Vitor Antonio Fortuna

β‐Carotene storage, conversion to retinoic acid, and induction of the lipocyte phenotype in hepatic stellate cells

Hepatic stellate cells (HSCs) are the major site of retinol (ROH) metabolism and storage. GRX is a permanent murine myofibroblastic cell line, derived from HSCs, which can be induced to display the fat‐storing phenotype by treatment with retinoids. Little is known about hepatic or serum homeostasis of β‐carotene and retinoic acid (RA), although the direct biogenesis of RA from β‐carotene has been described in enterocytes. The aim of this study was to identify the uptake, metabolism, storage, and release of β‐carotene in HSCs. GRX cells were plated in 25 cm2 tissue culture flasks, treated during 10 days with 3 μmol/L β‐carotene and subsequently transferred into the standard culture medium. β‐Carotene induced a full cell conversion into the fat‐storing phenotype after 10 days. The total cell extracts, cell fractions, and culture medium were analyzed by reverse phase high‐performance liquid chromatography for β‐carotene and retinoids. Cells accumulated 27.48 ± 6.5 pmol/L β‐carotene/106 cells, but could not convert it to ROH nor produced retinyl esters (RE). β‐Carotene was directly converted to RA, which was found in total cell extracts and in the nuclear fraction (10.15 ± 1.23 pmol/L/106 cells), promoting the phenotype conversion. After 24‐h chase, cells contained 20.15 ± 1.12 pmol/L β‐carotene/106 cells and steadily released β‐carotene into the medium (6.69 ± 1.75 pmol/ml). We conclude that HSC are the site of the liver β‐carotene storage and release, which can be used for RA production as well as for maintenance of the homeostasis of circulating carotenoids in periods of low dietary uptake.

Retinol uptake and metabolism, and cellular retinol binding protein expression in an in vitro model of hepatic stellate cells

Liver is a major site of retinoid metabolism and storage, and more than 80% of the liver retinoids are stored in hepatic stellate cells. These cells represent less than 1% of the total liver protein, reaching a very high relative intracellular retinoid concentration. The plasma level of retinol is maintained close to 2 μM, and hepatic stellate cells have to be able both to uptake or to release retinol depending upon the extracellular retinol status. In view of their paucity in the liver tissue, stellate cells have been studied in primary cultures, in which they loose rapidly the stored lipids and retinol, and convert spontaneously into the activated myofibroblast phenotype, turning a long-term study of their retinol metabolism impossible. We have analyzed the retinol metabolism in the established GRX cell line, representative of stellate cells. We showed that this cell line behaves very similarly, with respect the retinol uptake and release, to primary cultures of hepatic stellate cells. Moreover, we showed that the cellular retinol binding protein (CRBP-I) expression in these cells, relevant for both uptake and esterification of retinol, responds to the extracellular retinol status, and is correlated to the retinol binding capacity of the cytosol. Its expression is not associated with the overall induction of the lipocyte phenotype by other agents. We conclude that the GRX cell line represents an in vitro model of hepatic stellate cells, and responds very efficiently to wide variations of the extracellular retinol status by autonomous controls of its uptake, storage or release.

Acyl-CoA: retinol acyltransferase (ARAT) and lecithin:retinol acyltransferase (LRAT) activation during the lipocyte phenotype induction in hepatic stellate cells

We have examined retinol esterification in the established GRX cell line, representative of hepatic stellate cells, and in primary cultures of ex vivo purified murine hepatic stellate cells. The metabolism of [3H]retinol was compared in cells expressing the myofibroblast or the lipocyte phenotype, under the physiological retinol concentrations. Retinyl esters were the major metabolites, whose production was dependent upon both acyl-CoA:retinol acyltransferase (ARAT) and lecithin:retinol acyltransferase (LRAT). Lipocytes had a significantly higher esterification capacity than myofibroblasts. In order to distinguish the intrinsic enzyme activity from modulation of retinol uptake and CRBP-retinol content of the cytosol in the studied cells, we monitored enzyme kinetics in the purified microsomal fraction. We found that both LRAT and ARAT activities were induced during the conversion of myofibroblasts to lipocytes. LRAT induction was dependent upon retinoic acid, while that of ARAT was dependent upon the overall induction of the fat storing phenotype. The fatty acid composition of retinyl-esters suggested a preferential inclusion of exogenous fatty acids into retinyl esters. We conclude that both LRAT and ARAT participate in retinol esterification in hepatic stellate cells: LRATs activity correlates with the vitamin A status, while ARAT depends upon the availability of fatty acyl-CoA and the overall lipid metabolism in hepatic stellate cells.

Hepatic stellate cells uptake of retinol associated with retinol-binding protein or with bovine serum albumin

Retinol is stored in liver, and the dynamic balance between its accumulation and mobilization is regulated by hepatic stellate cells (HSC). Representing less than 1% total liver protein, HSC can reach a very high intracellular retinoid (vitamin‐A and its metabolites) concentration, which elicits their conversion from the myofibroblast to the fat‐storing lipocyte phenotype. Circulating retinol is associated with plasma retinol‐binding protein (RBP) or bovine serum albumin (BSA). Here we have used the in vitro model of GRX cells to compare incorporation and metabolism of BSA versus RBP associated [3H]retinol in HSC. We have found that lipocytes, but not myofibroblasts, expressed a high‐affinity membrane receptor for RBP–retinol complex (KD = 4.93 nM), and both cell types expressed a low‐affinity one (KD = 234 nM). The RBP–retinol complex, but not the BSA‐delivered retinol, could be dislodged from membranes by treatments that specifically disturb protein–protein interactions (high RBP concentrations). Under both conditions, treatments that disturb the membrane lipid layer (detergent, cyclodextrin) released the membrane‐bound retinol. RBP‐delivered retinol was found in cytosol, microsomal fraction and, as retinyl esters, in lipid droplets, while albumin‐delivered retinol was mainly associated with membranes. Disturbing the clathrin‐mediated endocytosis did not interfere with retinol uptake. Retinol derived from the holo‐RBP complex was differentially incorporated in lipocytes and preferentially reached esterification sites close to lipid droplets through a specific intracellular traffic route. This direct influx pathway facilitates the retinol uptake into HSC against the concentration gradients, and possibly protects cell membranes from undesirable and potentially noxious high retinol concentrations.