Histamine and acetylcholine receptor involvement in sensorimotor gating: an autoradiography study

Introduction

Sudden intense sensory stimuli can evoke a rapid muscular reaction called the startle response. The startle response can be inhibited by a milder sensory stimulus (prepulse) when presented shortly before the startling stimulus (pulse), a phenomenon termed prepulse inhibition (PPI) (Hoffman & Searle, 1965). PPI can involve any combination of visual, tactile, or auditory stimuli. PPI is believed to entail sensory gating mechanisms as well as direct motor reflex inhibition. An expansive network of brain regions underlies startle, PPI, and modulation of PPI. This network includes nuclei within the pontine reticular system, the superior and inferior colliculi, substantia nigra, basolateral amygdala, hippocampus, thalamus, prefrontal cortex, ventral pallidum, striatum, and ventral tegmental area (Swerdlow et al., 2001). In humans, PPI is believed to be related to an individual’s ability to filter incoming sensory information as well as inhibit resultant behavior (Braff et al., 2001; Rabin et al., 2009), and has been shown to be impaired in a number of neuropsychological disorders including schizophrenia and autism (Braff et al., 2001). PPI is a useful experimental technique used to investigate sensorimotor gating mechanisms and test therapeutic treatments in animal models of neurologic disease as well as in clinical populations, most extensively in patients with schizophrenia (Swerdlow et al., 2008).

The aim of this study was to investigate an established rat model of decreased PPI induced by administration of the NMDA antagonist, dizocilpine (Mansbach & Geyer, 1989), and the reversal of this PPI impairment by the histaminergic H1-antagonist, pyrilamine. H1-antagonism is a potential mechanism of the therapeutic effects of the atypical antipsychotic, clozapine, which improves PPI following dizocilpine administration in rats as well as in patients with schizophrenia (Kumari & Sharma, 2002; Levin et al., 2007; Roegge et al., 2007). In the present study we show that chronic pyrilamine administration prevents the PPI impairment induced by chronic dizocilpine administration, an effect that is correlated with a reduction in ligand-binding potential of H1 receptors in the anterior cingulate and an increase in nicotinic receptor α7 subunit binding in the insular cortex. In light of the functional anatomical connectivity of the anterior cingulate and insular cortex, both of which interact extensively with the core PPI network, our findings support the inclusion of both cortical areas in an expanded network capable of regulating sensorimotor gating.

Methods

Quantitative receptor autoradiography

Following behavioral analyses, the rats were anesthetized with Euthasol (100 mg/kg, Virbac, Inc., Fort Worth, TX, USA) and then sacrificed by exsanguination. Brains were immediately removed and stored at -80°C. To prepare histological sections, each brain was sliced coronally at 20 µm thickness using a cryostat (Bright Instruments, Cambridgeshire, UK) and sections were mounted on gelatin-subbed slides (precleaned Superfrost Fisherbrand slides treated with 0.5% gelatin and 0.05% chromium potassium sulphate). All radioligands were purchased from American Radiolabeled Chemicals (Saint Louis, MO). Specific binding of the α7 subunit of the acetylcholine receptor was determined using [125I]α-bungarotoxin (5 nM, 147 Ci/mmol), and specific binding of the histamine H1 receptor was determined using [3H]pyrilamine (5 nM, 20 Ci/mmol). Radioligands were diluted in a chilled buffer containing 50 nM Tris-HCl and 1 mg/ml BSA, pH 7.4 and the entire radioligand binding protocol was performed on ice to limit nonspecific binding. Slides were first pre-incubated in buffer for 1 hr, after which they were incubated with the appropriate radioligand solution for 1 hr. Non-specific binding was assessed by co-incubation with l-nicotine (4 µM, Acros Organics, NJ) or nonradioactive pyrilamine (2 µM, MP Biomedicals, CA) for the α7 subunit and H1-receptor, respectively. Following incubation in radioligand solution, the slides were washed for 15 min in three changes of chilled buffer, and then subsequently dried with blown air at room temperature. Sections were then exposed to film (Kodak, Hyperfilm) for 22 days to visualize α7 subunit labeling and approximately 6 weeks to visualize pyrilamine labeling. Films were scanned (Epson, Perfection V750 Pro) and analyzed densitometrically using Image-J (Schneider et al., 2012). Each film was exposed to the appropriate set of radioactive standards, and densitometric data were converted to fmol/mg tissue equivalent.

Radioligand binding was quantified within a distributed network of brain regions (Figure 1) previously implicated in regulating PPI of startle (Swerdlow et al., 2001). The regions we investigated within this network included the inferior colliculus, amygdala, and hippocampus. Within the amygdaloid complex, we sampled within specific groups of nuclei: the basolateral group, the cortical group, and the centromedial group (Sah et al., 2003), and within the hippocampal complex, we sampled CA1-3 of the hippocampus proper, as well as the dentate gyrus. Additionally, we qualified radioligand binding in the anterior cingulate and insular cortex, both of which are known to interact with the PPI network and have been preliminarily implicated in fMRI investigations of PPI in humans (Campbell et al., 2007; Kumari et al., 2007). A stereotaxic atlas (Paxinos, 2005) was used to confirm that each brain region was sampled consistently, and data from each brain hemisphere were averaged prior to statistical analysis.

Figure 1. Diagram of brain regions investigated.

Each of the regions investigated with autoradiography are shown in light grey and are labeled. The colored highlighted portions of each region demonstrate the precise location within which autoradiographic analyses were performed in this study. This figure was generated from images acquired in a 3-D stereotaxic rat brain atlas software (Brain Navigator release 2.0, 2009).

Results

Pyrilamine treatment reverses PPI impairment in dizocilpine-treated rats

Although prepulse intensity did not significantly affect inhibition of the startle response, we did observe a trend towards increased PPI with increased prepulse intensity (Figure 2, Dataset 1). For trials using the 77 dB prepulse, the highest intensity employed, ANOVA demonstrated a significant effect of pharmacologic treatment (p = 0.026), with pyrilamine administration significantly attenuating the dizocilpine-induced PPI impairment (p = 0.021).

Figure 2. Pyrilamine treatment attenuates dizocilpine-induced reduction in PPI.

At each prepulse decibel (dB) level employed, the general pattern of pharmacologic manipulation of PPI was maintained: pyrilamine treatment increased PPI and attenuated dizocilpine-induced PPI impairment. Furthermore, in agreement with prior reports, increased prepulse intensity was associated with stronger prepulse inhibition. Error bars represent standard error of the mean.

Pyrilamine treatment decreases H1-receptor binding in the anterior cingulate, an effect correlated with PPI improvement

Within the anterior cingulate, pyrilamine treatment induced a 37 ± 19% decrease in H1-receptor binding that was nearly significant (p = 0.058) (Figure 3, Dataset 2). This change correlated with PPI such that decreased H1-receptor binding was associated with increased mean PPI at all three prepulse intensities (Figure 4, R² = -0.15, p = 0.05). Note as demonstrated in Figure 4, Figure 6, and Figure 7 that some rats (irrespective of pharmacologic treatment) did not demonstrate the PPI phenomenon: this is a common occurrence in PPI studies and reflects inter-individual differences.

Figure 3. Pyrilamine treatment results in decreased H1 receptor binding in the anterior cingulate.

An overall effect of pyrilamine treatment was a reduction in H1-receptor binding in the anterior cingulate (AC). This was a nearly significant trend that will require further investigation to confirm (p = 0.058).

Figure 4. H1 receptor binding in the anterior cingulate correlated with PPI.

There was a negative correlation between H1 receptor binding in the anterior cingulate (AC) and PPI, such that lower H1 binding in the AC was associated with higher PPI. Data are displayed for the 77 dB prepulse trials for which PPI was greatest.

Acetylcholine receptors in the insular cortex are likely involved in pyrilamine-induced PPI improvement

Within the insular cortex, pyrilamine treatment induced a 9 ± 3% increase in α7 acetylcholine (ACh)-receptor binding (p = 0.002) (Figure 5, Dataset 2). Increased α7 ACh-receptor binding correlated with increased mean PPI at all three prepulse intensities (Figure 6, R² = 0.24, p = 0.004).

Figure 5. Pyrilamine treatment results in increased α7-nicotinic receptor binding in the insula.

An overall effect of pyrilamine treatment was an increase in α7-nicotinic receptor binding in the insular cortex (p = 0.002).

Figure 6. α7 receptor binding in the insular cortex correlated with PPI.

We observed a positive correlation between α7-nicotinic receptor binding in the insular cortex and PPI, such that greater α7 binding in the insula was associated with higher PPI. Data are displayed for the 77 dB prepulse trials for which PPI was greatest.

Multiple mechanisms of pyrilamine-induced PPI improvement

The results from this study imply that pyrilamine modulates PPI of the startle response through diverse mechanisms that are regionally discrete. In the anterior cingulate, α7 ACh-receptor binding was positively correlated with mean PPI (Figure 7, R² = 0.14, p = 0.027); however, α7 ACh-receptor binding in this region was not significantly affected by pyrilamine treatment (p = 0.507, Dataset 2). Likewise, pyrilamine treatment did not affect H1-receptor binding in the insular cortex (p = 0.237), and H1-receptor binding in the insula was not associated with mean PPI (p = 0.702). In the other brain regions investigated (inferior colliculus, amygdala, and dorsal hippocampus), neither H1-receptor binding nor α7 ACh-receptor binding was associated with PPI levels.

Figure 7. α7 receptor binding in the anterior cingulate correlated with PPI.

Although pyrilamine treatment did not affect α7 receptor binding in the anterior cingulate (AC), we did observe a positive correlation between α7 receptor binding in this region and PPI. This indicates that α7 receptors in the AC may contribute to PPI independently of H1 receptor-mediated signaling.

Discussion

Sensorimotor gating impairment has been associated with a wide variety of neurological and psychiatric disorders. It has been observed in patients with sensory processing disorder (Davies et al., 2009), Parkinson’s disease (Nakashima et al., 1993), schizophrenia (Braff et al., 2001), non-epileptic seizures (Pouretemad et al., 1998), Tourette’s syndrome and ADHD (Castellanos et al., 1996), nocturnal enuresis (Ornitz et al., 1992), blepharospasm (Gómez-Wong et al., 1998), obsessive-compulsive disorder (Swerdlow et al., 1993; Hoenig et al., 2005), panic disorder (Ludewig et al., 2002), bipolar disorder (Perry et al., 2001; Rich et al., 2005), Huntington’s disease (Swerdlow et al., 1995), Fragile X syndrome (Hessl et al., 2009), and autism (Perry et al., 2007). In humans, sensory gating measures such as PPI have been found to correlate with a growing list of behavioral symptoms, including neuroticism, disinhibition on the go/no-go task, premonitory urges in Tourette’s syndrome, high trait anxiety in panic disorder, restricted/repetitive behavior in autism, and in schizophrenia: positive and negative symptoms, semantic priming abnormalities, perseveration, information-processing deficits, thought disorder, distractibility, and violent behavior (Perry & Braff, 1994; Karper et al., 1996; Vinogradov et al., 1996; Braff et al., 1999; Corr et al., 2002; Ludewig et al., 2002; Perry et al., 2007; Rabin et al., 2009; Yadon et al., 2009). It is important to investigate the neural bases that underlie dysfunctions of sensory gating in order to develop effective therapeutic treatments.

The present study employed a PPI paradigm in which an auditory startle stimulus was immediately preceded by a milder auditory prepulse expected to induce significant gating of the startle response. Chronic administration of the NMDA antagonist, dizocilpine, caused a considerable deficit in sensorimotor gating, demonstrated by the low levels of PPI of the startle response in dizocilpine-treated rats (mean PPI of 5.6%). This deficit was entirely reversed by co-administration of the selective histaminergic H1-receptor antagonist, pyrilamine (Figure 2).

Dizocilpine disruption of PPI has been previously used to model sensorimotor gating impairments (Mansbach & Geyer, 1989; Geyer et al., 2001). The predictive validity of this model is supported by the fact that the atypical antipsychotic, clozapine, reverses PPI impairment in this model (Levin et al., 2007; Lim et al., 2012) as it does similarly in humans (Nagamoto et al., 1996; Adler et al., 1998). Here we show that chronic pyrilamine treatment mimics the effect of clozapine, reversing PPI impairment in the dizocilpine-treated rat. As clozapine has been shown to saturate the brain’s H1-receptors at therapeutic concentrations (Humbert-Claude et al., 2012), this may provide an important mechanism of clozapine’s therapeutic action.

In the mammalian brain, histaminergic fibers originating in the tuberomammillary nucleus of the posterior hypothalamus regulate the response to noxious stimuli (Itoh et al., 1989) and infection (Saper et al., 2012), and also regulate the excitability of arousal circuits (Tasaka et al., 1989). One mechanism of histamine’s effects on arousal is through regulation of cholinergic transmission in the nucleus basalis of Myenert in the basal forebrain (Bacciottini et al., 2001; Dringenberg & Kuo, 2003). Histamine’s cholinergic effects have been shown to differ regionally in the brain: in the frontoparietal cortex, local histamine administration inhibited acetylcholine release by 50% (Blandina et al., 1996), whereas histamine application to the nucleus basalis has been shown to double cholinergic output to the cortex in rats (Cecchi et al., 1998). The complicated interactions between histaminergic signaling and cholinergic tone throughout the brain awaits further elucidation; however, it has been shown that tuberomammillary lesions or systemic administration of the H1 antagonists, chlorpheniramine and pyrilamine, are similarly capable of significantly increasing cortical acetylcholine dose-dependently (Dringenberg et al., 1998). Further, members of our group have recently shown that pyrilamine treatment reduces nicotine self-administration in rats (Levin et al., 2011). So it is believed that systemic H1 antagonists modify behavior in part by increasing cortical cholinergic transmission.

The role of histaminergic signaling in PPI has not been extensively studied, but of the disorders investigated to date that are associated with sensorimotor gating abnormalities, all have displayed increased histaminergic neurotransmission. These include: Parkinson’s disease (Rinne et al., 2002), Tourette’s syndrome (Fernandez et al., 2012), bipolar disorder (Jin et al., 2009), Huntington’s disease (van Wamelen et al., 2011), and schizophrenia (Ito, 2004; Arrang, 2007). In Alzheimer’s dementia, interestingly, researchers found significantly decreased histamine levels in the frontal and temporal cortices (Mazurkiewicz-Kwilecki & Nsonwah, 1989), and normal PPI (Hejl et al., 2004).

Prior work by members of our group demonstrated the ability of a single dose of pyrilamine to attenuate PPI impairment in rats acutely administered dizocilpine or amphetamine (Roegge et al., 2007; Larrauri & Levin, 2010). The current study was designed to further explore this phenomenon pharmacologically in rats by chronically administering dizocilpine and/or pyrilamine, a scenario more analogous to long-term therapeutic enhancement of sensorimotor gating in humans. Our aim was to determine which components of the distributed PPI network were impacted by pyrilamine treatment to improve PPI.

Pyrilamine treatment resulted in decreased H1-receptor binding in the anterior cingulate, which was correlated with PPI improvement (Figure 3 and Figure 4). Functional involvement of the anterior cingulate in PPI has been previously demonstrated through lesion as well as fMRI and PET studies (Hazlett et al., 1998; Yee, 2000; Goldman et al., 2006; Campbell et al., 2007; Neuner et al., 2010), and this region is known to express a high concentration of H1-receptors (Tagawa et al., 2001). As a site of limbic and cortical integration, the anterior cingulate modulates conditioned fear responses and arousal (Hamner et al., 1999). The anterior cingulate has been shown to send efferent projections to the amygdala (Wang et al., 2009), a wide distribution of thalamic nuclei (Fujii, 1983), substantia nigra (Beckstead, 1979), nucleus accumbens (Sesack et al., 1989), globus pallidus (Beckstead, 1979), and superior colliculus (Sesack et al., 1989), all regions implicated in regulating PPI of the startle response (Yamada et al., 1998; Fendt et al., 2001; Hazlett et al., 2001; Takahashi et al., 2007; Forcelli et al., 2012). Efferents of the anterior cingulate have even been traced to the giant neurons of the caudal pontine reticular nucleus, which display prepulse-inhibited membrane potential and are believed to be an integral component of the acoustic startle response (Sesack et al., 1989; Lingenhöhl & Friauf, 1994). Similarly, anterior cingulate efferents were traced to the adjacent pedunculopontine tegmental nucleus (Sesack et al., 1989), which is an established component of the core brainstem PPI circuitry that directly modulates the pontine reticular giant neurons through cholinergic innervation (Fendt et al., 2001). We have shown that α7 ACh-receptor binding in the anterior cingulate is positively correlated with PPI (Figure 7), so it is possible that cholinergic signaling in this region plays a role in PPI. Indeed, nicotine has been shown to increase activity in the anterior cingulate while improving PPI (Postma et al., 2006). However, we found no effect of pyrilamine on the α7 ACh-receptor binding in the anterior cingulate, implicating an alternate mechanism of pyrilamine enhancement of PPI in this brain region.

Contrary to the anterior cingulate, the insular cortex displayed a significant increase in α7 ACh-receptor binding in pyrilamine-treated rats that was positively correlated with PPI (Figure 5 and Figure 6). However, there was no change in H1-receptor binding in the insula of rats that were treated with pyrilamine. Because H1 antagonists have been shown to be ineffective in regulating insular cortical excitability when directly applied to this region (Takei et al., 2012), pyrilamine treatment may modulate insular acetylcholine signaling indirectly. The insular cortex contains a high concentration of nicotinic receptors, and has shown increased activation following nicotine administration in patients with schizophrenia whose PPI was improved by nicotine (Postma et al., 2006). Agonism of α7 ACh-receptors has been demonstrated to improve sensory gating in patients with schizophrenia (Martin & Freedman, 2007), and has proven beneficial in animal models of not only gating impairments (Cilia et al., 2005; Hajós et al., 2005; Thomsen et al., 2009) but also positive and negative symptoms of schizophrenia (Hauser et al., 2009). Therefore, it is likely that pyrilamine-induced increases in insular α7 ACh-receptor expression contribute to the observed improvement in PPI.

Due to its anatomic connections, the insular cortex is believed to integrate the processing of autonomic responses with that of ongoing behavioral plans and emotional states (Allen et al., 1991; Gu et al., 2013). The insula is anatomically poised to influence top-down processing of sensorimotor gating, with both direct as well as indirect connections to the pontine reticular startle circuit (Wiesendanger & Wiesendanger, 1982). The insula send largely reciprocal projections to the amygdala and mediodorsal nucleus of the thalamus (Shi & Cassell, 1998), substantia nigra, raphe nucleus, ventral pallidum, and ventral striatum (Reep & Winans, 1982), all of which have been associated with regulation of PPI (Young et al., 1995; Fendt et al., 2001; Adams et al., 2008; Baldan et al., 2011; Forcelli et al., 2012). The insula have furthermore been shown to be involved in distinguishing successive stimuli presented with a short interstimulus time interval as is required to elicit PPI (Kosillo & Smith, 2010).

Although the anterior cingulate and insular cortex have been shown to be highly interconnected functionally (Di Martino et al., 2009; Medford & Critchley, 2010; Cauda et al., 2011), there was no correlation between the changes we report in pyrilamine binding in the anterior cingulate and α-bungarotoxin binding in the insular cortex. The anterior cingulate and insular cortex both send projections to regions that are known to directly modulate the startle response through interaction with the pontine reticular nuclei responsible for this reflex. It is therefore possible that systemic pyrilamine treatment modified the activity of the anterior cingulate and insular cortex independently to improve PPI, although further research is necessary to understand the interaction between anterior cingulate and insular cortical networks in modulating startle inhibition. This study demonstrates the wide distribution of networks capable of influencing PPI of startle, supporting the importance of the PPI measure as a tool to analyze interactions between multiple hierarchies of neuronal processing in disparate brain regions, particularly in pathological states. It will be important to further investigate effects of pharmacological treatments on PPI in humans as well as animal models of disease to elucidate the neuronal machinery underlying sensory filtering and behavioral inhibition. The involvement of H1 receptors in PPI has not only been demonstrated in pharmacologically induced PPI disruption models, but also in models of developmental disturbance that show PPI impairment. In a mouse model of isolation rearing, pyrilamine treatment or H1 receptor knockout were capable of preventing PPI impairment (Dai et al., 2005) similarly to clozapine (Möller et al., 2011). Along these lines, further research into the histaminergic regulation of PPI is warranted, as H1-antagonists have demonstrated a low incidence of side effects in clinical trials (Pearlman et al., 1997; Lankford et al., 2012). A recent initial clinical study has shown that the antihistamine meclizine significantly improves PPI in people who have less than typical PPI and exaggerated startle response (Larrauri et al., 2014). H1 antagonists may therefore prove useful in reversing sensory gating dysfunctions. Further evaluation should determine the possible efficacy and side effects of this novel line of treatment.

Data availability

figshare: Prepulse inhibition of the startle response and radioligand binding assays, doi: 10.6084/m9.figshare.1060215 (Skefos et al., 2014)