![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
, 5
* Division of Nutrition, Department of Biochemistry, MCP·Hahnemann School of Medicine, Allegheny University, Philadelphia, PA 19129 and Department of Nutrition, Pennsylvania State University, University Park, PA 16802
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
LITERATURE CITED
ABSTRACT 
The cellular distribution of enzymes that esterify retinol and hydrolyze retinyl esters (RE) was studied in liver of vitamin A-sufficient,-deficient, and deficient rats treated with retinoic acid or N-(4-hydroxyphenyl)-retinamide. Livers were perfused and cell fractions enriched in hepatocytes, and nonparenchymal cells were obtained for assays of RE and enzyme activity. The specific activity of lecithin:retinol acyltransferase (LRAT) was approximately 10-fold greater in the nonparenchymal cell than the hepatocyte fraction from both vitamin A-sufficient and retinoid-treated rats. Total RE mass, newly synthesized [3H]RE and LRAT activity were positively correlated in liver and isolated cells of both normal (P < 0.0001) and retinoid-treated rats (P < 0.0002). In nonparenchymal cells, these three constituents were nearly equally enriched as evaluated by their relative specific activity values (RSA, defined as the percentage of recovered activity divided by the percentage of recovered protein), which were each significantly greater than 1.0, with values of 4.3 for total RE mass (P < 0.05), 3.6 for newly synthesized [3H]RE (P < 0.01) and 3.8 for LRAT activity (P < 0.01). In contrast, the specific activities of neutral and acid bile salt-independent retinyl ester hydrolases (REH) did not vary with vitamin A status, and their RSA values were close to 1.0 in both hepatocytes and nonparenchymal cells. These data show that LRAT and REH are differentially regulated by retinoids and that these enzymes also differ in their spacial distribution between liver parenchymal and nonparenchymal cells.
Key words: retinoic acid, retinamide, stellate cells, vitamin A, rats.INTRODUCTION
Nearly 50-80% of the total body vitamin A is stored in liver as long-chain fatty acid esters of retinol (Goodman and Blaner 1984
). Two types of liver cells, hepatocytes (parenchymal cells) and stellate cells, are known to play important roles in the assimilation, storage and mobilization of vitamin A (Blomhoff and Wake 1991, Blomhoff et al. 1991). Chylomicron remnants containing newly absorbed dietary retinol esters (RE)6 are taken into hepatocytes through receptor-mediated endocytosis. During or shortly after uptake, these RE are hydrolyzed and the unesterified retinol is re-esterified with long-chain fatty acids, predominantly palmitic, stearic and oleic acid, to form hepatic RE stores (Goodman et al. 1965). Within a few hours, vitamin A is transferred from hepatic parenchymal cells to the stellate cells (Blomhoff et al. 1991) by mechanisms that have not yet been elucidated. The stellate cells [also known as fat-storing cells, Ito cells, and lipocytes (Blomhoff and Wake 1991)] are nonparenchymal cells situated in the perisinusoidal space. In normal rat liver, these cells are characterized morphologically by numerous cytoplasmic lipid droplets and biochemically by a high concentration of esterified retinol. Hendriks et al. (1985) estimated that in normal rat liver greater than 75% of RE resides in stellate cells. Cell culture studies with isolated stellate cells have shown that the number and size of the lipid droplets vary directly with vitamin A uptake and storage (Blaner et al. 1985, Matsuura et al. 1993).
Research in several laboratories (see references in Blomhoff and Wake 1991) has demonstrated the importance of stellate cells in hepatic RE storage. However, there is still little information concerning the cellular location(s) of the enzymes responsible for RE formation and hydrolysis. Lecithin:retinol acyltransferase (LRAT), the principal activity involved in hepatic retinol esterification (MacDonald and Ong 1988, Randolph and Ross 1991, Randolph et al. 1991), catalyzes the transfer of the sn-1 fatty acid from membrane phosphatidyl choline to retinol bound to the cellular retinol-binding proteins, CRBP or CRBP-II (MacDonald and Ong 1988). This enzyme has been localized to the endoplasmic reticulum (microsomal) fraction of liver, and, in liver, its activity has been shown to be strongly regulated by vitamin A and related retinoids (Matsuura and Ross 1993, Matsuura et al. 1996, Randolph and Ross 1991). Whereas livers of vitamin A-deficient rats had negligible LRAT activity (Randolph and Ross 1991) and lacked the ability to esterify a test dose of [3H]retinol in vivo (Matsuura and Ross 1993), LRAT activity increased rapidly after these rats were treated with retinol (Randolph and Ross 1991), retinoic acid or 4-hydroxyphenyl retinamide (4-HPR) (Matsuura and Ross 1993, Matsuura et al. 1996).
The processes of chylomicron clearance and the mobilization of stored RE from intracellular lipid droplets both require RE hydrolysis. Chylomicron RE clearance takes place in hepatocytes, whereas most stored RE is present in stellate cells. A number of biochemically distinct RE hydrolase (REH) activities have been described in various subcellular fractions of rat liver (Harrison 1993). A neutral, bile salt-dependent REH that closely resembles pancreatic carboxyl ester lipase is present in the cytoplasmic and nuclear fractions (Harrison and Gad 1989); this REH activity is distributed broadly among hepatocytes and stellate cells, with little activity in endothelial or Kupffer cells (Blaner et al. 1985). Membrane-associated REH also have been described that are maximally active at neutral or acid pH and that are capable of hydrolyzing RE in the absence of bile salts (Gad and Harrison 1991, Harrison 1993). These bile salt-independent REH activities were localized in plasma membranes and/or endosomes (Gad and Harrison 1991, Harrison and Gad 1989, Harrison et al. 1995), sites that suggest their involvement in the initial catabolism of chylomicron remnant RE. However, their distribution between hepatocytes and nonparenchymal cells has not yet been reported.
The present studies had four related goals: 1) to determine the distribution of LRAT among hepatocytes and nonparenchymal cells in the livers of normal, vitamin A-sufficient rats; 2) to determine whether the LRAT induced in the livers of vitamin A-deficient rats by either retinoic acid or 4-HPR has the same cellular distribution as LRAT in vitamin A-sufficient rats; 3) to examine the relationship between LRAT activity measured in vitro and the ability of parenchymal and nonparenchymal cells to synthesize and store RE in vivo; and 4) to compare these data with the cellular distribution of neutral and acid bile salt-independent REH activities in order to provide a comprehensive picture of the location of liver enzymes responsible for RE synthesis and hydrolysis in both vitamin A-sufficient, vitamin A-deficient and retinoid-treated rats.
MATERIALS AND METHODS
) approved by the Institutional Animal Use and Care Committee of the Medical College of Pennsylvania. The control, vitamin A-sufficient diet was a nutritionally adequate semipurified diet, described previously (Bowman et al. 1990), that contained 4 mg retinol/kg. The same diet without vitamin A was used to induce vitamin A deficiency. The vitamin A-free diet was fed to the lactating mothers of the rats used in this study beginning at 8-10 d of lactation; their male and female offspring were fed either the control or vitamin A-free diet from the time of weaning (d 20-21). With this protocol, rats fed the latter diet became vitamin A deficient (assessed by a plasma retinol <0.2 µmol/L) by approximately 45 d (males) and 55 d of age (females) (Matsuura et al. 1996). The methods used to prepare 4-HPR and retinoic acid for administration in vivo were described by Matsuura et al. (1996). The amounts of retinoic acid (20 µg intraperitoneally) and 4-HPR (0.5 mg intragastrically) were chosen based on doses that produced maximum increases in LRAT in activity in previous experiments (Matsuura and Ross 1993, Matsuura et al. 1996). To determine the esterification of [3H]retinol in vivo, rats received an intravenous injection of 0.4 µg of [3H]retinol 20 h before liver perfusion. To prepare this dose, tritiated retinol (Du Pont NEN Products, Boston, MA) was mixed with unlabeled retinol and purified on a column of aluminum oxide (Ross 1982). The resulting [3H]retinol (approximately 1 µBq/0.4 µg) was stored in ethanol under nitrogen at 
20°C. Just prior to administration, ethanol was evaporated, and the [3H]retinol was mixed with 100 µL of ethanol and 10 µL of Tween 80 and then diluted with PBS. The concentration was determined by spectrophotometry (Matsuura and Ross 1993), and 200 µL (0.4 µg) was then was injected into a caudal vein.
Isolation and separation of liver cells. Rats were anesthetized with diethyl ether, and cannulae were inserted into the portal vein and inferior vena cava. The preparation of liver cell fractions is outlined in Figure 1. The liver was perfused first without collagenase at 37°C at approximately 15 mL/min with 100 mL of calcium-free perfusate (0.14 mol/L NaCl, 5 mmol/L KCl, 0.5 mmol/L NaH2PO4, 0.4 mmol/L Na2HPO4, 9 mmol/L HEPES, 0.5 mmol/L EGTA, 4 mmol/L NaHCO3, 0.09% glucose). Following this first perfusion, the perfusate was discarded, and perfusion continued in a circulating fashion for 5-7 min with 100 mL of collagenase solution [60,000 U/L (Worthington Biochemicals, Freehold, NJ) containing 0.14 mol/L NaCl, 5 mmol/L KCl, 5 mmol/L CaCl2, 0.5 mmol/L MgCl2, 0.4 mmol/L MgSO4, 0.5 mmol/L NaH2PO4, 0.3 mmol/L Na2HPO4, 9 mmol/L HEPES, 4 mmol/L NaHCO3, 0.05 g/L trypsin inhibitor and 10 g/L glucose (Seglen 1973
)]. After perfusion, a small portion of liver was removed and quickly frozen in liquid nitrogen for later determination of enzyme activities and RE mass in whole liver. The remaining tissue, which was very soft at this time, was dissociated by gentle filtration through a wire mesh (mesh size 500 µm) and a fine nylon sieve (mesh size 105 µm; Small Parts Inc., Miami Lakes, FL) to obtain a suspension of dispersed well-separated cells that was centrifuged at 60 × g for 1 min and then washed in PBS by recentrifugation under the same conditions. The washed cell pellet, termed the "first hepatocyte" fraction (HP-1), was highly enriched in viable parenchymal cells as judged by light microscopy (cells were approximately 20-25 µm in diameter and had distinct nuclei, and >80-90% excluded trypan blue dye). To maximize recovery of the nonparenchymal cells, the liver tissue that remained on the wire mesh was dissociated by shaking in PBS. All fractions were evaluated by light microscopy. This fraction, which contained dispersed cells of the size of both hepatocytes and smaller, nonparenchymal cells, was centrifuged at 60 × g for 1 min and washed in PBS by suspension and recentrifugation; it is referred to as the "second hepatocyte" fraction (HP-2). The trypan blue dye exclusion of cells in this fraction was lower (30-40%) than in HP-1. For the purposes of accurately assessing the recoveries of constituents, tissue that did not pass through the filter as a suspension of dispersed cells was collected and termed the "liver remainder"; this fraction contained parts of liver not well digested by collagenase. Following this, the supernatants from the two hepatocyte fractions were combined and centrifuged at 300 × g for 3 min to obtain the nonparenchymal cell-enriched fraction; this fraction consisted of intact cells smaller than and easily distinguished from hepatocytes. The remaining fraction, termed "combined supernatants and washes," consisted of broken cells, cell sap and membranes as judged by light microscopy, but was essentially free of intact cells. It was also saved for analysis so that recoveries could be calculated. Liver tissue and isolated cells were frozen rapidly in liquid nitrogen and stored at 70°C prior to analysis. The enzymes of interest were found in preliminary studies to be stable to storage under these conditions, whereas, in another preliminary study, collagenase perfusion was shown to have no effect on the hepatic LRAT activity of normal rats.
Preparation of liver and cell homogenates.
Portions of liver tissue and isolated liver cells were thawed and homogenized with a Potter-Elvejhem homogenizer in 2.5 volumes of ice-cold homogenization buffer (0.28 mol/L sucrose, 0.01 mol/L K2HPO4, 1 mmol/L dithiothreitol, pH 7.25). Protein concentrations were determined by the method of Markwell et al. (1978).
Assay of lecithin:retinol acyltransferase and retinyl ester hydrolase activities.
In most experiments, LRAT activity was assayed in homogenates of whole liver and cell fractions (Randolph and Ross 1991, Matsuura and Ross 1993) to avoid the loss of enzyme activity that sometimes occurred when microsomes were prepared from the isolated cell fractions. Homogenates of each sample containing 1 mg of protein were incubated in duplicate for 4 min with 5 µmol/L of [3H]retinol-CRBP in 0.15 mol/L K2HPO4 buffer, pH 7.4, containing 2 mmol/L dithiothreitol at 37°C; a boiled sample of each preparation served as background, which was subtracted. [3H]Retinyl ester was separated from [3H]retinol on columns of aluminum oxide, and radioactivity was determined by liquid scintillation counting (Ross 1982).
). All assays were conducted for 30 min in the absence of exogenous detergents or bile salts and contained 40 µmol/L retinyl palmitate as substrate. Neutral REH activity was assessed in reaction mixtures containing 50 mmol/L Tris-maleate, pH 8, and acid REH activity was assayed in reaction mixtures containing 50 mmol/L sodium acetate, pH 5. All enzyme assays were conducted under conditions where the extent of product formation was proportional to the amount of enzyme (protein) in the assay mixture and time of incubation.
). Esterified retinol was estimated as the difference between total retinol determined after saponification and unesterified retinol determined in nonsaponified plasma or liver cell homogenates. After in vivo administration of [3H]retinol, the [3H]RE in liver cell extracts was isolated by column chromatography, as for the LRAT assay.
RESULTS
|
Table 1. Lecithin:retinol acyltransferase (LRAT) specific activities in liver hepatocyte and nonparenchymal cell fractions of vitamin A-sufficient, -deficient and vitamin A-deficient, retinoid-treated rats1 |
|
Table 2. Distributions of protein, retinyl esters (RE) and lecithin:retinol acyltransferase (LRAT) activity in cell fractions of vitamin A-sufficient rat liver1 |
Table 3.
Distributions and relative specific activity (RSA) values of neutral and acid retinyl ester hydrolase (REH)
activites in cells from vitamin A-sufficient rat liver1
DISCUSSION
, Hendriks et al. 1987), are the major cellular site of vitamin A storage. Using preparations of highly purified hepatocytes and sinusoidal cells (comprised of stellate cells, Kupffer and endothelial cells) Hendriks et al. (1985) found that, per milligram of protein, stellate cells of vitamin A-sufficient rats contained approximately 300 times as much RE as hepatocytes. The composition of stellate cell RE, mainly retinyl palmitate, stearate and oleate, was indistinguishable from that found in whole liver. In contrast to stellate cells, neither Kupffer nor endothelial cells, which also are constituents of the nonparenchymal cell fraction, have been shown to play an important role in RE storage or retinoid metabolism. Blaner et al. (1985) confirmed stellate cells to be the major site of RE deposition. These investigators calculated the quantity of liver retinoids and retinoid-related proteins in stellate cells of normal rat liver, taking into account the lower number and mass of nonparenchymal cells compared with hepatocytes. They estimated that stellate cells contain ~88% of total retinoid, <1% of retinol-binding protein, 8% of CRBP, 21% of cellular retinoic acid-binding proteins, and 10% of a bile salt-dependent retinyl palmitate hydrolase (which differs from the REH enzymes that we have measured).
protein, RE mass, new [3H]RE, LRAT, acidic REH or neutral REHdid not differ significantly from one another (P > 0.05) or from the recovery of protein, and therefore the RSA values for RE, LRAT and REH enzymes in these fractions also did not differ from 1.0. These data indicate that although these fractions contained a substantial amount of liver protein, they were neither depleted nor enriched in any of the constituents measured and therefore do not affect the calculations of distribution or RSA for hepatocytes or nonparenchymal cells. Thus, in comparison to the RSA values for these constituents in hepatocytes, their RSA values in nonparenchymal cells are approximately five- to sixfold greater. Previous work (Blaner et al. 1985, Hendriks et al. 1985) demonstrated that, among cells of the nonparenchymal fraction, only stellate cells have an important role in vitamin A metabolism. In agreement, we found that the LRAT specific activity of Kupffer cells purified by Percoll density gradient centrifugation (Matsuura et al. 1989) was very low (less than 6% of that in the stellate cells from the same preparation; T. Matsuura and A. C. Ross, unpublished results). Thus, even though endothelial and Kupffer cells are present in the nonparenchymal cell fraction, it is unlikely that they affected the results for retinoid content or LRAT activity other than by contributing protein and thereby lowering (by dilution) the calculated RSA values for LRAT activity and retinoids. If RE is present nearly entirely in stellate cells of liver, then the co-enrichment of LRAT activity and RE implies that nearly all LRAT activity in the nonparenchymal cell fraction is also present in stellate cells. Although the RSA for LRAT activity in nonparenchymal cells exceeds that in hepatocytes by five- to sixfold, this may be a conservative estimate of its enrichment, because if nearly all RE are found in stellate cells and LRAT and RE are equally co-enriched in the nonparenchymal cell fraction (Table 2), then, based on co-localization, LRAT also is located primarily in stellate cells. This conclusion differs from that of Blaner et al. (1990), who interpreted their data from pronase E-Nycodenz centrifugation studies as showing that approximately 85% of LRAT activity resides in parenchymal cells. These authors did not, however, apply methods of analytical differential centrifugation that included recoveries of protein and the various cell constituents assayed. We believe that the analytical approach used in the present study is preferable for drawing inferences about the localization and distribution of LRAT and REH enzymes between liver parenchymal and nonparenchymal cells.
, Horn et al. 1986), in which the liver pathology includes a type of cytoskeletal intermediate filament, desmin, and a form of microfilament, alpha-smooth muscle actin, in stellate cells (Gressner 1995, Yokoi et al. 1984). However, these cytoskeletal elements are also present in fibroblasts and smooth muscle cells, and desmin could not be detected in stellate cells in human liver (Enzan et al. 1994, Nouchi et al. 1991). Therefore, the utility of these proteins as markers of stellate cells is questionable. The enrichment of LRAT activity in rat liver stellate cells and the ability to detect LRAT in human liver (McDonald and Ong 1988) suggest that, if LRAT in human liver also is enriched in stellate cells, then the assay of LRAT activity may prove useful as a marker of stellate cells in future studies of human liver disease.
). Although the exact physiological roles of the bile salt-dependent and -independent REH have yet to be clarified, the subcellular location of the bile salt-independent REH in liver plasma membranes and endosomes suggests that these enzymes may be involved in the initial metabolism of chylomicron RE (Gad and Harrison 1991, Harrison and Gad 1989, Harrison et al. 1995). Chylomicron uptake occurs almost exclusively in hepatocytes. The localization of both acid and neutral bile salt-independent REH in both parenchymal and nonparenchymal cell suggests that these enzymes also may have a role in mobilizing stored vitamin A esters.
FOOTNOTES
Manuscript received 19 July 1996. Initial reviews completed 6 September 1996. Revision accepted 7 October 1996.
LITERATURE CITED
This article has been cited by other articles:
|
|
 |
|
  L. Liu and L. J. Gudas Disruption of the Lecithin:Retinol Acyltransferase Gene Makes Mice More Susceptible to Vitamin A Deficiency J. Biol. Chem., December 2, 2005; 280(48): 40226 - 40234. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Ross and R. Zolfaghari Regulation of Hepatic Retinol Metabolism: Perspectives from Studies on Vitamin A Status J. Nutr., January 1, 2004; 134(1): 269S - 275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Ross Retinoid Production and Catabolism: Role of Diet in Regulating Retinol Esterification and Retinoic Acid Oxidation J. Nutr., January 1, 2003; 133(1): 291S - 296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Cerignoli, X. Guo, B. Cardinali, C. Rinaldi, J. Casaletto, L. Frati, I. Screpanti, L. J. Gudas, A. Gulino, C. J. Thiele, et al. retSDR1, a Short-Chain Retinol Dehydrogenase/Reductase, Is Retinoic Acid-inducible and Frequently Deleted in Human Neuroblastoma Cell Lines Cancer Res., February 1, 2002; 62(4): 1196 - 1204. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Kelleher and B. Lonnerdal Long-Term Marginal Intakes of Zinc and Retinol Affect Retinol Homeostasis without Compromising Circulating Levels during Lactation in Rats J. Nutr., December 1, 2001; 131(12): 3237 - 3242. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Zolfaghari and A. C. Ross Lecithin:retinol acyltransferase from mouse and rat liver: cDNA cloning and liver-specific regulation by dietary vitamin A and retinoic acid J. Lipid Res., December 1, 2000; 41(12): 2024 - 2034. [Abstract] [Full Text]
|
|
|
|
 |
|
 X. Guo, A. Ruiz, R. R. Rando, D. Bok, and L. J. Gudas Esterification of all-trans-retinol in normal human epithelial cell strains and carcinoma lines from oral cavity, skin and breast: reduced expression of lecithin:retinol acyltransferase in carcinoma lines Carcinogenesis, November 1, 2000; 21(11): 1925 - 1933. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Vogel, R. Piantedosi, J. Frank, A. Lalazar, D. C. Rockey, S. L. Friedman, and W. S. Blaner An immortalized rat liver stellate cell line (HSC-T6): a new cell model for the study of retinoid metabolism in vitro J. Lipid Res., June 1, 2000; 41(6): 882 - 893. [Abstract] [Full Text]
|
|
|
|
|
|
H. D. Dawson, Y. Yamamoto, R. Zolfaghari, F. J. Rosales, J. Dietz, T. Shimada, N.-q. Li, and A. C. Ross Regulation of Hepatic Vitamin A Storage in a Rat Model of Controlled Vitamin A Status during Aging J. Nutr., May 1, 2000; 130(5): 1280 - 1286. [Abstract] [Full Text]
|
|
|
|
|
|
E. H. Harrison Lipases and Carboxylesterases: Possible Roles in the Hepatic Utilization of Vitamin A J. Nutr., February 1, 2000; 130(2): 340 - 340. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
