Original Article

Structure and functions of human oxysterol 7-hydroxylase cDNAs and gene CYP7B1

Correspondence to: J. Y. L. Chiang

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Oxysterol 7-hydroxylase has broad substrate specificity for sterol metabolites and may be involved in many metabolic processes including bile acid synthesis and neurosteroid metabolism. The cloned human oxysterol 7-hydroxylase (CYP7B1) cDNA encodes a polypeptide of 506 amino acid residues that shares 40% sequence identity to human cholesterol 7-hydroxylase (CYP7A1), the rate-limiting enzyme in the conversion of cholesterol to bile acids in the liver. In contrast to the liver-specific expression of CYP7A1, CYP7B1 mRNA transcripts were detected in human tissues involved in steroid genesis (brain, testes, ovary, and prostate) and in bile acid synthesis (liver) and reabsorption (colon, kidney, and small intestine). The human oxysterol 7-hydroxylase transiently expressed in 293/T cells was able to catalyze 7-hydroxylation of 27-hydroxycholesterol and dehydroepiandrosterone (DHEA). The human CYP7A1 and CYP7B1 both contain six exons and five introns. However, CYP7B1 spans at least 65 kb of the genome and is about 6-fold longer than CYP7A1. The transcription start site (+1) was localized 204 bp upstream of the initiation codon. No TATA box-like sequence was found near the transcription start site. Transient transfection assays of CYP7B1 promoter/luciferase reporter constructs in HepG2 cells revealed that the promoter was highly active. The 5' upstream region from nt -83 to +189 is the core promoter of the gene.—Wu, Z., K. O. Martin, N. B. Javitt, and J. Y. L. Chiang. Structure and functions of human oxysterol 7-hydroxylase cDNAs and gene CYP7B1. J. Lipid Res. 1999. 40: 2195;–2203.

Supplementary key words: bile acid synthesis, cholesterol 7-hydroxylase, dehydroepiandrosterone (DHEA), neurosteroids, cytochrome P450

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The recent discovery of a sterol 7-hydroxylase that differs significantly in tissue distribution and substrate specificity from cholesterol 7-hydroxylase has introduced new concepts with respect to sterol metabolism and regulation of bile acid biosynthesis in humans (1) (2) (3) (4) (5) (6) (7) (8) (9). This sterol 7-hydroxylase was initially cloned from a cDNA library of rat hippocampus (10). It was later shown to have 7-hydroxylase activity for 27-hydroxycholesterol (11) and for dehydroepiandrosterone (DHEA) and pregnenolone in the brain (12). Thus the sterol 7-hydroxylase with broad substrate specificity toward oxysterols is now referred as oxysterol 7-hydroxylase.

The classic pathway of bile acid synthesis is initiated by cholesterol 7-hydroxylase in the liver and is tightly regulated by bile acid feedback, hormones, and diurnal rhythm (2). In contrast, the pathway initiated with sterol 27-hydroxylase is located mainly in extrahepatic tissues, such as kidney, macrophage, and vascular endothelium. The 27-hydroxycholesterol is then 7-hydroxylated by oxysterol 7-hydroxylase and is finally converted to chenodeoxycholic and cholic acid in the liver. The absence of 7-hydroxylation would lead to formation of monohydroxy bile acids. Although in adults this alternative (or acidic) pathway makes a quantitatively minor contribution to total bile acid synthesis, it may represent a significant mechanism for preventing the development of atherosclerosis (13).

Oxysterol 7-hydroxylase activity expressed after 18 days of life might be responsible for the synthesis of 7-hydroxylated bile acids in cyp7a (-/-) mice (14). The accumulation of a high level of monohydroxy bile acids in newborn cyp7a (-/-) mice (5 to 10 days), the result of absence of both cholesterol 7-hydroxylase and oxysterol 7-hydroxylase activities, might cause severe neonatal cholestasis (15). We previously identified a metabolic defect in the 7-hydroxylation of 3ß-hydroxy-⁵-cholestenoic acid in a child with neonatal cholestasis (16). This has led us to isolate cDNA and the gene encoding human oxysterol 7-hydroxylase. Our study revealed the structure of the human CYP7B1 and its functions in the metabolism of neurosteroids and oxysterols and provided important molecular tools for studying the mechanism of oxysterol metabolism and the cause of cholestatic liver disease in humans.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

cDNA library screening
Plasmid pcDNA3.1 (+)-mcyp7b (11) containing a mouse cyp7b1 cDNA was digested with Hind III and Bgl II to release a 477 bp fragment from the 5' region of the coding sequence. A random primer labeling kit (RadPrimer DNA Labeling System, Life Technology, Gaithersburg, MD) was used to label the probe with [-³²P]dCTP. A human hippocampus gt10 cDNA library (Clontech, Palo Alto, CA) and a human liver ZAP cDNA library (Stratagene, La Jolla, CA) were plated. About 10 ⁶ plaque-forming units of each library were screened in each study. Positive clones were plaque-purified and phage DNAs were isolated using a Nucleobond Lambda kit (Clontech). The cDNA inserts were released with EcoR I and were subcloned into pBlueScript KS (II+) vector. The purified ZAP clones were excised in vivo from phage DNA and circularized to form a phagemid by co-infection of a f1 helper phage with host E. coli strain XL1-Blue.

Cloning the human CYP7B1
A human placenta genomic FIX II library (Stratagene) was screened for CYP7B1 genomic clones. The cloned phage DNAs were cut with XbaI and run on a 1% agarose gel. The size of each clone was estimated by adding up all XbaI fragments. A restriction map of the gene was determined by analysis of number XbaI fragments of each individual clone by Southern blot hybridization using cDNA probes. XbaI fragments containing exons were then subcloned into pBlueScript KS (II+) vector, cut with XbaI, and sequenced. The sizes of the introns were estimated by adding up all XbaI fragments between two exons and were later confirmed by genomic PCR. Several rounds of screening obtained clones that contained intron 1. The 5'-XbaI fragment of a genomic clone (G17) containing exon 2 and closest to the 5' end was identified and was then used as a probe to again screen the genomic library for upstream clones. Those clones were analyzed by restriction analysis to identify the XbaI fragment closest to the 5' end. This screening was repeated to obtain a genomic clone (G30), the 5' end of which overlaps the 3' end of the genomic clone (G43) that contains exon 1 and the 5'-flanking sequences.

Determining the CYP7B1 structure
An Advantage Genomic PCR kit (Clontech) was used for polymerase chain reactions (PCR) to locate exons and to determine the sizes of introns. Genomic clones that contained two adjacent exons were used as templates for PCR. Pairs of primers approximately 30 nucleotides each were designed according to the donor and acceptor exon sequences. The intron sequence was amplified and PCR products were analyzed on agarose gels. Intron/exon junction sequences were determined by DNA sequencing. Intron 1 was too big to be determined by PCR. The size of intron 1 was then estimated by restriction analysis of overlapping intron 1 clones obtained as described above.

DNA preparation and sequencing
All plasmids and phage DNA samples were prepared with Nucleobond AX columns (Clontech). A Sequenase Version 2.0 DNA sequencing kit (USB, Cleveland, OH) was used for nucleotide sequencing. Sequencing primers were designed from previously determined sequences and were used for further sequencing of both strands of DNA. Sequences in the GC-rich region were determined using both dITP and dGTP.

Northern blot hybridization analysis
Two human multiple tissue mRNA blots MTN-I (#7760-1) and II (#7759-1)(Clontech) were used to detect CYP7B1 mRNA transcripts in human tissues. The full-length human CYP7B1 cDNA was used as a probe and hybridization was carried out according to manufacturer's instructions. Radioactivity was detected by scanning with a phosphorimager (Model 455si, Molecular Dynamic, Sunnyvale, CA). Radioactivity was then washed out and the membrane was rehybridized with a CYP7A1 cDNA probe. ß-Actin was probed as an internal standard for normalization of RNA levels in each tissue.

Expression vector construction
A full-length human CYP7B1 cDNA was isolated from a phage clone H107 and was subcloned into pBluescript KS (II+). The recombinant plasmid pH107 was digested with Sma I and EcoR V to release a 1.6 kb fragment, which contains all but the codons for the first 20 amino acids. This fragment was then cloned into pcDNA3.1/His A plasmid (Invitrogene, Portland, OR) cut with EcoR V. Restriction analysis and DNA sequencing determined the orientation of the insert in the plasmid. The clone containing the insert in the forward direction was named p20hCYP7B1 and that with the sequence in the reverse direction was called p20hCYP7B1R. The DNA fragment containing the full-length coding sequence was released from pH107 with KpnI and XbaI and was subsequently subcloned into pcDNA3 (Invitrogene, Portland, OR) cut with KpnI and XbaI. The recombinant plasmid was named phCYP7B1.

Determination of transcription start site
Oligonucleotide primer (5'-CTAGTCCAGGGCCGGAGAGGC TGGCCTGC-3') corresponding to the complimentary sequence of nt 105 to 77 was used for primer extension. Primers were labeled with [-³³P]dATP (NEN Life Science, Beverly, MA). Assays were done using a Primer Extension kit from Promega (Madison, WI). Ten pmol of radiolabeled primer was annealed to 1 µg of human hippocampus poly A (+) RNA (Clontech) at 60°C for 1 h in primer extension buffer (50 mM Tris, pH 8.3, 50 mM KCl, 10 mM MgCl₂, 10 mM DTT, 1 mM each of dNTP, 0.5 mM spermidine). Reverse transcription was carried out by avian myeloblastosis virus (AMV) reverse transcriptase (Promega) for 30 min at 42°C. Loading dye was added into each reaction and the reverse transcription products were analyzed on 8% polyacrylamide gel electrophoresis in 8 M urea, along with sequencing reactions using the same primer. The gel was run until bromophenol blue was close to the bottom of the gel. The gel was dried and radioactivity was scanned with phosphorimager.

Computer analysis
All the sequence alignment, restriction enzyme mapping, PCR primer design, and amino acid translation studies were done with MacVector software (Oxford Molecular Biology, Oxford, England). Binding sites for putative transcription factor were identified using the TRANSFAC program (http://transfac.gbf.de/Transfac/MatInspector).

Expression of hCYP7B1 and enzyme activity assay
Human kidney 293/T cells derived from ATCC293 were grown and maintained in DMEM with 10% fetal bovine serum. 293/T cells were plated in 6-well plates 1 day before the transfection (11). For transfection, freshly prepared calcium phosphate precipitate (1.25 µg plasmid/well) was added to 293/T cells that were 60;–70% confluent and was allowed to remain in contact with the cells for 24 h. The media containing the DNA were removed. The cells were washed with phosphate-buffered saline, 2 ml of fresh media containing 15% D₂O and 10% delipidated fetal calf serum, and 5 mM 27-hydroxycholesterol, cholesterol, or 13.4 µM DHEA were added, and the mixture was incubated for an additional 24 h. The media were then used for HPLC analysis. The amounts of the substrate and product were calculated from the peak area in HPLC spectra. Activity was calculated in terms of nmole product per mg protein per 24 h.

Construction of promoter/luciferase reporter genes and transient transfection assay
The upstream sequence from HindIII (nt - 1082) to Sac II (+189) was released from pBluescript KS (II+) with Kpn I and SacI and was subcloned into pGL3 Basic luciferase reporter plasmid (Promega). The 5' deletion mutants (p-343/+189-Luc, p-291/+189-Luc, and p-144/+189-Luc) were generated with the pBluescript II Exo/Mung DNA Sequencing System (Stratagene). Three constructs (p1082/-84Luc, p-857/-84Luc, and p-291/-84-Luc) were also made by deletion of a SmaI fragment from the 3' end of the constructs. Transient transfection assays were performed as described previously with some modification (17). HepG2 cells were cultured in DMEM supplemented with 10% fetal bovine serum in 12-well plates. At 70% confluence, the cells were transfected with 2.5 µg DNA together with 150 ng of pCMVß, an expression plasmid for ß-galactosidase, as an internal control for transfection efficiency. After incubation for 2 days, the cells were harvested and lysed with lysate buffer. A 20-µl aliquot of lysate was used to assay luciferase activity with Luciferase Assay System (Promega). Ten µl of each lysate was also assayed for ß-galactosidase activity by incubating with o-nitrophenyl-ß-D-galactopyranoside at 37°C for 30 min and measuring absorbance at 405 nm. Luciferase reporter activity was expressed as random light units (RLU) divided by ß-galactosidase activity.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cloning human CYP7B1 cDNAs
To clone human CYP7B1 cDNA, a 5'-fragment of mouse cyp7b1 cDNA was used as a hybridization probe to screen a gt 10 cDNA library of human hippocampus. Three full-length cDNA clones H20 (5kb), H25 (1.7 kb), H107 (2.1 kb) were obtained and restriction maps were determined, as well as the nucleotide sequence of clone H107 (GenBank Accession #AF127090). H107 contains an open reading frame of 1521 bp, which encodes a peptide of 506 amino acids. The longest clone (H20) obtained is 5 kb in length and contains more than 3 kb of the 3' untranslated region (UTR). Sequence analysis revealed that human CYP7B1 shared 64.6% amino acid sequence identity with mouse cyp7b1 (10) and 40.4% identity with human CYP7A1 (18). The N-terminal membrane-binding domain (amino acid residues 1 to 38) is highly hydrophobic. The nucleotide coding sequences show 72.6% and 49.2% identity with the mouse cyp7b1 and human CYP7A1, respectively. The clones obtained from a cDNA library of human hippocampus all lack a poly A (+) tail because the library was constructed by a random priming method. Ten partial-length CYP7B1 clones were also isolated from a cDNA library of human liver constructed by oligo dT-priming. These clones lacked 5'-coding sequences and all contained a poly (A+) tail. The nucleotide sequences of these liver cDNA clones were identical to the corresponding sequences of the hippocampus cDNA clones.

Setchell et al. (19) reported cloning a human CYP7B1 cDNA. Comparison with the cDNA and amino acid sequences found by these investigators and those found in our studies revealed one base-pair difference at codon 324, which codes for His (CAT) and Arg (CGT), respectively.

The GC content of the 281-bp nucleotide sequences of the 5'-region of human cDNA is 74.4%. This is well above the average GC content of 56% for human genes around the initiation codon (20). Two putative polyadenylation signals were noticed in 3'UTR. One (AATAAA) was located 19 bp and the other (AATAAA) 284 bp downstream of the stop codon. Indeed, nucleotide sequences of the cDNA clones obtained revealed that both polyadenylation signals were utilized in poly A (+) mRNA synthesis. However, all liver clones lacked the GC-rich 5' sequence. This could be because the secondary structures of the 5' GC-rich region prevented the continuation of synthesis of the full-length first-strand cDNA from mRNA templates by reverse transcriptase when oligo dT was used for priming.

Cloning the human CYP7B1
As the CYP7A1 we had previously cloned were approximately 10 kb, we had expected that the CYP7B1 would be similar in size. The same human placenta FIXII genomic library was screened using a full-length human CYP7B1 cDNA. A total of 18 positive clones were obtained by screening approximately 2 million plaque-forming units. Six overlapping genomic clones used to construct the gene structure are shown in Figure 1. The XbaI fragments containing exons were identified using cDNA fragments as hybridization probes. The exon/intron junction sequences in these fragments were then determined using primer pairs designed according to the exon sequences. The exon/intron junction sequences and the sizes of each exon and intron are presented in Table 1. All donor and acceptor sites of exon/intron junction sequences follow the GT/AG rule. The sizes of introns were determined as described above in Material and Methods. The size of intron 1 was estimated to be approximately 30 kb by adding up all XbaI fragments between exon 1 and exon 2. The minimum size of CYP7B1 was estimated to be 65 kb.

View larger version (11K): [in this window] [in a new window]   Figure 1. Gene structure of human CYP7B1. The CYP7B1 contains 6 exons (boxes) and 5 introns (lines between two exons), and spans about 65 kb. The intron sizes are drawn to scale, but the exons are not. Six genomic clones, which were used to determine gene structure, are shown below the gene. XbaI (X) restriction fragments used to determine the exon/intron junction structure are indicated.

Multiple tissue Northern blot analysis
Human multiple tissue Northern blots were hybridized with the full-length human CYP7B1 cDNA probe. A single band approximately 9 kb long was seen in all tissues ( Figure 2 A, MTN-I). The levels of CYP7B1 mRNA expression were estimated after normalizing with ß-actin signal and were in the order of liver > brain = kidney = pancreas > heart > lung > skeletal muscle > placenta. In another tissue blot (Figure 2b, MTN-II), 9 kb mRNA transcripts were detected in ovary, testes, prostate, colon, small intestine, and thymus. mRNA species of approximately 2.5 kb in length were also detected in human liver and testes. The human CYP7B1 mRNA transcripts are much longer than those found in mouse and rat, which were reported to be about 2kb (10) (21). The same blot was stripped of radioactivity and was then re-probed with human CYP7A1 cDNA. A band of approximately 4 kb was found only in liver.

View larger version (206K): [in this window] [in a new window]   Figure 2. Human multiple tissue northern blot hybridization. The blots were hybridized with human CYP7B1 cDNA probe and then rehybridized with CYP7A1 and human ß-actin cDNA probes. A: MTN-I, Ht: heart, Br: brain, Pl: placenta, Lu: lung, Li: liver, Sm: skeleton muscle, Kd: kidney, Pc: pancreas. B: MTN-II, Sp, spleen; Th, thymus; Pr, prostate; Ts, testes; Ov, ovary; Si, small intestine; Cl, colon; and Phl, peripheral blood lymphocyte.

Cloning the 5' flanking upstream region of the human CYP7B1 and mapping the transcription start site
The genomic clone G43 contained 19 kb of the gene. A 10 kb XbaI fragment containing exon 1 was isolated and further mapped ( Figure 3a). Nucleotide sequences from Hind III to the initiation codon ATG are shown in Figure 3b. Using human hippocampus mRNA as a template, the transcription start site was mapped to a G located 204 bp upstream of the initiation codon ( Figure 4). A perfect GC box sequence is located just upstream of the transcription start site.

View larger version (72K): [in this window] [in a new window]   Figure 3. The nucleotide sequences of the 5'-upstream region of the human CYP7B1. Top: Restriction map of a 10 kb XbaI fragment which contains exon 1 and 9 kb upstream sequence was obtained from genomic clone G43. The 1.3 kb Hind III, SacII fragment was further subcloned and sequenced. X: XbaI, E: EcoRI, H: Hind III, S: Sac II. Bottom: The 5'-upstream sequences of human CYP7B1 gene. The transcription start site (+) G is located 204 nucleotides upstream of ATG start codon. A reversed CAAT box and a GC box are located. Putative binding sites for transcription factors were identified by computer analysis and are underlined and labeled. Sterol response element (SRE)-like sequences are also identified. GenBank accession #127089.

View larger version (111K): [in this window] [in a new window]   Figure 4. Primer extension assay of the transcription start site. Human hippocampus poly A (+) RNA was used as a template for primer extension assay as described under Experimental Procedures. Sequencing reactions with the same primer used for primer extension assay were run on the same gel and are shown. Lane 1, primer extension reaction with 1 µg of human hippocampus poly A (+) RNA. Lane 2, a positive control reaction with a 1.2 kb Kanamycin mRNA and primer provided in primer extension kit.

The clone G43 contained approximately 9 kb of the 5'upstream sequence. Analysis of the 1.3 kb sequence with the MatInspector program (http:/transfac.gbf.de/cgi-bin/matSearch/matsearch.pl) (22) revealed several putative transcription factor-binding sites. No canonical TATA box sequence was found in the region approximately 50 bp upstream of the transcription start site (+1), but CAAT and GC box sequences were found at nt -105 and -26, respectively. A typical TATAAAA sequence was found between nt -709 and -703, which was far upstream of the transcription start site we had determined. Two overlapping sterol response element (SRE)-like sequences (CACCCCAC) were found to be located between nt -327 and -314 (23). Other putative transcription factor binding sequences such as NFAT (24), C/EBPß, NFB (25), GATA box (26), HNF3ß (27), NF1 (28), and AP4 were identified.

Alignment of the corresponding 5'-upstream regions of the human CYP7B1 and CYP7A1 showed a very low sequence identity, less than 30%. The CYP7B1 promoter was found to have a GC content of 65% in the region of 500 bp upstream of the transcription start site, whereas that of the CYP7A1 was only 45%.

Transient transfection assays of CYP7B1/luciferase constructs
To identify the promoter and map the cis-regulatory elements in the 5'-flanking region, eight deletion constructs of CYP7B1 promoter/luciferase reporter genes were transiently transfected into HepG2 cells. Figure 5 shows that the CYP7B1 promoter was very active, comparable with the SV40 promoter. Deletion of the 5'-flanking sequence from -1082 to -291 did not affect reporter activity. Deletion of the sequences between -291 and -87 decreased activity by about 45%, indicating that the deleted region may contain binding sites for the trans-activator. Deletion of the sequence from -83 to +189 resulted in complete loss of transcriptional activity for these reporter constructs, indicating that the sequences from -84 to +189 are essential for promoter activity. A perfect GC box is located in this region (Figure 3). The GC box binds SP-1, a ubiquitous transcription activator involved in transcriptional regulation of many genes. These results suggest that a TATA box sequence located at -719 is not likely to play a functional role in initiation of transcription by RNA polymerase II. Nucleotide sequences from -291 to -87 contain positive cis-regulatory elements. Putative binding sites for C/EBPß, CAAT box, NF-B are located in this region. These sequences are likely important for transcriptional activity of the CYP7B1 promoter. Experiments are in progress to further identify cis-regulatory elements in the 5'-flanking region of the gene in different tissues.

View larger version (30K): [in this window] [in a new window]   Figure 5. Transient transfection assays of CYP7B1 promoter/luciferase reporter gene constructs in HepG2 cells. CYP7B1 promoter/luciferase reporter gene constructs were cotransfected with pCMVß into HepG2 cells. The luciferase activity and ß-galactosidase activity were measured 2 days later. Luciferase activity is expressed as random light unit (RLU) divided by ß-galactosidase activity. Error bar represents standard deviation of triplicate samples.

Transfection assays of enzyme activities expressed by CYP7B1 cDNA
To provide evidence that the human CYP7B1 cDNA we cloned does express oxysterol 7-hydroxylase activity, expression plasmids carrying a full-length (phCYP7B1) and a truncated (p20hCYP7B1) CYP7B1 cDNA were transiently transfected into 293/T cells and enzyme activities toward 27-hydroxycholesterol and DHEA were measured in lysates of the transfected cells. With 27-hydroxycholesterol as a substrate, the full-length and truncated enzymes expressed in 293/T cells produced 0.6 and 0.8 nmol 7, 27-dihydroxycholesterol per mg protein per 24 h, respectively ( Table 2). In contrast, with DHEA as a substrate, the full-length and truncated enzymes produced 85 and 89 nmol 7-hydroxyDHEA/mg protein/24 h, respectively, activities that were approximately 100-fold greater than those with 27-hydroxycholesterol. The plasmid carrying a cDNA cloned in the reverse direction (phCYP7B1R) did not express any enzyme activity at all. Activities of the truncated and full-length enzymes were similar toward both the above substrates, indicating that the N-terminal membrane-binding domain may not be required for catalytic activity, similar to the finding for cholesterol 7-hydroxylase (29). These data provided further evidence that the human CYP7B1 clone we isolated does encode oxysterol 7-hydroxylase.

View this table: [in this window] [in a new window]   Table 2. Expression of oxysterol 7-hydroxylase activity in 293/T cells

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

It is rather surprising from our study that the CYP7B1 mRNA and its gene are much larger than CYP7A1 although the intron/exon structures of the two are very similar. These two genes are in close proximity on human chromosome 8 (19). It is likely that the more compact CYP7A1 may be derived from the CYP7B1 by an ancient gene duplication or gene conversion event, and cholesterol 7-hydroxylase was selectively expressed only in the liver for bile acid synthesis. On the other hand, oxysterol 7-hydroxylase mRNA transcripts are expressed in tissues involved in steroidogenesis and neurosteroid metabolism (testes, ovary, prostate, kidney, liver, and brain) and in tissues involved in bile acid synthesis (liver) and absorption (small intestine, kidney, and colon). This may imply that oxysterol 7-hydroxylase may have multiple functions, including bile acid synthesis, neurosteroid metabolism, sex hormone synthesis, and detoxification of oxysterols.

The size of oxysterol 7-hydroxylase mRNAs in all tissues is approximately 9 kb, which is much longer than cholesterol 7-hydroxylase. The coding sequences of both hydroxylases are about the same. As the transcription start site is located 204 bp upstream of the initiation codon, the exon 6 of oxysterol 7-hydroxylase could be as long as 7 kb. The longest cDNA we isolated was 5 kb. It would be extremely difficult to isolate a full-length cDNA clone of 9 kb long. The human CYP7B1 shows a 64.6% amino acid sequence identity with the mouse cyp7b1. This is typical for cytochrome P450 genes in the same subfamily (30). The similarity in enzyme activity and substrate specificity between human and mouse enzyme supports the concept that the human CYP7B1 we cloned is the mouse homologue. The presence of multiple oxysterol 7-hydroxylase activities has been suggested (31). However, all human CYP7B1 clones we isolated from cDNA libraries of the liver and hippocampus using mouse cyp7b1 and human CYP7B1 cDNA as probes are identical in nucleotide sequences. The human CYP7B1 cDNAs we cloned and that were isolated by PCR are identical except for one nucleotide difference (19). It is unlikely that the human oxysterol 7-hydroxylase we isolated is a third isoform of the CYP7 family. This does not rule out the presence of a third form of oxysterol 7-hydroxylase, however.

Setchell et al. (19) recently reported an inborn error of bile acid metabolism caused by a point mutation in the Arg 388 codon of oxysterol 7-hydroxylase that resulted in premature termination. This neonate had severe cholestasis and cirrhosis. The defect in 7-hydroxylation was revealed by the absence of primary bile acid conjugates and the presence of high levels of 3ß-hydroxy-⁵-cholenoic and cholestenoic acids in the serum and urine, the metabolites formed via the acidic pathway of bile acid biosynthesis. Cholesterol 7-hydroxylase activity was also absent in this patient. The probable cause of the liver injury was the accumulation of high levels of hepatotoxic monohydroxy bile acids, 24, 25, and 27-hydroxycholesterols. Thus a physiological function of oxysterol 7-hydroxylase is to protect liver from cholestasis in neonatal life (16) (19).

Oxysterols are found in human serum, macrophages, vascular endothelium, and atherosclerotic plaques (32) and are potent inhibitors of expression of the genes encoding low density lipoprotein receptor, 3-hydroxy-3-methylglutaryl-CoA reductase, and other enzymes in the mevalonate pathway (33). Recently, oxysterols were identified as ligands for the liver orphan receptor (LXR), which has been suggested to function in a signaling pathway for regulating cholesterol and bile acid biosynthesis (34). Thus, sterol hydroxylases in bile acid biosynthesis pathways play important roles in cholesterol homeostasis. As oxysterol 7-hydroxylase has broad substrate specificity, it may metabolize cytotoxic oxysterols to dihydroxy metabolites, which are more soluble and readily excreted from extrahepatic tissues, diffused into circulation, transported into liver, and converted to bile acids, presumably by acidic bile acid biosynthesis pathway.

DHEA and pregnenolone are active neurosteroids that play roles in neuronal development and activation of N-methyl-D-aspartate receptor (35) (36). The 7-hydroxylation of DHEA is a quantitatively important metabolic step in brain tissue and may contribute to the regulation of the concentration and activity of neurosteroids in brain (37).

Although the recognition of a novel 7-hydroxylase enzyme resulted from studies of pathways of bile acid synthesis, it is unlikely that oxysterol 7-hydroxylase functions as a key regulatory enzyme like cholesterol 7-hydroxylase. In contrast to cholesterol 7-hydroxylase, oxysterol 7-hydroxylase does not initiate bile acid synthesis and its activity is much higher than that of cholesterol 7-hydroxylase in human liver (4). It is not known whether oxysterol 7-hydroxylase activity is regulated by bile acids, oxysterols, or other factors. Results from transient transfection assay of the CYP7B1/luciferase reporters in HepG2 cells have defined the nucleotide sequences essential for the transcriptional activity of the promoter to a 272 bp fragment in the 5'-upstream region. Recently, FXR has been identified as a bile acid receptor, which mediates the negative feedback regulation of CYP7A1 gene transcription (38) (39) (40). Preliminary study from this laboratory showed that FXR and bile acids do not regulate the CYP7B1 transcription. This study provides us with important molecular tools for studying the mechanism of regulation of the acidic bile acid biosynthesis pathway in the liver, the pathogenesis of cholestatic liver diseases, and the regulation of steroid metabolisms in brain and other tissues.

	ACKNOWLEDGMENTS

This research was supported by grants GM31584 and DK44442 from the National Institutes of Health. The technical assistance and discussion of Dr. Diane Stroup and editorial assistance of Mrs. Suzanne Javitt are gratefully acknowledged.

Manuscript received May 13, 1999; and in revised form August 24, 1999

Abbreviations: CYP7B1, oxysterol 7-hydroxylase gene; CYP7A1, cholesterol 7-hydroxylase gene; DHEA, dehydroepiandrosterone; DMEM, Dubecco's modified Eagle's medium; Luc, luciferase; CMV, cytomegalovirus; PCR, polymerase chain reaction; RLU, random light units; kb, kilobases; nt, nucleotide

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

1. Javitt, N. B. 1994. Bile acid synthesis from cholesterol: regulatory and auxiliary pathways. FASEB J. 8:1308-1311[Abstract].

2. Chiang, J. Y. L. 1998. Regulation of bile acid synthesis. Front. Biosci. 3:D176-193.

3. Shoda, J., Toll, A., Axelson, M., Pieper, F., Wikvall, K., Sjovall, J. 1993. Formation of 7- and 7ß-hydroxylated bile acid precursors from 27-hydroxycholesterol in human liver microsomes and mitochondria. Hepatology. 17:395-403[Medline].

4. Bjorkhem, I., Nyberg, B., Einarsson, K. 1992. 7-hydroxylation of 27-hydroxycholesterol in human liver microsomes. Biochim. Biophys. Acta. 1128:73-76[Medline].

5. Axelson, M., Shoda, J., Sjovall, J., Toll, A., Wikvall, K. 1992. Cholesterol is converted to 7-hydroxy-3-oxo-4-cholestenoic acid in liver mitochondria. Evidence for a mitochondrial sterol 7-hydroxylase. J. Biol. Chem. 267:1701-1704[Abstract/Free Full Text].

6. Martin, K. O., Budai, K., Javitt, N. B. 1993. Cholesterol and 27-hydroxycholesterol 7-hydroxylation: evidence for two different enzymes. J. Lipid Res. 34:581-588[Abstract].

7. Toll, A., Shoda, J., Axelson, M., Sjovall, J., Wikvall, K. 1992. 7-Hydroxylation of 26-hydroxycholesterol, 3ß-hydroxy-5-cholestenoic acid and 3ß-hydroxy-5-cholenoic acid by cytochrome P-450 in pig liver microsomes. FEBS Lett. 296:73-76[Medline].

8. Zhang, J., Larsson, O., Sjovall, J. 1995. 7-Hydroxylation of 25-hydroxycholesterol and 27-hydroxycholesterol in human fibroblasts. Biochim. Biophys. Acta. 1256:353-359[Medline].

9. Zhang, J., Akwa, Y., Baulieu, E. E., Sjovall, J. 1995. 7-Hydroxylation of 27-hydroxycholesterol in rat brain microsomes. C. R. Acad. Sci. Ser. III, Sci. Vie. 318:345-349[Medline].

10. Stapleton, G., Steel, M., Richardson, M., Mason, J. O., Rose, K. A., Morris, R. G. M., Lathe, R. 1995. A novel cytochrome P450 expressed primarily in brain. J. Biol. Chem. 270:29739-29745[Abstract/Free Full Text].

11. Martin, K. O., Reiss, A. B., Lathe, R., Javitt, N. B. 1997. 7-Hydroxylation of 27-hydroxycholesterol: biologic role in the regulation of cholesterol synthesis. J. Lipid Res. 38:1053-1058[Abstract].

12. Rose, K. A., Stapleton, G., Dott, K., Kieny, M. P., Best, R., Schwarz, M., Russell, D. W., Bjorkhem, I., Seckl, J., Lathe, R. 1997. Cyp7b, a novel brain cytochrome P450, catalyzes the synthesis of neurosteroids 7-hydroxy dehydroepiandrosterone and 7-hydroxypregnenolone. Proc. Natl. Acad. Sci. USA. 94:4925-4930[Abstract/Free Full Text].

13. Bjorkhem, I., Andersson, O., Diczfalusy, U., Sevastik, B., Xiu, R-j., Duan, C., Lund, E. 1994. Atherosclerosis and sterol 27-hydroxylase: evidence for a role of this enzyme in elimination of cholesterol from human macrophages. Proc. Natl. Acad. Sci. USA. 91:8592-8596[Abstract/Free Full Text].

14. Schwarz, M., Lund, E. G., Setchell, K. D. R., Kayden, H. J., Zerwekh, J. E., Bjorkhem, I., Hertz, J., Russell, D. W. 1996. Disruption of cholesterol 7-hydroxylase gene in mice: II. Bile acid deficiency is overcome by induction of oxysterol 7-hydroxylase. J. Biol. Chem. 271:18024-18031[Abstract/Free Full Text].

15. Arnon, R., Yoshimura, T., Reiss, A., Budai, K., Lefkowitch, J. H., Javitt, N. B. 1998. Cholesterol 7-hydroxylase knockout mouse: a model for monohydroxy bile acid-related neonatal cholestasis. Gastroenterology. 115:1223-1228[Medline].

16. Javitt, N. B., A. Kok, and M. Gut. 1984. Neonatal cholestasis: identification of a metabolic error in bile acid synthesis. Pediatr. Res. 18: Abstract #627.

17. Crestani, M., Stroup, D., Chiang, J. Y. L. 1995. Hormonal regulation of the cholesterol 7-hydroxylase gene (CYP7). J. Lipid Res. 36:2419-2432[Abstract].

18. Karam, W. G., Chiang, J. Y. L. 1992. Polymorphisms of human cholesterol 7-hydroxylase. Biochem. Biophys. Res. Commun. 185:588-595[Medline].

19. Setchell, K. D. R., Schwarz, M., O'Connell, N. C., Lund, E. G., Davis, D. L., Lathe, R., Thompson, H. R., Weslie Tyson, R., Sokol, R. J., Russell, D. W. 1998. Identification of a new inborn error in bile acid synthesis: mutation of the oxysterol 7-hydroxylase gene causes severe neonatal liver disease. J. Clin. Invest. 102:1690-1703[Medline].

20. Mizuno, M., Kanehisa, M. 1994. Distribution profiles of GC content around the translation initiation site in different species. FEBS Lett. 352:7-10[Medline].

21. Schwarz, M., Lund, E. G., Lathe, R., Bjorkhem, I., Russell, D. W. 1997. Identification and characterization of a mouse oxysterol 7-hydroxylase cDNA. J. Biol. Chem. 272:23995-24001[Abstract/Free Full Text].

22. Quandt, K., Frech, K., Karas, H., Wingender, E., Werner, T. 1995. MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res. 23:4878-4884[Abstract/Free Full Text].

23. Wang, X., Sato, R., Brown, M. S., Hua, X., Goldstein, J. L. 1994. SREBP-1, a membrane-bound transcription factor released by sterol-regulated proteolysis. Cell. 77:53-62[Medline].

24. Flanagan, W. M., Corthesy, B., Bram, R. J., Crabtree, G. R. 1991. Nuclear association of a T-cell transcription factor blocked by FK-506 and cyclosporin A. Nature. 352:803-807[Medline].

25. Lenardo, M. J., Baltimore, D. 1989. NF-KB: a pleiotropic mediator of inducible and tissue-specific gene control. Cell. 58:227-229[Medline].

26. Whyatt, D. J., deBoer, E., Grosveld, F. 1993. The two zinc finger-like domains of GATA-1 have different DNA binding specificities. EMBO J. 12:4993-5005[Medline].

27. Xanthopoulos, K. G., Prezioso, V. R., Chen, W. S., Sladek, F. M., Cortese, R., Darnell, J. E. 1991. The different tissue transcription patterns of genes for HNF1, C/EBP, HNF3, and HNF4 protein factors that govern liver-specific transcription. Proc. Natl. Acad. Sci. USA. 88:3807-3811[Abstract/Free Full Text].

28. Rosenfeld, P. J., Kelly, T. J. 1986. Purification of nuclear factor I by DNA recognition site affinity chromatography. J. Biol. Chem. 261:1398-1408[Abstract/Free Full Text].

29. Li, Y. C., Chiang, J. Y. L. 1991. The expression of a catalytically active cholesterol 7a-hydroxylase cytochrome P450 in E. coli. J. Biol. Chem. 266:19186-19191[Abstract/Free Full Text].

30. Nelson, D. R., Kamataki, T., Waxman, D. J., Guengerich, F. P., Estabrook, R. W., Feyereisen, R., Gonzalez, F. J., Coon, M. J., Gunsalus, I. C., Gotoh, O., Okuda, K., Nebert, D. W. 1993. The P450 superfamily: update on new sequences, gene mapping, accession number, early trivial names of enzymes, and nomenclature. DNA Cell Biol. 12:1-51[Medline].

31. Norlin, M., Wikvall, K. 1998. Biochemical characterization of the 7-hydroxylase activities towards 27-hydroxycholesterol and dehydroepiandrosterone in pig liver. Biochim. Biophys. Acta. 1390:269-281[Medline].

32. Brown, A. J., Jessup, W. 1999. Oxysterols and atherosclerosis. Atherosclerosis. 142:1-28[Medline].

33. Goldstein, J. L., Brown, M. S. 1990. Regulation of the mevalonate pathway. Nature. 343:425-430[Medline].

34. Janowski, B. A., P. J. Willy, T. R. Devi, J. R. Falck, D. and J. Mangelsdorf. 1996. An oxysterol signaling pathway mediated by the nuclear receptor LXR. Nature. 383: 728;–731.

35. Baulieu, E. E., Robel, P. 1998. Dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS) as neuroactive neurosteroids. Proc. Natl. Acad. Sci. USA. 95:4089-4091[Free Full Text].

36. Compagnone, N. A., Mellon, S. H. 1998. Dehydroepiandrosterone: a potential signaling molecule for neocortical organization during development. Proc. Natl. Acad. Sci. USA. 95:4678-4683[Abstract/Free Full Text].

37. Robel, P., Baulieu, E. E. 1995. Dehydroepiandrosterone (DHEA) is a neuroactive neurosteroid. Ann. NY Acad. Sci. 774:82-110[Medline].

38. Wang, H., Chen, J., Hollister, K., Sowers, L. C., Forman, B. M. 1999. Endogenous bile acids are ligands for the nuclear receptor FXR/BAR. Mol. Cell. 3:543-553.

39. Makishima, M., Okamoto, A. Y., Repa, J. J., Tu, H., Learned, R. M., Luk, A., Hull, M. V., Lustig, K. D., Mangelsdorf, D. J., Shan, B. 1999. Identification of a nuclear receptor for bile acids. Science. 284:1362-1365[Abstract/Free Full Text].

40. Parks, D. J., Blanchard, S. G., Bledsoe, R. K., Chandra, G., Consler, T. G., Kliewer, S. A., Stimmel, J. B., Willson, T. M., Zavacki, A. M., Moore, D. D., Lehmann, J. M. 1999. Bile acids: Natural ligands for an orphan nuclear receptor. Science. 284:1365-1368[Abstract/Free Full Text].