elevation in type i interferons inhibits hcn1 and slows cortical neuronal oscillations

Elevation in Type I Interferons Inhibits HCN1 and Slows Cortical Neuronal Oscillations

Konstantin Stadler1,, Claudia Bierwirth4, Luminita Stoenica1, Arne Battefeld1, Olivia Reetz1, Eilhard Mix4,Sebastian Schuchmann2,5, Tanja Velmans1, Karen Rosenberger3, Anja U. Bräuer1, Seija Lehnardt1,3, Robert Nitsch6,Matthias Budt7, Thorsten Wolff7, Maarten H.P. Kole8 and Ulf Strauss1,4

1Institute of Cell Biology and Neurobiology, 2Neuroscience Research Center (NWFZ), 3Department of Neurology, Charité-Universitaetsmedizin Berlin, Berlin, Germany 4Deptartment of Neurology, University of Rostock, Rostock, Germany, 5Klinik fürIntensiv- und Rettungsmedizin, Helios Klinikum Emil von Behring, Berlin, Germany, 6Institute for Microanatomy andNeurobiology, University Medicine, Johannes-Gutenberg-University Mainz, Mainz, Germany 7Division of Influenza/RespiratoryViruses, Robert-Koch Institute, Berlin, Germany and 8Netherlands Institute of Neuroscience, Amsterdam, The Netherlands

Address correspondence to Ulf Strauss, Institute of Cell Biology and Neurobiology, Charité-Universitätsmedizin Berlin, Charitéplatz 1, 10117Berlin, Germany. Email: [email protected]. and C.B. contributed equally to this work.

Central nervous system (CNS) inflammation involves the generationof inducible cytokines such as interferons (IFNs) and alterations inbrain activity, yet the interplay of both is not well understood. Here,we show that in vivo elevation of IFNs by viral brain infectionreduced hyperpolarization-activated currents (Ih) in cortical pyrami-dal neurons. In rodent brain slices directly exposed to type I IFNs,the hyperpolarization-activated cyclic nucleotide (HCN)-gatedchannel subunit HCN1 was specifically affected. The effect requiredan intact type I receptor (IFNAR) signaling cascade. Consistentwith Ih inhibition, IFNs hyperpolarized the resting membrane poten-tial, shifted the resonance frequency, and increased the membraneimpedance. In vivo application of IFN-β to the rat and to the mousecerebral cortex reduced the power of higher frequencies in the cor-tical electroencephalographic activity only in the presence of HCN1.In summary, these findings identify HCN1 channels as a novelneural target for type I IFNs providing the possibility to tune neuralresponses during the complex event of a CNS inflammation.

Keywords: EEG, ion channels, neuroinflammation, viral infection

Introduction

The cytokines interferons (IFN)-α and -β belong to type I IFNsand share receptor and signaling pathways. They are keyelements in the mammalian first-line innate immune responsebecause they activate immune cells, stimulate surface mol-ecules, and regulate the differentiation of monocytes (Pestka2007). In the context of central nervous system (CNS) inflam-mation, IFN-β is increased in myeloid cells (Prinz et al. 2008)and locally produced by neurons (Delhaye et al. 2006). Inturn, all cell types of the CNS respond to type I IFNs (Delhayeet al. 2006; Paul et al. 2007; Detje et al. 2009). Besides theseestablished immunological and antiviral effects, evidencefor a neuromodulatory potential of IFNs is growing. Forexample, during medical treatment, IFNs can lead to variousbehavioral changes, for example, fatigue, cognitive dysfunc-tion, depressed mood, condensed as sickness behavior(Dantzer et al. 2008). On the cellular level, type I IFNsenhance neuronal excitability in neurons of the cortex, thehippocampus, and the amygdala (Dafny et al. 1996).However, the mechanisms by which type I IFNs impact onneuronal excitability are poorly understood.

In our previous in vitro study, application of IFN-β lead toan increased firing rate and input resistance in corticalneurons (Hadjilambreva et al. 2005). Indirect evidence

suggested that the hyperpolarization-activated nonselectivecation current (Ih) may serve as a possible molecular target ofIFNs: IFN-β decreased neuronal resting conductance andblocking Ih prevented the increase in input resistance afterIFN-β application (Hadjilambreva et al. 2005). In addition,inflammation altered Ih in myenteric neurons (Linden et al.2003). However, interactions between ion channels and IFNshave not been investigated in detail.

Ih is mediated by HCN channels, which derive from fourgenes (HCN1–4) found throughout the brain (Santoro et al.1998). HCN channel subunits generate channels with distinctbiophysical properties by assembling to homo- or heterotetra-meric complexes (Frere et al. 2004). In neocortical pyramidalneurons, Ih is mainly mediated by HCN1 and HCN2. HCN1 ishighly expressed in distal dendrites where it regulates actionpotential firing patterns (Kole et al. 2006), synaptic inte-gration (Stuart and Spruston 1998; Strauss et al. 2004) andcontributes to the subthreshold somato-dendritic voltageattenuation (Zhang 2004) and membrane resonance (Naraya-nan and Johnston 2008). Modulation of Ih provides a cellularmechanism to alter network oscillations of neural circuitsduring cognitive processing (Wahl-Schott and Biel 2009).HCN channels are sensitive to a number of intra- and extra-cellular modulators, which in many cases act by shifting thechannel’s voltage sensitivity via either cAMP, intracellularprotons, phosphatidylinositol-4,5-phosphate or acidic lipids(for review: Wahl-Schott and Biel 2009).

To address whether and, if so, how the neuromodulation byIFNs is mediated via Ih changes, we investigated the effect oftype I IFNs on HCN channels of rat neocortical layer 5 neurons.The results show a reduction and deceleration of Ih with conse-quences for the neuronal frequency behavior. This effect ismediated by the IFN-signaling cascade and specifically gov-erned by a modulation of the fastest activating HCN subunit,HCN1. The data emphasize the role of cytokines in determiningthe single neuron and network states and present the first directevidence for a type I IFN action on a neuronal ion channel.

Materials and Methods

Interferons and Signaling Pathway ModulatorsFor all experiments, Chinese hamster ovary-derived recombinant IFN(rat IFN-α,-β: U-CyTech, Utrecht, The Netherlands; mouse IFN-β: Hy-cultec GmbH, Beutelsbach, Germany) was used. The lyophilizedproduct was reconstituted in sterile double-distilled water, and small

© The Author 2012. Published by Oxford University Press. All rights reserved.For Permissions, please e-mail: [email protected]

Cerebral Cortexdoi:10.1093/cercor/bhs305

Cerebral Cortex Advance Access published October 4, 2012 at R

obert Koch-Institut on N

ovember 27, 2012

http://cercor.oxfordjournals.org/D

ownloaded from

http://cercor.oxfordjournals.org/

aliquots were stored according to the data sheets. The activation ofJanus protein tyrosine kinase (JAK)1 or JAK2 was prevented by theselective blocker substances JAK inhibitor1 and AG-490, respectively(both from Calbiochem, San Diego, CA, United States of America).Blockers were dissolved in dimethyl sulfoxide to 75 or 500 µM,respectively, and stored at −20 °C.

AnimalsMale Wistar rats (Forschungseinrichtung für experimentelleMedizin (FEM), Berlin, Germany) and mice (C57/B6/J and B6/129-HCN1tm2Kndl/J, Jackson Laboratory) were bred at the local animal facil-ity. For HCN1−/− experiments B6/129-HCN1tm2Kndl/J were crossedwith C57/B6/J, heterozygote offspring were further crossed, and geno-typing was performed from tail cuts via polymerase chain reaction(PCR) according to the available protocol (Jackson Laboratory).Animals were kept under standard laboratory conditions, and all pro-cedures were performed in agreement with the European Commu-nities Council Directive of 24 November 1986 (86/609/EEC).

Viral InfectionAnesthetized (ketamine, 100 mg kg−1 intraperitoneal (i.p.); DeltaSelectand xylazine 20 mg kg−1 i.p.; Bayer Health Care, Berlin, Germany)21-day-old male C57/B6 mice were intrathecally (L2 or L3) injected(Hoffmann et al. 2007) with 20 µL of phosphate-buffered saline (PBS; 3mice) or 25 µL virus stock (9 mice) containing 106 PFU Theiler’s murineencephalomyelitis virus (TMEV) GDVII (Delhaye et al. 2006). Postopera-tively, adequate waking and the absence of paresis were verified. Exper-imental procedures were reviewed by institutional and state authorities(G0175/07).

Slice Preparation and Culture of Cortical Neuronsand HEK293 CellsFor a detailed description see Supplementary Methods.

Patch-Clamp RecordingsIndividual slices, HEK293 cells, or primary cultures were transferred toa submerged recording chamber. Cortical neurons and HEK293 cellswere visualized with either an Axioskop 2 FS Zeiss or an Axiovert S100(both from Carl Zeiss MicroImaging GmbH, Göttingen, Germany), andwhole-cell patch-clamp experiments were carried out at room tempera-ture (RT; 22–24 °C). For the visualization of cortical pyramidal neurons,infrared differential interference contrast video microscopy was used.These experiments were accomplished at 32–34 °C. Patch pipetteswere pulled to a final resistance of 3–5 MΩ. In somatic whole-cellvoltage clamp, a maximal series resistance of 15 MΩ with changes< 20% during recordings was tolerated. A fast (pipette) capacitive tran-sient (τ < 1.5 μs, 6–13 pF) was compensated. The pipette solution con-tained (in mM): 120 KMeSO4 (ICN Biomedicals, Eschwege, Germany),20 KCl (Merck), 14 Na-phospocreatine, 0.5 ethylene glycol tetraaceticacid, 4 NaCl, 10 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, 4Mg2+-ATP, 0.3 Tris3-GTP with or without 0.1 cAMP (all fromSigma-Aldrich, pH 7.25, 288 mOsm). A liquid junction potential of−10 mV has not been corrected for. Voltage-clamp recordings ofpharmacologically isolated Ih were obtained by blocking IK(IR)with 200–400 µM Ba2+ added to a modified artificial cerebrospinalfluid: 10 mM K+, MgCl2 replacing MgSO4 (all from Merck), andNaH2PO4 omitted. Sodium currents were suppressed with 1 µM tetro-dotoxin (Tocris, Bristol, United Kingdom), low-threshold Ca2+ currentswere blocked with 1 mM Ni2+, glutamate receptors were blocked with20 µM 6-cyano-7-nitroquinoxaline-2,3-dione (both Sigma-Aldrich), and25 µM D-(-)-2-amino-5-phosphonopentanoic acid (Tocris), GABAA re-ceptors were blocked with 10 µM bicuculline (Tocris), K+ currents(fast-inactivating A-type, IA and delayed rectifier type, IK(DR)) wereblocked with 5 mM 4-aminopyridine and 10 mM tetraethylammonium(both from Sigma-Aldrich). Data from all patch-clamp recordings werecollected with an EPC-10 (HEKA Elektronik GmbH, Lambrecht,Germany), digitized (10 kHz, after Bessel filtering at 2.5 kHz), andstored using PATCHMASTER software (HEKA). The electrical resonance

of the neurons was analyzed with the impedance (Z) amplitude profile(ZAP) method (Narayanan and Johnston 2008). For details see Sup-plementary Experimental Procedures.

SimulationsFor simulations of membrane resonance, we implemented an estab-lished model (Hutcheon et al. 1996) in LabVIEW 8.6 (National Instru-ments Inc., TX, United States of America) and Scilab 5.0.1 (ScilabConsortium, INRIA, ENPC and Contributors, 1989–2008) and utilized aNEURON model (version 7.1; see Supplementary Experimental Pro-cedures) using a reconstructed morphology of a layer 5 pyramidalneuron (Stuart and Spruston 1998, Fig. 1A therein) with properties de-scribed in detail in Supplementary Experimental Procedures. The latterwas also applied to test the contribution of the different HCN subtypes.

Quantitative Real-Time PCRFor details see Supplementary Methods.

ImmunochemistrySagittal brain sections (30 µm) were prepared as described (Kole et al.2007). The endogenous peroxidase was quenched by 0.3% hydrogenperoxide diluted in PBS for 30 min at RT. After washing, brainsections were blocked with 10% fetal bovine serum (FBS) inPBS overnight at 4 °C. Sections were incubated with anti-IFNAR1(Abcam, Cambridge, United Kingdom, diluted 1:50 in blockingsolution) at 4 °C for 48 h. After washing in PBS, sections were incu-bated with a biotinylated secondary anti-rabbit antibody (Invitrogen,Darmstadt, Germany) diluted 1:400 in PBS overnight at 4 °C and thenin avidin–biotin peroxidase complex reagent (Vectastain ABC Kit,Vector Labs, Burlingame, CA, United States of America) for 2 h at RT.Immunoreaction was visualized with 3,3′-diaminobenzidine as achromogen. Cultured primary neurons were fixed with 4% parafor-maldehyde and 15% sucrose for 20 min at RT. Cells were treated with20% FBS and incubated with anti-IFNAR1 antibody diluted 1:100 in 5%FBS. After washing with PBS, cells were incubated with a secondaryAlexa 488-labeled goat anti-rabbit antibody (Invitrogen; diluted1:1500). Afterwards, cells were treated with 0.1% Triton X-100 for3 min and again blocked with 20% FBS. Anti-mitogen-activated protein2 (MAP2; Sigma-Aldrich) was incubated in a dilution of 1:1500 in 5%FBS. The cells were washed thrice, and Alexa 568-labeled goat anti-mouse antibody (Invitrogen) was added in a dilution of 1:1500. Allincubations were performed overnight at 4 °C. Cells were imaged usingan upright confocal microscope (DM250, Leica Microsystems CMSGmbH, Mannheim, Germany). For further details and cultures of glialcells see Supplementary Experimental Procedures.

Western BlottingIFN-β–treated primary neurons and respective controls were lysed inice-cold buffer (150 mM NaCl, 1% NP-40, 25 mM MgCl2, 10% glycerin,50 mM Tris, pH 7.4) containing complete protease inhibitor cocktailand phosphatase inhibitors (both Roche, Mannheim, Germany).Protein extracts were separated on SDS–PAGE and electroblotted to anitrocellulose membrane. Membranes were blocked for 3 h at RT with5% BSA and incubated with the following antibodies: Anti-p38 MAPkinase 1:1000; anti-phospho-p38 MAP kinase (Thr180/Tyr182) 1:250,anti-phospho-Tyk2 (Tyr1054/1055) 1:1000 (all Cell Signaling,Danvers, MA, United States of America), or anti-Tyk2 1:200 (SantaCruz Biotechnology, Santa Cruz, CA, United States of America) over-night at 4 °C. Blots were incubated with horseradish peroxidase-conjugated secondary antibodies (1:5000, GE Healthcare, Chalfont StGiles, United Kingdom) overnight at 4 °C, and immunoreactive bandswere visualized with chemoluminescence.

EEGCortical activity from freely moving Wistar rats was recorded on post-natal days (P)11–12 and from freely moving C57/B6/J and B6/129-HCN1tm2Kndl/J mice on P32–40. For rats: Cortical electrodes were

2 IFN Modulation of Ih • Stadler et al.

at Robert K

och-Institut on Novem

ber 27, 2012http://cercor.oxfordjournals.org/

Dow

nloaded from

http://cercor.oxfordjournals.org/lookup/suppl/doi:10.1093/cercor/bhs305/-/DC1









placed and fixed at P10–11 as described previously (Schuchmannet al. 2006) at the following coordinates: 3 mm posterior from thebregma, 2 mm lateral from the midline, with a subdural referenceelectrode above the cerebellum. Close to the recording electrode, aguiding plastic tube (conical, tip outer diameter 0.9 mm) was im-planted above the dura (2 mm posterior from the bregma, 2 mm lateralfrom the midline; Fig. 8A). After recovery from anesthesia, the pupswere returned into their original litter; experiments started earliest 24 hafter surgery. The recordings were performed in a specific heated area(32 °C; Supplementary Fig. S8A). The signals were sampled at 0.5–3kHz using a 12-bit data acquisition board (AD Instruments GmbH,Spechbach, Germany). After a period of exploration (30–60 min), theanimals started to show increasing silent periods, which were recordedfor at least 2 h. Using the implanted guiding plastic tube, we appliedrat IFN-β. To reach a large number of neurons and an immediate onsetof IFN-β effect, we used a time delayed application technique: Thedura was perforated through the implanted guiding tube with a26-Gauge needle, then we applied 5000 IU IFN-β in 5 µL saline via a30-Gauge Hamilton syringe into the tubing on top of the perforateddura mater. Using this technique, we avoided any volume effect to thecortex. Subsequently, we recorded again for at least 1 h. For the analy-sis, we only used periods with motorically silent behavior. In controlexperiments, only saline was applied.

For mice, electroencephalography (EEG) experiments were per-formed as in rats with slight modifications: Between P30–34 corticalelectrodes were placed and fixed 1.5 mm posterior from the bregma and2 mm lateral from the midline. Saline and IFN-β injections were done inidentical animals at least 48 h apart (Fig. 9A) to increase comparabilityand to reduce animal numbers. Because the mice never were motoricallysilent, periods showing the least motor artifacts were analyzed.

Statistical AnalysisFor all data, statistics were performed depending on the dataset asappropriate in Origin7 (OriginLab, Northhampton, MA, United Statesof America) or Statview v.457 (Abacus Concepts Inc., CA, UnitedStates of America). In detail, for normal distributed datasets (Shapiro-Wilk W-test), we used 2-tailed Student’s t-tests. In the case of signifi-cant deviations from the normal distribution (P≤ 0.05) or if thesample size was too small (n≤ 8) for a reliable test of normal distri-bution, nonparametric tests were used: Wilcoxon signed-rank test for

paired sample sets and Mann–Whitney U-test for unpaired samplesets. Data are presented as mean ± SD.

Results

Viral Infection Inhibits Neocortical IhNeurons in mice infected with the neurovirulent GDVII strainof TMEV act as producers of type I IFNs after intrathecal virusapplication (Delhaye et al. 2006). To test whether in vivo IFNinduction by a CNS infection is influencing Ih, we used theTMEV model and performed whole-cell recordings on soma-tosensory cortex layer 5 neurons at the third or fourth daypostinfection (Fig. 1A,B), when animals started to show signsof sickness indicating the elevation of early induced type IIFN levels. TMEV infection reduced the amplitudes ofpharmacologically isolated Ih over a broad-voltage range(Supplementary Fig. S1A). At maximum activation (−130 mV),TMEV-reduced Ih amplitudes to 66% in layer 5 neurons(n = 29), compared with such neurons in sham-injected mice(n = 25, P < 0.003, Fig. 1C). Comparing Ih densities to excludeconfounding cell-size effects showed a likewise reduction to77% (P < 0.05, Supplementary Fig. S1B). The current reductionwas accompanied by a slightly hyperpolarized voltage depen-dence of Ih activation (ctrl: V1/2 = –85.8 ± 3.3 mV, TMEV: V1/

2 = –90.4 ± 5.5 mV, P < 0.001, Fig. 1D). This shift cannot causethe reduction of maximal Ih, but may increase it at physiologi-cal membrane potentials (Fig. 1D, Supplementary Fig. S1A).Thus, type I IFN elevation in a pathophysiological contextsuch as viral CNS infection appears sufficient to inhibit Ih.

Type I IFNs Reduce Neocortical Ih when Applied Acutelyand DirectlyTo study the IFN effect in more detail, we acutely appliedrecombinant type I IFNs derived from Chinese hamster ovarycells to rat slices containing the somatosensory cortex. The

Figure 1. Infection with neurotopic GDVII strain of the TMEV-reduced hyperpolarization-activated currents (Ih) in mouse pyramidal cortical neurons. (A) Scheme of in vivo IFNinduction by intrathecal TMEV/GDVII injection in mice at P21 followed by decapitation and prompt in vitro slice recordings after the first signs of sickness (3–4 days postinjection).(B) Exemplified responses to a family of graded voltage steps (bottom) in neurons from infected (green) and noninfected (black) animals under pharmacological Ih isolation. (C)Top: Comparison of averaged Ih traces of all layer 5 neurons from TMEV- (green) and sham- (black) injected animals at −130 mV. Bottom: Population data and 25–75% box plots.The mean Ih amplitudes (“thick lines through boxes and data”) differed between neurons in slices from noninfected and infected animals (sham: Ih = 1.18 ± 0.50 nA; n= 25 vs.TMEV: Ih = 0.78 ± 0.43 nA; n= 29). Thin horizontal lines within boxes represent the median. (D) Top: Steady-state activation curves constructed from mean relative tail currentsupon return to −65 mV from the activation of Ih at different voltages for 2 s, plotted against preceding test voltage in sham- (black) or TMEV- (green) injected animals. Fits of theBoltzmann function are superimposed. Note that the V1/2 shift is not sufficient to effectively reduce the channel availability at voltages below −120 mV. Bottom: Population data ofIh voltage dependence (ctrl: V1/2 =−85.8 ± 3.3 mV, TMEV: V1/2 =−90.4 ± 5.5 mV). *P<0.05, **P<0.01, ***P< 0.001 throughout all figures.

Cerebral Cortex 3

at Robert K



Dow

nloaded from






concentration we used (1000 IU mL−1) had an assured effecton neurons (Hadjilambreva et al. 2005) and is in the rangeobserved after systemic viral infections (Heremans et al.1980). The treatment reduced Ih peak amplitude in allneurons, on average to 80 ± 14% for IFN-α (n = 10, P < 0.05)and to 75.5 ± 19.1% for IFN-β (n = 16, P < 0.001, Fig. 2A) at−130 mV for an example of the entire voltage range (Sup-plementary Fig. S2A,B).

Acute type I IFN induced amplitude reduction was not dueto a change in voltage sensitivity. Both the half-maximum acti-vation, V1/2 and the steepness of the activation curve, k were

not affected; neither by IFN-α (ctrl: V1/2 = –88.3 ± 3.4 mV vs.IFN-α: V1/2 = –86.6 ± 4.3 mV, P > 0.16; ctrl: k = 10.6 ± 1.0 vs.IFN-α: k = 10.0 ± 2.1, P > 0.06; n = 10) nor by IFN-β (ctrl: V1/2=–88.3 ± 7.5 mV vs. IFN-β: V1/2 = –89.6 ± 7.9 mV, P > 0.16; ctrl:k = 10.6 ± 2.4 vs. IFN-β: k = 10.8 ± 2.2, P > 0.62; n = 16; Fig. 2B,C). As the effect of both type I IFNs appeared similar, resultsof further experiments are shown just for IFN-β. Thereduction in peak amplitude developed surprisingly rapid fora cytokine, starting at about 10 min after IFN-β arrived atthe slice and took about 15 further minutes to reach themaximum (Fig. 2D). The peak amplitude partially recovered

Figure 2. Application of type I IFN (1000 IU mL−1) reduced Ih in rat layer 5 neocortical neurons of the somatosensory cortex. (A) Left: Traces of pharmacologically isolated Ih(bottom: Voltage step). Bath application of IFN-α led to an Ih attenuation (blue trace). Middle: Population data, the mean Ih amplitude changed from 1.22 ± 0.57 to 1.00 ± 0.50nA (n=10). Right: IFN-β led to a comparable Ih reduction from 681± 626 to 521 ± 460 pA (n= 16, for responses to a family of graded voltage steps see Supplementary Fig.S1A). Modulation of Ih increased with the IFN dose and occurred over a concentration range from 100 to 10 000 IU (shown for IFN-β in Supplementary Fig. S2D). (B) Steady-stateactivation curves constructed from mean relative tail currents, plotted against the preceding test potential without (black circles) and with type I IFN (IFN-α blue circles, IFN-β redcircles) application. Fits of the Boltzmann function are superimposed. Note, that for statistical comparison steady-state activation curves of each individual neuron wereconstructed. Inset: Family of tail currents recorded upon return to −65 mV from 2 s voltage steps from −40 to −130 mV. (C) The mean V1/2 did not change upon type I IFNapplication (left: IFN-α, right: IFN-β). (D) Exemplified time course of the Ih reduction induced by IFN-β (red circles) compared with the time course of Ih in a nontreated neuron(black circles). For long-term stability voltage pulses of −100 mV for 2 s were applied (instead of the maximum of –130 mV shown for most other experiments). (E) The remaininghyperpolarization-activated current after blocking of HCN channels with ZD7288 (50 μM) was stable upon IFN-β application (n=9), indicating no accompanying change in non-Ihcurrent components. (F) The membrane potential hyperpolarized after IFN-β application from −69.1 ± 0.8 to −72.4 ± 2 mV (n= 9).


at Robert K



Dow

nloaded from







when IFN-β was washed out (Supplementary Fig. S2A–C). Inthe presence of the specific Ih blocker ZD7288 (50 µM), IFN-βdid not change the residual non-Ih mediated, hyperpolariz-ation-evoked currents (at –130 mV: 495 ± 130 vs. 502 ± 154pA, n = 9, P > 0.86; Fig. 2E), further supporting the idea thatthe current reduction was specific to HCN channels. Consist-ent with the contribution of Ih to the resting membrane poten-tial, application of IFN-β hyperpolarized layer 5 neurons by3.3 ± 2.4 mV (n = 9, P < 0.01; Fig. 2F). The Ih reversal poten-tial, determined from the fully reconstructed current-voltagerelationships, did not change due to IFN (tested for IFN-β:ctrl: –20.5 ± 1.4 mV vs. IFN-β: –22.1 ± 5.2 mV; n = 16, P > 0.6),and the effective maximum Ih conductance reduced signifi-cantly from 13.1 ± 11.3 to 9.9 ± 8.2 nS (P < 0.001, n = 16, Sup-plementary Fig. S2E). On the whole, these data suggest thatIFNs rapidly and reversibly decrease the HCN channel con-ductance, either by reducing channel number or single-channel conductance.

Ih Reduction is due to a Pronounced Inhibitionof HCN1 ChannelsIn addition to the reduction of the amplitude, type I IFNs de-celerated Ih kinetics in rat cortical neurons. The Ih activationwas best described by double-exponential fits (Fig. 3A, Sup-plementary Fig. S3A). IFN-β induced a selective increase inthe faster time constant by a factor of 1.31 ± 0.42 (n = 16,P < 0.05), whereas the slower time constant remained thesame (Fig. 3B). This was further reflected in the reduction ofthe relative amplitude contribution of the fast component(0.70 ± 0.09 in ctrl vs. 0.61 ± 0.06 under IFN-β; P < 0.01,Fig. 3C). In conjunction with the slowing of the activation,IFNs prolonged the deactivation of Ih by a factor of 1.5 ± 0.65(n = 16, P < 0.01, Fig. 3D, Supplementary Fig. S3C–E). Theseeffects were qualitatively reproduced by IFN-α (Supplemen-tary Fig. S3C) and partly by TMEV infection (SupplementaryFig. S3H).

Because acutely applied type I IFNs predominantly de-creased the fast component of Ih activation, we assumed anion channel subtype specificity of the effect, namely areduction of the fastest activating subtype, HCN1. To test thishypothesis, we used a NEURON model based on a

reconstruction of a layer 5 neuron’s morphology. In the firstinstance, we fitted the model parameters (SupplementaryExperimental Procedures) to our experimental results beforeIFN application. We then simulated 2 putative mechanisms ofHCN channel inhibition by IFNs: 1) a uniform reduction ofboth HCN1 and HCN2 channel subtype peak conductances,�gh, versus 2) a specific reduction of the HCN1 channel peakconductance, �gHCN1. To attenuate the Ih amplitude to 76%, asin vitro measured at the soma, we reduced �gh for both set-tings. This required a combined reduction of �gh to 61.2% or areduction of �gHCN1 to 42.5%. The specific HCN1 reduction mi-micked our in vitro experimental results (Fig. 4A): The fasttime constant of activation decelerated by a factor of 1.2 (ctrl:τfast = 78 ms; IFN-β: τfast = 93 ms; relative amplitude of τfast:0.61 vs. 0.59), whereas the slow time constant of activationremained almost unchanged (ctrl: τslow = 451 ms; IFN-β: τslow =491 ms) and the deactivation time constant increased by 1.6(ctrl: τ = 212 ms; IFN-β: τ = 340 ms). In contrast, modeling auniform reduction of both HCN subunit models only margin-ally increased the time constants of activation (τfast = 88 ms,τslow = 482 ms; relative amplitude of τfast: 0.74) correspondingto an increase by the factor of 1.12 and 1.06, respectively, andthe time constant of deactivation rather decreased (τ = 210ms).

To test directly whether the HCN1 subunit is required forthe IFN-mediated modulation of Ih, we first isolated and thenmeasured Ih in HCN1−/− mice (Nolan et al. 2003) before andafter the application of recombinant mouse IFN-β. In accord-ance with previous studies (Chen, Shu, Kennedy et al. 2009),Ih in neocortical neurons in these mice amounted to one-thirdof the one in their HCN1+/+ litters (Fig. 4B,C). In line with ourmodel predictions, IFN-β did not affect the remaining Ih inHCN1−/− mice, presumably mediated by HCN2 channels,whereas in control HCN1+/+ mice we observed a similarattenuation as in rats. In detail, 1000 IU mL−1 IFN-β reducedIh in HCN1+/+ to 71.9 ± 15.4% (P < 0.01, n = 8), but left it un-changed in HCN1−/− (99.7 ± 19.2%, P > 0.4, n = 8).

If HCN channels were composed of homomers, one wouldexpect a distinct V1/2 shift upon selective HCN1 reduction, inparticular when the intracellular cAMP levels are low. To testthis assumption, we removed cAMP from the intracellular

Figure 3. IFN-β (1000 IU mL−1) preferentially reduces fast Ih components in neocortical pyramidal neurons. (A) Ih activation was best described by a double-exponential fit(superimposed fit and corresponding residuals in light gray for a 2 vs. dark gray for a 1 exponential fit; for a statistical comparison see Supplementary Fig. S3A). The correspondingparameters for the displayed double-exponential fits are Afast = 190.9 pA, τfast = 178 ms, Aslow = 105.2 pA, τslow = 1.248 s. (B) IFN-β decelerates Ih activation as shown byquantitative comparison of the fast (τfast, left) and slow (τslow, middle) activation time constants before and after IFN-β obtained by a −130 mV step. The mean τfast increasedfrom 102± 43 to 125 ± 47 ms (n= 16). For an example fit see Supplementary Figure S3B. (C) IFN-β reduced the relative fast component of Ih (Afast/Afast + Aslow) from0.70 ± 0.09 in control to 0.61 ± 0.06 (n=16). (D) IFN-β led to a deceleration of the Ih deactivation kinetics estimated from the exponential decline of the tail current at −65 mVafter a 2-s hyperpolarizing step to −130 mV (Supplementary Fig. S3C,D). The mean deactivation time constant changed from 211± 56 to 310 ± 129 ms.

Cerebral Cortex 5

at Robert K



Dow

nloaded from

















solution. Also under this condition, IFN-β reduced Ih to73 ± 16% (P < 0.05, n = 6; Supplementary Fig. S3F) withoutchanging the voltage dependence of channel activation (ctrl:V1/2 = –91.2 ± 4 mV vs. IFN-β: V1/2 = –89.1 ± 3 mV, P > 0.24,n = 6; Supplementary Fig. S3G). When rerunning our modeladjusted to the data obtained in vitro, where omitting cAMPshifted the control V1/2 by −2.9 mV without reaching the sig-nificance levels (P = 0.14, compared with 100 µM cAMP), re-ducing the HCN1 conductance to 42.5% led to the predictionof a V1/2 shift by −0.88 mV. This is below our experimentalresolution limit.

Taken together, these results are consistent with a specificinhibition of HCN1-mediated Ih by type I IFNs.

IFN-β Inhibits Neuronal Ih without Glial IntermediationTo distinguish a direct neuronal modulation from indirectglial effects, we used primary cultured cortical neuronsmeasured at 9–14 days in vitro. In these cultures, the amountof confounding glial cells was reduced to < 5%. Similar to theobservations in slices, application of IFN-β to these culturesled to a reduction of the Ih peak amplitude in all corticalneurons to 75.7 ± 14.9% (n = 5, P < 0.05; Fig. 5A). Consistentwith the decreased HCN conductance, the input resistance in-creased on average 14% in current-clamp recordings(P < 0.001, n = 6; Fig. 5A). These results suggest that the sub-threshold effects of IFN-β on cortical excitability are directlyattributable to neuronal mechanisms.

IFN-β-Induced Ih Inhibition Requires an IntactCanonical Type I IFN Receptor PathwayDoes IFN-β act directly on HCN channels? To address thisquestion, we expressed the prominent cortical HCN channelsubunit HCN1 alone and/or together with HCN2 under con-ditions where the rat IFN-β signaling pathway is negligible.Given the relative insensitivity of human IFN receptors to ratIFN (Novick et al. 1994) and no confounding currents in thehyperpolarizing range (Supplementary Fig. S5B), we chosethe HEK293 expression system. This approach revealed thatrat IFN-β did not directly affect currents mediated by rat HCN1(ctrl: 2.25 ± 1.6 nA vs. IFN-β: 2.20 ± 1.5 nA, n = 6, P = 0.35,Fig. 5B left), by rat HCN1/2 (ctrl: 2.39 ± 1.65 nA vs. IFN-β:

2.40 ± 1.67 nA, n = 9, P > 0.93, Fig. 5B right), and by rat HCN2(Supplementary Fig. S5A) in cells lacking the specific rat typeI IFN receptors. The result points to an IFN-β effect mediatedby intracellular signaling cascades rather than a direct confor-mational change in the channel.

IFN-β actions are mediated by 2 transmembrane receptorchains (IFNAR1 and 2) that form the specific type I IFN mem-brane receptors (IFNARs; Takaoka and Yanai 2006). First, weinvestigated whether both IFNAR chains are actually ex-pressed by resident cells in the cerebral cortex. Quantitativereal-time PCR from cortical tissue showed expression ofIFNAR1 and IFNAR2mRNA. A detailed analysis in primary cul-tures indicated that both receptor chains are present inneurons, astrocytes, and microglia (Fig. 6A). Immunohisto-chemical staining of rat cortex revealed IFNAR1 in neurons(Fig. 6B, for antibody specificity see Supplementary Fig. S6A).IFNAR1 staining on primary cortical neurons showed a punc-tuated immunoreaction in the cell bodies and the dendrites(Fig. 6C). To determine whether this neuronal expression ofIFNAR1 is functional, we tested the activation of signaling mol-ecules downstream of the intracellular domain of the IFNARsubsequent to IFN-β binding. The signaling cascade involvesTyk2 and JAK1, both enzymes of the JAK family, and thesignal transducers and activators of transcription (STAT)1 andSTAT2 (Takaoka and Yanai 2006). Alternate type I IFN signal-ing pathways operate via theCa2+-independent di-acylglycerol-mediated protein kinase Cpathway or via the MAP kinase pathway, which include p38and the extracellular signal-regulated kinase 2 (Takaoka andYanai 2006). For a selective signal cascade analysis, we incu-bated primary neocortical neurons for 30 min with bathsolution containing 1000 IU mL−1 IFN-β. We first probed thecanonical pathway, represented by Tyk2, for phosphorylation.Subsequent to receptor activation by IFN-β application, we ob-served a marked phosphorylation of Tyk2 (Fig. 6D), whichpoints to a proper activation of the early signaling steps.However, the alternative p38 MAP kinase pathway was not ac-tivated in neurons, even after 30 min of IFN-β treatment(Fig. 6E). Taken together, we conclude that the IFNAR and itscanonical pathway are present and functionally active inneurons.

Figure 4. IFN-β acts via HCN1-mediated current reduction. (A) Simulating IFN-β effects in a morphologically realistic model of a layer 5 neuron (Stuart and Spruston 1998)adjusted to our in vitro control values (black) favor a specific HCN1 reduction: Reducing HCN1 (dashed) mimicked the pronounced increase in τfast, whereas a uniform attenuationof the peak conductance of both channel subtypes (dotted) failed to do so. Inset: HCN density (2/3 HCN1 and 1/3 HCN2) was distributed exponentially across compartmentswith a 40-times increasing density starting at the soma with 0.95 pS μm−2. (B) Reduction of Ih by 1000 IU mL−1 mouse (m) IFN-β (gray) in HCN1+/+ mice (top) was similar tothe one in rat neurons but absent in HCN1−/− mice (middle). (C) Population data of the mouse IFN-β effect on mouse neurons. Whereas mouse IFN-β reduced Ih in HCN1

+/+

mice from 829± 314 to 590 ± 274 pA (n= 8), it failed to do so in HCN1−/− mice (213 ± 119 pA in ctrl vs. 202 ± 98 pA under mouse IFN-β; n= 8).


at Robert K



Dow

nloaded from







Both the existence of neuronal IFNAR and the lack of adirect effect of IFN-β on the HCN channel subunits promptedus to investigate whether IFNAR downstream signaling cas-cades are required to inhibit Ih by testing neuronal IFN-βmodulation in the presence and absence of JAK selectiveblockers. Inhibition of the first signal transduction step withinthe JAK/STAT pathway prevented the Ih attenuation by IFN-β,as tested by application of IFN-β after pretreatment with 75nM JAK1 inhibitor (Thompson et al. 2002). Upon such treat-ment, the amplitude of Ih remained at 101.6 ± 15.9% (n = 12,P > 0.96; Fig. 6F). In contrast, AG-490, a preferentially type IIreceptor associated JAK2 inhibitor (Meydan et al. 1996), wasnot effective in blocking the IFN-β modulation of Ih (Sup-plementary Fig. S6B). These data show that JAK1/Tyk2, butnot JAK2 blockade, prevents the IFN-β effect, supporting thehypothesis that the IFN-β effect on neurons requires theJAK1/Tyk2 receptor-signaling pathway.

IFN-β Shifts Neuronal Resonance by Ih ModulationTo test whether the IFN-β–mediated HCN1 channel reductionleads to functionally relevant changes in cellular excitability,we explored the impact of IFN-β on resonance. One functionof Ih in pyramidal neurons is to complement passive cellproperties in generating a membrane resonance leading to a

Figure 5. IFN-β (1000 IU mL−1) modulates Ih independently of the presence of aproper glial environment but fails to act directly on heterologously expressed corticalHCN channels. (A) Left: Recordings of Ih in a primary cultured cortical neuron showingan amplitude reduction after exposition to IFN-β from 577 (black trace) to 353 pA(gray trace). Ih was elicited by a −130 mV voltage step (bottom). Inset: Time courseof the Ih reduction. Middle: Quantitative comparison of Ih amplitudes at −130 mVbefore and after the application of IFN-β for 6–30 min. The mean Ih amplitudedecreased from 389± 134 to 283 ± 63 pA (n= 5). Right: Population data on inputresistance obtained by current-clamp recordings on primary cultured neocorticalneurons revealed an increase from 645± 518 to 721 ± 549 MΩ (7–30 min of IFN-βapplication, n= 6). (B) Representative traces of Ih mediated by rat HCN1 (left) and ratHCN1/2 (right) overexpressed in HEK293 cells. Ih was elicited by a voltage step to−130 mV (respective bottom). Application of IFN-β (gray trace) did not change Ihwhen compared with control (black trace) for rat HCN1 (n= 6) or rat HCN1/2(n= 9). Population data for the Ih amplitudes from rat HCN1 and rat HCN1/2overexpressing HEK293 cells demonstrate no direct effect of IFN-β on the respectivechannels (right of the respective traces). Scale bars: 0.25 s; 2 nA.

Figure 6. The type I IFN receptor (IFNAR) is functionally present on neocorticalneurons and constitutes a prerequisite of the Ih attenuation by IFN-β. (A) Quantitativereal-time PCR revealed mRNA expression of both IFNAR subunits (IFNAR1 andIFNAR2) in the rat neocortex, primary cultured neurons, astrocytes, and microglia.Data were normalized to glyceraldehyde 3-phosphate dehydrogenase. Normalizationto hypoxanthine phosphoribosyltransferase gained similar results. (B) Top and bottomleft: Immunohistochemical analysis in the rat neocortex revealed IFNAR1 proteinexpression in neurons. Top: Arrows indicate neuronal structures, scale bar = 20 µm.Bottom: Slices processed without primary IFNAR1 antibody (scale bar = 20 µm). Forantibody confirmation see Supplementary Figure S6A. (C) Co-localization studies withthe neuronal marker MAP2 (red) confirmed the IFNAR1 (green) expression in somataand proximal dendritic structures in cultured cortical neurons. As control, neuronswere processed without primary IFNAR1 antibody (scale bar = 10 µm). (D)Activation of IFNAR after treatment of primary cortical neurons with 1000 IU mL−1

IFN-β for 30 min was confirmed by phosphorylation of the receptor-associatedtyrosine kinase 2 (Tyk2). For (D and E) molecular weight (MW) is given in kDa. (E)IFN-β for 30 min did not change the activation state of p38 MAP kinase, whereastreatment with 2 mM CaCl2 for 10 min readily phosphorylated the enzyme. (F)Inhibition of IFNAR-associated JAK1 prevented the IFN-β–induced change in Ih. Left:Ih traces from a layer 5 neuron under control conditions with 75 nM JAK inhibitor 1(blue trace) and 30 min after the application of 1000 IU mL−1 IFN-β (red trace) inresponse to a 2 s hyperpolarizing voltage step to −130 mV (bottom). Right:Population data of Ih before (mean: 795 ± 253 pA) and after (mean: 797 ± 259 pA,n=12/12) the superfusion with IFN-β.

Cerebral Cortex 7

at Robert K



Dow

nloaded from





distinct frequency preference of incoming inputs at rest(Ulrich 2002; Narayanan and Johnston 2008).

To examine whether the magnitude of changes in Ih afterrising levels of IFN-β is sufficient to change the resonance be-havior, we employed a 3-step approach, starting with an estab-lished mathematical model of neuronal resonance (Hutcheonet al. 1996), followed by the application of a morphologicallyrealistic model of a layer 5 neuron, and finalized by exper-imental testing of the achieved prediction. For the mathemat-ical model, we adjusted parameter values to our experimentaldata (Supplementary Experimental Procedures). We modeledthe effect of IFN-β according to the observed alterations: We1) reduced �gh to 9.9 nS, 2) increased τfast to 0.312 s, and 3)decreased pf to 0.61. This resulted in a resonance frequencyshift from 2.23 to 1.76 Hz and an increase in peak impedance(|Z|) from 121.3 to 131.2 MΩ, accompanied by a shift in Φ0

(the frequency where the phase shift plot passes zero) from1.55 to 1.16 Hz. The subsequently used morphological realis-tic model of the layer 5 neuron (Supplementary ExperimentalProcedures) showed similar results when HCN1 was specifi-cally reduced (Fig. 7A,B). Here, the resonance frequency ofthe membrane shifted from f = 4.52 to 3.3 Hz, the peak impe-dance increased from |Z| = 36 to 40 MΩ, and Φ0 decreasedfrom Φ0 = 3.0 to 1.95 Hz.

Finally, we tested the prediction of the above models inlayer 5 pyramidal neurons of acute brain slices. As predictedby our modeling, IFN-β application shifted the membrane res-onance to lower frequencies (ctrl: f = 2.3 ± 0.7 Hz, IFN-β:

f = 1.9 ± 0.7 Hz, P < 0.001, n = 9) and increased the maximalmembrane impedance (ctrl: |Z| = 117 ± 62 MΩ, IFN-β: |Z| =133 ± 70 MΩ, P < 0.05, n = 9; Fig. 7C,D). In further accordancewith the model, we observed an apparent reduction of the in-ductive phase component along with a phase shift of theinput impedance (ctrl: Φ0 = 1.3 ± 0.5 Hz, IFN-β: Φ0 = 1.1 ± 0.5Hz, P < 0.05, n = 7; Fig. 7E). The shift of the resonance par-ameters to lower values was associated with a narrower rangeof preferred intrinsic frequencies (half-band width [HBW]ctrl =3.9 ± 1.4 Hz, HBWIFN-β = 3.2 ± 0.9 Hz, n = 9, P < 0.05; Fig. 7F).The strength of the resonance (Q-value), represented by theratio of the impedance at the resonant peak to the impedanceat rest, remained unaffected (Qctrl = 1.3 ± 0.24 vs. QIFN-β = 1.3 ±0.15, n = 9, P > 0.32). On the single cell level, IFN-β led to apreference of lower frequencies.

IFN-β Slows the Cortical EEGGiven the impact of IFN-β on resonance behavior in single cer-ebral pyramidal neurons in our in vitro slice preparation andthe role of resonance in setting network activity, we hypoth-esized that IFN-β affects the oscillatory dynamics of neuronalnetworks, which appear at the cerebral cortex as EEG rhythms(Karameh et al. 2006). To test this assumption, we applied aneffectual amount of IFN-β directly to the rat cerebral cortexin vivo (Fig. 8A, Supplementary Fig. S8A). A deposit volumeof 5 µL containing 5000 IU recombinant rat IFN-β was placedright above the cortex while recording the surface EEG inawake motorically silent rats. The diffusion-driven exogenous

Figure 7. IFN-β changed the subthreshold membrane resonance of neocortical layer 5 neurons corresponding to in silico predictions based on the in vitro estimated Ih reduction.(A) In silico prediction of changes due to the Ih reduction (gray trace; reduction modeled as estimated in vitro for IFN-β) using the morphologically realistic NEURON model of alayer 5 neuron as presented in Figure 2. Ih reduction decreased the peak impedance frequency and raised the peak impedance. (B) Simulated phase shift plot before and after Ihreduction. Φ0, the frequency where the plot passes zero, shifted to lower frequencies. (C) |IFN-β (1000 IU mL−1) shifted the peak resonance of in vitro whole-cell recordedneurons and increased the maximum impedance. Top: Injected ZAP current (Supplementary Experimental Procedures). Middle: Corresponding voltage recordings from a neuronwith a distinct frequency maximum (arrow), which was markedly shifted under the influence of IFN-β (bottom). (D) Left: Respective impedance profile plot (black line) showing ashift of the resonance peak after IFN-β (gray line) from 3.62 Hz/67 MΩ to 3.14 Hz/81 MΩ (n=9). The population data reveal a marked shift of the resonance frequency in allneurons after IFN-β treatment (middle) accompanied by an increase in the impedance magnitude (right). (E) Left: Respective example of a phase shift plot. The inductivecomponent (i.e. the positive part of the plot, where the voltage leads current) decreases after IFN-β. In this experiment, Φ0 shifted from 2.2 to 1.6 Hz (n=7). Right: Quantitativecomparisons of Φ0 before and after application of IFN-β. (F) The HBW of the impedance profile narrowed after IFN-β treatment.


at Robert K



Dow

nloaded from







IFN-β application significantly slowed the cortical EEG activityby ∼0.6 Hz in the frequency range between 2 and 6 Hz(Fig. 8B–D). Furthermore, the slowing was associated with adecrease in the power decline from 2.44 ± 0.16 to 1.98 ± 0.18Hz (n = 6, P < 0.05). The EEG alterations were not due to theexperimental methodology, as normal saline without IFN-βhad no effect on the EEG power (fctrl = 2.19 ± 0.29 Hz, fsaline =2.21 ± 0.19 Hz, P = 0.72, n = 4; Supplementary Fig. S8B–D).

We repeated cortical EEGs in older mice litters (P32–40)from HCN1+/− matings. The effects of exogenous IFN-β appli-cation to cortices of HCN1+/+ mice resembled the ones in rats.Here, IFN-β reduced the power of the higher frequencies inthe cortical EEG activity, and this was associated with a de-crease in the power decline from 2.23 ± 0.62 to 1.40 ± 0.25 Hz(n = 6, P < 0.01; Fig. 9B–D, upper row) comparable with theone found in the rats. When applied to cortices of litter

Figure 8. IFN-β slows the cortical EEG in rats. (A) Schematic drawing of the experimental setting for rat EEG recordings. Arrows point to the positions of the electrodes, thestar indicates the application tube. (B) Representative traces from a P11 rat pup under control conditions (black trace) and after application of 5000 IU IFN-β (red trace). (C)Characteristic power spectrum from motorically silent periods in a rat pup under control conditions (black) and within 30 min after application of IFN-β (red). The decrease of thepower was well fitted by single exponential functions (green lines). Inset: Presentation of the frequency difference at a certain power level as a function of frequency undercontrol conditions. (D) Population data on the frequency course given as the e-fold frequency decrease show that IFN-β treatment (n= 6) led to a power decline of the EEG.

Figure 9. IFN-β–induced cortical EEG in mice is prevented by HCN1 channel deficiency. (A) Scheme of the experimental setting for mouse EEG recordings. Arrows point to thepositions of the electrodes, the star indicates the application tube. (B) Representative traces from a HCN1+/+ mouse (upper traces) and a HCN1−/− mouse (lower traces) afterapplication of saline (gray trace) and 5000 IU mouse IFN-β (red trace). (C) Characteristic power spectrum from the mice shown in B within 30 min after application of saline(gray) and of IFN-β (red). The decrease of the power was well fitted by single exponential functions (green lines). (D) Population data on the frequency course given as the e-foldfrequency decrease demonstrate IFN-β–induced power decline of the EEG in HCN1+/+ (n= 6), but not in HCN1−/− (n= 7) mice.

Cerebral Cortex 9

at Robert K



Dow

nloaded from



HCN1−/− mice, however, IFN-β did not alter the EEG (powerdecline: 1.09 ± 0.20 vs. 1.11 ± 0.23 Hz, n = 7, P = 0.49; Fig. 9B–D, lower row).

These findings suggest that IFN-β modulates spontaneousEEG slow-wave activity depending on the presence of HCN1and can reversibly alter the physiological responses of corticalneuronal networks.

Discussion

The present study demonstrates that Ih, a major component ofintrinsic neuronal excitability, is instrumental in mediatingtype I IFN-induced changes in neuronal excitability. Extendedtype I IFN presence as after in vivo induction of IFNs in theCNS by Theiler’s virus infection led to an Ih reduction. Also,the acute application of type I IFNs rapidly reduces rodent Ihby about one-fourth. The effect is attributable to a modifi-cation of HCN channels, because, first, after blocking Ih nochange in current amplitude upon IFN application could beobserved, and secondly, synaptic influences or other voltagesensitive currents were excluded by pharmacologically isolat-ing Ih. Based on modeling results and measurements inHCN1−/− mice, we conclude that the effect is predominantlymediated by the fast activating HCN subunit, HCN1. Datafrom primary cultures and the detection of functionally intactIFNAR in neurons revealed that IFN-β directly acts onneurons, that is, without glial mediation.

Interactions of the Signaling Cascades of IFNARwith HCN ChannelsClassical signaling in response to IFNAR activation includes anumber of phosphorylation steps. Here, we show that corticalneurons possess an essential component of the IFN-signalingcascade, functional IFNAR. Furthermore, the effect of IFN-βon HCN channels requires the activation of the JAK/STATpathway since IFN-β did not interact directly with HCN chan-nels and was ineffective after disrupting the signaling cascade.

However, the connection between IFNAR activation andHCN channels remains to be demonstrated. One possible linkis the p38 MAP kinase, as it can be activated by IFNARs(Takaoka and Yanai 2006) and exerts a direct influence on Ih(Poolos et al. 2006). Nevertheless, 2 lines of evidence in ourexperiments argue against such an interaction: First, IFN-βfailed to phosphorylate neuronal p38 MAP kinase in corticalneurons. This also excludes a contribution of MAPkinase-induced arachidonic acid to be involved in the de-crease of maximal current (Fogle et al. 2007). Secondly, IFN-βdoes not cause a hyperpolarization in the voltage dependenceof Ih activation, whereas an upregulation of p38 MAP kinaseresulted in an ∼11-mV depolarization of in hippocampal pyra-midal neurons (Poolos et al. 2006). Due to the lack of a shiftof V1/2, we also exclude an IFN-β/HCN channel interactionthrough allosteric regulators of voltage dependence of acti-vation such as cAMP, H+, 4,5-PIP2, or signaling lipids (Fogleet al. 2007). A sole protein–protein interaction (i.e. exclusivephosphorylation) appears also unlikely because of the some-what delayed onset of the effect. Further, the changes of neur-onal properties such as input resistance and membrane timeconstant, which may be linked to HCN channels, were depen-dent on protein synthesis (Beyer et al. 2009). Therefore,future research on the exact molecular mechanism may be

focused on proteins involved in both the IFNAR signalingpathway and HCN modulation and newly transcribed uponIFNAR activation.

The IFN-β Modulation of Ih is Sufficient to CauseAlterations in Functional Properties of CorticalPyramidal NeuronsIh is involved in the regulation of neuronal excitability particu-larly due to its partially open state at resting membrane poten-tial (Wahl-Schott and Biel 2009). Accordingly, the inputresistance was markedly increased by IFN-β in neocorticalneurons in brain slices (Hadjilambreva et al. 2005) and inprimary cultured cortical neurons (this study). Given the dis-tance dependence of the HCN channel density in dendrites ofpyramidal neurons (Magee 1998; Kole et al. 2006), an inhi-bition of HCN channels might lead to an even greater augmen-tation in dendritic excitability by decreasing the local restingconductance and increasing the summation of excitatory post-synaptic potentials (Magee 1998; Huang et al. 2009). Theirlocal action in dendrites remains to be tested. Of importancefor dendritic integration and neuronal computation is thefrequency-dependent response of neurons at and belowresting membrane potential, for which Ih is the main respon-sible active current (Ulrich 2002; Narayanan and Johnston2008). Eventually, resonance describes the frequency depen-dence of impedance and the band-pass filter characteristics ofdendrites for incoming signals (Ulrich 2002). Furthermore, theaccompanying phase shift endows neurons to scale the arrivaltimes of signals on the soma (Narayanan and Johnston 2008).Our data suggest that layer 5 neurons favor responses to lowerfrequency inputs when exposed to elevations in IFN-β. Takentogether, these changes point to a significant modulation ofneuronal excitability by IFN-β–induced Ih reduction.

Connection between Ih Reduction and EEG SlowingA number of points suggest that Ih modulation considerablycontributed to observed changes in the EEG. As strongest ar-gument, we regard the lack of an IFN-β effect on the corticalEEG of HCN1−/− mice (this study). Further, HCN channelsexert a global control of neuronal network rhythms(Wahl-Schott and Biel 2009), and the appearance of aberrantEEG activity and seizures is associated with HCN channelreduction (Strauss et al. 2004; Kole et al. 2007; Marcelin et al.2009). We observed a reduced EEG power in the frequencyrange between 1 and 8 Hz associated with IFN-β, as it was alsoreported for Ih inhibition in frontal lobe epilepsy (Marcelinet al. 2009). Karameh et al. (2006) suggested with a modelingapproach that cortical alpha rhythms depend on intrinsic cur-rents in layer 5 cells, namely Ih and T-type calcium current.Their study also predicts a pronounced shift to the delta rangeupon Ih blockade. Furthermore, the sensitivity to synchronizedsynaptic inputs is promoted by hyperpolarization (Carr et al.2007) as it is triggered by IFN-β (this study). Interestingly, EEGchanges have long been recognized after treatment with IFN-α(Dafny 1983; Birmanns et al. 1990; Kamei et al. 2005), theother type I IFN that also activates signaling via IFNAR.

Pathophysiological RelevanceIn the process of inflammation, IFN-β levels are dynamic anddependent on the local environment. Therefore, the concen-tration used in this study might only represent one within the


at Robert K



Dow

nloaded from


range of local IFN-β concentrations triggered by inflammation.At present, the precise extracellular type I IFN protein levelsthat might be expected during a CNS inflammation areunknown, although there are some hints. For example, neur-onal IFN-β production in viral infection has been investigated inmice by RT-PCR providing quantitative values of IFN-mRNAproduction (Delhaye et al. 2006), and a time line of nonspeci-fied IFN tissue content was studied by a 3H-uridine-based assay(Heremans et al. 1980). Quantitative protein analysis of spinalcord homogenate during experimental autoimmune encephalo-myelitis, an animal model of multiple sclerosis, showed thatIFN-β is expressed at markedly higher amounts in the CNS thanin the periphery (Prinz et al. 2008). Given that in inflammationcytokines generally act together, they may combine theirimpact in vivo, in particular concerning the type I IFNs acting atthe IFNAR. Due to the local production (Delhaye et al. 2006),IFNs may activate their respective receptors, even if the tissueconcentrations are still quite low. By utilizing the CNS inflam-mation caused by the neurovirulent GDVII strain of TMEV atthe height of type I IFN production (Delhaye et al. 2006), wehere showed that viral IFN induction mimic the amplitudereduction in Ih. This supports a pathophysiological role of thetype I IFN neuron interaction in CNS inflammation.

Phenomenological similarities between Ih reduction andinflammation provide indirect evidence for a link betweenboth. Pharmacological HCN1 reduction with compounds suchas ketamine or several volatile anesthetics (Chen, Shu, Baylisset al. 2009; Chen, Shu, Kennedy et al. 2009) produces aninflammation-like deteriorated mental state, including dis-turbed vigilance. Likewise, subtle modulation of HCN channelactivity, as observed with physiological changes of cAMPlevels, contributes substantially to altered network activitycorrelated with behavioral states (Wahl-Schott and Biel 2009).In human inflammation states, such functional changes mayassociate with sickness behavior and an increased suscepti-bility to depression (Amodio et al. 2005). It was suggestedthat certain cytokines play a causal role in the genesis of psy-chosocial alterations (Kent et al. 1992; Dantzer et al. 2008)and that IFN-α therapy can produce such side effects and maycontribute to cytokine-related subtypes of affective disorders(Anisman et al. 2008; Dantzer et al. 2008).

Ih modulation by IFN-β may even be of broader importanceunder physiological conditions given the basic IFN-β level re-cently reported in the uninfected and noninflamed CNS (Prinzet al. 2008). This would imply a physiological neuromodu-latory role of IFN-β.

In summary, our data imply a major role of IFNs in alteringthe neuronal state during inflammation. This put IFNs in linewith previously recognized neuromodulatory cytokines suchas IL-2, IL-1β, IL-6, and TNF-α (Mendoza-Fernandez et al.2000; Dantzer et al. 2008). Taking the contribution of Ih forthe proper function of neuronal cells in the nervous system(central and peripheral) into account, our experimental dataopen up new fields of investigation. These results might shedlight on the involvement of Ih alterations in the adaptive pro-cesses during acute and chronic neuronal inflammation (Kentet al. 1992; Johnson 2002). Of particular interest will be todetermine whether IFNs counteract the frequency responsetuning capabilities of neurons, or if they act as a natural pro-tector of brain tissue from inflammation (Prinz et al. 2008) byproviding an auxiliary mechanism to adapt neuronal compu-tation to the state of inflammation.

Supplementary MaterialSupplementary material can be found at: http://www.cercor.oxford-journals.org/.

Funding

This study, in particular the work of K.S. and U.S., was sup-ported by the German Research Foundation (DFG STR865/3-1) and the DAAD/Go8 program (U.S. and M.H.P.K.). Someof the equipment used was donated by the Sonnenfeld-Stiftung, Berlin (A.U.B. and U.S.).

NotesWe thank Bettina Brokowski, Carla Strauss, and Rike Dannenberg forexpert technical assistance, Shigetada Nakanishi for the donation ofrHCN2, and Thomas Michiels for advice on the induction of the viralencephalomyelitis and for providing TMEV GDVII. We also thankArndt Rolfs for providing us with laboratory space, consumables, andequipment in the initial phase.

ReferencesAmodio P, De Toni EN, Cavalletto L, Mapelli D, Bernardinello E,

Del Piccolo F, Bergamelli C, Costanzo R, Bergamaschi F, Poma SZ.2005. Mood, cognition and EEG changes during interferon alpha(alpha-IFN) treatment for chronic hepatitis C. J Affect Disord.84:93–98.

Anisman H, Merali Z, Hayley S. 2008. Neurotransmitter, peptide andcytokine processes in relation to depressive disorder: comorbiditybetween depression and neurodegenerative disorders. Prog Neu-robiol. 85:1–74.

Beyer S, Raether G, Stadler K, Hoffrogge R, Scharf C, Rolfs A, Mix E,Strauss U. 2009. Interferon-beta modulates protein synthesis in thecentral nervous system. J Neuroimmunol. 213:31–38.

Birmanns B, Saphier D, Abramsky O. 1990. Alpha-interferon modifiescortical EEG activity: dose-dependence and antagonism by nalox-one. J Neurol Sci. 100:22–26.

Carr DB, Andrews GD, Glen WB, Lavin A. 2007. alpha2-Noradrenergic receptors activation enhances excitability and sy-naptic integration in rat prefrontal cortex pyramidal neurons viainhibition of HCN currents. J Physiol. 584:437–450.

Chen X, Shu S, Bayliss DA. 2009. HCN1 channel subunits are a mol-ecular substrate for hypnotic actions of ketamine. J Neurosci.29:600–609.

Chen X, Shu S, Kennedy DP, Willcox SC, Bayliss DA. 2009. Subunit-specific effects of isoflurane on neuronal Ih in HCN1 knockoutmice. J Neurophysiol. 101:129–140.

Dafny N. 1983. Interferon modifies morphine withdrawal phenomenain rodents. Neuropharmacology. 22:647–651.

Dafny N, Prieto-Gomez B, Dong WQ, Reyes-Vazquez C. 1996. Inter-feron modulates neuronal activity recorded from the hypothala-mus, thalamus, hippocampus, amygdala and the somatosensorycortex. Brain Res. 734:269–274.

Dantzer R, O’Connor JC, Freund GG, Johnson RW, Kelley KW. 2008.From inflammation to sickness and depression: when the immunesystem subjugates the brain. Nat Rev Neurosci. 9:46–56.

Delhaye S, Paul S, Blakqori G, Minet M, Weber F, Staeheli P,Michiels T. 2006. Neurons produce type I interferon during viralencephalitis. Proc Natl Acad Sci USA. 103:7835–7840.

Detje CN, Meyer T, Schmidt H, Kreuz D, Rose JK, Bechmann I,Prinz M, Kalinke U. 2009. Local type I IFN receptor signaling pro-tects against virus spread within the central nervous system.J Immunol. 182:2297–2304.

Fogle KJ, Lyashchenko AK, Turbendian HK, Tibbs GR. 2007. HCNpacemaker channel activation is controlled by acidic lipids down-stream of diacylglycerol kinase and phospholipase A2. J Neurosci.27:2802–2814.

Cerebral Cortex 11

at Robert K



Dow

nloaded from





Frere SG, Kuisle M, Luthi A. 2004. Regulation of recombinant andnative hyperpolarization-activated cation channels. Mol Neurobiol.30:279–305.

Hadjilambreva G, Mix E, Rolfs A, Muller J, Strauss U. 2005. Neuro-modulation by a cytokine: interferon-beta differentially augmentsneocortical neuronal activity and excitability. J Neurophysiol.93:843–852.

Heremans H, Billiau A, De Somer P. 1980. Interferon in experimentalviral infections in mice: tissue interferon levels resulting from thevirus infection and from exogenous interferon therapy. InfectImmun. 30:513–522.

Hoffmann O, Braun JS, Becker D, Halle A, Freyer D, Dagand E,Lehnardt S, Weber JR. 2007. TLR2 mediates neuroinflammationand neuronal damage. J Immunol. 178:6476–6481.

Huang Z, Walker MC, Shah MM. 2009. Loss of dendritic HCN1subunits enhances cortical excitability and epileptogenesis.J Neurosci. 29:10979–10988.

Hutcheon B, Miura RM, Puil E. 1996. Models of subthreshold mem-brane resonance in neocortical neurons. J Neurophysiol.76:698–714.

Johnson RW. 2002. The concept of sickness behavior: a brief chrono-logical account of four key discoveries. Vet Immunol Immuno-pathol. 87:443–450.

Kamei S, Oga K, Matsuura M, Tanaka N, Kojima T, Arakawa Y, Matsu-kawa Y, Mizutani T, Sakai T, Ohkubo H. 2005. Correlationbetween quantitative-EEG alterations and age in patientswith interferon-alpha-treated hepatitis C. J Clin Neurophysiol.22:49–52.

Karameh FN, Dahleh MA, Brown EN, Massaquoi SG. 2006. Modelingthe contribution of lamina 5 neuronal and network dynamics tolow frequency EEG phenomena. Biol Cybern. 95:289–310.

Kent S, Bluthe RM, Kelley KW, Dantzer R. 1992. Sickness behavior asa new target for drug development. Trends Pharmacol Sci.13:24–28.

Kole MH, Brauer AU, Stuart GJ. 2007. Inherited cortical HCN1channel loss amplifies dendritic calcium electrogenesis andburst firing in a rat absence epilepsy model. J Physiol.578:507–525.

Kole MH, Hallermann S, Stuart GJ. 2006. Single Ih channels in pyra-midal neuron dendrites: properties, distribution, and impact onaction potential output. J Neurosci. 26:1677–1687.

Linden DR, Sharkey KA, Mawe GM. 2003. Enhanced excitability ofmyenteric AH neurones in the inflamed guinea-pig distal colon.J Physiol. 547:589–601.

Magee JC. 1998. Dendritic hyperpolarization-activated currentsmodify the integrative properties of hippocampal CA1 pyramidalneurons. J Neurosci. 18:7613–7624.

Marcelin B, Chauviere L, Becker A, Migliore M, Esclapez M,Bernard C. 2009. h channel-dependent deficit of theta oscillationresonance and phase shift in temporal lobe epilepsy. NeurobiolDis. 33:436–447.

Mendoza-Fernandez V, Andrew RD, Barajas-Lopez C. 2000.Interferon-alpha inhibits long-term potentiation and unmasks along-term depression in the rat hippocampus. Brain Res.885:14–24.

Meydan N, Grunberger T, Dadi H, Shahar M, Arpaia E, Lapidot Z,Leeder JS, Freedman M, Cohen A, Gazit A. 1996. Inhibition ofacute lymphoblastic leukaemia by a Jak-2 inhibitor. Nature.379:645–648.

Narayanan R, Johnston D. 2008. The h channel mediates location de-pendence and plasticity of intrinsic phase response in rat hippo-campal neurons. J Neurosci. 28:5846–5860.

Nolan MF, Malleret G, Lee KH, Gibbs E, Dudman JT, Santoro B, YinD, Thompson RF, Siegelbaum SA, Kandel ER et al. 2003. Thehyperpolarization-activated HCN1 channel is important for motorlearning and neuronal integration by cerebellar Purkinje cells.Cell. 115:551–564.

Novick D, Cohen B, Rubinstein M. 1994. The human interferonalpha/beta receptor: characterization and molecular cloning. Cell.77:391–400.

Paul S, Ricour C, Sommereyns C, Sorgeloos F, Michiels T. 2007. TypeI interferon response in the central nervous system. Biochimie.89:770–778.

Pestka S. 2007. The interferons: 50 years after their discovery, there ismuch more to learn. J Biol Chem. 282:20047–20051.

Poolos NP, Bullis JB, Roth MK. 2006. Modulation of h-channels inhippocampal pyramidal neurons by p38 mitogen-activated proteinkinase. J Neurosci. 26:7995–8003.

Prinz M, Schmidt H, Mildner A, Knobeloch KP, Hanisch UK, Raasch J,Merkler D, Detje C, Gutcher I, Mages J. 2008. Distinct and nonre-dundant in vivo functions of IFNAR on myeloid cells limit autoim-munity in the central nervous system. Immunity. 28:675–686.

Santoro B, Liu DT, Yao H, Bartsch D, Kandel ER, Siegelbaum SA,Tibbs GR. 1998. Identification of a gene encoding a hyperpolariz-ation-activated pacemaker channel of brain. Cell. 93:717–729.

Schuchmann S, Schmitz D, Rivera C, Vanhatalo S, Salmen B, MackieK, Sipila ST, Voipio J, Kaila K. 2006. Experimental febrile seizuresare precipitated by a hyperthermia-induced respiratory alkalosis.Nat Med. 12:817–823.

Strauss U, Kole MH, Brauer AU, Pahnke J, Bajorat R, Rolfs A, NitschR, Deisz RA. 2004. An impaired neocortical Ih is associated withenhanced excitability and absence epilepsy. Eur J Neurosci.19:3048–3058.

Stuart G, Spruston N. 1998. Determinants of voltage attenuation inneocortical pyramidal neuron dendrites. J Neurosci.18:3501–3510.

Takaoka A, Yanai H. 2006. Interferon signalling network in innatedefence. Cell Microbiol. 8:907–922.

Thompson JE, Cubbon RM, Cummings RT, Wicker LS, Frankshun R,Cunningham BR, Cameron PM, Meinke PT, Liverton N, Weng Yet al. 2002. Photochemical preparation of a pyridone containingtetracycle: a Jak protein kinase inhibitor. Bioorg Med Chem Lett.12:1219–1223.

Ulrich D. 2002. Dendritic resonance in rat neocortical pyramidal cells.J Neurophysiol. 87:2753–2759.

Wahl-Schott C, Biel M. 2009. HCN channels: structure, cellular regu-lation and physiological function. Cell Mol Life Sci. 66:470–494.

Zhang ZW. 2004. Maturation of layer V pyramidal neurons in the ratprefrontal cortex: intrinsic properties and synaptic function. J Neu-rophysiol. 91:1171–1182.


at Robert K



Dow

nloaded from


elevation in type i interferons inhibits hcn1 and slows cortical neuronal oscillations

Documents