Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Human lung might contain several secretory phospholipases A₂ (sPLA₂s) that mediate proinflammatory responses in diseases such as the acute respiratory disease syndrome (ARDS) (1) and asthma (2). However, the expression of sPLA₂s in the lung, especially their cells of origin, are unknown.

Phospholipases A₂ (PLA₂s) hydrolyze the sn-2 position of membrane phospholipids, causing release of a free fatty acid (such as arachidonic acid [AA]) and formation of lysophospholipid. PLA₂s have many roles in phospholipid biochemistry within cells, but their roles in release of lipid mediators of inflammation have captured intense interest during the past two decades (reviewed in References 3 and 4). Multiple PLA₂s exist within cells and have been divided into groups based on structural homology and numbered by their order of discovery (reviewed in Reference 5). These include a Group IV 85-kD "cytosolic" PLA₂ (cPLA₂), which is activated by elevation of cytosolic Ca²⁺ and by phosphorylation after receptor-mediated stimulation of diverse cell types (6, 7). cPLA₂ appears to have a central role in AA release leading to eicosanoid formation (8). Two paralogs of the cPLA₂ have recently been described; their roles have yet to be determined (9). A cytosolic calcium-independent Group VI enzyme has also been described that may play a role in membrane phospholipid homeostasis (10, 11). In contrast, secretory PLA₂s are smaller (14 to 17 kD) proteins with multiple disulfide bridges defining a conserved tertiary structure even though homology at the amino acid level is low (about 35%). Over 70 sPLA₂s are known, mainly from snake venoms and bee venoms as well as mammalian cells. The roles of sPLA₂s in humans are currently being defined.

Several sPLA₂ isotypes have been described in mammalian tissues that could be important in human lung. Group IB sPLA₂ was the first sPLA₂ isotype to be cloned from human lung tissue, by Seilhamer and colleagues in 1986 (12). It is better known as the pancreatic enzyme and has largely been thought to play a role in digestion, although increased levels in serum and other organs may arise as a consequence of pancreatitis, which is a recognized cause of ARDS. Group IIA sPLA₂ was identified in and cloned from synovial fluid from patients with rheumatoid arthritis, from peritoneal exudate cells (13), and from platelets (14); hence it has been designated the inflammatory sPLA₂. The Group IIA sPLA₂ may be especially important in antibacterial defense (15). More recently, Group V sPLA₂ was cloned from a human stomach complementary DNA (cDNA) library (16) after its initial discovery in mouse and rat (17). It has been found in mouse bone marrow-derived mast cells (18) and P388D₁ macrophages (19). In these cells it appears to play a significant role in eicosanoid production, previously ascribed only to the Group IIA enzyme. Tissue Northern blots from that first description suggested that Group V enzyme messenger RNA (mRNA) might be found in lung.

Cupillard and colleagues recently described a novel Group X enzyme that was discovered when they analyzed sequence databases for demonstrated homology with other sPLA₂s in a search for the natural ligand to the mammalian sPLA₂ receptor (20). They found a novel enzyme from a human fetal lung cDNA library that, when cloned, possessed calcium-dependent enzymatic activity typical for sPLA₂. Fetal lung was the original source of the cloned cDNA; however, tissue Northern blots in that report showed only a faint reaction in adult lung.

In addition to their well recognized roles in eicosanoid production, certain sPLA₂s could hydrolyze surfactant phospholipids, either as a normal reaction in surfactant turnover or as part of the pathology of lung injury leading to surfactant dysfunction (21, 22). sPLA₂s have been shown to have different binding affinities and activities against different membrane phospholipids (23, 24) and thus sPLA₂ isotypes might impact differently on lung pathology.

To begin to understand the role of sPLA₂s in human lung, we wished to determine whether each sPLA₂ would have a unique cellular distribution within lung tissue. We were especially interested in the expression of the Group X sPLA₂ that has been described only once to date. Because monospecific antibodies are not yet available to differentiate the human sPLA₂s, we chose a reverse transcriptase/polymerase chain reaction (RT-PCR)-based strategy to capitalize on the unique nucleic acid sequence of each mRNA. By sequencing the cDNA products we have been able to identify unequivocally the sPLA₂s expressed in human lung. We have also synthesized complementary RNA (cRNA) probes to determine the differential cellular distribution of each sPLA₂ within lung tissue.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents

Unless noted, all reagents, including high-performance liquid chromatography-grade solvents, were purchased from Fisher (Suwanee, GA) or Sigma Chemical (St. Louis, MO).

Human Lung Tissue RNA

Dr. Timothy Oaks (Wake Forest University School of Medicine) donated human lung biopsy samples according to a protocol approved by the Institutional Review Board. The samples of macroscopically normal tissue were obtained as the extreme peripheral tissue margins of tumor biopsies. Tissues were snap-frozen in liquid nitrogen and stored at  70°C until used.

Total RNA was extracted from lung samples with a modified technique of Chomczynski and Sacchi (25) using RNA STAT-60 (Tel-Test, Inc., Friendswood, TX). Lung tissues were homogenized using 1 ml RNA STAT-60 per 50 to 100 mg tissue in a glass tube with a motorized Dounce homogenizer. The RNA in the aqueous phase was precipitated with isopropanol, washed with 70% ethanol, resuspended in ribonuclease-free water, and stored at 20°C until used.

BEAS Cell Line

The bronchial airway epithelial cell line (BEAS) was obtained from American Type Culture Collection (Bethesda, MD) and grown in serum-free, hormonally supplemented bronchial epithelial cell basal media (Clonetics, Walkersville, MD) in T-75 flasks on a surface of bovine serum albumin (BSA) and fibronectin. Cells were trypsinized for harvest and total RNA was prepared with RNA STAT-60.

RT-PCR

The amount of 1 µg of total RNA was used for each reaction using an RT-PCR Core kit from Perkin-Elmer (Foster City, CA). The RT step was performed at 42°C for 60 min with oligo-dT primers in a volume of 20 µl. The resultant cDNAs were amplified by PCR in a volume of 100 µl using oligonucleotide primers chosen from the published sequences for each sPLA₂ isotype and synthesized in the Wake Forest University School of Medicine Molecular Biology Core Lab. These primers overlapped the natural start and stop codons for each message. Sequences were: Group IB sense primer 5'-CCTTGACTGCAAGATGAAACTCC-3', antisense primer 5'-TCAACTCTGACAATACTTCTTGGTGTCCAGG-3'; Group IIA sense primer 5'-CCCAAGAACTCTTACCATGAAGACCC-3', antisense primer 5'-GACTCAGAACGAGGGGTGCTCCC-3'; Group V sense primer 5'-CCATCGATCCCCAGAGATGAAGGCCTC-3', antisense primer 5'-GCTCTAGATGGTCTGGGAGGAGCTCGC-3'; and Group X sense primer 5'-CCATCGATTCCCACCTCTGCCACCCTCCG-3', antisense primer 5'-GGTCT- AGAGTCAGTCACACTTGGGCG-3'. The PCR consisted of 35 cycles at 94°C for 1 min, 55°C for 30 s, and 72°C for 1.25 min for Groups IB and IIA, with a change in annealing temperature to 63°C for Groups V and X. The resultant cDNAs were analyzed by ethidium bromide-stained agarose gel electrophoresis.

Sequence Confirmation of sPLA₂ cDNAs

Each cDNA was excised from the low-melt agarose gels, purified with Wizard PCR Preps DNA Purification System (Promega, Madison, WI), and sequenced by automated sequencing (ABI-PRISM; Perkin-Elmer) in the Wake Forest University School of Medicine Molecular Biology Core Lab.

cDNA Cloning and Expression of Group X sPLA₂

The Group X cDNA band was excised and isolated from the PCR gel. This cDNA was ligated into a T-overhang vector, pTarget (Promega), transformed into JM109 Escherichia coli for selection on LB/Amp, IPTG/X-gal agar plates, and recombinant plasmids purified by alkaline lysis. Plasmids were screened for inserts with EcoRI digestion. Two Group X clones, designated 6AA and 6B6 were found to have sequences identical to the Group X sequence published by Cupillard and associates (20). Plasmids from each clone were transfected into COS-1 cells using the diethylaminoethyl-dextran method as described elsewhere (26). Culture media were collected 72 to 96 h after transfection and assayed for sPLA₂ enzyme activity. Similar cloning and expression of Group V sPLA₂ was also performed. All four sPLA₂ cDNAs were also cloned into pGEM-T bacterial expression vectors (Promega) for synthesis of probes for in situ hybridization.

sPLA₂ Enzymatic Assay

sPLA₂ was assayed by measuring the hydrolysis of radiolabeled E. coli phospholipids. Assays contained 50 to 100 µl of COS cell supernate and were incubated for 60 min at 37°C in 400 µl final volume containing 50 mM Tris/NaCl, pH 8.5, with 5 mM CaCl₂ and 5 nmol (40 nCi) [³H]arachidonate-labeled E. coli phospholipids (New England Nuclear, Boston, MA). The reaction was terminated after 60 min by acidified Bligh and Dyer extraction (27). Dried reaction extracts were resuspended in a 1:1 mixture of CHCL₃/ CH₃OH and samples spotted on silica gel G thin-layer chromatography plates, resolved with a mobile phase of hexane/ether/formic acid, 90:60:6, vol/vol/vol, and visualized by iodine vapor relative to cold standards. Bands were scraped for scintillation counting. Preliminary experiments determined that these assay conditions were optimal for linearity regarding enzyme protein concentration and assay duration. Dipalmitoyl glycerophosphocholine (DPPC) was used as a model of surfactant phospholipid. DPPC, 34 nmol per 400-µl assay tube spiked with a trace of ³[H]DPPC (L-A-dipalmitoyl [2-palmitoyl-9, 10-³H(N)]-phosphatidyl-choline), was sonicated in a buffer of 5 mM Tris/HCl (pH 8.5), 5 mM CaCl₂, and 150 mM NaCl. The assay was otherwise run as described for E. coli.

In Situ Hybridization

For probing of mRNA, slides of formalin-fixed, paraffin-embedded sections of human lung were dewaxed by heating to 60°C, transferred to a fresh xylene-substitute bath, and rehydrated into phosphate-buffered saline (PBS) with 0.1 M glycine. Tissues were permeabilized by sequential treatment in 0.3% Triton X-100 (twice for 5 min each) and 0.1% pepsin in 0.2 M HCl (20 min at 37°C), rinsed in PBS, and postfixed for 5 min in 4% paraformaldehyde. Sections were acetylated in 0.1 M triethanolamine buffer containing 0.25% acetic anhydride. After prehybridization in 4× sodium chloride/sodium citrate (SSC) with 50% deionized formamide, sections were incubated overnight at 47°C in a moist environment with cRNA probes in hybridization buffer containing 1× SSC with 40% deionized formamide, 10% dextran sulfate, 1× Denhardt's solution, 1 mg/ml yeast transfer RNA, 10 mM diothiothreitol, and 1 mg/ml denatured and sheared salmon sperm DNA. Digoxigenin-labeled cRNA probes (digoxigenin-deoxyuridine triphosphate; Boehringer Mannheim, Indianapolis, IN) were synthesized using SP6- or T7-primed in vitro transcription of the sPLA₂ pGEM-T clones according to the manufacturer's instructions. Probes were quantitated with DIG Quantification Test Strips (Boehringer Mannheim) and used at equal concentrations. Lung sections from four different biopsy samples were probed with antisense cRNA probe (e.g., T7 promoter in pGEM-T clone of Group X). Results were confirmed using sense probe (SP6 transcripts from the same plasmid) as negative controls. Sections were covered with 50 µl each and coverslipped. After hybridization, slides were washed with a series of SSC buffers with increased stringency, with a final wash in 2× SSC with 50% formamide, and transferred into an antibody blocking buffer (Tris- buffered saline containing 0.5% wt/vol casein plus 0.1% Tween 20). Sections were probed with antidigoxigenin alkaline phosphatase-conjugated Fab₂ fragments (1:50 in PBS containing 0.1% BSA) at room temperature for 2 h, rinsed, and color-developed with nitroblue tetrazolium (NBT)/5-bromo-4-chloro-3-indolyl-phosphate (BCIP) solution for visualizing stained cells by brightfield microscopy. Sections were visualized on a Zeiss Axioplan microscope (Zeiss, Thornwood, NY) and photographed with a Spot CCD camera and computer software from Diagnostic Instruments (Sterling Heights, MI). For positive control tissues, sections of African green monkey pancreas were used for Group IB (a gift from Dr. Manuel Jayo, Department of Pathology). Sections of synovial tissue from a patient with rheumatoid arthritis were used as positive controls for Group IIA sPLA₂ (a gift from Dr. Paul DiCesare, New York University School of Medicine, New York, NY). Designated sections of these tissues were individually stained with Diff-Quik (Baxter Scientific Products, McGaw Park, IL) to resolve tissue structure.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Four sPLA₂s Are Expressed in Human Lung

We selected an RT-PCR approach to identify each isotype of sPLA₂ in human lung because it allows unambiguous sequence identification of cDNA products, it is more sensitive than Northern blotting, and it was a first step to cloning of the proteins for further study. A result of RT-PCR for Groups IB, IIA, V, and X sPLA₂ from adult human lung is shown in Figure 1. cDNAs at the expected base-pair (bp) sizes are seen in each lane when primers are selected just distal to the natural start and stop codons, respectively. The quantitative relationship between these mRNAs has not been established because RT-PCR was not run with quantitated competitive internal standards. However, Group IIA cDNA was consistently observed as the signal with the most variable intensity, even though the RT-PCR was optimized for amplification of the Group IIA sequence using these primers with HEP G2 hepatoma cell mRNA as a positive control (not shown).

View larger version (85K): [in this window] [in a new window]   Figure 1.   Four sPLA2 genes are expressed in human lung tissue. RT-PCR was performed to identify sPLA2 expression in 1 µg of total human lung RNA using primer sequences specific for each of Groups IB, IIA, V, and X. The resultant cDNAs were analyzed by agarose gel electrophoresis with ethidium bromide staining. Each lane depicts typical results (n = 4-6 experiments for each isotype) showing cDNA products at the expected bp size for the full coding region of each mRNA (just under 500 bp). Conditions for the RT-PCR are given in MATERIALS AND METHODS.

Each band was excised from the agarose gel, purified, and sequenced in both directions using the original PCR primers. Sequence pairs were compared using the Wisconsin GCG GAP program and any sequence discrepancies resolved visually from the ABI Prism graphic output. Sequences from three to six experiments for each sPLA₂ cDNA were then compared with those published sequences registered with GenBank/EMBL database using BLAST programs from the NIH. In each case, the sequences were identical to the published sequences for all four sPLA₂ coding regions (not shown). This identification is especially important for the Group V and Group X sPLA₂s, which were only faintly positive on lung Northern blots in the original reports, respectively (16, 20). Also, this is novel independent confirmation of the existence and sequence of the Group X enzyme described by Cupillard and colleagues (20), and extends the localization of Group X and V enzymes firmly to adult human lung.

Expression of sPLA₂s in BEAS Bronchial Epithelial Cells

We used the normal human bronchial epithelial airway cell line BEAS as a model to test the hypothesis that sPLA₂ isotypes may follow cell-specific expression in human lung. BEAS cells were grown as an adherent monolayer in fibronectin-coated flasks in their normal hormonal-supplemented, serum-free culture media. Total RNA was extracted and analyzed by RT-PCR using the same sPLA₂ group-specific primers shown earlier for total lung. The results in Figure 2 show that Group X and Group IIA sPLA₂ mRNAs were readily amplified from the airway cells. The cDNA bands were excised and sequenced and the results confirmed that the cDNAs were identical to the known Group X and Group IIA sPLA₂. We did not detect Group IB in BEAS cells. The Group V cDNA band was seen consistently in each preparation, but it has not been recovered in sufficient quantity to confirm its sequence identity.

View larger version (80K): [in this window] [in a new window]   Figure 2.   The BEAS epithelial cell line expresses Group IIA, V, and X sPLA2. The normal human airway epithelial cell line (BEAS) was examined by RT-PCR for each sPLA2 gene product. The resultant cDNAs were analyzed by agarose gel electrophoresis as described in MATERIALS AND METHODS. Note that prevalent bands are seen for Group IIA and Group X. Group V products were consistent, though weaker. Group IB products were absent in all three cell preparations examined.

Detection of Human Group X and Group V sPLA₂ mRNAs in Lung Epithelial Cells by In Situ Hybridization

Using the cloned cDNA for each human sPLA₂, we synthesized cRNA probes for in situ hybridization. Preliminary experiments revealed a pattern of staining consistent with an epithelial cell expression of Group X in adult human lung (Figure 3). The dark purple stain marks the localization of the digoxigenin-cRNA Group X probe:antidig alkaline phosphatase complex, which was visualized with an NBT/BCIP substrate without counterstains (arrow in Figure 3B). The stained cells are localized in the columnar-to-cuboidal-to-squamous cell transitions of distal airways. These cells did not stain with the sense cRNA probe (arrow in Figure 3A). The faint staining of muscle and connective tissue is typical for this substrate. The nonspecific staining (brown-tinged) in the leukocytes in the small airway (upper right of Figures 3A and 3B) was present for either probe. The figures shown are representative of four different lung specimens. A similar pattern of staining was obtained for Group V sPLA₂ mRNA, which was found distinctly in airway epithelial cells (arrow in Figure 3D).

View larger version (135K): [in this window] [in a new window]   Figure 3.   Detection of human Group X and Group V sPLA2 mRNA in lung epithelial cells by in situ hybridization. Sequential paraffin-embedded sections of human lung were incubated with digoxigenin-labeled cRNA probes for human Group X sPLA2 (A and B) or Group V sPLA2 (C and D) and developed with alkaline phosphatase-conjugated, antidigoxigenin antibodies. A and C show the lack of reactivity when sense probes were used. The dark stain in B (Group X) and D (Group V) marks the localization of NBT/BCIP substrate reaction products that were most prevalent in the epithelial cells of distal airways (arrows). Some nonspecific staining of granular leukocytes was noted in some airways (upper right of A and B). Two different lung specimens are shown here, which are typical of four lungs examined. A and B are approximately 1,200 microns across; C and D are 400 microns across.

In contrast, probes for Group IB and IIA sPLA₂ mRNA failed to stain airway epithelium (Figures 4B and 4D, respectively) and did not show remarkable differences between antisense and sense probes in any lung cells. The integrity of the antisense probes was demonstrated by testing them with positive control tissues known to express the Group IB gene (pancreas; Figures 5A-5C) and the Group IIA gene (synovial epithelium from a patient with rheumatoid arthritis; Figures 5D-5F) Both the pancreas and the synovial cells reacted positively with the antisense probes for Group IB sPLA₂ (Figure 5A) and Group IIA sPLA₂ (Figure 5D), respectively. Again, the sense probes provided negative controls for these experiments (Figures 5B and 5E). Thus, airway epithelium expressed the human Group X and Group V sPLA₂ genes but not Group IB or Group IIA.

View larger version (133K): [in this window] [in a new window]   Figure 4.   Lung epithelial cells do not contain detectable Group IB or Group IIA sPLA2 mRNA. Airway epithelial cells (arrows) did not bind antisense cRNA probes for Group IB sPLA2 (B) or Group IIA sPLA2 (D). Negative controls with sense probes are shown in A and C for Groups IB and IIA, respectively. Panels are 1,200 microns each.

View larger version (87K): [in this window] [in a new window]   Figure 5.   Positive control cells demonstrate the integrity of the antisense cRNA probes for Group IB (A) and Group IIA (D) sPLA2 mRNAs. Pancreatic acinar tissue stains for Group IB mRNA in A, whereas the islet cells do not (nonstaining cells in A and light cells in Diff-Quik stain in C). Synovial epithelium stains positively for Group IIA mRNA (D). The synovial epithelium is an irregular epithelium shown both in longitudinal (midline diagonal) and transverse sections (upper right-hand portion of D-F; note the cell structure in the Diff-Quik-stained section shown in F). Negative control panels with sense strands are shown in B and E. All panels are about 500 microns each.

sPLA₂ Isotypes Hydrolyze Substrates with Different Efficiencies

The genetic redundancy of lung tissue expression of several sPLA₂ enzymes suggests a functional specificity that could result not only from the cell-specific expression seen earlier, but also from substrate specificity in vivo. As a test of this hypothesis, the recombinant proteins were examined for their ability to hydrolyze two different substrates commonly reported in sPLA₂ assays, in vitro. Recombinant Groups IIA, V, and X were expressed in COS-1 cells for 72 h and the supernates assayed for the ability of the secreted recombinant enzymes to hydrolyze phospholipids from either E. coli membranes (comprised of mixed phospholipid labels, especially phosphatidylethanolamine) versus sonicated vesicles of pure DPPC (Figure 6). For comparison, we utilized pig pancreatic sPLA₂ as the mammalian Group IB protein. The assays were run simultaneously for all enzymes and substrates, and were optimized for linear hydrolytic rates to adjust for differences in enzyme concentrations from each transfection experiment (using 20 to 50 µl of transfected cell supernates). Mock-transfected COS-1 cells were used as controls and found to express no sPLA₂ activity (not shown). The activity of each sPLA₂ enzyme is shown for both E. coli hydrolysis and for pure DPPC vesicles. The ratio of hydrolysis of DPPC to E. coli phospholipid provides an index of substrate hydrolysis, independent of differences in enzyme concentration. Key metabolic differences were seen among these enzymes. For example, Group V sPLA₂ preferentially hydrolyzed DPPC, with the least activity against E. coli phospholipids. In contrast, recombinant Group IIA sPLA₂ had the least activity against DPPC, yet it readily hydrolyzed E. coli phospholipids, reflected in the low hydrolysis ratio. Groups IB and X readily hydrolyzed either substrate. These findings support the hypothesis that each sPLA₂ isotype may have distinct functional roles in lung tissue related to substrates hydrolyzed in vivo. These data also indicate that recently described Group V and X enzymes are capable of significant hydrolytic rates of DPPC, the major phospholipid of surfactant.

View larger version (31K): [in this window] [in a new window]   Figure 6.   sPLA2 isotypes demonstrated substrate specificity. Recombinant human Groups IIA, V, and X sPLA2 enzyme activities were measured in transfected COS-1 cell supernates using two different substrates (DPPC vesicles [striped bars] or E. coli membranes [gray bars]) to obtain a profile of differential activity. (Black bars indicate ratio.) Under identical conditions of assay linearity (see MATERIALS AND METHODS), Group V hydrolyzed DPPC much more efficiently than did E. coli membranes, whereas Group IIA demonstrated the opposite profile of activity. Group X was slightly more active against DPPC. As a reference, pig pancreatic Group IB was also active against both substrates.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Definition of the role(s) of sPLA₂s in humans has become complicated by the discovery of an expanding family of sPLA₂ enzymes. The sPLA₂ isotypes may differ in ability to bind to and hydrolyze different phospholipid substrates (23). They might also differ in cellular localization and mechanisms of release. It would not be surprising to find that they have different roles in lung diseases. Multiple roles in pathogenesis are already becoming apparent. sPLA₂s have been implicated in the mobilization of AA and formation of eicosanoids in murine mast cells (18), P388D macrophages (19), and transfected cell lines (28, 29). They may also interact with either of two sPLA₂ receptors (30) to initiate intracellular events which might include, but are not necessarily limited to, eicosanoid production (31). sPLA₂s could possibly have other effects by direct damage to airway epithelial cells or to airway surfactant.

Data support the concept that these enzymes may have such roles in lung diseases. It is established that ARDS is associated with surfactant dysfunction and that one cause is the release into airways of sPLA₂(s) (32). It has recently become apparent that sPLA₂s are also released into airways in milder forms of lung disease, such as asthma. Using an inhaled antigen provocation of asthmatic patients, Bowton and colleagues (2) recently found that sPLA₂ enzymatic activity was elevated in bronchoalveolar lavage fluid (BALF) within 4 h of challenge and the elevation paralleled an increased release of AA into the airways. Using similar models, Chilton and colleagues (33) found that products of sPLA₂ activity, especially 1-palmitoyl-2-lyso-phosphatidylcholine, appeared in the BALF after endobronchial antigen challenge in asthmatics. Preliminary data suggest that palmitic acid levels were also elevated after endobronchial challenge (Seeds, unpublished observation). Palmitic acid is the predominant fatty acid in the sn-2 position in pulmonary surfactant, and thus would be released during sPLA₂-mediated surfactant hydrolysis. Moreover, surfactant dysfunction, measured as an increased minimal surface tension of BALF, occurred in a surprisingly large proportion (six of 11) of mild asthmatics after endobronchial antigen challenge (Hite, Seeds, Bowton, and Bass, in preparation). The combination of surfactant dysfunction, elevated palmitic acid, and elevated lyso-phosphatidylcholine suggests that hydrolysis of airway surfactant phospholipids occurred in antigen-challenged asthmatic lungs. To determine the mechanisms of such damage, it is critical to determine which sPLA₂s are expressed in lung and their cellular localizations within the lungs.

In the recent reports of cloning the human Group V and Group X sPLA₂s, Northern blots of lung tissue RNA extracts revealed faint or no reactivity for Group V (16). Northern blots probed for Group X detected only a faint band for the smaller of two possible mRNA sizes (20). Our initial studies used RT-PCR to identify the sPLA₂s expressed in human lung. The results show that adult human lung tissue expresses mRNA for the most recently described sPLA₂ isotype, Group X, as well as for sPLA₂ Groups IB, IIA, and V. Very recently, Valentin and associates (34) and Ishizaki and coworkers (35) have reported several variants of the Group IIA enzyme, but their functional relationship to lung tissue remains to be explored. We have amplified the new Group IId sPLA₂ mRNA from human lung total RNA extracts by RT-PCR, but expression was the weakest thus far of the sPLA₂ transcripts (not shown).

If there are unique functions for each sPLA₂ isotype, then different cells are likely to express different combinations of these proteins. Initial studies employed the transformed cell line BEAS, originally derived from normal bronchial airway epithelial cells. By RT-PCR, these cells expressed Groups X and IIA, with small amounts of putative Group V and no expression of Group IB sPLA₂. These transformed cells could differ in expression from normal lung cells in situ, but the data support the concept that the Group X sPLA₂ might be expressed in airway epithelial cells.

Localization of all sPLA₂ mRNAs was next examined in human lung in situ. We used cloned cDNA to synthesize cRNA probes for in situ hybridization. Group X sPLA₂ was expressed primarily in airway epithelial cells. The mRNA appeared in ciliated columnar epithelial cells and expression extended distally in the airways to the nonciliated cuboidal cells at the opening of alveolar spaces. No other cell type in lung clearly displayed reactivity for the Group X sPLA₂ cRNA probe. Group V mRNA was also detected in the airway columnar epithelial cells. Group IB and IIA mRNAs, although detected in total lung RNA extracts by the more sensitive RT-PCR experiments, were not detectable by in situ hybridization. The verification of the Group IB and IIA probes in other tissues known to express those genes suggests that mRNAs for Group IB and Group IIA sPLA₂ are minimally expressed in human lung, possibly not at all in the epithelial cells. Thus, airway epithelium may specifically express the Group X and Group V sPLA₂ genes. Little is known about the possible induction of Group V or Group X sPLA₂ mRNA, although some may have occurred in these pathologically normal margins of lung biopsy tissue.

During review of this manuscript, Hanasaki and coworkers also described the cloning of human Group X sPLA₂ (36). They reported that the Group X enzyme mediated release of both AA and oleic acid, and also produced formation of prostalandin (PG)E₂ from human myeloid leukemia (THP-1) cells. The same laboratory also recently reported that although Group IIA and Group V sPLA₂ appear linked to eicosanoid production via induction of COX-2, Group X distinctly did not induce COX-2 or cPLA₂, and overexpression of COX-2 proteins were required for PGE₂ release in human embryonic kidney 293 cell transformants (29). Hanasaki and associates also reported localization of immunoreactive Group X sPLA₂ protein in human alveolar epithelial cells within the lung (36). This corroborates our finding and suggests that functional gene expression occurs in both distal airways and alveolar spaces in normal human lung. Only one other published report has addressed sPLA₂ distribution in human lung, and that was directed to the Group IIA enzyme that was found in the fibroblast layer surrounding epitheloid cell granulomas (37).

The function of the sPLA₂ in these lung epithelial cells is unknown. Group X could impact on surfactant if airway cells are damaged or are stimulated to release the sPLA₂. Phosphatidylcholine (PC) is the major phospholipid component of surfactant, specifically DPPC. Both Group X and Group V sPLA₂ demonstrated robust enzymatic activity against PC vesicles (20, 36) as shown here. The combination of enzymatic activity against PC and anatomic localization in airway epithelial cells suggests that the Group X or Group V isotypes of sPLA₂ may provide one mechanism contributing to lung injury, both to cells and possibly to surfactant.