Rapid characterization of cellular retinoid-binding proteins by high-performance size-exclusion chromatography

A high-performance size-exclusion chromatographic method was developed for the analysis of cellular retinol and retinoic acid-binding proteins. The chromatographic analysis of the retinol-retinol-binding protein complex or the retinoic acid-retinoic acid-binding protein complex requires 15 min. The use of high-specific-radioactivity retinoid ligand (30-40 Ci/mmol) allows routine detection of 25 fmol of retinoid-binding protein/mg of cytosolic protein. Thus, this method provides a rapid alternative to sucrose density gradient sedimentation analysis of the cellular retinoid-binding proteins. High-performance size-exclusion chromatography is well suited to screening novel tissues, tumors, and cell lines for the presence of retinoid-binding proteins and to the further characterization of these cellular proteins. This method was applied to the characterization of cellular retinol- and retinoic acid-binding proteins in fetal rabbit tissues. Both retinol- and retinoic acid-binding proteins were detected in fetal rabbit brain, intestine, kidney, and lung at a gestational age of 28 days. Neither retinoid-binding protein was detected in 28-day-old fetal rabbit placenta. Specific retinoic acid binding in fetal intestinal cytosol decreased as a function of increasing gestational age.     

Novel retinoid-binding proteins from filarial parasites.

The present study deals with the discovery and partial characterization of specific binding proteins for retinol and retinoic acid from filarial parasites (worms of the superfamily Filarioidea), including those from two species of Onchocerca. These binding proteins, which are distinct in their physicochemical properties and in the mode of ligand interactions from the host-tissue retinoid-binding proteins, may be involved in the mediation of the putative biological roles of retinoids in the control of parasitic growth, differentiation and reproduction. Parasite retinol-binding protein and retinoic acid-binding protein exhibited specificity for binding retinol and retinoic acid respectively. Both the binding proteins showed an s20,w value of 2.0 S. On gel filtration, both proteins were retarded to a position corresponding to the same molecular size (19.0 kDa). On preparative columns, the parasite binding proteins exhibited isoelectric points at pH 5.7 and 5.75. Unlike the retinoid-binding proteins of mammalian and avian origin, the parasite retinoid-binding proteins showed a lack of mercurial sensitivity in ligand binding. The comparative amounts of retinoic acid-binding protein in five parasites, Onchocerca volvulus, Onchocerca gibsoni, Dipetalonema viteae, Brugia pahangi and Dirofilaria immitis, were between 2.7 and 3.1 pmol of retinoic acid bound/mg of extractable protein. However, the levels of parasite retinol-binding protein were between 4.8 and 5.8 pmol/mg, which is considerably higher than the corresponding levels of cellular retinol-binding protein of mammalian and avian origin. Both retinol- and retinoic acid-binding-protein levels in O. volvulus-infected human nodules and O. gibsoni-infected bovine nodules were similar to their levels in mammalian tissues. Also, these nodular binding proteins, like the host-binding proteins, exhibited mercurial sensitivity to ligand interactions.     

Determination of apo and holo retinoic acid-binding protein levels in retinoid-responsive transformed cells by high-performance size-exclusion chromatography

A method to measure the endogenous levels of apo and holo cellular retinoic acid-binding proteins was developed using calf testis cytosol as the source of retinoic acid-binding protein. [3H]Retinoic acid-retinoic acid-binding protein complexes were assayed by high-performance size-exclusion chromatography. Preincubation of cytosol with 10 mM p-hydroxymercuribenzoate at 4 degrees C resulted in complete inhibition of retinoic acid binding to apo retinoic acid-binding protein. In addition, total dissociation of preformed holo retinoic acid-binding protein complexes was noted within 20 min after mercurial addition. Thus, p-hydroxymercuribenzoate converted the total pool of cellular retinoic acid-binding protein (apo plus holo) to mercurial-protein complexes unable to bind retinoic acid in vitro. Mercurial inhibition of retinoic acid-retinoic acid-binding protein complex formation was totally reversed upon the addition of 50 mM dithiothreitol. Total cytosolic retinoic acid-binding protein was determined from specific retinoic acid binding after treatment with p-hydroxymercuribenzoate and dithiothreitol. Apo cellular retinoic acid-binding protein concentration was measured by determining specific radioligand binding prior to p-hydroxymercuribenzoate treatment, and correcting for exchange of endogenously bound retinoid with exogenous tritiated retinoic acid. Holo cellular retinoic acid-binding protein concentration was derived from the difference between total and apo retinoic acid-binding protein concentrations. Using this method, we have demonstrated that retinoid-responsive EJ and T24 human bladder carcinoma cell lines and AT3A and AT3B rat pancreatic acinar carcinoma cell lines lack detectable levels of either apo or holo cellular retinoic acid-binding protein. These results established that retinoid inhibition of transformed bladder and acinar cell proliferation in culture was mediated by a cellular retinoic acid-binding protein-independent mechanism.     

Retinoic acid-binding protein in rat tissue. Partial purification and comparison to rat tissue retinol-binding protein.

When the 100,000 X g supernatant fractions of several rat organs are incubated with all-trans-[3H]retinoic acid, a binding component for retinoic acid with a sedimentation coefficient of 2 S can be detected by sucrose gradient centrifugation. This tissue binding protein for retinoic acid is distinct from the tissue binding protein for retinol which has been previously described. The tissue retinoic acid-binding protein has been partially purified from rat testis and this partially purified protein would appear to have a molecular weight of 14,500 as determined by gel filtration and high binding specificity for all-trans-retinoic acid. Binding of [3H]retinoic acid is not diminished by a 200-fold molar excess of retinal, retinol, or oleic acid but is reduced by a 200-fold excess of unlabeled retinoic acid. Tissue retinoic acid-binding protein can be detected in extracts of brain, eye, ovary, testis, and uterus but is apparently absent in heart muscle, small intestine, kidney, liver, lung, gastrocnemious muscle, serum, and spleen. This distribution is different than that observed for the tissue retinol-binding protein. Tissue retinol-binding protein was also purified extensively from rat testis. The partially purified protein has an apparent molecular weight of 14,000 and high binding specificity for all-trans-[3H]retinol as only unlabeled all-trans-retinol but not retinal, retinoic acid, retinyl acetate, retinyl palmitate, or oleic acid could diminish binding of the 3H ligand under the conditions employed. The partially purified protein has a fluorescence excitation spectrum with lambda max at 350 nm. In contrast, the retinol-binding protein isolated from rat serum and described by others has a fluorescence excitation spectrum with lambda max at 334 nm and an apparent molecular weight of 19,000. When partially purified tissue retinol-binding protein is extracted with heptane, the heptane extract has a fluorescence excitation spectrum similar to that of all-trans-retinol.     

Absence of phosphorylation of retinoid-binding proteins by protein kinase C in vitro

Cellular retinol-binding protein, cellular retinoic acid-binding protein, and fetal cellular retinol-binding protein were purified to homogeneity and each polypeptide had a molecular weight of 16,000. Their apoproteins were not phosphorylated under the same conditions. Their holoproteins did not inhibit the phosphorylation of histone III-S by protein kinase C. Each of these observations is contrary to the results reported by Cope et al. (Biochem. Biophys. Res. Commun., 120, 593-601, 1984).     

3H]thymidine incorporation into whole liver as an alternative to [3H]thymidine incorporation into DNA as a parameter of cell proliferation in regenerating liver tissue in rats

OBJECTIVE: To monitor liver regeneration following partial hepatectomy, liver cell proliferation can be measured by assaying in vivo [3H]thymidine incorporation into liver cell DNA. We hypothesized that [3H]thymidine incorporation into whole liver tissue parallels [3H]thymidine incorporation into liver cell DNA, both in high proliferating and low proliferating liver. STUDY DESIGN: Liver cell proliferation in rats after partial hepatectomy or a sham operation was studied by measuring incorporation of [3H]thymidine into various fractions of liver tissue on days 1, 2, 3, 4 and 10 after surgery. RESULTS: [3H]thymidine incorporation into whole liver tissue and in the protein fraction correlated well with DNA-specific [3H]thymidine incorporation into regenerating (r > .80, P < .0001) and nonregenerating liver (r > .69, P < .005). [3H]thymidine incorporation into DNA was < 5% of the total amount of administered [3H]thymidine in both sham-operated and hepatectomized rats. Significant differences in [3H]thymidine incorporation into partially hepatectomized livers as compared to sham-operated rat livers were found on days 1 and 2 (whole liver tissue and protein fraction) or day 1 (DNA) after surgery. CONCLUSION: [3H]thymidine incorporation into whole liver tissue is a simple technique that can be used for the study of liver cell proliferation after partial hepatectomy in rats.     

Effects of dietary retinoic acid on cellular retinol- and retinoic acid-binding protein levels in various rat tissues

A study was conducted to explore the effects of retinoic acid, fed to retinol-deficient rats, on the tissue distribution and levels of cellular retinol-binding protein (CRBP) and cellular retinoic acid-binding protein (CRABP). Sensitive and specific radioimmunoassays were employed to measure the levels of both CRBP and CRABP. Two groups of six male rats each were fed a purified retinoid-deficient diet supplemented with either: i) retinyl acetate (control group); or ii) retinoic acid (30 mg/kg diet) (retinol deficient-retinoic acid group). The retinoic acid supplementation was begun after 38 days on the retinoid-deficient diet alone, and was continued for 52-54 days. Analysis of the data indicated that only the CRBP level of the proximal epididymis in the retinol-deficient/retinoic acid group differed significantly from (was lower than) the corresponding control level, at the 1% confidence level. CRABP tissue levels did not differ significantly between the two groups. Thus, a moderately large intake of retinoic acid, as the only source of retinoids, had very little effect on the tissue distribution or levels of either its own cellular binding protein (CRABP) or of CRBP. This study provides further information showing that the tissue levels of the cellular retinoid-binding proteins are highly regulated and maintained in rats, even in the presence of marked changes in retinoid nutritional status.     

Promotion of the release of 11-cis-retinal from cultured retinal pigment epithelium by interphotoreceptor retinoid-binding protein.

This study investigates whether the interphotoreceptor retinoid-binding protein (IRBP) is necessary for the release of 11-cis-retinaldehyde (RAL) or if the retinoid is constitutively released from the retinal pigment epithelium (RPE) following synthesis. The strategic location of IRBP in the interphotoreceptor matrix (IPM) and its retinoid-binding ability make it a candidate for a role in 11-cis-RAL release. Fetal bovine RPE cells were grown in permeable chambers, and their apical surfaces were incubated with medium containing either apo-IRBP, the apo form of cellular retinaldehyde-binding protein (CRALBP), the apo form of serum retinol-binding protein (RBP), or bovine serum albumin (BSA) or with medium devoid of binding proteins. [3H]-all-trans-Retinol (ROL) was delivered to the basal surface of the cells by RBP. High-performance liquid chromatography demonstrated that [3H]-11-cis-RAL was optimally released into the apical medium when apo-IRBP was present. The most surprising result was the diminished level of [3H]-11-cis-RAL when apo-CRALBP was in the apical medium. Circular dichroism demonstrated that CRALBP had not been denatured by the photobleaching required for endogenous ligand removal. Therefore, apo-CRALBP should have been able to bind [3H]-11-cis-RAL if it was constitutively released into the apical medium. In addition, when proteins other than apo-IRBP were present, or if the cells were incubated with medium alone, the observed decrease in apical [3H]-11-cis-RAL was concomitant with a buildup of intracellular [3H]-all-trans-retinyl palmitate and [3H]-all-trans-ROL in the basal culture medium.(ABSTRACT TRUNCATED AT 250 WORDS)     

Purification and partial characterization of a novel cellular retinol-binding protein, type three, from the piscine eyes

A novel cellular retinol-binding protein, termed type three (CRBP III), was isolated from eyes of the bigeye of tuna. CRBP III showed a molecular weight of 15,400, an isoelectric point of 4.80, alpha 1-mobility in electrophoresis, and a lambda max of 350 nm. All-trans-retinol, the endogenous ligand, could be competitively displaced by retinoic acid but not by retinal. CRBP III was differentiated from purified piscine and rat cellular retinol-binding proteins (CRBP) and cellular retinoic acid-binding proteins (CRABP) by its amino-acid composition, electrophoretic mobility, fluorescence spectra and ligand-binding specificity.     

Enzymatic esterification of vitamin a in the pigment epithelium of bovine retina

The kinetic properties and subcellular distribution of an esterifying enzyme in the pigment epithelium of bovine retina have been studied using both [1-³H]retinol and [³H]retinol bound to cellular retinol-binding protein as substrates. The most active esterifying fraction in pigment epithelial cell preparations was the microsomes, but the lysosome plus mitochondria fraction also showed some activity, probably due to endoplasmic reticulum present as an impurity. The microsomal enzyme showed optimum activity at pH 7.5, and the reaction was linear up to 30 μg protein and for the first 10–15 min. The apparent K_m values were 16.6 · 10^?6 and 5.5 · 10^?6 M for [³H]retinol and bound [³H]retinol, respectively. This is the first time that retinol bound to cellular retinol-binding protein has been shown to undergo metabolic stransformation. The microsomal esterifying activity was destroyed by boiling for 1 min, or after freezing for 2 months. No clear requirement for ATP, CoA or fatty acid could be demonstrated.Of all the other tissues examined under the same experimental conditions as those used for the pigment epithelium, onlt intestine showed measurable activity. With larger amounts of tissue protein and longer incubation periods, activity was also detectable in microsomes of liver, testis and retina     

Separable binding proteins for retinoic acid and retinol in bovine retina.

Binding proteins for retinoic acid and retinol were separated from a supernatant prepared from bovine retina. Fraction IV from DEAE-cellulose chromatography bound exogenous [3H] retinoic acid which could not be effectively displaced by retinol, retinal, retinyl acetate or palmitate, but which was readily displaced with excess retinoic acid. [3H] Retinol was bound by fraction V from DEAE-cellulose chromatography and was not displaced by retinal, retinoic acid, retinyl acetate or retinyl palmitate, but was readily displaced by excess retinol. Unlike bovine serum retinol-binding protein, neither intracellular binding protein formed a complex with purified human serum prealbumin. The supernatant from bovine retinas was estimated to contain five times more retinoic acid binding than retinol binder.     

Metabolism of retinoids by embryonal carcinoma cells

Several embryonal carcinoma (EC) cell lines were tested in culture for their ability to metabolize all-trans-[3H]retinol, all-trans-[3H]retinyl acetate, and all-trans-[3H]retinoic acid. There was little, if any, metabolism of all-trans-retinol to more polar compounds; we failed to detect conversion to acidic retinoids by reverse-phase high performance liquid chromatography and derivatization. We also did not observe [3H]retinoic acid when EC cells were incubated with [3H]retinyl acetate. Unlike the other retinoids, all-trans-[3H]retinoic acid, even at micromolar levels, was almost totally modified by cells from several EC lines within 24 h. Most of the labeled products were secreted into the medium. Some EC lines metabolized retinoic acid constitutively, whereas others had an inducible enzyme system. A differentiation-defective line, which contains little or no cellular retinoic acid-binding protein activity, metabolized retinoic acid poorly, even after exposure to inducers. At least eight retinoic acid metabolites were generated; many contain hydroxyl residues. Our data lead us to propose that retinol does not induce differentiation of EC cells in vitro via conversion to retinoic acid. Also, the relatively rapid metabolism of retinoic acid by EC cells suggests either that the induction of differentiation need involve only a transient exposure to this retinoid or that one or more of the retinoic acid metabolites can also promote differentiation.     

Differentiation-dependent expression of retinoid-binding proteins in BFC-1 beta adipocytes.

Recently, we demonstrated that adipose tissue plays an important role in retinol storage and retinol-binding protein (RBP) synthesis. Our data suggested that RBP expression in adipose tissue is dependent on the state of adipocyte differentiation. To examine this possibility, we explored the differentiation-dependent expression of RBP using BFC-1 beta preadipocytes, which can be stimulated to undergo adipose differentiation. Total RNA was isolated from undifferentiated (preadipocytes) and differentiated (adipocytes) BFC-1 beta cells and analyzed by Northern blotting. RBP mRNA was not detected in the preadipocytes, but considerable RBP mRNA was present in differentiated BFC-1 beta cells. In BFC-1 beta cells, induced to differentiate with insulin and thyroid hormone, RBP mRNA was first detected after 4 days, reached a maximum level by day 10, and remained at this maximum level for at least 2 more days. Cellular retinol-binding protein was expressed at low levels in the BFC-1 beta preadipocytes and the level of expression increased for 6 days after induction to differentiate and slowly declined on later days. Neither the maximum level of RBP expression nor the day on which this level was reached was influenced by the level of retinol provided in the BFC-1 beta culture medium. BFC-1 beta cells secreted newly synthesized RBP into the culture medium at a rate of 43 +/- 14 ng RBP/24 h/10(6) adipocytes. When the BFC-1 beta adipocytes were provided 1.0 microM retinol in the medium, they accumulated the retinol and synthesized retinyl esters. These studies with BFC-1 beta cells confirm that RBP synthesis and secretion and retinol accumulation are intrinsic properties of differentiated adipocytes. Furthermore, they suggest that RBP and cellular retinol-binding protein gene expression are regulated as part of a package of genes which are modulated during adipocyte differentiation.     

Retinoic acid-induced modifications in the growth and cell surface components of a human carcinoma (HeLa) cell line

The addition of retinoic acid to cultures of HeLa-S3 cells caused a reduction in cell proliferation rate which became apparent after 72 h and was linearly dependent on retinoic acid concentration in the range 10⁻⁹–10⁻⁵ M. After 72 h of exposure to retinoic acid, the cells assumed a flattened appearance and no longer formed multilayers. These changes were reversed within 48 h after removal of retinoic acid from the medium. Structural analogs of retinoic acid with a free ---COOH group at C-15 were usually more potent in growth inhibition than compounds with an alcohol, aldehyde, ether or ester group. A cellular retinoic acid-binding protein was detected in cell homogenates, and the binding of [³H]retinoic acid to the binding protein was inhibited by most, but not all, analogs possessing a free terminal ---COOH group. For example, the 4-oxo analog of retinoic acid, while capable of inhibiting cellular proliferation, failed to bind to the retinoic acid-binding protein. Analysis of cell surface and cellular glycoproteins by lactoperoxidase-catalysed ¹²⁵I iodination and by metabolic labeling with [³H]glucosamine revealed that a 190000 D glycoprotein which was labeled by both methods and a 230000 D glycoprotein which was labeled only with [³H]glucosamine were labeled more intensely in retinoic acid-treated cells compared with untreated cells. The electrophoretic mobility of the 230000 D glycoprotein could be modified by treatment of intact cells with either neuraminidase or proteolytic enzymes, suggesting that this glycoprotein is also exposed on the cell surface. The cell surface alterations were detected much earlier than the onset of growth inhibition and appeared as early as 24 h after exposure to retinoic acid. The possible relationship between retinoic acid-induced changes in cell membrane structure, cell morphology, and cell proliferation is discussed.     

The primary structure of bovine cellular retinoic acid-binding protein

The complete amino acid sequence of bovine adrenal gland cellular retinoic acid-binding protein (CRABP) has been determined. The primary structure was established by analyses of cyanogen bromide fragments and peptides obtained by trypsin and Staphylococcus aureus protease digestions. The polypeptide chain of bovine CRABP comprises 136 amino acid residues. From partial sequence information, CRABP has been shown to be homologous to cellular retinol-binding protein, myelin protein P2, and the fatty acid-binding Z-protein. A comparison of the complete amino acid sequences of the members of this protein family, which also includes the rat intestinal fatty acid-binding protein, shows that CRABP is more similar to cellular retinol-binding protein and protein P2 than to the fatty acid-binding proteins. All five proteins are very similar in their NH2-terminal regions, suggesting that this part is important for a property common to the members of this protein family. This is the first report of a complete amino acid sequence of a CRABP.     

Separable binding proteins for retinoic acid and retinol in bovine retina

Binding proteins for retinoic acid and retinol were separated from a supernatant prepared from bovine retina. Fraction IV from DEAE-cellulose chromatography bound exogenous [³H] retinoic acid which could not be effectively displayed by retinol, retinal, retinyl acetate or palmitate, but which was readily displaced with excess retinoic acid. [³H] Retinol was bound by fraction V from DEAE-cellulose chromatography and was not displaced by retinal, retinoic acid, retinyl acetate or retinyl palmitate, but was readily displaced by excess retinol. Unlike bovine serum retinol-binding protein, neither intracellular binding protein formed a complex with purified human serum prealbumin. The supernatant from bovine retinas was estimated to contain five times more retinoic acid binding than retinol binder.     

Purification and characterization of a murine cellular retinoic acid-binding protein

A cellular retinoic acid-binding protein from 1-day-old mouse pups has been purified to homogeneity. The isolation procedure included gel filtration on Sephadex G-75, ion exchange chromatography on DEAE cellulose, and chromatofocusing on PBE9-4 ion exchange resin. The chromatofocusing step was most useful in removing the major contaminants, which were otherwise difficult to remove. The binding protein was finally subjected to two cycles of high performance liquid chromatography on a DEAE-5PW column to achieve homogeneity. The protein has an isoelectric point of 4.75 and consists of a single polypeptide, migrating with an apparent Mr of 14,600 in SDS--polyacrylamide gel electrophoresis. Amino-terminal sequence analysis showed that the mouse cellular retinoic acid-binding protein has a high percentage of amino acid identity with other retinoid-binding proteins. However, it is immunologically distinct from the cellular retinol-binding protein.     

The effect of dibutyryl cyclic AMP and butyrate on F9 teratocarcinoma cellular retinoic acid-binding protein activity

F9 teratocarcinoma cells contain a cellular retinoic acid-binding protein (CRABP) that may mediate the retinoic acid-induced differentiation of this cell line. Specific [3H]retinoic acid binding to CRABP in F9 stem cell cytosol is protein-dependent, reaches equilibrium within 4 h at 4 degrees C, and yields 643 +/- 105 fmol of [3H]retinoic acid per mg of protein with an apparent dissociation constant of 9.2 +/- 1.1 nM. When F9 stem cells are grown in the presence of either dibutyryl cyclic AMP or sodium butyrate, CRABP activity is stimulated 2-4-fold. The effect of these drugs on CRABP activity is both time and concentration-dependent, resulting in an increase in the number of binding sites for [3H]retinoic acid with no change in their affinity. The new [3H]retinoic acid-binding sites have a sedimentation coefficient of 2 S and are not displaced by excess retinol. When F9 stem cells are grown in the presence of cyclic 8-bromo-AMP or cholera toxin, no increase in CRABP activity is observed. We conclude that the stimulation of CRABP activity by dibutyryl cyclic AMP may result from the action of butyrate. In addition, the stimulation of retinoic acid-induced F9 cell differentiation by cyclic AMP analogs (Strickland, S., Smith, K.K., and Marotti, K.R. (1980) Cell 21, 347-355) and the inhibition of this differentiation by butyrate (Levine R. A., Campisi, J., Wang, S.-Y., and Gudas, L. J. (1984) Dev. Biol. 105, 443-450) are not correlated with increases or decreases, respectively, in the level of CRABP activity.     

The level of CRABP-I expression influences the amounts and types of all-trans-retinoic acid metabolites in F9 teratocarcinoma stem cells.

The CRABP-I and CRABP-II proteins are high affinity cytoplasmic retinoic acid-binding proteins. In undifferentiated F9 teratocarcinoma stem cells, only the CRABP-I protein is expressed at detectable levels. We have previously shown that overexpression of the CRABP-I protein in stably transfected F9 stem cell lines results in a lower sensitivity to a given external concentration of retinoic acid relative to that of untransfected F9 cells; in contrast, reduced CRABP-I expression in CRABP-I cDNA anti-sense transfected lines is associated with increased sensitivity of these lines to retinoic acid. These three types of cell lines were cultured in the presence of 50 nM [3H]retinoic acid, and the metabolism of retinoic acid was followed over the next 24 h. The results demonstrate that CRABP-I has the ability to alter both the levels and types of RA metabolites produced in the cytoplasm of differentiating embryonic stem cells. Moreover, the level of CRABP-I determines the rate of RA metabolism to 4-oxo-RA such that the higher the CRABP-I level, the faster the metabolism of [3H]retinoic acid. This is the first reported connection between the level of CRABP-I expression and intracellular RA metabolism.     

Retinoids, retinoid-binding proteins, and retinyl palmitate hydrolase distributions in different types of rat liver cells

A study was conducted to determine the levels and distributions of retinoids, retinol-binding protein (RBP), retinyl palmitate hydrolase (RPH), cellular retinol-binding protein (CRBP), and cellular retinoic acid-binding protein (CRABP) in different types of isolated liver cells. Highly purified fractions of parenchymal, fat-storing (stellate), endothelial, and Kupffer cells were isolated in high yield from rat livers. The retinoid content of each fraction was measured by HPLC analysis. RBP, CRBP, and CRABP were measured by sensitive and specific radioimmunoassays, and RPH activity was measured by a sensitive microassay. The concentrations of each parameter expressed per 10(6) parenchymal or fat-storing cells were, respectively: retinoids, 1.5 and 83.9 micrograms of retinol equivalents; RBP, 138 and 7.4 ng; RPH, 826 and 1152 pmol FFA formed hr-1; CRBP, 470 and 236 ng; and CRABP, 5.6 and 8.7 ng. When these data were expressed on the basis of per unit mass of cellular protein, the concentrations of RPH, CRBP, and CRABP in the fat-storing cells, which contain 10-fold less protein than the large parenchymal cells, were seen to be greatly enriched over parenchymal cells. The parenchymal cells contained approximately 9% of the total retinoids, 98% of the total RBP, 90% of the total RPH activity, 91% of the total CRBP, and 71% of the total CRABP found in the liver. The fat-storing cells accounted for approximately 88% of the total retinoids, 0.7% of the total RBP, 10% of the RPH activity, 8% of the total CRBP, and 21% of the CRABP in the liver. The endothelial and Kupffer cell fractions contained very low levels of all of these parameters. Thus, the large and abundant parenchymal cells account for greater than 70% of the liver's RBP, RPH, CRBP, and CRABP; but the much smaller and less abundant fat-storing cells contain the majority of hepatic retinoids and greatly enriched concentrations of RPH, CRBP, and CRABP.