Human intestinal glucose transporter and localization of GLUT5 NICHOLAS 0. DAVIDSON, ANNALISE JOHN B. BUSE, GWYN W. GOULD,
expression
M. L. HAUSMAN, CARMEN CHARLES F. BURANT, AND
A. IFKOVITS, GRAEME I. BELL
Departments of Medicine, Biochemistry, and Molecular Biology and Howard Hughes Medical Institute, The University of Chicago, Chicago, Illinois 60637; and Molecular Pharmacology Group, Department of Biochemistry, The University of Glasgow, Glasgow G12 8&Q, United Kingdom Davidson, Nicholas O., Annalise M. L. Hausman, Carmen A. Ifkovits, John B. Buse, Gwyn W. Gould, Charles F. Burant, and Graeme I. Bell. Human intestinal glucose transporter expression and localization of GLUT5 Am. J. Physiol. 262 (Cell Physiol. 31): C795C800, 1992.-We have
studied the developmental and regional expressionof mRNAs encodingsodium-dependentand facilitative glucosetransporter proteins in human fetal and adult small intestine. The abundance of mRNAs encoding the Na’-glucose cotransporter isoform SGLTl and the facilitative glucosetransporter isoforms GLUT2 and GLUT5 is developmentally modulated with highest levels in adult small intestine. By contrast, the levels of GLUT1 mRNA are higher in fetal than adult small intestine. Immunohistochemicalanalysis of adult small intestine localized GLUT5 to the luminal surface of mature enterocytes, a finding confirmed by Western blot analysis of purified human jejunal brush-bordermembranes.By contrast, in the fetal small intestine, GLUT5 was localized along the intercellular junctions of the developingvillus, indicating that both its expression and localization are developmentally regulated. The localization of GLUT5 to the luminal surface of mature absorptive epithelial cells implies that this protein participates in the uptake of dietary sugars. brush border; plasma membrane; immunohistochemistry; Western blot
and transcellular delivery of dietary glucose is an important function of the small intestinal enterocyte. Adult mammalian small intestine expresses a number of genes involved in glucose transport including Na+dependent as well as facilitative glucose transporters (reviewed in Ref. 2). The intestinal brush-border Na+glucose cotransporter isoform SGLTl functions in the uptake of glucose across the microvillus membrane (4, 10, 11). The key role played by SGLTl in the active transport of glucose in the small intestine is evident from studies of individuals with glucose/galactose malabsorption syndrome (20). This recessive genetic disorder which is characterized by an inability to absorb dietary glucose and galactose from the small intestine can arise from mutations in SGLTl, as recently demonstrated (20). In addition to SGLTl, biochemical studies suggest that other proteins may also participate in the transport of glucose from the lumen of the small intestine into the enterocyte (1, 14). Glucose taken up by the enterocyte is released across the basolateral membrane into the interTHE UPTAKE
stitium via facilitative glucose transport. Recent studies have revealed that the facilitative glucose transporters comprise a family of structurally related proteins. cDNAs encoding five members of this family have been cloned and characterized (reviewed in Ref. 2). The five facilitative glucose transporters have been designated as GLUTl/erythrocyte type, GLUT2/ liver type, GLUT3/brain type, GLUT4/muscle-fat type, and GLUT5/smail intestine type, and all are expressed at varying levels in adult human small intestine (5-7,12, 13). However, only GLUT2 and GLUT5 represent relatively abundant mRNA species in the small intestine (7, 12), implying that these two isoforms may participate in the transepithelial movement of glucose. GLUT2 has been localized to the basolateral membrane of fully differentiated absorptive epithelial cells in the adult rat small intestine, consistent with it mediating the efflux of absorbed glucose from these cells (19). The localization of GLUT5 within the intestine has not been determined. In the present study, the regional and developmental regulation of glucose transporter mRNAs has been characterized in human fetal and adult small intestine. In addition, the localization of GLUT5 protein has been determined by immunohistochemical techniques and Western blotting. GLUT5 is confined to the brush border of mature enterocytes populating the upper half of the villus in adult small intestine, consistent with this protein acting in concert with SGLTl to promote the uptake of sugars from the lumen of the small intestine. MATERIALS
AND
METHODS
Tissue procurement, RNA extraction, and analysis. Human fetal intestine was obtained during the course of elective or spontaneousterminations of first and secondtrimester pregnancies as approved by the Institutional Review Board of the University of Chicago Hospitals. Adult tissue was obtained during diagnostic upper or lower endoscopicexamination or from organ donor referrals (National DiseaseResearchInterchange, Philadelphia, PA). Tissue wasobtained within 30 min of removal and was either snap frozen in liquid nitrogen or usedimmediately for RNA extraction. RNA was extracted into a solution of 4 M guanidinium thiocyanate and 0.3 M of 2-mercaptoethanolas described(18) with the following modifications. After the first precipitation with isopropanol, sampleswere extracted six times with ice-
0363-6143/92 $2.00 Copyright 0 1992 the American Physiological Society
c795
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
C796
GLUCOSE TRANSPORTER
EXPRESSION
cold 4 M LiCl and pelleted. After resuspension in freshly prepared 4 M guanidinium thiocyanate and precipitation with isopropanol, samples were suspended in 10 mM tris(hydroxymethyl)aminomethane (Tris) . HCl, pH 7.4, 1 mM EDTA, and 0.5% sodium dodecyl sulfate (SDS) and extracted with chloroform-isoamyl alcohol (24:l). The RNA was ethanol precipitated and then finally resuspended in ribonuclease (RNase)-free water and stored at -80°C. Northern blots were prepared using 20 rg of total RNA that was fractionated through 1% agarose-2% formaldehyde gels before capillary transfer onto nylon membranes (18). The following human facilitative glucose transporter cDNA clones were used as probes: GLUTl, phGT2-1 (6); GLUT2, pGEM4ZHTL3 (7); GLUT3, PBS-MGT3 (13); GLUT4, pGEM4ZAMT7 (5); and GLUT5, phJHT5 (12). The human SGLTl cDNA clone phNaGTB-2 was isolated from an adult human jejunum cDNA library and includes nucleotides 201-2,343 of the published cDNA sequence (11). The human sucrase-isomaltase cDNA clone pSl-22 was generously supplied by Dr. D. Swallow (8). The cDNA inserts from these clones were labeled by random priming. The hybridizations and washes were conducted as described (18), with a final wash in 15 mM NaCl-1.5 mM sodium citrate and 0.1% SDS at 65°C for 30 min. Antibody production. Antibodies against a peptide corresponding to the COOH-terminus of human GLUT5 (amino acids 488-501) were prepared by immunizing rabbits with a GLUT5-specific peptide (KEELKELPPVTSEQ) which had a cysteine added at the NHz-terminus to allow coupling to keyhole limpet hemocyanin (Sigma). Rabbits (New Zealand White) were immunized with 250 pg of conjugated peptide in a 40:60 mixture of phosphate-buffered saline (PBS)/Freund’s complete adjuvant at eight intradermal sites. One month later, animals were boosted with 250 pg of conjugated peptide in a 60:40 mixture of PBS/Freund’s incomplete adjuvant. Twenty milliliters of blood were collected 2 wk later, and anti-human GLUT5 antibodies were affinity purified over a peptide-coupled immunopurification column (Immunopure Ag/Ab, Pierce, Rockland, IL) as described by the manufacturer. Absorption of GLUT5 immunoreactivity was achieved by incubating antiserum with GLUT5 peptide (1 mg/ml) at 4°C for 12 h. Immunolocalization of GLUT5 in small intestine. Tissue samples were fixed in Bouin’s solution for 2 h, and 5-pm sections were prepared from paraffin-embedded blocks. Samples were blocked at 23°C for 20 min in blocking buffer (PBS containing 2% bovine serum albumin, 0.5% nonfat dry milk, 0.3% Triton X-100, and 0.05% sodium azide) containing 10% normal goat serum. Tissue sections were then incubated at 4°C overnight with a 1:50 dilution (in blocking buffer) with the anti-COOHterminus affinity-purified GLUT5 antibody. After five washes with PBS, samples were reblocked as above. They were then incubated for 1 h at 23°C with a 1:40 dilution (in blocking buffer) of gold-labeled goat anti-rabbit immunoglobulin G (Aurodye, Amersham, Arlington Heights, IL). After five washes in PBS and glass-distilled water, silver enhancement was carried out as suggested by the manufacturer. Sections were visualized under epifluoresence using the Olympus IGS cube and photographed with Kodak Ektachrome 400 film. Plasma membrane protein preparation and Western blotting. Human jejunal and both proximal and distal colonic brushborder membranes, prepared from mucosal scrapings as described (9, 16), were provided by Drs. T. Brasitus and J. Harig, Section of Gastroenterology, University of Chicago. Small intestinal brush-border membranes demonstrated at least 16-fold enrichment of sucrase activity in comparison with the starting homogenate. Protein (100 fig) from each preparation was separated by 10% polyacrylamide-SDS gel electrophoresis and transferred electophoretically to Immobilon membranes (Millipore, Bedford, MA). The membranes were blocked using a
IN HUMAN
INTESTINE
solution of 5% nonfat dry milk in TTBS (0.05% Tween 20, 10 mM Tris. HCl, pH 7.4, and 500 mM NaCl) and then incubated for 2 h at room temperature with a 1:500 dilution (in TTBS) of the rabbit polyclonal antibody to human GLUT5 described above. The membranes were washed three times in TTBS, and antibody binding was detected by incubation with iz51-protein A and autoradiography using Kodak XAR-5 film and an intensifying screen at -80°C for 16 h. RESULTS
Glucose transporter expression in human small intestine. SGLTl, GLUT2, and GLUT5 mRNAs are readily detected by Northern blotting in adult small intestine (Fig. 1). Multiple transcripts encode each of these glucose transporters as noted previously (7, 11, 12). These different transcripts are believed to be due to alternative A
SGLTl
FETAL 8 11.513.515 17 19’
ADULT ’ D J I I
- 2.2
kb
18SrRNA
B GLUT1
FETAL
t 15
ADULT
17 ’ D
J
ie I Co”D
J
I
. ..
ma
-3.0
kb
15SrRNA
C GLUT2
FETAL ’ 11 12.5
ADULT 15 ’ ’ J I ’ -5.4
kb
-3.4 -2.3
kb kb
FETAL ’ 11 12.5 15’
D GLUT5
ADULT ’ J I ’
(a) -5.1
kb
-2.3
kb
-2.0
kb
18SrRNA
Fig. 1. Expressionof glucose transportermRNAsin humanfetal and adult small intestine. Total RNA was extracted from various gestational age (weeks) fetal and adult small intestines. Northern blots were preparedusing20pg total RNA. Membranes were hybridizedsequentially with the cDNA probes indicated. For GLUT5 (D), exposure was extended from 2 (a) to 14 days (b) to show the low levels of fetal mRNA. For eachset of blots, hybridizationto 18s rRNA confirmed equivalence of loading and transfer. D, duodenum; J, jejunum; I, ileum;
and Co, colon. Adult intestinal RNA was obtainedfrom the same individual for the regions shown.
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
GLUCOSE
TRANSPORTER
A
EXPRESSION
IN HUMAN
B J
I
Co
-6.0
*
lsp
-
‘%I D ADULT
c797
C D
170
INTESTINE
kb
*
J ,am
A’
Fig. 2. Developmental and regional expression of sucrase-isomaltase mRNA in human intestine. A: total small intestinal RNA (5,2.5, and 1 pg) from fetal (gestational age indicated in weeks) and adult (D, duodenum; J, jejunum; I, ileum) intestine was immobilized directly onto nylon membranes using a slot-blot format. Where sample permitted, aliquots of both proximal (P) and distal (D) fetal small intestine were analyzed. Blot was hybridized with a 32P-labeled human sucrase-isomaltase cDNA probe. B: 20 pg of total RNA from different regions of the same 17-wk human fetal intestine (D, duodenum; J, jejunum; I, ileum; Co, colon) were fractionated on a 1% agarose-formaldehyde gel, blotted, and hybridized with a human sucrase-jsomaltase cDNA probe. Probe specifically hybridizes to an mRNA of 6 kb, as noted previously (8). C: slot-blot shown in A was stripped and rehybridized with a cDNA to human B-actin.
polyadenylation generating transcripts having 3’-untranslated regions of different sizes (12). In fetal small intestine, SGLTl mRNA could be detected by Northern blotting at 17 wk, and by 19 wk, the levels were comparable to those found in the adult (Fig. 1A). As shown in Fig. lA, SGLTl mRNA abundance appears higher in adult ileum than in duodenum or jejunum in tissue samples obtained from the same individual. However, in a second individual, there were comparable levels of SGLTl mRNA in ileum and jejunum (data not shown). Both individuals were male, aged 37 and 32 yr, and were otherwise healthy. Based on this limited sample, the data suggest that there may be interindividual variation in levels of SGLTl mRNA in different regions of the small intestine, the molecular basis for which is unknown. Low levels of GLUT2 mRNA could be detected by Northern blotting at 11 wk of gestation (Fig. lC), the earliest time studied. The levels then increase with gestational age and are highest in adult small intestine. The levels of GLUT2 mRNA in adult jejunum and ileum appear to be similar. By contrast, the levels of GLUT5 mRNA are much lower in fetal small intestine (and only evident with long autoradiographic exposure, Fig. 1D) than those seen in the adult. The levels of GLUT5 mRNA in jejunal and ileal RNA preparations from the same individual were similar. Thus, while not discounting the potential importance of individual variability, mRNAs encoding all the abundant glucose transporter genes appear to be expressed throughout the adult small intestine with no evidence of a proximal to distal gradient of gene expression. We have previously demonstrated low levels of GLUTl, GLUT3, and GLUT4 mRNA in adult human small intestine (5,6,13). The present study reveals levels of GLUT1 mRNA to be higher in fetal than in adult
small intestine with expression additionally demonstrated in fetal colon (Fig. 1B). The decrease observed in GLUT1 mRNA abundance in fetal as compared with adult small intestine has been previously noted for rat liver, heart, lung, and muscle (21). Under the experimental conditions of this study, GLUT3 mRNA could not be detected by Northern blotting, implying that the levels of GLUT3 mRNA are much lower than those of the other isoforms. Additionally, GLUT4 mRNA, which is believed to originate from smooth muscle cells and possibly adherent adipocytes (7), was only barely detectable after prolonged autoradiographic exposure (data not shown). Sucrase-isomaltase expression in human small intestine. For purposes of comparison, the abundance of
mRNA encoding a member of another class of proteins involved in luminal carbohydrate absorption was also examined. Sucrase-isomaltase mRNA was first detectable in 13-wk fetal intestine, and its abundance increased in the proximal (but not distal) intestine at 15 and 17 wks of gestation (Fig. 2A). For comparison purposes, the blot was rehybridized with a @-actin cDNA to demonstrate equivalent RNA loading (Fig. 2C). By 17 wk, sucrase-isomaltase mRNA was detectable in all regions of the fetal intestine including colon (Fig. 2B). Adult intestine demonstrated roughly equal abundance of suerase mRNA in all regions of the small intestine, but no signal was detectable in the adult colon (data not shown). The changing pattern of expression of sucrase-isomaltase mRNA in the small intestine thus appears to parallel that seen for SGLTl and GLUT2. Immunolocalization of GLUT5 in human small intestine. The distribution of GLUT5 in fetal and adult small
intestine was studied in tissue sections using antibodies directed against the COOH-terminus of this protein. Fetal small intestine showed immunoreactivity princi-
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
C798
GLUCOSE
TRANSPORTER
EXPRESSION
IN
HUMAN
INTESTINE
Fig. 3. Immunogold epifluorescence localization of human GLUT5 A-D: fetal small intestine (12 wk). E-H: adult duodenum. Z and J, adult colon. Samples were incubated with affinity-purified rabbit anti-GLUT5 antibodies (A, E, and I) or preimmune rabbit immunoglobulin G (C and G) followed by gold-conjugated goat anti-rabbit IgG as described in MATERIALS AND METHODS. Note staining pattern between cells of the fetal intestine (arrows, A). Specific GLUT5 staining in the adult duodenal sample (E) was confined to the brush border of the top third of the villus with diminished intensity noted in the middle to lower region (arrow). Phase-contrast pictures of the respective sections are also shown in B, D, F, H, and J.
pally between cells of the immature villus (Fig. 3A, arrows), while immunoreactivity in the adult small intestine was predominantly along the brush border of cells lining the upper third of the villus (Fig. 3E). The immunoreactive material in the lamina propria was also seen with the preimmune serum (Fig. 3G) and presumably reflects nonspecific binding of the antibody to structures in this region of the villus. No antibody staining was detected in the colon (Fig. 31), which is consistent with Northern blotting data showing the absence of GLUT5 mRNA in this region of the human intestine. The localization of GLUT5 to the luminal surface of the absorptive epithelial cells was also examined by Western blotting of brush-border membranes isolated from human small intestine and colon. The GLUT5 antibody specifically reacted with a jejunal brush-border
protein of -50 kDa (Fig. 4), similar to the predicted size of human GLUT5 (M, = 54,983; Ref. 12). The absence of a signal in either homogenate or a crude microsomal fraction (containing a mixture of organellar, brush-border, and plasma membranes) is consistent with the polarized expression of GLUT5 in jejunal brush-border membranes. There was also no signal detected using either proximal or distal colonic brush-border membranes, findings consistent with the distribution of GLUT5 determined by both the Northern blotting and immunohistochemical studies. DISCUSSION
The uptake and release of glucose by epithelial the small intestine is carried out by members
cells of of two
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
GLUCOSE
Absorbed
aGLUT
TRANSPORTER
EXPRESSION
Non-Absorbed
aGLUT
Jejunum
Colon
IN HUMAN
INTESTINE
c799
mRNAs suggests that they may be coordinately regulated in preparation for the changes in nutrition that occur I postnatally. II Such adaptive changes in glucose trans‘H M BBM Prox Dist ’ H M BBM’ porter activity have been inferred from biochemical studkDa ies in rat enterocytes, providing some indirect support 106for this hypothesis (3). The changing pattern of expres60sion of these genes occurs in the setting of morphological differentiation of the human intestine which occurs during the second trimester. During the period from 10 to 49.5 20 wk gestation, the villus-crypt axis undergoes organization and discrete cell-types emerge, each with a distinct morphology (15). Demonstration of GLUT5 immuno32.5 reactivity in the intercellular regions of 12-wk fetal small intestine may reflect the relative immaturity of the 27.5brush-border membrane at this stage of gestation. The roles of SGLTl and GLUT2 in transepithelial transport of glucose seem firmly established. However, the function of GLUT5 in this process is still a matter Fig. 4. Western blot analysis of GLUT5 distribution in human jeof conjecture. The present data suggest that GLUT5 junum and proximal and distal colon. Protein (100 pg) from each mRNA and protein are relatively abundant in the small preparation of homogenate (H), total microsomal (M), or brush-border membranes (BBM) was separated by 10% polyacrylamide-sodium do- intestine, and the localization of GLUT5 to the luminal decyl sulfate electrophoresis and transferred to an Immobilon memmembrane of absorptive epithelial cells is consistent with brane. Half of the membrane was incubated with absorbed GLUT5 a role for this protein in the transport of sugars from the antibodies, and the other half was incubated with unabsorbed GLUT5 antibodies. Antibody binding was detected by incubation with iz51- lumen of the small intestine. It is tempting to speculate protein A followed by autoradiography. that GLUT5 may have a substrate specificity different from that of the other facilitative glucose transporters, different families of glucose transporter proteins, the and this possibility is currently under investigation. In SGLT family of Na+-glucose cotransporters and the this context, studies are underway to define the biochemGLUT fam.ily of facilitative glucose transporters. In the ical properties of GLUT5 in more detail to understand small intestine, SGLTl participates in the active uptake how it participates in sugar uptake by the small intestine. and accumulation of glucose across the luminal membrane of absorptive epithelial cells, and GLUT2 mediates We thank Drs. T. Brasitus and J. Harig, Section of Gastroenterology, its release across the basolateral membrane of these cells University of Chicago, for providing human small intestinal and colonic brush-border membranes for localization of GLUT5 into the interstitium. The pattern and relative abunThese studies were supported by National Institutes of Health dance of SGLTl mRNA in the human intestine pre- Grants DK-42086, DK-20595, HL-02166, and HL-38180, the Howard sented in this report are identical to those reported by Hughes Medical Institute, and the Juvenile Diabetes Foundation, InHediger et al. (11) but differ from the pattern of expres- ternational. Address for reprint requests: N. 0. Davidson, Section of Gastroension noted in the rabbit small intestine (4). This latter terology, Dept. of Medicine, Univ. of Chicago, 5841 S. Maryland Ave., report suggested a twofold increase in SGLTl mRNA Box 400, Chicago, IL 60637. abundance between duodenum and jejunum, while the current studies demonstrate no such regional trend. Ad- Received 26 September 1991; accepted in final form 26 November 1991. ditionally, the demonstration that the distal small intesREFERENCES tine has comparable if not higher levels of SGLTl mRNA than the proximal adult small intestine is consistent with 1. Brot-Laroche, E., M. T. Dao, A. I. Alcalde, B. Delhomme, studies examining glucose uptake in isolated human N. Triadou, and F. Alvarado. Independent modulation by food supply of two distinct sodium-activated D-ghCOSe transport sysbrush-border membrane vesicles, which showed higher tems in the guinea pig-_ieiunal brush-border membrane. Proc. N&l. _ transport rates in distal small intestine (9). Acad. Sci. &‘A 85: 6370-6373,1988. The expression of SGLTl, GLUTl, GLUT2, and 2. Burant. C. F.. W. I. Sivitz. H. Fukumoto. T. Kavano. S. GLUT5 is developmentally regulated in the human small Nagamatsu, S: Seino, J. E. Bessin, and G. I. ‘Bell. Mammalian glucose transporters: structure and molecular regulation. Recent intestine. However, whereas the levels of GLUT1 mRNA Prog. Horn. Res. In press. decrease during fetal development, those of SGLTl, 3. Cheeseman, C. I., and B. Harley. Adaptation of glucose transGLUT2, and GLUT5 increase. The levels of sucraseport across rat enterocyte basolateral membrane in response to isomaltase mRNA which encodes a protein also involved altered dietary carbohydrate intake. J. Physiol. Lord. 437: 563575,199l. in luminal carbohydrate absorption also increase during 4. Coady, M. J., A. M. Pajor, and E. M. Wright. Sequence fetal development. Previous work has suggested that homologies among intestinal and renal Na+/glucose cotransporters. sucrase-isomaltase activity and mRNA abundance are Am. J. Physiol. 259 (Cell Physiol. 28): C605-C610, 1990. regulated coordinately in the human and rodent fetal 5. Fukumoto, H., T. Kayano, J. B. Buse, Y. Edwards, P. F. small intestine (17). Whether GLUT5 mRNA abundance Pilch, G. I. Bell, and S. Seino. Cloning and characterization of the major insulin-responsive glucose transporter expressed in huand protein accumulation demonstrate a similar pattern man skeletal muscle and other insulin-responsive tissues. J. Biol. of developmental induction is unknown. Chem. 264: 7776-7779,1989. The temporal coincidence of the changing patterns of 6. Fukumoto, H., S. Seino, H. Imura, Y. Seino, and G. I. Bell. expression of SGLTl, GLUTB, and sucrase-isomaltase Characterization and expression of human HepG2/erythrocyte gluJejunum
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
C800
GLUCOSE
TRANSPORTER
EXPRESSION
case-transporter gene. Diabetes 37: 657-661, 1988. 7. Fukumoto, H., S. Seino, H. Imura, Y. Seino, R. L. Eddy, Y. Fukushima, M. G. Byers, T. B. Shows, and G. I. Bell. Sequence, tissue distribution, and chromosomal localization of mRNA encoding a human glucose transporter-like protein. Proc. Natl. Acad. Sci. USA 85: 54345438,198s. 8. Green, F., Y. Edwards, H. P. Hauri, S. Povey, M. W. Ho, M. Pinto, and D. Swallow. Isolation of a cDNA probe for a human jejunal brush-border hydrolase, sucrase-isomaltase, and assignment of the gene locus to chromosome 3. Gene 57: 101-110, 1987. 9. Harig, J., J. A. Barry, V. M. Rajendran, K. H. Soergel, and K. Ramaswamy. D-Glucose and L-leucine transport by human intestinal brush-border membrane vesicles. Am. J. Physiol. 256 (Gustrointest. Liver Physiol. 19): G618-G623, 1989. 10. Hediger, M. A., T. Ikeda, M. Coady, C. B. Gundersen, and E. M. Wright. Expression of size-selected mRNA encoding the intestinal Na+/glucose cotransporter in Xenopus luevis oocytes. Proc. Natl. Acad. Sci. USA 84: 2634-2637, 1987. 11. Hediger, M. A., E. Turk, and E. M. Wright. Homology of the human intestinal Na’/glucose and Eschericia coli Na+/proline cotransporters. Proc. Natl. Acad. Sci. USA 86: 5748-5752, 1989. 12. Kayano, T., C. F. Burant, H. Fukumoto, G. W. Gould, Y. Fan, R. L. Eddy, M. G. Byers, T. B. Shows, S. Seino, and G. I. Bell. Human facilitative glucose transporters: isolation, functional characterization and gene localization of cDNAs encoding an isoform (GLUT5) expressed in small intestine, kidney, muscle and adipose tissue and an unusual glucose transporter pseudogene-like sequence (GLUTG). J. Biol. Chem. 265: 1327613282,199O. R. L. Eddy, Y.-S. Fan, M. G. 13. Kayano, T., H. Fukumoto, Byers, T. B. Shows, and G. I. Bell. Evidence for a family of
14.
15.
16.
17.
18.
19.
20.
21.
IN
HUMAN
INTESTINE
human glucose transporter-like proteins: sequence and gene localization of a protein expressed in fetal skeletal muscle and other tissues. J. Biol. Chem. 263: 15245-15248, 1988. Meddings, J. B., D. DeSouza, M. Goel, and S. Thiesen. Glucose transport and microvillus membrane physical properties along the crypt-villus axis of the rabbit. J. Clin. Invest. 85: 10991107,199o. Moxey, P. C., and J. S. Trier. Development of villus absorptive cells in the human fetal small intestine: a morphological and morphometric study. Anat. Rec. 195: 463-482, 1979. Rajendran, V. M., S. A. Ansari, J. M. Harig, M. B. Adams, A. H. Khan, and K. Ramaswamy. Transport of glycyl-L-proline by human intestinal brush border membrane vesicles. Gastroenterology 89: 1298-1304,1985. Sebastio, G., W. Hunziker, B. O’Neill, C. Malo, S. Auricchio, and G. Semenza. The biosynthesis of intestinal sucrase-isomaltase in human embryo is most likely controlled at the level of transcription. Biochem. Biophys. Res. Commun. 149: 830-839,1987. Teng, B., M. Verp, J. Salomon, and N. 0. Davidson. Apolipoprotein B messenger RNA editing is developmentally regulated and widely expressed in human tissues. J. Biol. Chem. 265: 2061620620,199O. Thorens, B., Z. Q. Cheng, D. Brown, and H. F. Lodish. Liver glucose transporter: a basolateral protein in hepatocytes and intestine and kidney cells. Am. J. Physio1259 (Cell Physiol. 28): C279C285,1990. Turk, E., B. Zabel, S. Mundolos, J. Dyer, and E. M. Wright. Glucose/galactose malabsorption caused by a defect in the Na’/ glucose cotransporter. Nature Lond. 350: 354-356, 1991. Werner, H., M. Adamo, W. L. Lowe, C. T. Roberts, and D. LeRoith. Developmental regulation of rat brain/HepG2 glucose transporter gene expression. Mol. Endocrinol. 3: 273-279, 1989.
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
expression
M. L. HAUSMAN, CARMEN CHARLES F. BURANT, AND
A. IFKOVITS, GRAEME I. BELL
Departments of Medicine, Biochemistry, and Molecular Biology and Howard Hughes Medical Institute, The University of Chicago, Chicago, Illinois 60637; and Molecular Pharmacology Group, Department of Biochemistry, The University of Glasgow, Glasgow G12 8&Q, United Kingdom Davidson, Nicholas O., Annalise M. L. Hausman, Carmen A. Ifkovits, John B. Buse, Gwyn W. Gould, Charles F. Burant, and Graeme I. Bell. Human intestinal glucose transporter expression and localization of GLUT5 Am. J. Physiol. 262 (Cell Physiol. 31): C795C800, 1992.-We have
studied the developmental and regional expressionof mRNAs encodingsodium-dependentand facilitative glucosetransporter proteins in human fetal and adult small intestine. The abundance of mRNAs encoding the Na’-glucose cotransporter isoform SGLTl and the facilitative glucosetransporter isoforms GLUT2 and GLUT5 is developmentally modulated with highest levels in adult small intestine. By contrast, the levels of GLUT1 mRNA are higher in fetal than adult small intestine. Immunohistochemicalanalysis of adult small intestine localized GLUT5 to the luminal surface of mature enterocytes, a finding confirmed by Western blot analysis of purified human jejunal brush-bordermembranes.By contrast, in the fetal small intestine, GLUT5 was localized along the intercellular junctions of the developingvillus, indicating that both its expression and localization are developmentally regulated. The localization of GLUT5 to the luminal surface of mature absorptive epithelial cells implies that this protein participates in the uptake of dietary sugars. brush border; plasma membrane; immunohistochemistry; Western blot
and transcellular delivery of dietary glucose is an important function of the small intestinal enterocyte. Adult mammalian small intestine expresses a number of genes involved in glucose transport including Na+dependent as well as facilitative glucose transporters (reviewed in Ref. 2). The intestinal brush-border Na+glucose cotransporter isoform SGLTl functions in the uptake of glucose across the microvillus membrane (4, 10, 11). The key role played by SGLTl in the active transport of glucose in the small intestine is evident from studies of individuals with glucose/galactose malabsorption syndrome (20). This recessive genetic disorder which is characterized by an inability to absorb dietary glucose and galactose from the small intestine can arise from mutations in SGLTl, as recently demonstrated (20). In addition to SGLTl, biochemical studies suggest that other proteins may also participate in the transport of glucose from the lumen of the small intestine into the enterocyte (1, 14). Glucose taken up by the enterocyte is released across the basolateral membrane into the interTHE UPTAKE
stitium via facilitative glucose transport. Recent studies have revealed that the facilitative glucose transporters comprise a family of structurally related proteins. cDNAs encoding five members of this family have been cloned and characterized (reviewed in Ref. 2). The five facilitative glucose transporters have been designated as GLUTl/erythrocyte type, GLUT2/ liver type, GLUT3/brain type, GLUT4/muscle-fat type, and GLUT5/smail intestine type, and all are expressed at varying levels in adult human small intestine (5-7,12, 13). However, only GLUT2 and GLUT5 represent relatively abundant mRNA species in the small intestine (7, 12), implying that these two isoforms may participate in the transepithelial movement of glucose. GLUT2 has been localized to the basolateral membrane of fully differentiated absorptive epithelial cells in the adult rat small intestine, consistent with it mediating the efflux of absorbed glucose from these cells (19). The localization of GLUT5 within the intestine has not been determined. In the present study, the regional and developmental regulation of glucose transporter mRNAs has been characterized in human fetal and adult small intestine. In addition, the localization of GLUT5 protein has been determined by immunohistochemical techniques and Western blotting. GLUT5 is confined to the brush border of mature enterocytes populating the upper half of the villus in adult small intestine, consistent with this protein acting in concert with SGLTl to promote the uptake of sugars from the lumen of the small intestine. MATERIALS
AND
METHODS
Tissue procurement, RNA extraction, and analysis. Human fetal intestine was obtained during the course of elective or spontaneousterminations of first and secondtrimester pregnancies as approved by the Institutional Review Board of the University of Chicago Hospitals. Adult tissue was obtained during diagnostic upper or lower endoscopicexamination or from organ donor referrals (National DiseaseResearchInterchange, Philadelphia, PA). Tissue wasobtained within 30 min of removal and was either snap frozen in liquid nitrogen or usedimmediately for RNA extraction. RNA was extracted into a solution of 4 M guanidinium thiocyanate and 0.3 M of 2-mercaptoethanolas described(18) with the following modifications. After the first precipitation with isopropanol, sampleswere extracted six times with ice-
0363-6143/92 $2.00 Copyright 0 1992 the American Physiological Society
c795
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
C796
GLUCOSE TRANSPORTER
EXPRESSION
cold 4 M LiCl and pelleted. After resuspension in freshly prepared 4 M guanidinium thiocyanate and precipitation with isopropanol, samples were suspended in 10 mM tris(hydroxymethyl)aminomethane (Tris) . HCl, pH 7.4, 1 mM EDTA, and 0.5% sodium dodecyl sulfate (SDS) and extracted with chloroform-isoamyl alcohol (24:l). The RNA was ethanol precipitated and then finally resuspended in ribonuclease (RNase)-free water and stored at -80°C. Northern blots were prepared using 20 rg of total RNA that was fractionated through 1% agarose-2% formaldehyde gels before capillary transfer onto nylon membranes (18). The following human facilitative glucose transporter cDNA clones were used as probes: GLUTl, phGT2-1 (6); GLUT2, pGEM4ZHTL3 (7); GLUT3, PBS-MGT3 (13); GLUT4, pGEM4ZAMT7 (5); and GLUT5, phJHT5 (12). The human SGLTl cDNA clone phNaGTB-2 was isolated from an adult human jejunum cDNA library and includes nucleotides 201-2,343 of the published cDNA sequence (11). The human sucrase-isomaltase cDNA clone pSl-22 was generously supplied by Dr. D. Swallow (8). The cDNA inserts from these clones were labeled by random priming. The hybridizations and washes were conducted as described (18), with a final wash in 15 mM NaCl-1.5 mM sodium citrate and 0.1% SDS at 65°C for 30 min. Antibody production. Antibodies against a peptide corresponding to the COOH-terminus of human GLUT5 (amino acids 488-501) were prepared by immunizing rabbits with a GLUT5-specific peptide (KEELKELPPVTSEQ) which had a cysteine added at the NHz-terminus to allow coupling to keyhole limpet hemocyanin (Sigma). Rabbits (New Zealand White) were immunized with 250 pg of conjugated peptide in a 40:60 mixture of phosphate-buffered saline (PBS)/Freund’s complete adjuvant at eight intradermal sites. One month later, animals were boosted with 250 pg of conjugated peptide in a 60:40 mixture of PBS/Freund’s incomplete adjuvant. Twenty milliliters of blood were collected 2 wk later, and anti-human GLUT5 antibodies were affinity purified over a peptide-coupled immunopurification column (Immunopure Ag/Ab, Pierce, Rockland, IL) as described by the manufacturer. Absorption of GLUT5 immunoreactivity was achieved by incubating antiserum with GLUT5 peptide (1 mg/ml) at 4°C for 12 h. Immunolocalization of GLUT5 in small intestine. Tissue samples were fixed in Bouin’s solution for 2 h, and 5-pm sections were prepared from paraffin-embedded blocks. Samples were blocked at 23°C for 20 min in blocking buffer (PBS containing 2% bovine serum albumin, 0.5% nonfat dry milk, 0.3% Triton X-100, and 0.05% sodium azide) containing 10% normal goat serum. Tissue sections were then incubated at 4°C overnight with a 1:50 dilution (in blocking buffer) with the anti-COOHterminus affinity-purified GLUT5 antibody. After five washes with PBS, samples were reblocked as above. They were then incubated for 1 h at 23°C with a 1:40 dilution (in blocking buffer) of gold-labeled goat anti-rabbit immunoglobulin G (Aurodye, Amersham, Arlington Heights, IL). After five washes in PBS and glass-distilled water, silver enhancement was carried out as suggested by the manufacturer. Sections were visualized under epifluoresence using the Olympus IGS cube and photographed with Kodak Ektachrome 400 film. Plasma membrane protein preparation and Western blotting. Human jejunal and both proximal and distal colonic brushborder membranes, prepared from mucosal scrapings as described (9, 16), were provided by Drs. T. Brasitus and J. Harig, Section of Gastroenterology, University of Chicago. Small intestinal brush-border membranes demonstrated at least 16-fold enrichment of sucrase activity in comparison with the starting homogenate. Protein (100 fig) from each preparation was separated by 10% polyacrylamide-SDS gel electrophoresis and transferred electophoretically to Immobilon membranes (Millipore, Bedford, MA). The membranes were blocked using a
IN HUMAN
INTESTINE
solution of 5% nonfat dry milk in TTBS (0.05% Tween 20, 10 mM Tris. HCl, pH 7.4, and 500 mM NaCl) and then incubated for 2 h at room temperature with a 1:500 dilution (in TTBS) of the rabbit polyclonal antibody to human GLUT5 described above. The membranes were washed three times in TTBS, and antibody binding was detected by incubation with iz51-protein A and autoradiography using Kodak XAR-5 film and an intensifying screen at -80°C for 16 h. RESULTS
Glucose transporter expression in human small intestine. SGLTl, GLUT2, and GLUT5 mRNAs are readily detected by Northern blotting in adult small intestine (Fig. 1). Multiple transcripts encode each of these glucose transporters as noted previously (7, 11, 12). These different transcripts are believed to be due to alternative A
SGLTl
FETAL 8 11.513.515 17 19’
ADULT ’ D J I I
- 2.2
kb
18SrRNA
B GLUT1
FETAL
t 15
ADULT
17 ’ D
J
ie I Co”D
J
I
. ..
ma
-3.0
kb
15SrRNA
C GLUT2
FETAL ’ 11 12.5
ADULT 15 ’ ’ J I ’ -5.4
kb
-3.4 -2.3
kb kb
FETAL ’ 11 12.5 15’
D GLUT5
ADULT ’ J I ’
(a) -5.1
kb
-2.3
kb
-2.0
kb
18SrRNA
Fig. 1. Expressionof glucose transportermRNAsin humanfetal and adult small intestine. Total RNA was extracted from various gestational age (weeks) fetal and adult small intestines. Northern blots were preparedusing20pg total RNA. Membranes were hybridizedsequentially with the cDNA probes indicated. For GLUT5 (D), exposure was extended from 2 (a) to 14 days (b) to show the low levels of fetal mRNA. For eachset of blots, hybridizationto 18s rRNA confirmed equivalence of loading and transfer. D, duodenum; J, jejunum; I, ileum;
and Co, colon. Adult intestinal RNA was obtainedfrom the same individual for the regions shown.
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
GLUCOSE
TRANSPORTER
A
EXPRESSION
IN HUMAN
B J
I
Co
-6.0
*
lsp
-
‘%I D ADULT
c797
C D
170
INTESTINE
kb
*
J ,am
A’
Fig. 2. Developmental and regional expression of sucrase-isomaltase mRNA in human intestine. A: total small intestinal RNA (5,2.5, and 1 pg) from fetal (gestational age indicated in weeks) and adult (D, duodenum; J, jejunum; I, ileum) intestine was immobilized directly onto nylon membranes using a slot-blot format. Where sample permitted, aliquots of both proximal (P) and distal (D) fetal small intestine were analyzed. Blot was hybridized with a 32P-labeled human sucrase-isomaltase cDNA probe. B: 20 pg of total RNA from different regions of the same 17-wk human fetal intestine (D, duodenum; J, jejunum; I, ileum; Co, colon) were fractionated on a 1% agarose-formaldehyde gel, blotted, and hybridized with a human sucrase-jsomaltase cDNA probe. Probe specifically hybridizes to an mRNA of 6 kb, as noted previously (8). C: slot-blot shown in A was stripped and rehybridized with a cDNA to human B-actin.
polyadenylation generating transcripts having 3’-untranslated regions of different sizes (12). In fetal small intestine, SGLTl mRNA could be detected by Northern blotting at 17 wk, and by 19 wk, the levels were comparable to those found in the adult (Fig. 1A). As shown in Fig. lA, SGLTl mRNA abundance appears higher in adult ileum than in duodenum or jejunum in tissue samples obtained from the same individual. However, in a second individual, there were comparable levels of SGLTl mRNA in ileum and jejunum (data not shown). Both individuals were male, aged 37 and 32 yr, and were otherwise healthy. Based on this limited sample, the data suggest that there may be interindividual variation in levels of SGLTl mRNA in different regions of the small intestine, the molecular basis for which is unknown. Low levels of GLUT2 mRNA could be detected by Northern blotting at 11 wk of gestation (Fig. lC), the earliest time studied. The levels then increase with gestational age and are highest in adult small intestine. The levels of GLUT2 mRNA in adult jejunum and ileum appear to be similar. By contrast, the levels of GLUT5 mRNA are much lower in fetal small intestine (and only evident with long autoradiographic exposure, Fig. 1D) than those seen in the adult. The levels of GLUT5 mRNA in jejunal and ileal RNA preparations from the same individual were similar. Thus, while not discounting the potential importance of individual variability, mRNAs encoding all the abundant glucose transporter genes appear to be expressed throughout the adult small intestine with no evidence of a proximal to distal gradient of gene expression. We have previously demonstrated low levels of GLUTl, GLUT3, and GLUT4 mRNA in adult human small intestine (5,6,13). The present study reveals levels of GLUT1 mRNA to be higher in fetal than in adult
small intestine with expression additionally demonstrated in fetal colon (Fig. 1B). The decrease observed in GLUT1 mRNA abundance in fetal as compared with adult small intestine has been previously noted for rat liver, heart, lung, and muscle (21). Under the experimental conditions of this study, GLUT3 mRNA could not be detected by Northern blotting, implying that the levels of GLUT3 mRNA are much lower than those of the other isoforms. Additionally, GLUT4 mRNA, which is believed to originate from smooth muscle cells and possibly adherent adipocytes (7), was only barely detectable after prolonged autoradiographic exposure (data not shown). Sucrase-isomaltase expression in human small intestine. For purposes of comparison, the abundance of
mRNA encoding a member of another class of proteins involved in luminal carbohydrate absorption was also examined. Sucrase-isomaltase mRNA was first detectable in 13-wk fetal intestine, and its abundance increased in the proximal (but not distal) intestine at 15 and 17 wks of gestation (Fig. 2A). For comparison purposes, the blot was rehybridized with a @-actin cDNA to demonstrate equivalent RNA loading (Fig. 2C). By 17 wk, sucrase-isomaltase mRNA was detectable in all regions of the fetal intestine including colon (Fig. 2B). Adult intestine demonstrated roughly equal abundance of suerase mRNA in all regions of the small intestine, but no signal was detectable in the adult colon (data not shown). The changing pattern of expression of sucrase-isomaltase mRNA in the small intestine thus appears to parallel that seen for SGLTl and GLUT2. Immunolocalization of GLUT5 in human small intestine. The distribution of GLUT5 in fetal and adult small
intestine was studied in tissue sections using antibodies directed against the COOH-terminus of this protein. Fetal small intestine showed immunoreactivity princi-
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
C798
GLUCOSE
TRANSPORTER
EXPRESSION
IN
HUMAN
INTESTINE
Fig. 3. Immunogold epifluorescence localization of human GLUT5 A-D: fetal small intestine (12 wk). E-H: adult duodenum. Z and J, adult colon. Samples were incubated with affinity-purified rabbit anti-GLUT5 antibodies (A, E, and I) or preimmune rabbit immunoglobulin G (C and G) followed by gold-conjugated goat anti-rabbit IgG as described in MATERIALS AND METHODS. Note staining pattern between cells of the fetal intestine (arrows, A). Specific GLUT5 staining in the adult duodenal sample (E) was confined to the brush border of the top third of the villus with diminished intensity noted in the middle to lower region (arrow). Phase-contrast pictures of the respective sections are also shown in B, D, F, H, and J.
pally between cells of the immature villus (Fig. 3A, arrows), while immunoreactivity in the adult small intestine was predominantly along the brush border of cells lining the upper third of the villus (Fig. 3E). The immunoreactive material in the lamina propria was also seen with the preimmune serum (Fig. 3G) and presumably reflects nonspecific binding of the antibody to structures in this region of the villus. No antibody staining was detected in the colon (Fig. 31), which is consistent with Northern blotting data showing the absence of GLUT5 mRNA in this region of the human intestine. The localization of GLUT5 to the luminal surface of the absorptive epithelial cells was also examined by Western blotting of brush-border membranes isolated from human small intestine and colon. The GLUT5 antibody specifically reacted with a jejunal brush-border
protein of -50 kDa (Fig. 4), similar to the predicted size of human GLUT5 (M, = 54,983; Ref. 12). The absence of a signal in either homogenate or a crude microsomal fraction (containing a mixture of organellar, brush-border, and plasma membranes) is consistent with the polarized expression of GLUT5 in jejunal brush-border membranes. There was also no signal detected using either proximal or distal colonic brush-border membranes, findings consistent with the distribution of GLUT5 determined by both the Northern blotting and immunohistochemical studies. DISCUSSION
The uptake and release of glucose by epithelial the small intestine is carried out by members
cells of of two
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
GLUCOSE
Absorbed
aGLUT
TRANSPORTER
EXPRESSION
Non-Absorbed
aGLUT
Jejunum
Colon
IN HUMAN
INTESTINE
c799
mRNAs suggests that they may be coordinately regulated in preparation for the changes in nutrition that occur I postnatally. II Such adaptive changes in glucose trans‘H M BBM Prox Dist ’ H M BBM’ porter activity have been inferred from biochemical studkDa ies in rat enterocytes, providing some indirect support 106for this hypothesis (3). The changing pattern of expres60sion of these genes occurs in the setting of morphological differentiation of the human intestine which occurs during the second trimester. During the period from 10 to 49.5 20 wk gestation, the villus-crypt axis undergoes organization and discrete cell-types emerge, each with a distinct morphology (15). Demonstration of GLUT5 immuno32.5 reactivity in the intercellular regions of 12-wk fetal small intestine may reflect the relative immaturity of the 27.5brush-border membrane at this stage of gestation. The roles of SGLTl and GLUT2 in transepithelial transport of glucose seem firmly established. However, the function of GLUT5 in this process is still a matter Fig. 4. Western blot analysis of GLUT5 distribution in human jeof conjecture. The present data suggest that GLUT5 junum and proximal and distal colon. Protein (100 pg) from each mRNA and protein are relatively abundant in the small preparation of homogenate (H), total microsomal (M), or brush-border membranes (BBM) was separated by 10% polyacrylamide-sodium do- intestine, and the localization of GLUT5 to the luminal decyl sulfate electrophoresis and transferred to an Immobilon memmembrane of absorptive epithelial cells is consistent with brane. Half of the membrane was incubated with absorbed GLUT5 a role for this protein in the transport of sugars from the antibodies, and the other half was incubated with unabsorbed GLUT5 antibodies. Antibody binding was detected by incubation with iz51- lumen of the small intestine. It is tempting to speculate protein A followed by autoradiography. that GLUT5 may have a substrate specificity different from that of the other facilitative glucose transporters, different families of glucose transporter proteins, the and this possibility is currently under investigation. In SGLT family of Na+-glucose cotransporters and the this context, studies are underway to define the biochemGLUT fam.ily of facilitative glucose transporters. In the ical properties of GLUT5 in more detail to understand small intestine, SGLTl participates in the active uptake how it participates in sugar uptake by the small intestine. and accumulation of glucose across the luminal membrane of absorptive epithelial cells, and GLUT2 mediates We thank Drs. T. Brasitus and J. Harig, Section of Gastroenterology, its release across the basolateral membrane of these cells University of Chicago, for providing human small intestinal and colonic brush-border membranes for localization of GLUT5 into the interstitium. The pattern and relative abunThese studies were supported by National Institutes of Health dance of SGLTl mRNA in the human intestine pre- Grants DK-42086, DK-20595, HL-02166, and HL-38180, the Howard sented in this report are identical to those reported by Hughes Medical Institute, and the Juvenile Diabetes Foundation, InHediger et al. (11) but differ from the pattern of expres- ternational. Address for reprint requests: N. 0. Davidson, Section of Gastroension noted in the rabbit small intestine (4). This latter terology, Dept. of Medicine, Univ. of Chicago, 5841 S. Maryland Ave., report suggested a twofold increase in SGLTl mRNA Box 400, Chicago, IL 60637. abundance between duodenum and jejunum, while the current studies demonstrate no such regional trend. Ad- Received 26 September 1991; accepted in final form 26 November 1991. ditionally, the demonstration that the distal small intesREFERENCES tine has comparable if not higher levels of SGLTl mRNA than the proximal adult small intestine is consistent with 1. Brot-Laroche, E., M. T. Dao, A. I. Alcalde, B. Delhomme, studies examining glucose uptake in isolated human N. Triadou, and F. Alvarado. Independent modulation by food supply of two distinct sodium-activated D-ghCOSe transport sysbrush-border membrane vesicles, which showed higher tems in the guinea pig-_ieiunal brush-border membrane. Proc. N&l. _ transport rates in distal small intestine (9). Acad. Sci. &‘A 85: 6370-6373,1988. The expression of SGLTl, GLUTl, GLUT2, and 2. Burant. C. F.. W. I. Sivitz. H. Fukumoto. T. Kavano. S. GLUT5 is developmentally regulated in the human small Nagamatsu, S: Seino, J. E. Bessin, and G. I. ‘Bell. Mammalian glucose transporters: structure and molecular regulation. Recent intestine. However, whereas the levels of GLUT1 mRNA Prog. Horn. Res. In press. decrease during fetal development, those of SGLTl, 3. Cheeseman, C. I., and B. Harley. Adaptation of glucose transGLUT2, and GLUT5 increase. The levels of sucraseport across rat enterocyte basolateral membrane in response to isomaltase mRNA which encodes a protein also involved altered dietary carbohydrate intake. J. Physiol. Lord. 437: 563575,199l. in luminal carbohydrate absorption also increase during 4. Coady, M. J., A. M. Pajor, and E. M. Wright. Sequence fetal development. Previous work has suggested that homologies among intestinal and renal Na+/glucose cotransporters. sucrase-isomaltase activity and mRNA abundance are Am. J. Physiol. 259 (Cell Physiol. 28): C605-C610, 1990. regulated coordinately in the human and rodent fetal 5. Fukumoto, H., T. Kayano, J. B. Buse, Y. Edwards, P. F. small intestine (17). Whether GLUT5 mRNA abundance Pilch, G. I. Bell, and S. Seino. Cloning and characterization of the major insulin-responsive glucose transporter expressed in huand protein accumulation demonstrate a similar pattern man skeletal muscle and other insulin-responsive tissues. J. Biol. of developmental induction is unknown. Chem. 264: 7776-7779,1989. The temporal coincidence of the changing patterns of 6. Fukumoto, H., S. Seino, H. Imura, Y. Seino, and G. I. Bell. expression of SGLTl, GLUTB, and sucrase-isomaltase Characterization and expression of human HepG2/erythrocyte gluJejunum
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
C800
GLUCOSE
TRANSPORTER
EXPRESSION
case-transporter gene. Diabetes 37: 657-661, 1988. 7. Fukumoto, H., S. Seino, H. Imura, Y. Seino, R. L. Eddy, Y. Fukushima, M. G. Byers, T. B. Shows, and G. I. Bell. Sequence, tissue distribution, and chromosomal localization of mRNA encoding a human glucose transporter-like protein. Proc. Natl. Acad. Sci. USA 85: 54345438,198s. 8. Green, F., Y. Edwards, H. P. Hauri, S. Povey, M. W. Ho, M. Pinto, and D. Swallow. Isolation of a cDNA probe for a human jejunal brush-border hydrolase, sucrase-isomaltase, and assignment of the gene locus to chromosome 3. Gene 57: 101-110, 1987. 9. Harig, J., J. A. Barry, V. M. Rajendran, K. H. Soergel, and K. Ramaswamy. D-Glucose and L-leucine transport by human intestinal brush-border membrane vesicles. Am. J. Physiol. 256 (Gustrointest. Liver Physiol. 19): G618-G623, 1989. 10. Hediger, M. A., T. Ikeda, M. Coady, C. B. Gundersen, and E. M. Wright. Expression of size-selected mRNA encoding the intestinal Na+/glucose cotransporter in Xenopus luevis oocytes. Proc. Natl. Acad. Sci. USA 84: 2634-2637, 1987. 11. Hediger, M. A., E. Turk, and E. M. Wright. Homology of the human intestinal Na’/glucose and Eschericia coli Na+/proline cotransporters. Proc. Natl. Acad. Sci. USA 86: 5748-5752, 1989. 12. Kayano, T., C. F. Burant, H. Fukumoto, G. W. Gould, Y. Fan, R. L. Eddy, M. G. Byers, T. B. Shows, S. Seino, and G. I. Bell. Human facilitative glucose transporters: isolation, functional characterization and gene localization of cDNAs encoding an isoform (GLUT5) expressed in small intestine, kidney, muscle and adipose tissue and an unusual glucose transporter pseudogene-like sequence (GLUTG). J. Biol. Chem. 265: 1327613282,199O. R. L. Eddy, Y.-S. Fan, M. G. 13. Kayano, T., H. Fukumoto, Byers, T. B. Shows, and G. I. Bell. Evidence for a family of
14.
15.
16.
17.
18.
19.
20.
21.
IN
HUMAN
INTESTINE
human glucose transporter-like proteins: sequence and gene localization of a protein expressed in fetal skeletal muscle and other tissues. J. Biol. Chem. 263: 15245-15248, 1988. Meddings, J. B., D. DeSouza, M. Goel, and S. Thiesen. Glucose transport and microvillus membrane physical properties along the crypt-villus axis of the rabbit. J. Clin. Invest. 85: 10991107,199o. Moxey, P. C., and J. S. Trier. Development of villus absorptive cells in the human fetal small intestine: a morphological and morphometric study. Anat. Rec. 195: 463-482, 1979. Rajendran, V. M., S. A. Ansari, J. M. Harig, M. B. Adams, A. H. Khan, and K. Ramaswamy. Transport of glycyl-L-proline by human intestinal brush border membrane vesicles. Gastroenterology 89: 1298-1304,1985. Sebastio, G., W. Hunziker, B. O’Neill, C. Malo, S. Auricchio, and G. Semenza. The biosynthesis of intestinal sucrase-isomaltase in human embryo is most likely controlled at the level of transcription. Biochem. Biophys. Res. Commun. 149: 830-839,1987. Teng, B., M. Verp, J. Salomon, and N. 0. Davidson. Apolipoprotein B messenger RNA editing is developmentally regulated and widely expressed in human tissues. J. Biol. Chem. 265: 2061620620,199O. Thorens, B., Z. Q. Cheng, D. Brown, and H. F. Lodish. Liver glucose transporter: a basolateral protein in hepatocytes and intestine and kidney cells. Am. J. Physio1259 (Cell Physiol. 28): C279C285,1990. Turk, E., B. Zabel, S. Mundolos, J. Dyer, and E. M. Wright. Glucose/galactose malabsorption caused by a defect in the Na’/ glucose cotransporter. Nature Lond. 350: 354-356, 1991. Werner, H., M. Adamo, W. L. Lowe, C. T. Roberts, and D. LeRoith. Developmental regulation of rat brain/HepG2 glucose transporter gene expression. Mol. Endocrinol. 3: 273-279, 1989.
Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 9, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
