Human sPLA2-III [group III secreted PLA2 (phospholipase A2)] is an atypical sPLA2 isoenzyme that consists of a central group III sPLA2 domain flanked by unique N- and C-terminal domains. In the present study, we found that sPLA2-III is expressed in neuronal cells, such as peripheral neuronal fibres, spinal DRG (dorsal root ganglia) neurons and cerebellar Purkinje cells. Adenoviral expression of sPLA2-III in PC12 cells (pheochromocytoma cells) or DRG explants facilitated neurite outgrowth, whereas expression of a catalytically inactive sPLA2-III mutant or use of sPLA2-III-directed siRNA (small interfering RNA) reduced NGF (nerve growth factor)-induced neuritogenesis. sPLA2-III also suppressed neuronal death induced by NGF deprivation. Lipid MS revealed that sPLA2-III overexpression increased the cellular level of lysophosphatidylcholine, a PLA2 reaction product with neuritogenic and neurotropic activities, whereas siRNA knockdown reduced the level of lysophosphatidylcholine. These observations suggest the potential contribution of sPLA2-III to neuronal differentiation and its function under certain conditions.
PLA2 (phospholipase A2) enzymes hydrolyse glycerophospholipids to produce lysophospholipids and free fatty acids. These lipid products or their metabolites, such as arachidonic acid-derived eicosanoids, LPA (lysophosphatidic acid) and platelet-activating factor, exert various effects on cells by acting on their cognate receptors. PLA2s have also been implicated in membrane remodelling as well as other biological events. To date, a number of mammalian PLA2s have been identified and subdivided into several groups, including the sPLA2 (secreted PLA2), cPLA2 (cytosolic PLA2) and iPLA2 (Ca2+-independent PLA2) families [1,2].
The sPLA2 family (sPLA2-IB, -IIA, -IIC, -IID, -IIE, -IIF, -III, -V, -X and -XIIA) represents a group of structurally related, disulfide-rich, Ca2+-dependent, low-molecular-mass enzymes with a histidine–aspartic acid residue catalytic dyad [1,2]. Individual sPLA2s exhibit unique tissue and cellular localizations and enzymatic properties, suggesting that they perform distinct biological roles, probably by acting on different target substrates. Previous studies utilizing gene-manipulated mice have uncovered some biological functions of sPLA2s: sPLA2-IB plays a role in the digestion of dietary phospholipids [3,4], sPLA2-IIA in antibacterial defence [5,6], sPLA2-V in eicosanoid production and phagocytosis by macrophages [7,8], atherosclerosis development  and lung surfactant hydrolysis , and sPLA2-X in allergen-induced airway inflammation and remodelling . Nevertheless, the tissue- or cell-specific roles of each sPLA2 are still a subject of debate.
Current evidence suggests that sPLA2s may affect some neuronal functions, such as neuritogenesis, neurotoxicity, neurotransmitter release and neuronal survival [12–17]. Exogenous sPLA2s from fungi and snake or bee venom can exhibit neuritogenic or neurotropic actions through their products (fatty acids and lysophospholipids), which often act synergistically [18–21]. Work by our group and by others has shown that, among mammalian sPLA2s, sPLA2-X is the most potent hydrolyser of the PC (phosphatidylcholine)-rich outer plasma membrane [22–24]. sPLA2-X is also expressed in peripheral neurons at a low but significant level, and exhibits a neuritogenic effect which is dependent on the production of LPC (lysophosphatidylcholine) [19,25,26]. It has been proposed that these neuronal activities of LPC are mediated by the L-type Ca2+ channel , G-protein-coupled receptor  or by altered membrane curvature and fluidity .
Human sPLA2-III (group III sPLA2) is an atypical sPLA2 that consists of unique N-terminal and C-terminal domains and a central sPLA2 domain, the S domain, which has higher homology with bee venom sPLA2 (a prototypic group III enzyme) than with other mammalian sPLA2s . The full-length sPLA2-III is processed to the mature form, which contains only the S domain, and this domain is sufficient for enzymatic function [27–29]. Overexpression studies have indicated that the cellular arachidonate-releasing function of sPLA2-III is, in general terms, less than those of sPLA2-X and sPLA2-V and superior to that of sPLA2-IIA [28,29]. Since bee venom sPLA2 exerts various pharmacologic effects [13,17,30–32], it is anticipated that sPLA2-III could function as an endogenous counterpart of bee venom sPLA2 in physiological milieu. However, beyond its arachidonate-releasing capacity as noted above, the true biological functions of sPLA2-III are still largely unknown.
Given that bee venom PLA2 has neuritogenic and neurotropic capacities [13,17], we were interested in the neuronal expression and function of sPLA2-III. In the present study, we show that sPLA2-III is expressed in the central nervous system and peripheral neuronal fibres, and that it has the ability to promote neuronal outgrowth and survival, most probably through the production of LPC.
MATERIALS AND METHODS
Rat PC12 cells were obtained from the Riken Cell Bank (Tsukuba, Ibaraki, Japan) and were cultured in RPMI 1640 medium (Nissui Pharmaceutical, Tokyo, Japan) containing 10% (v/v) fetal horse serum (Sigma), 5% (v/v) fetal calf serum (Sigma), 2 mM L-glutamine (Invitrogen), 100 i.u./ml penicillin (Invitrogen), 100 μg/ml streptomycin (Invitrogen) and 100 mM non-essential amino acids (Invitrogen) at 37 °C in humidified air with 5% CO2. For neural differentiation, the cells (1×104 cells/ml) were cultured with 1 ng/ml (suboptimal) or 10 ng/ml (optimal) rat NGF (nerve growth factor; Sigma) in type I collagen-coated 6- or 12-well plates (Iwaki Glass, Funabashi, Chiba, Japan) as described previously . The cells were cultured with 10 ng/ml NGF for 3 days and then deprived of NGF for additional periods. Where required, cell viability was assessed by the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide] assay using cell counting kit 8 (Dojin Biochemicals, Kumamoto, Japan), following the manufacturer's instructions.
Isolation and culture of mouse DRG (dorsal root ganglia) explants
Mouse DRG explants were prepared from C57BL/6 mouse fetuses at E12.5 (embryonic day 12.5)–13 (Shizuoka Laboratory Animal Centre, Hamamatsu, Japan) as described previously . Briefly, mouse fetuses were incised longitudinally along the spinal cord from the upper cervical vertebrae to the tail. The spinal cord sections were separated and removed, and DRG neurons attached to the spinal cord were picked up and harvested. DRG explants were then cultured for up to 36 h in Dulbecco's modified Eagle's medium (Nissui Pharmaceutical) supplemented with the appropriate concentration of NGF as required in poly-D-lysine-coated glass-bottomed dishes (MatTek, Ashland, MA, U.S.A.) pre-coated with 20 μg/ml laminin (Sigma).
RT–PCR (reverse transcription–PCR)
Synthesis of cDNA was performed using 0.5 μg of total RNA from cells and tissues and AMV (avian myeloblastosis virus) reverse transcriptase, according to the manufacturer's instructions as supplied with the RNA PCR kit (TaKaRa Biomedicals, Osaka, Japan). Subsequent amplification of the cDNA fragments was performed using 0.5 μl of the reverse-transcribed mixture as a template with specific primers for each sPLA2 (Fasmac, Atsugi City, Kanagawa, Japan). For amplification of sPLA2-V and -X, a set of 23 bp oligonucleotide primers corresponding to 5′- and 3′-nucleotide sequences of the open reading frames were used as primers as described previously . The PCR conditions for sPLA2-V and -X were: 94 °C for 30 s, followed by 30 cycles of amplification at 94 °C for 5 s and 68 °C for 4 min, using the Advantage cDNA polymerase mix (Clontech). The PCR conditions for sPLA2-III were 94 °C for 30 s, followed by 30 cycles of amplification at 94 °C for 30 s, 58 °C for 30 s, and 72 °C for 30 s, using ExTaq polymerase (TaKaRa Biomedicals) and the primers 5′-CTTGCAGTATAACCTATGGCATCCG-3′ and 5′-CTCCAGCACGTTGAAGAAGGCCAC-3′ (Fasmac). The PCR products were analysed by 1% (w/v) agarose gel electrophoresis and visualized by use of ethidium bromide.
Lysates from 2×105 cultured cells or 25 μg protein equivalent of tissue homogenate in TBS (Tris-buffered saline) cell lysis buffer [10 mM Tris/HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1 mM PMSF, 2 μg/ml leupeptin and 2 μg/ml pepstatin] were subjected to SDS/PAGE under reducing conditions with 2-mercaptoethanol. Protein concentrations in the samples were determined using a BCA (bicinchoninic acid) protein assay kit (Pierce), with BSA (Sigma) used as a standard. The separated proteins were electroblotted on to nitrocellulose membranes (Schleicher and Schuell) using a semi-dry blotter (MilliBlot-SDE system; Millipore). After blocking with 5% (w/v) skimmed milk in TBS-Tween buffer [10 mM Tris/HCl (pH 7.4), 150 mM NaCl and 0.05% (v/v) Tween 20], the membranes were probed with anti-sPLA2-III antibody (1:5000 dilution in TBS-Tween buffer) for 2 h. Membranes were also probed using antibodies against p-ERK [phosphorylated ERK (extracellular-signal-regulated kinase)] and p-Akt (phosphorylated Akt) (BD Biosciences), neurofilament M (Chemicon), and caspase 3 (Santa Cruz Biotechnology), following the manufacturer's instructions. The membranes were then incubated for 2 h with horseradish peroxidase-conjugated rabbit anti-IgG antibody (Zymed Laboratories) at a concentration of 1:5000 in TBS-Tween buffer and visualized using the enhanced chemiluminescence Western blot system (PerkinElmer) as described previously .
Immunohistochemical staining of human tissue sections was performed with approval from the ethics committee of our Faculty, as described previously [25,29,33]. Briefly, the tissues were embedded in paraffin, sectioned and mounted on slides, deparaffinized in 100% xylene for 30 min and rehydrated for 5 min in 100, 90, 80 and 70% (v/v) ethanol. The tissue sections (4 μm) were incubated with Target Retrieval Solution (Dako, Carpinteria, CA, U.S.A.) as required, incubated for 10 min with 3% (v/v) H2O2, washed three times for 5 min with TBS, incubated with 5% (v/v) skimmed milk in TBS for 30 min, washed 3 times with TBS for 5 min each, and incubated for 2 h with anti-sPLA2 antibodies at 1:200 dilution in TBS. The sections were then treated with a CSA (catalysed signal amplification) system staining kit (Dako) with diaminobenzidine substrate. The cell type was identified by conventional haematoxylin and eosin staining of serial sections adjacent to the specimen used for immunohistochemistry.
Expression of sPLA2-III using the adenovirus system
Preparation of adenovirus was performed as described previously . cDNAs for sPLA2-III-S and its catalytically dead mutant sPLA2-III-S-HQ were subcloned into the pENTER/D-TOPO vector using the pENTER Directional TOPO Cloning kit (Invitrogen). After purification of the plasmids from transformed Top10 competent cells (Invitrogen), the inserts were subcloned into the pAd/CMV/V5-DEST vector (Invitrogen) using the Gateway system and LR clonase (Invitrogen). The plasmids were purified and digested with PacI (New England Biolabs). The linearized plasmids (1–2 μg) were then mixed with 4 μl of Lipofectamine™ 2000 (Invitrogen) in 200 μl of OptiMEM medium (Invitrogen) and transfected into subconfluent HEK-293A cells (human embryonic kidney 293A cells) (Invitrogen) in 1 ml of OptiMEM in 6-well plates (Iwaki Glass). HEK-293A cells were cultured for 1–2 weeks in RPMI 1640 medium containing 10% (v/v) fetal calf serum, with replacement of the medium every 2 days. When most of the cells had become detached from the plates, the cells and culture medium were harvested, freeze–thawed twice and centrifuged to obtain the adenovirus-enriched supernatants. Aliquots of the supernatants were added to fresh HEK-293A cells and cultured for 2–3 days to amplify the adenoviruses. After 2–4 rounds of amplification, the resulting adenovirus-containing medium was used as virus stock. Viral titres were determined using a plaque-forming assay with HEK-293A cells. As a control, the pAd/CMV/V5-GW/LacZ vector (Invitrogen) was digested with PacI and transfected into HEK-293A cells to produce LacZ-bearing adenovirus. Aliquots of the adenovirus-containing medium were added to PC12 cells and cultured for appropriate periods with appropriate concentrations of NGF for subsequent analyses.
Experiments with siRNA (small interfering RNA)
A set of synthetic hairpin-forming oligonucleotides directed against sPLA2-III [5′-CACCGCTATGGCATCCGAAACTACTTCAAGAGAGTAGTTTCGGATGCCATAG-3′ (sense) and 5′-AAAACTATGGCATCCGAAACTACTCTCTTGAAGTAGTTTCGGATGCCATAGC-3′ (antisense)] were prepared. After annealing, the oligonucleotides were subcloned into the pENTER™/U6 vector using the BLOCK-iT™ U6 RNAi Entry Vector kit (Invitrogen). After transformation into Top10 competent cells and plasmid purification, the insert was transferred to the pAd/BLOCK-iT™-DEST vector using the Gateway system and LR clonase (Invitrogen). After transformation into Top10 competent cells and plasmid purification, the plasmid was linearized by digestion with PacI and transfected using Lipofectamine™ 2000 into HEK-293A cells, in order to produce adenovirus carrying the sPLA2-III siRNA construct. After 2–4 rounds of amplification, the resulting adenovirus-containing medium was used for subsequent analyses, as described previously .
ESI (electrospray ionization)–MS
MS spectra were obtained on a Quattro Micro tandem quadrupole mass spectrometer (Micromass, Manchester, U.K.) equipped with an electrospray ion source, as described previously [10,25]. Lipid extracts obtained from PC12 cells (100 μg protein equivalent) were reconstituted in chloroform/methanol (2:1, v/v) (100–300 μmol phosphorus/l), and 2 μl of the sample was injected per run. As an internal standard, 100 pmol of LPC (C12:0) was added to the samples. The samples were introduced by means of a flow injector into the ESI chamber at a flow rate of 4 μl/min in a solvent system of acetonitrile/methanol/water (2:3:1, by vol.) 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 litres/min at 80 °C. The capillary and cone voltages were set at 3.7 kV and 30 V respectively, argon at (3–4)×104 torr (1 torr=0.133 kPa) was used as the collision gas, and a collision energy of 30–40 V was used for obtaining fragment ions for precursor ions.
sPLA2-III is expressed in neuronal cells and tissues
To assess the expression of sPLA2-III protein in mouse tissues, homogenates of C57BL/6 mouse tissues were subjected to immunoblotting using an anti-sPLA2-III antibody. An 18 kDa immunoreactive band, which appeared to represent sPLA2-III-S, was detected faintly in the brain and intensely in the spinal cord (Figure 1A), revealing expression of sPLA2-III in the central nervous system. The same 18 kDa immunoreactive band was also strongly detected in the lung and stomach (Figure 1A). When DRG explants isolated from fetal C57/BL6 mice (E12.5) were subjected to RT–PCR for various sPLA2s, sPLA2-III transcript and, to a lesser extent, sPLA2-X (as reported previously ) were detected, but not sPLA2-V (Figure 1B). Immunoblotting also showed the presence of the sPLA2-III protein (mainly as the sPLA2-III-S form) in mouse DRG (Figure 1C). In addition, immunohistochemistry of human DRG demonstrated marked staining with anti-sPLA2-III antibody (Figure 1D, panel a), but not with pre-immune serum (Figure 1D, panel b). Furthermore, expression of the sPLA2-III-S form of sPLA2-III was weakly detected in undifferentiated PC12 cells, and this was increased following culture for 3 days with 10 ng/ml NGF (Figure 1E), which resulted in marked neurite extension (see below) .
Expression of sPLA2-III in neuronal tissues and cells
Since sPLA2-III expression was intensely detected in mouse stomach by immunoblotting (Figure 1A), we performed immunohistochemistry on human stomach with anti-sPLA2-III antibody to identify in which particular cell types sPLA2-III was present. We found that sPLA2-III was preferentially localized in neuronal fibres (Figure 1F, black arrows) as well as in microvascular endothelium (Figure 1F, red arrows). The distribution of sPLA2-III in the endothelium is consistent with our previous results . Localization of sPLA2-III immunoreactivity to peripheral neuronal fibres was also observed in serial sections of human aorta, which were markedly stained with two distinct anti-sPLA2-III antibodies (Figure 1G, panels b and c), but not with pre-immune serum (Figure 1G, panel a). Moreover, sPLA2-III immunoreactivity was clearly localized in Purkinje cells from human cerebellum (Figure 1H, panel b). In comparison, anti-sPLA2-X antibody stained neuronal fibres, but not Purkinje cells (Figure 1H, panel c), whereas the control antibody provided no appreciable staining in the cerebellum (Figure 1H, panel a).
sPLA2-III induces neurite outgrowth in PC12 cells
Having established that sPLA2-III is widely expressed in neuronal cells, we next investigated the impact of sPLA2-III on cellular functions by adenovirally transfecting sPLA2-III-S or sPLA2-III-S-HQ (in which the catalytic-centre histidine residue is replaced with a glutamine residue) into PC12 cells. Lysates of cells infected with adenovirus expressing sPLA2-III-S or sPLA2-III-S-HQ contained an intense 18 kDa band, which was exactly the same size as the band detected faintly in cells infected with control adenovirus (endogenous sPLA2-III-S; see Figure 1E) (Figure 2A). Notably, transfection of sPLA2-III-S facilitated neurite extension in PC12 cells that were cultured in the presence of a suboptimal concentration (1 ng/ml) of NGF (Figures 2B and 2C). This neuritogenic effect of sPLA2-III required catalytic activity, because no neurite outgrowth was induced in cells transfected with the catalytically inactive sPLA2-III-S-HQ mutant (Figures 2B and 2C).
Overexpression of sPLA2-III induces neuritogenesis of PC12 cells
To address whether endogenous sPLA2-III would participate in NGF-induced neuritogenesis, we attempted to knockdown endogenous sPLA2-III, using siRNA. As assessed by RT–PCR, expression of endogenous sPLA2-III was almost totally ablated in cells infected with adenovirus expressing sPLA2-III-directed siRNA when compared with cells treated with control adenovirus expressing either LacZ or a control oligonucleotide (sPLA2-IIA siRNA) (Figure 3A). When the cells were cultured with 10 ng/ml NGF, marked neuritogenesis occurred in cells infected with adenovirus expressing LacZ or a control oligonucleotide (sPLA2-IIA siRNA), whereas neurite outgrowth was scarcely observed in cells infected with adenovirus expressing sPLA2-III siRNA (Figures 3B and 3C). These results suggest that endogenous sPLA2-III profoundly affects neuritogenesis of PC12 cells.
siRNA for sPLA2-III suppresses NGF-induced neuritogenesis of PC12 cells
The observation that sPLA2-III enzyme activity is required for neurite outgrowth (Figure 2) suggests that some catalytic product (or products) generated by sPLA2-III is responsible for the neuritogenic effect of this enzyme. Previous reports have suggested that LPC, a hydrolytic product of membrane PC, has the ability to induce neurite extension and that sPLA2-X does so via the production of LPC [25,26]. Therefore we attempted to assess whether sPLA2-III could indeed generate LPC molecular species in PC12 cell membranes, using ESI–MS to detect LPC species. As shown in Figure 4(A), LPC subclasses with C16:0 and C18:0 were increased in sPLA2-III-S-transfected cells, whereas no increase in these LPC species was evident in sPLA2-III-S-HQ-transfected cells when compared with those in control cells. Conversely, LPC molecular species with C16:0 and C18:0 were decreased by approx. 50% in cells transfected with siRNA against sPLA2-III relative to those species in control cells (Figure 4B). These results suggest that sPLA2-III (both overexpressed and endogenously expressed) hydrolyses PC to generate LPC in cellular membranes. Considering that expression of endogenous sPLA2-III was almost undetectable after treatment with sPLA2-III siRNA (Figure 3A), it appears that the remaining LPC in the siRNA-treated cells (Figure 4B, lower panel) may be produced by other PLA2s.
ESI–MS analysis of cellular LPC molecular species
In addition, when PC12 cells were infected with sPLA2-III-S-HQ, the neurite extension induced by 10 ng/ml NGF was significantly reduced in terms of both neurite length and number (Figures 5A and 5B). Such an inhibitory effect was not observed when inactive mutants for other sPLA2s, such as sPLA2-IIA and -X, were transfected into PC12 cells (results not shown). A possible interpretation of these results is that there may be a specific binding site for sPLA2-III on PC12 cells, and that the HQ mutant blocked the interaction of endogenous sPLA2-III with this site in a dominant-negative manner, although this possibility requires further elucidation.
The catalytically inactive sPLA2-III mutant suppresses NGF-induced neuritogenesis of PC12 cells
sPLA2-III promotes neuronal cell survival on NGF deprivation
Once PC12 cells undergo neuronal differentiation, removal of NGF from the culture leads to neurite retraction and cell death [34–36]. Therefore we next examined the neurotropic effect of sPLA2-III following NGF withdrawal. PC12 cells markedly developed neurites after culture for 3 days with 10 ng/ml NGF, but 24–36 h after NGF deprivation, the cells had lost the majority of neurites and became detached from the culture dishes (Figures 6A and 6B). In sPLA2-III-S-transfected PC12 cells, which displayed more neurites with branches than control cells in culture containing 10 ng/ml NGF (in agreement with the results shown above), the neurite retraction on NGF depletion occurred slowly, with ∼80% and ∼50% of the cells still retaining neurites (albeit retracted) 24 h and 36 h after depletion respectively (Figures 6A and 6B), suggesting that sPLA2-III supports neuronal survival. Cell viability was determined, and sPLA2-III-transfected cells displayed an increase in cell viability compared with control cells 12 h after NGF removal, which agreed with this notion (Figure 6C). The survival-supporting ability of sPLA2-III appeared to depend on its catalytic activity, since the sPLA2-III-S-HQ mutant failed to suppress cell death (Figure 6D). Moreover, a lower level of pro-caspase 3 activation, an indicator of apoptosis, was observed in sPLA2-III-S-transfected cells than in mock-transfected cells (Figure 6E), further supporting the neuronal survival-promoting capacity of sPLA2-III.
sPLA2-III promotes neuronal survival upon NGF deprivation in PC12 cells
NGF-induced neuronal differentiation and survival signals are known to link to the mitogen-activated protein kinase pathway and the phosphoinositide 3-kinase/Akt pathway [34–36]. As shown by immunoblotting for phosphorylated (and thereby activated) forms of ERKs (p-ERK1 and p-ERK2) and Akt (p-Akt), both signalling pathways were activated in the presence of NGF and were down-regulated after the removal of NGF in mock-transfected cells (Figure 6F). The levels of both p-ERK-1 and p-Akt were modestly higher (1.4- and 1.6-fold respectively, as determined by densitometric analysis) in sPLA2-III-S-transfected cells than in control cells in the presence of NGF. Strikingly, in sPLA2-III-S-transfected cells after NGF deprivation, the phosphorylation of ERKs was considerably increased, whereas that of Akt was decreased, but still detectable (Figure 6F). Thus sPLA2-III transmits the signals essential for cell survival in a manner distinct from those induced by NGF.
Supporting the above observations, marked expression of neurofilament M, a marker of neurite outgrowth and neuronal differentiation, was detected in control cells in the presence of NGF and entirely disappeared after NGF removal (Figure 6F). Consistent with the neuritogenic effect of sPLA2-III, neurofilament M expression was significantly increased in sPLA2-III-S-transfected cells, in comparison with expression in control cells under NGF-supplemented conditions. Moreover, sPLA2-III-S-transfected cells retained a high level of neurofilament M expression, even after NGF depletion (Figure 6F). Additionally, expression of βIII tubulin, an immature neuron marker, was decreased in sPLA2-III-S-transfected cells relative to that in control cells in the presence of NGF (Figure 6F), providing additional support for the hypothesis that sPLA2-III facilitates neuronal differentiation. After NGF deprivation, βIII tubulin expression disappeared in both mock- and sPLA2-III-S-transfected cells.
sPLA2-III induces neurite outgrowth in primary DRG explants
In order to assess whether the neuritogenic action of sPLA2-III was exclusive to a specific cell line (PC12) or whether it also occured in other neuronal cells, we examined mouse DRG explants (E12.5), exploiting the fact that when such explants are cultured with NGF for 24–48 h on laminin-coated plates, there is a marked extension of neuronal fibres . When these explants were infected with sPLA2-III-S or mock adenovirus (LacZ) and cultured with a suboptimal concentration of NGF (1 ng/ml), neurite extension from the explants was markedly accelerated in the presence of sPLA2-III-S (Figure 7A). On the other hand, DRG explants infected with adenovirus expressing sPLA2-III siRNA had fewer neurites with branches than those infected with control adenovirus in culture containing an optimal concentration of NGF (10 ng/ml) (Figure 7B), suggesting that endogenous sPLA2-III significantly contributes to the axonal branching of DRG neurons in this experimental system.
sPLA2-III induces neurite outgrowth in mouse DRG neurons
Previous studies have demonstrated that sPLA2-X is uniquely expressed in peripheral neurons and has the ability to promote neuronal outgrowth and survival through the production of LPC [19,25,26]. Here, we have expanded on those studies and found that sPLA2-III, an atypical sPLA2 isoenzyme that is structurally related to bee venom PLA2 (a prototypic group III enzyme), is also expressed in neuronal cells, exhibits neuritogenic and neurotropic actions and produces LPC from neuronal membranes. These findings delineate an unexplored, lysophospholipid-dependent, neurotrophin-like action of particular sPLA2 isoenzymes that are intrinsically expressed in the nervous system.
Immunohistochemistry of several human tissues revealed the localization of sPLA2-III in peripheral neuronal fibres (Figures 1F and 1G), where sPLA2-X (but not other sPLA2s) is also detected . Expression of sPLA2-III in peripheral neurons is further supported by detection of sPLA2-III protein and mRNA in mouse and human DRG neurons (Figures 1B–1D) and in rat PC12 cells undergoing NGF-induced neuronal differentiation (Figure 1E). As established by optimized RT–PCR, the expression of sPLA2-III in mouse DRG is >10-fold higher than that of sPLA2-X (Figure 1B). In human cerebellum, sPLA2-III is predominantly located in Purkinje cells and sPLA2-X in neuronal fibres, but not vice versa (Figure 1H), suggesting that the neuronal expression profiles of sPLA2-III and sPLA2-X are not entirely identical and therefore their neuronal functions in vivo may not be fully redundant.
However, as has been reported for sPLA2-X , we found that sPLA2-III is capable of supporting neuronal outgrowth (Figures 2–4 and 7) and survival (Figure 6). Under conditions where sPLA2-III-S exhibited neuritogenic (in the presence of suboptimal concentrations of NGF) and neurotropic (after NGF deprivation) effects, sPLA2-III-S-HQ (the catalytically inactive mutant) failed to do so (Figures 2B, 2C and 6D). In this context, the cellular content of LPC, a PLA2 reaction product with neuritogenic activity, was increased in PC12 cells transfected with sPLA2-III-S, but not in those transfected with sPLA2-III-S-HQ (Figure 4A). Indeed, sPLA2-III has a potent hydrolysing activty for the PC-rich outer leaflet of the plasma membrane, even though it is less efficient than sPLA2-X [28,29]. Bee venom PLA2, when added exogenously, also elicits neuritogenesis in PC12 cells in a catalytic-activity-dependent manner . Thus it appears that the neuritogenic action of sPLA2-III depends primarily on the lipolytic production of LPC. The possibility that LPA, a breakdown product of LPC through the lysophospholipase D/autotaxin-mediated mechanism, mediates the neuritogenic action of sPLA2-III (as well as sPLA2-X) is unlikely, since LPA fails to induce neuritogenesis [18,25] or facilitate neurite retraction [36a,36b].
Several possibilities have been proposed for the mechanism of the neuronal action of LPC. An increase in the intracellular Ca2+ level through LPC-directed activation of L-type Ca2+ channels leads to activation of protein kinase C and mitogen-activated protein kinases . Ca2+ influx through the voltage-gated channels also triggers a variety of cellular events, leading to neurite outgrowth via the concerted actions of various Ca2+-binding proteins  and cell adhesion molecules [38,39]. Although it has been reported that the neuritogenic action of LPC is mediated by the G-protein-coupled receptor G2A in PC12 cells , the true contribution of this receptor remains enigmatic, since its physiological ligand is now recognized to be protons and not LPC . Interestingly, LPC in combination with fatty acids induces enlargement of nerve terminals, closely mimicking the changes observed in sPLA2-treated neurons . In this scenario, LPC remains confined mainly to the external leaflet of the presynaptic membrane, whereas fatty acids have a high rate of transbilayer movement and will partition between the two membrane leaflets. In view of the fact that cell membrane expansion is achieved through fusion of transport organelles with the plasma membrane, such a configuration, with the inverted cone-shaped LPC in trans and the cone-shaped fatty acid in cis with respect to the membrane fusion site, may allow axonal membrane expansion. Moreover, ω-3 and ω-6 fatty acids stimulate neuronal cell membrane expansion by acting on syntaxin 3, a SNARE (soluble N-ethylmaleimide-sensitive factor-attachment protein receptor) protein that is a prerequisite for axonal growth . Thus LPC and polyunsaturated fatty acids, which are produced simultaneously by PLA2, may act in synergy to facilitate the addition of transport vesicle material to the plasma membrane.
Importantly, siRNA-directed knockdown of endogenous sPLA2-III markedly attenuated the NGF-induced neuronal outgrowth in PC12 cells (Figure 3) and in mouse DRG neurons (particularly affecting neurite branching in the latter case) (Figure 7B). Considering that the suppressive effect of sPLA2-X-directed siRNA on neuritogenesis is only modest but significant , the profound suppression of neuritogenesis by sPLA2-III-directed siRNA is noteworthy and implies that sPLA2-III may represent a critical component of the neuritogenic machinery. Notably, the level of LPC in sPLA2-III siRNA-treated PC12 cells was reduced by as little as 50% when compared with mock-treated control cells (Figure 4B), suggesting that endogenous sPLA2-III contributes to LPC generation in PC12 cells, that the residual LPC, which is likely to be produced by other PLA2s, is barely accessible to the neuritogenic pathway, and that the sPLA2-LPC axis leading to neuritogenesis may be spatiotemporally controlled, probably at the leading edges of the growth cones [20,25].
It was significant that NGF-induced neuronal outgrowth was markedly attenuated (∼50% reduction) by the catalytically inactive mutant, sPLA2-III-S-HQ (Figure 5). This seemingly dominant-negative action of the sPLA2-III-S-HQ mutant leads us to hypothesize that endogenous sPLA2-III binds to a particular cellular component which governs, at least in part, the neuritogenic activity of this enzyme. In this regard, the neuronal action of sPLA2-III may require two sequential steps: its binding to the receptor-like molecule and hydrolysis of membrane PC. This mode of action is reminiscent of some snake venom presynaptic PLA2 neurotoxins, which bind to nerve terminals via receptors whose nature may vary for different enzymes, gaining access to membrane phospholipids for hydrolysis . sPLA2-IIA and -V also demonstrate this action in binding to cellular heparan sulfate proteoglycans which, in turn, facilitate membrane hydrolysis [41,42]. In addition, the M-type sPLA2 receptor binds to a subset of snake venom and mammalian sPLA2s with high affinity and can transduce signals independently of enzyme activity . However, neither heparan sulfate proteoglycans nor M-type receptor bind to group III sPLA2s, such as bee venom PLA2 and human sPLA2-III, with high affinity . Since bee venom PLA2 binds to the N-type sPLA2 receptor (which has not yet been molecularly identified) , it is possible that endogenous sPLA2-III interacts with the N-type receptor or related molecules that transmit particular signals independently of, or in concert with, enzymatic activity (i.e. LPC production), even though the contribution of the N-type receptor to the neuritogenic action of bee venom PLA2 has been reported to be negligible in PC12 cells . Nevertheless, identification of such a putative sPLA2-III-binding component in future studies will shed further light on the neuronal function of sPLA2-III.
In summary, we have shown that sPLA2-III is expressed in neuronal cells in the peripheral and central nervous systems and exhibits a neuritogenic/neurotropic effect in cell culture. sPLA2s have also been previously implicated in neurotoxicity [13–17,45]. For instance, sPLA2-IIA is induced in ischaemic rat brain  and can promote apoptosis of primary rat cortical neurons . sPLA2-IB also promotes neuronal cell death, an event that depends on its binding to the M-type sPLA2 receptor . Bee venom PLA2, in synergy with glutamate, exerts a neurotoxic effect through the N-type sPLA2 receptor [13,14]. In addition, sPLA2-IIA enhances neurotransmitter release from PC12 cells and cultured rat hippocampal neurons [12,46]. To circumvent these complexities, studies using sPLA2-isoenzyme-specific inhibitors or gene-manipulated mice may help to clarify the physiological relevance of these observations.
We thank Dr G. Lambeau (CNRS-UPR 411, Sophia Antipolis, France) for providing human sPLA2-III cDNA, and Dr Y. Takanezawa and Dr H. Arai (University of Tokyo, Tokyo, Japan) for ESI–MS analysis. We thank Dr A. Maehama (National Institute of Infectious Diseases, Tokyo, Japan) for his helpful advice on immunoblotting for p-ERK and p-Akt. This work was supported by grants-in-aid for scientific research from the Ministry of Education, Science, Culture, Sports and Technology of Japan. M. M. was supported by PRESTO (Precursory Research for Embryonic Science and Technology) from the Japan Science and Technology Agency, the Tokyo Biochemical Research Foundation, the Mitsubishi Pharma Research Foundation, and NOVARTIS Foundation for the Promotion of Science. S. M. is a Research Fellow supported by the Japan Society for the Promotion of Science.
dorsal root ganglia
embryonic day 12.5
- HEK-293A cells
human embryonic kidney 293A cells
nerve growth factor
small interfering RNA
group III sPLA2
S domain-only construct of human sPLA2-III
kinase-dead mutant of human sPLA2-III-S