G Model
BCP-12251; No. of Pages 6 Biochemical Pharmacology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm
Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia Kathy L. Kohlhaas *, Holly M. Robb, Victoria A. Roderwald, Lynne E. Rueter AbbVie, Inc., 1 N. Waukegan Road, North Chicago, IL 60064-6115, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 March 2015 Accepted 21 May 2015 Available online xxx
Schizophrenia is a chronic disease that has been hypothesized to be linked to neurodevelopmental abnormalities. Schizophrenia patients exhibit impairments in basic sensory processing including sensory gating deficits in P50 and mismatch negativity (MMN). Neuronal nicotinic acetylcholine receptor (nAChR) agonists have been reported to attenuate these deficits. Gestational exposure of rats to methylazoxymethanol acetate (MAM) at embryonic day 17 leads to developmental disruption of the limbic-cortical system. MAM exposed offspring show neuropathological and behavioral changes that have similarities with those seen in schizophrenia. In this study, we aimed to assess whether N40 auditory sensory gating (the rodent form of P50 gating) and MMN deficits as measures of auditory evoked potential (AEP) electroencephalography (EEG) are present in MAM rats and whether nAChR agonists could attend the deficit. E17 male MAM and sham rats were implanted with cortical electrodes at 2 months of age. EEG recordings evaluating N40 gating and MMN paradigms were done comparing effects of vehicle (saline), nicotine and the a7 agonist ABT-107. Deficits were seen for MAM rats compared to sham animals in both N40 auditory sensory gating and MMN AEP recordings. There was a strong trend for N40 deficits to be attenuated by both nicotine (0.16 mg/kg i.p. base) and ABT-107 (1.0 mg/kg i.p. base). MMN deficits were significantly attenuated by ABT-107 but not by nicotine. These data support the MAM model as a useful tool for translating pharmacodynamic effects in clinical medicine studies of novel therapeutic treatments for schizophrenia. ß 2015 Elsevier Inc. All rights reserved.
Keywords: EEG Mismatch negativity Sensory gating MAM Nicotine a7nAChR
1. Introduction Schizophrenia is a chronic disease that is linked to neurodevelopmental abnormalities [1,2]. Schizophrenia patients exhibit impairments in basic sensory processing and higher cognitive functions, such as language, reasoning, and planning. There is an ever growing interest in the development of animal models that represent impairments found in schizophrenia. Gestational exposure of rats to methylazoxymethanol acetate (MAM) at embryonic day 17 leads to developmental disruption of the limbic-cortical system [3–5]. MAM exposed offspring show neuropathological and behavioral changes including impairment of cognitive tasks such as spatial working memory, attentional set-shifting, and reversal learning that have similarities with those seen in schizophrenia
* Corresponding author. Tel.: +1 847 937 4066. E-mail addresses: [email protected] (K.L. Kohlhaas), [email protected] (H.M. Robb), [email protected] (V.A. Roderwald), [email protected] (L.E. Rueter).
[3,6]. Electroencephalography (EEG) and imaging studies done in MAM rats suggest abnormalities in structure and synchronized oscillatory activity that parallel disruptions of frontotemporal connectivity in schizophrenia [5,7–10]. Similar to schizophrenia patients, MAM rats also demonstrate a deficit in prepulse inhibition (PPI) of the startle response, a form of sensorimotor gating. To date, whether MAM rats also show other forms of sensory gating deficits typically seen in schizophrenia patients such as P50 and mismatch negativity has not been studied. Sensory gating describes neurological processes for filtering out redundant or unnecessary stimuli in the brain from all possible environmental stimuli [11] by preventing an overload of irrelevant information in the higher cortical centers of the brain. Although sensory gating is largely automatic, it also occurs within the context of attentional processes. Sensory gating is mediated by a network in the brain which involves the auditory cortex, prefrontal cortex and hippocampus. Other areas of the brain associated with sensory gating include the amygdala, striatum, medial prefrontal cortex, and the GABAergic neurons of the midbrain dopamine cell region [12]. Using a paired-click paradigm is a common
http://dx.doi.org/10.1016/j.bcp.2015.05.011 0006-2952/ß 2015 Elsevier Inc. All rights reserved.
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 2
K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
non-invasive technique used to measure sensory gating, a type of event-related potential (ERP). For normal sensory gating, if a person hears a pair of clicks within 500 ms of one another, the person will gate out the second click because it is perceived as being redundant. In humans, evidence of the gating can be seen in the P50 wave, occurring in the brain 50 ms after the click. Low values of the P50 wave indicate that sensory gating has occurred. High values of the P50 wave indicate a lack of sensory gating [11]. Individuals with schizophrenia only reduce the amplitude of the second stimulus by 10–20%, whereas individuals without schizophrenia reduce the amplitude of second stimulus by 80–90%. The P50 auditory gating deficit is one of the best established endophenotypes associated with schizophrenia. Most antipsychotics used clinically do not normalize P50 gating deficits in schizophrenia patients, except clozapine [13,14]. In rodents, a similar paradigm is measured in EEG denoted by amplitude differences between the P20 and N40 waves [15]. Cognitive decline associated with schizophrenia can be indexed by temporofrontal functional deficiency expressed as deficient auditory discrimination and orienting [16]. This functional deficiency can be objectively assessed by means of an electrophysiological measure known as mismatch negativity (MMN). MMN is presumed to reflect the existence of a short-term memory trace of the standard stimulus at the moment of presentation of a deviant stimulus. The MMN is considered an automatic (pre-attentive) response. It can be recorded in the absence of the subject’s attention to auditory stimuli and even in unconscious comatose patients [17]. Since MMN is a change–detection response of the brain that can be elicited in the absence of conscious attention or behavioral task, an auditory ERP wave is generated for the responses to both the ‘standard’ and ‘deviant’ stimuli. As frequencies become divergent, a normal individual can detect the differences between the standard and the deviant. ERPs identify the mismatch. In the human auditory cortex, a detectable change (deviant stimulus) in the physical parameters of a repeated (standard) stimulus evokes a negative component in the auditory ERP within a 100–200 ms time window from stimulus onset. MMN has traditionally been recorded in the so-called auditory odd-ball paradigm involving randomized presentation of repetitive standard sounds and rare occasional deviant sounds. MMN is best visualized in a subtractive or difference waveform. Attenuated MMN amplitudes are a robust finding in patients with schizophrenia and studies done with either typical or atypical antipsychotic medications do not abolish MMN abnormality in patients with schizophrenia [18–21]. MMN has been demonstrated to be recorded in humans along with various animals, including rats [22,23]. Activation of nicotinic acetylcholine receptors (nAChRs) has been shown to improve attention, learning, and memory. Several subtypes of nAChRs are expressed in the mammalian brain, each with distinct physiological and pharmacological properties. Nicotine, which acts as an agonist at multiple nAChR subtypes, can normalize temporal aspects of sensory processing in patients with schizophrenia. Nicotine has also been demonstrated to normalize P50 [24] and MMN amplitude [25,26] in normal healthy subjects. A series of human and animal investigations has suggested that altered expression and function of the a7 nAChR may be responsible for the P50 auditory sensory gating deficit characterized in schizophrenia patients and their relatives [26,27]. a7 nAChRs are widely expressed in the hippocampus, an area of the brain closely associated with memory [28]. a7 nAChR agonists have been shown to improve performance in preclinical learning and memory tasks including pre-attention as measured by sensory gating [29–32]. They have also been shown to improve attention, working and episodic memory in humans. DMXB-A, a partial a7 nAChR agonist, demonstrated significant
improvements in P50 inhibition in schizophrenia patients [33]. Recently, it has also been demonstrated that EVP-6124, a selective and potent a7 nAChR agonist can attenuate P50 and MMN deficits in schizophrenia patients [34]. In this study we evaluate the measures of N40 and MMN in MAM versus control rats and the effects of nicotine and the selective a7 nAChR agonist ABT-107. 2. Methods 2.1. Animals Sprague Dawley timed pregnant female rats were received at E12 from Charles River Laboratories (Raleigh, NC). On E17 pregnant dams were injected intraperitoneally (i.p.) with either sterile phosphate buffered saline (PBS) or a 25 mg/kg methylazoxymethanol acetate (MAM). Pups were born 5–6 days later. On PD21, pups were weaned and separated by gender and litter into groups of 3–5 rats. Rat pups were maintained in group housing on a 12 h light:dark cycle (home cage and lab lights on 06:00–18:00; off 18:00–06:00) with food and water ad libitum. Pups were allowed to mature until they were 2 months old (PD60) before any behavior evaluation was completed. Based upon knowledge from within our lab, the MAM deficit phenotype can be characterized by spontaneous locomotor hyperactivity in early adulthood and postmortem brain weight. Behavioral profiling identified litters to have a good MAM effect prior to assignment for EEG surgery (data not shown). Male rats from two litters for an inclusive total of eight rats each of MAM exposed or control (PBS exposed) groups were surgically implanted with electrodes for recording EEG signals from both frontal and parietal cortices. 2.2. Surgical procedure Under isoflurane anesthesia, male rats were implanted with stainless steel recording electrodes in the frontal (AP + 3.0; ML 2.0) and parietal (AP 3.0; ML 4.0) cortices. Reference electrodes were implanted over the frontal sinus (AP + 6.0; ML + 1.0) and cerebellum. Recording electrodes were grounded to cerebellar reference. Peri-operatively, rats received 2.5 mg/kg subcutaneously (s.c.) flunixin meglumine (QD dosing; Banamine-S, Merck Animal Health, NJ). Flunixin was given for 2 additional days post-operatively. After surgery, the rats were single housed, maintained on a normal light:dark cycle and on a restricted diet to control weight (4 pellets/day). 2.3. EEG recording EEG recordings did not begin for 2 week post-surgery to allow sufficient recovery from the surgical procedure. Rats were habituated to recording chambers for 1 week before start of studies. During habituation, spontaneous EEG was recorded for 2 h periods. EEG was recorded between 9:00 a.m. and 3:00 p.m. Freely moving rats were tethered in sound attenuating chambers to electrical swivels to convey EEG biosignals (Plastics One, Roanoke, VA, USA). EEG biosignals were amplified with differential AC amplifiers (A-M Systems, Sequim, WA, USA) and filtered at 1 and 100 Hz. The same rats were tested in both N40 and MMN paradigms. Rats were administered with either vehicle or investigational compound(s) in a within subject randomized Latin square design, allowing for at least 3 days wash out between test sessions. N40 sensory gating was presented as paired auditory white noise stimuli (500 ms interstimulus interval) every 15 s for a total of 120 trials. Mismatch negativity (MMN) potentials were generated by presenting auditory stimuli every 500 ms using an odd-ball paradigm where there were no successive repeats of the deviant stimulus. The MMN sound stimulus set was a white-noise
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
frequency range (standard = 100–5000 Hz, deviant = 600– 9000 Hz), with a deviant probability of 0.20. For each study day, MMN was run first, followed by a session of standard stimuli and deviant stimuli alone. For N40 and MMN studies, rats were injected with test compound and allowed 30 min before start of session. All studies were coded for drug treatment and un-coded after data sets were analyzed. EEG was acquired and quantified using SciWorks (Datawave Technologies, Loveland, CO, USA).
P20
A
P20
CS
25 µv
2.4. Chemicals
2.5. Statistical analysis An a priori t-test was done to compare control (PBS treated) rats to MAM rats for both N40 and MMN analysis to define whether there was indeed an effect due to condition. Statistical analysis used for N40 and MMN during compound treatment was a twoway repeated measure ANOVA. Post hoc analysis included a Bonferroni’s multiple comparison.
B
*
0.4 0.3 0.2 0.1 0.0
PBS-control
MAM
condition
C
0.6
0.4
T:C ratio
3.1. N40 auditory gating in control and MAM rats
Fig. 2A and B shows the waveform generated by running the MMN paradigm in both control (PBS treated) and MAM rats. The difference waveform between the standard and deviant evoked response potential (ERP) was generated by subtracting the deviant wave from the standard wave. Only values for standards that were immediately preceding deviants were analyzed. MMN was measured by calculating absolute values for area above or below zero reference for the difference waves. Difference waves that lie at or around zero indicate a deficit in MMN, compared to normal in which there is measurable area above or below the zero reference (Fig. 2A and B). Because the window for MMN has not been clearly defined in rats, waveform data were analyzed for AUC at three latency windows; 20–50 ms, 50–180 ms, and 20–180 ms. The windows were defined by the approximate times when the
PBS MAM
N40
0.5
0.5
3.2. Mismatched negativity in control and MAM rats
N40
0.6
3. Results
Fig. 1A shows the grand average traces for N40 recording comparing control (PBS treated) to MAM rats. N40 values were measured by comparing the peak amplitudes from P20 to N40 of test stimulus (TS) to those elicited by condition stimulus (CS). Amplitudes were measured from the response to the first stimulus (condition amplitude, CAMP) and second stimulus (test stimulus, TAMP). Sensory gating was measured by calculating a T:C ratio (TAMP/CAMP). Ratios closer to 1 were considered a deficit, whereas ratios less than 0.5 were considered normal gating. An a priori comparison of control (PBS-treated) rats dosed with saline to MAM rats also dosed with saline show a significant difference in N40 gating [F(2, 16) = 1.114, p = 0.045 by unpaired Student’s t-test] (Fig. 1B). A two-way repeated measure ANOVA for rats treated with either saline, nicotine (0.16 mg/kg i.p.) or ABT-107 (0.32, 1.0 mg/kg i.p.) showed no significant interaction of condition and treatment [F(3, 42) = 1.844, p = 0.1539]. The effect of treatment were not quite significant [F(3, 42) = 2.64, p = 0.0618] (Fig. 1C).
TS
500 msec
T:C ratio
Nicotine hydrogen tartrate salt was purchased from Sigma Aldrich (St Louis, MO, USA). ABT-107, [5-(6-[(3R)-1-azabicyclo[2.2.2]oct-3-yloxy]pyridazin-3-yl)-1H-indole] was synthesized at AbbVie, Incorporated. Compound solutions were prepared in saline. Compounds were administered intraperitoneally in solution in a volume of 1.0 ml/kg body weight for rats and reported as free base weight.
3
saline nicotine 0.32 ABT 1.0 ABT
0.3 0.2 0.1 0.0
PBS-control
MAM
condition Fig. 1. (A) Grand average traces for the evoked response to paired auditory stimuli (ISI = 500 ms; CS = conditioning stimulus, TS = test stimulus) recorded from frontal cortex surface electrodes in MAM and control (PBS) rats. (B) T:C ratio for P20–N40 amplitudes from PBS control rats vs. MAM rats. Data represent mean SEM, n = 8. F(2, 16) = 1.114, *p = 0.045 by unpaired Student’s t-test. (C) Effects of nicotine (0.16 mg/kg) and ABT-107 (0.32 and 1.0 mg/kg) on T:C ratios for P20–N40 amplitudes from PBS control and MAM rats. Data represent mean SEM, n = 7–8. Two-way repeated measures ANOVA show a non-significant trend of effect for treatment. F(3, 42) = 2.64, p = 0.0618.
standard wave crosses the zero reference. An a priori comparison of control (PBS-treated) rats dosed with saline to MAM rats also dosed with saline show a significant difference in MMN. During the 20–50 ms window, MAM rats have a significant decrease in AUC [F(2, 14) = 1.509, p = 0.0189 by unpaired Student’s t-test] (Fig. 3A). During the 50–180 ms window, MAM rats have a significant decrease in AUC [F(2, 14) = 1.727, p = 0.0276 by unpaired Student’s t-test] (Fig. 3B). During the 20–180 ms window, MAM rats have a non-significant decrease in AUC [F(2, 14) = 3.179, p = 0.0651 by unpaired Student’s t-test] (Fig. 3C). Two-way repeated measures ANOVA analysis of MMN for rats treated with either saline, nicotine (0.16 mg/kg i.p.), or ABT-107 (1.0 mg/kg i.p.) during 20–50 ms latency revealed a significant interaction of condition and treatment [F(2, 42) = 5.954, p = 0.0079]. The effects of treatment were not significant [F(2, 42) = 2.649, p = 0.0913] (Fig. 3D). Two-way repeated measures ANOVA analysis of MMN for rats treated with either saline,
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
4
A
standard deviant difference
5 µV 50 msec
B
standard deviant difference
5 µV 50 msec Fig. 2. Grand average traces for waveform of MMN paradigm in control (PBS treated) (A) and MAM (B) rats for standard (frequency range = 100–5000 Hz (dashed line)), deviant (frequency range = 600–9000 Hz (light gray line)); and difference wave (black line). Deviant probability = 0.20. Waveforms indicate deficit in MMN detection in MAM treated animals as denoted by difference wave being lying at or around zero reference.
nicotine (0.16 mg/kg i.p.), or ABT-107 (1.0 mg/kg i.p.) during 50–180 ms latency revealed a non-significant trend for interaction of condition and treatment [F(2, 42) = 3.056, p = 0.0657]. The effects of treatment were not significant [F(2, 42) = 1.482, p = 0.2472] (Fig. 3E). Two-way repeated measures ANOVA analysis of MMN for rats treated with either saline, nicotine (0.16 mg/kg i.p.), or ABT-107 (1.0 mg/kg i.p.) during 20–180 ms latency revealed a significant interaction of condition and treatment [F(2, 42) = 5.072, p = 0.0146]. The effects of treatment were not significant [F(2, 42) = 2.860, p = 0.0769] (Fig. 3F). Post hoc analyses using Bonferroni’s multiple comparison indicate a significant effect of ABT-107 versus saline on MMN recorded in MAM rats during 20–180 ms window. 4. Discussion In this investigation, we compared frontal cortical EEG recordings in cohorts of MAM-exposed adult rats and sham (PBS-exposed) control rats. In particular we were interested in whether we could record ERPs using auditory-evoked N40 and MMN paradigms in awake, freely moving rats implanted with surface electrodes. Most often, N40 is recorded using electrodes placed in the hippocampus or the auditory cortex of anesthetized animals. In our instance, we did not want to use these preclinical investigational methods, but rather wanted to establish whether a more translational measure could be used in rodents for both N40 and MMN measures. Thus, all of the EEG recordings were captured using surface electrodes over frontal cortex. N40 was tested first, since using the paired-click auditory paradigm would additionally indicate that the rats had normal hearing. For MMN, we used a white-noise frequency range for our standard and deviant stimuli
rather than a single frequency or tone to allow for differing hearing spectrum in rodents. Our initial objective was to determine whether there was face/ construct validity to MAM rats as a preclinical model of schizophrenia-like symptoms. Key to this objective would be whether there was a measurable difference in N40 and/or MMN in the MAM-exposed rats compared to controls. With a group size of 7–8 rats per condition, we were indeed able to show a significant difference in N40 signaling (Fig. 1B), with a deficit being driven by the responses to both the condition and the test stimuli (Fig. 1A). We were also able to show significant differences in the MMN difference waves (Figs. 2 and 3A–C). As little information is available on MMN measures from rats and we could only identify one other publication involving MMN recording from awake, freemoving animals [35], it was not clear as to what window should be used appropriately for MMN analysis. We decided that it would be best to analyze and report three latency windows of ERP recording, 20–50 ms, 50–180 ms, and 20–180 ms. MMN analysis 20–50 ms encompasses the N1 (N40) window, thought to be an attentional component that is deficient in schizophrenia and may impact MMN. Given that there was indeed a MAM effect in rat AEP EEG in both N40 and MMN, we decided to test the model further with the introduction of pharmacology that has been reported to attenuate ERPs in schizophrenia patients. We did not test any antipsychotic medications, such as clozapine largely because of difficulties in reproducing its effects in other behavioral assays in MAM-exposed rats (unpublished data). We also preferred to test pharmacology that would have potential efficacy in both N40 and MMN. We tested the effects of two nAChR agonists, (1) nicotine, which serves as a broad spectrum nAChR agonist which binds with high affinity to a4b2 and at a lower affinity to a7 nAChRs and (2) ABT-107 a selective a7 nAChR agonist. Not unexpectedly, nicotine, tested at a dose that is relevant for cognitive enhancement (0.16 mg/kg i.p.), demonstrated a positive effect on attenuating an N40 deficit seen in MAM rats. Although the effect was not quite significant (p = 0.0618), with the small group size and cross-over study design, a strong trend is present. Nicotine however did not attenuate MMN deficits seen in MAM rats. While this was unexpected, it is not dissimilar to nicotine’s effects in patients. As reported by Baldeweg et al. [25], acute nicotine treatment is only partially able to restore MMN deficits in schizophrenia patients suggesting nicotine may only be weakly affecting the neurotransmitter system modulated in MMN. Doses for ABT-107 were chosen based upon previous behavioral studies done in normal rats [31]. In contrast to nicotine’s effects, ABT-107 (1.0 mg/ kg i.p.), demonstrated the ability to normalize both N40 sensory gating and MMN in MAM rats. This preclinical efficacy for ABT-107 concurs with ERP clinical data generated for the a7 agonist EVP6124 in schizophrenia patients. Nicotine has weak affinity at a7 nAChRs, and it may be due to the selectivity and potency of ABT107 and EVP-6124 on a7 receptors that allow for MMN attenuation [33]. Neither compound that we tested had any negative effects in control (PBS-exposed) rats. ABT-107 did show a non-significant trend for a reduced T:C ratio in N40 signaling of control (PBSexposed) rats whereas nicotine had no effect. Of the limitations to this study and the data generated, the primary limitation would be group size. As a standard procedure with preclinical EEG animals from our lab, we test for pharmacological effects in a Latin square cross-over design. This allows for comparison of baseline response and treatment response from the same animal. Most of the evaluation was done with n = 7–8 for condition (MAM or PBS control). We were able to detect a main effect of condition and some intriguing novel pharmacology data with the small cohorts. However, the statistical analysis using AUC was chosen to be of substantial rigor and, in doing so, does not
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
D A
300
*
200 100 0
PBS-control
E
300 200 100
PBS-control
MAM
condition
1600
saline nicotine ABT-107
1400
amplitude (µV)
2000
amplitude ( µV)
saline nicotine ABT-107
400
0
MAM
condition
B
500
amplitude ( µV)
amplitude ( µV)
400
5
1500
*
1000 500
1200 1000 800 600 400 200
0
PBS-control
C
0
MAM
condition
F
2000
*
1750
amplitude (µV)
amplitude (µV)
MAM
condition
2000 1500 1000 500 0
PBS-control
1500 1250
saline nicotine ABT-107
1000 750 500 250
PBS-control
MAM
condition
0
PBS-control
MAM
condition Fig. 3. (A) AUC amplitude for MMN paradigm in 20–50 ms window. F(2, 14) = 1.509, *p = 0.0189 by unpaired Student’s t-test. (B) AUC amplitude for MMN paradigm in 50– 180 ms window. F(2, 14) = 1.727, *p = 0.0276 by unpaired Student’s t-test. (C) AUC amplitude for MMN paradigm in 20–180 ms window. F(2, 14) = 3.179, p = 0.0651 by unpaired Student’s t-test. (D) Effect of nicotine (0.16 mg/kg) and ABT-107 (1.0 mg/kg) on AUC amplitude for MMN paradigm in 20–50 ms window. Two-way repeated measures ANOVA analysis revealed a significant interaction of condition and treatment [F(2, 42) = 5.954, p = 0.0079]. The overall effects of treatment were not significant [F(2, 42) = 2.649, p = 0.0913]. (E) Effect of nicotine (0.16 mgkg) and ABT-107 (1.0 mg/kg) on AUC amplitude for MMN paradigm in 50–180 ms window. Two-way repeated measures ANOVA analysis revealed a non-significant trend for interaction of condition and treatment [F(2, 42) = 3.056, p = 0.0657]. The overall effect of treatment were not significant [F(2, 42) = 1.482, p = 0.2472]. (F) Effect of nicotine (0.16 mg/kg) and ABT-107 (1.0 mg/kg) on AUC amplitude for MMN paradigm in 20–180 ms window. Two-way repeated measures ANOVA analysis revealed a significant interaction of condition and treatment [F(2, 42) = 5.072, p = 0.0146]. The overall effects of treatment were not significant [F(2, 42) = 2.860, p = 0.0769] however post hoc analyses using Bonferroni’s multiple comparison indicates a significant effect of ABT-107 versus saline (*p < 0.05) on MMN recorded in MAM rats during 20–180 ms window.
show the pharmacological effects to be significant. We believe that replication and an increased group size is necessary to lend greater significance to our data. This investigation shows that we are able to measure AEP EEG in awake, free-moving rats and that MAM-exposed rats show deficits in both N40 and MMN recordings. We also show that these recordings can be done using surface electrodes from the frontal cortex. The pharmacology studies suggest that we can attenuate the ERP deficits in the MAM rats with nAChR modulation. Translational investigations such as this are useful in defining the usefulness, construct validity and predictive validity of a neurodevelopmental model of schizophrenia, such as the MAM model. Using EEG could serve as an important biomarker in the development of pharmacological treatments for schizophrenia. The pharmacology studies detail the underlying usefulness of auditory-evoked EEG to define target engagement for schizophrenia treatment. Disclosure All authors are employees of AbbVie. The design, study conduct, and financial support for this research were provided by AbbVie. AbbVie participated in the interpretation of data, review, and approval of the publication.
References [1] L.S. Pilowsky, R.W. Kerwin, R.M. Murray, Schizophrenia: a neurodevelopmental perspective, Neuropsychopharmacology 9 (1) (1993) 83–91. [2] D.A. Lewis, P. Levitt, Schizophrenia as a disorder of neurodevelopment, Annu. Rev. Neurosci. 25 (2002) 409–432. [3] G. LePen, R. Gourevitch, F. Hazane, C. Hoareau, T.M. Jay, M.-O. Krebs, Peripubertal maturation after developmental disturbance: a model for psychosis onset in the rat, Neuroscience 143 (2006) 395–445. [4] H. Moore, J.D. Jentsch, M. Ghajarnia, M.A. Geyer, A.A. Grace, A neurobehavioral systems analysis of adult rats exposed to methylazoxymethanol acetate on E17: implications for the neuropathology of schizophrenia, Biol. Psychiatry 60 (3) (2006) 253–264. [5] D.J. Lodge, A.A. Grace, Gestational methylazoxymethanol acetate administration: a developmental disruption model of schizophrenia, Behav. Brain Res. 204 (2) (2009) 306–312. [6] R.E. Featherstone, Z. Rizos, J.N. Nobrega, S. Kapur, P.J. Fletcher, Gestational methylazoxymethanol acetate treatment impairs select cognitive functions: parallels to schizophrenia, Neuropsychopharmacology 32 (2) (2007) 483–492. [7] C.L. Chin, P. Curzon, A.J. Schwartz, E.M. O’Connor, L.E. Rueter, G.B. Fox, et al., Structural abnormalities revealed by magnetic resonance imaging in rats prenatally exposed to methylazoxymethanol acetate parallel pathology in schizophrenia, Synapse 65 (2011) 393–403. [8] K.G. Phillips, M.C. Cotel, A.P. McCarthy, D.M. Edgar, M. Tricklebank, M.J. O’Neill, et al., Differential effects of NMDA antagonists on high frequency and gamma EEG oscillations in a neurodevelopmental model of schizophrenia, Neuropharmacology 62 (2012) 1359–1370. [9] M. Hajos, Targeting information-processing deficit in schizophrenia: a novel approach to psychotherapeutic drug discovery, Trends Pharmacol. Sci. 27 (7) (2006) 391–398.
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 6
K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
[10] P.J. Uhlhaas, W. Singer, Abnormal neural oscillations and synchrony in schizophrenia, Nat. Rev. 11 (2010) 100–113. [11] R. Freedman, L.E. Adler, G.A. Gerhardt, N. Baker, G.M. Rose, C. Dreging, et al., Neurobiological studies of sensory gating in schizophrenia, Schizophr. Bull. 13 (4) (1987) 669–676. [12] A.R. Mayer, F.M. Hanlon, A.R. Franco, T.M. Teshiba, R.J. Thoma, V.P. Clark, et al., The neural networks underlying auditory sensory gating, Neuroimage 44 (2009) 182–189. [13] R. Freedman, L.E. Adler, M.C. Waldo, E. Pachtman, R.D. Franks, Neurophysiological evidence for a defect in inhibitory pathways in schizophrenia: comparison of medicated and drug-free patients, Biol. Psychiatry 18 (1983) 537–551. [14] L.E. Adler, A. Olincy, E.M. Gawthra, K.A. McRae, J.G. Harris, H.T. Nagamoto, et al., Varied effects of atypical neuroleptics on P50 auditory gating in schizophrenia patients, Am. J. Psychiatry 161 (2004) 1822–1828. [15] P.C. Bickford-Wimer, H. Nagamoto, R. Johnson, L.E. Adler, M. Egan, G.M. Rose, et al., Auditory sensory gating in hippocampal neurons: a model system in the rat, Biol. Psychiatry 27 (1990) 183–192. [16] R. Na¨a¨ta¨nen, T. Kujala, K. Kreegipuu, S. Carlson, C. Escera, T. Baldeweg, et al., The mismatch negativity: an index of cognitive decline in neuropsychiatric and neurological diseases and aging, Brain 134 (12) (2011) 3435–3453. [17] D. Tikhonravov, T. Neuvonen, A. Pertovaara, K. Savioja, T. Ruusuvirta, R. Naatanen, et al., Dose-related effects of memantine on a mismatch negativitylike response in anesthetized rats, Neuroscience 167 (2010) 1175–1182. [18] U. Schall, S.V. Catts, S. Chaturvedi, B. Liebert, J. Redenbach, F. Karayanidis, et al., The effect of clozapine therapy on frontal lobe dysfunction in schizophrenia: neuropsychology and event-related potential measures, Int. J. Neuropsychopharmacol. 1 (1998) 19–29. [19] D. Umbricht, D.C. Javitt, G.P. Noval, J.A. Bates, S. Pollack, J. Lieberman, et al., Effects of clozapine on auditory event-related potentials in schizophrenia, Biol. Psychiatry 44 (1998) 716–725. [20] D. Umbricht, D.C. Javitt, G.P. Noval, J.A. Bates, S. Pollack, J. Lieberman, et al., Effects of risperidone on auditory event-related potentials in schizophrenia, Int. J. Neuropsychopharmacol. 2 (1998) 229–304. [21] D. Umbricht, S. Krljes, Mismatch negativity in schizophrenia: a meta-analysis, Schizophr. Res. 76 (2005) 1–23. [22] P. Astikainen, T. Ruustuvirta, J. Wikgren, M. Penttonen, Memory-based detection of rare sound feature combinations in anesthetized rats, Neuroreport 17 (14) (2006) 1561–1564. [23] T. Ruusuvirta, K. Koivisto, J. Wikgren, P. Astikainen, Processing melodic contours in urethane anesthetized rats, Eur. J. Neurosci. 26 (2007) 701–703. [24] L.E. Adler, L.J. Hoffer, J. Griffith, M.C. Waldo, R. Freedman, Normalization by nicotine of deficient auditory sensory gating in the relatives of schizophrenia, Biol. Psychiatry 32 (1992) 607–616.
[25] T. Baldeweg, D. Wong, K. Stephan, Nicotinic modulation of human auditory sensory memory: evidence from mismatch negativity potentials, Int. J. Psychophysiol. 59 (2006) 49–58. [26] L. Dulude, A. Labelle, V.J. Knoot, Acute nicotine alteration of sensory memory impairment in smokers with schizophrenia, J. Clin. Psychopharmacol. 30 (5) (2010) 541–548. [27] L.E. Adler, A. Olincy, M. Waldo, J.G. Harris, J. Griffith, K. Stevens, et al., Schizophrenia, sensory gating, and nicotinic receptors, Schizophr. Bull. 24 (2) (1998) 189–202. [28] J. Prickaerts, N.P. van Goethem, R. Chesworth, G. Shapiro, F.G. Boess, C. Methfessel, et al., EVP-6124, a novel and selective a7 nicotinic acetylcholine receptor partial agonist, improves memory performance by potentiating the acetylcholine response to a7 nicotinic acetylcholine receptors, Neuropharmacology 62 (2012) 1099–1110. [29] M. Hajos, R.S. Hurst, W.E. Hoffman, M. Krause, T.M. Wall, N.R. Higdon, et al., The selective a7 nicotinic acetylcholine receptor agonist PNU-282987 [N[(3R)-1-azabicyclo[2.2.2]oct-3-yl]-4-chlorobenzamide hydrochloride] enhances GABAergic synaptic activity in brain slices and restores auditory gating deficits in anesthetized rats, J. Pharmacol. Exp. Ther. 312 (3) (2005) 1213–1222. [30] K.R. Tieje, D.J. Anderson, R.S. Bitner, E.A. Blomme, P.J. Brackemeyer, C.A. Briggs, et al., Preclinical characterization of A-582941: a novel a7 neuronal nicotinic receptor agonist with broad spectrum cognition-enhancing properties, CNS Neurosci. Ther. 14 (1) (2008) 65–82. [31] R.J. Radek, H.M. Robb, K.E. Stevens, M. Gopalakrishnan, R.S. Bitner, Effects of the novel a7 nicotinic acetylcholine receptor agonist ABT-107 on sensory gating in DBA/2 mice: pharmacodynamic characterization, J. Pharmacol. Exp. Ther. 343 (3) (2012) 736–745. [32] R.S. Bitner, W.H. Bunnelle, M.W. Decker, K.U. Drescher, K.L. Kohlhaas, S. Markosyan, et al., In vivo pharmacological characterization of a novel selective a7 neuronal nicotinic acetylcholine receptor agonist ABT-107: preclinical considerations in Alzheimer’s disease, J. Pharmacol. Exp. Ther. 334 (3) (2010) 875–886. [33] A. Olincy, J.G. Harris, L.L. Johnson, V. Pender, S. Kongs, D. Allensworth, et al., Proof-of-concept trial of an a7 nicotinic agonist in schizophrenia, Arch. Gen. Psychiatry 63 (2006) 630–638. [34] S.H. Preskorn, M. Gawryl, N. Dgteluck, M. Palfreyman, L. Bauer, D.C. Hilt, Normalizing effects of EVP-6124, an alpha-7 nicotinic partial agonist, on event-related potentials and cognition: a proof of concept, randomized trial in patients with schizophrenia, J. Psychiatr. Pract. 20 (1) (2014) 12–24. [35] S. Digavalli, P. Chen, N. Lodge, Deviance-based negativity in the conscious rat: modulation by nicotinic agonists, Biochem. Pharmacol. 82 (8) (2011) 1039.
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
BCP-12251; No. of Pages 6 Biochemical Pharmacology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm
Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia Kathy L. Kohlhaas *, Holly M. Robb, Victoria A. Roderwald, Lynne E. Rueter AbbVie, Inc., 1 N. Waukegan Road, North Chicago, IL 60064-6115, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 March 2015 Accepted 21 May 2015 Available online xxx
Schizophrenia is a chronic disease that has been hypothesized to be linked to neurodevelopmental abnormalities. Schizophrenia patients exhibit impairments in basic sensory processing including sensory gating deficits in P50 and mismatch negativity (MMN). Neuronal nicotinic acetylcholine receptor (nAChR) agonists have been reported to attenuate these deficits. Gestational exposure of rats to methylazoxymethanol acetate (MAM) at embryonic day 17 leads to developmental disruption of the limbic-cortical system. MAM exposed offspring show neuropathological and behavioral changes that have similarities with those seen in schizophrenia. In this study, we aimed to assess whether N40 auditory sensory gating (the rodent form of P50 gating) and MMN deficits as measures of auditory evoked potential (AEP) electroencephalography (EEG) are present in MAM rats and whether nAChR agonists could attend the deficit. E17 male MAM and sham rats were implanted with cortical electrodes at 2 months of age. EEG recordings evaluating N40 gating and MMN paradigms were done comparing effects of vehicle (saline), nicotine and the a7 agonist ABT-107. Deficits were seen for MAM rats compared to sham animals in both N40 auditory sensory gating and MMN AEP recordings. There was a strong trend for N40 deficits to be attenuated by both nicotine (0.16 mg/kg i.p. base) and ABT-107 (1.0 mg/kg i.p. base). MMN deficits were significantly attenuated by ABT-107 but not by nicotine. These data support the MAM model as a useful tool for translating pharmacodynamic effects in clinical medicine studies of novel therapeutic treatments for schizophrenia. ß 2015 Elsevier Inc. All rights reserved.
Keywords: EEG Mismatch negativity Sensory gating MAM Nicotine a7nAChR
1. Introduction Schizophrenia is a chronic disease that is linked to neurodevelopmental abnormalities [1,2]. Schizophrenia patients exhibit impairments in basic sensory processing and higher cognitive functions, such as language, reasoning, and planning. There is an ever growing interest in the development of animal models that represent impairments found in schizophrenia. Gestational exposure of rats to methylazoxymethanol acetate (MAM) at embryonic day 17 leads to developmental disruption of the limbic-cortical system [3–5]. MAM exposed offspring show neuropathological and behavioral changes including impairment of cognitive tasks such as spatial working memory, attentional set-shifting, and reversal learning that have similarities with those seen in schizophrenia
* Corresponding author. Tel.: +1 847 937 4066. E-mail addresses: [email protected] (K.L. Kohlhaas), [email protected] (H.M. Robb), [email protected] (V.A. Roderwald), [email protected] (L.E. Rueter).
[3,6]. Electroencephalography (EEG) and imaging studies done in MAM rats suggest abnormalities in structure and synchronized oscillatory activity that parallel disruptions of frontotemporal connectivity in schizophrenia [5,7–10]. Similar to schizophrenia patients, MAM rats also demonstrate a deficit in prepulse inhibition (PPI) of the startle response, a form of sensorimotor gating. To date, whether MAM rats also show other forms of sensory gating deficits typically seen in schizophrenia patients such as P50 and mismatch negativity has not been studied. Sensory gating describes neurological processes for filtering out redundant or unnecessary stimuli in the brain from all possible environmental stimuli [11] by preventing an overload of irrelevant information in the higher cortical centers of the brain. Although sensory gating is largely automatic, it also occurs within the context of attentional processes. Sensory gating is mediated by a network in the brain which involves the auditory cortex, prefrontal cortex and hippocampus. Other areas of the brain associated with sensory gating include the amygdala, striatum, medial prefrontal cortex, and the GABAergic neurons of the midbrain dopamine cell region [12]. Using a paired-click paradigm is a common
http://dx.doi.org/10.1016/j.bcp.2015.05.011 0006-2952/ß 2015 Elsevier Inc. All rights reserved.
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 2
K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
non-invasive technique used to measure sensory gating, a type of event-related potential (ERP). For normal sensory gating, if a person hears a pair of clicks within 500 ms of one another, the person will gate out the second click because it is perceived as being redundant. In humans, evidence of the gating can be seen in the P50 wave, occurring in the brain 50 ms after the click. Low values of the P50 wave indicate that sensory gating has occurred. High values of the P50 wave indicate a lack of sensory gating [11]. Individuals with schizophrenia only reduce the amplitude of the second stimulus by 10–20%, whereas individuals without schizophrenia reduce the amplitude of second stimulus by 80–90%. The P50 auditory gating deficit is one of the best established endophenotypes associated with schizophrenia. Most antipsychotics used clinically do not normalize P50 gating deficits in schizophrenia patients, except clozapine [13,14]. In rodents, a similar paradigm is measured in EEG denoted by amplitude differences between the P20 and N40 waves [15]. Cognitive decline associated with schizophrenia can be indexed by temporofrontal functional deficiency expressed as deficient auditory discrimination and orienting [16]. This functional deficiency can be objectively assessed by means of an electrophysiological measure known as mismatch negativity (MMN). MMN is presumed to reflect the existence of a short-term memory trace of the standard stimulus at the moment of presentation of a deviant stimulus. The MMN is considered an automatic (pre-attentive) response. It can be recorded in the absence of the subject’s attention to auditory stimuli and even in unconscious comatose patients [17]. Since MMN is a change–detection response of the brain that can be elicited in the absence of conscious attention or behavioral task, an auditory ERP wave is generated for the responses to both the ‘standard’ and ‘deviant’ stimuli. As frequencies become divergent, a normal individual can detect the differences between the standard and the deviant. ERPs identify the mismatch. In the human auditory cortex, a detectable change (deviant stimulus) in the physical parameters of a repeated (standard) stimulus evokes a negative component in the auditory ERP within a 100–200 ms time window from stimulus onset. MMN has traditionally been recorded in the so-called auditory odd-ball paradigm involving randomized presentation of repetitive standard sounds and rare occasional deviant sounds. MMN is best visualized in a subtractive or difference waveform. Attenuated MMN amplitudes are a robust finding in patients with schizophrenia and studies done with either typical or atypical antipsychotic medications do not abolish MMN abnormality in patients with schizophrenia [18–21]. MMN has been demonstrated to be recorded in humans along with various animals, including rats [22,23]. Activation of nicotinic acetylcholine receptors (nAChRs) has been shown to improve attention, learning, and memory. Several subtypes of nAChRs are expressed in the mammalian brain, each with distinct physiological and pharmacological properties. Nicotine, which acts as an agonist at multiple nAChR subtypes, can normalize temporal aspects of sensory processing in patients with schizophrenia. Nicotine has also been demonstrated to normalize P50 [24] and MMN amplitude [25,26] in normal healthy subjects. A series of human and animal investigations has suggested that altered expression and function of the a7 nAChR may be responsible for the P50 auditory sensory gating deficit characterized in schizophrenia patients and their relatives [26,27]. a7 nAChRs are widely expressed in the hippocampus, an area of the brain closely associated with memory [28]. a7 nAChR agonists have been shown to improve performance in preclinical learning and memory tasks including pre-attention as measured by sensory gating [29–32]. They have also been shown to improve attention, working and episodic memory in humans. DMXB-A, a partial a7 nAChR agonist, demonstrated significant
improvements in P50 inhibition in schizophrenia patients [33]. Recently, it has also been demonstrated that EVP-6124, a selective and potent a7 nAChR agonist can attenuate P50 and MMN deficits in schizophrenia patients [34]. In this study we evaluate the measures of N40 and MMN in MAM versus control rats and the effects of nicotine and the selective a7 nAChR agonist ABT-107. 2. Methods 2.1. Animals Sprague Dawley timed pregnant female rats were received at E12 from Charles River Laboratories (Raleigh, NC). On E17 pregnant dams were injected intraperitoneally (i.p.) with either sterile phosphate buffered saline (PBS) or a 25 mg/kg methylazoxymethanol acetate (MAM). Pups were born 5–6 days later. On PD21, pups were weaned and separated by gender and litter into groups of 3–5 rats. Rat pups were maintained in group housing on a 12 h light:dark cycle (home cage and lab lights on 06:00–18:00; off 18:00–06:00) with food and water ad libitum. Pups were allowed to mature until they were 2 months old (PD60) before any behavior evaluation was completed. Based upon knowledge from within our lab, the MAM deficit phenotype can be characterized by spontaneous locomotor hyperactivity in early adulthood and postmortem brain weight. Behavioral profiling identified litters to have a good MAM effect prior to assignment for EEG surgery (data not shown). Male rats from two litters for an inclusive total of eight rats each of MAM exposed or control (PBS exposed) groups were surgically implanted with electrodes for recording EEG signals from both frontal and parietal cortices. 2.2. Surgical procedure Under isoflurane anesthesia, male rats were implanted with stainless steel recording electrodes in the frontal (AP + 3.0; ML 2.0) and parietal (AP 3.0; ML 4.0) cortices. Reference electrodes were implanted over the frontal sinus (AP + 6.0; ML + 1.0) and cerebellum. Recording electrodes were grounded to cerebellar reference. Peri-operatively, rats received 2.5 mg/kg subcutaneously (s.c.) flunixin meglumine (QD dosing; Banamine-S, Merck Animal Health, NJ). Flunixin was given for 2 additional days post-operatively. After surgery, the rats were single housed, maintained on a normal light:dark cycle and on a restricted diet to control weight (4 pellets/day). 2.3. EEG recording EEG recordings did not begin for 2 week post-surgery to allow sufficient recovery from the surgical procedure. Rats were habituated to recording chambers for 1 week before start of studies. During habituation, spontaneous EEG was recorded for 2 h periods. EEG was recorded between 9:00 a.m. and 3:00 p.m. Freely moving rats were tethered in sound attenuating chambers to electrical swivels to convey EEG biosignals (Plastics One, Roanoke, VA, USA). EEG biosignals were amplified with differential AC amplifiers (A-M Systems, Sequim, WA, USA) and filtered at 1 and 100 Hz. The same rats were tested in both N40 and MMN paradigms. Rats were administered with either vehicle or investigational compound(s) in a within subject randomized Latin square design, allowing for at least 3 days wash out between test sessions. N40 sensory gating was presented as paired auditory white noise stimuli (500 ms interstimulus interval) every 15 s for a total of 120 trials. Mismatch negativity (MMN) potentials were generated by presenting auditory stimuli every 500 ms using an odd-ball paradigm where there were no successive repeats of the deviant stimulus. The MMN sound stimulus set was a white-noise
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
frequency range (standard = 100–5000 Hz, deviant = 600– 9000 Hz), with a deviant probability of 0.20. For each study day, MMN was run first, followed by a session of standard stimuli and deviant stimuli alone. For N40 and MMN studies, rats were injected with test compound and allowed 30 min before start of session. All studies were coded for drug treatment and un-coded after data sets were analyzed. EEG was acquired and quantified using SciWorks (Datawave Technologies, Loveland, CO, USA).
P20
A
P20
CS
25 µv
2.4. Chemicals
2.5. Statistical analysis An a priori t-test was done to compare control (PBS treated) rats to MAM rats for both N40 and MMN analysis to define whether there was indeed an effect due to condition. Statistical analysis used for N40 and MMN during compound treatment was a twoway repeated measure ANOVA. Post hoc analysis included a Bonferroni’s multiple comparison.
B
*
0.4 0.3 0.2 0.1 0.0
PBS-control
MAM
condition
C
0.6
0.4
T:C ratio
3.1. N40 auditory gating in control and MAM rats
Fig. 2A and B shows the waveform generated by running the MMN paradigm in both control (PBS treated) and MAM rats. The difference waveform between the standard and deviant evoked response potential (ERP) was generated by subtracting the deviant wave from the standard wave. Only values for standards that were immediately preceding deviants were analyzed. MMN was measured by calculating absolute values for area above or below zero reference for the difference waves. Difference waves that lie at or around zero indicate a deficit in MMN, compared to normal in which there is measurable area above or below the zero reference (Fig. 2A and B). Because the window for MMN has not been clearly defined in rats, waveform data were analyzed for AUC at three latency windows; 20–50 ms, 50–180 ms, and 20–180 ms. The windows were defined by the approximate times when the
PBS MAM
N40
0.5
0.5
3.2. Mismatched negativity in control and MAM rats
N40
0.6
3. Results
Fig. 1A shows the grand average traces for N40 recording comparing control (PBS treated) to MAM rats. N40 values were measured by comparing the peak amplitudes from P20 to N40 of test stimulus (TS) to those elicited by condition stimulus (CS). Amplitudes were measured from the response to the first stimulus (condition amplitude, CAMP) and second stimulus (test stimulus, TAMP). Sensory gating was measured by calculating a T:C ratio (TAMP/CAMP). Ratios closer to 1 were considered a deficit, whereas ratios less than 0.5 were considered normal gating. An a priori comparison of control (PBS-treated) rats dosed with saline to MAM rats also dosed with saline show a significant difference in N40 gating [F(2, 16) = 1.114, p = 0.045 by unpaired Student’s t-test] (Fig. 1B). A two-way repeated measure ANOVA for rats treated with either saline, nicotine (0.16 mg/kg i.p.) or ABT-107 (0.32, 1.0 mg/kg i.p.) showed no significant interaction of condition and treatment [F(3, 42) = 1.844, p = 0.1539]. The effect of treatment were not quite significant [F(3, 42) = 2.64, p = 0.0618] (Fig. 1C).
TS
500 msec
T:C ratio
Nicotine hydrogen tartrate salt was purchased from Sigma Aldrich (St Louis, MO, USA). ABT-107, [5-(6-[(3R)-1-azabicyclo[2.2.2]oct-3-yloxy]pyridazin-3-yl)-1H-indole] was synthesized at AbbVie, Incorporated. Compound solutions were prepared in saline. Compounds were administered intraperitoneally in solution in a volume of 1.0 ml/kg body weight for rats and reported as free base weight.
3
saline nicotine 0.32 ABT 1.0 ABT
0.3 0.2 0.1 0.0
PBS-control
MAM
condition Fig. 1. (A) Grand average traces for the evoked response to paired auditory stimuli (ISI = 500 ms; CS = conditioning stimulus, TS = test stimulus) recorded from frontal cortex surface electrodes in MAM and control (PBS) rats. (B) T:C ratio for P20–N40 amplitudes from PBS control rats vs. MAM rats. Data represent mean SEM, n = 8. F(2, 16) = 1.114, *p = 0.045 by unpaired Student’s t-test. (C) Effects of nicotine (0.16 mg/kg) and ABT-107 (0.32 and 1.0 mg/kg) on T:C ratios for P20–N40 amplitudes from PBS control and MAM rats. Data represent mean SEM, n = 7–8. Two-way repeated measures ANOVA show a non-significant trend of effect for treatment. F(3, 42) = 2.64, p = 0.0618.
standard wave crosses the zero reference. An a priori comparison of control (PBS-treated) rats dosed with saline to MAM rats also dosed with saline show a significant difference in MMN. During the 20–50 ms window, MAM rats have a significant decrease in AUC [F(2, 14) = 1.509, p = 0.0189 by unpaired Student’s t-test] (Fig. 3A). During the 50–180 ms window, MAM rats have a significant decrease in AUC [F(2, 14) = 1.727, p = 0.0276 by unpaired Student’s t-test] (Fig. 3B). During the 20–180 ms window, MAM rats have a non-significant decrease in AUC [F(2, 14) = 3.179, p = 0.0651 by unpaired Student’s t-test] (Fig. 3C). Two-way repeated measures ANOVA analysis of MMN for rats treated with either saline, nicotine (0.16 mg/kg i.p.), or ABT-107 (1.0 mg/kg i.p.) during 20–50 ms latency revealed a significant interaction of condition and treatment [F(2, 42) = 5.954, p = 0.0079]. The effects of treatment were not significant [F(2, 42) = 2.649, p = 0.0913] (Fig. 3D). Two-way repeated measures ANOVA analysis of MMN for rats treated with either saline,
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
4
A
standard deviant difference
5 µV 50 msec
B
standard deviant difference
5 µV 50 msec Fig. 2. Grand average traces for waveform of MMN paradigm in control (PBS treated) (A) and MAM (B) rats for standard (frequency range = 100–5000 Hz (dashed line)), deviant (frequency range = 600–9000 Hz (light gray line)); and difference wave (black line). Deviant probability = 0.20. Waveforms indicate deficit in MMN detection in MAM treated animals as denoted by difference wave being lying at or around zero reference.
nicotine (0.16 mg/kg i.p.), or ABT-107 (1.0 mg/kg i.p.) during 50–180 ms latency revealed a non-significant trend for interaction of condition and treatment [F(2, 42) = 3.056, p = 0.0657]. The effects of treatment were not significant [F(2, 42) = 1.482, p = 0.2472] (Fig. 3E). Two-way repeated measures ANOVA analysis of MMN for rats treated with either saline, nicotine (0.16 mg/kg i.p.), or ABT-107 (1.0 mg/kg i.p.) during 20–180 ms latency revealed a significant interaction of condition and treatment [F(2, 42) = 5.072, p = 0.0146]. The effects of treatment were not significant [F(2, 42) = 2.860, p = 0.0769] (Fig. 3F). Post hoc analyses using Bonferroni’s multiple comparison indicate a significant effect of ABT-107 versus saline on MMN recorded in MAM rats during 20–180 ms window. 4. Discussion In this investigation, we compared frontal cortical EEG recordings in cohorts of MAM-exposed adult rats and sham (PBS-exposed) control rats. In particular we were interested in whether we could record ERPs using auditory-evoked N40 and MMN paradigms in awake, freely moving rats implanted with surface electrodes. Most often, N40 is recorded using electrodes placed in the hippocampus or the auditory cortex of anesthetized animals. In our instance, we did not want to use these preclinical investigational methods, but rather wanted to establish whether a more translational measure could be used in rodents for both N40 and MMN measures. Thus, all of the EEG recordings were captured using surface electrodes over frontal cortex. N40 was tested first, since using the paired-click auditory paradigm would additionally indicate that the rats had normal hearing. For MMN, we used a white-noise frequency range for our standard and deviant stimuli
rather than a single frequency or tone to allow for differing hearing spectrum in rodents. Our initial objective was to determine whether there was face/ construct validity to MAM rats as a preclinical model of schizophrenia-like symptoms. Key to this objective would be whether there was a measurable difference in N40 and/or MMN in the MAM-exposed rats compared to controls. With a group size of 7–8 rats per condition, we were indeed able to show a significant difference in N40 signaling (Fig. 1B), with a deficit being driven by the responses to both the condition and the test stimuli (Fig. 1A). We were also able to show significant differences in the MMN difference waves (Figs. 2 and 3A–C). As little information is available on MMN measures from rats and we could only identify one other publication involving MMN recording from awake, freemoving animals [35], it was not clear as to what window should be used appropriately for MMN analysis. We decided that it would be best to analyze and report three latency windows of ERP recording, 20–50 ms, 50–180 ms, and 20–180 ms. MMN analysis 20–50 ms encompasses the N1 (N40) window, thought to be an attentional component that is deficient in schizophrenia and may impact MMN. Given that there was indeed a MAM effect in rat AEP EEG in both N40 and MMN, we decided to test the model further with the introduction of pharmacology that has been reported to attenuate ERPs in schizophrenia patients. We did not test any antipsychotic medications, such as clozapine largely because of difficulties in reproducing its effects in other behavioral assays in MAM-exposed rats (unpublished data). We also preferred to test pharmacology that would have potential efficacy in both N40 and MMN. We tested the effects of two nAChR agonists, (1) nicotine, which serves as a broad spectrum nAChR agonist which binds with high affinity to a4b2 and at a lower affinity to a7 nAChRs and (2) ABT-107 a selective a7 nAChR agonist. Not unexpectedly, nicotine, tested at a dose that is relevant for cognitive enhancement (0.16 mg/kg i.p.), demonstrated a positive effect on attenuating an N40 deficit seen in MAM rats. Although the effect was not quite significant (p = 0.0618), with the small group size and cross-over study design, a strong trend is present. Nicotine however did not attenuate MMN deficits seen in MAM rats. While this was unexpected, it is not dissimilar to nicotine’s effects in patients. As reported by Baldeweg et al. [25], acute nicotine treatment is only partially able to restore MMN deficits in schizophrenia patients suggesting nicotine may only be weakly affecting the neurotransmitter system modulated in MMN. Doses for ABT-107 were chosen based upon previous behavioral studies done in normal rats [31]. In contrast to nicotine’s effects, ABT-107 (1.0 mg/ kg i.p.), demonstrated the ability to normalize both N40 sensory gating and MMN in MAM rats. This preclinical efficacy for ABT-107 concurs with ERP clinical data generated for the a7 agonist EVP6124 in schizophrenia patients. Nicotine has weak affinity at a7 nAChRs, and it may be due to the selectivity and potency of ABT107 and EVP-6124 on a7 receptors that allow for MMN attenuation [33]. Neither compound that we tested had any negative effects in control (PBS-exposed) rats. ABT-107 did show a non-significant trend for a reduced T:C ratio in N40 signaling of control (PBSexposed) rats whereas nicotine had no effect. Of the limitations to this study and the data generated, the primary limitation would be group size. As a standard procedure with preclinical EEG animals from our lab, we test for pharmacological effects in a Latin square cross-over design. This allows for comparison of baseline response and treatment response from the same animal. Most of the evaluation was done with n = 7–8 for condition (MAM or PBS control). We were able to detect a main effect of condition and some intriguing novel pharmacology data with the small cohorts. However, the statistical analysis using AUC was chosen to be of substantial rigor and, in doing so, does not
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
D A
300
*
200 100 0
PBS-control
E
300 200 100
PBS-control
MAM
condition
1600
saline nicotine ABT-107
1400
amplitude (µV)
2000
amplitude ( µV)
saline nicotine ABT-107
400
0
MAM
condition
B
500
amplitude ( µV)
amplitude ( µV)
400
5
1500
*
1000 500
1200 1000 800 600 400 200
0
PBS-control
C
0
MAM
condition
F
2000
*
1750
amplitude (µV)
amplitude (µV)
MAM
condition
2000 1500 1000 500 0
PBS-control
1500 1250
saline nicotine ABT-107
1000 750 500 250
PBS-control
MAM
condition
0
PBS-control
MAM
condition Fig. 3. (A) AUC amplitude for MMN paradigm in 20–50 ms window. F(2, 14) = 1.509, *p = 0.0189 by unpaired Student’s t-test. (B) AUC amplitude for MMN paradigm in 50– 180 ms window. F(2, 14) = 1.727, *p = 0.0276 by unpaired Student’s t-test. (C) AUC amplitude for MMN paradigm in 20–180 ms window. F(2, 14) = 3.179, p = 0.0651 by unpaired Student’s t-test. (D) Effect of nicotine (0.16 mg/kg) and ABT-107 (1.0 mg/kg) on AUC amplitude for MMN paradigm in 20–50 ms window. Two-way repeated measures ANOVA analysis revealed a significant interaction of condition and treatment [F(2, 42) = 5.954, p = 0.0079]. The overall effects of treatment were not significant [F(2, 42) = 2.649, p = 0.0913]. (E) Effect of nicotine (0.16 mgkg) and ABT-107 (1.0 mg/kg) on AUC amplitude for MMN paradigm in 50–180 ms window. Two-way repeated measures ANOVA analysis revealed a non-significant trend for interaction of condition and treatment [F(2, 42) = 3.056, p = 0.0657]. The overall effect of treatment were not significant [F(2, 42) = 1.482, p = 0.2472]. (F) Effect of nicotine (0.16 mg/kg) and ABT-107 (1.0 mg/kg) on AUC amplitude for MMN paradigm in 20–180 ms window. Two-way repeated measures ANOVA analysis revealed a significant interaction of condition and treatment [F(2, 42) = 5.072, p = 0.0146]. The overall effects of treatment were not significant [F(2, 42) = 2.860, p = 0.0769] however post hoc analyses using Bonferroni’s multiple comparison indicates a significant effect of ABT-107 versus saline (*p < 0.05) on MMN recorded in MAM rats during 20–180 ms window.
show the pharmacological effects to be significant. We believe that replication and an increased group size is necessary to lend greater significance to our data. This investigation shows that we are able to measure AEP EEG in awake, free-moving rats and that MAM-exposed rats show deficits in both N40 and MMN recordings. We also show that these recordings can be done using surface electrodes from the frontal cortex. The pharmacology studies suggest that we can attenuate the ERP deficits in the MAM rats with nAChR modulation. Translational investigations such as this are useful in defining the usefulness, construct validity and predictive validity of a neurodevelopmental model of schizophrenia, such as the MAM model. Using EEG could serve as an important biomarker in the development of pharmacological treatments for schizophrenia. The pharmacology studies detail the underlying usefulness of auditory-evoked EEG to define target engagement for schizophrenia treatment. Disclosure All authors are employees of AbbVie. The design, study conduct, and financial support for this research were provided by AbbVie. AbbVie participated in the interpretation of data, review, and approval of the publication.
References [1] L.S. Pilowsky, R.W. Kerwin, R.M. Murray, Schizophrenia: a neurodevelopmental perspective, Neuropsychopharmacology 9 (1) (1993) 83–91. [2] D.A. Lewis, P. Levitt, Schizophrenia as a disorder of neurodevelopment, Annu. Rev. Neurosci. 25 (2002) 409–432. [3] G. LePen, R. Gourevitch, F. Hazane, C. Hoareau, T.M. Jay, M.-O. Krebs, Peripubertal maturation after developmental disturbance: a model for psychosis onset in the rat, Neuroscience 143 (2006) 395–445. [4] H. Moore, J.D. Jentsch, M. Ghajarnia, M.A. Geyer, A.A. Grace, A neurobehavioral systems analysis of adult rats exposed to methylazoxymethanol acetate on E17: implications for the neuropathology of schizophrenia, Biol. Psychiatry 60 (3) (2006) 253–264. [5] D.J. Lodge, A.A. Grace, Gestational methylazoxymethanol acetate administration: a developmental disruption model of schizophrenia, Behav. Brain Res. 204 (2) (2009) 306–312. [6] R.E. Featherstone, Z. Rizos, J.N. Nobrega, S. Kapur, P.J. Fletcher, Gestational methylazoxymethanol acetate treatment impairs select cognitive functions: parallels to schizophrenia, Neuropsychopharmacology 32 (2) (2007) 483–492. [7] C.L. Chin, P. Curzon, A.J. Schwartz, E.M. O’Connor, L.E. Rueter, G.B. Fox, et al., Structural abnormalities revealed by magnetic resonance imaging in rats prenatally exposed to methylazoxymethanol acetate parallel pathology in schizophrenia, Synapse 65 (2011) 393–403. [8] K.G. Phillips, M.C. Cotel, A.P. McCarthy, D.M. Edgar, M. Tricklebank, M.J. O’Neill, et al., Differential effects of NMDA antagonists on high frequency and gamma EEG oscillations in a neurodevelopmental model of schizophrenia, Neuropharmacology 62 (2012) 1359–1370. [9] M. Hajos, Targeting information-processing deficit in schizophrenia: a novel approach to psychotherapeutic drug discovery, Trends Pharmacol. Sci. 27 (7) (2006) 391–398.
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
G Model
BCP-12251; No. of Pages 6 6
K.L. Kohlhaas et al. / Biochemical Pharmacology xxx (2015) xxx–xxx
[10] P.J. Uhlhaas, W. Singer, Abnormal neural oscillations and synchrony in schizophrenia, Nat. Rev. 11 (2010) 100–113. [11] R. Freedman, L.E. Adler, G.A. Gerhardt, N. Baker, G.M. Rose, C. Dreging, et al., Neurobiological studies of sensory gating in schizophrenia, Schizophr. Bull. 13 (4) (1987) 669–676. [12] A.R. Mayer, F.M. Hanlon, A.R. Franco, T.M. Teshiba, R.J. Thoma, V.P. Clark, et al., The neural networks underlying auditory sensory gating, Neuroimage 44 (2009) 182–189. [13] R. Freedman, L.E. Adler, M.C. Waldo, E. Pachtman, R.D. Franks, Neurophysiological evidence for a defect in inhibitory pathways in schizophrenia: comparison of medicated and drug-free patients, Biol. Psychiatry 18 (1983) 537–551. [14] L.E. Adler, A. Olincy, E.M. Gawthra, K.A. McRae, J.G. Harris, H.T. Nagamoto, et al., Varied effects of atypical neuroleptics on P50 auditory gating in schizophrenia patients, Am. J. Psychiatry 161 (2004) 1822–1828. [15] P.C. Bickford-Wimer, H. Nagamoto, R. Johnson, L.E. Adler, M. Egan, G.M. Rose, et al., Auditory sensory gating in hippocampal neurons: a model system in the rat, Biol. Psychiatry 27 (1990) 183–192. [16] R. Na¨a¨ta¨nen, T. Kujala, K. Kreegipuu, S. Carlson, C. Escera, T. Baldeweg, et al., The mismatch negativity: an index of cognitive decline in neuropsychiatric and neurological diseases and aging, Brain 134 (12) (2011) 3435–3453. [17] D. Tikhonravov, T. Neuvonen, A. Pertovaara, K. Savioja, T. Ruusuvirta, R. Naatanen, et al., Dose-related effects of memantine on a mismatch negativitylike response in anesthetized rats, Neuroscience 167 (2010) 1175–1182. [18] U. Schall, S.V. Catts, S. Chaturvedi, B. Liebert, J. Redenbach, F. Karayanidis, et al., The effect of clozapine therapy on frontal lobe dysfunction in schizophrenia: neuropsychology and event-related potential measures, Int. J. Neuropsychopharmacol. 1 (1998) 19–29. [19] D. Umbricht, D.C. Javitt, G.P. Noval, J.A. Bates, S. Pollack, J. Lieberman, et al., Effects of clozapine on auditory event-related potentials in schizophrenia, Biol. Psychiatry 44 (1998) 716–725. [20] D. Umbricht, D.C. Javitt, G.P. Noval, J.A. Bates, S. Pollack, J. Lieberman, et al., Effects of risperidone on auditory event-related potentials in schizophrenia, Int. J. Neuropsychopharmacol. 2 (1998) 229–304. [21] D. Umbricht, S. Krljes, Mismatch negativity in schizophrenia: a meta-analysis, Schizophr. Res. 76 (2005) 1–23. [22] P. Astikainen, T. Ruustuvirta, J. Wikgren, M. Penttonen, Memory-based detection of rare sound feature combinations in anesthetized rats, Neuroreport 17 (14) (2006) 1561–1564. [23] T. Ruusuvirta, K. Koivisto, J. Wikgren, P. Astikainen, Processing melodic contours in urethane anesthetized rats, Eur. J. Neurosci. 26 (2007) 701–703. [24] L.E. Adler, L.J. Hoffer, J. Griffith, M.C. Waldo, R. Freedman, Normalization by nicotine of deficient auditory sensory gating in the relatives of schizophrenia, Biol. Psychiatry 32 (1992) 607–616.
[25] T. Baldeweg, D. Wong, K. Stephan, Nicotinic modulation of human auditory sensory memory: evidence from mismatch negativity potentials, Int. J. Psychophysiol. 59 (2006) 49–58. [26] L. Dulude, A. Labelle, V.J. Knoot, Acute nicotine alteration of sensory memory impairment in smokers with schizophrenia, J. Clin. Psychopharmacol. 30 (5) (2010) 541–548. [27] L.E. Adler, A. Olincy, M. Waldo, J.G. Harris, J. Griffith, K. Stevens, et al., Schizophrenia, sensory gating, and nicotinic receptors, Schizophr. Bull. 24 (2) (1998) 189–202. [28] J. Prickaerts, N.P. van Goethem, R. Chesworth, G. Shapiro, F.G. Boess, C. Methfessel, et al., EVP-6124, a novel and selective a7 nicotinic acetylcholine receptor partial agonist, improves memory performance by potentiating the acetylcholine response to a7 nicotinic acetylcholine receptors, Neuropharmacology 62 (2012) 1099–1110. [29] M. Hajos, R.S. Hurst, W.E. Hoffman, M. Krause, T.M. Wall, N.R. Higdon, et al., The selective a7 nicotinic acetylcholine receptor agonist PNU-282987 [N[(3R)-1-azabicyclo[2.2.2]oct-3-yl]-4-chlorobenzamide hydrochloride] enhances GABAergic synaptic activity in brain slices and restores auditory gating deficits in anesthetized rats, J. Pharmacol. Exp. Ther. 312 (3) (2005) 1213–1222. [30] K.R. Tieje, D.J. Anderson, R.S. Bitner, E.A. Blomme, P.J. Brackemeyer, C.A. Briggs, et al., Preclinical characterization of A-582941: a novel a7 neuronal nicotinic receptor agonist with broad spectrum cognition-enhancing properties, CNS Neurosci. Ther. 14 (1) (2008) 65–82. [31] R.J. Radek, H.M. Robb, K.E. Stevens, M. Gopalakrishnan, R.S. Bitner, Effects of the novel a7 nicotinic acetylcholine receptor agonist ABT-107 on sensory gating in DBA/2 mice: pharmacodynamic characterization, J. Pharmacol. Exp. Ther. 343 (3) (2012) 736–745. [32] R.S. Bitner, W.H. Bunnelle, M.W. Decker, K.U. Drescher, K.L. Kohlhaas, S. Markosyan, et al., In vivo pharmacological characterization of a novel selective a7 neuronal nicotinic acetylcholine receptor agonist ABT-107: preclinical considerations in Alzheimer’s disease, J. Pharmacol. Exp. Ther. 334 (3) (2010) 875–886. [33] A. Olincy, J.G. Harris, L.L. Johnson, V. Pender, S. Kongs, D. Allensworth, et al., Proof-of-concept trial of an a7 nicotinic agonist in schizophrenia, Arch. Gen. Psychiatry 63 (2006) 630–638. [34] S.H. Preskorn, M. Gawryl, N. Dgteluck, M. Palfreyman, L. Bauer, D.C. Hilt, Normalizing effects of EVP-6124, an alpha-7 nicotinic partial agonist, on event-related potentials and cognition: a proof of concept, randomized trial in patients with schizophrenia, J. Psychiatr. Pract. 20 (1) (2014) 12–24. [35] S. Digavalli, P. Chen, N. Lodge, Deviance-based negativity in the conscious rat: modulation by nicotinic agonists, Biochem. Pharmacol. 82 (8) (2011) 1039.
Please cite this article in press as: Kohlhaas KL, et al. Nicotinic modulation of auditory evoked potential electroencephalography in a rodent neurodevelopmental model of schizophrenia. Biochem Pharmacol (2015), http://dx.doi.org/10.1016/j.bcp.2015.05.011
