cells and in BBM vesicles to evaluate the delivery of the symporter to the apical surface of the enterocyte. (b) Densitometric quantification of the SGLT-1 band expression in BBM vesicles reported in (a). Results are representative of three repetitive experiments.

Tat effect on α-tubulin staining

We analysed the effect of Tat on the cytoskeleton organization in Caco-2 cells using α-tubulin as a marker. Following a 1-h exposure to Tat, cells displayed a dramatic decrease in α-tubulin staining, consistent with substantial dysruption in cytoskeleton organization. The maximal effect was observed with 0.1 nM (Fig. 5).

Fig. 5
Fig. 5:
Effect of Tat protein on α-tubulin staining in Caco-2 cells. Enterocytes were analysed by confocal microscopy at × 100 magnification. In these experiments microtubules appeared in green upon indirect immunofluorescence staining with an anti-α-tubulin monoclonal antibody and secondary fluorescein-conjugated anti-mouse IgG. When compared with controls (a), cells exposed to 0.1 nM Tat for 1 h displayed a marked decrease in α-tubulin labelling (b). Results are representative of three repetitive experiments.

Discussion

Glucose malabsorption is a major feature of the complex picture defined as intestinal dysfunction in HIV-1 infected children [3]. Our data indicate that Tat peptide directly impairs intestinal glucose absorption by inhibiting the SGLT-1 activity on enterocyte brush border. Decreased activity of this symporter may result not only in sugar malabsorption but also in diarrhoea, as SGLT-1 has the properties of a water channel [23]. Regardless of the route of transmission, HIV-1 selects CD4 cells that have surface receptors known as CCR5. The vast majority of CD4/CCR5 cells reside in the gut, which is considered a major target of HIV infection and replication and CD4 T-cell depletion predominantly occurs in the gastrointestinal tract [24]. It has been estimated that in the human small intestine, SGLT-1-mediated active fluid transport can account for as much as 5 l per day [23]. Our results support the concept of a direct etiologic role for Tat in the well-known pathogen-negative AIDS-related diarrhoea. Interestingly, the effects of Tat on SGLT-1 were dose-dependent with a maximal effective dose of 0.1 nM, which is well within the range of what generally measured in the sera of patients with HIV-1 infection [16]. Similar to the Tat effects on ion transport and on cell proliferation [16], also the inhibition of glucose uptake involves L-type Ca2+ channels as suggested by the experiments with the specific L-type Ca2+ channels agonist Bay K8644. Interestingly, Tat and Bay K8644, compete for binding to dendritic cells, which reinforces the concept that the effects of Tat on glucose uptake are L-type Ca2+ channel-dependent [25].

Microtubules are normally present in enterocytes and are important for intracellular transport [8]. Microtubule-disrupting drugs such as colchicine and nocodazole cause acute diarrhoea and missorting of several apical proteins in the enterocytes, including SGLT-1 [12,26]. Enteric microtubule depolymerization occurs in HIV-1 infected individuals [8]. In addition, α-tubulin staining was dramatically decreased in the intestinal HT29 cell line after exposure for 1 h to HIV-1, probably as a consequence of direct Gp120 action [12]. The decreased α-tubulin staining is consistent with a major change in cytoskeleton organization which, in turn, could lead to SGLT-1 missorting. The total SGLT-1 expression remained stable within the cell, whereas the symporter expression at BBM level was significantly decreased suggesting a functional rather than direct structural damage. The similar dose response profile of the effects exerted by Tat on ion and water transport, sugar absorption and cell structural damage suggest that Tat-induced enterocyte alterations occur via a single pathway. This pathway recalls that induced by the non-structural peptide 4 (NSP4) produced by Rotavirus [27]. Like NSP4, Tat is a protein capable of inducing Ca2+ dependent enteropathogenic and enterotoxigenic effects and of inhibiting glucose uptake by causing changes in the enterocyte cytoskeleton [16,27–30]. Such peptides are called ‘virotoxins’ [27]. Collectively, our previous findings on the Tat-enterocyte interaction [16] together with the results of the present study suggest that glucose malabsorption in AIDS patients results from the following cascade: 1) binding of Tat to plasma membrane of the enterocyte, 2) increase in intracellular Ca2+ concentration, 3) depolymerization of microtubules, 4) accumulation of transporting vesicles containing brush border proteins, 5) missorting of SGLT-1,which results in inhibition to glucose uptake. Calcium-dependent pathway is one of the four established intracellular signal transduction mechanisms leading to water and electrolyte secretion, the other three being cAMP, cGMP and Nitric oxide intracellular concentrations [31]. Following the increase in intracellular Ca2+ concentration and in parallel with structural damage, Tat also induces net Cl secretion [16]. The clinical manifestations of Tat effects are nutrient malabsorption and large volumes of diarrhoea, commonly observed in the advanced stages of HIV-1 infection. With the rapid turnover of intestinal cells, the acute impairment of the symporter by circulating Tat may induce constant damage to maturing enterocytes thereby causing chronic sugar malabsorption that is observed in patients with AIDS. Whether not all patients with high viral load have diarrhoea and/or intestinal malabsorption is not known. However, diarrhoea may not be evident in all patients with high circulating Tat due to the variable role of homeostatic pathway. However the observed restoration of intestinal digestive-absorptive functions, in parallel with the decrease of viral load (4), and the decreased risk of diarrhea in patients undergoing HAART support the direct involvement of HIV in both ion secretion and enterocyte damage [32].

There are at least two therapeutic implications of this work: first, the classic glucose-containing oral rehydration solution may be relatively ineffective for treatment of dehydration in these patients. This is supported by the clinical data reporting a high rate of parenteral rehydration in HIV-1-infected subjects [33]. Second it is of clinical relevance that all Tat effects on the enterocytes are substantially blocked by specific antibodies, which suggests that interdiction of extracellular Tat by active or passive immunization would reduce its pathogenic effects in the intestine.

Sponsorship: Supported by a grant from Ministero della Salute 4th AIDS Research Project, Program 50 D.28.

References

1. Janoff EN, Smith PD. Emerging concepts in gastrointestinal aspects of HIV-1 pathogenesis and management. Gastroenterology 2001; 120:607–621.
2. Sharpstone D, Gazzard B. Gastrointestinal manifestations of HIV infection. Lancet 1996; 348:379–383.
3. Wittenberg D, Benìtez CV, Berni Canani R, Hadigan C, Medeiros Perin N, Rabinowitz S, et al. HIV Infection: Working Group Report of the Second World Congress of Pediatric Gastroenterology, Hepatology, and Nutrition. J Pediatr Gastroenterol Nutr 2004; 39:S640–S646.
4. Berni Canani R, Spagnuolo MI, Cirillo P, Guarino A. Ritonavir combination therapy restores intestinal function in children with advanced HIV disease. J Acquir Immune Defic Syndr 1999; 21:307–312.
5. Guarino A, Spagnuolo MI, Giacomet V, Berni Canani R, Bruzzese E, Giaquinto C, et al. Effect of nutritional rehabilitation on intestinal function and on CD4 cell number in children with HIV. J Pediatr Gastronterol Nutr 2002; 34:366–371.
6. The Italian Paediatric Intestinal/HIV Study Group. Intestinal malabsorption of HIV-infected children: relationship to diarrhoea failure to thrive, enteric micro-organisms and immune impairment.AIDS 1993, 7:1435–1440.
7. Winter H. Gastrointestinal tract function and malnutrition in HIV-1 infected children. J Nutr 1996; 126:S2620–S2622.
8. Clayton F, Kapetanovic S, Kotler DP. Enteric microtubule depolymerization in HIV infection: a possible cause of HIV-associated enteropathy. AIDS 2001; 15:123–124.
9. Kotler DP, Gaetz HP, Lange M, Klein EB, Holt PR. Enteropathy associated with the acquired immunodeficiency syndrome. Ann Intern Med 1984; 101:421–428.
10. Seideman EG, Russo P. Gastrointestinal manifestation of human immunodeficiency virus infection and other secondary immunodeficiencies. In: Pediatric Gastrointestinal Disease, 3rd ed. Walker WA, Durie PR, Hamilton JR, et al. Hamilton, ON: BC Decker Inc. 2000:548–568.
11. Rubartelli A, Poggi A, Sitia R, Zocchi MR. HIV-1 Tat: a polypeptide for all seasons. Immunol Today 1998; 19:543–545.
12. Delezay O, Yahi N, Tamalet C, Baghdguian S, Boudier JA, Fantini J. Direct effect of type 1 human immunodeficiency virus (HIV-1) on intestinal epithelial cell differentiation: relationship to HIV-1 enteropathy. Virology 1997; 238:231–242.
13. Gallo RC. Tat as one key to HIV-induced immune pathogenesis and Tat toxoid as an important component of a vaccine. Proc Natl Acad Sci USA 1999; 96:8324–8326.
14. Zocchi MR, Rubartelli A, Morgavi P, Poggi A. HIV-1 Tat inhibits human natural killer cell function by blocking L-type calcium channels. J Immunol 1998; 161:2938–2943.
15. Haughey NJ, Holden CP, Nath A, Geiger JD. Involvement of inositol 1,4,5-trisphosphate-regulated stores of intracellular calcium in calcium dysregulation and neuron cell death caused by HIV-1 protein Tat. J Neurochem 1999; 73:1363–1374.
16. Berni Canani R, Cirillo P, Mallardo G, Buccigrossi V, Secondo A, Annunziato L, et al. Effects of HIV-1 Tat protein on ion secretion and on cell proliferation in human intestinal epithelial cells. Gastroenterology 2003; 124:368–376.
17. Miller TL, Orav EJ, Martin SR, Cooper ER, McIntsh K, Winter HS. Malnutrition and carbohydrate malabsorption in children with vertically trasmissed human immunodeficiency virus 1 infection. Gastroenterology 1991; 100:1296–1302.
18. Thomson AB, Wild G. Adaption of intestinal nutrient transport in health and disease. Dig Dis Sci 1997; 42:453–488.
19. Wright EM, Hirsch JR, Loo DD, Zampighi GA. Regulation of Na+/glucose cotransporters. J Exp Biol 1997; 200:287–293.
20. Ferraris RP. Dietary and developmental regulation of intestinal sugar transport. Biochem J 2001; 360:265–276.
21. Kellett GL. The facilitated component of intestinal glucose absorption. J Physiol 2001; 531:585–595.
22. Hauser H, Howell K, Dawson RM, Boyer DE. Rabbit small intestinal brush border membrane preparation and lipid composition. Biochim Biophys Acta 1980; 602:567–577.
23. Wright EM, Loo DD, Turk E, Hirayama BA. Sodium cotransporters. Curr Opin Cell Biol 1996; 8:468–473.
24. Brenchley JM, Schacker TW, Ruff LR, Price DA, Taylor JH, Beilman GJ, et al. CD4+ T cell depletion during all stages of HIV disease occurs predominantly in the gastrointestinal tract. J Exp Med 2004; 200:749–759.
25. Poggi A, Rubartelli A, Zocchi MR. Involvment of dihydropyridine-sensitive calcium channels in human dendritic cell function. Competition by HIV-1 Tat. J Biol Chem 1998; 273:7205–7209.
26. Glickman RM, Pernotto JL, Kirsch K. Intestinal lipoprotein formation: effect of colchicine. Gastroenterology 1976; 70:347–352.
27. Morris AP, Estes MK. Microbes and microbial toxins: paradigms for microbial-mucosal interactions. VIII. Pathological consenquences of rotavirus infection and its enterotoxin. Am J Physiol Gastrointest Liver Physiol 2001; 281:G303–G310.
28. Halaihel N, Lievin V, Alvarado F, Vasseur M. Rotavirus infection impairs intestinal brush-border membrane Na+-solute cotransport activities in young rabbits. Am J Gastroenterol Liver Physiol 2000; 279:G587–G596.
29. Halaihel N, Lievin V, Ball JM, Estes MK, Alvarado F, Vasseur M. Direct inhibitory effect of Rotavirus NSP4 (114–135) peptide on the Na+-D-Glucose symporter of rabbit intestinal brush border membrane. J Virol 2000; 74:9464–9470.
30. Jourdan N, Brunet JP, Sapin C, Blais A, Cotte-Laffitte J, Forestier F, et al. Rotavirus infection reduces sucrase-isomaltase expression in human intestinal epithelial cells by perturbing protein targeting and organization of microvillar cytoskeleton. J Virol 1998; 72:7228–7236.
31. Fasano A. Toxins and the gut: role in human disease. Gut 2002; 50(Suppl 3):9–14.
32. Guarino A, Bruzzese E, De Marco G, Buccigrossi V. Management of gastrointestinal disorders in children with HIV infection. Pediatr Drugs 2004; 6:347–362.
33. Smith PD, Janoff EN. Gastrointestinal complications of the acquired immunodeficiency syndrome. In: Textbook of Gastroenterology, 4th edn. Yamada T, Alpers DH, Kaplowitz N, et al. (eds). Philadelphia: Lippincott Williams & Wilkins; 2003: 2567–2589.
Keywords:

AIDS; intestinal glucose absorption; intestinal dysfunction; diarrhoea

© 2006 Lippincott Williams & Wilkins, Inc.