A Novel Hepatointestinal Leukotriene B4 Receptor

Leukotriene B4 (LTB4) is a product of eicosanoid metabolism and acts as an extremely potent chemotactic mediator for inflammation. LTB4 exerts positive effects on the immigration and activation of leukocytes. These effects suggest an involvement of LTB4 in several diseases: inflammatory bowel disease, psoriasis, arthritis, and asthma. LTB4 elicits actions through interaction with one or more cell surface receptors that lead to chemotaxis and inflammation. One leukotriene B4 receptor has been recently identified (LTB4-R1). In this report we describe cloning of a cDNA encoding a novel 358-amino acid receptor (LTB4-R2) that possesses seven membrane-spanning domains and is homologous (42%) and genetically linked to LTB4-R1. Expression of LTB4-R2 is broad but highest in liver, intestine, spleen, and kidney. In radioligand binding assays, membranes prepared from COS-7 cells transfected with LTB4-R2 cDNA displayed high affinity (K d = 0.17 nm) for [3H]LTB4. Radioligand competition assays revealed high affinities of the receptor for LTB4 and LTB5, and 20-hydroxy-LTB4, and intermediate affinities for 15(S)-HETE and 12-oxo-ETE. Three LTB4 receptor antagonists, 14,15-dehydro-LTB4, LTB4-3-aminopropylamide, and U-75302, had high affinity for LTB4-R1 but not for LTB4-R2. No apparent affinity binding for the receptors was detected for the CysLT1-selective antagonists montelukast and zafirlukast. LTB4 functionally mobilized intracellular calcium and inhibited forskolin-stimulated cAMP production in 293 cells. The discovery of this new receptor should aid in further understanding the roles of LTB4 in pathologies in these tissues and may provide a tool in identification of specific antagonists/agonists for potential therapeutic treatments.

Leukotriene B 4 (LTB 4 ) 1 is derived as a product of eicosanoid metabolism and is a pro-inflammatory lipid mediator that potently stimulates neutrophil chemotaxis to sites of inflammation (1)(2)(3). LTB 4 is involved in the following events: stimulating immigration of leukocytes from the blood stream (4,5); neutrophil activation leading to degranulation and release of noxious mediators, enzymes, and superoxides (6); inflammatory pain (7); host defense against infection (3); and increased interleukin production (8) and transcription (9). These processes have been implicated in the pathogenesis of a variety of diseases such as inflammatory bowel disease (IBD), psoriasis, arthritis, and asthma (10,11). Considerable efforts have been devoted in the development of antagonists targeting the cell surface receptors, by screening compounds with radioligand binding assays utilizing membrane preparations from cells such as neutrophils. Potential treatments of various inflammatory conditions with these antagonists have been recently illustrated in human and animal models (11)(12)(13)(14)(15).
Extensive studies of LTB 4 and the search for the molecular identity of its receptors have resulted in the recent cloning of a LTB 4 receptor (16) (LTB 4 -R1). This protein is a cell surface receptor and belongs to the G-protein-coupled receptor superfamily containing seven membrane-spanning domains. The LTB 4 receptor binds LTB 4 with high affinity, which in turn leads to intracellular signaling and chemotaxis. Among the major tissues tested, the receptor is expressed abundantly only in peripheral leukocytes (16). In this report, we describe the identification of a novel LTB 4 receptor (LTB 4 -R2) that shares homology with LTB 4 -R1, and the finding that the two receptors are genetically linked. This novel receptor is highly expressed in several peripheral tissues such as liver, spleen, and intestine and binds LTB 4 with high affinity. The ligand-receptor interaction activates the receptor leading to intracellular signal transduction.
Cloning and Sequencing of the New LTB 4 Receptor-The amino acid sequences of known G-protein-coupled receptors were used to conduct a BLAST search against expressed sequence tag data bases. The search identified a 397-base sequence as a putative GPCR fragment (HDPYA90R, see Fig. 1A). A phylogenetic analysis (Wisconsin Package, Genetics Computer Group, Madison, WI) suggested that the sequence was related to a leukotriene receptor covering transmembrane domains 3 through 6. Further computational survey of public data bases identified contiguous sequences that resulted in a composite 2451-base sequence that contained a 888-base sequence at the 3Ј-end, which appeared to be a portion of an open reading frame of a GPCR (Met to TM6, see Fig. 1A). A 3Ј-RACE was performed to obtain the missing 3Ј-portion of the putative open reading frame (ORF) by PCR using the Marathon RACE kit for PCR reactions and human liver Marathon-ready cDNA (CLONTECH) as a template. Primary PCR using oligo347 and AP1 (35 cycles of 94°C for 30 s and 68°C for 3 min), secondary PCR using primers AP2 and oligo348A (35 cycles of 94°C for 30 s, 65°C for 30 s, and 72°C for 2 min), and tertiary PCR using primers AP2 and oligo348B (35 cycles of 94°C for 30s, 65°C for 30 s, and 72°C for 2 min) resulted in a ϳ400-base pair (bp) 3Ј-RACE product (see Fig. 1A). To obtain a full ORF, a 5Ј-primer (oligo358) containing the ATG codon in the 888-base sequence and a 3Ј-primer (oligo359) containing the putative stop codon in the 3Ј-RACE sequence were generated. A PCR with this pair of primers and human liver cDNA as a template (35 cycles of 94°C for 30 s, 65°C for 30 s, and 72°C for 2 min) yielded a PCR product of ϳ1.1 kb (SP9030, Fig. 1A).
Isolation of Genomic Clone-A genomic clone containing both LTB 4 -R2 (SP9030) and LTB 4 -R1 receptors was obtained by PCR screening a human PAC PCRable DNA pool (Genome Systems, St. Louis, MO) with primers 63U and 480L. PCR was performed using PCR Supermix (Life Technologies, Inc) with a thermal cycling of 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s (35 cycles). The size of the intron was determined by PCR using Supermix HiFidelity (Life Technologies) with primers oligo347 and MM311 and the PAC DNA as a template (94°C for 30 s, 55°C for 30 s, and 68°C for 5 min). The resulting PCR product (ϳ4.0 kb) was gel-purified with a Qiaex II gel extraction kit (Qiagen) and partially sequenced (ϳ600 bp) at each of the two ends of the fragment.
Transfection of Cells and Membrane Preparations-COS-7 cells grown in DMEM/10% FCS at 80 -90% confluency were transfected with SuperFect agent (Qiagen) at 20 g of DNA/150-mm plate. 48 h after transfection, medium was changed to Opti-MEM or DMEM/Opti-MEM (1:1)/5%FCS. 72 h after transfection, the cells were washed with 20 ml of phosphate-buffered saline without Ca 2ϩ /Mg 2ϩ and incubated with 10 ml of 10 mM Hepes, pH 7.4, 0.5 mM phenylmethylsulfonyl fluoride, 20 g/ml aprotinin at room temperature for 30 min. The cells were scraped off the plate and vortexed. The cell suspension was then centrifuged at 13,000g at 4°C for 15 min. The pellets were resuspended in 1.8 ml of 50 mM Tris-Cl, pH 7.5, and vortexed. The membranes were homogenized with a 23-gauge needle. The protein concentration of the membrane preparations was determined with the BCA agents (Pierce).
Radioligand Binding Assay-For saturation binding, 150 l of binding assay buffer (30 mM Hepes, pH 7.4, 10 mM CaCl 2 , 10 mM MgCl 2 , 0.05% fatty acid free BSA (w/v), kept cold on ice) containing 24 g of membranes were mixed with 50 l of binding assay buffer containing 0 or 1 M leukotriene in 2% (v/v) Me 2 SO. [ 3 H]LTB 4 (PerkinElmer Life Sciences, 50 nM) was added to the assays at increasing concentrations. The reactions were incubated for 1 h at 4°C with slow rotation. Binding solutions were filtered through Multiscreen FB filters (Millipore) presoaked with 50 l of binding assay buffer for 1 h at room temperature, and the filters were washed twice with 100 l of 50 mM Tris-Cl, pH 7.5 (ice-cold). Fifty l of Microscint fluid was added to the filters and counted for the bound radioligands. For radioligand competition assays, 160 l of binding assay buffer containing membranes were mixed with 20 l of binding assay buffer containing various concentrations of competing compounds in 6% Me 2 SO (v/v). A final 20 l of binding assay buffer containing 1 l of [ 3 H]LTB 4 (PerkinElmer Life Sciences, 50 nM, final concentration 0.25 nM) in 6% (v/v) Me 2 SO was added to start the binding reaction. The final concentration of ethanol as the solvent in the stock LTB 4 solution was 0.5% (v/v). Binding data were analyzed with non-linear regression software (Prism; GraphPad, San Diego, CA).
Intracellular Ca 2ϩ Concentration Measurement-293-EBNA cells grown in DMEM/10% FCS at 80 -90% confluency were transfected with the SuperFect agent (Qiagen). On the next day the cells were trypsinized off the plate and washed with phosphate-buffered saline without Ca 2ϩ /Mg 2ϩ . The cells were then seeded at a cell density of 35,000 cells/100 l of medium into 96-well plates precoated with poly-D-lysine (Becton Dickinson). On the third day, the medium was removed, and 100 l of Hanks' balanced salt solution (without phenol red) containing 4 M of Fluo-3, AM (Molecular Probes), 20 mM Hepes, pH 7.4, 0.1% (w/v) BSA, and 250 mM probenecid were added, and the cells were incubated at 37°C, 5% CO 2 for 1 h. The cells were then washed three times with 150 l of buffer containing Hanks' balanced salt solution, 40 mM Hepes, pH 7.4, and 250 mM probenecid. One hundred l of the wash buffer was added after the final wash, and Ca 2ϩ flux was measured with a Fluorometric Imaging Plate Reader (Molecular Devices) after addition of 40 l of the buffer containing appropriate concentration of ligands.
Cyclic AMP Assay-293 cells were transfected with plasmid DNA and the SuperFect agent. Forty-eight hours post-transfection, 100 ng/ml pertusis toxin (PTX) was added to the cells, which were incubated at 37°C overnight. Immediately prior to assay, the cells were released from the plate by the cell dissociation buffer (Sigma), and pelleted by centrifuge at 2000 rpm, 5 min, at 4°C in a 10-ml Ficon tube. The cells were resuspended and seeded at 50,000 cells/50 l of stimulation buffer/ per well of FlashPlate. Fresh LTB 4 was prepared in Hanks'-HEPES buffer (Hanks' balanced salt solution with 10 mM HEPES, pH 7.4, and 0.2% BSA (w/v), filtered, and stored at 4°C). Fifty l of LTB 4 plus or minus 10 M forskolin was added to the FlashPlate. Standard cAMP ranging from 0 to 1000 pmol/ml were arranged in the same plate. The plate was incubated on a rotating shaker at room temperature for 0.5 h. At the end of the incubation, the assay was terminated by addition of 100 l of the adenylyl cyclase activation FlashPlate detection mixture (PerkinElmer Life Sciences). The FlashPlate was covered and gently agitated on a shaker for 3-5 h. After the development, the FlashPlate was counted on a 96-well counter. A standard curve was prepared, and the counts were converted to mass (pmol of cAMP/ml). Cyclic AMP production data were analyzed with non-linear regression.
Analyses of Northern Blots/Dot Blots-Hybridization to Northern blots and dot blots (CLONTECH) was carried out using a PCR-generated 440-bp DNA fragment from the 5Ј-untranslated region of the open reading frame (ORF) of LTB 4 -R2. The DNA fragment was random prime-labeled with [ 32 P]dCTP, and the blots were hybridized for 14 h in ExpressHyb (CLONTECH) containing ϳ2 million cpm/ml radiolabeled probe. The following day the blots were washed according to the manufacturer's protocol and exposed to Kodak Biomax MS film for 6 days at Ϫ70°C. The films were analyzed for relative expression levels using the MCID M4 image analysis system (Imaging Research, Ontario, Canada). For the known receptor LTB 4 -R1, a 425-bp fragment corresponding to nucleotides 49 -474 of the published LTB 4 open reading frame (16) was generated by PCR. The Northern blot and dot blot analyses were carried out with the random-prime [ 32 P]dCTP-labeled fragment, hybridized overnight at 65°C with multiple tissue dot blots and Northern blots (CLONTECH).

RESULTS
Sequence analysis of the 1.1-kb PCR product resulting from multiple RACE amplification steps identified a putative ORF of 1077 bp (Fig. 1A), which encodes a protein of 358 amino acids (Fig. 1B). Hydrophobicity analysis of the 358-amino acid sequence suggested that there are seven transmembrane-spanning regions. BLAST analysis with the amino acid sequence against the GenBank data base revealed homology of the amino acid sequence to the human leukocyte LTB 4 receptor (LTB 4 -R1, 42%) (16) (Fig. 1B), the human CRTH2 (32%) (17), and the human somatostatin receptor SSTR4 (27%) (18,19). The amino acid sequence of the receptor is only distantly related (ϳ18% homology) to the recently cloned leukotriene D 4 receptor (20,21). The high amino acid sequence homology to LTB 4 -R1 and the presence in all seven transmembrane domains of conserved amino acid motifs indicated that this was a G-protein-coupled receptor. Thus the novel receptor was tentatively termed as LTB 4 -R2.
Alignment of the cDNA of LTB 4 -R2 with the ORF of LTB 4 -R1 revealed that the 5Ј-untranslated region (UTRs) of the transcript (GenBank accession number D89079 (16)) is identical to the coding region at the 3Ј-portion and the immediate downstream sequence in the 3Ј-UTR of the ORF of LTB 4 -R2. This analysis suggests that portions of both LTB 4 receptors could exist on a single mRNA and the two LTB 4 receptors are in the same chromosomal region. A genomic clone containing coding regions of LTB 4 -R2 and LTB 4 -R1 was obtained (PAC clone 159K10) by PCR screening of a human PAC library using primers 63U and 480L in the coding region of LTB 4 -R2 ( Fig.  2A). Direct sequencing of the PAC clone revealed no intervening sequences in the coding region of either receptor, thus both receptors are encoded by intronless ORFs (Fig. 2A). A second PCR using primers oligo347 and MM311 revealed a single ϳ3.6-kb intron 3Ј downstream of LTB 4 -R2 and 5Ј upstream of LTB 4 -R1 ( Fig. 2A). GenBank entry AL096870 is a genomic sequence from chromosome 14. This fragment contains both the LTB 4 -R1 and LTB 4 -R2 genes, as determined by BLAST.
Dot blot and Northern blot analyses were performed to determine the expression of LTB 4 -R2 mRNA in human tissues. A dot blot containing mRNAs from 56 human tissues (CLON-TECH) was hybridized to a 440-bp fragment derived from the 5Ј-UTR of LTB 4 -R2 cDNA. The highest expression of LTB 4 -R2 was detected in liver followed by small intestine, spleen, and fetal liver (20 -40% of that of liver; Fig. 3A). Adrenal gland and pituitary had expression levels between 10 and 20% of those in the liver. All of the other 49 tissues expressed LTB 4 -R2 at 10% of or less than that of the liver level (Fig. 3A), including peripheral leukocytes in which LTB 4 -R1 is highly expressed (16). Using the same fragment as probe to hybridize a Northern blot, a mRNA of approximately 1.6 kb with high abundance was seen in the liver and a weak band in kidney (Fig. 3B). No expression was detected in heart, brain, placenta, skeletal muscle, and pancreas. Employing a quantitative PCR method (Taqman, PE Biosystems) with 27 cDNA preparations generated primarily from human fetal and diseased tissues as templates, the highest expression of LTB 4 -R2 was again detected in fetal small intestine, Crohn's colon, fetal liver, and fetal lung (not shown). As is the case for the novel LTB 4 receptor, the LTB 4 -R1 mRNA appears to be widely distributed in human tissues based on the results of the dot blot. LTB 4 -R1 is most abundant in immune-related tissues, including spleen, peripheral blood leukocytes, and bone marrow. Although there is also low expression of the LTB 4 -R1 mRNA in liver, it is not as prominent as that for LTB 4  in which high expression was seen in peripheral leukocytes and low or no mRNA was detected in other tissues.
Radioligand binding assays were performed to directly test the ability of LTB 4 -R2 to bind LTB 4 . The ORF of LTB 4 -R2 was cloned in expression vector pCR3.1 to form construct pCR3.1-LTB 4 -R2. COS-7 cells were transfected with the construct, and membranes were prepared for a [ 3 H]LTB 4 binding assay. As shown in Fig. 4A, specific binding was observed with the membranes prepared from cells transfected with pCR3.1-LTB 4 -R2; in contrast, no specific binding was seen with membranes prepared from cells transfected with vector alone. Because serum used in the cell cultures may carry low concentrations of LTB 4 (23), radioligand binding assays were performed in parallel with membranes prepared from cells grown for the last 24 h in either serum-free (Opti-MEM) or medium containing 5% (v/v) FCS. No difference in the abilities of the two membrane preparations to bind [ 3 H]LTB 4 were found (Fig. 4A), indicating that the serum used in the experiments did not affect ligand-receptor interaction through potential effects of either desensitization or receptor down-regulation. Saturation radioligand binding assays employing membranes from cells cultured in serum yielded a K d of 0.17 Ϯ 0.07 nM and B max of 70 Ϯ 8 fmol/mg of membrane protein (n ϭ 3) (Fig. 4B). Similar K d and B max values were obtained when membranes were prepared from cells cultured in serum-free medium: K d ϭ 0.21 Ϯ 0.06 nM and B max ϭ 64 Ϯ 7 fmol/mg of protein (not shown).
Other test compounds displayed generally low affinities for the LTB 4 -R2 and LTB 4 -R1 receptors in the competition assay (Table I) Table I), two arachidonic acid derivatives (5-oxo-ETE and 5,6-dehydroarachidonic acid, relative affinities Ͼ 435), and lipoxin-A 4 with affinities of ϳ1400 nM (K i ) for the two receptors (Table I). 6-trans-LTB 4 , a trans stereoisomer of LTB 4 , also bound the receptors with relatively low affinity (Table I). Leukotrienes in the C, D, E, and F families and methyl ester of the leukotriene As, displayed weak or no affinities for LTB 4 -R2, as well as LTB 4 -R1 (Table I). Two CysLT1-antagonists, montelukast (Singulair) and zafirlukast (Accolate) that have high affinities for the LTD 4 receptor (20,21), had affinities at least 4000 times lower than that of LTB 4 for LTB 4 -R2 and LTB 4 -R1 (Fig. 5, Table I).
The ability of LTB 4 -R2 receptor to mediate intracellular signal transduction was examined by measurements of fluorescence of Fluo-3, AM as intracellular Ca 2ϩ flux and by assays of forskolin-induced cAMP production. Interaction of LTB 4 -R2 expressed in 293-EBNA cells with LTB 4 activated cellular Ca 2ϩ release (Fig. 6A), suggesting a functional coupling of LTB 4 -R2 with intracellular G-proteins. In contrast, cells that were mocktransfected with vector alone did not respond to incubation with LTB 4 (Fig. 6A). The ability of the receptor to mediate inhibition of forskolin-stimulated intracellular cAMP was tested in 293 cells expressing LTB 4 -R2. LTB 4 caused a dosedependent inhibition of the cAMP production (Fig. 6B). Nonlinear regression analysis of the data yielded a maximum inhibition of ϳ60% and EC 50 of 58 Ϯ 20 nM (n ϭ 2). Incubation of the cells with 100 ng/ml PTX reversed most of the inhibitory activity, indicating that this is a PTX-sensitive pathway (Fig. 6B). DISCUSSION A thorough understanding of the roles of LTB 4 requires identification and characterization of all the LTB 4 receptors. We have cloned and characterized a novel LTB 4 receptor that is different from and genetically linked to the previously cloned LTB 4 receptor (LTB 4 -R1). The two receptors vary in primary structures displaying only 42% homology and in tissue expression patterns, but both are able to bind LTB 4 with high affinity and are functional in stimulating intracellular signaling. In addition, the two receptors are genetically linked at one locus of the genome.
The cloned LTB 4 -R2 binds LTB 4 with highest affinity among the 31 ligands tested. The binding affinity is comparable to that of LTB 4 for the previously published LTB 4 -R1 receptor ( Table  I), indicating that the receptor is a LTB 4 receptor subtype.
Consistent with this conclusion are the observations that LTB 4 -R2 shares a high homology to LTB 4 -R1 and that the two receptors are genetically co-localized ( Figs. 1 and 2). The ligand affinity of LTB-R2 extends from LTB 4 only to LTB 5 and 20hydroxy-LTB 4 , whose structures are closely related to LTB 4 , suggesting a specific ligand recognition (Table I). High affinity binding of 20-hydroxy-LTB 4 to human PMNs has been observed previously (28); it is more soluble than and expresses functional activities similarly to LTB 4 , suggesting a more important role of 20-hydroxy-LTB 4 in inflammation than LTB 4 (28). Except the moderate affinity of 12-oxo-ETE and 15(S)-HETE for LTB 4 -R2, all the other ligands tested, including analogs of LTB 4 or 5-lipoxygenase products, do not bind the receptor with high affinity (Table I). We note that clinically used LTD 4 receptor antagonists montelukast and zafirlukast have no affinity for the receptor (Fig. 5), consistent with the finding that the two LTB 4 receptors share low homology with the LTD 4 receptor. Although LTB 4 and LTD 4 are derived from a common metabolic pathway, their receptors are nevertheless phylogenetically distant, suggesting the use of a common pathway in regulations of diverse physiological functions.
Activation of the LTB 4 -R2 receptor leads to variable signaling events, including the mobilization of Ca 2ϩ and modulation of intracellular cAMP levels. Activation of the receptor leads to the interaction of the receptor with two classes of G-protein: the Gq class, which mediates cellular Ca 2ϩ release, or the Gi class, which mediates inhibition of forskolin-stimulated cAMP production in a PTX-sensitive manner. The activation of multiple signal transduction pathways suggests different roles for LTB 4 through the interaction with LTB 4 -R2, depending on the availability of G-protein reserves in different cell types. LTB 4 activities may be regulated by differential expression of the receptors in different tissues. The new LTB 4 receptor is highly expressed in the liver, intestine, and spleen, and to some degree, in the kidney (Fig. 3), suggesting that this receptor regulates LTB 4 -mediated functions in these tissues. This expression pattern is in contrast to that of LTB 4 -R1, which is expressed only in the peripheral leukocytes (16). The differential expression patterns (Ref. 16, Fig. 3) suggest the functions of the two receptors are tissue-specific and may be involved in different inflammatory processes associated with immune or hepatointestinal systems. The characteristics of LTB 4 -R2 as an active cell surface receptor that binds radiolabeled LTB 4 saturatably (Fig. 4B) and stimulates intracellular signaling (Fig. 6) support LTB 4 acting as stimulant for cellular functions other than merely being passively metabolized by liver cells.
Relatively high expression of LTB 4 -R2 is consistent with existing evidence implicating functions of LTB 4 in these tissues. Hepatic macrophages secreting LTB 4 attract neutrophils to the liver in rats with septic liver injury (30). In a hepatic ischemia-reperfusion injury model, rat liver LTB 4 levels were increased to levels 50-fold those in control liver, accompanied with increase of plasma alanine aminotransferase activities and polymorphonuclear leukocyte accumulation in the liver (31). Only the concentration of LTB 4 , not LTC 4 and LTE 4 , in plasma and stimulated peripheral blood leukocyte superna- tants of children with hepatitis A infection was elevated, suggesting that LTB 4 may be a critical mediator of hepatitis A virus-induced hepatocellular injury (32). Use of 5-lipoxygenase inhibitor significantly lowers blood LTB 4 level and promotes liver regeneration after hepatectomy with obstructive jaundice (33). Liver non-parenchymal cell production of LTB 4 was higher in rats fed corn oil and ethanol (alcoholic liver) than in animals fed saturated fat and ethanol (no liver injury) (34). Finally, enhanced production of LTB 4 by peripheral blood mononuclear cells in patients was found with fulminant hepatitis (35).
In spleen, intraperitoneal LTB 4 increased the survival rate of mice infected with methicillin-resistant Staphylococcus aureus (36). Guinea pig spleen membrane preparations have high affinity for LTB 4 and contain moderate number of LTB 4 binding sites (37); this receptor has high affinity for LTB 4 and 20-hydroxy-LTB 4 but low affinity for 20-carboxy-LTB 4 (37), a profile similar to LTB 4 -R2 (Table I). In a rat renal ischemiareperfusion injury model, Chinese hamster ovary cells expressing a LTB 4 receptor accumulate along with neutrophils in the postischemic kidney; use of LTB 4 antagonists led to the marked decrease in accumulation of Chinese hamster ovary cells and neutrophils (38). Spontaneously hypercholesterolemic rats, characterized by glomerular infiltration of macrophages, fed a normal diet developed end-stage renal failure in 26 weeks, while those fed a diet supplemented with a LTB 4 antagonist showed normal renal function (39). In the intestinal system, high LTB 4 levels in colonic mucosa in patients with IBD were detected (10), more than 10-fold chemotactic activity was found in homogenates of IBD mucosa than in those of normal colonic mucosa, and only lipid extract fraction co-eluted with LTB 4 was chemotactically active (40). Furthermore, the response to IBD mucosa was inhibited by anti-LTB 4 antisera, suggesting that LTB 4 is an important stimulus to neutrophil chemotaxis in the disease (40). Given that LTB 4 -R1 is not found in either small  4 ] is the concentration of the radioligand used in the assay, K D is the affinity of the radioligand for the receptor (0.2 nM), and EC 50 is determined by non-linear regression analysis of the binding data. Results are represented as mean Ϯ S.E. from two to four independent experiments performed in duplicate. K i /K i (LTB 4 ) denotes relative affinity of ligands in respect to that of LTB 4 . The cDNA of LTB 4 -R1 was generated in a PCR with primers oligo417 and oligo418 with human spleen cDNA as a template. The thermal cycling profile was 93°C for 30 s, 63°C for 30 s, 72°C for 90 s (35 cycles). The cDNA was then cloned in expression vector pCR3.1. intestine or colon (16), LTB 4 -R2 may serve as a more promising therapeutic target for treatment of IBD.
The close genetic linkage between the two LTB 4 receptors ( Fig. 2A) is unique to other GPCR families and may play a role in regulating receptor activation. The genomic organization illustrated in Fig. 2A was confirmed by a more recent Gen-Bank entry of a human genomic fragment, AL096870, which contains both LTB 4 receptor genes. Because the fragment is located on chromosome 14, we mapped the LTB 4 -R1 and LTB 4 -R2 receptors to this chromosome. Although we have observed various transcripts that may contain single receptor of LTB 4 -R1 or LTB 4 -R2, or may contain coding regions of both LTB 4 -R1 and LTB 4 -R2, the molecular mechanism that regulates the formation of those transcripts from the two genes and the regulatory roles of the transcripts in specific physiological functions are still not clear.
Identification of multiple LTB 4 receptors with distinct primary structures, differential pharmacological profiles, and expression patterns points to the potential of differential regulation of LTB 4 effects on a variety of inflammatory diseases. Delineation of pleiotropic LTB 4 receptor activation is essential for developments of disease-and receptor subtype-specific an-tagonists/agonists in various therapeutic areas. It will be necessary to define the specificity of the previously reported LTB 4 antagonists (13-15, 41) toward the two LTB 4 receptor subtypes, should these compounds be successfully developed and safely used in clinic settings. The discovery of the hepatointestinal receptor and illustration of the cellular mechanisms mediated by this receptor should aid in design of further studies to understand the roles of LTB 4 in functions of the tissues and identification of receptor subtype-specific antagonists and agonists.