HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 33, 30191-30197, August 16, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/33/30191    most recent M205046200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Quadro, L. Articles by Gottesman, M. E. Search for Related Content PubMed PubMed Citation Articles by Quadro, L. Articles by Gottesman, M. E. Social Bookmarking                      What's this?

Muscle Expression of Human Retinol-binding Protein (RBP)

SUPPRESSION OF THE VISUAL DEFECT OF RBP KNOCKOUT MICE^*

Loredana Quadro^§, William S. Blaner^¶, Leora Hamberger^¶, Russell N. Van Gelder^**, Silke Vogel^¶, Roseann Piantedosi^¶, Peter Gouras, Vittorio Colantuoni^§, and Max E. Gottesman

From the  Institute of Cancer Research and the Departments of ^¶ Medicine and  Ophthalmology, Columbia University, College of Physicians and Surgeons, New York, New York 10032, the ^§ Faculty of Biological Sciences, University of Sannio, 82100 Benevento, Italy, and the ^** Department of Ophthalmology and Visual Sciences, Washington University School of Medicine, St. Louis, Missouri 63110

Received for publication, May 22, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Mice lacking retinol-binding protein (RBP) have low circulating retinol levels. They have severe visual defects due to a low content of retinol or retinyl esters in the eye. A transgenic mouse strain that expresses human RBP under the control of the muscle creatine kinase promoter in the null background was generated. The exogenous protein bound retinol and transthyretin in the circulation and effectively delivered retinol to the eye. Thus, RBP expressed from an ectopic source suppresses the visual phenotype, and retinoids accumulate in the eye. No human RBP was found in the retinal pigment epithelium of the transgenic mice, indicating that retinol uptake by the eye does not entail endocytosis of the carrier RBP.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Retinoids are needed to maintain normal growth and development, immunity, reproduction, vision, and other essential physiological process (1). Although not biologically active per se, retinol (vitamin A alcohol) is the major circulating form of vitamin A. Within tissues, retinol is oxidized to all-trans- and 9-cis-retinoic acids. These bind to nuclear receptors and regulate transcription of >300 diverse target genes (2-7) whose expression is influenced by retinoic acid availability (8, 9). In the eye, retinol is converted to 11-cis-retinal, the chromophore for the visual pigment opsin (10, 11).

The transport of retinol from liver stores to target tissues requires a specific 21-kDa transport protein, retinol-binding protein (RBP)¹ (12). Crystallographic analysis of the retinol·RBP complex from different species reveals a single retinol molecule buried within a highly conserved eight-strand -barrel structure (13). The retinol-binding site of RBP consists principally of amino acid residues coded by exons III-V. The genomic organization of the RBP gene is also highly conserved across species (12).

Retinol·RBP is found in a 1:1 molar complex with a 55-kDa protein, transthyretin (TTR) (14). Complex formation prevents retinol·RBP excretion by the kidney (12, 15). The mechanism through which tissues acquire retinol from the circulating retinol·RBP·TTR complex is subject to considerable debate. Membrane receptors for holo-RBP have been reported in retinal pigment epithelium (RPE), in placenta, and in several other tissues and cells (16-19). Such receptors could provide cells with a mechanism to regulate retinol uptake.

Most RBP is synthesized in hepatocytes. However, other adult organs and tissues have been reported to synthesize RBP, including, oddly, the eye (12). Expression of RBP in rat, bovine, and human RPE has been described (20-25). This RBP is believed to be secreted from the RPE into the interphotoreceptor matrix (26). Little is known about the function that RBP serves in the eye, one of the two organs (testis is the other) that are highly dependent on retinol (11, 27, 28).

We generated a mutant mouse that lacks RBP (RBP^/) by targeted disruption of the RBP genomic locus (29). RBP^/ mice, although viable and fertile, have reduced blood retinol and eye retinoid levels and markedly impaired retinal function during the first months of life. The visual impairment does not arise from abnormal eye development during embryogenesis. Given a retinoid-sufficient diet, the mutant mice acquire normal vision by 5 months of age, even though blood retinol levels remain low. Nevertheless, RBP is required for efficient eye uptake of retinol from the circulation, as shown by the near absence of retinyl ester reserves in the eyes of the RBP^/ mutants (29). Deprived of dietary vitamin A, vision remains abnormal, and blood retinol declines to undetectable levels. Our studies also indicate that the livers of RBP^/ mice acquire retinol normally from the diet and establish hepatic retinyl ester stores. However, these mice are unable to mobilize stored retinol from the liver efficiently. Inefficient uptake of dietary retinol into the eye and inability to mobilize hepatic retinol account for the visual phenotype of the mice.² Thus, the retinoid status of the eyes of RBP^/ mice is extremely tenuous and dependent on a regular vitamin A intake from the diet.

In this work, we further explore the role of RBP in assuring that the eye acquires and maintains normal vitamin A levels. Specifically, we investigated whether RBP expressed in an extrahepatic tissue can fulfill this role. Accordingly, we introduced a transgene that expresses human RBP (hRBP) under the control of the mouse muscle creatine kinase (MCK) promoter into the RBP null mouse. We found that circulating hRBP derived from muscle fully suppresses the visual defect of the RBP null mouse.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Generation of the MCK/hRBP Recombinant Vector-- A 3.3-kb PstI/ClaI DNA fragment containing the regulatory sequence of the MCK gene (30) was cloned into pBluescript. In the second step, a 1-kb EcoRI hRBP cDNA fragment (31) was filled in and blunt end-cloned into a filled ClaI site of the pBluescript/MCK vector. Subsequently, a 350-bp SacI fragment containing a bovine polyadenylation signal was filled in and blunt end-cloned into a filled XhoI site of the pBluescript/MCK/hRBP vector. The complete recombinant construct, encompassing 4.65 kb of DNA, was linearized with BssHII, isolated by electroelution, and purified with an Elutip D column (Schleicher & Schüll, Dasse, Germany).

Generation of Transgenic Mice-- The linearized complete recombinant construct (pBluescript/MCK/hRBP/poly(A)) was injected into the male pronucleus of fertilized eggs from superovulated (C57BL/6J × CBA/J)F₁ females that had been mated with males of the same genetic background. Microinjected eggs were transferred into the oviducts of surrogate females (32). Founder animals were bred with C57BL/6J mice, and three transgenic mouse lines were established.

DNA and RNA Analysis-- 10 µg of genomic DNA were prepared from mouse tails, digested with EcoRI or BamHI, and analyzed by Southern blotting (33). The integrated recombinant DNA was detected using, as probes, the hRBP cDNA (31), the EcoRI mRBP cDNA (29), and a 246-bp BamHI MCK fragment (30). Estimation of the integrated copies of the transgene was performed by Southern blotting and comparing the intensity (determined using a PhosphorImager) of the transgenic fragment with that of the fragment corresponding to the endogenous -actin locus. Tissue RNA isolation and Northern blotting were performed as described (33). The EcoRI hRBP cDNA fragment, which hybridizes to both mRBP and hRBP mRNAs, was used as probe. Because the transgenic hRBP mRNA is longer than endogenous mRBP due to the 350-bp SacI fragment of the bovine polyadenylation signal cloned into the recombinant construct, we were able to discriminate the two transcripts.

Protein Analysis-- Rabbit polyclonal anti-rat RBP serum (34) and rabbit polyclonal anti-rat TTR serum (35) were used for Western blot analysis performed according to standard procedures (33). Sheep polyclonal anti-rat RBP serum (36), rabbit polyclonal anti-hRBP serum (37), and rabbit polyclonal anti-rat TTR serum were used for radioimmunoassay performed as described (37).

HPLC Analysis-- Retinol and retinyl ester concentrations in plasma and tissues were measured by reverse-phase HPLC as described (15, 38). 11-cis-Retinal was measured in eye cup homogenates by normal-phase HPLC (39, 40).

Electroretinography-- Dark-adapted electroretinography was performed as previously described (29, 41). Dark-adapted ERG responses were obtained from anesthetized mice after their pupils were dilated with 1% phenylepinephrine HCl. A 30-gauge needle was placed subcutaneously on the forehead to serve as a reference electrode, and a ground electrode was placed subcutaneously on the trunk. A saline-moistened cotton wick recording electrode was positioned to contact the cornea. Stimulating light flashes were obtained with a stroboscope (Grass Instruments Inc.). Neutral density filters were placed in front of the aperture to vary the intensity of the flashes. Responses were detected using an oscilloscope and an evoked response-detecting computer in parallel (Nicolet Instruments CA-100). All mice were dark-adapted overnight before electroretinography was performed. Stimulation was begun with 4.8 log units of neutral density filtering, and the responses were averaged to one flash/s. At high flash intensity, each flash was repeated every 20 s, an interval that, in preliminary experiments, was found to be sufficiently long to exclude interference between flashes. The duration of one flash was nominally 10 µs.

Eye Immunohistochemistry-- 5-µm thick sections were collected on Polysine CTD slides (Fisher) and allowed to dry overnight at 37 °C before staining. Slides were deparaffinized in xylene and placed in 100% alcohol. Endogenous peroxidase was blocked with 3% H₂O₂ in methanol for 30 min. Sections were rehydrated through a graded series of alcohol washes. Endogenous melanin was bleached with 0.25% potassium permanganate solution for ~30 min (based on checking individual slides every 5 min for adequate bleaching). Sections were rinsed in distilled H₂O and placed in 5% oxalic acid until clear (5 s to 1 min). Slides were then washed in distilled H₂O. Antigen retrieval was performed using a citrate buffer system of 1.8 mM citric acid and 8.2 mM sodium citrate. Sections were submerged in this solution and placed in an electric pressure cooker (Decloaking Chamber, Biocare Medical) for 3 min. Sections were allowed to cool in solution for 30 min and then washed in distilled water. Slides were blocked with 20% normal donkey serum for 30 min. Sections were drained, and the primary antibody (37) diluted in Da Vinci Green antibody diluent (Biocare Medical) was added at a dilution of 1:100 to 1:500 overnight at 4 °C. Slides were washed with phosphate-buffered saline (4 × 15 min) and then incubated with 20% normal goat serum for 20 min. Secondary antibody (biotinylated goat-anti rabbit Ig; Vectastain ABC Elite kit, Vector Labs, Inc.) was prepared according to the kit instructions and incubated for 1 h at room temperature, followed by washing with phosphate-buffered saline (4 × 5 min). Slides were incubated in ABC reagent, prepared according to the kit instructions for 1 h. Slides were washed with phosphate-buffered saline (4 × 5 min). 3,3'-Diaminobenzidine was prepared according to kit instructions (Sigma Fast 3,3'-diaminobenzidine tablet sets), and slides were incubated with 3,3'-diaminobenzidine solution until a color reaction appeared. Slides were washed with phosphate-buffered saline for 15 min and then washed with distilled water. Slides were counterstained with hematoxylin, washed, dehydrated in a graded series of alcohols, cleared in xylene, and coverslipped with Cytoseal 60 (Richard Allan Scientific). Hematoxylin and eosin staining was performed according to standard procedures (40).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Generation of Transgenic Mice That Express hRBP from a Muscle-specific Promoter-- An MCK/hRBP recombinant vector was constructed from an EcoRI hRBP cDNA fragment (31) spanning the first to the sixth exon of the hRBP gene. To establish muscle-specific expression, the hRBP cDNA was fused to a 3.3-kb DNA fragment containing the promoter region of the MCK gene (Fig. 1A) (30). A linearized 4.65-kb BssHII fragment containing this construct (MCK/hRBP) was microinjected into pronuclei, and 46 newborn were obtained from pseudopregnant recipients. Three transgenic animals (MCK/hRBP2/4, MCK/hRBP2/5, and MCK/hRBP2/8) were obtained as founders, each of which gave rise to an independent line, as demonstrated by Southern blot analysis (data not shown). Because the three strains showed no marked differences in copy number of the integrated transgene or in their biochemical characteristics (see below), we chose one strain, MCK/hRBP2/4, for most subsequent studies. To detect DNA corresponding to the integrated recombinant construct, mouse genomic DNA was prepared from tail clips, digested with EcoRI, and hybridized with full-length hRBP cDNA (Fig. 1B) and with a 246-bp BamHI MCK fragment (Fig. 1C). These probes were used to screen the founder animals and their progeny (Fig. 1, B and C) for the presence of the transgene. DNAs from mice carrying the hRBP transgene and probed with the ³²P-labeled hRBP cDNA yielded a 4.65-kb band (derived from a partial digest) and a 1.35-kb band (derived from the hRBP cassette) (Fig. 1B). Using the ³²P-labeled MCK cDNA promoter fragment as probe, a 4.65-kb band (derived from a partial digestion) and a 3.3-kb band (derived from the MCK cassette) were obtained (Fig. 1C).

View larger version (45K): [in this window] [in a new window]   Fig. 1.   Generation of the hRBP+/+ mouse strain. A, map of the MCK/hRBP recombinant construct (not to scale). The positions of the relevant restriction sites are indicated. Filled-in restriction sites are indicated in italics. B and C, DNA analysis by Southern blotting. 10 µg of genomic DNA extracted from mouse tails were digested with EcoRI and hybridized either with the hRBP cDNA (B) or with a 246-bp BamHI MCK fragment (C). The molecular sizes of the expected bands are indicated to the right of each panel. The hRBP cDNA probe detected a band at 4.65 kb (from a partial digestion) and one at 1.35 kb (derived from the hRBP cDNA cassette) (B). The BamHI MCK probe detected a 4.65-kb band (from a partial digestion) and a 3.3-kb band (derived from the MCK cassette) (C). V ctr, EcoRI-digested MCK/hRBP recombinant vector.

Transgene Integration and Expression-- The band patterns described above indicated that the transgene was primarily arranged in head-to-tail arrays in the three different strains (data not shown). Southern blot analysis showed that each strain carried between 5 and 10 integrated copies of the transgene (data not shown). Northern blot analysis using hRBP cDNA as probe showed high concentrations of hRBP mRNA of the expected size (1350 nucleotides) in skeletal and cardiac muscle of the transgenic mice (Fig. 2A). Although the hRBP probe reacted with both hRBP and mRBP RNAs, the two species could be distinguished. The hRBP mRNA includes a 350-bp fragment containing the bovine polyadenylation signal and is thus longer than the mRBP mRNA (Fig. 2A). As expected, no hRBP mRNA signal was detected in non-transgenic control mice (Fig. 2A). A weak hRBP mRNA signal was detected in the eye. This signal may derive from contaminating muscle tissue because reverse transcription-PCR analysis using MCK-specific primers revealed muscle-specific RNA in our eye preparation (data not shown). Note that no hRBP mRNA was found in the liver, the predominant source of RBP expression in wild-type mice.

View larger version (47K): [in this window] [in a new window]   Fig. 2.   Analysis of mRNA and serum proteins. A, Northern blot analysis of total RNA (15 µg) extracted from mouse tissues performed with hRBP cDNA as probe (31). An hRBP+/+ transgenic mouse (+) and a control littermate () are shown. M, muscle; H, heart; E, eye; L, lung; A, adipose tissue; K, kidney. The positions of 18 S and 28 S mRNAs are indicated to the left. The molecular sizes of the expected bands are indicated to the right. B, analysis of the human protein in the plasma of different mouse strains by Western blotting. 5 µl of a 1:10 dilution of plasma from +/+, /, hRBP+/+ and hRBP/ mice were used. Rabbit polyclonal anti-rat RBP serum and rabbit polyclonal anti-rat TTR serum were used for immunodetection. 100-ng aliquots of rat RBP serum and hRBP serum purified to homogeneity were loaded as controls. The positions of protein markers are indicated to the left. Wild-type mRBP, hRBP, and the TTR monomer are indicated to the right.

The Transgene Expresses Functional hRBP-- To verify that hRBP produced in muscle and heart was secreted into peripheral blood, we immunoblotted blood samples from the hRBP^+/+ mouse strain (reported as MCK/hRBP (29)) with rabbit polyclonal anti-rat RBP serum. This serum cross-reacts with both endogenous mouse protein and exogenous human protein, which migrated more slowly on an SDS-12% polyacrylamide gel (Fig. 2B). Both mRBP and hRBP were seen in blood from hRBP^+/+ mice, whereas only mRBP was detected in the circulation of wild-type animals. As previously reported, RBP^/ mice have no circulating RBP (29). The levels of circulating TTR were equivalent in wild-type, RBP^/, and hRBP^+/+ mice (Fig. 2B). To quantify RBP levels in plasma, liver, muscle, and eye for the three transgenic strains (MCK/hRBP2/4, MCK/hRBP2/5, and MCK/hRBP2/8), radioimmunoassays were performed. We used rabbit polyclonal anti-hRBP serum, which interacts with hRBP, but not with mRBP. The levels of mRBP and mTTR were also determined with sheep polyclonal anti-rat RBP serum and rabbit polyclonal anti-rat TTR serum, respectively. No statistically significant differences among the three strains were observed (data not shown). High levels of hRBP were detected in blood (average of 10.1 mg/dl or 4.8 µM) and in muscle (average of 56.2 µg/g), suggesting specific expression and secretion of transgenic hRBP. No hRBP was detected in perfused liver, as expected. Plasma levels of mRBP and TTR in hRBP^+/+ mice were similar to those in wild-type mice (data not shown).

Liver and serum levels of retinol and retinyl esters were also determined for the MCK/hRBP2/4, MCK/hRBP2/5, and MCK/hRBP2/8 strains by HPLC. Table I shows the retinol and retinyl ester levels in serum and tissues of hRBP^+/+ transgenic mice (MCK/hRBP2/4) at 13 weeks of age compared with wild-type littermates. Serum retinol levels for hRBP^+/+ mice averaged 57 µg/dl compared with 18 µg/dl for wild-type mice, suggesting that circulating hRBP binds and transports retinol efficiently. Circulating hRBP was bound to mTTR in all the transgenic strains (29). Interestingly, retinol and retinyl ester levels were significantly increased in the muscle of the transgenic animals. An increase in the levels of retinol and retinyl esters was observed in other tissues tested. Note the relatively high total retinol levels in the eyes of the MCK/hRBP mice.

Introduction of the MCK/hRBP Transgene into RBP^/ Mice-- Having demonstrated that hRBP was physiologically functional, we investigated whether it could suppress the visual defect of the RBP knockout mice (29). Accordingly, we crossed the hRBP^+/+ and RBP^/ mutants to generate hRBP^/ mice in which the sole source of RBP was hRBP expressed from muscle. The RBP^/ mice were crossed with mice homozygous for the MCK/hRBP transgene to yield F₁ mice that were heterozygous for the mRBP knockout allele and the transgene. The F₁ mice were then crossed to give F₂ progeny, which were screened to identify homozygous mRBP null mice carrying the hRBP transgene (hRBP^/). Such mice were identified by Southern blot analysis of mouse tail genomic DNA digested with BamHI and hybridized with ³²P-labeled EcoRI mRBP cDNA. This probe recognized both the mRBP and hRBP genes (Fig. 3). The hRBP^/ mice showed the following band pattern: a doublet of a 7.5-kb band and a 7.0-kb band, indicating disruption of the endogenous mRBP gene, as described (29), and a 1.35-kb band corresponding to the transgene.

View larger version (89K): [in this window] [in a new window]   Fig. 3.   Generation of hRBP/ mice. Genomic DNA extracted from mouse tails was digested with BamHI and hybridized with mRBP cDNA (29). The genotypes of littermates are indicated at the top. The molecular sizes of the expected bands are indicated to the left. The hRBP/ mice showed the following band pattern: a doublet of a 7.5-kb band and 7.0-kb band arising from knockout (ko) of the endogenous mRBP gene (29) and a 1.35-kb band corresponding to the transgene. wt, wild-type.

Biochemical Characterization of the hRBP^/ Strain-- We analyzed the plasma RBP content of the hRBP^/ mice by Western blot analysis using rabbit polyclonal anti-rat RBP serum. This analysis clearly revealed that the only RBP present in the circulation of the hRBP^/ strain was hRBP (Fig. 2B). The concentration of TTR was not affected by the hRBP transgene, as shown by probing the blood samples with rabbit polyclonal anti-rat TTR serum (Fig. 2B).

We also measured the amount of hRBP and mTTR protein by radioimmunoassay in serum, muscle, and liver from eight male hRBP^/ mice at 14 weeks of age using rabbit polyclonal anti-hRBP serum and rabbit polyclonal anti-rat TTR serum (Table II). As expected, mRBP was detected in none of the tissues tested (data not shown). High levels of hRBP were detected in muscle and blood, consistent with muscle-specific expression of the transgene and secretion of hRBP into the circulation. Retinol and retinyl ester concentrations in tissues and serum from the hRBP^/ animals were determined by HPLC analysis (Table III). Tissue retinoid levels were comparable to those for the hRBP^+/+ strain (see Table I). Note the relatively high concentrations of serum retinol and of retinol and retinyl ester in muscle and eye.

View this table: [in this window] [in a new window]   Table II Serum levels of hRBP and mTTR in hRBP/ mice Serum levels of hRBP and mTTR were determined by radioimmunoassay and expressed as means ± S.D. n, number of 14-week-old male mice analyzed.

View this table: [in this window] [in a new window]   Table III Total retinol tissue levels in the hRBP/ strain Retinol levels were determined by reverse-phase HPLC and expressed as means ± S.D. n, number of 14-week-old male mice analyzed.

The Visual Defect of RBP^/ Mice Is Suppressed by Extrahepatic hRBP-- To demonstrate the role of RBP in the visual process, it was necessary to demonstrate that the eye phenotype of RBP^/ mice is due entirely to the absence of RBP. We therefore performed dark-adapted electroretinography on hRBP^/ mice at 4 weeks of age (29). The ERG profiles of these mice were indistinguishable from those of age-matched wild-type mice (Fig. 4A). This contrasts with the highly abnormal ERGs of the RBP^/ mice. Adult hRBP^/ mice also showed a normal ERG profile (data not shown), consistent with the observation that retinol (Fig. 4B), 11-cis-retinal, and retinyl ester (data not shown) levels were in the normal range in the eyes of hRBP^/ mice. Thus, hRBP expression entirely suppresses the visual impairment of the RBP^/ mice, confirming that this phenotype is the result of RBP insufficiency. Moreover, these experiments indicate that RBP derived from extrahepatic tissue can deliver retinol to the eye.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Analysis of retinal function and retinol eye cup levels of hRBP/ mice. A, corneal ERG responses to light stimuli of 4-week-old /, +/+, and hRBP/ mice that had been dark-adapted overnight. The scale indicates 150 mV vertically and 50 ms horizontally. The relative logarithmic light intensity of each flash is indicated to the left. B, eye cup levels of retinol for 13-week-old +/+, /, and hRBP/ mice. Whole eye homogenates were analyzed by reverse-phase HPLC to determine the levels of retinol in the eye cups of four /, five +/+, and five hRBP/ age- and sex-matched mice. Error bars indicate S.E.

hRBP Is Not Found in the RPE of hRBP^/ Mice-- We next investigated whether retinol delivery to the eye by RBP entails uptake of the carrier protein. Accordingly, we looked for hRBP in the eyes of wild-type, RBP^/, and hRBP^/ mice. Fig. 5 presents the results of an immunohistochemical analysis using rabbit polyclonal anti-hRBP serum. No hRBP could be detected in the RPE of hRBP^/ mice. Staining of the photoreceptor layer was probably an artifact due to cross-reaction of the antibody with purpurin, which shares epitopes with RBP (42). Thus, circulating hRBP efficiently delivers retinol to the eye without concomitant endocytosis of the retinol·RBP complex. Furthermore, these results imply that RBP expressed in the RPE is not essential for maintaining vision.

View larger version (43K): [in this window] [in a new window]   Fig. 5.   Eye immunohistochemistry. Immunostaining of the RPE of +/+, /, and hRBP/ mice with rabbit polyclonal anti-hRBP serum. The first panel shows hematoxylin and eosin (H&E) staining of a 5-µm section from a wild-type eye. Background staining due to cross-reaction of the antibody with purpurin (42) can be seen in the photoreceptor layer of all mice analyzed. GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer; OS, outer segment.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The visual cycle is driven by retinal derived from circulating retinol (10, 43). Retinol is water-insoluble and is carried in blood dissolved in postprandial chylomicron remnants as retinyl esters or bound to a specific carrier protein, RBP (12, 43). RBP is expressed principally in the liver, the major body site of retinol storage, from which it is secreted bound to retinol. However, the synthesis of RBP in other tissues, notably the RPE, has been reported (12, 20-25). The role of RBP of extrahepatic origin is not known.

In this work, we have shown that hRBP expressed ectopically in muscle can deliver retinol to the eye. This was demonstrated using a genetically engineered mouse strain that lacks functional RBP (29) and that carries a transgene that expresses hRBP from the MCK promoter. As expected, the major tissue sites of hRBP expression are skeletal muscle and heart. No hRBP is expressed in the liver (Fig. 2A).

Unlike RBP^/ animals, hRBP^/ mice do not show visual defects in the first months of life, as determined by electroretinography, and have copious optic stores of retinol and retinyl esters (Fig. 4). The amount of hRBP secreted into the bloodstream is very high; and correspondingly, the levels of serum retinol are 3-fold higher than in wild-type animals and 30-fold higher than in RBP null mice. The high concentrations of retinol may account for the suppression of the visual defect of RBP^/ animals. However, we have not excluded a more direct role of RBP in the delivery of retinol to the eye. Interestingly, we noted that the eye takes up retinol bound to chylomicron remnants very poorly in comparison to other tissues² and may rely on a mechanism involving a receptor for RBP. Such a receptor has been reported for the eye (19) and other tissues (16-18). We found, however, no detectable hRBP in the RPE of hRBP^/ mice (Fig. 5). Thus, the delivery of retinol to the eye does not entail endocytosis of a retinol·RBP complex.

Retinol·RBP is secreted from the liver in association with TTR, a serum protein that prevents renal filtration of the RBP complex (12, 44). Binding between holo-RBP and TTR occurs within the hepatocyte (44, 45). Detectable levels of TTR mRNA have been reported in rat skeletal muscle (46). Thus, the appearance in the blood of hRBP derived from muscle could be the result of a co-secretion process. However, the elevated concentration of hRBP in the muscle of transgenic animals (see Table II) may circumvent a requirement for co-secretion with TTR.

Note that the muscle of the hRBP^/ animals also contains substantially greater levels of retinoids than wild-type muscle (see Tables I and III). Because muscle is not a depot for retinoids in wild-type mice, this accumulation must be facilitated by hRBP present in this tissue. Neither the subcellular location of this retinoid nor the protein to which it is complexed is known.

Among the tissues where the synthesis of extrahepatic RBP has been shown, the eye is one of the most difficult to understand. RBP is expressed in the RPE and from there secreted into the interphotoreceptor matrix (26). Little is known about the function that RBP serves in the eye. Our data indicate that circulating hRBP, derived from muscle, efficiently delivers vitamin A to the eye. The hRBP^/ animals have wild-type eye retinoid levels and do not display the visual impairment of RBP^/ mice. Thus, RBP expressed in the eye appears to play no role in maintaining the visual cycle.

The data reported in this study leave open the issue of how the retinol·hRBP complex is taken up by the RPE. We found no hRBP in the RPE of the transgenic animals. Thus, if there is an RBP receptor, it does not function through an endocytic pathway. We do not believe, however, that hRBP acts only to keep circulating retinol levels high. Our experimental evidence suggests that the eye has a special affinity for retinol·RBP.² Efforts to elucidate the role of RBP in the delivery of retinol to the eye and to identify a possible RBP receptor are ongoing.

	ACKNOWLEDGEMENT

We thank Dr. J. L. Breslow for the mouse creatine kinase promoter plasmid.

	FOOTNOTES

^* This work was supported by Grants R01 EY12858 and R01 DK52444 from the National Institutes of Health and Grant 9900693 from the United States Department of Agriculture.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed. Tel.: 212-305-5429; Fax: 212-305-2801; E-mail: wsb2@columbia.edu.

Published, JBC Papers in Press, June 4, 2002, DOI 10.1074/jbc.M205046200

² S. Vogel, R. Piantedosi, S. M. O'Byrne, Y. Kako, L. Quadro, I. J. Goldberg, M. E. Gottesman, and W. S. Blaner, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: RBP, retinol-binding protein; hRBP, human RBP; mRBP, mouse RBP; RBP^/ mice, RBP knockout mice; hRBP^+/+ mice, hRBP-overexpressing mice in the wild-type background; hRBP^/ mice, hRBP-overexpressing mice in the mRBP knockout background; TTR, transthyretin; mTTR, mouse transthyretin; RPE, retinal pigment epithelium/epithelia; MCK, muscle creatine kinase; HPLC, high performance liquid chromatography; ERG, electroretinogram.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				C. Redondo, M. Vouropoulou, J. Evans, and J. B. C. Findlay Identification of the retinol-binding protein (RBP) interaction site and functional state of RBPs for the membrane receptor FASEB J, April 1, 2008; 22(4): 1043 - 1054. [Abstract] [Full Text] [PDF]

				A. C. Ross and N.-q. Li Retinol Combined with Retinoic Acid Increases Retinol Uptake and Esterification in the Lungs of Young Adult Rats when Delivered by the Intramuscular as well as Oral Routes J. Nutr., November 1, 2007; 137(11): 2371 - 2376. [Abstract] [Full Text] [PDF]

				L. Quadro, W. S. Blaner, L. Hamberger, P. M. Novikoff, S. Vogel, R. Piantedosi, M. E. Gottesman, and V. Colantuoni The role of extrahepatic retinol binding protein in the mobilization of retinoid stores J. Lipid Res., November 1, 2004; 45(11): 1975 - 1982. [Abstract] [Full Text] [PDF]

				L. Quadro, L. Hamberger, M. E. Gottesman, V. Colantuoni, R. Ramakrishnan, and W. S. Blaner Transplacental delivery of retinoid: the role of retinol-binding protein and lipoprotein retinyl ester Am J Physiol Endocrinol Metab, June 1, 2004; 286(5): E844 - E851. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/33/30191    most recent M205046200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Quadro, L. Articles by Gottesman, M. E. Search for Related Content PubMed PubMed Citation Articles by Quadro, L. Articles by Gottesman, M. E. Social Bookmarking                      What's this?