Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

We have reported that 20-hydroxyeicosatetraenoic acid (20-HETE) is produced in human lungs and is a potent cyclooxygenase-dependent dilator of isolated pulmonary arteries (1). These data raise the possibility that 20-HETE could play a role in control of pulmonary vascular tone. However, many questions remain. The cell types responsible for the formation of 20-HETE in the lung are unknown and this information is essential in determining any possible role(s) of this product in modulating pulmonary vascular or other lung function. Previous immunohistochemical studies of rabbit lungs by Masters and colleagues (2) have indicated the presence of cytochrome P450 (cP450) 4A4 protein in nonciliated cells of the proximal airways and in end capillary endothelial cells. However, a systematic examination of individual cell or tissue types for immunospecific cP450 4A protein and/or messenger RNA (mRNA), with particular attention to small pulmonary arteries (< 500 µm outer diameter), has not been reported. Furthermore, we are unaware of any data addressing the capacity of individual tissue or cell types in the lung to produce 20-HETE when incubated with arachidonic acid (AA). Because 20-HETE is a vasodilator of pressurized pulmonary arteries, we were particularly interested in investigating the capacity of lung vasculature (large and small) and vascular smooth muscle cells (VSMs) to generate 20-HETE. We examined ¹⁴C-AA metabolites and cP450 4A-immunospecific protein of microsomes from large and small pulmonary arteries, airways, peripheral lung tissue, and isolated VSMs and lavaged macrophages. Our data demonstrate that airways, peripheral lung tissue, small arteries, and individual VSMs from small arteries metabolize AA into 20-HETE. Semiquantitative comparisons of cP450 4A microsomal protein revealed greater levels of immunospecific protein in small (external diameter 400 µm and less) versus large vessels. Alveolar macrophages produced a number of products when incubated with ¹⁴C-AA, but they did not produce 20-HETE. Immunospecific protein (detected by a primary antibody raised against rat liver cP450 4A1 that cross-reacts with 4A2 and 4A3, and homologous regions in rabbit cP450 4A4, 4A6, and 4A7) was identified in all of the above tissues. Polymerase chain reaction (PCR) products confirmed that cP450 4A6 and/or 4A7 mRNA are expressed in small arterial tissue.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Reagents

The 17-octadecynoic acid (17-ODYA) used in these experiments was synthesized and kindly provided by Dr. Ortiz de Montellano (UCSF, San Francisco, CA). ¹⁴C-AA was acquired from DuPont New England Nuclear, Wilmington, DE (NEC-661). Isocitric acid (I 1252) and isocitric dehydrogenase (I 2516) were obtained from Sigma Chemical (St. Louis, MO). Western blots were visualized using Renaissance Western Blot Chemiluminescence Reagent (NEL-102A; Dupont New England Nuclear). All chemicals were analytical grade unless otherwise stated.

Separation of Tissue/Cell Types

The protocol for the use of animals was approved by the Animal Care and Use Committee of the Medical College of Wisconsin. Young (2-3 kg; 2-4 mo of age) male New Zealand White rabbits were anesthetized with sodium pentobarbital (50 mg/kg). A tracheostomy tube was placed, and the animals were ventilated with room air 10 ml/kg at a rate of 30 times per minute with a Harvard pump. The chest was opened, and the lungs flushed with 40 ml of low calcium, N-[2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid] (Hepes)-buffered saline injected into the pulmonary artery through the right ventricle (lavage and perfusion solution; see Table 1 for constituents). The flush was followed by 20 ml of buffered saline containing 0.1% Evans Blue. The lungs were lavaged with 40 ml of Hepes- buffered saline (in 2 aliquots), and alveolar macrophages were harvested from the lavaged fluid. After the lavage, the heart and lungs were removed from the chest en bloc and pulmonary arteries and airways were dissected using a microscope. Large pulmonary arteries were identified based upon their relation to the right ventricle and dissected distally in order that small arteries could be distinguished from veins. Small airways were differentiated from vasculature based upon continuity with the trachea and mainstem bronchi, and on coloration (all pulmonary vessels were identified by the Evans Blue, whereas the airways were white). Main pulmonary arteries and vessels proximal to three divisions were considered "large" for purposes of this study, whereas vessels beyond the fifth division (less than 400 µm external diameter) were categorized as small. Vessels and airways were assiduously cleaned of adventitial or other attached lung tissue before microsomal preparation or enzymatic dispersion in order to avoid contamination from other cell types.

Enzymatic Dispersion of Isolated VSMs

Dispersion of vessels into individual VSMs was accomplished using modifications of our previously published protocol (3). Vessels were cut into small pieces and placed in a vial containing bovine serum albumin (BSA; 1 mg/ ml) in low calcium-buffered saline solution at room temperature for 10 min (holding solution; Table 1). The vessel pieces were then transferred to a second vial containing 1.5 mg/ml papain (Worthington Biochemical, Lakewood, NJ) and 1 mg/ml dithiothreitol (DTT; D-0632; Sigma) in holding solution and incubated for 10 min at 37°C. Finally, vessel pieces were transferred to a third vial containing 2 mg/ml collagenase (Blend C-8051; Sigma), 0.5 mg/ml type IV elastase (E-0258; Sigma), and 1 mg/ml type 1-S trypsin inhibitor (T-9003; Sigma) in holding solution. The vial containing the enzyme solution and vascular pieces was placed in a water-jacketed incubator at 37°C and gently stirred. Supernatant fractions (0.5 ml) were collected at 5-min intervals and diluted to 1 ml cold buffered saline solution containing calcium (see Table 1). The procedure was repeated by incubating the remaining vessel fragments with fresh enzyme solution. Cells were identified by specific binding of monoclonal anti- smooth muscle actin Cy3 conjugate (C-1698; Sigma), and absence of anti Factor VIII antigen (Atlantic antibodies #80286; Stillwater, MN).

cP450 Metabolism of AA by Lung and Vessel Microsomes, Isolated VSMs, and Alveolar Macrophages

Microsomes were prepared from homogenates of rabbit lung tissues by differential centrifugation (1). Sequential centrifugations of homogenized tissue at 3,000 × g for 15 min, 9,000 × g for 30 min, and 100,000 × g for 1.5 h were performed to generate a microsomal pellet. Protein was quantified according to the method of Bradford (4). Micro-somal proteins (1 mg/ml, 200 µl final volume) were resuspended in assay buffer (100 mM KPO₄, 1 mM ethylenediaminetetraacetic acid, and 10 mM MgCl₂) and incubated for 60 min at 37°C with 1-¹⁴C-AA (0.125 µCi/ml; 5 µM). VSMs were suspended in Dulbecco's modified Eagle's medium (DMEM; GIBCO #12100-012; GIBCO BRL, Grand Island, NY) containing 0.01% fatty acid-free BSA (A-3350; Sigma) and incubated at 37°C for 20 min with 50 nM A23187 (calcium ionophore; C-7522; Sigma) to deplete endogenous AA. After incubation with ionophore, the cell suspension was centrifuged at 3,500 × g for 15 min, and the cell pellet was washed twice with DMEM before incubation with ¹⁴C-AA. Similarly, lavaged cells were centrifuged and resuspended in DMEM for incubation with labeled arachidonate. Cytochrome P450 assays on isolated cells (VSMs and alveolar macrophages) were performed for 3 h rather than 60 min (which was used for microsomal proteins) due to the comparatively small amount of tissue. Nicotinamide adenine dinucleotide phosphate, reduced (NADPH; 1 mM) and an NADPH-regenerating system containing 10 mM isocitrate and 0.1 U/ml isocitrate dehydrogenase (5) were included in each assay. Assays were performed in the presence of room air. Reactions were terminated by acidification with 1 M formic acid and product was extracted twice with ethyl acetate. The organic phase was back-extracted with 1 ml distilled water, evaporated under nitrogen, and reconstituted in ethanol. Reaction products were separated on a C-18 reverse phase high-pressure liquid chromatography (RP-HPLC) column (Supelco, Bellefonte, PA) using a linear gradient ranging from 100% solvent A (acetonitrile:water:acetic acid, 30:70:1) to 100% solvent B (acetonitrile:acetic acid, 100:1) over 40 min. ¹⁴C-labeled products were detected using a radioactive detector (HPLC, Beckman System Gold Programmable Detector Module #171; Fullerton, CA). Identification of metabolites was based upon coelution with authentic standards. Authentic 20-HETE (derived from pregnant rabbit lungs incubated with ¹⁴C-AA, separated by HPLC, and verified by gas chromatography/mass spectrometry) had a retention time of ~ 22 min in our system.

Western Blot Identification of cP450 4A Protein

Microsomal suspensions were separated by electrophoresis on 10% sodium dodecyl sulfate (SDS)-polyacrylamide gels and transferred to a nitrocellulose membrane. In the case of lavaged cells only, total cell lysates (obtained by sonication of lavaged, centrifuged, and washed cell pellets) rather than microsomal proteins were used for Western blot studies. Nonspecific binding was blocked by incubating the membrane overnight in Tris-buffered saline containing 0.05% Tween-20 (TBS-T) plus 10% nonfat milk. The membrane was then incubated for 1.5 h at room temperature with (1) a polyclonal antibody (1:1,000) to rat liver cP450 4A omega hydroxylase enzyme which cross- reacts with cP450 4A1, 4A2, and 4A3 isoforms (5); or (2) a polyclonal goat antibody raised against rat liver cP450 4A1 and 4A3 (Gentests, Woburn, MA), followed by six washes with TBS-T. Rat isoforms 4A1, 4A2, and 4A3 exhibit high amino-acid sequence homology with rabbit 4A4, 4A6, and 4A7. Because studies with the two antibodies yielded similar results, only data obtained with our primary antibody (5) are shown in results. The membrane was incubated with horseradish peroxidase-labeled secondary antibody (1:2,000), and then visualized using enhanced chemiluminescence. X-ray film was developed on the Kodak (Rochester, NY) XOMAT developer; the X-ray image of the film and the bands within it were scanned on a densitometer. The bands corresponding to the 50-kD molecular weight marker were selected on the computer representation of the scan and, after background correction, the pixel density within each selected band was measured by the computer, providing a means for relative quantitation.

mRNA Detection and Quantitation

A total 100 mg of tissue were added directly to 1 ml of TRIzol reagent (GIBCO), homogenized, and centrifuged. RNA in the aqueous phase was extracted with chloroform and precipitated with isopropanol. The RNA pellet was washed in 75% ethanol and reconstituted in 50 µl diethylpolycarbonate-treated (RNAse-free) water. The concentration of RNA was determined spectrophotometrically. Total RNA, 5 µg, was diluted in RNAse-free water to a final volume of 25 µl. After heating at 65°C for 10 min, the entire mixture was added to 25 µl of reverse transcription (RT) reaction buffer containing: 50 mM Tris (pH 8.3), 75 mM KCl, 7.5 mM DTT, 10 mM MgCl₂, 0.08 mg/ml BSA, and 2.4 mM each dNTP (Ready-to-Go You-Prime First Strand beads; Pharmacia Biotech, Piscataway, NJ). NotI oligo dt primer (5 µg) was added to the reaction and incubated for 1 h at 37°C. Cytochrome P450 4A complementary DNAs (cDNAs) were amplified from the RT product by PCR using 5 µl of RT reaction solution added to 50 µl of PCR reaction buffer containing: 50 mM KCl, 10 mM Tris (pH 8.3), 1.5 mM MgCl₂, and 50 pM of specific primer. Reactions were "hot-started" at 94°C for 2 min followed by 80°C for 5 min, at which point Taq polymerase and primers were added. Reactions were then cycled 40 times between 94°C for 30 s, 60°C for 1 min, and 68°C for 2 min. Final extension was accomplished at 68°C for 7 min. Cytochrome P450 4A primer sequences were: 4A6 forward 5' CCC CAG CCT TCC ACT ACG AC 3' and reverse 5' GAC CTT CTC CAG CTC CCC CTC 3'; and 4A7 forward 5' CTC ACC CCA GCC TTC CAC TAC G 3' and reverse 5' GTC TCA GCG CCT CCT TGA TGC 3' (6). Despite an attempt to identify specific primers for 4A isoforms, homology between nucleotide sequences of 4A6 and 4A7 as determined by GCG (Wise Package Version 9.1; Genetics Computer Group, Madison, WI) searches implies potential for cross-amplification.

Aliquots of PCR reaction products were separated on a 1% agarose gel containing ethidium bromide (0.4 µg/ml) and visualized under ultraviolet light. Gels were imaged using a Fluoroimager (VISTRA SI, Cleveland, OH). For Southern blotting, P450 4A cDNA generated by RT-PCR was separated on an agarose gel, soaked for 10 min in 1 N NaOH and 30 min in 10× standard saline citrate (SSC) (see Table 1 for constituents), and transferred to a 0.2-µm nylon membrane using a vacuum blotter. Following transfer, the membrane was prehybridized at 42°C for 1 h in 6× SSC with 50% formamide, 25 mM sodium phosphate, 10% dextran sulphate, 0.1% SDS, 1× Denhardts solution, 40 µl transfer RNA, and 50 mg/ml single-stranded salmon sperm DNA. The membrane was then hybridized overnight in the same buffer with a fluorescein-labeled internal oligo 4A probe. After hybridization, the membrane was washed twice using 5× SSC plus 0.1% SDS, then twice more with 1× SSC with 0.1% SDS. The membrane was then rinsed in a buffer containing 0.15 M NaCl and 0.1 M Tris base (T 1503; Sigma), and incubated in a blocking buffer for 30 min. The blot was developed using enhanced chemiluminescence (Amersham Life Sciences) which includes a primary antibody to fluorescein tagged with horseradish peroxidase, and visualized using an enhanced luminol reaction which is captured on X-ray film.

Statistics

Data are presented as mean ± standard error of the mean (SEM). Differences between conversion rates of tissues were assessed using one-way analysis of variance for repeated measures followed by a Student-Newman-Keuls test when significant differences were identified. Differences between immunospecific band density in large and small vessel microsomes were determined using a two-tailed paired t test. A P value of  0.05 using a two-tailed test was considered significant.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

The average cellular yield of lavage fluid was 14.1 ± 0.75 × 10⁶ cells per rabbit with 98.6 ± 0.5% viability by trypan blue exclusion (n = 6) and 95 ± 3% macrophages by differential count (n = 4), yields and purities reasonably consistent with those reported by other investigators (9). The yield for isolated VSMs was 4.0 ± 1.1 × 10⁶ with 81 ± 5% viability by trypan blue exclusion.

Representative chromatograms of authentic 20-HETE (purified from pregnant rabbit lungs incubated with ¹⁴C-AA, fraction collected and verified by gas chromatography/ mass spectrometry) and peripheral lung microsomes incubated with ¹⁴C-AA are shown in Figures 1a and 1b. A prominent eicosanoid metabolite in peripheral lung micro-somes had a retention time of ~ 22 min (identical to that of authentic 20-HETE standard; see inset). The production of 20-HETE was potently blocked by the omega hydroxylase inhibitor 17-ODYA (Kd 2.9 µM; data not shown). Conversion rates of substrate into 20-HETE by peripheral lung microsomes were 2.7 ± 0.4 pM/mg protein/min.

View larger version (15K): [in this window] [in a new window]   View larger version (18K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   View larger version (19K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   Figure 1.   Representative chromatograms of cP450 4A assays from different tissue types. (a) Elution of our 20-HETE standard (obtained by incubating pregnant rabbit lung homogenates with 14C-AA and fraction-collecting the peak which was inhibited by inclusion of 17-ODYA in the assay media, verified by gas chromatography/mass spectrophotometry) along with 14C-AA. 20-HETE has an elution time of ~ 22 min in our system. Similarly, a chromatogram of 14C-AA metabolites of peripheral lung tissue (b) shows a peak with an elution time consistent with authentic 20-HETE, as we have previously reported (1). (c) A small peak with an elution time consistent with that of 20-HETE in an assay performed on large artery microsomes. (d) A similar product in small vessel microsomes. (e) 14C-AA products from isolated VSMs; a peak that coelutes with 20-HETE is evident. ( f ) Conversion of substrate to 20-HETE was robust in microsomal proteins from airway tissue. In assays from alveolar macrophages harvested from five rabbits (g), numerous AA products were evident, but none of the preparations exhibited a product with a retention time consistent with 20-HETE. Other assays from these same tissue or cell types showed similar results; see text for quantitative comparisons.

Representative chromatograms of AA metabolites of microsomes from large and small vessels appear in Figures 1c and 1d. When microsomes prepared from large pulmonary arteries (outer diameter ~ 1-5 mm) of four rabbits were incubated with AA, we detected no production of 20-HETE. In the chromatogram shown, a small peak which coelutes with 20-HETE is observed; similar results were observed in microsomes from large pulmonary arteries of one other rabbit. In contrast, microsomes from small pulmonary arteries of all rabbits studied demonstrated robust conversion of ¹⁴C-AA (Figure 1d; 1.1 ± 0.36 pM/mg protein/min).

Figure 1e shows a representative chromatogram of metabolites formed when individual VSMs isolated from small pulmonary arteries were incubated with ¹⁴C-AA. A product with a retention time consistent with that of 20-HETE (~ 22 min) was observed in cells isolated from arteries of four rabbits. A second product which eluted at ~ 7 min was also evident in chromatograms from individual VSMs and small vessel microsomes. The formation of the metabolite which eluted at 7 min was blocked by the inclusion of 10 µM indomethacin, a cyclooxygenase inhibitor (data not shown).

Representative chromatograms of AA metabolites from airway microsomes and lavaged macrophages are presented in Figures 1f and 1g, respectively. Airway tissue converted AA to 20-HETE at a rate of 3.9 ± 0.3 pM/mg protein/min; n = 4, a rate which was roughly 1.5 times that of peripheral lung tissue and 3 times that of microsomes prepared from small vessels. A metabolite which eluted at ~ 7 min, as was observed with small artery microsomes and isolated VSMs, was also evident. In contrast, alveolar macrophages converted AA into a number of products, but none of these metabolites coeluted with 20-HETE (n = 6).

Immunospecific Protein Identification

Microsomal proteins from airway, peripheral lung, large and small vessels, and cell lysates of alveolar macrophages separated electrophoretically and probed with cP450 4A polyclonal antibody appear in Figure 2a. Immunospecific protein bands between 48 and 52 kD were identified in microsomes from all tissue studied, though density of proteins appears to differ among tissues. The densest band appears in microsomal protein from airways, with the least amount of immunospecific protein being present in large vessel microsomes. To compare semiquantitatively cP450 4A proteins in large and small pulmonary arteries, micro-somal proteins of the two tissues were separated, and immunospecific band densities were measured and compared. A representative gel is shown in Figure 2b. Immuno-specific band density was greater in microsomal proteins derived from small compared with large arteries (P = 0.03 paired t test; n = 5).

View larger version (20K): [in this window] [in a new window]   View larger version (30K): [in this window] [in a new window]   Figure 2.   (a) A Western blot of airway, peripheral lungs, large and small vessel microsomes, and alveolar macrophages. A total of 5 µg of protein from each of the samples was loaded onto the gel. Following separation, the membrane was probed with a primary antibody raised against rat cytochrome P450 4A1, 4A2, and 4A3 and visualized with chemiluminescence. Bands of ~ 50-kD size were observed in all samples studied. We did not perform semiquantitative densitometric comparisons of all tissue types because large numbers of samples and studies would be required for reliable comparisons. However, qualitatively similar results were observed with proteins of the same tissue and cell types from two other rabbits, with the highest apparent density of immunospecific protein being evident in the airways, and the lowest band density being observed in microsomal proteins from large pulmonary arteries. (b) A representative Western blot of large and small vessel microsomes (10 µg protein each lane) from two rabbits. Immunospecific protein band density was greater in small vessel than large vessel preparations (n = 5; two paired samples of large and small artery microsomes from two rabbits appear in b).

PCR products amplified from RNA extracted from small arteries were separated on an agarose gel and appear along with a base-pair ladder in Figure 3a. We detected cDNA products around 400 and 700 bp using isoform-specific primers for 4A6 and 4A7. These bands are consistent with the expected size of cDNA products for these isoforms based on published mRNA sequences (10) and PCR product sizes determined by FASTA (Genetics Computer Group). A Southern blot of PCR products probed with an internal oligonucleotide sequence (29 bp) to rabbit cP450 4A7 yielded a single band of the anticipated size (~ 700 bp; data not shown).

View larger version (48K): [in this window] [in a new window]   Figure 3.   PCR products separated on an agarose gel from mRNA derived from small pulmonary arteries. Distinct bands of the predicted size (400 and 700 base pairs, respectively) for 4A6 and 4A7 products are evident. No PCR products for 4A4 were detected. A Southern blot using a distinct internal oligonucleotide probe was positive for the 4A7 PCR product (data not shown).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Cytochrome P450 4A enzymes, which catalyze the NADPH-dependent  hydroxylation of prostaglandins, were first identified in rabbit lungs nearly 20 years ago (11, 12). Expression of cP450 4A4 enzyme and mRNA is markedly induced (> 100-fold) by pregnancy or by progesterone administration (8, 11), but conversion of AA to 20-HETE can be demonstrated in microsomes prepared from the lungs of untreated male rabbits (e.g., Reference 13 and present work). We recently reported that 20-HETE is also produced by human lung tissue, and that it is a potent cyclooxygenase-dependent vasodilator of isolated pulmonary arteries (1). These data support the hypothesis that 20-HETE may function as an endogenous vasodilator in the lung, but raise important questions regarding the cellular source(s) of cP450 4A enzymes and their products. Immunohistochemical studies of cP450 4A4 in rabbit lungs have demonstrated that cP450 4A protein is present in nonciliated cells of the proximal airways. The majority of the labeling in rabbit lungs was found in end capillary endothelial cells (2). These studies did not address the capacity of individual tissue or cell types to convert AA to 20-HETE. In the present study, we examined immunospecific protein, mRNA, and the potential of isolated tissue and cell types in rabbit lungs to synthesize 20-HETE, with specific emphasis on the vasculature because our data implicated 20-HETE as a potential pulmonary vasodilator. We demonstrated the presence of immunospecific protein in airways, lavaged macrophages, peripheral lung tissue, and small and large pulmonary arteries; the greatest cP450 4A protein was found in the airways. Cytochrome P450 4A4, 4A6, and 4A7 proteins all have the capacity to convert AA into 20-HETE (7, 8, 10). Production of 20-HETE was observed when microsomes prepared from airway, peripheral lung, and large and small pulmonary arteries were incubated with ¹⁴C-AA. Of these tissues, the highest conversion rates were seen in microsomes prepared from airway tissues (3.9 ± 0.3 in airway compared with 2.7 ± 0.4 pM/mg protein/min peripheral lung; P = 0.04). Micro-somes prepared from large pulmonary arteries yielded inconsistent and modest synthesis of 20-HETE. These data are in line with semiquantitative estimates of cP450 4A specific protein levels in these same tissue types. Finally, isolated VSMs derived from small pulmonary arteries metabolized AA into 20-HETE. Western blot studies of immunospecific protein in individual VSMs were not performed due to the small yield of tissue from our isolation procedure. However our Western blot experiments confirm that cP450 4A protein is expressed in small pulmonary arteries, presumably in the VSMs or endothelial cells. RT-PCR results indicating that mRNA for 4A6 and 4A7 are expressed in small pulmonary arteries supports the view that these vessels metabolize AA to 20-HETE via the cP450 4A pathway (10). Our investigations do not directly address the contribution of endothelial cells to cP450 4A immunospecific protein or 20-HETE synthesis in the rabbit lung. As stated above, immunohistochemical studies suggest that pulmonary capillary endothelial cells are a rich source of the cP450 4A protein. Capillary endothelial cells necessarily require culturing to obtain enough tissue to assay ¹⁴C-AA products, an intervention which profoundly alters the metabolism of AA (14, 15), thus we did not pursue this line of investigation. However, our observations of relatively dense immunospecific cP450 4A protein and robust conversion rates of ¹⁴C-AA to 20-HETE in peripheral lung tissue (which are a rich source of capillary endothelium) are consistent with a significant endothelial cell contribution to total peripheral lung cP450 4A protein. Our data demonstrate that VSMs, in addition to the previously implicated endothelial cells, have the capacity to synthesize 20-HETE. Similar results have recently been reported using isolated VSMs from the cat cerebral circulation; VSMs from this vascular bed also produce 20-HETE (16). Smooth muscle cells from other vascular beds are also reported to synthesize and metabolize epoxyeicosatrienoic acids (17) and a variety of cyclooxygenase products (18). With respect to cyclooxygenase products, chromatograms from isolated VSMs (as well as airway and peripheral lung) exhibited a peak with retention time of ~ 7 min, which was blocked by the inclusion of indomethacin in the assay media and which eluted 1-2 min later than our 6-keto-PGF₁ standard. This cyclooxygenase product is a candidate for the mediator of indomethacin-inhibitable vasodilatory effect of 20-HETE in isolated pressurized pulmonary arteries (1), though such a function and the identity of this metabolite are not addressed by data in the present work.

Unlike other tissue and cell types we studied, alveolar macrophages exhibited an apparent discrepancy between the abundance of immunospecific cP450 4A protein and no 20-HETE production. The competence of lavaged macrophages in our hands to metabolize AA via any pathway appears intact because our cells produced a number of ¹⁴C-AA metabolites (see Figure 1g). Products with retention times coincident with 6-keto-PGF₁, TxB₂, PGF₂, LTB4, and 5-HETE were observed. These products are consistent with cyclooxygenase and lipoxygenase metabolites of AA in alveolar macrophages reported by other investigators (19). It is possible that a cP450 4A-like protein in alveolar macrophages cross-reacted with our primary antibody, or that the cP450 4A6 and 4A7 synthetic function of macrophage proteins was somehow damaged in our harvesting procedures. However, we speculate that another explanation of our inability to demonstrate 20-HETE synthesis in alveolar macrophages exists: macrophages are a rich source of NO synthase (22), and NO appears to be a potent inhibitor of cP450 4A enzymes (23). Thus NO generated by alveolar macrophages may autoinhibit the P450 4A enzymes and production of 20-HETE in these cells. Additional studies are needed to determine the reasons for disparate immunospecific protein identification and metabolic synthesis in rabbit alveolar macrophages.

In conclusion, we have demonstrated that several tissue types and cells within rabbit lungs express cP450 4A4, 4A6, or 4A7 protein and convert AA into 20-HETE, including large and small pulmonary arteries, airways, and peripheral lung tissue. Lavaged macrophages also express immunoactive cP450 4A protein but do not synthesize 20-HETE when incubated with ¹⁴C-AA. Small pulmonary arteries express cP450 4A6 and 4A7 mRNA which codes for two proteins that produce 20-HETE from AA in rabbit lungs (7). With respect to vasodilatory effects of 20-HETE which we have observed in isolated vessels, we have demonstrated that small pulmonary arteries have cP450 4A immunospecific protein and mRNA, and metabolic capacity to synthesize 20-HETE. Isolated VSMs also synthesize 20-HETE from AA. Beyond the vasculature, our data demonstrate a surprisingly widespread distribution of cP450 4A activity in non-pregnant rabbit lung tissue under basal conditions. These observations are consistent with the high degree of preservation and extensive representation of cP450 4A proteins in other tissues and species (24). Our observations argue that 20-HETE may have more than one physiologic function in the lung or that 20-HETE derived from nonvascular cells could act in a paracrine manner to modify vascular tone.