Differential subcellular distribution of glucose transporters GLUT1-6 and GLUT9 in human cancer: ultrastructural localization of GLUT1 and GLUT5 in breast tumor tissues

It has been proposed that the enhanced metabolic activity of tumor cells is accompanied by an increased expression of facilitative hexose transporters (GLUTs). However, a previous immunohistochemical analysis of GLUT1 expression in 154 malignant human neoplasms failed to detect the GLUT1 isoform in 87 tumors. We used 146 normal human tissues and 215 tumor samples to reassess GLUT1 expression. A similar number of samples were used to compare the expression of GLUT2-6 and 9. The classical expression of GLUT1-5 in different normal human tissues was confirmed, however, we were unable to detect GLUT2 in human pancreatic islet cells. GLUT6 was principally detected in testis germinal cells and GLUT9 was localized in kidney, liver, heart, and adrenal. In tumor samples, GLUT1, 2, and 5 were the main transporters detected. GLUT1 was the most widely expressed transporter, however, 42% of the samples had very low-to-negative expression levels. GLUT2 was detected in 31% of the samples, being mainly expressed in breast, colon, and liver carcinoma. GLUT5 was detected in 27% of breast and colon adenocarcinoma, liver carcinoma, lymphomas, and testis seminoma samples. In situ RT-PCR and ultrastructural immunohistochemistry confirmed GLUT5 expression in breast cancer. GLUT6 and 9 are not clearly over-expressed in human cancer. The extensive expression of GLUT2 and 5 (glucose/fructose and fructose transporters, respectively) in malignant human tissues indicates that fructose may be a good energy substrate in tumor cells. Our functional data obtained in vitro in different tumor cells support this hypothesis. Additionally, these results suggest that fructose uptake could be used for positron emission tomography imaging and, may possibly represent a novel target for the development of therapeutic agents in different human cancers.     

A preformed basal lamina alters the metabolism and distribution of hyaluronan in epidermal keratinocyte "organotypic" cultures grown on collagen matrices

A rat epidermal keratinocyte (REK) line which exhibits histodifferentiation nearly identical to the native epidermis when cultured at an air-liquid interface was used to study the metabolism of hyaluronan, the major intercellular macromolecule present in basal and spinous cell layers. Two different support matrices were used: reconstituted collagen fibrils with and without a covering basal lamina previously deposited by canine kidney cells. REKs formed a stratified squamous, keratinized epithelium on both support matrices. Hyaluronan and its receptor, CD44, colocalized in the basal and spinous layers similar to their distribution in the native epidermis. Most (approximately 75%) of the hyaluronan was retained in the epithelium when a basal lamina was present while most (approximately 80%) diffused out of the epithelium in its absence. While REKs on the two matrices synthesized hyaluronan at essentially the same rate, catabolism of this macromolecule was much higher in the epithelium on the basal lamina (half-life approximately 1 day, similar to its half-life in native human epidermis). The formation of a true epidermal compartment in culture bounded by the cornified layer on the surface and the basal lamina subjacent to the basal cells provides a good model within which to study epidermal metabolism.     

Heterogeneity of keratin distribution in the oral mucosa and skin of mammals as determined using monoclonal antibodies

The immunohistochemical localization of keratins in the oral epithelia of several mammals was investigated using the monoclonal antibodies to keratins, PKK1 (41-56 kilodaltons) and KL1 (55-57 kilodaltons). The staining patterns obtained in different locations of the oral mucosa and of the skin epidermis were compared. In the papillae on the dorsal surface of the tongue, some areas exhibited marked PKK1 staining, while other area were PKK1 negative. In general, rodent oral epithelia were negative for PKK1 in the basal layer, while comparatively strong PKK1 staining was observed in cells of the upper spinous layer. In the epidermis, positive PKK1 reactions were confined to the basal layer, while KL1 staining was occasionally seen in the basal layer of oral epithelia. In cats, dogs, and monkeys, different PKK1 and KL1 binding patterns were observed in oral epithelia. Also, the distribution in oral epithelia differed from that seen in the epidermis of these animals. In the epidermis, the distribution of PKK1 and KL1 was regular, with PKK1 usually being confined to the basal layer, while KL1 binding was found in the spinous and granular cell layers, and was dependent on the degree of keratinization. In the animals studies, keratin expression--as detected by PKK1 and KL1--was different in the skin epidermis and oral epithelia, and the localization of these keratins differed in the various types of oral mucosa.     

Pep-1 as a novel probe for the in situ detection of hyaluronan.

Hyaluronan (HA) is expressed by most tissues, including skin. Localization of HA in the skin is assessed by histology with HA-binding protein (HABP) serving as the probe. Reports have suggested that HA expression in skin is altered in a number of diseases. However, interlaboratory variations in HABP staining profiles, even in normal skin, suggest a need to standardize methods and/or identify new probes. We report the staining patterns of a HA-binding peptide (termed "Pep-1") in human and mouse skin. After acetone fixation, Pep-1 stained HA in the intercellular spaces of the epidermis, whereas staining in the dermis was weak and diffuse in both human and mouse skin. HABP staining of the epidermis and dermis were comparable in human skin but failed to stain the vital epidermis of mouse skin. In human skin, Pep-1 stained the basal, spinous, and granular layers, whereas HABP failed to stain the basal layer. Precipitation of HA in situ resulted in dermal staining but weak staining in the epidermis for HABP and Pep-1. Our results may suggest that Pep-1 is sensitive to HA conformation. Furthermore, Pep-1 may represent a new probe to study HA expression in the skin.     

Rab4b Is a Small GTPase Involved in the Control of the Glucose Transporter GLUT4 Localization in Adipocyte

Background

Endosomal small GTPases of the Rab family, among them Rab4a, play an essential role in the control of the glucose transporter GLUT4 trafficking, which is essential for insulin-mediated glucose uptake. We found that adipocytes also expressed Rab4b and we observed a consistent decrease in the expression of Rab4b mRNA in human and mice adipose tissue in obese diabetic states. These results led us to study this poorly characterized Rab member and its potential role in glucose transport.

Methodology/Principal Findings

We used 3T3-L1 adipocytes to study by imaging approaches the localization of Rab4b and to determine the consequence of its down regulation on glucose uptake and endogenous GLUT4 location. We found that Rab4b was localized in endosomal structures in preadipocytes whereas in adipocytes it was localized in GLUT4 and in VAMP2-positive compartments, and also in endosomal compartments containing the transferrin receptor (TfR). When Rab4b expression was decreased with specific siRNAs by two fold, an extent similar to its decrease in obese diabetic subjects, we observed a small increase (25%) in basal deoxyglucose uptake and a more sustained increase (40%) in presence of submaximal and maximal insulin concentrations. This increase occurred without any change in GLUT4 and GLUT1 expression levels and in the insulin signaling pathways. Concomitantly, GLUT4 but not TfR amounts were increased at the plasma membrane of basal and insulin-stimulated adipocytes. GLUT4 seemed to be targeted towards its non-endosomal sequestration compartment.

Conclusion/Significance

Taken our results together, we conclude that Rab4b is a new important player in the control of GLUT4 trafficking in adipocytes and speculate that difference in its expression in obese diabetic states could act as a compensatory effect to minimize the glucose transport defect in their adipocytes.     

The effect of age on glucose uptake and GLUT1 and GLUT4 expression in rat skeletal muscle

During the life span, phenotypic and structural modifications on skeletal muscle contribute to a reduction on glucose uptake either in basal state or triggered by insulin, but the underlying mechanisms for this decline are not entirely identified. A reduction in the expression of skeletal muscle glucose transporters (GLUTs), glucose transporter type 1 (GLUT1) and glucose transporter type 4 (GLUT4), has been associated to such phenomena, but unlike the case of insulin, only few studies have addressed the effect of age on muscle-contraction-induced glucose uptake. The aim of the study was to investigate the influence of age on GLUT1 and GLUT4 expression in skeletal muscle and its relation to the glucose uptake induced by muscle contraction. For this purpose, soleus muscle from Wistar rats aged 4, 10, 22 and 42 weeks were isolated and electrically stimulated (30 min, 10 Hz, 20 V, 0.2 ms). After stimulation, glucose uptake and GLUT1 and GLUT4 expression and localisation were evaluated. Muscle contraction caused an increase in glucose uptake in all studied groups. In addition, the absolute rates of glucose uptake were negatively correlated with age. The expression of GLUT4 was lower in older animals, whereas no relation between age and GLUT1 expression was found. Immunohistochemistry confirmed the ontogenic effect on GLUT4 expression and suggested an age-related modification on GLUT1 distribution within the muscle fibres; for instance, this protein seems to be present mainly out of the sarcoplasm. The present findings demonstrate that the ability of muscle contraction to increase glucose uptake is not influenced by age, whereas glucose uptake under basal conditions decreases with age.     

Heterogeneity of keratin distribution in the oral mucosa and skin of mammals as determined using monoclonal antibodies

Summary The immunohistochemical localization of keratins in the oral epithelia of several mammals was investigated using the monoclonal antibodies to keratins, PKK1 (41–56 kilodaltons) and KL1 (55–57 kilodaltons). The staining patterns obtained in different locations of the oral mucosa and of the skin epidermis were compared. In the papillae on the dorsal surface of the tongue, some areas exhibited marked PKK1 staining, while other area were PKK1 negative. In general, rodent oral epithelia were negative for PKK1 in the basal layer, while comparatively strong PKK1 staining was observed in cells of the upper spinous layer. In the epidermis, positive PKK1 reactions were confined to the basal layer, while KL1 staining was occasionally seen in the basal layer of oral epithelia. In cats, dogs, and monkeys, different PKK1 and KL1 binding patterns were observed in oral epithelia. Also, the distribution in oral epithelia differed from that seen in the epidermis of these animals. In the epidermis, the distribution of PKK1 and KL1 was regular, with PKK1 usually being confined to the basal layer, while KL1 binding was found in the spinous and granular cell layers, and was dependent on the degree of keratinization. In the animals studies, keratin expression as detected by PKK1 and KL1-was different in the skin epidermis and oral epithelia, and the localization of these keratins differed in the various types of oral mucosa.     

Glucose and cationic amino acid transporter expression in growing chickens (Gallus gallus domesticus)

Tissue glucose transporter (GLUT1-3) and cationic amino acid transporter (CAT1-3) mRNA expression was determined in growing broiler chicks posthatch. In two experiments, tissues were either collected on days 1, 3 and 7 or days 1 and 14 posthatch. Heart and liver were the only tissues expressing a GLUT isoform on day 1. All tissues expressed a GLUT isoform on day 7 except for the thymus. Most tissues expressing a CAT isoform on day 1 decreased mRNA levels through day 7 (P<0.05), except for bursa CAT-1 which tended to increase (P=0.05). The thymus and spleen did not express any CAT isoform mRNA until day 7. The liver was the only tissue expressing GLUT-2 mRNA through day 14. On day 14, GLUT-1, CAT-1 and CAT-2 mRNA were differentially expressed across tissues (P<0.05). High-affinity GLUT and CAT mRNA expression was highest in the heart and bursa, respectively (P<0.05). Total CAT mRNA expression was greatest in the bursa (P<0.05). The thymus had the lowest high affinity GLUT and total CAT mRNA expression on day 14 posthatch. Therefore, T lymphocytes within the thymus may be most susceptible to glucose and cationic amino acid supply.     

The distribution of blood group antigens in rodent epithelia

Summary The pattern of distribution of antigens cross-reacting with antibodies to human blood group antigens A and B and two precursor molecules was examined by immunofluorescence in the epidermis, oral mucosa and forestomach of rats and mice. Staining for blood group antigen A was negative. In all epithelia examined, blood group antigen B was present at the surface of basal and parabasal cells, and the H antigen at the surface of spinous cells. N-acetyllactosamine was present on the cell membranes in the upper spinous and granular cell layers of epidermis and forestomach epithelium and was not expressed in the oral epithelia except for a limited area in the dorsal tongue epithelium.Thus, the expression of antigen varies both regionally and, as earlier shown in human epithelium, with the stage of maturation of cells within a given epithelium. The observed sequence of expression of these antigens during maturation differs from that of human epithelia, but the present study provides a basis for further experimental studies of the role of cell surface antigens in epithelial homeostasis and maturation.     

Insulin regulates GLUT1-mediated glucose transport in MG-63 human osteosarcoma cells

Osteosarcoma is the most common type of malignant bone cancer, accounting for 35% of primary bone malignancies. Because cancer cells utilize glucose as their primary energy substrate, the expression and regulation of glucose transporters (GLUT) may be important in tumor development and progression. GLUT expression has not been studied previously in human osteosarcoma cell lines. Furthermore, although insulin and insulin-like growth factor (IGF-I) play an important role in cell proliferation and tumor progression, the role of these hormones on GLUT expression and glucose uptake, and their possible relation to osteosarcoma, have also not been studied. We determined the effect of insulin and IGF-I on GLUT expression and glucose transport in three well-characterized human osteosarcoma cell lines (MG-63, SaOs-2, and U2-Os) using immunocytochemical, RT-PCR and functional kinetic analyses. Furthermore we also studied GLUT isoform expression in osteosarcoma primary tumors and metastases by in situ hybridization and immunohistochemical analyses. RT-PCR and immunostaining show that GLUT1 is the main isoform expressed in the cell lines and tissues studied, respectively. Immunocytochemical analysis shows that although insulin does not affect levels of GLUT1 expression it does induce a translocation of the transporter to the plasma membrane. This translocation is associated with increased transport of glucose into the cell. GLUT1 is the main glucose transporter expressed in osteosarcoma, furthermore, this transporter is regulated by insulin in human MG-63 cells. One possible mechanism through which insulin is involved in cancer progression is by increasing the amount of glucose available to the cancer cell.     

Role of LIM kinases in normal and psoriatic human epidermis

We present evidence that LIM kinases can control cell adhesion and compaction in human epidermis. LIMK2 is expressed in the epidermal basal layer and signals downstream of the GTPase Rac1 to promote extracellular matrix adhesion and inhibit terminal differentiation. Conversely, LIMK1 is expressed in the upper granular layers and phosphorylates and inhibits cofilin. Expression of LIMK1 is lost in psoriatic lesions and other skin disorders characterized by lack of cell compaction in the differentiating cell layers. In psoriatic lesions down-regulation of LIMK1 correlates with up-regulation of Myc. Expression of constitutively active cofilin or Myc in reconstituted human epidermis blocks cell compaction. Overexpression of LIMK1 leads to down-regulation of Myc, whereas inhibition of Rho kinase, an upstream activator of LIMK1, stimulates Myc expression. Inhibition of Myc by LIMK1 is via inhibition of Stat3 phosphorylation, because constitutively active cofilin or inhibition of Rho kinase results in Stat3 phosphorylation and increased Myc levels, whereas dominant negative Stat3 abolishes the effect. In conclusion, we have uncovered a novel antagonistic relationship between the LIMK1/phosphocofilin and Myc/Stat3 pathways in the differentiating layers of human epidermis and propose that down-regulation of LIMK1 contributes to one of the pathological features of psoriatic epidermal lesions.     

Expression patterns of loricrin in various species and tissues

Abstract. In this study we analyzed the expression patterns of loricrin in various species and tissues using immunohistochemistry, immunoblotting and Northern blots. Loricrin is a glycine-, serine- and cysteine-rich protein expressed very late in epidermal differentiation in the granular layers of normal mouse and human epidermis. Later on in differentiation, loricrin becomes cross-linked as a major component into the cornified cell envelope by the formation of N^ɛ -(γ-glutamyl)lysine isopeptide bonds. This process either occurs directly or by the intermediate accumulation in L-keratohyaline granules of mouse epidermis and human acrosyringia. Loricrin was identified in all mammalian species analyzed by virtue of its highly conserved carboxy-terminal sequences revealing an electric mobility of ∼60 kDa in rodents, rabbit and cow and of ∼35 kDa in lamb and human on sodium dodecyl sulfate polyacrylamide gel electrophoresis. Loricrin is expressed in the granular layer of all mammalian orthokeratinizing epithelia tested including oral, esophageal and fore-stomach mucosa of rodents, tracheal squamous metaplasia of vitamin A deficient hamster and estrogen induced squamous vaginal epithelium of ovary ectomized rats. Loricrin is also expressed in a few parakeratinizing epithelia such as BBN [N-butyl-N-(4–hydroxybutyl)nitrosamine]-induced murine bladder carcinoma and a restricted subset of oral and single vaginal epithelial cells in higher mammals. Our results provide further evidence that the program of squamous differentiation in internal epithelia of the upper alimentary tract in rodents and higher mammals differ remarkably. In addition, we also have noted the distinct distribution patterns of human loricrin and involucrin, another major precursor protein of the cornified cell envelope.     

Overexpression of GLUT4 in mice causes up-regulation of UCP3 mRNA in skeletal muscle

Mitochondrial uncoupling protein 3 (UCP3) is expressed in skeletal muscles. We have hypothesized that increased glucose flux in skeletal muscles may lead to increased UCP3 expression. Male transgenic mice harboring insulin-responsive glucose transporter (GLUT4) minigenes with differing lengths of 5'-flanking sequence (-3237, -2000, -1000 and -442 bp) express different levels of GLUT4 protein in various skeletal muscles. Expression of the GLUT4 transgenes caused an increase in UCP3 mRNA that paralleled the increase of GLUT4 protein in gastrocnemius muscle. The effects of increased intracellular GLUT4 level on the expression of UCP1, UCP2 and UCP3 were compared in several tissues of male 4 month-old mice harboring the -1000 GLUT4 minigene transgene. In the -1000 GLUT4 transgenic mice, expression of GLUT4 mRNA and protein in skeletal muscles, brown adipose tissue (BAT), and white adipose tissue (WAT) was increased by 1.4 to 4.0-fold. Compared with non-transgenic littermates, the -1000 GLUT4 mice exhibited about 4- and 1.8-fold increases of UCP3 mRNA in skeletal muscle and WAT, respectively, and a 38% decrease of UCP1 mRNA in BAT. The transgenic mice had a 16% increase in oxygen consumption and a 14% decrease in blood glucose and a 68% increase in blood lactate, but no change in FFA or beta-OHB levels. T3 and leptin concentrations were decreased in transgenic mice. Expression of UCP1 in BAT of the -442 GLUT4 mice, which did not overexpress GLUT4 in this tissue, was not altered. These findings indicate that overexpression of GLUT4 up-regulates UCP3 expression in skeletal muscle and down-regulates UCP1 expression in BAT, possibly by increasing the rate of glucose uptake into these tissues.     

Regulation of Human Trophoblast GLUT1 Glucose Transporter by Insulin-Like Growth Factor I (IGF-I)

Glucose transport to the fetus across the placenta takes place via glucose transporters in the opposing faces of the barrier layer, the microvillous and basal membranes of the syncytiotrophoblast. While basal membrane content of the GLUT1 glucose transporter appears to be the rate-limiting step in transplacental transport, the factors regulating transporter expression and activity are largely unknown. In view of the many studies showing an association between IGF-I and fetal growth, we investigated the effects of IGF-I on placental glucose transport and GLUT1 transporter expression. Treatment of BeWo choriocarcinoma cells with IGF-I increased cellular GLUT1 protein. There was increased basolateral (but not microvillous) uptake of glucose and increased transepithelial transport of glucose across the BeWo monolayer. Primary syncytial cells treated with IGF-I also demonstrated an increase in GLUT1 protein. Term placental explants treated with IGF-I showed an increase in syncytial basal membrane GLUT1 but microvillous membrane GLUT1 was not affected. The placental dual perfusion model was used to assess the effects of fetally perfused IGF-I on transplacental glucose transport and syncytial GLUT1 content. In control perfusions there was a decrease in transplacental glucose transport over the course of the perfusion, whereas in tissues perfused with IGF-I through the fetal circulation there was no change. Syncytial basal membranes from IGF-I perfused tissues showed an increase in GLUT1 content. These results demonstrate that IGF-I, whether acting via microvillous or basal membrane receptors, increases the basal membrane content of GLUT1 and up-regulates basal membrane transport of glucose, leading to increased transepithelial glucose transport. These observations provide a partial explanation for the mechanism by which IGF-I controls nutrient supply in the regulation of fetal growth.     

Differential expression of glucose transporter isoforms during embryonic stem cell differentiation

In mouse blastocysts six facilitative glucose transporter isoforms (GLUT)1-4, 8 and 9 are expressed. We have used the mouse embryonic stem (ES) cell line D3 and spontaneously differentiating embryoid bodies (EB) to investigate GLUT expression and the influence of glucose during differentiation of early embryonic cells. Both ES cells and EBs (2d-20d) expressed GLUT1, 3, and 8, whereas the isoforms 2 and 4 were detectable exclusively in EBs. Differentiation-associated expression of GLUT was analyzed by double staining with stage-specific embryonic antigen (SSEA-1), cytokeratins (CK18, 19), nestin, and desmin. Similar to trophoblast cells in mouse blastocysts the outer cell layer of endoderm-like cells showed a high GLUT3 expression in early EBs. In 20-day-old EBs no GLUT3 protein and only minor GLUT3 mRNA amounts could be detected. A minimal glucose concentration of 5 mM applied during 2 and 8 days of EB culture resulted in up-regulated GLUT4, Oct-4 and SSEA-1 levels and a delay in EB differentiation. We conclude that GLUT expression depends on cellular differentiation and that the expression is modulated by glucose concentration. The developmental and glucose-dependent regulation of GLUT strongly suggests a functional role of glucose and glucose transporters in ES cell differentiation and embryonic development.