Transgenic Expression of Group V, but Not Group X, Secreted Phospholipase A2 in Mice Leads to Neonatal Lethality because of Lung Dysfunction*

In an effort to elucidate the functions of secreted phospholipase A2 (sPLA2) enzymes in vivo, we generated transgenic (Tg) mice for group V sPLA2 (sPLA2-V) and group X sPLA2 (sPLA2-X), which act potently on phosphatidylcholine in vitro.We found that sPLA2-V Tg mice died in the neonatal period because of respiratory failure. The lungs of sPLA2-V Tg mice exhibited atelectasis with thickened alveolar walls and narrow air spaces, accompanied by infiltration of macrophages and only modest changes in eicosanoid levels. This severe pulmonary defect in sPLA2-V Tg mice was attributable to marked reduction of the lung surfactant phospholipids, phosphatidylcholine and phosphatidylglycerol. Given that the expression of sPLA2-V is greatly elevated in human lungs with severe inflammation, our present results raise the intriguing possibility that this isozyme may contribute to ongoing surfactant hydrolysis often observed in the lungs of patients with respiratory distress syndrome. In contrast, sPLA2-X Tg neonates displayed minimal abnormality of the respiratory tract with normal alveolar architecture and surfactant composition. This unexpected result was likely because sPLA2-X protein existed as an inactive zymogen in most tissues. The active form of sPLA2-X was detected in tissues with inflammatory granulation in sPLA2-X Tg mice. These results suggest that sPLA2-X mostly remains inactive under physiological conditions and that its proteolytic activation occurs during inflammation or other as yet unidentified circumstances in vivo.

In an effort to elucidate the functions of secreted phospholipase A 2 (sPLA 2 ) enzymes in vivo, we generated transgenic (Tg) mice for group V sPLA 2 (sPLA 2 -V) and group X sPLA 2 (sPLA 2 -X), which act potently on phosphatidylcholine in vitro. We found that sPLA 2 -V Tg mice died in the neonatal period because of respiratory failure. The lungs of sPLA 2 -V Tg mice exhibited atelectasis with thickened alveolar walls and narrow air spaces, accompanied by infiltration of macrophages and only modest changes in eicosanoid levels. This severe pulmonary defect in sPLA 2 -V Tg mice was attributable to marked reduction of the lung surfactant phospholipids, phosphatidylcholine and phosphatidylglycerol. Given that the expression of sPLA 2 -V is greatly elevated in human lungs with severe inflammation, our present results raise the intriguing possibility that this isozyme may contribute to ongoing surfactant hydrolysis often observed in the lungs of patients with respiratory distress syndrome. In contrast, sPLA 2 -X Tg neonates displayed minimal abnormality of the respiratory tract with normal alveolar architecture and surfactant composition. This unexpected result was likely because sPLA 2 -X protein existed as an inactive zymogen in most tissues. The active form of sPLA 2 -X was detected in tissues with inflammatory granulation in sPLA 2 -X Tg mice. These results suggest that sPLA 2
The sPLA 2 family represents a group of structurally related, disulfide-rich, Ca 2ϩ -dependent, low molecular weight enzymes with a catalytic histidine (1,2). Individual sPLA 2 s exhibit unique tissue and cellular localizations and enzymatic properties, suggesting their distinct roles in various pathophysiological events. Several sPLA 2 enzymes have the ability to hydrolyze cellular phospholipids to release arachidonic acid (AA), a precursor of eicosanoids, and lysophospholipids (13)(14)(15)(16)(17)(18). Among them, group V sPLA 2 (sPLA 2 -V) and group X sPLA 2 (sPLA 2 -X) represent the most potent sPLA 2 isoforms that have the ability to act on the phosphatidylcholine (PC)-rich outer plasma membrane of various cells (14 -17). Because of the secreted moiety of sPLA 2 s, it is likely that their target substrates also exist in the extracellular milieu. In fact, several sPLA 2 s hydrolyze phospholipids in bacterial membranes (19 -21), plasma lipoproteins (22)(23)(24)(25)(26), lung surfactant (27)(28)(29)(30)(31)(32)(33)(34)(35), and so on with distinct poten-cies in vitro, suggesting that they play divergent roles in various biological events. However, the proposed roles of sPLA 2 enzymes have largely been inferred from their enzymatic and biochemical properties, the effects of their overexpression or exogenous addition in in vitro experimental settings, or the use of pharmacological inhibitors (whose specificity may not be strictly defined).
Given the complexity and diversity of the PLA 2 family, transgenic (Tg) overexpression (gain-of-function) and gene targeting (loss-of-function) strategies seem to be more appropriate for elucidating the functions of individual enzymes in vivo. Indeed, these approaches have proved fruitful for determining the crucial roles of cPLA 2 ␣ in eicosanoid and platelet-activating factor generation in a wide variety of pathological and physiological states (36 -39); iPLA 2 ␤ in ischemic heart damage, sperm motility, nitric oxide-induced antiviral response, and diet-induced insulin secretion (40 -43); iPLA 2 (adipose triglyceride lipase) in fat mobilization in adipose tissue (44); intracellular PAF-AH-I in spermatogenesis (45); and two lysosomal PLA 2 s in steady-state lung surfactant turnover (11,12). With regard to sPLA 2 s, sPLA 2 -IIA Tg mice (beyond that endogenous sPLA 2 -IIA in mice is largely restricted to intestinal Paneth cells (e.g. BALB/c) or not expressed at all (e.g. C57BL/6 and 129/Sv)) show marked resistance to bacterial infection (19,20), thus substantiating the in vitro observations that this enzyme efficiently hydrolyzes bacterial membranes (21). In addition, sPLA 2 -IIA Tg mice show increased lipoprotein hydrolysis and atherosclerosis (22)(23)(24), although sPLA 2 -V and sPLA 2 -X have also been implicated in these processes (25,26). Mouse strains intrinsically deficient in sPLA 2 -IIA because of a frameshift mutation in its gene are susceptible to intestinal tumorigenesis (45). Recent gene targeting studies of sPLA 2 -IB and sPLA 2 -V have provided unequivocal evidence for their roles in digestion of dietary phospholipids (47,48) and eicosanoid production and phagocytosis by macrophages (49,50), respectively. Furthermore, a gain-of-function study in Xenopus has revealed the role of sPLA 2 -XIIA in olfactory sensory neuron structure in a manner independent of its catalytic activity (51).
In this study, we generated Tg mice overexpressing sPLA 2 -V or sPLA 2 -X, the two most potent PC-hydrolyzing isozymes, in an effort to search for their suitable substrates in vivo. We found that all sPLA 2 -V Tg neonates died within several hours after birth because of respiratory failure. This severe, lethal phenotype of sPLA 2 -V Tg mice is attributable to aberrant hydrolysis of lung surfactant phospholipids, an event that has often been observed in the lungs of patients with respiratory distress syndrome (RDS). In contrast, sPLA 2 -X Tg mice do not display such detrimental lung damage, because sPLA 2 -X exists as an inactive zymogen in most tissues under physiological conditions. Proteolytic activation of sPLA 2 -X in vivo occurs in particular situations such as inflammation.

EXPERIMENTAL PROCEDURES
Generation of sPLA 2 Tg Mice-All mice were housed in climate-controlled (21°C) specific pathogen-free facilities with a 12-h light-dark cycle, with free access to standard laboratory food (Picolab mouse diet 20; Laboratory Diet, Brentwood, MO) and water. All procedures involving animals were performed according to protocols approved by the faculties.
The strategy for the generation of sPLA 2 Tg mice is illustrated in Fig. 1A. The cDNA for mouse sPLA 2 -V or human sPLA 2 -X was inserted into the EcoRI site of the pCALNL5 vector (52) (Fig. 1). The plasmid, in which the transgene was flanked downstream of a neomycin cassette with LoxP sites at both ends, was excised at the HindIII and SalI sites to produce a 5.1-kb CAG-loxP-neo r -loxP-sPLA 2 (LNL-sPLA 2 ) fragment. Female 6 -8-week-old C57BL/6 mice (Japan SLC) were treated with 5 units of horse gonadotropin (Teikokuzoki Pharmaceutical) and 48 h later with 5 units of human chorionic gonadotropin (Teikokuzoki Pharmaceutical) and then mated with male C57BL/6 mice. The purified LNL-sPLA 2 fragment was microinjected into 0.5-day fertilized eggs using a micromanipulator (Nikon), and the eggs were transferred to the fimbriae of uterine tubes in female ICR mice (Japan SLC) that had been mated with vasoligated male mice the previous day. The tails (5 mm in length) of the pups (4-week-old) were cut off and lysed in 500 l of tissue lysis buffer comprising 50 mM Tris-HCl (pH 8.0), 100 mM EDTA, 0.5% SDS, and 0.4 mg/ml proteinase K (Sigma) at 55°C overnight. DNA was purified from the lysates using an automated DNA isolation system (NA-2000; Kurabo Industries), and aliquots were taken for PCR for genotyping (see below). Male founders were mated with female C57BL/mice to confirm germ line transmission by PCR genotyping, and those with successful germ line transmission were then crossed with female CAG-Cre Tg mice, which carry the Cre recombinase transgene under the control of the ubiquitously active cytomegalovirus immediate early enhancer-chicken ␤-actin hybrid (CAG) promoter (53,54). This step resulted in removal of the neo r cassette from the LNL-sPLA 2 transgene, thereby allowing activation of the sPLA 2 -V or sPLA 2 -X transgene in the whole body of the offspring (sPLA 2 Tg mice) (Fig. 1). As required, the male founders were mated with female pgk2-Cre Tg mice, in which the transcription of the Cre recombinase transgene was controlled by the spermatogenic cell-specific promoter pgk2, allowing generation of the testis-specific sPLA 2 Tg mice (55). All the sPLA 2 Tg mice were inbred with C57BL/6 mice.
Northern Blotting-Equal amounts (ϳ10 g) of total RNA obtained from tissues by use of TRIzol reagent (Invitrogen) were applied to separate lanes of 1.2% (w/v) formaldehyde-agarose gels, electrophoresed, and transferred to Immobilon-N membranes (Millipore). The resulting blots were then probed with appropriate cDNA probes that had been labeled with [ 32 P]dCTP (PerkinElmer Life Sciences) by random priming (Takara Biomedicals). Hybridization and subsequent membrane washing were carried out as described previously (14).
RT-PCR-Synthesis of cDNA was performed using 0.5 g of total RNA from tissues and avian myeloblastosis virus reverse transcriptase, according to the manufacturer's instructions supplied with the RNA PCR kit (Takara Biomedicals). Subsequent amplification of the cDNA fragments was performed using 0.5 l of the reverse-transcribed mixture as a template with specific primers (23-bp oligonucleotide primers corresponding to 5Ј-and 3Ј-nucleotide sequences of the open reading frame) for mouse sPLA 2 -V. The PCR conditions were 94°C for 30 s and then 35 cycles of amplification at 94°C for 5 s and 68°C for 4 min, using the Advantage cDNA polymerase mix (Clontech). RT-PCR for ␤-actin as a control was performed according to the manufacturer's instructions (Sigma). The PCR products were analyzed by 1% agarose gel electrophoresis with ethidium bromide.
Immunoblotting-Tissues (100 mg) were soaked in 500 l of SET buffer comprising 20 mM Tris-HCl (pH 7.4), 0.25 M sucrose, 1 mM EDTA, and 1 mM phenylmethylsulfonyl fluoride (Sigma) and then homogenized with a Polytron homogenizer and sonicator. The homogenates (20 g of protein equivalents) were subjected to SDS-PAGE on 13% gels under reducing conditions with 2-mercaptoethanol. Protein concentrations in the samples were determined with a BCA protein assay kit (Pierce) using bovine serum albumin (Sigma) as a standard. The separated proteins were electroblotted onto nitrocellulose membranes (Schleicher & Schuell) with a semi-dry blotter (Transblot SD; Bio-Rad). After blocking with 5% (w/v) skim milk in TBS containing 0.05% (v/v) Tween 20 (TBS-T), the membranes were probed with rabbit antibody against each sPLA 2 at 1:5,000 dilution in TBS-T for 2 h, followed by incubation with horseradish peroxidase-conjugated anti-rabbit IgG (Zymed Laboratories Inc.) at 1:5,000 dilution in TBS-T for 2 h, and then visualized using the ECL Western blot system (PerkinElmer Life Sciences), as described previously (14).
Measurement of sPLA 2 Activity-sPLA 2 activities in tissue homogenates were assayed by measuring the amounts of radiolabeled linoleic acid or AA released from the substrate 1-palmitoyl-2-[ 14 C]linoleoyl-phosphatidylethanolamine (PE) or 1-palmitoyl-2-[ 14 C]arachidonoyl-PC (PerkinElmer Life Sciences). The substrate in ethanol was dried under a N 2 stream and dispersed in water by sonication. Each reaction mixture (total volume 250 l) consisted of appropriate amounts of the required samples, 100 mM Tris-HCl (pH 7.4), 4 mM CaCl 2 , and 2 mol of substrate. After incubation for 30 -60 min at 37°C, [ 14 C]linoleic acid or [ 14 C]AA was extracted, and the radioactivity was quantified using a liquid scintillation counter, as described previously (14).
Plethysmography-Respiration monitoring of embryonic Tg mice (E18.5) obtained by cesarean section was carried out by whole-body plethysmography (56). Briefly, we used a small glass chamber that was connected to one side of a differential pressure transducer (TP602T; Nihon Kohden). The measured pressure signal was amplified with a DC amplifier (AR601G; Nihon Kohden). At the time of measurement, the chamber was sealed, and instantaneous pressure within the chamber with reference to ambient pressure was measured by a pressure transducer. Respiratory frequency, tidal volume, and minute volume were calculated from 50 to 70 respiratory cycles during quiet breathing.
Histochemistry-Immunohistochemistry of mouse tissue sections was performed as described previously (57,58). In brief, formalin-fixed tissues were embedded in paraffin, sectioned, mounted on glass slides, deparaffinized in xylene, and rehydrated in ethanol with increasing concentrations of water. The tissue sections (4 m thick) were incubated with Target Retrieval Solution (DAKO) as required, incubated for 10 min with 3% (v/v) H 2 O 2 , washed three times with TBS for 5 min each, incubated with 5% (v/v) skim milk in TBS for 30 min, washed three times with TBS for 5 min each, and incubated with anti-sPLA 2 -V antibody at 1:200 dilution in TBS overnight at 4°C. Alternatively, anti-F4/80 (Serotec) and anti-surfactant protein C (SP-C; Santa Cruz Biotechnology) antibodies were used at 1:50 dilution, respectively. The sections were then treated with a catalyzed signalamplified system staining kit (DAKO) with diaminobenzidine substrate, followed by counterstaining with hematoxylin and eosin. The cell type was identified from conventional hematoxylin and eosin staining of serial sections adjacent to the specimen used for immunohistochemistry.
For periodic acid-Schiff (PAS) staining (59), tissue sections were deparaffinized, rehydrated, and placed in 0.5% periodic acid solution for 5 min. After washing in water, the sections were placed in Coleman's Schiff reagent for 15 min, washed in water for 10 min, and counterstained with hematoxylin solution for 15 min. After washing in water for 15 min, the sections were dehydrated and mounted with resinous medium.
Electrospray Ionization/Mass Spectrometry (ESI/MS)-Lungs of newborn mice were lavaged with 1 ml of SET buffer, and the resulting bronchoalveolar lavage fluid (BALF) was used after centrifugation to remove the cells and tissue debris. Lipids were extracted from the BALF by the method of Bligh and Dyer (60). Before the lipid extraction, PC with C14:1-14:1 (diacyl) and phosphatidylglycerol (PG) with C14:0 -14:0 (diacyl) were added to the BALF (1 nmol of internal standard per 1 mg of protein equivalents) as internal standards. MS spectra were obtained on a Quattro Micro tandem quadrupole mass spectrometer (Micromass) equipped with an ESI, as described previously (61). Lipid extracts obtained from BALF were reconstituted in 2:1 chloroform/methanol (100 -300 mol phosphorus/liter), and 2 l of the sample was injected per run. The samples were introduced by means of a flow injector into the ESI chamber at a flow rate of 4 l/min in a solvent system of acetonitrile/methanol/water (2:3:1; v/v) containing 0.1% (v/v) ammonium formate (pH 6.4). The mass spectrometer was operated in the positive and negative scan modes. The flow rate of the nitrogen drying gas was 12 liters/min at 80°C. The capillary and cone voltages were set at 3.7 kV and 30 V, respectively; argon at 3-4 ϫ 10 4 torr was used as the collision gas, and a collision energy of 30 -40 V was used for obtaining fragment ions for precursor ions.
Measurement of Eicosanoids-The contents of prostaglandin E 2 (PGE 2 ) and leukotriene C 4 (LTC 4 ) in the BALF were quantified with PGE 2 and LTC 4 enzyme immunoassay kits (Cayman Chemicals).
AA Release-Culture of the human bronchial epithelial cell line BEAS-2B and its stable transfectants for various PLA 2 s (produced by the lentivirus method (Invitrogen)) was done as described previously (58). Cells grown to near confluency in 48-well plates (Iwaki Glass) were incubated with [ 3 H]AA (Amersham Biosciences) (0.1 Ci/ml) overnight. After three washes with fresh medium, 100 l of culture medium with or without tert-butylhydroperoxide was added to each well, and the radioactivities released into the supernatants after incubation for appropriate periods were measured. The percentage release was calculated using the formula (S/(S ϩ P)) ϫ 100, where S and P are the radioactivities measured in the supernatant and cell pellet, respectively, as described (14).
Cotton Thread-induced Granulation Tissue Formation-Cotton threads (model no. 8; Araiwa) were washed overnight with ethyl acetate, dried at room temperature before being cut into 1-cm lengths (3 mg weight), and sterilized by dry heat at 180°C for 2 h. The cotton threads were implanted subcutaneously into the dorsum of anesthetized mice by using a 13-gauge implant needle (Natume). After appropriate periods, the mice were anesthetized and killed, and the granulation tissues were dissected together with the cotton threads and weighed. The isolated granulation tissues were washed, cut into small pieces with scissors, and homogenized with a Polytron homogenizer in SET buffer. The homogenates were taken for immunoblotting and PLA 2 enzyme assay, as mentioned above.

RESULTS
Generation of sPLA 2 Tg Mice-Tg constructs for sPLA 2 -V or sPLA 2 -X ( Fig. 1A) were microinjected into the pronuclei of fertilized eggs of C57BL/6 females and transferred into the oviducts of ICR pseudopregnant females. The offspring were examined for expression of the transgenes (LNL-V or LNL-X) by PCR genotyping. Tail biopsies were taken on day 28, and the genomic DNAs isolated were subjected to PCR genotyping using a CAG forward primer and reverse primers specific for individual sPLA 2 s. In each case, two mice (both males) showed the presence of 1.7-and 2-kb bands for LNL-V and LNL-X, respectively (Fig. 1B). These founder mice, in which each sPLA 2 transgene was still silent, were mated with female CAG-Cre Tg mice to allow the removal of the neo r cassette from the LNL-V or LNL-X transgene by the Cre/LoxP reaction. Following this, the transcription of each sPLA 2 transgene is directly regulated by the CAG promoter and thereby activated in all tissues of the offspring. Representative results of the PCR genotyping for the F1 progeny are shown in Fig. 1C. As a result of the Cre/LoxP reaction, the band for the sPLA 2 -V transgene shifted from 1.7 (LNL-V) to 0.3 kb (Fig. 1C, left panel) and that for the sPLA 2 -X transgene from 2 (LNL-X) to 0.6 kb (Fig. 1C, right panel), whereas no band was detected in the littermate siblings. The ratio of the transgene-positive to -negative pups was ϳ1:1 in both cases, and these pairs (hereafter designated as sPLA 2 Tg mice and wild-type (WT) littermate controls) were used in each experiment.
Neonatal Death of sPLA 2 -V Tg Mice Because of Respiratory Failure-Although the litter size was normal (8 pups per delivery on average), we noted that about half of LNL-V Tg ϫ CAG-Cre Tg neonates were gasping, exhibited tachypnea and cyano-sis, and died within the first 8 h on average after birth ( Fig. 2A). PCR genotyping showed that all the dying pups carried the active sPLA 2 -V transgene, whereas the survivors did not. Pups from two independent lines of LNL-V Tg (see Fig. 1B) mated with CAG-Cre Tg mice displayed a similar phenotype, whereas the LNL-V Tg ϫ C57BL/6 offspring grew healthy. There was no evidence for abnormality in the LNL-V Tg ϫ CAG-Cre Tg pups before birth (E18.5), yet the sPLA 2 -V transgene-positive, but not transgene-negative, pups revived after cesarean section again all died within ϳ8 h. These results clearly indicate that Tg overexpression of sPLA 2 -V in the whole body leads to neonatal death in mice.
In agreement with the PCR genotyping, RNA blotting demonstrated marked expression of sPLA 2 -V mRNA in all the tissues of transgene-positive neonates (sPLA 2 -V Tg mice), whereas it was below the detection limit in transgene-negative WT neonates in this experimental setting (Fig. 2B). More sen- sitive RT-PCR demonstrated the expression of endogenous sPLA 2 -V mRNA in the hearts of newborn WT mice (an organ in which sPLA 2 -V shows the highest expression in various species (1, 2)), and sPLA 2 -V expression in the heart was increased ϳ15fold in Tg mice compared with that in WT mice (supplemental Fig. S1). Increases of PLA 2 enzymatic activity in the homogenates of tissues from sPLA 2 -V Tg mice toward an exogenous substrate (1-palmitoyl-2-linoleoyl-PE) (Fig. 2C) roughly correlated with the results of RNA blotting (Fig. 2B). Thus, as compared with the PLA 2 activities in the brain, heart, and lung of WT mice, which reflect the combined activities of various PLA 2 enzymes intrinsically present in each tissue, those of sPLA 2 -V Tg mice were increased ϳ2-, 2.5-, and 7-fold, respectively. Judging from the enzyme activity (in comparison with pure recombinant sPLA 2 -V expressed by Escherichia coli and under the assumption the activity was not perturbed by endogenous substances), and given that the tissue homogenates had been diluted 5-fold with the buffer during preparation, the concentration of sPLA 2 -V in the Tg mice was estimated to be within the range 10 -100 ng/ml, depending on the tissue.
The short gasping breaths, tachypnea, and cyanotic skin color (indicative of impaired oxygenation) observed in new-born sPLA 2 -V Tg mice suggested that they had a profound defect of the respiratory system. To confirm this, we monitored the respiration rates of sPLA 2 -V Tg and control mice using plethysmography. As compared with newborn WT mice that exhibited a regular and constant rate of respiration that declined gradually to reach a steady state after 5-6 h, the breathing rate of sPLA 2 -V Tg mice was shallow at birth and became more rapid after 3-4 h, thereafter becoming labored, irregular, and gasping after 5-6 h ( Fig. 3 and Table 1). However, ventilation (in terms of minute volume) in Tg and WT mice was similar over the first 6 h (Table 1).
Histological Abnormalities in the Lungs of sPLA 2 -V Tg Mice-Histological examination of the lungs of sPLA 2 -V Tg mice and their littermate controls (8 h after birth) revealed pronounced defects in the former (Fig. 4). Tissues other than the lung appeared normal in sPLA 2 -V Tg mice, except for signs of skin inflammation such as dermal edema and macrophage infiltration (supplemental Fig. S2). Because of the marked respiratory defect in Tg mice, we subsequently focused on lung histology.
Remarkably, thinning of the alveolar septa at birth, a prerequisite for blood-gas exchange, was impaired in sPLA 2 -V Tg mice (Fig. 1A). Thus, as compared with the well organized lung architecture of WT mice (Fig. 4A, panels a and b), sPLA 2 -V Tg mice exhibited extensive atelectasis with formation of hyaline membranes lining the airway surfaces, and the alveolar walls were abnormally thickened (Fig. 4A, panels c and d). Although microvessels were embedded within the thickened alveolar walls, there was no sign of vascular leakage or endothelial cell abnormalities. Immunostaining of WT lungs with anti-sPLA 2 -V antibody revealed the localization of endogenous sPLA 2 -V in some, if not all, bronchial epithelial cells (Fig. 4B,  panel a), a staining pattern similar to that in normal human lungs (58). In Tg mice, the whole lung area was markedly stained with anti-sPLA 2 -V (Fig. 4B, panel b) but not with con-  trol (Fig. 4B, panel c) antibody, with the alveolar surfaces being most intensely stained. Although the mechanism whereby overexpressed sPLA 2 -V tended to accumulate on the alveolar surfaces is unclear, one possible explanation is that sPLA 2 -V secreted from alveolar cells might be preferentially associated with heparan sulfate proteoglycan on the alveolar surface, because sPLA 2 -V has high affinity for heparan sulfate proteoglycan (14,17) and because the alveolar surface is enriched in the proteoglycan that plays a crucial role in the regulation of alveolar differentiation (62).
Although there were few F4/80positive alveolar macrophages in the lungs of WT neonates (Fig. 4C,  panel a), scattered staining of these cells was found in the lungs of sPLA 2 -V Tg mice (Fig. 4C, panel b), suggesting the presence of inflammation in the latter. However, no migration of granulocytes and lymphocytes was observed in the lungs of either Tg or WT mice. Immunohistochemistry for SP-C, a surfactant apolipoprotein produced by type II alveolar epithelial cells, showed equivalent staining in the lungs of WT (Fig. 4D, panels a and  b) and Tg (Fig. 4D, panels c and d) mice, unlike several gene-manipulated mice that exhibit neonatal death because of respiratory failure resulting from impaired differentiation of alveolar epithelial cells (62)(63)(64)(65)(66)(67), suggesting that terminal differentiation of type II pneumocytes did occur normally in the lungs of sPLA 2 -V Tg mice. To further address this issue, lung sections were subjected to PAS staining, because immature epithelial cells contain abundant PAS-positive gly-  cogen stores that are subsequently converted to surfactant phospholipids following cell maturation (64). As expected, alveoli of control neonates showed no PAS staining (Fig. 4E, panel a), because glycogen is almost entirely converted to surfactant phospholipids at this developmental stage (64).
Although most areas of the lungs in Tg mice were largely PASnegative (data not shown), weak staining of a few epithelial cells was focally evident (Fig. 4E, panel b). Thus, although most type II epithelial cells were terminally differentiated in the lungs of sPLA 2 -V Tg mice, there was a small population of PAS-positive cells that remained immature.
Modest Changes in Eicosanoid Levels in BALF of sPLA 2 -V Tg Mice-Eicosanoids, particularly LTs, are known to profoundly influence lung pathology (68 -71), and sPLA 2 -V has the ability to promote the generation of eicosanoids in various cells in vitro (14,17). We therefore analyzed the levels of PGE 2 , a product of the cyclooxygenase pathway, and LTC 4 , a product of the lipoxygenase pathway, in BALF obtained from sPLA 2 -V Tg and WT neonates at 8 h after birth. As shown in Fig. 5, the BALF of sPLA 2 -V Tg mice contained about twice as much PGE 2 as that of WT mice. This result is in line with our in vitro observations using cultured lung epithelial cells that sPLA 2 -V, in comparison with other PLA 2 s, has the particular ability to augment AA release from cells undergoing membrane-oxidative stress relative to that produced by undamaged cells (supplemental Fig.  S3). In contrast, the levels of LTC 4 in BALF were somewhat low and showed no difference between Tg and WT mice (Fig. 5). This characteristic of sPLA 2 -V Tg mice differs from that of mice or humans with asthma or lung fibrosis, in which levels of cysteinyl LTs produced from infiltrating leukocytes and mast cells are markedly elevated and participate in the pathogenesis (69 -71). Overall, it is unlikely that only modest changes in prostanoid levels in the lungs could fully account for the lethal phenotype of sPLA 2 -V Tg mice.
Aberrant Reduction of Lung Surfactant in sPLA 2 -V Tg Mice-Recent reports have shown that the neonatal death observed in several gene-manipulated mice because of pulmonary collapse is attributable to disturbed homeostasis of pulmonary surfactant (62)(63)(64)(65)(66)(67). The principal function of lung surfactant, which is a mixture of phospholipids (90%) and surfactant proteins (10%), is to reduce the work of breathing by lowering alveolar surface tension during respiration. PC with saturated and monosaturated fatty acids (palmitic acid in particular) is predominant in surfactant phospholipids (ϳ80% of total lipid) (27), and PG (ϳ10% of total lipid) plays a role in phospholipid-protein interactions that maintain the alveolar surfactant layer, especially through interaction with SP-B (72,73). Because changes in the ratio of the surfactant components can dramatically alter the surface tension in small airways and alveoli, compromising airway patency, gas exchange, and host defense, any surfactant abnormality can have severe pathological consequences in terms of lung function. Interestingly, previous studies have indicated a role of sPLA 2 s in inflammationmediated surfactant malfunction through hydrolysis of surfactant phospholipids (27)(28)(29)(30)(31)(32)(33)(34)(35).
We therefore extracted and analyzed surfactant phospholipids in BALF (1 ml) from sPLA 2 -V Tg and WT neonates. Protein concentrations in BALF from WT and Tg mice were similar (0.22 Ϯ 0.05 mg/ml and 0.28 Ϯ 0.09 mg/ml (n ϭ 9 -12), respectively) over 8 h after birth (i.e. just prior to death). Both genotypes contained 15-25 nmol of phospholipids in their BALF. In a primary study, phospholipids extracted from BALF of both genotypes at term were separated on two-dimensional thin layer chromatography, followed by quantification of phosphorus in a spot for PC as well as for PE, another minor surfactant phospholipid. Interestingly, there was 41.6 Ϯ 0.1% (n ϭ 2) reduction of PC in BALF of sPLA 2 -V Tg mice as compared with that of control mice. PE was also decreased substantially (19.1 Ϯ 0.6% reduction, n ϭ 2) in BALF of sPLA 2 -V-Tg mice relative to that of WT mice, even though the decrease in PE was less obvious than that in PC.
To gain more insight into the change in surfactant of V-Tg mice, we carried out ESI/MS analysis. Before lipid extraction from BALF, internal standards (see "Experimental Procedures") were added to BALF (1 nmol of standards per 1 mg of protein equivalents) for calibration. ESI/MS revealed that surfactant obtained from WT neonates just at birth (0 h) contained PC molecular species with C16:0-C16:0, C16:0 -16:1, and to a lesser extent C16:0 -18:1 fatty acyl chains as well as other minor molecular species (Fig. 6A, upper panel), consistent with a previous report (65). These PC species were markedly reduced (by ϳ75, 60, and 35% in PC species with C16:0 -16:1, C16:0 -16:0, and C16:0 -18:1, respectively, relative to those in littermate WT mice) in the surfactant of sPLA 2 -V Tg mice (Fig. 6A, lower  panel). Comparable internal standard peaks in individual MS spectra between both genotypes indicate that the lipid extraction from BALF was successfully done. Thus, the overall reduction of PC species in V-Tg mice relative to WT mice revealed by ESI/MS seemed consistent with that shown by the thin layer chromatographic study (see above). Reduction of surfactant phospholipids in sPLA 2 -Tg mice at 8 h after birth was more prominent (Fig. 6, B and C). Thus, in contrast to WT littermates in which surfactant PC levels were nearly constant (Fig. 6B,  upper panel), sPLA 2 -V Tg neonates at 8 h had lost most of their surfactant PC (Fig. 6B, lower panel), indicating that the hydrolysis of surfactant PC paralleled the ongoing lung damage in Tg mice. It is noteworthy that the marked reduction of surfactant PC over time was not accompanied by a concomitant increase in lyso-PC in Tg mice (data not shown), even though the ratio of lyso-PC to PC increased from Ͻ5% in WT mice to Ͼ90% in Tg mice because of a marked decrease of PC in the latter. This is probably because lyso-PC is taken up rapidly and re-acylated by alveolar cells, as has been reported for patients with inflammatory lung diseases in which increased sPLA 2 activity and surfactant dysfunction have typically not been associated with concomitant elevation of lyso-PC (74). Fig. 6C shows the results of ESI/MS analysis of PG, a minor but important component of surfactant, in BALF from mice 8 h after birth. As in the case of PC above, PG molecular species in the surfactant were markedly decreased in sPLA 2 -V Tg mice (Fig. 6C, lower panel) as compared with those in WT littermates, which showed PG species with C16:0 -16:0, C16:0 -18:1, and C18:0 -18:2 (Fig. 6C, upper panel).
Taken together, we conclude that the lethal lung dysfunction observed in sPLA 2 -V Tg mice is caused by aberrant hydrolysis of lung surfactant by sPLA 2 -V. The reduction in both PC and PG (as well as PE as assessed by thin-layer chromatography) in Tg mice (Fig. 6) agrees with a recent study that sPLA 2 -V hydrolyzes these phospholipids derived from lung surfactant in vitro (33,35).
Testis-specific sPLA 2 -V Tg Mice-Because a previous study has shown that sPLA 2 -IIA Tg mice are infertile because of impaired spermatogenesis (75), we generated testis-specific sPLA 2 -V Tg mice to ask if Tg overexpression of sPLA 2 -V in the testis would also result in the same phenotype. For this purpose, we crossed LNL-V Tg mice with pgk2-Cre Tg mice, in which Cre recombinase is expressed only in spermatogenic cells, allowing Cre/LoxP recombination and thereby activation of the sPLA 2 -V transgene under control of the CAG promoter only in these cells.
The LNL-V Tg ϫ pgk2-Cre Tg litter size was normal, and all the pups were viable and grew normally over 1 year. The transgene-positive pups (pgk2-V Tg mice) overexpressed sPLA 2 -V mRNA only in the testis, as assessed by RNA blotting in this setting (supplemental Fig. S4A). Furthermore, RT-PCR demonstrated increased expression of sPLA 2 -V mRNA in the testis but not in other tissues of the Tg mice in comparison with respective WT tissues with (e.g. spleen) or without (e.g. liver) endogenous sPLA 2 -V expression (supplemental Fig. S4B), thus confirming that the sPLA 2 -V transgene was activated by the CAG promoter only in germ cells of pgk2-V Tg mice. These results indicate that overexpression of sPLA 2 -V in spermatogenic cells does not lead to infertility and does not affect the life of male mice.
It was expected that if male pgk2-V-Tg mice were crossed with female C57BL/6 mice, the CAG promoter-regulated sPLA 2 -V transgene would be transferred to the whole tissues/ cells of their progeny, thereby causing early death of the transgene-positive neonates because of lung collapse. Indeed, the transgene-positive F1 neonates exhibited gasping behavior, tachypnea, and cyanosis, and all died soon after birth because of respiratory failure (data not shown), similar to the outcome in the LNL-V Tg ϫ CAG-Cre Tg pups described above. These results further support our conclusion that overexpression of sPLA 2 -V in the lungs facilitates surfactant hydrolysis, eventually leading to respiratory failure and thereby early death.
Lack of Respiratory Abnormality in sPLA 2 -X Tg Mice-Current biochemical and cell biological studies have established that sPLA 2 -X is the most active sPLA 2 isozyme (sPLA 2 -V is the next) on PC-rich membranes (14 -16). We therefore anticipated that, as in the case of sPLA 2 -V Tg mice, sPLA 2 -X Tg mice would also display a lethal respiratory defect because of aberrant surfactant hydrolysis. Most unexpectedly, however, all the pups of LNL-X Tg (both of two independent founders; see Fig.  1B) ϫ CAG-Cre Tg showed no neonatal death and grew to adulthood. RNA blotting clearly indicated the expression of sPLA 2 -X mRNA in all tissues from the sPLA 2 -X transgene-positive pups (sPLA 2 -X Tg mice) (Fig. 7A). When PLA 2 enzymatic activities in the tissue homogenates from sPLA 2 -X Tg and WT mice were measured using 1-palmitoyl-2-arachidonoyl-PC as a substrate, the activities in tissues of the Tg mice were similar to those of WT mice, with only a modest (1.3-fold) increase in the lung (Fig. 7B). Also, there was no increase in PLA 2 activity in sera of sPLA 2 -X Tg mice compared with that of WT mice (supplemental Fig. S5).
To assess whether sPLA 2 -X mRNA would be transcribed to its protein in sPLA 2 -X Tg mice, we carried out immunoblotting of these homogenates with anti-sPLA 2 -X antibody, using the culture supernatants and cell lysates of PC12 cells transfected with sPLA 2 -X as a positive control. We have reported previously that, in cultured cells (e.g. PC12 cells as shown in Fig. 7C), the active form of sPLA 2 -X, which was produced after proteolytic elimination of the N-terminal propeptide, and its highly glycosylated form, which gave higher and more heterogeneous molecular masses, were present mainly in the culture supernatant, whereas the inactive proenzyme before processing, as well as its glycosylated form, resided in the cells (61). As shown in Fig. 7C, two main bands (ϳ23 and ϳ27 kDa) appeared in most tissues of sPLA 2 -X Tg mice, verifying that the Tg overexpression of sPLA 2 -X in mice was indeed successful. Comparing these bands with those in the positive control, together with the result of the PLA 2 enzyme assay (Fig. 7B), we realized that the two bands found in the Tg-derived tissues corresponded to the inactive zymogen and its glycosylated form (Fig. 7C). Thus, it is reasonable to consider that sPLA 2 -X exists mainly as a zymogen and is barely processed to the active form in most tissues under physiological conditions. An additional point to note was that only a minor but significant fraction of sPLA 2 -X was converted to the active form in the lung homogenate from Tg mice. In addition, the active form appeared to predominate in the Tg brain, even though the level of the overexpressed sPLA 2 -X in this tissue was lower than that in other tissues (Fig. 7C).
The histology of the lungs of sPLA 2 -X Tg and WT mice is shown in Fig. 7D. As in the case of WT mice (Fig. 7D, left panel), the alveolar architecture of Tg mice appeared well organized, and atelectasis was scarcely observed, even though moderately thickened alveolar walls were evident locally (Fig. 7D, right  panel). Furthermore, analysis of lung surfactant by ESI/MS showed that there were minimal changes in the levels of surfactant PC (Fig. 8, upper panel) and PG (Fig. 8, lower panel) of sPLA 2 -X Tg mice compared with those of WT mice. Thus, the presence of almost intact lung surfactant could account for the absence of the lethal respiratory defect in sPLA 2 -X Tg mice, consistent with the notion that sPLA 2 -X remained largely inactive in vivo.
Activation of sPLA 2 -X during Inflammation-As the amino sequence of the N-terminal cleavage site for sPLA 2 -X is similar to that cleaved by several inflammatory proteases (76), we next examined if proteolytic activation of sPLA 2 -X would occur at inflamed sites. We employed a model in which subcutaneous implantation of a cotton thread in the dorsum of mice induces FIGURE 7. Detection of sPLA 2 -X in sPLA 2 -X Tg mice. A, detection of sPLA 2 -X mRNA in tissues of WT and sPLA 2 -X Tg mice by Northern blotting. The membrane was exposed to x-ray film for 1 day. B, PLA 2 enzymatic activity in homogenates of tissues from WT and sPLA 2 -X Tg mice shown in B. C, detection of sPLA 2 -X protein in tissues of WT and sPLA 2 -X Tg mice by immunoblotting. sPLA 2 -X expressed in PC12 cells was used as a positive control. Positions of mature sPLA 2 -X, its proform, and its glycosylated form are shown in the left margin. D, histology of the lungs from WT (left) and sPLA 2 -X Tg (right) mice. The structure of the Tg lung was normal overall, although areas with mildly thickened alveolar walls and narrow air spaces were focally evident (arrows). formation of granulation tissue within 5 days (77). As assessed by immunoblotting, we found that, in addition to the bands for the proenzyme and its glycosylated form, a band corresponding to the active enzyme appeared in the granulation tissue of sPLA 2 -Tg mice on day 5, where nearly half of the overexpressed sPLA 2 -X was processed to the active form (Fig. 9A). Accordingly, the PLA 2 enzymatic activity was significantly increased (3-4-fold) in the homogenate of the granulation tissue in sPLA 2 -Tg mice compared with that in WT mice (Fig. 9B), confirming the activation of sPLA 2 -X in this inflammation model.

DISCUSSION
In an effort to gain new insights into the in vivo functions of sPLA 2 enzymes, we analyzed Tg mice for sPLA 2 -V and sPLA 2 -X, two very potently active enzymes on mammalian cell membranes because of their high interfacial affinity for PC (14 -17). We have obtained unequivocal evidence that sPLA 2 -V targets lung surfactant, leading to catastrophic lung damage similar to RDS, a phenotype that has not been observed in sPLA 2 -IIA Tg mice (19, 20, 22-24, 75, 78). We have also shown that sPLA 2 -X, an enzyme whose activation requires cleavage of an N-terminal propeptide, remains inactive in a physiological milieu and that its proteolytic activation occurs in particular situations such as inflammation.
The major role of surfactant is to reduce the surface tension at the air-liquid interface of the alveolus (27). Reduction of sur-factant results in loss of alveolar stability and severe deterioration of gas exchange, leading to alveolar collapse. In this study, we found that sPLA 2 -V Tg mice died immediately after birth because of respiratory failure resulting from aberrant hydrolysis of lung surfactant phospholipids (Figs. 2-6). This neonatal death of sPLA 2 -V Tg mice is in many aspects similar, even if not entirely identical, to that observed in mutant mice manipulated for a panel of genes that have been implicated in surfactant homeostasis either directly or indirectly. These include, for example, mice deficient in Foxa2 (63), a transcription factor prerequisite for the expression of ABCA3 (for ATP-binding cassette A3) that plays a critical role in proper trafficking of surfactant in alveolar type II cells (79), mice ablated for hypoxia-inducible transcription factor-2␣, in which maturation of type II cells and thereby surfactant synthesis are prevented (64), and mice null for endothelial nitric-oxide synthase, in which alveolar capillary dysplasia eventually causes marked thickening of saccular septa and reduction of surfactant material (65). Mice manipulated for surfactant apolipoprotein genes also display neonatal RDS, including mice with targeted disruption of SP-B, in which the routing, storage, and secretion of surfactant by type II cells are mitigated (67), and mice overexpressing SP-C, in which lung development is delayed or arrested because of cytotoxic aggregation of SP-C in the early surfactant secretory pathway (66). In all of these cases (as in the case of sPLA 2 -V Tg mice), mice are born normally because oxygen is supplied from the mother in utero, but they die shortly after birth because of RDS due to surfactant insufficiency. Likewise, newborn human infants with hereditary SP-B deficiency (80) or ABCA3 mutation (81) develop unremitting lethal RDS. A difference between these previously reported mouse models and the sPLA 2 -Tg mice analyzed here is that the former mostly arises from defects in surfactant synthesis and/or secretion, whereas the latter involves sPLA 2 -V-directed hydrolysis of surfactant after secretion. Although it remains obscure if sPLA 2 -V overexpression also affects surfactant homeostasis before secretion, virtually normal expression of SP-C in sPLA 2 -V Tg mice (Fig. 4D) suggests that terminal maturation of, and thereby surfactant synthesis by, alveolar type II cells may not be profoundly (even if not entirely because PAS-positive immature type II cells are focally present (Fig. 4E)) influenced by sPLA 2 -V.
Acute RDS is a clinically and pathologically complex syndrome because of acute life-threatening lung injury involving an alteration of pulmonary surfactant, and numerous predisposing factors can be involved in the etiology, including pneumonia and sepsis (27). Hydrolysis of surfactant phospholipids is an early pathological event of RDS, and hydrolysis of as little as 10 -15% of surfactant can cause this serious condition (82,83). Levels of PLA 2 activity in BALF of patients with RDS are often positively correlated with disease severity (27)(28)(29)(30)(31), and a chemical sPLA 2 inhibitor (S-5920/LY315920Na), which was initially developed to target sPLA 2 -IIA but subsequently shown to inhibit several sPLA 2 isozymes with varied potency (84), protects animals against experimental RDS (32). These observations have led to the hypothesis that certain sPLA 2 isozymes do contribute to surfactant hydrolysis associated with lung inflammation and injury. Several sPLA 2 isozymes, including IB, IIA, IID, III, V, and X, are expressed in the lung (58,85). Until recently, sPLA 2 -IIA has been thought to be primarily responsible for RDS pathogenesis, because it is markedly elevated in BALF of various animal species with RDS or related lung injury (27)(28)(29)(30)(31)(32), and it hydrolyzes surfactant phospholipids, particularly PG, in vitro (35) and in vivo following intratracheal instillation (33). In contrast, however, SP-A, which is a homolog of sPLA 2 inhibitory protein in snakes and does inhibit the activity of sPLA 2 -IIA (28), exerts an inhibitory effect on surfactant hydrolysis even in mouse strains that do not intrinsically express sPLA 2 -IIA (33), and it is now generally recognized that the hydrolytic action of sPLA 2 -IIA on PC is very weak (1,2). Moreover, unlike sPLA 2 -V Tg mice (as used in this study), sPLA 2 -IIA Tg mice reportedly do not display any lung disorder (19, 20, 22-24, 75, 78). These observations suggest that sPLA 2 -IIA may not be the sole isozyme responsible for the hydrolysis of surfactant, particularly PC.
To our knowledge, this study is the first clear demonstration that sPLA 2 -V is capable of potently degrading surfactant phospholipids in vivo, and it lends support to the intriguing possibility that sPLA 2 -V may contribute profoundly, if not solely, to the hydrolysis of surfactant during RDS pathogenesis. In support of this notion, the expression of sPLA 2 -V is markedly elevated in bronchial and alveolar epithelial cells as well as in alveolar macrophages in the lungs of patients with severe pneumonia (58). sPLA 2 -V efficiently hydrolyzes PC, PG, and PE in surfactant reconstituted in vitro (35), consistent with our present in vivo study. sPLA 2 -V has a unique enzymatic property in showing a preference for PC with palmitic acid at the sn-2 position to that with AA in vitro (86) (even though it can release AA from cellular membranes in culture (14,17)), which is in line with the fatty acid composition of surfactant PC. Moreover, beyond the limitations of the Tg overexpression strategy, only Tg mice for sPLA 2 -V, but not those for sPLA 2 -IIA (19, 20, 22-24, 75, 78), for sPLA 2 -X (Figs. 7 and 8), and for sPLA 2 -III, 4 exhibit neonatal lethality because of RDS, implying the specific action of sPLA 2 -V on surfactant phospholipids in vivo. Nonetheless, definitive evidence that sPLA 2 -V does, in fact, play a role in inflammation-associated surfactant destruction in the lung awaits further studies employing proper RDS models in sPLA 2 -V-null mice.
On the other hand, the involvement of sPLA 2 -V in surfactant metabolism under "physiological" conditions is unlikely, because its expression in normal lung is low and limited to a population of bronchial epithelial cells in both mice (Fig. 4B,  panel a) and humans (58), and more importantly, because our preliminary ESI/MS analysis has shown that surfactant levels are comparable between control mice and sPLA 2 -V-null mice. 5 With regard to this issue, two lysosomal PLA 2 s, namely aiPLA 2 / peroxiredoxin 6 in pneumocytes (11) and LPLA 2 /1-O-acylceramide synthase in alveolar macrophages (12), have recently been implicated in physiological surfactant metabolism. Thus, gene targeting of aiPLA 2 /peroxiredoxin 6 has revealed that this enzyme is responsible for ϳ40% of steady-state surfactant PC degradation in the lysosomes of alveolar cells (11), and targeting of LPLA 2 /1-O-acylceramide synthase results in marked accumulation of lung surfactant, accompanied by macrophage phospholipidosis, by 1 year of age (12).
Application of the acute RDS model to cPLA 2 ␣-null mice has demonstrated the crucial role of cPLA 2 ␣-derived lipid eicosanoids in lung injury (38). Although sPLA 2 -V has the ability to promote eicosanoid generation by various cells in vitro (14,17) and by macrophages in vivo as revealed in sPLA 2 -V-null mice (49), it is unlikely that eicosanoids participate in the lung injury of sPLA 2 -V Tg mice because the increases in eicosanoids in BALF of sPLA 2 -V Tg mice are only modest (for PGE 2 ) or absent (for LTC 4 ) (Fig. 5). This is probably because massive infiltration of eicosanoid-producing leukocytes does not occur in the lungs of sPLA 2 -V Tg neonates. A small increase in PGE 2 might result from the direct action of sPLA 2 -V on macrophages, whose number is increased moderately in Tg mice (Fig. 4C), or on airway epithelial cells, particularly those undergoing membrane damage, as sPLA 2 -V is potently active on membranedamaged cells following oxygen stress (supplemental Fig. S3). Alternatively, sPLA 2 -V might increase AA release from airway cells in concert with cPLA 2 ␣, an event that has been observed in several cell culture studies (87)(88)(89). We do not rule out the possibility that sPLA 2 -V-derived eicosanoids do participate in the process of some pulmonary diseases such as those involving eosinophil infiltration, because sPLA 2 -V released from bronchial epithelial cells facilitates membrane hydrolysis of adjacent eosinophils in co-culture through the transcellular route (90).
Importantly, comparison between sPLA 2 -V Tg mice and sPLA 2 -IIA Tg mice sheds light on the distinct roles of these two enzymes in vivo. As mentioned above, the lethal lung phenotype is peculiar to Tg mice for sPLA 2 -V but not to those for sPLA 2 -IIA (and several other sPLA 2 s), emphasizing a specific role of sPLA 2 -V in the lung. Besides the severe lung injury, sPLA 2 -V Tg mice show a symptom of dermal inflammation accompanied by edema and macrophage infiltration (supplemental Fig. S2). In contrast, no inflammatory phenotype has been reported in the skin (or any other tissue) of sPLA 2 -IIA Tg mice, which show permanent hair loss associated with hyperkeratosis (78). This difference might be related to the well established view that sPLA 2 -V has the ability to elicit eicosanoid production more potently than does sPLA 2 -IIA in various cells (14,17). Although male infertility because of impaired spermatogenesis has been reported for testis-specific sPLA 2 -IIA Tg mice (75), testis-specific sPLA 2 -V Tg mice are fertile. So far, no reproductive abnormality has been observed in sPLA 2 -V knock-out and Tg mice, even though sPLA 2 -V is expressed in the acrosomes of spermatogenic cells (57) and a potential role of the PLA 2 reaction product, lyso-PC, in the acrosome reaction and fertilization has been proposed (91). By contrast, whether sPLA 2 -V does play a role in defense against bacterial infection and increased atherosclerosis, which are prominent phenotypes of sPLA 2 -IIA Tg mice that may be relevant to the physiological functions of this enzyme (19,20,(22)(23)(24), is difficult to assess with sPLA 2 -V Tg mice because all die in the neonatal period. Because sPLA 2 -V exhibits bactericidal activity in vitro (less efficiently than sPLA 2 -IIA) (20, 21), promotes plasma lipoprotein hydrolysis (more potently than sPLA 2 -IIA) (26), and is detected in atherosclerotic lesions (as is sPLA 2 -IIA) (26), it is possible that sPLA 2 -IIA and sPLA 2 -V share redundant roles or act in a coordinated manner in anti-bacterial innate immunity and atherosclerosis development. Future studies using tissue-specific sPLA 2 -V Tg mice, as well as sPLA 2 -V knock-out mice, would be helpful for clarifying the roles of sPLA 2 -V in these situations.
Transfection of sPLA 2 -X in several types of mammalian cell results in spontaneous release of its active form into the culture medium (14,58,61), although the entity of the protease(s) responsible for the proteolytic processing of this enzyme is unknown. The amino acid sequence of the proteolytic cleavage site in sPLA 2 -X is similar to that cleaved by subtilisin-like intracellular processing proteases (76), and in fact, intracellularly activated sPLA 2 -X promotes AA release prior to secretion in HEK293 transfectants (90,92). However, subsequent studies have shown that the active form of sPLA 2 -X is predominantly present in the culture supernatants of sPLA 2 -X-transfected bronchial epithelial cells, fibroblasts, and neuronal cells, leading to the suggestion that certain protease(s) present in serum, secreted from cells, or on cell surfaces may play a major role in the activation of sPLA 2 -X (61). This idea is supported by our present finding that sPLA 2 -X exists as a zymogen in most tissues of sPLA 2 -X Tg mice under physiological conditions (Fig.  7), thus largely explaining why sPLA 2 -X Tg mice do not exhibit the neonatal RDS phenotype (Figs. 7D and 8), regardless of the reported capacity of this enzyme to hydrolyze surfactant in vitro (33,35). Moreover, the observation that sPLA 2 -X, but not sPLA 2 -V, is inhibited by SP-A (33) may also provide an additional explanation for the fact that only sPLA 2 -V Tg mice exhibit the RDS phenotype. Interestingly, activation (processing) of sPLA 2 -X was evident in the inflammatory granulation tissue of Tg mice (Fig. 9). Several inflammation-associated extracellular proteases, such as cathepsins and kallikreins, as well as type II transmembrane serine proteases, such as hepsin of which expression is elevated in cancer (93), have a potential to cleave the propeptide sequence of sPLA 2 -X. sPLA 2 -IB, another sPLA 2 that possesses an N-terminal propeptide, can be converted to its active form by extracellular proteases such as trypsin and serum plasmin (94). We therefore propose that the activation of sPLA 2 -X occurs following the action of these disease-associated proteases, in particular pathological circumstances such as inflammation, tissue injury, or even cancer. This tightly regulated activation mechanism seems to be of great biological importance, as unregulated activation of sPLA 2 -X would be detrimental for the host because of its marked activity on cell membranes.
From this standpoint, we do not rule out the possibility that sPLA 2 -X could also participate in surfactant degradation and associated lung injury, in which the processing of sPLA 2 -X, and thereby surfactant hydrolysis, might occur. Indeed, some histological abnormality (alveolar wall thickness) was evident locally in sPLA 2 -X Tg neonates (Fig. 7D). This may be related to the presence of a small fraction of the active form of sPLA 2 -X in the Tg lung (Fig. 7C), even though the bulk surfactant composition in the Tg mice appears normal (Fig. 8). The fact that endogenous sPLA 2 -X is localized in type II epithelial cells (15) also supports a possible relationship of this enzyme with surfactant metabolism in certain situations. Another notable feature is that the active form of sPLA 2 -X was present in the brain of sPLA 2 -X Tg mice (Fig. 7C). Considering that sPLA 2 -X is located in neuronal fibers and that it is capable of facilitating neuritogenesis in cultured neuronal cells (61), processing of sPLA 2 -X might occur constitutively in neuronal cells. Likewise, endogenous sPLA 2 -X has been reported to be expressed in skin keratinocytes (95), leukocytes (76), and the gastrointestinal tract (96), and we have recently found that sPLA 2 -X Tg mice display some phenotypes unique to these tissues. 6 Further studies using sPLA 2 -X Tg mice as well as sPLA 2 -X knock-out mice should help to reveal the tissue-specific roles of sPLA 2 -X, in association with its proteolytic processing and the identity of its activating protease(s).