Herpes Simplex Virus type-1 infection induces synaptic dysfunction in cultured cortical neurons via GSK-3 activation and intraneuronal amyloid-β protein accumulation

Increasing evidence suggests that recurrent Herpes Simplex Virus type 1 (HSV-1) infection spreading to the CNS is a risk factor for Alzheimer’s Disease (AD) but the underlying mechanisms have not been fully elucidated yet. Here we demonstrate that in cultured mouse cortical neurons HSV-1 induced Ca2+-dependent activation of glycogen synthase kinase (GSK)-3. This event was critical for the HSV-1-dependent phosphorylation of amyloid precursor protein (APP) at Thr668 and the following intraneuronal accumulation of amyloid-β protein (Aβ). HSV-1-infected neurons also exhibited: i) significantly reduced expression of the presynaptic proteins synapsin-1 and synaptophysin; ii) depressed synaptic transmission. These effects depended on GSK-3 activation and intraneuronal accumulation of Aβ. In fact, either the selective GSK-3 inhibitor, SB216763, or a specific antibody recognizing Aβ (4G8) significantly counteracted the effects induced by HSV-1 at the synaptic level. Moreover, in neurons derived from APP KO mice and infected with HSV-1 Aβ accumulation was not found and synaptic protein expression was only slightly reduced when compared to wild-type infected neurons. These data further support our contention that HSV-1 infections spreading to the CNS may contribute to AD phenotype.


Ca 2+ -mediated activation of GSK-3 induced by HSV-1 is critical for APP phosphorylation at
Thr668. We previously demonstrated that HSV-1 infection induces Ca 2+ -dependent phosphorylation of APP at Thr668 (pAPP) in rat cortical neurons in vitro 5 . Immunoreactivity for pAPP was found on the membrane and in the nuclei of infected cells, thus suggesting that the antibody raised against pAPP also recognized the C-terminus fragment of APP, AICD, that accumulates in the nucleus and is phosphorylated at the same threonine residue 6 .
Therefore, we further investigated the effects of HSV-1 on GSK-3 activation by studying its phosphorylation at Tyr279 and 216 (pGSK-3) that are well known activatory sites for α and β isoforms of this kinase, respectively. At 18 h p.i. pGSK-3 immunoreactivity was 1.25 ± 0.07 times greater than that of mock-infected neurons (n = 71 and 244, respectively; P = 1.3 × 10 −71 , Student's t test; Fig. 2a,b,d). This effect, that preceded the marked increase in APP phosphorylation we observed at 24 h p.i., was counteracted by SB216763 (10 μ M), thus confirming the specificity of pGSK-3 immunostaining (data not shown).
Given that many studies pointed out the importance of the GSK-3β isoform in mediating the action of this kinase, we performed Western blot experiments to determine if GSK-3β was activated following HSV-1 infection. Notably, we found a marked increase of pGSK-3β in HSV-1-infected neurons (Fig. 2g,h) whereas the total expression of GSK-3β was not significantly modified, as also demonstrated by immunofluorescence (total GSK-3 immunoreactivity was: 1.00 ± 0.06 and 1.09 ± 0.07 in mock and HSV-1-infected cells, n = 22 and 28 respectively; P = 0.07 Student's t test; Fig. 2e-h). Finally, to determine whether the observed effects depended on HSV-1 binding to the cell membrane, we repeated Western blot experiments in cells treated with UV-inactivated HSV-1 that retains the ability to bind the membrane and activate the signaling cascade leading to increased [Ca 2+ ] i but it does not replicate inside cells. No significant increases in pGSK-3β /GSK-3β ratio were induced by UV-inactivated HSV-1 (P = 0.10 vs. mock; Fig. 2g,h).
In agreement with our recent findings in rat cortical neurons 28 and other literature reports 29,30 , HSV-1 infection also slightly increased GSK phosphorylation at Ser9 (Suppl. Fig. 1). However, the effect of the GSK-3 inhibitor SB216763 on APP phosphorylation suggests that the "net" result of HSV-1-induced GSK-3 phosphorylation is the kinase activation.

GSK-3 activation mediates intraneuronal accumulation of Aβ.
We previously demonstrated that the culture medium of HSV-1-infected neurons contained small soluble Aβ oligomers (i.e., dimers and trimers), and lysates of infected neuronal cells subjected to Western blotting exhibited a band recognized by the 4G8 antibody (that specifically reacts with murine high molecular weight Aβ aggregates) at 35 kDa, likely corresponding to a Aβ nonamer 5,8 . Here we showed that mouse cortical neurons infected with HSV-1 exhibited an increased immunoreactivity for Aβ 40/42 with respect to mock-infected cells Combination of SB216763 and roscovitine prevented the effects of HSV-1 on pAPP immunoreactivity. (h) Western blot analysis of pAPP T668 , APP and GAPDH carried out on lysates of mock-infected cells, HSV-1-infected cells at 24 h p.i., and infected cells treated with the specific inhibitor of GSK-3 SB216763 for all the time p.i. (i) Densitometric analysis of three independent Western blot experiments. Scale bar: 20 μ m in panels (a-c) and 50 μ m in panels (d-f). *P < 0.05 vs. mock; **P < 0.001 vs. mock; #P < 0.05 vs. HSV-1; ##P < 0.001 vs. HSV-1. In this, and the following figures, statistical significance was assessed by Student's t test for comparison between mock and HSV-1, and by one-way ANOVA with Bonferroni post-hoc test for multiple comparisons. For experiments that included fewer than 10 observations (e.g. densitometric analysis of WB data), the Mann-Whitney (Wilcoxon) statistic was used.
Scientific RepoRts | 5:15444 | DOi: 10.1038/srep15444 ( Fig. 3a,b,d,e). We then specifically looked for intracellular accumulation of Aβ 42 in infected neurons, that was assessed by an antibody recognizing the C-terminus region of the protein (Fig. 3g,h). In agreement with our previous reports, the specificity of Aβ 42 C-t immunolabeling was assessed by adding β -and γ -secretase inhibitors to the culture medium during the time p.i. 5,8 . Under these experimental conditions Aβ 42 C-t immunoreactivity was significantly reduced (data not shown).
In agreement with previous reports 31 , addition of the GSK-3 inhibitor to the culture medium of mock-infected neurons significantly increased the expression of synapsin-1 (+ 36% of mock-infected neurons without SB216763; n = 13 microscopic fields; P = 7.3 × 10 −3 ; Student's t test), but it did not affect synaptophysin expression (n = 12 microscopic fields; P = 0.67; Student's t test). Finally, we determined if the intracellular Aβ accumulation that we observed following infection was involved in the GSK-3-dependent reduction of synaptic protein expression. To address this issue, we exposed neuronal cultures to 4G8 antibody (3.3 μ g/mL) during the post-infection period. It is known that extracellularly-applied 4G8 interacts with the extracellular site of APP, it is internalized and accumulated intracellularly 20 . Once inside neurons, 4G8 antibody may bind Aβ and counteract its effects. In our experimental conditions, extracellular 4G8 was indeed internalized and it recognized HSV-1-produced Aβ (Fig. 5a-c). We found that the expression of synapsin-1 and synaptophysin in HSV-1-infected neurons (24 h p.i.) exposed to 4G8 was significantly different from that of HSV-1-infected cells. In the presence of 4G8 the reduction of synapsin-1 and synaptophysin immunoreactivity passed from − 58% to − 33% and from -39% to -5%, respectively (P = 3.3 × 10 −3 and P = 3.2 × 10 −3 , respectively vs. HSV-1; Bonferroni post-hoc test; Fig. 4d,h,i). These findings were supported by results of Western blot analyses (Fig. 4j,k).
To further investigate the dependence of the HSV-1-induced synaptic effects on APPF production, we infected cortical neurons isolated from APP-KO mice. Given that these cells do not express APP, HSV-1 failed to produce intracellular accumulation of Aβ 42 (compare Fig. 5d with Fig. 3h) although APP-KO neurons exhibited a clear HSV-1-induced activation of GSK-3 (Fig. 5e,f). When the immunolabeling of synaptophysin and synapsin-1 was evaluated in APP-KO-infected neurons, we found slight but not significant differences respect to APP-KO-mock-infected cells (for synapsin-1: P = 0.12, Student's t test; n = 13 and 11 fields analyzed for mock and HSV-1-infected cells, respectively, Fig. 5h,i,g; for synaptophysin: P = 0.12, Student's t test; n = 9 and 14 fields analyzed for mock and HSV-1-infected cells, respectively, Fig. 5j,k,g), thus supporting our contention that HSV-1-induced Aβ accumulation is responsible for synaptic protein downregulation.

Synaptic transmission is affected by HSV-1 infection.
We then evaluated whether the molecular alterations induced by HSV-1 at the presynaptic level impaired synaptic transmission. To address this issue we studied the amplitude of evoked excitatory post-synaptic currents (EPSCs) and the amplitude and the frequency of spontaneous miniature EPSCs (mEPCS). When autaptic cultures of mouse cortical neurons 23,24 were infected with HSV-1 for 24 hours, EPSC amplitude was significantly reduced if compared to that of mock-infected neurons (− 36%; n = 10 and 19 for mock-and HSV-1-infected neurons, respectively; P = 1.7 × 10 −3 ; Student's t test; Fig. 6a,b). The same treatment also markedly reduced the frequency of mEPSC (− 49% vs. mock-infected neurons; n = 10 for mock and 18 for HSV-1-infected neurons; P = 6.0 × 10 −4 ; Student's t test; Fig. 6c,d), whereas their mean amplitude was only slightly reduced (− 15% vs. mock infected neurons, P = 0.15; Fig. 6d). The inhibitory effects of HSV-1 on both EPSC amplitude and mEPCS frequency were significantly reversed by the presence of 10 μ M SB216763 in the culture medium, and they depended on Aβ accumulation. Indeed, according to what we observed for presynaptic protein expression, 4G8 rescued synaptic transmission from the inhibitory effects of HSV-1. In particular, no significant differences were found between EPSC amplitudes recorded in mock-infected neurons and in neurons infected with HSV-1 and treated with either SB216763 or 4G8 for all the time p.i. (one way ANOVA: F 2,28 = 0.83, P = 0.44; Fig. 6a,b). Similar results were found for the mEPSCs frequency that was increased from 3.7 Hz in HSV-1-infected neurons to 6.7 Hz (n = 12) in the presence of the GSK-3 inhibitor and to 6.0 Hz (n = 10) in the presence of 4G8 antibody (P = 2.6 × 10 −3 and P = 3.6 × 10 −2 , respectively; Bonferroni post-hoc test; Fig. 6c,d).

Discussion
In this study we demonstrated that HSV-1 infection impairs synaptic function in cultured mouse cortical neurons through GSK-3 activation and intracellular accumulation of Aβ . These events were responsible for the HSV-1-induced molecular and functional alterations we observed at the synaptic level consisting of: i) reduced expression of the presynaptic proteins synapsin-1 and synaptophysin; ii) depressed synaptic transmission. Finally, we found that Aβ -and GSK-3-mediated inhibition of CREB is likely involved in the HSV-1-induced synaptic effects.
In a previous report we demonstrated that HSV-1 binding to neuronal membrane of rat cortical neurons in vitro induces membrane depolarization due to activation of persistent Na + currents and inhibition of leak K + currents that, in turn, triggers intracellular Ca 2+ signals (due to both Ca 2+ entry through Ca v 1 channels and Ca 2+ release from IP 3 receptors) leading to increased [Ca 2+ ] i 5 . Ca 2+ transients triggered by HSV-1 binding to the membrane were responsible for APP phosphorylation at Thr668 as well as for phosphorylation of the same amino acid in the C-terminus region of APP, named AID or AICD 7 . The former is believed to be a key event in APP processing and Aβ production 10,35 , whereas the latter enables AICD to interact with the Fe65 adapter protein 36 and enter the nucleus, where it affects gene transcription and induces neurodegeneration 6,37 . Very recently, we also demonstrated that the HSV-1-induced nuclear accumulation of ACID regulates the expression of the Aβ clearance/degrading enzyme neprilysin and GSK-3 28 . Here we found that HSV-1 infection induces an intense activation of GSK-3, as revealed by immunoreactivity for phospho-GSK-3α /β (at Tyr279/216). Western blot experiments also indicated that after viral infection GSK-3β phosphorylation at Tyr216 was significantly increased whereas the total amount of the protein did not significantly change, thus suggesting the specific activation of this kinase rather than upregulation of its expression following infection. HSV-1 infection also slightly increased Ser9 phosphorylation in mouse cortical neurons. In cultured human fibroblasts Naghavi and colleagues demonstrated that the viral Ser/Thr kinase Us3 inhibits GSK-3β activity by increasing its phosphorylation at Ser9 29 . However, in our experimental model, the "net" resulting activity of GSK-3 following 18-24 h of infection was increased. Indeed, the specific GSK-3 inhibitor, SB216763, reduced the HSV-1-induced increase in pAPP immunoreactivity by more than 50%. SB216763 has been reported to inhibit GSK-3 by specifically reducing pTyr279/216 rather than increasing pSer21/9 38  Our data also indicate that GSK-3 is involved in the HSV-1-induced production and intracellular accumulation of APPFs including Aβ 40 and 42. Indeed, neuronal immunostainings with either antibodies labeling all APPFs containing the Aβ sequence 17-24 (i.e., anti-Aβ 40/42 and 4G8) or the Aβ 42 C-terminus antibody, were significantly reduced in HSV-1-infected cells treated with SB216763 during the p.i. period, and similar results were also obtained with Western blot experiments. The involvement of GSK-3 in Aβ production is fully supported by previous literature reports showing that in vitro: i) GSK-3 interacts with presenilins 15 ; ii) GSK-3β promotes the production and intracellular accumulation of Aβ by enhancing pAPP 10 as well as β -secretase activity 40 . Conflicting results have been reported in vivo where neither GSK-3α nor GSK-3β were found to modulate APP processing 41 , and GSK-3 inhibition reduced Aβ levels independently of APP processing modulation by acting on Aβ degradation/sequestration 42 .
Our findings provide novel evidence that HSV-1 infection affects synaptic function. To our knowledge this is the first paper documenting that in HSV-1-infected neurons there is a GSK-3/Aβ -dependent decrease in the expression of presynaptic proteins that is associated with depressed synaptic transmission.
The role of intracellular Aβ accumulation in synaptic function dysregulation underlying cognitive impairment is well known [19][20][21][22][23][24] . Here we demonstrated that using two different approaches counteracting or preventing HSV-1-induced Aβ production and accumulation (i.e., intracellular loading of 4G8 antibody and APP KO mice, respectively), HSV-1 lost its synaptotoxicity. The residual effects of HSV-1 that we found at the synaptic level under these experimental conditions might depend on several factors. First, 4G8 might be unable to fully counteract the APPFs produced following HSV-1 infection. In fact, Tampellini et al. 20 showed that 24-h lasting 4G8 treatment of primary neurons from Tg2576 mice reduced intracellular Aβ levels by 30-40% only. We also found a slight reduction of presynaptic protein expression in HSV-1-infected APP null neurons. These findings allow us to speculate that other GSK-3-dependent and Aβ -independent mechanisms compromising synaptic function might play a minor role in HSV-1 effects. Increased GSK-3 expression has been shown to negatively modulate neurotransmitter release via phosphorylation of VGCCs 43 , and GSK-3 inhibition has been found to increase neurotransmission 44 and synapsin-1 expression 31 . GSK-3 has been also found to be involved in the modulation of other ion channels including voltage-gated Na + channels that may influence neuronal firing and synaptic transmission [45][46][47][48][49] . Moreover, GSK-3 up-regulation has been reported to impair synaptic plasticity at both functional and structural levels by reducing synapsin-1 clustering 50 and dendritic spine density 17 , whereas GSK-3 inhibition by SB216763 or TDZD-8 reverted these effects. Finally, another possibility is that activated GSK-3 causes synaptic dysfunctions via Tau protein hyperphosphorylation [51][52][53] . All these data are in agreement with our finding that HSV-1 acts upstream GSK-3 activation and Aβ production. Aβ -independent effects due to virus binding to neuronal membrane did not play a role in the observed alterations of synaptic proteins. In fact, when neurons were infected with UV-pretreated virus retaining the ability to bind cell membrane but not that of replicating inside cells, no changes in synaptic protein expression were observed.
Our data also suggest that CREB is critically involved in the HSV-1-activated molecular pathway leading to synaptic dysfunction. CREB is a transcription factor whose activity is modulated by GSK-3β via phosphorylation at Ser129 33,34,54 . It also plays a pivotal role in structural and functional synaptic plasticity 17,55,56 . Here we found that CREB is inactivated following HSV-1 infection and that this inhibition depends on GSK-3 activation and Aβ accumulation. These findings allow us to hypothesize that GSK-3-induced CREB inhibition is a critical determinant of HSV-1-induced synaptic dysfunction although further studies are needed to gain in-depth understanding of its role in the observed effects.
Collectively, findings reported here suggest that HSV-1 alters the expression of synaptic proteins via activation of GSK-3 and intraneuronal accumulation of Aβ . Although the production and accumulation of Aβ has been reported in the temporal brain cortex of mice 5 days after intranasal inoculation with HSV-1 57 , further studies on wild-type and AD transgenic mice are needed to demonstrate the cause-and-effect relationships among recurrent HSV-1 infections, Aβ overproduction and synaptic dysfunction underlying cognitive impairment in AD.

Material and Methods
Primary cortical neuron cultures. Primary cultures of cortical neurons were obtained from E18 C57/ bl6 mice and B6.129S7-App tm1 Dbo/J mice as previously described 5 . Briefly, after removing brains, dissected cortices were incubated for 10 min at 37 °C in PBS containing trypsin-ethylenediaminetetraacetic acid (0.025%/0.01% w/v; Biochrom AG, Berlin, Germany), and the tissue was mechanically dissociated at room temperature (RT) with a fire-polished Pasteur pipette. The cell suspension was harvested and centrifuged at 235 × g for 8 min. The pellet was suspended in 88.8% Minimum Essential Medium (MEM, Biochrom), 5% fetal bovine serum, 5% horse serum, 1% glutamine (2 mM), 1% penicillin-streptomycin-neomycin antibiotic mixture (PSN, Invitrogen, Carlsbad, CA), and glucose (25 mM). Cells were plated at a density of 10 5 cells on 20-mm coverslips (for immunocytochemical studies) and 10 6 cells/well on 35-mm six-well plates (for Western blot studies), precoated with poly-L-lysine (0.1 mg/ml; Sigma). Twenty-four hours later, the culture medium was replaced with a mixture of 96.5% Neurobasal medium (Invitrogen), 2% B-27 (Invitrogen), 0.5% glutamine (2 mM), and 1% PSN. After 72 h, this medium was replaced with a glutamine-free version of the same medium, and the cells were grown for 10 more days before carrying-out experiments.
HSV-1 infection. Infection was performed as previously described 5 . Fourteen days after plating, the culture medium was replaced with Neurobasal medium containing particles of HSV-1 strain F [5 × 10 8 plaque forming units (PFU)/mL] at a multiplicity of infection (MOI) of 5, and neuronal cultures were incubated for 1 h at 37 °C (adsorption period). The HSV-1-containing medium was then removed and after 2 washes in PBS, the cells were returned to the original medium and cultured for 18-24 h in the presence or not of specific inhibitors left in the culture medium for all the time p.i. and then fixed in paraformaldehyde (PFA, 4% in PBS, Sigma). The supernatants of infected cells were then subjected to standard plaque assay 58 to quantify virus production.
In all the experiments, the effects observed in infected cells were compared with those obtained in mock-infected controls that had been exposed to the vehicle alone. Additional control experiments were performed with virus that had been exposed on ice for 5 min to a 30 W, 254 nm, germicidal UV light placed at a distance of 15 cm (UV-inactivated HSV-1). UV-inactivated HSV-1 retains the capacity for plasma membrane binding and cell entry although it is incapable of replication 5 , as confirmed by standard plaque assay.
Images (512 × 512 pixels) were acquired at 63 × magnification with a confocal laser scanning system (TCS-SP2, Leica Microsystem, Wetzlar, Germany) and an oil-immersion objective (N.A. 1.4; physical pixel size: 233 nm). For some images, additional 2 × magnification was applied. Fluorescent dyes were excited with Ar/ArKr laser (for 488 nm) or HeNe lasers (for 543 and 633 nm). DAPI staining was imaged after two-photon excitation with an ultrafast, tunable, mode-locked titanium:sapphire laser (Chamaleon, Coherent Inc., Santa Clara, CA). All experiments were repeated at least 3 times, and at least 10 randomly chosen microscopic fields were analyzed for each condition. Immunoreactivity for pGSK3α /β , pAPP, Aβ 42-C-terminus, 4G8 and pCREB was quantified by drawing regions of interest (ROIs) in the acquired field and expressing fluorescence as the sum of intensities (8-bit depth; 256 levels) measured for every pixel in the ROI. Immunofluorescence for synaptophysin and synapsin-1 were quantified as "density" that is the total fluorescence intensity of synaptophysin/synapsin-1 labeling (quantified as above) divided by the total area in the field that was occupied by neurons (identified by MAP2 immunoreactivity and calculated by ImageJ software [available at http://rsbweb.nih.gov/ij/]). In every studied condition, negative controls were obtained by omitting the primary antibody. The operator was blinded to the study conditions. Western Blot. Infected or mock-infected mouse cortical neurons (12)(13)(14)(15) were washed twice with PBS and scraped in cold RIPA buffer containing 1 mM phenylmethylsulfonyl fluoride, phosphatase and protease inhibitor mixtures (Sigma), and 0,1% sodium dodecyl sulfate. After incubation for 30 min on ice, cellular suspensions were centrifuged (10,000 × g for 30 min at 4 °C) and the supernatants were collected and assayed to determine their protein concentration (Micro BCA protein assay, Thermo Fisher Scientific, Waltham, MA). Equivalent amounts of proteins were loaded onto either 8% tris-glycine polyacrylamide gels or 10-20% Novex Tricine precast gels (Invitrogen) for electrophoretic separation, and then electroblotted onto nitrocellulose membranes for Western blot analysis. Membranes were blocked with 5% nonfat dry milk in tris-buffered saline containing 0.1% Tween-20 for 1 h at RT and incubated with primary antibodies (rabbit polyclonal anti-GSK-3β , Cell Signaling; rabbit polyclonal anti-pGSK3 α /β [Tyr279/Tyr216], Signalway antibody; rabbit polyclonal anti-pGSK3 α β [Ser9] Cell Signaling; rabbit polyclonal anti APP, Cell Signaling; rabbit polyclonal anti pAPP [Thr668], Cell Signaling; rabbit polyclonal anti synapsin-1 XP, Cell Signaling; mouse monoclonal anti synapthophysin, Immunological Science; mouse monoclonal anti Beta-Amyloid, 17-24 [4G8], Covance; mouse monoclonal anti-α Tubulin, Sigma; mouse monoclonal anti-GAPDH, Abcam) at a final concentration of 1 μ g/mL. After incubation with appropriate secondary horseradish peroxidase-conjugated antibodies (1:2,000; Cell Signaling), visualization was performed with ECL plus (GE Healthcare, Amersham Place, Buckinghamshire) using either UVItec Cambridge Alliance or Kodak X-OMAT films. In the latter case,images of the blots were acquired using a high-resolution scanner. When necessary, membranes were stripped by heating at 56 °C in 62.5 mM Tris-HCl, pH 6.7, with 100 mM 2-mercaptoethanol and 2% SDS and re-probed. Molecular weights for immunoblot analysis were determined using BenchMark ™ Pre-Stained Protein Ladder (Invitrogen). Densitometric analysis was carried out with Image J software. Experiments were repeated at least 3 times.

Electrophysiological measurements in autaptic microcultures. Autaptic cultures of cortical
neurons were prepared as previously described 23,24 . Glass coverslips coated with agarose, which does not allow cell growth, were sprayed with a mixture of poly-D-lysine and collagen (both from Sigma) to create microislands where glial cells could be grown. Cortical astrocytes from the brains of postnatal day 0-to-2 C57/Bl6 mice were cultured according to standard procedure. Dissociated cells (suspended in Dulbecco's MEM supplemented with 10% fetal bovine serum and antibiotics) were plated onto the coverslips. After 4-6 days, when glial microislands had formed, half the medium volume was replaced with neuronal medium (consisting of Neurobasal medium, 2% B-27, 0.5% glutamine, and 1% PSN). Cortical neurons from postnatal day 0-to-2 C57/Bl6 mouse brains were prepared according to standard procedure 5 and suspended in neuronal medium. Neurons were then plated onto glial microislands at low density (25 × 10 3 /cm 2 ) to obtain a ratio of one neuron per island. Every 4 days half the neuronal medium volume was replaced with fresh neuronal medium supplemented with 2 μ M cytosine arabinoside. Autapses were studied from 9 to 21 DIV.
Synaptic transmission was studied using the patch-clamp technique in the whole-cell configuration as previously described 24,59,60 . An Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA) was used, and stimulation and data acquisition were performed with the Digidata 1200 series interface and pCLAMP 10 software (Molecular Devices). Recording electrodes were fabricated from borosilicate glass capillaries with the aid of a micropipette puller (P-97, Sutter Instruments, Novato, CA). The extracellular solution (pH 7.4) contained (in mM): 140 NaCl; 2 KCl; 10 HEPES; 10 glucose; 4 MgCl 2 ; and 4 CaCl 2 . The internal solution (pH 7.4) contained (in mM): 136 KCl, 17.8 HEPES, 1 EGTA, 0.6 MgCl 2 , 4 ATP, 0.3 GTP, 12 creatine phosphate, and 50 U/ml phosphocreatine kinase. Neurons were voltage-clamped at a membrane potential of − 70 mV, and stimuli mimicking action potentials (2 ms at 0 mV) were delivered to evoke EPSCs, which were recorded every 6 or 20 s. Input resistance, access resistance, and membrane capacity were monitored before and at the end of the experiments to ensure recording stability and the health of studied cells. The amplitudes and frequency of spontaneous mEPSCs were evaluated in 60-s recordings. The detection threshold was set to 3.5 times the baseline standard deviation.
Statistics. Statistical comparisons were performed with Systat 10.2 software. All data were expressed as mean ± standard error of the mean (SEM) and followed normal distribution, as assessed by the same software. In order to asses whether HSV-1 infection significantly affected the measured parameters, the two-tailed Student's t test was used. One-way ANOVA with Bonferroni's post-hoc test was used for multiple-comparisons. For experiments that included fewer than 10 observations (e.g. densitometric analysis of WB data), the Mann-Whitney (Wilcoxon) statistic was used. The level of significance was set at 0.05.