Genomic organization and regulation of expression of the lectin-like oxidized low-density lipoprotein receptor (LOX-1) gene.

Lectin-like oxidized low-density lipoprotein receptor (LOX-1) is a recently identified receptor for oxidized low-density lipoprotein, one of the major atherogenic substances. Although LOX-1 was reported to be expressed abundantly in endothelial cells, including atheromatous lesions, the regulation of LOX-1 gene has not yet been clarified. In the present study, we isolated the rat LOX-1 gene and investigated the regulation of gene expression. The rat LOX-1 gene was encoded by a single copy gene spanning over 19 kilobases and consisted of eight exons. Exon boundaries correlated well with the functional domain boundaries of the receptor protein. The promoter region contained putative TATA and CAAT boxes and multiple cis-elements such as NF-kappaB, AP-1 and AP-2 sites, and a shear stress response element. Northern blot analysis revealed that LOX-1 gene expression was up-regulated 9-fold by shear stress, 21-fold by lipopolysaccharide, and 4-fold by tumor necrosis factor-alpha, in cultured vascular endothelial cells. LOX-1 was also expressed in macrophages but not in vascular smooth muscle cells. These data provide important information for elucidating the molecular mechanisms of LOX-1 gene regulation and suggest a role for LOX-1 in the pathophysiology of atherosclerotic cardiovascular disease.

Oxidized low-density lipoprotein (OxLDL) 1 is implicated in the pathogenesis of atherosclerotic cardiovascular disease (1,2). Previous studies indicate that OxLDL is present in atherosclerotic lesions (3) and that antioxidant drugs slow the progression of atherosclerosis (4,5). OxLDL possesses many atherogenic properties. First, OxLDL is thought to be taken up into macrophages via scavenger receptors, which promotes the deposition of lipid-laden foam cells in the vascular walls and leads to fatty streaks (1). Second, recent data indicate that OxLDL alters various endothelial functions. OxLDL induces endothelial expression of several proteins, including adhesion molecules (6,7), monocyte chemotactic protein-1 (8,9), smooth muscle growth factors (10), and colony-stimulating factors (11), some of which might be involved in endothelial cell-mediated recruitment of monocytes/macrophages into the intima. Ox-LDL also attenuates the endothelium-dependent vasodilatory response through reduced production of nitric oxide (12,13).
It has long been thought that there is a specific endothelial receptor for OxLDL. In 1997, Sawamura et al. (14) identified a novel receptor for OxLDL (LOX-1) using expression cloning with bovine cultured endothelial cells. LOX-1 is a membrane protein abundantly expressed in endothelial cells. It binds, internalizes, and degrades OxLDL, but not native LDL or acetylated LDL. Its mRNA was shown to be expressed in human atheromatous lesions. This endothelial receptor might mediate some of the actions of OxLDL in the endothelium. The biologic roles of LOX-1, however, remain to be determined.
In a previous study, we performed rat LOX-1 cDNA cloning and demonstrated that it encodes a single-transmembrane protein with its N terminus in the cytoplasm (15). The extracellular region consists of a spacer, 46-amino acid triple repeats, and C-type lectin-like domains. Quite unexpectedly, LOX-1 expression was markedly (Ͼ20-fold) up-regulated in the aorta of hypertensive rats (16), which implies a pathophysiologic role for LOX-1 in hypertension or in hypertensive vascular remodeling. However, the genomic structure of LOX-1 or its regulation of expression in vitro has not yet been reported in any species. Here we report the genomic organization of rat LOX-1 and identified several consensus sequences in the 5Ј-flanking region. We demonstrated that the LOX-1 gene expression was markedly up-regulated by shear stress (9-fold), bacterial lipopolysaccharide (LPS) (21-fold), and tumor necrosis factor (TNF)-␣ (4-fold), in cultured vascular endothelial cells. We also examined its expression in cells other than endothelial cells, such as macrophages and vascular smooth muscle cells.

EXPERIMENTAL PROCEDURES
Screening of the Genomic Library-A rat genomic DNA library (Sau3AI partial digest constructed in the phage, EMBL3 SP6/T7) (CLONTECH) was screened with a rat LOX-1 cDNA probe (432-bp XhoI/HindIII fragment, probe 1 in Fig. 1A). Approximately 5 ϫ 10 5 phages were plated at a density of 30,000 plaques/15-cm plate, and two replica nitrocellulose filters were prepared from each plate. High stringency screening was performed with hybridization in 50% formamide and final washes in 0.1 ϫ SSC (1 ϫ SSC is 150 mM NaCl, 15 mM sodium citrate, pH 7.0) containing 0.1% SDS at 60°C. Plaques that produced positive signals on both replicas were selected and purified. The second and the third rounds of screening were carried out under the same conditions to isolate positive clones. Positive clones were digested with XhoI, EcoRI, SacI, HindIII, and ApaI. Each restriction fragment was subcloned into pBluescript II SK Ϫ .
Sequence Analysis-Nucleotide sequences were determined on both strands using the dideoxynucleotide chain-termination method with a Sequi Therm Long-Read cycle sequencing kit (Epicentre Technologies, Madison, WI) and an Automated Laser Fluorescent DNA Sequencer (model 4000; LI-COR, Lincoln, NE). The sequence and consensus nucleotide motif analyses were performed using the GENETYX-MAC software (Software Development, Tokyo, Japan).
Southern Blot Analysis-Southern blot analysis was carried out according to Sambrook et al. (17). Rat genomic DNA (20 g each) was digested with restriction enzymes (PstI, EcoRV, EcoRI, and BamHI), electrophoresed on a 0.7% agarose gel, and transferred onto a nylon membrane filter. An EcoRI/BamHI genomic DNA fragment (1.8 kb, probe 2 in Fig. 1A) was used as a probe. Hybridization and wash were performed as described for the screening.
Primer Extension-Primer extension analysis was performed according to the manufacturer's recommendation (Promega, Madison, WI). In brief, a synthetic oligonucleotide, 5Ј-GCCACATGACTTCTGATCAG-GCTGGCCATT-3Ј (P1) (10 pmol), was end-labeled with [␥-32 P]dATP using T4 polynucleotide kinase. Labeled primer (1 pmol) was annealed to poly(A) ϩ RNA (20 g) from the rat cultured vascular endothelial cells (described below) at 58°C for 1 h and then allowed to cool slowly to room temperature. The hybridized primer-RNA complex was extended using 1 unit of avian myeloblastosis virus reverse transcriptase at 42°C for 1 h. Following ethanol precipitation, the extension products were separated on a 7 M urea/6% polyacrylamide sequencing gel and visualized with autoradiography. The mobility of the extended products was compared with that of a [␥-32 P]dATP dideoxy sequence ladder obtained using the same primer. S1 Nuclease Mapping-A 3.2-kb HindIII fragment of RLG3, which contained the translation initiation site, was subcloned in the HindIII site of pBluescript II and used as a template. A single-stranded antisense DNA was synthesized with this template, end-labeled P1 primer, and T7 DNA polymerase. The products were digested with NheI, purified using alkaline gel electrophoresis, and used as a probe. The probe was hybridized to 20 g of rat endothelial poly(A) ϩ RNA at 30°C overnight. Nonannealed nucleic acids were digested with 1000 units/ml S1 nuclease. The nuclease-resistant products were ethanol-precipitated and electrophoresed in parallel with the primer extension product on a 7 M urea/6% polyacrylamide sequencing gel followed by autoradiography.
Cell Culture-Primary cultures of endothelial cells (ECs) were obtained from the rat aorta or the descending thoracic aorta of a bovine fetus by brief collagenase digestion of the intimal lining, as described previously (18). An established cultured vascular smooth muscle cell line (A10 cells) were purchased from the American Type Culture Collection (Rockville, MD). Cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 50 units/ml penicillin, and 50 g/ml streptomycin in a controlled atmosphere of 5% CO 2 /95% air at 37°C. ECs within the 20th passage were exposed to shear stress (see below), bacterial LPS (Escherichia coli 055:B5) (Difco) or TNF-␣ (Sigma).
Preparation of Alveolar and Peritoneal Macrophages-Six-week-old male Wistar rats (n ϭ 10) were purchased from Tokyo Laboratory Animal Center (Tokyo, Japan). The rats were anesthetized with pentobarbital (50 mg/kg). A 16-gauge elastatic needle was cannulated into the trachea. Alveolar macrophages were recovered by bronchoalveolar lavage with phosphate-buffered saline, 0.1% EDTA. Peritoneal macrophages were harvested by washing the peritoneal cavities with phosphate-buffered saline, 0.1% EDTA.
Flow Loading-ECs were exposed to a well defined steady laminar flow using a parallel plate flow chamber, as described previously (19). In brief, the flow chamber consisted of upper acrylic and lower glass plates, which were separated by a rectangular silicon rubber gasket (250 m thick). ECs were cultured on the glass plate (100 ϫ 70 mm) prior to assembly of the chamber. The inner dimensions of the chamber were 85 ϫ 55 ϫ 0.25 mm. The chamber was filled with culture medium. A peristaltic pump and silicone tubing (ATTO Co., Tokyo, Japan) were used to generate a constant flow. The entire system was placed in an incubator maintained in an atmosphere of 5% CO 2 /95% air at 37°C. The intensity of fluid shear stress (, dyne/cm 2 ) was calculated as follows: ϭ 6Q/a 2 b, where is the fluid viscosity (0.0094 poise at 37°C); Q is the velocity of the fluid (ml/s); a (0.02 cm) and b (1.4 cm) are cross-sectional dimensions of the flow path. Fluid shear stress (20 dynes/cm 2 ) was applied to the ECs for 0 -24 h. Control experiments were simultaneously performed in a static condition with cells derived from the same pool.
Northern Blot Analysis-Total RNA was prepared using the acid guanidinium thiocyanate/phenol/chloroform method (20). Poly(A) ϩ RNA was purified using an mRNA purification kit (Amersham Pharmacia Biotech, Uppsala, Sweden). RNA was fractionated on a formaldehyde-denatured 1.2% agarose gel and transferred to a nylon membrane filter. The probe was synthesized using reverse transcription-PCR with total RNA from ECs and the following primers: 5Ј-CTGGC-TCTGGCATGAAGAAA-3Ј and 5Ј-CGCCTTCTTTTGACATATACTG-3Ј. The amplification was carried out for 30 cycles of 95°C for 1 min, 43°C for 1 min, and 75°C for 3 min with Pfu polymerase (Stratagene, La Jolla, CA). The probe was labeled with [␣-32 P]dCTP using the random primer labeling method. After prehybridization, nylon membranes were hybridized with the 32 P-labeled cDNA probe (1 ϫ 10 6 cpm/ml) in a solution containing 50% formamide at 42°C for 16 h. Blots were then washed in 0.2 ϫ SSC containing 0.1% SDS at 60°C. Filters were exposed to Kodak X-Omat AR5 film with an intensifying screen at Ϫ80°C. For quantitative analysis, the Northern filters were exposed to an imaging plate and the radioactivity of the bands was quantified as photostimulated luminescence with a Bioimage Analyzer (model BAS 2000; Fuji Film, Tokyo, Japan). The photo-stimulated luminescence value for LOX-1 was standardized with that of glyceraldehyde-3-phosphate dehydrogenase (GAPDH).
Anti-LOX-1 Polyclonal Antibody-Rabbit antiserum was raised against rat LOX-1. In brief, the extracellular region of LOX-1 (amino acid residues 60 -364) was subcloned into a (His) 6 -tagged vector. Expression of the recombinant proteins was induced by the addition of isopropyl-1-thio-␤-D-galactopyranoside. The fusion proteins were then purified using a nickel-agarose column. Rabbits were injected subcutaneously with the purified protein (250 g) emulsified with an equal volume of complete Freund's adjuvant (Sigma). For boost injections, the rabbits were immunized with 250 g of the protein emulsified with an equal volume of incomplete Freund's adjuvant every 2 weeks. Bleeding was performed 10 days after the third boost.
Western Blot Analysis-Proteins from extracts of ECs were separated by SDS-polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane. The membrane was incubated with 1:1000 dilution of the rabbit antiserum against rat LOX-1. After washing with Tris-buffered saline containing 0.5% Tween 20, the membrane was incubated with 1:5000 dilution of goat anti-rabbit immunoglobulin conjugated with alkaline phosphatase (Boehringer Mannheim) for 3 h. Proteins were visualized by adding 5-bromo-4-chloro-3-indolyl phosphate and nitro blue tetrazolium.

RESULTS
Isolation of Genomic Clones of Rat LOX-1-We isolated genomic clones encoding rat LOX-1 using a combination of screening and PCR. Fig. 1A summarizes the cloning strategy, genomic structure, and partial restriction map. A rat genomic library was screened with a rat LOX-1 cDNA probe (probe 1 in Fig. 1A). Four genomic clones (RLG1-4) were isolated from 6 ϫ 10 5 plaques. The four clones did not cover the 3Ј end of the LOX-1 gene, therefore, the missing portion was determined using PCR (RLG11-14). The nucleotide sequence was determined using the restriction fragments of the six overlapping clones (RLG2, RLG3, RLG11-14). Comparison of the genomic and cDNA sequences revealed that the coding region  1B). Fig. 1C shows the exon organization in relation to the protein structure. Exon 1 encodes the 5Ј-untranslated region and the N-terminal 25 amino acids in the intracellular domain. Exon 2 encodes 34 amino acids, which is mostly the transmembrane domain. Exon 3 encodes 82 amino acids and corresponds to the juxtamembrane region and repeat 1 in the extracellular domain. Both exon 4 and exon 5 encode 46 amino acids, corresponding to all of repeat 2 (exon 4) and repeat 3 (exon 5). Exons 6 -8 encode 131 amino acids, which is mostly the lectin-like domain. Exon 8 also encodes the 3Ј-untranslated region.
Southern Blot Analysis-Rat genomic DNA was digested with the restriction enzymes PstI, EcoRV, EcoRI, and BamHI and hybridized at high stringency with probe 2 shown in Fig.  1A. A single band was detected in each lane, indicating that rat LOX-1 is encoded as a single copy gene (Fig. 2).
Determination of Transcription Initiation Site-To identify the transcription start site, we first performed primer extension analysis using a 30-mer oligonucleotide P1 (complementary to nucleotides ϩ31 to ϩ60, relative to the ATG codon). The oligonucleotide was end-labeled with 32 P, hybridized to RNA, and extended by reverse transcription. As shown in Fig. 3, a single extension product of 122 bases was obtained with endothelial RNA (lane 1) but not with control yeast tRNA (lane 2). The accompanying sequence ladder revealed that the band was located 62 nucleotides upstream of the translation start site and corresponded to a guanine residue. To confirm this result, we next performed an S1 nuclease mapping assay using the same RNA and primer used for primer extension. The result from S1 nuclease mapping was in accordance with that from primer extension analysis; a protected fragment of 122 bases was observed with endothelial RNA (lane 3). Thus, the position 62 nucleotides upstream of the translation initiation site was most likely a transcription start site and therefore designated as ϩ1.
Sequence Analysis of Rat LOX-1 5Ј-Flanking Region- Fig. 4 shows the nucleotide sequence of 4.8 kilobases of the 5Ј-flanking region and the first exon of rat LOX-1. The promoter region contained a putative TATA box (tattta) at Ϫ30 to Ϫ25 and a CAAT box at Ϫ97 to Ϫ93. On the other hand, the promoter did not contain GC-rich regions. Computer analysis further identified several potential transcription factor binding sites; six

FIG. 2. Genomic Southern blot analysis of rat LOX-1.
Rat genomic DNA (20 g each) was digested with the indicated restriction enzymes. Southern blot analysis was performed at high stringency with a 32 P-labeled rat LOX-1 genomic DNA probe (probe 2 as illustrated in Fig. 1A).
Regulation of LOX-1 Gene Expression by Mechanical Stress, Endotoxin, and TNF-␣-We examined the effects of shear stress, bacterial LPS, and TNF-␣ on LOX-1 gene expression. In the first experiment, confluent monolayers of ECs were exposed to steady laminar shear flow (20 dyne/cm 2 ). Cells were harvested at 0, 1, 3, 6, 12, 24 h and assayed for LOX-1 mRNA with Northern blotting. As a control, cells were incubated without flow stimulus for 24 h. As shown in Fig. 5 (left), we detected a 2.4-kb band for LOX-1 mRNA in ECs. Incubation of ECs in a static condition for 24 h did not produce any significant changes in LOX-1 gene expression. In contrast, fluid shear stress induced a marked time-dependent increase in LOX-1 message levels. The amount of LOX-1 mRNA began to increase within 3 h of shear stress and reached a maximum (9-fold increase over basal levels) at 6 h. The mRNA level declined to a 3-fold increase over basal levels at 24 h.
Next, ECs were exposed to LPS (100 ng/ml) and TNF-␣ (10 ng/ml). LPS induced a dramatic increase in LOX-1 gene expression. The level increased up to 21-fold at 6 h and declined to 16-fold over basal levels at 24 h (Fig. 5, middle). TNF-␣, on the other hand, increased LOX-1 expression to a much lesser extent (4-fold). The level remained high, however, over 24 h (Fig.  5, right).
The LPS-induced LOX-1 regulation was also examined at the

FIG. 3. Identification of transcription initiation site of rat LOX-1 by primer extension and S1 nuclease mapping analyses.
For primer extension analysis, a synthetic oligonucleotide primer (30 bases) was labeled with [␣-32 P]dATP, hybridized to endothelial poly(A) ϩ RNA (20 g) (lane 1) or control yeast tRNA (lane 2), and cDNA was synthesized with avian myeloblastosis virus reverse transcriptase. For S1 nuclease mapping, the labeled probe was hybridized to endothelial poly(A) ϩ RNA (20 g) (lane 3). Nonannealed nucleic acids were digested with S1 nuclease. The extended products and the protected fragments were electrophoresed on a 7 M urea/6% polyacrylamide sequencing gel followed by autoradiography. The specific products are shown with an arrow. G, A, T, and C are DNA sequencing ladders obtained using the same primer for size estimation. protein level (Fig. 6). Western blot analysis with the rabbit antiserum against rat LOX-1 detected a band of ϳ45 kDa in the extract from control ECs (lane 3). The intensity of the band increased after 12 h of LPS exposure (lane 4). Control preimmune serum detected no band (lanes 1 and 2). We also confirmed the specificity of the band by preadsorption of antibody with antigen (data not shown).

LOX-1 Gene Expression in Macrophages and Smooth Muscle
Cells-To examine cell type specificity of LOX-1 gene expression, we performed Northern blot analysis with RNA from peritoneal and alveolar macrophages and cultured vascular smooth muscle cells as well as cultured ECs (Fig. 7). As expected, LOX-1 was expressed in ECs (lane 3), but not in cultured vascular smooth muscle cells (lane 4). Interestingly, LOX-1 expression was abundantly expressed in peritoneal (lane 1) and alveolar (lane 2) macrophages in vivo.

DISCUSSION
In the present study, we isolated and characterized the rat LOX-1 gene. The 5Ј-flanking region contained putative TATA and CAAT boxes and multiple cis-elements, such as NF-B, AP-1 and AP-2 sites, and a SSRE. LOX-1 gene expression was markedly up-regulated in response to hemodynamic mechanical force, bacterial endotoxin, and cytokine, suggesting a role for LOX-1 in the pathogenesis of atherosclerosis.
LOX-1 is a recently identified OxLDL receptor abundantly expressed in endothelial cells (14). It is a type II single-transmembrane protein with a cytoplasmic N terminus. It was first isolated in bovine and human, and subsequently our group identified the rat counterpart (16). Previous data revealed that the LOX-1 protein binds, internalizes, and degrades OxLDL specifically. The protein does not appear to couple to G protein, and does not possess definite kinase or cyclase domains or serine, threonine, or tyrosine phosphorylation sites. The functions of this receptor, including the signaling pathways, are yet to be elucidated. In the present study, we determined the genomic structure of the rat LOX-1 gene. ture was also reported in the extracellular domain of the LDL receptor (21). Deletion and mutation analyses of this unit are necessary to elucidate the structure-function relationship.
In this study, we determined a major transcription start site of the rat LOX-1 gene in ECs. Primer extension and S1 mapping analyses of RNA isolated from rat ECs clearly demonstrated that transcription began 62 bp upstream from the start of translation. The start site was 25 bp downstream of the TATA-like box and 93 bp downstream of the CAAT-like box. On the other hand, there was no GC box. Other start sites might exist in aorta, because one of the cDNA clones we isolated from a rat aortic cDNA library (RL9) had longer 5Ј-untranslated region (16). However, reverse transcription-PCR with multiple primer sets suggests that most transcripts were initiated from the site we identified in this study, at least in cultured ECs (data not shown).
Sequencing of genomic clones encoding up to 4.8 kb of the promoter region of the rat LOX-1 gene revealed the presence of several potential cis-elements, such as NF-B, AP-1 and AP-2 sites, and a SSRE. Although further studies are necessary to determine whether these elements are functional or not, the promoter structure suggests several transcriptional regulation of LOX-1 gene. First, the presence of the NF-B site suggests its regulation by LPS and cytokines (22). Second, the presence of AP-1 and NF-B binding sites as well as SSRE suggests the regulation by shear stress. The SSRE was the first identified cis-element responsible for the activation of PDGF-B gene by shear stress (23), which was subsequently found in other shearinducible genes. Transcription factors AP-1, NF-B, Sp1, and Egr-1 have also been shown to be involved in the shear-induced up-regulation (24 -27). Indeed, LOX-1 expression was markedly up-regulated by these stimuli. We reported previously that the rat LOX-1 cDNA contained multiple AϩU-rich elements in the 3Ј-untranslated region that may be involved in the rapid degradation of mRNA (16). However, we found that, in the presence of actinomycin D, neither LPS nor shear stress altered the half-life of LOX-1 mRNA (data not shown). Thus, it is likely that the up-regulation of LOX-1 mRNA in response to stimuli such as LPS and shear stress was caused by enhanced transcription, not by stabilization of LOX-1 mRNA.
The biologic role of LOX-1 has not been determined yet. It may be involved in atherosclerosis, via the actions of OxLDL in the endothelium, such as induction of adhesion molecules (6,7) and growth factors (10). The up-regulation of LOX-1 by the stimuli such as LPS, TNF-␣, and shear stress appears to be quite reasonable in this context. Unexpectedly, LOX-1 was expressed not only in the endothelium but also in macrophages in vivo. LOX-1 mRNA was previously demonstrated to be expressed in highly vascularized organs such as the placenta, lung, kidney, and vasculatures (14,16). On the other hand, the expression was very low in the brain, heart, adrenal, and other organs. It suggests that LOX-1 is specifically expressed in ECs and macrophages, which play important roles in the atherosclerotic lesion formation. Of interest, the expression of scavenger receptor (SR-AI), another OxLDL receptor abundantly expressed in macrophages, was reported to be down-regulated by LPS and TNF-␣ (28). It is in contrast to the up-regulation of LOX-1 by these stimuli. Recently, CD36, another receptor for OxLDL, was found to be involved in atherosclerosis through PPAR␥ (29,30). Furthermore, Adachi et al. (31) identified a new type of endothelial scavenger receptor (SREC). Like other scavenger receptors, it binds acetylated LDL with high affinity, which was partially displaced by OxLDL. Thus, it can be speculated that OxLDL and its receptors on ECs and macrophages constitute a complex network, which interactively leads to the atherosclerotic vascular lesion development.
We reported previously that LOX-1 gene expression is mark-edly enhanced in the aorta of hypertensive rats, stroke-prone spontaneously hypertensive rats, and salt-loaded Dahl salt-sensitive rats, as compared with the low level in control rats (15). This finding implicates a role for LOX-1 in the pathophysiology of hypertension. On the other hand, LOX-1 expression is not enhanced in prehypertensive young stroke-prone spontaneously hypertensive rats and salt-unloaded Dahl salt-sensitive rats (16). Thus, the up-regulation might be induced by hemodynamic, humoral, or other factors related to the hypertensive state. The present study demonstrated that, in vitro, LOX-1 expression was markedly increased by mechanical factor. Hence, it is possible that the mechanical factor has a strong influence on LOX-1 expression in hypertensive rats. Alternatively, enhanced LOX-1 expression might reflect the hypertensive vascular remodeling.
In conclusion, we determined the genomic structure of the rat LOX-1 gene. The up-regulation of LOX-1 gene expression by biomechanical stress and cytokines, as well as the presence of NF-B, AP-1 and AP-2 sites, and SSRE suggest a role for LOX-1 in the pathophysiology of atherosclerosis.