Address correspondence and reprint requests to Gary A. Weisman, Department of Biochemistry, 540E Life Sciences Center, 1201 Rollins Road, University of Missouri, Columbia, MO 65211-7310, USA. E-mail: firstname.lastname@example.org
The pro-inflammatory cytokine interleukin-1β (IL-1β), whose levels are elevated in the brain in Alzheimer's and other neurodegenerative diseases, has been shown to have both detrimental and beneficial effects on disease progression. In this article, we demonstrate that incubation of mouse primary cortical neurons (mPCNs) with IL-1β increases the expression of the P2Y2 nucleotide receptor (P2Y2R) and that activation of the up-regulated receptor with UTP, a relatively selective agonist of the P2Y2R, increases neurite outgrowth. Consistent with the accepted role of cofilin in the regulation of neurite extension, results indicate that incubation of IL-1β-treated mPCNs with UTP increases the phosphorylation of cofilin, a response absent in PCNs isolated from P2Y2R−/− mice. Other findings indicate that function-blocking anti-αvβ3/5 integrin antibodies prevent UTP-induced cofilin activation in IL-1β-treated mPCNs, suggesting that established P2Y2R/αvβ3/5 interactions that promote G12-dependent Rho activation lead to cofilin phosphorylation involved in neurite extension. Cofilin phosphorylation induced by UTP in IL-1β-treated mPCNs is also decreased by inhibitors of Ca2+/calmodulin-dependent protein kinase II (CaMKII), suggesting a role for P2Y2R-mediated and Gq-dependent calcium mobilization in neurite outgrowth. Taken together, these studies indicate that up-regulation of P2Y2Rs in mPCNs under pro-inflammatory conditions can promote cofilin-dependent neurite outgrowth, a neuroprotective response that may be a novel pharmacological target in the treatment of neurodegenerative diseases.
Neurodegenerative diseases are a serious cause of mortality in the United States with more than 5 million people currently afflicted (Duncan 2011; Thies and Bleiler 2011). Alzheimer's disease (AD) is the most prevalent of these conditions, and it is predicted that AD will affect 80 million people worldwide within 30 years (Blennow et al. 2006; Thies and Bleiler 2011). There are currently no effective treatments to prevent the onset or delay the progression of neurological deficits that degrade the quality of life of AD patients for many years prior to death (Thies and Bleiler 2011). It is now widely accepted that chronic inflammation plays a role in the progression of neurological changes observed in the AD brain, including neuronal loss and degeneration of neurological functions (Zilka et al. 2006; Lee et al. 2010; Obulesu et al. 2011; Wyss-Coray and Rogers 2012). Nevertheless, the initiating factors in AD remain obscure and whether neuroinflammation is primarily a neurodegenerative or a neuroprotective response in AD is an area of intense investigation (Zilka et al. 2006; Lee et al. 2010; Broussard et al. 2012).
Chronic inflammation in the central nervous system (CNS) is a conspicuous feature of many neurodegenerative diseases, including AD, Parkinson's disease and multiple sclerosis (Akiyama et al. 2000; Rothwell and Luheshi 2000; Broussard et al. 2012). A key cytokine associated with the neuroinflammatory phenotype is IL-1β, a pro-inflammatory cytokine produced by microglial cells and macrophages that regulates the production of other pro-inflammatory cytokines (e.g., TNF-α, IL-6, and interferons) and chemokines (e.g., CXCL1 and CXCL2) (Rothwell and Luheshi 2000; Shaftel et al. 2008). Although studies have investigated the neurodegenerative roles of IL-1β in AD progression (Rothwell and Luheshi 2000; Shaftel et al. 2008), observations in a mouse model of AD indicate that over-expression of IL-1β in the hippocampus can promote phagocyte recruitment and the clearance of β-amyloid plaques (Shaftel et al. 2007b). This suggests that IL-1β can also serve a neuroprotective role in the CNS that requires further investigation.
Recently, we demonstrated that the P2Y2 nucleotide receptor (P2Y2R), a G protein-coupled receptor that is activated equally well by ATP and UTP, is up-regulated in rat primary cortical neurons in response to IL-1β (Kong et al. 2009). Subsequent activation of the P2Y2R by extracellular nucleotides promotes the non-amyloidogenic processing of amyloid precursor protein (APP) (Camden et al. 2005; Kong et al. 2009). In mouse primary microglial cells, the P2Y2R is up-regulated by the neurotoxic β-amyloid (Aβ1-42) peptide associated with AD pathogenesis, whereupon activation of the microglial P2Y2R enhances Aβ phagocytosis and degradation (Kim et al. 2012), suggesting that P2Y2R up-regulation and P2Y2R-mediated non-amyloidogenic APP processing are neuroprotective responses that prevent excessive neurotoxic Aβ1-42 accumulation. Other studies have found that activation of ionotropic P2X7 receptors in microglial cells by extracellular ATP, a pathway that induces cell apoptosis, increases both IL-1β and ATP release from microglia (Takenouchi et al. 2009, 2011), thereby providing the agonists for both P2Y2R up-regulation and activation. Other potential neuroprotective responses to P2Y2R activation include the induction of intracellular calcium waves (Halassa et al. 2009), the up-regulation of anti-apoptotic protein expression in astrocytes (Chorna et al. 2004) and the enhancement of neuronal differentiation and survival (Arthur et al. 2005, 2006a, b; Pooler et al. 2005). Thus, P2Y2Rs in neurons, microglial cells, and astrocytes likely coordinately regulate neuroprotective responses to elevated levels of extracellular nucleotides that occur under pro-inflammatory, pro-apoptotic, and necrotic conditions (Peterson et al. 2010; Weisman et al. 2012a, b) and may prevent or delay neurodegeneration. Therefore, P2Y2Rs represent promising pharmacological targets in the treatment of AD and other diseases of the CNS.
This study was undertaken to further evaluate the role of P2Y2Rs in mouse primary cortical neurons (mPCNs), in particular the mechanism underlying the effect of extracellular nucleotides on neurite extension, a neuroprotective pathway that has not been characterized. It has been established that neurite outgrowth in response to activation of other G protein-coupled receptors requires the sequential activation of Rho, ROCK, LIMK, and cofilin (Meng et al. 2002; Bamburg et al. 2010; Bernstein and Bamburg 2010). Furthermore, we have previously demonstrated that activation of the P2Y2R can increase Rho and ROCK activities because of the presence of an Arg-Gly-Asp (RGD) sequence in the first extracellular loop of the P2Y2R that promotes its direct binding to αvβ3/5 integrins, an interaction required for extracellular nucleotides to activate heterotrimeric G12 protein and subsequently the small G protein Rho (Erb et al. 2001; Bagchi et al. 2005; Liao et al. 2007). In this study, we utilized mPCNs to demonstrate that up-regulation of the P2Y2R because of the pro-inflammatory cytokine IL-1β followed by P2Y2R activation with UTP increases both cofilin phosphorylation and neurite outgrowth. Furthermore, UTP-induced cofilin phosphorylation and neurite outgrowth was found to be absent in mPCNs from P2Y2R−/− mice and occurred via a pathway involving αvβ3/5 and CaMKII. These results strongly suggest that by virtue of P2Y2R interactions with αvβ3/5 integrins, nucleotides can activate Rho-dependent cofilin phosphorylation to regulate cytoskeletal rearrangements required for neurite extension and stabilization, which is critical for neuronal survival.
Fetal bovine serum was obtained from Hyclone (Logan, UT, USA). High glucose Dulbecco's modified Eagle's medium (HG-DMEM), Neurobasal medium, penicillin (100 units/mL), streptomycin (100 units/mL), and B27-AO were obtained from Gibco-BRL (Carlsbad, CA, USA). The Dual Color Protein Standards and nitrocellulose membranes (0.45 μm) were obtained from Bio-Rad (Hercules, CA, USA). LumiGLO chemiluminescent substrates were obtained from New England Biolabs (Beverly, MA, USA). The RNeasy Plus Mini Kit was obtained from Qiagen (Chatsworth, CA, USA). The First Strand cDNA Synthesis Kit was obtained from Roche (Indianapolis, IN, USA). Real-time PCR was performed on an Applied Biosystems 7500 Real-Time PCR machine with TaqMan Gene Expression assay probes (P2Y2R, NM_017255.1 and GAPDH, NM_017008.3) and TaqMan Universal PCR Master Mix (2X) from Applied Biosystems (Foster City, CA, USA). Function-blocking studies with the αV integrin utilized anti-integrin αVβ3 antibody (ab78289) produced by AbCam (Cambridge, MA, USA). All other reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA), unless stated otherwise.
C57BL/6 (wild type), C57BL/6-Tg(UBC-GFP)30Scha/J (GFP-wild type), and P2Y2R−/− mice on a C57BL/6 background were purchased from Jackson Laboratories (Bar Harbor, ME, USA) and bred at the Christopher S. Bond Life Sciences Center Animal Facility of the University of Missouri, Columbia, MO, USA. P2Y2R−/− mice were bred to GFP-wild type mice until a P2Y2R−/− mouse was established with the ubiquitous expression of GFP (GFP-P2Y2R−/−). Animals were housed in vented cages with 12 h light/dark cycles and received food and water ad libitum. All animals were handled using protocols approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Missouri (protocol no. 6728).
Primary cell culture of cortical neurons
Preparation of 95% pure mouse primary cortical neurons (mPCNs) was carried out as previously described with minor modifications (Kong et al. 2009). Briefly, cerebral cortices from 17-day-old embryos of wild type, GFP-wild type, P2Y2R−/−, or GFP- P2Y2R−/− mice were removed and the meninges discarded. The tissue was mechanically dissociated in HG-DMEM comprised of 10% (v/v) fetal bovine serum (FBS), 2 mM glutamine, 100 IU/mL penicillin, 100 mg/mL streptomycin, and 7.5 mg/mL fungizone. The tissue clumps were dispersed with a 10 mL pipette and suspended in 6 mL of 0.25% (w/v) trypsin at 37°C for 30 min. Then, 2 mL of heat-inactivated horse serum were added to neutralize trypsin activity and the suspension was triturated with a fire-polished Pasteur pipette. The cell suspension was centrifuged at 900 g for 2 min and the cell pellet was suspended in HG-DMEM. The resulting cell suspension was filtered through a sterilized 75 μm cell strainer (Becton Dickinson, Franklin Lakes, NJ, USA) and the cells were seeded at 500 cells/mm2 in plastic culture plates pre-coated with poly-d-lysine or poly-l-lysine (0.1 μg/mL). After 16–18 h and every 3 days thereafter, half the medium was replaced with B27-AO Neurobasal medium (2 mM glutamine, 100 IU/mL penicillin, 100 mg/mL streptomycin, 7.5 mg/mL fungizone, 10 mL of B27-AO and Neurobasal medium to 500 mL). After 6 days in vitro (DIV6) at 37°C in a humidified atmosphere of 95% air and 5% CO2, neurons were pre-treated for 24 h at 37°C in HG-DMEM containing 5% (v/v) FBS with or without IL-1β at the indicated concentration and used for experiments on DIV7.
Real-time and reverse transcription-PCR analysis of P2Y2R mRNA expression
Total RNA was isolated from mPCNs using the RNeasy Plus Mini Kit (Qiagen). cDNA was synthesized from 500 ng of purified RNA using the First Strand cDNA Synthesis Kit for RT-PCR (AMV; Roche). Ten percent of the synthesized cDNA was used as a template in 50 μL real-time PCR reactions. For TaqMan quantitative real-time PCR analysis, the probes were labeled at the 5′ end with 6-carboxy-fluorescein phosphoramidite (FAM) (for P2Y2R) or VIC® (for GAPDH; stable endogenous control; Applied Biosystems, Foster City, CA, USA) and at the 3′ end with minor groove binder (MGB) dye as the quencher. The samples were run in quadruplicate for the P2Y2R target and the endogenous GAPDH control. The relative levels of P2Y2R and GAPDH in each sample were determined and expressed as a ratio of P2Y2R to GAPDH (normalized to 1) using Applied Biosystems software.
As described previously, mPCNs were cultured on six-well plates and grown for 6 days in vitro (Kong et al. 2009). Cells were serum-starved and incubated in 1 mL of B27-AO Neurobasal medium with or without indicated compounds for specified time periods, as described in the figure legends. Samples were lysed in an equal volume of 2X Laemmli buffer (20 mM sodium phosphate, pH 7.0, 20% (v/v) glycerol, 4% (w/v) sodium dodecyl sulfate (SDS), 0.01% (w/v) bromophenol blue, and 100 mM DTT) and analyzed by western blot analysis. Total cofilin was used as a loading control. Cell lysates were heated for 5 min at 95°C, subjected to 7.5% (w/v) SDS-polyacrylamide gel electrophoresis, and transferred to nitrocellulose membranes for protein immunoblotting. After overnight blocking at 4°C with 5% (w/v) fat-free milk in TBS-T [10 mM Tris-HCl, pH 7.4, 120 mM NaCl and 0.1% (v/v) Tween-20], membranes were incubated with either 1 : 1000 dilution of rabbit anti-cofilin (#3312) or rabbit anti-phospho-cofilin (Ser3; 77G2; #3313) antibody (Cell Signaling, Danvers, MA, USA) overnight at 4°C followed by incubation with horseradish peroxidase-conjugated anti-rabbit IgG antibody [1 : 1000 dilution in TBS-T containing 5% (w/v) fat-free milk] for 1 h at 25°C. Protein immunoreactivity was visualized on autoradiographic film using chemiluminescence. The protein bands detected on X-ray film were quantified using a computer-driven scanner and Quantity One software (Bio-Rad, Hercules, CA, USA). The level of phospho-cofilin in each sample was expressed as a percentage of total cofilin.
Primary cortical neurons from GFP-wild type or GFP-P2Y2R−/− mice were pre-treated with IL-1β (100 ng/mL) or vehicle for 24 h in serum-free media. Cells were imaged in real time on a Zeiss LSM 510 META NLO (Carl Zeiss Microscopy, LLC Thornwood, NY, USA) inverted microscope equipped with an incubation chamber maintained at 37°C with 95% air and 5% CO2. UTP (100 μM) or vehicle was added and cells were imaged at 0, 30, 60, and 120 min. Neurite outgrowth was quantified at these time points by measuring the percent increase in neuron perimeter, using four to eight images per condition from three to six independent experiments. Perimeters were calculated using image analysis software (Adobe Photoshop CS4, Adobe Systems Inc., San Jose, CA, USA or ImageJ v. 1.46r, Rasband, W.S., ImageJ, U.S. National Institutes of Health, Bethesda, MD, USA). Perimeter distances in pixels or microns were used to calculate percent increases in the cell perimeter, as compared to the initial perimeter. Videos were captured on a Nikon Eclipse Ti inverted microscope (Nikon Inc., Melville, NY, USA) with an incubation chamber maintained at 37°C with 95% air and 5% CO2.
Results are expressed as the means ± SEM of data obtained from at least three experiments. Statistical analysis of data was performed using Graph Pad Prism version 5.0 (GraphPad Software Inc., La Jolla, CA, USA). Statistical significance was determined by a one tailed Student's t-test between groups. Differences were considered statistically significant when p <0.05.
IL-1β increases the expression of P2Y2R mRNA in mPCNs
Up-regulation of P2Y2R mRNA expression has been observed under a variety of pathophysiolgical conditions associated with inflammation and/or tissue damage (Turner et al. 1997; Seye et al. 2002; Schrader et al. 2005; Kong et al. 2009). P2Y2R expression is up-regulated by NF-κB binding to the P2Y2R promoter (Degagne et al. 2009), consistent with the established role of NF-κB in the induction of pro-inflammatory gene expression (Wullaert et al. 2011). Similar to our previous studies with rat primary neurons (Kong et al. 2009), we determined that P2Y2R mRNA expression increased in mPCNs treated for 24 h with IL-1β, and a maximal response was obtained at 100 ng/mL IL-1β (Fig. 1). Thus, subsequent experiments use mPCNs pre-treated with 100 ng/mL IL-1β for 24 h to up-regulate P2Y2R expression.
UTP increases neurite outgrowth in mPCNs
Using real-time microscopy, we determined that P2Y2R activation in IL-1β-pre-treated primary PCNs isolated from GFP-wild type mice caused extensive neurite outgrowth after a 120 min incubation with 100 μM UTP, as compared to GFP-wild type mPCNs treated with UTP for 0–120 min without IL-1β pre-treatment (Fig. 2a and c). Addition of the P2Y2R agonist UTP to mPCNs from GFP-P2Y2R−/− mice with or without IL-1β pre-treatment did not cause significant neurite outgrowth, as compared to GFP-wild type mice (Fig. 2b and c). In addition, we utilized live imaging of mPCNs isolated from GFP-wild type mice to generate time-lapse videos of neurite outgrowth induced by 100 μM UTP alone (Video S1) or 100 μM UTP following IL-1β pre-treatment (Video S2). Representative videos provide further evidence of increased UTP-induced neurite outgrowth following IL-1β pre-treatment, as compared to UTP treatment alone. These data are consistent with the conclusion that P2Y2R up-regulation induced by IL-1β is required to manifest the effect of UTP on neurite outgrowth in mPCNs.
P2Y2Rs mediate cofilin phosphorylation in IL-1β-pre-treated mPCNs
The regulation of neurite outgrowth and stabilization is a dynamic process that is necessary for the induction of long-term potentiation and requires a number of molecules that bind, sever, polymerize, or depolymerize actin filaments (Malenka and Nicoll 1999; Cingolani and Goda 2008; Okamoto et al. 2009). Neurite outgrowth is dependent on the phosphorylation/dephosphorylation of the actin-depolymerizing factor cofilin (Meberg et al. 1998; Meberg and Bamburg 2000; Gungabissoon and Bamburg 2003; Bamburg et al. 2010), a 19 kDa protein that regulates both G- and F-actin polymerization during cytoskeletal remodeling (Maciver and Hussey 2002). Therefore, we determined whether P2Y2R activation by UTP modifies the phosphorylation of cofilin required for neurite extension in IL-1β-pre-treated wild type mPCNs. Western blot analysis using an anti-phospho-cofilin-specific antibody from extracts of mPCNs pre-treated for 24 h with IL-1β (100 ng/mL) indicates that UTP causes significant time-dependent enhancement of cofilin phosphorylation, as compared to mPCNs not pre-treated with IL-1β (Fig. 3a and b). Cofilin phosphorylation was maximal within 10 min of UTP treatment and was sustained for up to 30 min. To determine unequivocally whether UTP induces cofilin phosphorylation via P2Y2R activation, mPCNs isolated from P2Y2R−/− mice were pre-treated with 100 ng/ml of IL-1β for 24 h, then stimulated with 100 μM UTP for up to 30 min, and levels of cofilin phosphorylation were compared to similarly treated mPCNs from wild type mice. In contrast to wild type mPCNs, the level of UTP-induced cofilin phosphorylation in IL-1β-pre-treated PCNs from P2Y2R−/− mice did not significantly increase over the basal level (Fig. 4). Taken together, these data indicate that P2Y2R expression and activation in mPCNs under pro-inflammatory conditions enhances cofilin phosphorylation that inactivates actin depolymerization, a pathway required for neurite outgrowth (Van Troys et al. 2008).
The Gq-coupled P2Y2R is unique among GPCRs in that it contains a RGD-sequence in its first extracellular loop that interacts directly with αvβ3/5 integrins and enables UTP to stimulate integrin-dependent activation of heterotrimeric Go and G12 proteins (Erb et al. 2001; Bagchi et al. 2005; Wang et al. 2005; Liao et al. 2007), which, respectively, activate Rac and Rho, monomeric GTPases known to regulate cytoskeletal rearrangements (Symons and Settleman 2000). It is well established that cofilin phosphorylation is mediated by Rho and Rac (Gungabissoon and Bamburg 2003; Van Troys et al. 2008); therefore, we investigated whether P2Y2R activation by UTP requires αvβ3/5 integrin to increase cofilin phosphorylation in mPCNs. P2Y2R-mediated activation of the αvβ3/5 integrin signaling pathway in IL-1β–pre-treated wild type mPCNs was inhibited, as in our previous studies (Liao et al. 2007), with function-blocking anti-αv integrin antibody. Western blot analysis of extracts from IL-1β–pre-treated mPCNs indicates that UTP-induced cofilin phosphorylation was reduced to near basal levels by anti-αv antibody (Fig. 5a), suggesting that interactions between the P2Y2R and αv integrin are required for UTP to increase phospho-cofilin levels. UTP-induced cofilin phosphorylation also was significantly attenuated by Y-27632 (20 μM) (Fig. 5a), an inhibitor of ROCK, the downstream target of Rho that regulates LIMK and cofilin phosphorylation (Meng et al. 2002; Bamburg et al. 2010). The function-blocking anti-αv antibody or Y-27632 alone had no significant effect on the phospho-cofilin/cofilin level in the absence of UTP (data not shown). These results strongly suggest a role for αvβ3/5 integrin-dependent Rho activation in P2Y2R-mediated increases in cofilin phosphorylation that enhance neurite outgrowth in IL-1β-pre-treated mPCNs.
CaMKII is a well-documented modulator of actin dynamics and dendritic spine growth (Malenka and Nicoll 1999; Matsuzaki et al. 2004; Takemura et al. 2009; Pi et al. 2010). In vitro studies have shown that active CaMKII induces neurite outgrowth (Jourdain et al. 2003; Pi et al. 2010) and the phosphorylation of cofilin (Takemura et al. 2009). The activation of CaMKII by calmodulin is dependent on rises in the intracellular [Ca2+] (Fink and Meyer 2002) and previous studies in several laboratories, including ours, have shown that P2Y2R activation increases the intracellular free calcium concentration, [Ca2+]i, through Gq-dependent activation of PLC leading to the generation of IP3 and the subsequent release of Ca2+ from intracellular stores (Lustig et al. 1992; Viana et al. 1998; Weisman et al. 1999). Once activated, CaMKII phosphorylates multiple downstream targets, including kalirin-7 (Xie et al. 2007) and CaMKK (Saneyoshi et al. 2008), ultimately leading to the activation of LIMK (Cingolani and Goda 2008; Okamoto et al. 2009) and the phosphorylation of cofilin (Endo et al. 2007). To investigate whether CaMKII regulates UTP-induced cofilin phosphorylation, IL-1β-pre-treated wild type mPCNs were treated with selective inhibitors of CaMKII (KN-93, 3 μM) or CaMKK (STO-609, 5 μM) for 30 min prior to UTP (100 μM) treatment. Cell lysates were subjected to immunoblot analysis to evaluate cofilin and phospho-cofilin levels. We found that KN-93 or STO-609 significantly attenuates cofilin phosphorylation induced by UTP in IL-1β-pre-treated mPCNs (Fig. 5b). KN-93 or STO-609 alone had no significant effect on the phospho-cofilin/cofilin level in the absence of UTP (data not shown). These results indicate that increases in CaMKII/CaMKK activities regulate P2Y2R-mediated cofilin phosphorylation, along with the αV integrin/Rho/ROCK pathway. The data presented in this study suggest a signaling pathway whereby the P2Y2R mediates cofilin-dependent neurite outgrowth in mPCNs under pro-inflammatory conditions (Fig. 6).
In this study, we demonstrate that IL-1β up-regulates the P2Y2R in mPCNs, where-upon activation of the P2Y2R promotes neurite extension through inactivation (phosphorylation) of the actin-depolymerizing protein cofilin, a known regulator of neurite extension (Meberg et al. 1998; Meberg and Bamburg 2000; Aizawa et al. 2001; Endo et al. 2003, 2007; Figge et al. 2012). Using an antibody that blocks αv integrin function and selective inhibitors, we also provide evidence that the P2Y2R mediates cofilin inactivation required for neurite outgrowth in mPCNs by a mechanism involving αv integrin, Rho/ROCK/LIMK, and CaMKII/CaMKK/LIMK. The P2Y2R also is known to modulate ADAM activity in neurons (Kong et al. 2009); however, we determined that TAPI-2, a matrix metalloprotease inhibitor, had no effect on UTP-induced cofilin phosphorylation in IL-1β-treated mPCNs (data not shown). Based on these findings, a model of the mechanisms underlying P2Y2R-mediated neurite extension is provided in Fig. 6.
IL-1β is thought to have both detrimental and beneficial effects on the pathogenesis of AD (Mrak and Griffin 2005; Shaftel et al. 2007b). Studies in humans have shown that the IL-1β gene (Colangelo et al. 2002) and IL-1β protein expression (Griffin et al. 1989) are elevated in AD brain tissue and experimental models demonstrate that IL-1β is involved in neuronal injury, degeneration, and loss (Rothwell and Luheshi 2000). Other studies, however, demonstrate that IL-1β promotes remyelination of neurons in the mouse CNS (Mason et al. 2001) and that chronic IL-1β expression causes an increase in blood–brain barrier permeability to leukocytes in mouse brain without causing neurodegeneration (Shaftel et al. 2007a). We show here that a 24 h pre-treatment of mPCNs with IL-1β causes an ~fourfold increase in P2Y2R mRNA expression (Fig. 1), suggesting that signaling cues from the P2Y2R in neurons are intensified under inflammatory conditions. Since P2Y2R activation in IL-1β-treated wild type mPCNs promotes neurite extension, in contrast to P2Y2R−/− mPCNs (Fig. 2), we postulate that P2Y2R up-regulation in the CNS may delay the progression of neurodegeneration that occurs with chronic inflammation. P2Y2R expression levels were found to be lower in post-mortem human AD brain tissue compared to normal tissue (Lai et al. 2008). This may be because of brain atrophy and the degeneration of P2Y2R-expressing cells in the CNS or the down-regulation of the P2Y2R in specific subsets of cells in the CNS during end stage AD, possibilities that we are currently evaluating in a mouse model of AD. A better understanding of the mechanisms of neurite extension may provide insights into the role of P2Y2Rs in the regulation of the transition from acute to chronic inflammation and neurodegeneration and the apparent neuroprotective role of P2Y2Rs during inflammation.
P2Y2R gene expression induced by IL-1β in mPCNs (Fig. 1) and rat cortical neurons (Kong et al. 2009) is inhibited by pre-treatment with Bay 11-7082, an inhibitor of IκB-α phosphorylation (Pierce et al. 1997). This suggests that the IκB/NF-κB signaling pathway, known to be enhanced by pro-inflammatory cytokines (DiDonato et al. 1997; Hacker and Karin 2006), regulates P2Y2R transcription in neurons. Consistent with this hypothesis, the P2Y2R promoter contains a NF-κB binding domain that regulates increased P2Y2R transcription in response to inflammatory agents (Degagne et al. 2009). In vascular smooth muscle cells, IL-1β was shown to up-regulate P2Y2R gene expression by a mechanism involving cyclooxygenase and protein kinase C (PKC), although NF-κB activation was not examined (Hou et al. 2000).
We found that activation of the P2Y2R in wild type mPCNs increases neurite extension, as indicated by an increase in the neurite perimeter of UTP-treated cells (Fig. 2), which is critical for establishing synaptic connections and for neuronal survival (Cline and Haas 2008). Neurite outgrowth in cultured neurons is considered an indication of neuroregenerative potential (Mitchell et al. 2007). Therefore, development of strategies to activate the P2Y2R in vivo may help retard neurodegeneration that occurs in AD and other neuroinflammatory diseases by promoting neurite extension and neuronal survival.
Neurite extension requires molecular signals that promote remodeling of the actin cytoskeleton within the growing neurite. The actin-binding protein cofilin regulates actin dynamics in essentially every type of eukaryotic cell (Maciver and Hussey 2002; Van Troys et al. 2008), and numerous studies indicate that cofilin is important for neurite extension (Meberg et al. 1998; Meberg and Bamburg 2000; Aizawa et al. 2001; Endo et al. 2003, 2007; Figge et al. 2012). Recent studies using Aplysia kurodai neurons found that microinjection of dephosphorylated cofilin led to rod formation, synapse loss (e.g., a decrease in the number of pre-synaptic varicosities) and, distal to the rod, impairment of synaptic plasticity measured by electrophysiological methods (Jang et al. 2005). In addition, phospho-cofilin administration impaired basal synaptic transmission, long-term potentiation (LTP), and the structure and dynamics of post-synaptic dendritic spines (Jang et al. 2005). Other studies indicate that blockade of calcineurin with FK506 or expression of a phosphomimetic mutant of cofilin (cof-S3D) prevented Aβ-induced spine loss in a hippocampal slice model of AD (Shankar et al. 2007). These findings are consistent with previous reports that cofilin signaling is perturbed in AD brain tissue and in neurons treated with synthetic Aβ (Maloney et al. 2005; Zhou et al. 2006). Thus, the ability of the P2Y2R to regulate cofilin phosphorylation in neurons (Figs. 3 and 4) likely has relevance to neurodegenerative disorders, such as AD, where cell damage or apoptosis would be anticipated to increase levels of extracellular nucleotides that activate the P2Y2R.
The activity of cofilin in the disassembly of actin filaments (F-actin) is regulated by several mechanisms, including phosphorylation of cofilin at serine3 (Ser3) (van Rheenen et al. 2009). Ser3 phosphorylation of cofilin prevents F-actin binding and severing of actin filaments by cofilin (van Rheenen et al. 2009), and thus it is thought that cofilin phosphorylation/dephosphorylation at Ser3 acts as a switch between actin assembly and disassembly (Huang et al. 2006). Several kinases and phosphatases have been identified that phosphorylate and dephosphorylate cofilin at Ser3, respectively; these include the actin-binding LIM kinases (LIMK1, LIMK2) (Arber et al. 1998; Yang et al. 1998), testicular protein kinases (Toshima et al. 2001a, b), slingshot (SSH) phosphatases (Niwa et al. 2002), and chronophin phosphatase (Niwa et al. 2002; Gohla et al. 2005). Here, we show that activation of the P2Y2R in IL-1β-pre-treated mPCNs causes phosphorylation of cofilin at Ser3 (Figs. 3 and 4) and that the Rho/ROCK/LIMK pathway is involved in this process, since inhibition of ROCK, which is activated by RhoA and controls the activation of LIMK2 (Gungabissoon and Bamburg 2003; Bernstein and Bamburg 2010), prevents P2Y2R-mediated cofilin phosphorylation at Ser3 (Fig. 5a). We also show that P2Y2R-mediated cofilin phosphorylation requires the activities of CaMKII and CaMKK (Fig. 5b). Furthermore, CaMKII controls cofilin phosphorylation and regulates F-actin dynamics through pathways that mainly converge on the Rho family of monomeric GTPases, such as RhoA and Rac1 (Okamoto et al. 2009). The activities of these GTPases are controlled by guanine-nucleotide-exchange factors (GEFs) and GTPase activating proteins (GAPs) (Symons and Settleman 2000; Newey et al. 2005). Although we did not explore which GEFs are activated by the P2Y2R in mPCNs, a likely candidate is kalirin-7, whose activity is reported to be essential for spine enlargement (Penzes and Jones 2008). Phosphorylation of kalirin-7 by CaMKII increases its GEF activity and leads to cofilin phosphorylation through the Rac1/PAK1/LIMK1 pathway (Penzes and Jones 2008). Other GEF candidates that mediate cofilin phosphorylation in dendritic spines and are controlled by CaMKII include βPIX, which signals through the Rac1 pathway, and Lcf, which binds to the actin-binding protein spinophilin and signals through the RhoA/ROCK/LIMK2 pathway (Okamoto et al. 2009).
Ionomycin-induced cofilin phosphorylation and neurite outgrowth are also blocked by KN-93, an inhibitor of Ca2+/calmodulin-dependent protein kinases, and STO-609, an inhibitor of CaMKK (Takemura et al. 2009). Other investigators have previously shown that ROCK and PAK activate LIMK1 by phosphorylation at Thr-508, which is in the activation loop of the kinase domain in LIMK1 (Edwards et al. 1999; Maekawa et al. 1999; Ohashi et al. 2000). Various signaling pathways, including Rac/PAK1 and CaMKIV/CaMKK activate Thr-508 phosphorylation in LIMK1 (Edwards et al. 1999; Maekawa et al. 1999; Ohashi et al. 2000; Takemura et al. 2009), suggesting a similar target for the Rho/ROCK and CaMKII/CaMKK pathways activated by the P2Y2R. This also suggests that LIMK1 activation is a point of convergence that links multiple signaling pathways to the regulation of actin cytoskeletal reorganization in cells (Takemura et al. 2009). Further studies are needed to evaluate cross-talk between the CaMKII/CaMKK, RhoA/ROCK, and other signaling pathways in the regulation of P2Y2R-mediated and LIMK-dependent cofilin phosphorylation and neurite extension.
This study was supported by NIH grant AG018357. The authors declare no conflict of interest. We are grateful to the Molecular Cytology Core of the Bond Life Sciences Center at the University of Missouri for their assistance in acquiring images.