Medline/PubMed Citation | Related Articles in PubMed | Download to Citation Manager

Biochem. J. (2001) 354 (521–529) (Printed in Great Britain)

Abstract

PDF

Other papers of interest

Papers by these authors

Expression pattern and localization of b,b-carotene 15,15´-dioxygenase in different tissues

Adrian WYSS*¹, Gabriele M. WIRTZ†, Wolf-D. WOGGON†, Roland BRUGGER*, Markus WYSS*, Arno FRIEDLEIN‡, Georges RISS*, Heinrich BACHMANN* and Willi HUNZIKER*

*F. Hoffmann–La Roche Ltd., Vitamins & Fine Chemicals Division, 4070 Basel, Switzerland, †Institute of Organic Chemistry, University of Basel, St. Johannsring 19, CH-4056 Basel, Switzerland, and ‡F. Hoffmann–La Roche Ltd., Roche Genetics, 4070 Basel, Switzerland

b,b-Carotene 15,15´-dioxygenase cleaves b,b-carotene into two molecules of retinal, and is the key enzyme in the metabolism of b,b-carotene to vitamin A. The enzyme has been known for more than 40 years, yet all attempts to purify the protein to homogeneity have failed. Recently, the successful cloning and sequencing of an enzyme with b,b-carotene 15,15´-dioxygenase activity from chicken, as well as from Drosophila, has been reported. Here, we describe in detail our attempt to enrich the chicken b,b-carotene 15,15´-dioxygenase to such an extent as to allow determination of partial amino acid sequences, which were then used to design degenerate oligonucleotides. Screening of a chicken duodenal expression library yielded a full-length clone containing a coding sequence of 1578bp. Functional expression in Escherichia coli and in eukaryotic cell lines confirmed that we had cloned the first vertebrate dioxygenase that cleaves b,b-carotene at the central 15,15´-double bond. By performing a sequence homology search, the cDNA sequence of the mouse homologue was found as an expressed sequence tag (EST) in the gene bank. At the amino-acid level, the degree of homology between the chicken and mouse sequences is 81%. Thus b,b-carotene 15,15´-dioxygenase can be considered as being an enzyme that is evolutionarily rather well conserved. We established the expression pattern of b,b-carotene 15,15´-dioxygenase in chicken and mouse tissues with a combination of Northern blots and in situ hybridization. The mRNA for b,b-carotene 15,15´-dioxygenase was localized primarily in duodenal villi, as well as in liver and in tubular structures of lung and kidney. These new findings demonstrate that b,b-carotene 15,15´-dioxygenase is also expressed in epithelial structures, where it serves to provide the tissue-specific vitamin A supply.

Key words: central cleavage, epithelial cell expression, retinal, vitamin A.

Abbreviations used: AP, alkaline phosphatase; BHK, baby hamster kidney; BHT, butylated hydroxytoluene; DTT, dithiothreitol; EST, expressed sequence tag; IEC, ion-exchange chromatography; poly(A)⁺, polyadenylated; RT, reverse transcriptase; TBME, t-butyl methyl ether.

¹To whom correspondence should be addressed (e-mail adrian.wyss@roche.com).

INTRODUCTION  

Carotenoids are a large family of natural pigments [1,2], of which 10% display provitamin A activity, b,b-carotene being the most important one for animal and human nutrition. Carotenoids serve two essential functions: first, as accessory pigments in the light-harvesting system of photosynthetic organisms [3], and, secondly, in photoprotection from light of short wavelength [4,5]. The characteristic polyene structure of carotenoids is responsible for light absorption, as well as for singlet oxygen quenching or inactivation of 'aggressive' free radicals [6,7]. In the human diet, it has been estimated that up to 80% of the daily vitamin A intake is derived from the provitamin A carotenoids in fruits and vegetables. Malnutrition in African and Asian countries leads to severe vitamin A deficiency among children [8]. Night blindness and xerophthalmia, as well as a higher susceptibility to infectious diseases, are the result. The benefits of vitamin A and b,b-carotene supplementation to combat vitamin A deficiency have been demonstrated in several studies [9–11,11a].

The b,b-carotene 15,15´-dioxygenase (EC 1.13.11.21) cleaves b,b-carotene into two molecules of retinal, and is the key enzyme in metabolism of carotene to vitamin A. Other carotenoids containing at least one unsubstituted b-ionone ring and a polyene chain attached to it also serve as substrates for the enzyme and, therefore, as precursors of retinoid formation [12]. However, using b,b-carotene as substrate resulted in the highest specific activity when various carotenoids and apo-carotenals were compared [13].

The enzyme was first isolated in the mid-1960s [14,15]. Subsequent experiments suggested a dioxygenase reaction mechanism [16], and the presence of a cofactor containing iron [17,18]. Thus the enzyme was termed carotene dioxygenase, although sound evidence for the detailed reaction mechanism is still lacking. Since then, many attempts have been made to purify and characterize this enzyme by biochemical means [15,17–21]. Experiments by Goodman et al. [16,19] and Fidge et al. [17] showed that b,b-carotene 15,15´-dioxygenase is a soluble cytosolic enzyme. Sklan [22] reported that enzyme activity is associated with a high-molecular-mass lipid–protein aggregate fraction. Highest dioxygenase activity was found in intestinal mucosa and in jejunal enterocytes, with a gradient of decreasing activity from the duodenum/jejunum to the colon [23,24]. In liver, lung, kidney and brain, b,b-carotene 15,15´-dioxygenase activity was also detected [25]. In addition, the mode of cleavage of b,b-carotene has been discussed controversially for many years [26]. Although Glover proposed 'excentric' cleavage as long ago as 1960 [27], the central cleavage pathway was widely accepted as the only metabolic pathway until Wang et al. [28–30] provided evidence for production both in vivo and in vitro of apo-carotenals and retinoic acid as main products of b,b-carotene cleavage. Nagao et al. [31] found stoichiometric formation of retinal with a ratio of 1.9:1, thus presenting evidence for the central cleavage pathway.

Recently, two reports describing the cloning of b,b-carotene 15,15´-dioxygenase have been published [32,33]. von Lintig and Vogt [32] cloned an enzyme displaying b,b-carotene 15,15´-dioxygenase activity from Drosophila melanogaster, while our group [33] published on the cloning of the chicken enzyme. In the present article, we describe in detail our attempts to purify b,b-carotene 15,15´-dioxygenase from extracts of chicken intestinal mucosa, to determine partial amino acid sequences, and to use this information to obtain cDNA sequences for the chicken and mouse b,b-carotene 15,15´-dioxygenase. In addition, we describe both the levels and pattern of expression in various chicken and mouse tissues.

EXPERIMENTAL  

Enrichment of b,b-carotene 15,15´-dioxygenase from chicken intestinal mucosa [14]

Upon preparation, the duodenum samples were kept on ice or at 4°C until further use. For isolation of mucosa, chicken (strain LSL Lohmann) intestines were thawed and opened in a Petri dish. The mucosa was scraped off, weighed and suspended in 4ml of buffer A [100mM KH₂PO₄/4mM MgCl₂/30mM nicotinamide (pH7.8), containing 4% protease-inhibitor cocktail (5.9mM benzamidine/HCl, 10mM 6-aminohexanoic acid and 5µM soya-bean trypsin inhibitor)] perg of mucosa. The suspension was homogenized with a Potter–Elvehjem homogenizer. After centrifugation at 62000g for 1h, the supernatant was subjected to 20–45% ammonium sulphate fractionation. The pellet obtained was resuspended in buffer B [10mM KH₂PO₄/1mM GSH/0.5M ammonium sulphate (pH7.8)] and loaded on a HiLoad 26/10 Phenyl Sepharose High Performance column (Amersham Pharmacia Biotech, Uppsala, Sweden). Proteins were eluted with a linear gradient with buffer C [10mM KH₂PO₄/1mM GSH/10% (v/v) glycerol (pH7.8)], and activity-containing fractions with a conductivity of < 1mS/cm were loaded on a 30ml Blue Sepharose 6 Fast Flow column (Amersham Pharmacia Biotech) equilibrated with buffer C. Dioxygenase activity was recovered in the breakthrough fractions, which were loaded on a 20ml Poros HQ/M column (PerSeptive Biosystems, Framingham, MA, U.S.A.), also equilibrated with buffer C. b,b-Carotene 15,15´-dioxygenase was eluted with a linear gradient of buffer D [10mM KH₂PO₄/1mM GSH/0.5M NaCl/10% (v/v) glycerol (pH7.8)] and recovered in fractions with a conductivity of 10–20mS/cm. The combined activity-containing fractions were concentrated 40–50-fold in Ultrafree-15 filter units (Millipore, Bedford, MA, U.S.A.), and aliquots of 500µl were passed over a Superdex 200 HR 10/30 column (Amersham Pharmacia Biotech), with buffer E [50mM KH₂PO₄/1mM GSH/150mM NaCl/10% glycerol (pH7.8)] as the eluent.

Amino acid sequencing

Multiple aliquots of activity-containing fractions from the final gel-filtration run were separated on a 10% Tris/glycine gel (Novex, Invitrogen Corp., Carlsbad, CA, U.S.A.). The two protein bands that correlated most closely with b,b-carotene 15,15´-dioxygenase activity were subjected directly to Edman sequencing, or were excised and digested in-gel with trypsin. The tryptic digest was separated by micro-bore reversed-phase HPLC on a 150mm×1.0mm Vydac C18 column (Vydac, Hesperia, CA, U.S.A.). Peptides were eluted with an acetonitrile gradient in 0.1% (v/v) trifluoroacetic acid, and peptide-containing fractions were collected for further analysis by matrix-assisted laser-desorption ionization–time of flight mass spectrometry ('MALDI–TOF-MS') (Voyager Elite, PerSeptive Biosystems). Two fractions contained one single peptide each, which were sequenced by Edman degradation.

Construction of a chicken duodenal cDNA library

Chickens were killed at 22 weeks of age. The duodenum was removed, washed and cut open. The mucosal layer was scraped off, weighed and homogenized immediately in Trizole reagent (Life Technologies, Rockville, MD, U.S.A.). For total RNA isolation, the standard protocol from Life Technologies was followed. Polyadenylated [poly(A)⁺] RNA was isolated with the polyATract mRNA isolation kit from Promega (Madison, WI, U.S.A.).

cDNA was prepared from chicken duodenal poly(A)⁺ RNA with the Copy Kit (Invitrogen) using a polyT/NotI primer and a modified Gubler–Hoffman procedure. BstXI linkers were ligated, the cDNA size was selected (0.9–5.5kb) and subsequently cloned into the BstXI/NotI site of the eukaryotic expression vector pcDNA1.1/Amp (Invitrogen). Electroporation into Escherichia coli Top10 was accomplished with a Gene Pulser II system (Bio-Rad Laboratories, Hercules, CA, U.S.A.), and this resulted in the generation of a cDNA library of 430000 individual clones.

PCR and reverse transcriptase (RT)-PCR

Two degenerate primers, the degeneracy of which was reduced by using inosine (I), were designed from the peptide sequence Asn-Lys-Glu-Glu-His-Pro-Glu-Pro-Ile-Lys-Ala-Glu-Val-Gln-Gly-Gln-Leu-Pro: 5´-primer, 5´-AAC AAR GAR GAS CAY CCI GA-3´ (20-mer with a degeneracy of ×16); 3´-primer, 5´-SAG CTG ICC CTG IAC YTC SGC-3´ (21-mer with a degeneracy of ×8), where R = A or G, S = C or G, and Y = C or T.

For PCR, 100ng of chicken duodenal cDNA were taken as the template, and the following PCR steps were performed (with 40 cycles): incubation at 94°C for 30s; at 52°C for 30s; and at 72°C for 1min. The resulting 51bp fragment was excised from a 10% TBE/polyacrylamide gel (where 1×TBE = 45mM Tris/borate/1mM EDTA), electroeluted, cloned into pGEM-T Easy (Promega), and then sequenced. From this DNA sequence, a homologous forward primer was derived: 5´-TCTCGGAGCCCATAAAAGC-3´ ('dioxy12'). At the 5´-end an EcoRI site (underlined sequence) was included; the subsequent 17nt are homologous with the previously determined dioxygenase sequence. RT-PCR was performed with a polyT/NotI reverse primer (Invitrogen) and with forward primer dioxy12 by using the Titan One Tube RT-PCR kit from Roche Molecular Biochemicals (Roche Diagnostics GmbH, Mannheim, Germany), according to the manufacturer's instructions, and by applying the following cycles: 50°C for 30s, then 94°C for 2min; 10 cycles at 94°C for 30s, 57°C for 30s and 68°C for 45s; then 25 cycles at 94°C for 30s, 62°C for 30s and 68°C for 45s+3s/cycle; additional extension at 68°C for 7min, then at 4°C overnight. The resulting 597bp cDNA fragment was purified on an agarose gel, isolated with the QIAquick gel extraction kit (Qiagen, Hilden, Germany) and sequenced.

Screening of the chicken duodenal cDNA library

Pools from the original library were plated on agar. Corresponding master and replica filters (nylon membranes; Gene Screen, NEN Life Science Products, Inc., Boston, MA, U.S.A.) were processed and hybridized with the radioactively labelled ([a-³²P]dCTP; Amersham Pharmacia Biotech) 597bp PCR fragment using 500000c.p.m./ml Church–Gilbert hybridization solution. Filters were washed with 2×SSC (where 1×SSC is 0.15M NaCl/0.015M sodium citrate) at room temperature for 20min, and then with 0.1×SSC/0.1% (w/v) SDS at room temperature for 20min, followed by exposure to X-OMAT AR films (Eastman Kodak, Rochester, NY, U.S.A.). Double-positive clones were identified, picked and sequenced on an automated fluorescence-based Vistra DNA sequencer 725 (Amersham Pharmacia Biotech).

Functional expression in eukaryotic cell lines with the Semliki Forest Virus system

The coding sequence of the b,b-carotene 15,15´-dioxygenase cDNA was amplified by PCR, using, as the 5´-primer, 5´-GAGCTAGAACGATGGAGACAATATTTAACAGAAAC-3´ (where the BglII site is underlined) and as the 3´-primer, 5´-CAAGTTTATTCCGTCTCAGCCCCCAAATCATTC-3´ (where the XhoI site is underlined and the His₆ tag is double-underlined). The resulting fragment of 1608bp was cloned into the BamHI/XhoI site of the plasmid pSFV₂gen (a gift from Dr K. Lundstrom, F. Hoffmann–La Roche Ltd, Basel, Switzerland). This plasmid was used for in vitro synthesis of recombinant RNA by combining 2.5µg of linearized DNA with 5µl of 10×SP6 buffer [400mM Hepes/KOH (pH7.4)/60mM magnesium acetate/20mM spermidine/HCl], 5µl of 10mM CAP [m7G(5´)ppp(5´)G; Amersham Pharmacia Biotech], 5µl of 50mM dithiothreitol (DTT), 5µl of rNTP mix (10mM ATP, 10mM CTP, 10mM UTP and 5mM GTP), 1.5µl of RNase inhibitor (50units/µl; Roche Molecular Biochemicals), 3.5µl of SP6 RNA polymerase (30units/µl; Amersham Pharmacia Biotech) and 20µl of water. Incubation was performed at 37°C for 60min. Subsequently, a Gene Pulser (Bio-Rad) was used to electroporate the RNA, together with a helper virus RNA, into baby hamster kidney (BHK) cells. The cells were incubated for 24h at 37°C/5% CO₂/95% humidity, before the medium was filtered through a 0.22µm membrane (Millex-GP; Millipore). This high-titre virus stock was activated with chymotrypsin (Roche Molecular Biochemicals), and was used to infect the mammalian cell lines HuTu80, CHO (Chinese Hamster Ovary cells) and BHK. After adding protease-inhibitor cocktail, cell extracts were prepared by two cycles of freezing and thawing, and by sonicating the suspension for 5min.

Dioxygenase activity assay

All work with carotenoids and retinoids was performed in the absence of light and under nitrogen. Measurements of activity were determined according to the method of During et al. [25,34].

Sample preparation

To the assay buffer (0.15mM Tricine, pH8.0), 3.0mg/ml of GSH and 0.21mg/ml sodium cholate were added before use. The protein pellets [(NH₄)₂SO₄ precipitates from intestinal mucosa] were thawed on ice, and then resuspended in cold assay buffer by gentle vortex-mixing at a protein concentration of approx. 10mg/ml. The suspension was centrifuged at 20000g for 10min at 4°C, and aliquots of the supernatant were used for the incubation with the substrate. Fractions of recombinant protein were dialysed against assay buffer containing 0.5mM FeSO₄, before activity was determined.

Substrate preparation

Before each assay, b,b-carotene was purified over an Alox (CAMAG Ltd, Muttenz, Switzerland) column: 1.0g of Alox (inactivated with 100µl of water) was poured into a Pasteur pipette and washed with hexane. Of the b,b-carotene stock solution [26.9mg of all-trans-b,b-carotene and 5µl of a-tocopherol stock solution (10µmol/ml in hexane) dissolved in 5.0ml of benzene], 40µl was poured on to the top of the column, and the hexane wash was initiated. The coloured band was eluted into a 5ml Pyrex vial containing 50µl of a-tocopherol stock solution and 200µl of Tween 40 stock solution (400µl in 10ml of acetone). The solvent was evaporated with N₂ at 45°C. Water (1ml at 37°C) was added, and the substrate was solubilized with gentle vortex-mixing. The solution was kept at 37°C in the dark to avoid crystallization. The concentration of the substrate solution was determined at 452nm in propan-2-ol (e_mol 118300M^-1·cm^-1).

Each sample was assayed in duplicate. In each assay series, a blank sample and a micellar solution of retinal were included as controls. Assay buffer [0.15mM Tricine (pH8.0), including 3.0mg/ml of GSH and 0.21mg/ml sodium cholate] (200µl) and 100µl of enzyme solution ( 1mg of protein) were placed in an Eppendorf tube and pre-incubated at 37°C for 5min. The reaction was started by addition of 75µl ( 18.8nmol) of the b-carotene substrate solution described above. After incubation for 60min at 37°C in the dark with shaking (1s^-1), the reaction was stopped by adding 25µl of formaldehyde. After an additional incubation period of 10min, 10µl of butylated hydroxytoluene (BHT) (0.1mg/ml in ethanol) was added as protectant to each tube, as well as 10µl of a Ro11–1430 [C₂₃H₃₁NO₂; all-trans-N-ethyl-9-(4-methoxy-2,3,6-trimethylphenyl)-3,7-dimethyl-2,4,6,8-nonatetraenamide] retinoid stock solution (0.213mg/ml in ethanol; F. Hoffmann–La Roche Ltd) as an internal standard. Each tube was vortex-mixed, and for detergent removal, transferred quantitatively to a 1ml Extralut column (Merck, Dietikon, Switzerland). The column was rinsed with 250µl of a NaCl solution (0.4g/ml), and was then flushed with 3.0ml of t-butyl methyl ether (TBME), followed by a second elution with 2.5ml of TBME. Solvents were evaporated under a gentle stream of N₂ at 37°C.

Of extraction mixture [ethanol/tetrahydrofurane/TBME (9:1:5, by vol.), including 0.025% BHT], 200µl was added and the sample was mixed while slowly turning the tube. A 10µl aliquot was then analysed on a reversed-phase C₁₈ column [Zorbax ODS, Stagroma, Reinach, Switzerland; 250mm×4mm; 1.5ml/min; isocratic elution with a mixture of methanol and 80mM ammonium acetate in water (9:1, v/v)]. Retinal was detected at 380nm, whereas retinol was determined with a fluorescence detector (excitation at 325nm, emission at 470nm). In case analysis of b-apocarotenal was required, an aliquot of the sample was additionally injected into a Vydac 218TP54, 250mm×4mm HPLC column [Paul Bucher AG, Basel, Switzerland; 1.5ml/min; elution with acetonitrile/TBME/(80mM ammonium acetate in water)/triethylamine, 73:20:7:0.05, by vol.]. A UV–visible light detector was used for detection at 450nm.

Northern blot analysis

Total RNA from chicken duodenum was separated on a 1% formaldehyde/agarose gel in 1×Mops buffer. The RNA gel was vacuum-blotted with 20×SSC on a Zeta-Probe membrane (Bio-Rad) for 4h. The membrane was cross-linked with a UV-Stratalinker (Stratagene, La Jolla, CA, U.S.A.) and pre-hybridized for 1h in 50% (v/v) formamide, 5×SSC, 5×Denhardt's solution (where 1×Denhardt's solution is 0.02% Ficoll 400/0.02% polyvinylpyrrolidone/0.002% BSA), 1% SDS and 100µg/ml denatured herring sperm DNA.

As a probe, a part of the coding sequence of b,b-carotene 15,15´-dioxygenase was amplified by PCR (5´-primer, 5´-GGTACTTCAATTGTTGATAAAGG-3´; 3´-primer, 5´-TTCTGTTGCATAGACATACTTG-3´) and purified over a S-400 spin column (Amersham Pharmacia Biotech). Denatured DNA (25ng) was labelled with 50µCi of [a-³²P]dCTP (6000Ci/mmol; Amersham Pharmacia Biotech) using the High Prime random labelling kit from Roche Molecular Biochemicals. The labelled DNA probe was purified over a S-200 spin column (Amersham Pharmacia Biotech). Hybridization was performed overnight at 42°C with 7×10⁵c.p.m./ml. Filters were washed with 2×SSC for 15min at room temperature, with 0.1×SSC/0.1% SDS for 15min at room temperature, and, when required, with 0.1×SSC/0.1% SDS for 15min at 65°C, and were then exposed overnight on X-OMAT AR films (Eastman Kodak) at -80°C with double intensifying screens.

The mouse multi-tissue Northern blot was purchased from Clontech. The probe was labelled as described above. Hybridization was performed using 1×10⁶c.p.m./ml in ExpressHyb hybridization solution (Clontech), according to the protocol supplied.

In situ hybridization

Chickens at an age of 18 weeks were fed with a low vitamin A diet for 4 weeks. Before killing (4 days), the animals were given daily 20µmol of the RARa antagonist Ro41–5253 (C₂₈H₃₆O₅S; p-{(E)-2-[3´,4´-dihydro-4´,4´-dimethyl-7´-(heptyloxy)-2´H-1-benzothiopyran-6´-yl]propenyl}benzoic acid 1´,1´-dioxide; F. Hoffmann–La Roche Ltd) in order to up-regulate the b,b-carotene 15,15´-dioxygenase mRNA. All in situ hybridization experiments were performed at frimorfo Ltd (Fribourg, Switzerland) according to the method of Fleming et al. [35]. Tissues were fixed immediately in 4% paraformaldehyde, and stored in a 5% sucrose/PBS solution for cryosections. The tissues were then frozen at -20°C, and sectioned at 10µm using a Microm HN 505M Cryostat.

For preparation of the hybridization probe, a 1120bp PstI fragment from the coding region of the cDNA was subcloned into pBSKII (Stratagene), and both the antisense and the sense RNA probes were synthesized using the in vitro transcription kit from Roche Molecular Biochemicals. Hybridization of sections was performed overnight at 50°C with 400ng of digoxygenin-labelled RNA probe per ml of hybridization buffer [300mM NaCl/10mM Tris/HCl (pH7.5)/1mM EDTA/1×Denhardt's solution/10% dextran sulphate/30mM DTT/50% formamide/200µg/ml tRNA/200µg/ml single-stranded DNA]. After a first wash step with 4×SSC (2×30min), the sections were treated with RNase in 0.5M NaCl/10mM Tris/HCl, pH7.5/1mM EDTA. Washing steps were performed consecutively with 2×SSC, 0.1×SSC/1mM DTT and 0.1% Tween 20 in PBS. Incubation with an a-digoxygenin/alkaline phosphatase (AP) antibody (1:500 dilution) was performed for 2h in blocking buffer (10% BSA/0.1% Tween 20 in PBS). The slides were rinsed with 0.1% Tween 20/PBS and then incubated with 5% levamisole in AP buffer (100mM Tris/HCl, 100mM NaCl, 5mM MgCl₂, pH9.5) for 5min. Incubation with the AP substrate Nitro-Blue Tetrazolium chloride/5-bromo-4-chloro-3´-indoly-phosphate p-toluidine salt ('NBT-BCIP') was performed overnight in AP buffer containing 5mM levamisole. Finally, the slides were washed with 96% ethanol until the background staining disappeared.

RESULTS  

Enrichment of b,b-carotene 15,15´-dioxygenase, identification of peptide sequences, and cloning and sequencing of its cDNA

Starting with a cytosolic fraction from chicken intestine, we have used a fast purification protocol encompassing ammonium sulphate fractionation, hydrophobic interaction chromatography, Blue Sepharose chromatography, ion-exchange chromatography (IEC) and gel filtration. Hydrophobic interaction chromatography, Blue Sepharose chromatography and IEC allowed a 226-fold enrichment of enzyme, with 45% recovery, yielding a preparation with a specific activity of 2.5nmol·h^-1·mgprotein^-1 (Table 1). However, gel filtration and SDS/PAGE of the concentrated IEC eluate demonstrated that the final preparation was far from being pure.

Table 1 Summary of the various steps in the purification of b,b-carotene 15,15´-dioxygenase from three chicken intestines

Shown are the means for 3–4 measurements.

Since we failed to achieve a better standard of purification at that time, an indirect approach had to be chosen in order to obtain amino acid sequence information concerning b,b-carotene 15,15´-dioxygenase. When the concentrated IEC eluate was subjected to gel filtration, b,b-carotene 15,15´-dioxygenase activity was eluted at a molecular mass of approx. 50kDa. SDS/PAGE analysis of the fractions revealed that out of 15 discrete protein bands, only two correlated reasonably well with b,b-carotene 15,15´-dioxygenase activity (see [33]). Edman degradation of the lower band yielded a sequence of 19 amino acids demonstrating homology with a thioesterase ([36]; GenBank/EMBL accession no. L05493) that was considered to be unlikely to display b,b-carotene 15,15´-dioxygenase activity. Since the second band was N-terminally blocked, the protein was excised from a gel and subjected to in-gel trypsin digestion. Upon separation of the tryptic peptides by micro-bore reversed-phase HPLC, the following two overlapping sequences were obtained by Edman degradation: (1) Ala-Glu-Val-Gln-Gly-Gln-Leu-Pro; and (2) Asn-Lys-Glu-Glu-His-Pro-Glu-Pro-Ile-Lys-Ala-Glu-Val-Gln-Gly-Gln-Leu-Pro.

Degenerate oligonucleotides were derived from these sequences and were used for PCR amplification of a 51bp DNA fragment within the longer peptide. This new sequence served as a template for the synthesis of a homologous primer. The RT-PCR fragment of 597bp obtained with this primer together with a polyT primer was radioactively labelled and used for the screening of a chicken duodenal expression library. The isolated full-length cDNA for b,b-carotene 15,15´-dioxygenase (EMBL accession no. AJ271386) has a length of 3.1kb. The coding sequence spans 1578bp and encodes a protein of 526 amino acids (Figure 1). The cDNA reveals a high similarity of b,b-carotene 15,15´-dioxygenase to RPE65 [37,38], a retinal pigment epithelium-specific protein of, at present, unknown function. By searching the EMBL gene bank, we identified a mouse expressed sequence tag (EST) (EMBL accession no. AW044715) containing the putative mouse homologue of b,b-carotene 15,15´-dioxygenase (resubmitted; EMBL accession no. AW278064). The regions of amino acids 97–108 and 132–155 are almost identical and, therefore, have been highly conserved during evolution. In addition, von Lintig and Vogt [32] have recently cloned and expressed the cDNA for the Drosophila enzyme displaying b,b-carotene 15,15´-dioxygenase activity. A PRETTYBOX alignment is shown in Figure 1. The chicken and mouse b,b-carotene 15,15´-dioxygenases show highest homology (81% on the amino acid level), whereas the Drosophila b,b-carotene 15,15´-dioxygenase sequence shows 50% homology with the chicken, and 48% with the mouse, amino acid sequences (Figure 1). Although the Drosophila protein has two additional stretches (amino acids 1–30 and 378–406), the rest of the sequence is very similar to the vertebrate dioxygenase sequences.

Functional expression in eukaryotic cells

Chicken b,b-carotene 15,15´-dioxygenase was overexpressed in CHO cells. For this purpose, the coding sequence of b,b-carotene 15,15´-dioxygenase was amplified by PCR, and the resulting fragment was cloned into the plasmid pSFV₂gen. This vector is part of the Semliki Forest Virus expression system [39], which can be used successfully with most mammalian cell types. The plasmid was used for production of a high-titre virus stock, and subsequently CHO cells were infected. The recombinant protein was purified using a Co²⁺-chelate affinity column, and the protein-containing fractions were dialysed against assay buffer and tested for b,b-carotene 15,15´-dioxygenase activity (Figure 2). When b,b-carotene was used as substrate, retinal was detected as the only reaction product. The resultant peak was identified by co-chromatography with a retinal standard solution (Figure 2), as well as by liquid chromatography–MS (results not shown). On the other hand, HPLC analysis revealed no b-apo-carotenals or other metabolites (results not shown).

Northern blotting

b,b-Carotene 15,15´-dioxygenase is expressed primarily in the digestive tract of chicken (Figure 3a). Highest expression was found in the duodenum, whereas in the ileum, the mRNA level was markedly lower. From the chicken Northern blot, the size of the message was estimated to be approx. 5.5kb. On a mouse multi-tissue Northern blot with eight poly(A)⁺ RNAs (Figure 3b), expression of b,b-carotene 15,15´-dioxygenase in three additional tissues was confirmed. In liver, quite a strong signal appeared, whereas expression in kidney was only marginal. No expression was observed in lung. Most interestingly, three distinct bands were clearly identified in mouse testis: a message of 2.4kb, a fainter band at 2kb and the most prominent band, at 1.7kb. Whether these three bands represent alternative splicing products of a single gene, or three distinct b,b-carotene 15,15´-dioxygenase genes, remains to be established. If the occurrence of particular isoenzymes of b,b-carotene 15,15´-dioxygenase in testis can be confirmed, this might point to an important function of this enzyme during maturation of germ cells.

In situ hybridization

To investigate the expression pattern of b,b-carotene 15,15´-dioxygenase, we performed in situ hybridization on 10µm cryosections from chicken tissues. In duodenum (Figure 4a), the epithelial cells of the crypts and villi gave rise to a strong signal. Stromal cells in the villi were also positively stained. These signals might arise, at least in part, from cells involved in the immune response (lymphocytes and plasma cells). Further identification as to whether these cells are immune cells has yet to be performed.

In liver (Figure 4b), a diffuse signal was detected in hepatocytes. Compared with the control staining with the sense probe (Figure 4e), these signals were considered to be clearly positive. The possibility that other cell types are also stained cannot be excluded. The 'foaminess' of the hepatocyte signal might have been caused by the presence of glycogen in the cytoplasm, or by the preparation of the sections. In lung sections (Figure 4c), a strong signal was seen in epithelial cells of bronchioles. Weaker signals were seen in alveolar walls. In kidney, the endothelial cells of glomeruli and tubuli were positively stained (Figure 4g). In this tissue, the endothelial cells of blood vessels also appeared to be stained. Ileum was positively stained (Figure 4h). In skin, dioxygenase staining was also increased compared with the control section. For all sections, counterstaining with haematoxylin and eosin was performed (results not shown), as were control incubations with a sense RNA probe (Figures 4d–f and 4j–l).

The hybridizations with the sense probe occasionally produced weak unspecific signals, which could be clearly distinguished from the positive signals obtained with the respective antisense probe. In conclusion, in all tissues examined, mainly the epithelial cells of tubular structures expressed b,b-carotene 15,15´-dioxygenase.

DISCUSSION  

Since 1965, a number of attempts have been made to purify b,b-carotene 15,15´-dioxygenase. In these studies, purification factors of up to ×240, and specific activities of up to 3.3nmol of retinal formed/h per mg of protein have been reported. The failure to obtain a more complete purification was hypothesized to be due to an amphiphilic character of b,b-carotene 15,15´-dioxygenase, which might render the protein unstable in aqueous solution [21].

In spite of using a fast purification protocol, encompassing ammonium sulphate fractionation, hydrophobic interaction chromatography, Blue Sepharose chromatography, IEC and gel filtration, our final preparation, with a specific activity of approx. 2.5nmol·h^-1·mgprotein^-1, was still far from being pure. Several lines of evidence suggest that, in our preparation too, amphiphilic characteristics of b,b-carotene 15,15´-dioxygenase might have prevented further enrichment of the protein: (a) major losses in enzyme activity were observed both during IEC (60%) and upon 50-fold concentration of the IEC eluate before gel filtration (75%) (both an increased ionic strength and a higher protein concentration are likely to promote aggregation and, presumably, precipitation of an amphiphilic protein); (b) inclusion of 10% glycerol in most of the elution buffers considerably stabilized the protein; and (c) incubation of the Blue Sepharose eluate with 1M ammonium sulphate resulted in 80% inactivation of b,b-carotene 15,15´-dioxygenase.

Despite the difficulties encountered in protein purification, we obtained sufficient amino acid sequence information to clone the chicken cDNA for this elusive enzyme. The isolated chicken cDNA clone had a length of 3.1kb, of which 1578bp are coding for the 60.4kDa protein. The mouse cDNA had a size of only 2.1kb, of which 1698bp are coding for a protein of 63.9kDa. Since gel-permeation chromatography suggested a molecular mass for the chicken enzyme of 50kDa, b,b-carotene 15,15´-dioxygenase most likely is a monomeric protein, at least in solution.

Northern blots showed that the mRNA is indeed longer in chicken ( 5.5kb; Figure 3a) than in mouse ( 2.4kb; Figure 3b). The alignment of the protein sequences of b,b-carotene 15,15´-dioxygenase from three different species (chicken, mouse and D. melanogaster) with chicken RPE65 (Figure 1) reveals high homology between all these proteins, which form a new family of carotenoid and/or retinoid binding/cleaving enzymes, as also suggested by von Lintig and Vogt [32]. In this family, the membrane receptor for the plasma retinol-binding protein [40] can also be included. Most of the highly conserved regions are located in the N-terminal part of the protein, e.g. a purported amphipathic helix (between amino acids 100 and 120 of the chicken sequence), which is highly conserved in all species examined. These conserved regions might well be important for substrate binding and/or catalysis. Another possibility is that this helix links the protein to the membrane, where the hydrophobic substrates, i.e. b,b-carotene, are normally located after uptake into the cell.

b,b-Carotene 15,15´-dioxygenase was found to be a cytosolic enzyme [17,19]. We also found cleavage activity exclusively in cytosolic fractions, and no b,b-carotene 15,15´-dioxygenase activity was associated with membrane fractions of the eukaryotic expression systems used. In addition, there is no evidence from the amino acid sequence for transmembrane segments of the protein.

Owing to the formation of inclusion bodies and to a high frequency of internal initiation products in E. coli, we have chosen the Semliki Forest Virus system and eukaryotic cell lines (BHK, CHO and HuTu80) for overexpression of b,b-carotene 15,15´-dioxygenase. In CHO cells, an expression level of 1–2mg of soluble and active protein per litre of medium was obtained.

In all of our experiments, retinal was identified as the only reaction product (Figure 2), thereby lending conclusive support for central cleavage of b,b-carotene at the 15,15´-bond. In contrast, neither b-apo-carotenals nor other metabolites were detected by HPLC. In the future, other provitamin A carotenoids will be tested, and the respective reaction products analysed.

Investigation of the expression pattern in different chicken and mouse tissues revealed, as expected, highest expression levels in the duodenum (Figure 3a), the tissue from which the cDNA was cloned. Ileum gave rise to a much weaker signal on Northern blots. Among eight mouse tissues analysed, liver, kidney and testis were positive.

In situ hybridization was used to analyse the expression pattern of b,b-carotene 15,15´-dioxygenase in selected chicken tissues in more detail (Figure 4). In duodenum, the epithelial cells of crypts and villi were positively stained. Some stromal cells were also stained, hypothetically representing T- or B-cells; however, this needs to be elucidated. There is growing evidence that vitamin A plays an important role in immunological processes. Thus the requisite vitamin A demand might be met by the dioxygenase in the corresponding cells and tissues.

In chicken-lung tissue, a clear signal was obtained with in situ hybridization on cryosections (Figure 4c); however, on the mouse multi-tissue Northern blot, no band was observed in lung. This could be explained by an interspecies difference of expression, or by the tissue-specific regulation and induction of the b,b-carotene 15,15´-dioxygenase by the RARa antagonist (see the Experimental section). In general, the results from in situ hybridization cannot be quantified, but show the localization within a tissue. All differences in expression levels have to be investigated further and confirmed by quantitative RT-PCR, which is the most accurate method available for measuring mRNA levels in tissues.

The cloning of the first vertebrate b,b-carotene 15,15´-dioxygenase is an important step forward in terms of acquiring a more detailed knowledge of b,b-carotene and vitamin A metabolism. We have cloned the chicken and mouse b,b-carotene 15,15´-dioxygenases, two new members of the family of carotene-cleaving enzymes, which appear to be highly conserved throughout evolution. Our studies have also demonstrated that many tissues express relatively high levels of this enzyme, leading to the conclusion that additional vitamin A supply has to be met by tissue-specific cleavage of b,b-carotene.

We thank U. Margelisch, M. Scott, M. Müller, P. Buchwald, M. Sedgewick and K. Flückiger for excellent technical support, and Chr. Schweitzer and K. Lundstrom for the great help with the Semliki Forest Virus expression system.

1 Olson, J. A. and Krinsky, N. I. (1995) Introduction: the colorful, fascinating world of the carotenoids: important physiologic modulators. FASEB J. 9, 1547–1550
Medline   1st Citation

2 Parker, R. S. (1996) Absorption, metabolism, and transport of carotenoids. FASEB J. 10, 542–551
Medline   1st Citation

3 Britton, G. (1993) Carotenoids in chloroplast pigment–protein complexes. In Pigment–Protein Complexes in Plastids: Synthesis and Assembly (Sundqvist, C. and Ryberg, M., eds.), pp. 447–483, Academic Press, New York
1st Citation

4 Mathews-Roth, M. M. (1993) Carotenoids in erythropoietic protoporphyria and other photosensitivity diseases. Ann. N.Y. Acad. Sci. 691, 127–138
Medline   1st Citation

5 Krinsky, N. I. (1994) The biological properties of carotenoids. Pure Appl. Chem. 66, 1003–1010
1st Citation

6 Palozza, P. and Krinsky, N. I. (1992) Antioxidant effects of carotenoids in vivo and in vitro: an overview. Methods Enzymol. 213, 403–420
Medline   1st Citation

7 Woodall, A. A., Lee, S. W., Weesie, R. J., Jackson, M. J. and Britton, G. (1997) Oxidation of carotenoids by free radicals: relationship between structure and reactivity. Biochim. Biophys. Acta 1336, 33–42
Medline   1st Citation

8 Sommer, A. (1997) Vitamin A deficiency, child health, and survival. Nutrition 13, 484–485
Medline   1st Citation

9 Sommer, A. (1989) New imperatives for an old vitamin (A). J. Nutr. 119, 96–100
Medline   1st Citation

10 West, K. P., LeClerq, S. C., Shrestha, S. R., Wu, L. S., Pradhan, E. K., Khatry, S. K., Katz, J., Adhikari, R. and Sommer, A. (1997) Effects of vitamin A on growth of vitamin A-deficient children: field studies in Nepal. J. Nutr. 127, 1957–1965
Medline   1st Citation

11 Carlier, C., Coste, J., Etchepare, M., Periquet, B. and Amedee-Manesme, O. (1993) A randomised controlled trial to test equivalence between retinyl palmitate and b carotene for vitamin A deficiency. Br. Med. J. 307, 1106–1110
1st Citation  2nd

11a Erratum, (1994) . Br. Med. J. 308, 6921
1st Citation

12 Wirtz, G. (1998) Über die Substratspezifität und Reinigung der b,b-Carotin-15,15´-Dioxygenase (EC 1.13.11.21). PhD Thesis, University of Basel
1st Citation

13 Singh, H. and Cama, H. R. (1974) Enzymatic cleavage of carotenoids. Biochim. Biophys. Acta 370, 49–61
Medline   1st Citation

14 Olson, J. A. and Hayaishi, O. (1965) The enzymatic cleavage of b-carotene into vitamin A by soluble enzymes of rat liver and intestine. Proc. Natl. Acad. Sci. U.S.A. 54, 1364–1370
Medline   1st Citation  2nd

15 Goodman, D. S. and Huang, H. S. (1965) Biosynthesis of vitamin A with rat intestinal enzymes. Science 149, 879–880
1st Citation  2nd

16 Goodman, D. S., Blomstrand, R., Werner, B., Huang, H. S. and Shiratori, T. (1966) The intestinal absorption and metabolism of vitamin A and b-carotene in man. J. Clin. Invest. 45, 1615–1623
Medline   1st Citation  2nd

17 Fidge, N. H., Smith, F. R. and Goodman, D. S. (1969) Vitamin A and carotenoids. The enzymic conversion of b-carotene into retinal in hog intestinal mucosa. Biochem. J. 114, 689–694
Medline   1st Citation  2nd  3rd  4th

18 Lakshmanan, M. R., Chansang, H. and Olson, J. A. (1972) Purification and properties of carotene 15,15´-dioxygenase of rabbit intestine. J. Lipid Res. 13, 477–482
Medline   1st Citation  2nd

19 Goodman, D. S., Huang, H. S. and Shiratori, T. (1966) Mechanism of the biosynthesis of vitamin A from b-carotene. J. Biol. Chem. 241, 1929–1932
Medline   1st Citation  2nd  3rd

20 Lakshman, M. R., Mychkovsky, I. and Attlesey, M. (1989) Enzymatic conversion of all-trans-b-carotene to retinal by a cytosolic enzyme from rabbit and rat intestinal mucosa. Proc. Natl. Acad. Sci. U.S.A. 86, 9124–9128
Medline   1st Citation

21 Devery, J. and Milborrow, B. V. (1994) b-Carotene-15,15´-dioxygenase (EC 1.13.11.21) isolation reaction mechanism and an improved assay procedure. Br. J. Nutr. 72, 397–414
Medline   1st Citation  2nd

22 Sklan, D. (1983) Carotene-cleavage activity in chick intestinal mucosa cytosol: association with a high-molecular-weight lipid-protein aggregate fraction and partial characterization of the activity. Br. J. Nutr. 50, 417–425
Medline   1st Citation

23 Duszka, C., Grolier, P., Azim, E. M., Alexandre-Gouabau, M. C., Borel, P. and Azais-Braesco, V. (1996) Rat intestinal b-carotene dioxygenase activity is located primarily in the cytosol of mature jejunal enterocytes. J. Nutr. 126, 2550–2556
Medline   1st Citation

24 van Vliet, T., van Schaik, F., Schreurs, W. H. and van den Berg, H. (1996) In vitro measurement of b-carotene cleavage activity: methodological considerations and the effect of other carotenoids on b-carotene cleavage. Int. J. Vitam. Nutr. Res. 66, 77–85
Medline   1st Citation

25 During, A., Nagao, A., Hoshino, C. and Terao, J. (1996) Assay of b-carotene 15,15´-dioxygenase activity by reverse-phase high-pressure liquid chromatography. Anal. Biochem. 241, 199–205
Medline   1st Citation  2nd

26 Wolf, G. (1995) The enzymatic cleavage of b-carotene: still controversial. Nutr. Rev. 53, 134–137
Medline   1st Citation

27 Glover, J. (1960) The conversion of b-carotene into vitamin A. Vitam. Horm. (San Diego) 18, 371–386
1st Citation

28 Wang, X. D., Tang, G. W., Fox, J. G., Krinsky, N. I. and Russell, R. M. (1991) Enzymatic conversion of b-carotene into b-apo-carotenals and retinoids by human, monkey, ferret, and rat tissues. Arch. Biochem. Biophys. 285, 8–16
Medline   1st Citation

29 Wang, X. D., Krinsky, N. I., Tang, G. W. and Russell, R. M. (1992) Retinoic acid can be produced from excentric cleavage of b-carotene in human intestinal mucosa. Arch. Biochem. Biophys. 293, 298–304
Medline   1st Citation

30 Wang, X. D., Russell, R. M., Marini, R. P., Tang, G., Dolnikowski, G. G., Fox, J. G. and Krinsky, N. I. (1993) Intestinal perfusion of b-carotene in the ferret raises retinoic acid level in portal blood. Biochim. Biophys. Acta 1167, 159–164
Medline   1st Citation

31 Nagao, A., During, A., Hoshino, C., Terao, J. and Olson, J. A. (1996) Stoichiometric conversion of all trans-b-carotene to retinal by pig intestinal extract. Arch. Biochem. Biophys. 328, 57–63
Medline   1st Citation

32 von Lintig, J. and Vogt, K. (2000) Filling the gap in vitamin A research. Molecular identification of an enzyme cleaving b-carotene to retinal. J. Biol. Chem. 275, 11915–11920
Medline   1st Citation  2nd  3rd  4th

33 Wyss, A., Wirtz, G., Woggon, W., Brugger, R., Wyss, M., Friedlein, A., Bachmann, H. and Hunziker, W. (2000) Cloning and Expression of b,b-Carotene 15,15´-Dioxygenase. Biochem. Biophys. Res. Commun. 271, 334–336
Medline   1st Citation  2nd  3rd

34 During, A., Nagao, A. and Terao, J. (1998) b-Carotene 15,15´-dioxygenase activity and cellular retinol-binding protein type II level are enhanced by dietary unsaturated triacylglycerols in rat intestines. J. Nutr. 128, 1614–1619
Medline   1st Citation

35 Fleming, A. J., Mandel, T., Roth, I. and Kuhlemeier, C. (1993) The patterns of gene expression in the tomato shoot apical meristem. Plant Cell 5, 297–309
Medline   1st Citation

36 Hwang, C. S. and Kolattukudy, P. E. (1993) Molecular cloning and sequencing of thioesterase B cDNA and stimulation of expression of the thioesterase B gene associated with hormonal induction of peroxisome proliferation. J. Biol. Chem. 268, 14278–14284
Medline   1st Citation

37 Hamel, C. P., Tsilou, E., Pfeffer, B. A., Hooks, J. J., Detrick, B. and Redmond, T. M. (1993) Molecular cloning and expression of RPE65, a novel retinal pigment epithelium-specific microsomal protein that is post-transcriptionally regulated in vitro. J. Biol. Chem. 268, 15751–15757
Medline   1st Citation

38 Marlhens, F., Griffoin, J. M., Bareil, C., Arnaud, B., Claustres, M. and Hamel, C. P. (1998) Autosomal recessive retinal dystrophy associated with two novel mutations in the RPE65 gene. Eur. J. Hum. Genet. 6, 527–531
Medline   1st Citation

39 Liljestrom, P. and Garoff, H. (1991) A new generation of animal cell expression vectors based on the Semliki Forest virus replicon. Bio/Technology 9, 1356–1361
Medline   1st Citation

40 Bavik, C. O., Busch, C. and Eriksson, U. (1992) Characterization of a plasma retinol-binding protein membrane receptor expressed in the retinal pigment epithelium. J. Biol. Chem. 267, 23035–23042
Medline   1st Citation

Received 25 September 2000/12 December 2000; accepted 11 January 2001

The Biochemical Society, London © 2001