Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Muscarinic cholinergic autoreceptors contribute to the regulation of ACh released by both peripheral and central neurons. Muscarinic autoreceptors inhibit ACh release (Starke et al., 1989) and modulate end organ responsiveness in vascular smooth muscle of the eye (Steinle and Smith, 2000), stomach (Ogishima et al., 2000), urinary bladder (D'Agostino et al., 1997), and upper airway (Ten Berge et al., 1996a). Central cholinergic neurotransmission is altered by muscarinic autoreceptors at the level of the brain stem (Baghdoyan et al., 1998), midbrain (Kitaichi et al., 1999), forebrain (Disko et al., 1999), and cortex (Quirion et al., 1994). Muscarinic autoreceptors modulate cholinergic neurotransmission in animals ranging from insects (Judge and Leitch, 1999) to humans (Ten Berge et al., 1996b).

Disorders of cognition and memory stimulate interest in cortical autoreceptors as potential targets for drug development (Buccafusco and Terry, 2000). For example, the depletion of cortical M2 receptors in Alzheimer's disease may reflect the loss of presynaptic autoreceptors on cholinergic terminals arising from the nucleus basalis of Meynert (Mesulam, 1998). One specific function of prefrontal cortex is manipulation of briefly stored information, or working memory (for review, see Goldman-Rakic, 1996; Gabrieli et al., 1998). Studies in rat report that ACh efflux in prefrontal cortex is enhanced during performance of a memory task (Hironaka et al., 2001). Cortical slice preparations taken from outbred strains of mice suggest that M2 autoreceptors modulate ACh outflow (Iannazzo and Majewski, 2000a, 2000b). No previous studies, however, have characterized autoreceptor modulation of ACh release from cortex of intact mouse.

Enthusiasm for murine models is reinforced by the fact that nearly every human gene has a mouse homolog (O'Brien et al., 1999). The mouse genomic sequence is anticipated to be 90% completed by 2003, enhancing genotyping and phenotyping relevant to human disease (Denny and Justice, 2000). Many mouse genes have been identified as the source of phenotypic deviants (Copeland et al., 1993; Silver, 1995) providing an important justification for the present effort to characterize central nervous system cholinergic neurotransmission as a phenotype. Chimeras crossed to the C57BL/6J (B6) mouse have been used to target genes coding for a wide range of receptors and signal transduction molecules (Gerlai, 1996). The B6 mouse provides the genomic background for many transgenic and knockout models (Silver, 1995). Therefore, the present study was designed to test the hypothesis that muscarinic autoreceptors modulate ACh release in prefrontal cortex of B6 mouse. This study also aimed to identify the muscarinic cholinergic receptor (mAChR) subtype regulating ACh release. In vivo microdialysis was used to deliver muscarinic antagonists with different affinities for the five mAChR subtypes and to collect endogenous ACh. The objective was to quantify the minimal antagonist concentration causing a significant increase in ACh release. This technique has been used successfully to identify M2 autoreceptors as regulators of ACh release in rat striatum (Billard et al., 1995) and cat pontine reticular formation (Baghdoyan et al., 1998). Portions of these data have been presented in abstract form (Douglas et al., 2000).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animal Care. Animal handling and procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (National Academy of Sciences Press, 1996). Adult male B6 mice (n = 45, 25-35 g; Jackson Laboratory, Bar Harbor, ME) were housed in a 12:12 light/dark cycle with free access to food and water. All experiments took place between 10 AM and 5 PM.

Drug Preparation. Scopolamine methyl bromide, pirenzepine dihydrochloride, tetrodotoxin (TTX), and pertussis toxin (PTX) (Sigma, St. Louis, MO) were dissolved in Ringer's solution (147 mM NaCl, 2.4 mM CaCl, 4.0 mM KCl, and 10 µM neostigmine bromide). AF-DX116 (a gift from Boehringer-Ingelheim, Ridgefield, CT) was dissolved initially in 0.05 N HCl and then diluted in Ringer's. The maximum HCl concentration in the final dilution (0.1 µM) previously was shown to have no effect on ACh release (Baghdoyan et al., 1998).

In Vivo Microdialysis and High Performance Liquid Chromatography with Electrochemical Detection (HPLC/EC). CMA/11 microdialysis probes (CMA/Microdialysis, Acton, MA) measuring 1.0 mm in length and 0.24 mm in diameter were used for all antagonist and TTX studies. The molecular weight cutoff of the CMA/11 cuprophane dialysis membrane is 6,000 Da. The molecular weight of ACh is less than 150 Da. IBR-2 combination microdialysis/microinjection probes (Bioanalytical Systems, Inc., West Lafayette, IN) measuring 1.0 mm in length and 0.36 mm in diameter with a 30,000 Da molecular weight cutoff were used for all PTX studies. Ringer's and drug solutions were delivered to the dialysis probe at a constant rate of 2.0 µl/min with a CMA/100 pump. Dialysis samples of 25 µl were collected every 12.5 min.

Anesthesia and Animal Preparation. At the beginning of each experiment, mice were weighed and anesthesia was induced by administering 1.5 to 2% isoflurane (Abbott Laboratories, North Chicago, IL) in 100% O₂. Anesthetized mice were placed in a David Kopf (Tujunga, CA) model 962 stereotaxic frame with a model 921 mouse adapter and rat ear bars. Anesthesia was maintained by 0.70 to 0.82% isoflurane delivered through a tube covering the nose. Delivered isoflurane concentration was measured using a Raman spectrometer (Ohmeda Rascal II, Louisville, CO). Anesthetic level was evaluated constantly by observation of spontaneous movements and by response to hindlimb pinch. Respiratory rate and core body temperature were monitored every 12.5 min. Body temperature was maintained in a normal range (35.5-36.5°C) with the use of a TP400 T/Pump Heat Therapy System (Gaymar, Orchard Park, NY). The scalp was opened to expose the skull from approximately 5 mm anterior to bregma to 5 mm posterior to lambda. A small craniotomy allowed access to the brain and the microdialysis probe was aimed for prefrontal cortex, designated as frontal association cortex by the mouse brain stereotaxic atlas (Franklin and Paxinos, 1997). The aim site was 3.0 mm anterior to bregma and 1.6 mm lateral from the midline. The dorsal-ventral position of each probe was determined visually with the membrane resting fully within cortex. Probes were dialyzed continuously with Ringer's solution.

Experimental Design. After probe insertion, several initial dialysis samples were collected to ensure stable baseline ACh release from prefrontal cortex. Six dialysis samples (75 min) then were collected and the ACh quantified from these samples was averaged to determine baseline ACh release. A CMA/110 liquid switch then was used to change the dialysate from Ringer's alone to Ringer's containing a drug, allowing uninterrupted dialysis. Prefrontal cortex was dialyzed with either scopolamine (1, 3, 10, 30, or 100 nM), AF-DX116 (0.3, 1, 3, or 30 nM), pirenzepine (100 or 300 nM), or TTX (1 µM) while an additional six dialysis samples (75 min) were collected. Drug-induced changes in ACh release are presented as percent change from control and are expressed as mean ± S.E.M. Only one drug concentration was tested per experiment and each mouse was used for only one experiment. At the conclusion of each experiment, the microdialysis probe was removed from the cortex, the scalp wound was closed with wound clips, and the mouse was allowed to recover overnight.

Histology. After each experiment, brains were removed for histological processing to confirm probe placements. Brains were frozen in a bromobutane/isopentane bilayer, then mounted for coronal sectioning at 40 µm on a Bright model OTF cryostat (Huntingdon, Cambs, England). Cut sections were thaw-mounted onto gelatin coated slides, dried for 2 h under vacuum desiccation, and then fixed with paraformaldehyde vapors at 80°C. The fixed sections were stained with cresyl violet, backlit with a Northern Light illuminator (Imaging Research, St. Catherine's, ON, Canada) and digitized with a Cohu (San Diego, CA) charge-coupled device camera connected to a Macintosh G-3 computer. These digitized images were used to localize microdialysis probe placements as compared with a stereotaxic atlas (Franklin and Paxinos, 1997). Only results from experiments in which the dialysis probes were localized fully within prefrontal cortex are reported.

Data Analysis. Fifteen mice were used for the scopolamine concentration-response study (three mice per concentration) and data were analyzed by repeated measures of one-way analysis of variance followed by a Dunnett's test. The scopolamine, AF-DX116, and pirenzepine minimum-releasing concentration studies were conducted using six mice each and analyzed by t test. Data collected for the TTX study were obtained from three mice and analyzed by t test. Pertussis toxin studies were conducted in nine mice, and statistical significance was tested with analysis of variance, t test, and Mann-Whitney U test. Degrees of freedom and probability values are provided under Results.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Histological examination verified that microdialysis probes were placed fully within the prefrontal cortex. Figure 1A shows a digitized cresyl violet-stained coronal mouse brain section. This image illustrates a typical microdialysis probe site. Stereotaxic coordinates for the center of each probe were localized along the rostral-caudal axis by comparison to the mouse brain atlas (Franklin and Paxinos, 1997). A metric ruler was digitized with each sectioned brain to calibrate the position of each dialysis probe along the medial-lateral and dorsal-ventral axes. Figure 1, B and C, shows the probe centers from 45 experiments in relation to a B6 brain viewed from above (Rosen et al., 2000). Rostral-caudal probe center placements ranged from 2.78 to 3.06 mm anterior to bregma. Medial-lateral probe center placements ranged from 1.4 to 2.1 mm from the midline. Dorsal-ventral probe center placements ranged from 0.5 to 0.8 mm below the surface of the cortex.

View larger version (43K): [in this window] [in a new window]   Fig. 1.   Histological localization of microdialysis probes. A, this representative cresyl violet-stained coronal section is localized to approximately 2.96 mm rostral to bregma. The arrow points to the approximate midpoint of the probe-induced lesion. B, dorsal view of the left olfactory bulb and prefrontal cortex. Filled circles show probe center placements from all 45 experiments (Dialysis Site) in relation to stereotaxic target (Aim Site). C, dorsal view of mouse brain showing location of enlargement in B, probe placements relative to mouse brain, and location of bregma. Rostral is to the left.

Figure 2 shows that microdialysis delivery of 1 µM TTX decreased ACh release in prefrontal cortex. Figure 2A illustrates a typical time course of ACh release as a function of sequential dialysis sample. Figure 2B shows data summarized from three experiments. TTX significantly decreased ACh release by 55% (t = 5.5; df = 34; p < 0.0001). These findings are consistent with the view that ACh measured in this study represents vesicular release.

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Effect of 1 µM TTX on ACh release. A, microdialysis samples collected every 12.5 min are plotted as picomoles of ACh per sample versus ordinal sample number. These data from one experiment show that ACh release began to decrease during the first 12.5 min of TTX delivery. B, average ACh release from three experiments shows that TTX significantly decreased ACh release from 0.30 ± 0.03 pmol/12.5 min to 0.14 ± 0.01 pmol/12.5 min (asterisk indicates p < 0.0001).

The effect of scopolamine on the time course of ACh release during three separate experiments is shown in Fig. 3. Microdialysis administration of 1 nM scopolamine caused no change in ACh levels. In contrast, dialysis delivery of 10 nM scopolamine began to increase ACh release by the 100-min time point. Dialysis delivery of 100 nM scopolamine caused an increase in ACh release within the first 12.5 min of dialysis delivery.

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Time course of ACh release as a function of dialysis time and scopolamine concentration for three typical experiments. The inset contains chromatographic peaks corresponding to ACh and illustrates the excellent signal to noise ratio. The left chromatographic peak represents control levels of ACh (0.26 pmol) measured during dialysis with Ringer's (0 nM scopolamine). The middle peak represents 0.36 pmol ACh and was quantified during dialysis delivery of Ringer's containing 10 nM scopolamine. The right ACh peak (0.57 pmol) was caused by dialysis delivery of 100 nM scopolamine.

Figure 4 summarizes results from experiments in 15 mice showing that dialysis delivery of scopolamine caused a concentration-dependent increase in ACh release (F = 30.5; df = 4, 72; p < 0.0001). Multiple comparisons statistic (Dunnett's) showed that microdialysis delivery of 3, 10, 30, and 100 nM scopolamine each significantly increased ACh release over control levels (asterisks indicate p < 0.05).

View larger version (16K): [in this window] [in a new window]   Fig. 4.   ACh release as a function of scopolamine concentration. Each concentration of scopolamine was tested in three mice. Histograms represent mean percent change compared with control. Delivery of 1 nM scopolamine caused no significant change in ACh release. Delivery of 3 nM scopolamine caused a 14 ± 3.8% increase over control levels. A concentration of 10 nM scopolamine caused a 24 ± 3.5% increase in ACh release. Higher concentrations of scopolamine delivered by microdialysis caused greater increases in ACh release: 30 nM scopolamine induced an average 87 ± 17.2% increase, whereas 100 nM scopolamine induced an average 180 ± 23.4% increase in ACh release. Asterisks indicate significant (p < 0.05) increases compared with control.

The minimum ACh-releasing concentrations and maximum non-ACh-releasing concentrations for three mAChR antagonists are shown in Fig. 5. The most potent antagonists were scopolamine and AF-DX116. The minimum ACh-releasing concentration for scopolamine was 3 nM, which produced a 14.2 ± 3.8% increase in ACh release (t = 3.2; df = 38; p < 0.01). The minimum ACh-releasing concentration for AF-DX116 was also 3 nM, which caused a 19.6 ± 2.4% increase in ACh release (t = 6.74; df = 34; p < 0.01). The least potent mAChR antagonist was pirenzepine, with a minimum ACh-releasing concentration of 300 nM, causing a 27.0 ± 2.17% increase in ACh release (t = 11.4; df = 34; p < 0.0001).

View larger version (22K): [in this window] [in a new window]   Fig. 5.   ACh release as a function of mAChR antagonist concentration. Histograms summarize the results of 18 experiments. Rows show the relative potency of three mAChR antagonists for enhancing ACh release. Asterisks indicate that ACh release was significantly increased above control levels. Scopolamine and AF-DX116 were most potent and pirenzepine was least potent in significantly increasing ACh release. This order of potency suggests that the M2 subtype functions as an autoreceptor in mouse prefrontal cortex.

Figure 6 summarizes the results of nine experiments designed to determine the effect of PTX on AF-DX116-stimulated ACh release. Delivery of 3 nM AF-DX116 into prefrontal cortex caused a significant (p < 0.001) increase in ACh release (Fig. 6A). Microinjection of PTX into the cortical dialysis site blocked the AF-DX116-induced increase in ACh release (Fig. 6B).

View larger version (32K): [in this window] [in a new window]   Fig. 6.   PTX blocked the AF-DX116-evoked increase in ACh release. A, the 3 nM minimum releasing concentration of AF-DX116 significantly increased ACh release in prefrontal cortex (asterisk indicates p < 0.01). B, microinjection of PTX (5 ng) into the microdialysis site before dialysis delivery of AF-DX116 blocked the antagonist-induced increase in ACh release.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Three main findings emerged from the present study. First, microdialysis delivery of scopolamine to prefrontal cortex caused a concentration-dependent increase in ACh release, consistent with the hypothesis that ACh release in mouse prefrontal cortex is modulated by muscarinic autoreceptors. Second, quantification of the minimum ACh-releasing concentrations for three mAChR antagonists revealed the order of potency to be scopolamine = AF-DX116 > pirenzepine. This finding suggests that the M2 subtype modulates ACh release in prefrontal cortex of B6 mouse. Third, the discovery that pertussis toxin blocked AF-DX116-evoked ACh release is consistent with M2 autoreceptors functioning as modulators of ACh release. These three findings are considered below after noting the limitations of this study.

Limitations. Potential methodological limitations include the presence of neostigmine in the dialysis solution (Ringer's), diffusion of antagonists away from the microdialysis probe, and the possibility of ACh release increasing as a result of antagonist binding to postsynaptic mAChRs. These issues have been discussed in detail elsewhere (Billard et al., 1995; Baghdoyan et al., 1998). It must be acknowledged that although the present study clearly demonstrates modulation of ACh release by M2 receptors, the data do not prove that these M2 receptors are autoreceptors. Postsynaptic heteroreceptors may have caused increased ACh release via feedback circuitry. Evidence from cortical slice preparations, however, supports locally mediated enhancement of cortical ACh release (Iannazzo and Majewski, 2000a). The decreased ACh levels caused by TTX (Fig. 2) are consistent with the likelihood that ACh measured by microdialysis reflects synaptic release, rather than changes in ACh turnover. This interpretation follows from the fact that TTX eliminates the propagation of action potentials by blocking sodium channels (Catterall, 1992).

Multiple Lines of Evidence for M2 Autoreceptor Subtype in Prefrontal Cortex of B6 Mouse. Scopolamine caused an increase in cortical ACh levels (Figs. 3 and 4). The scopolamine effects on ACh release were rapid in onset (Fig. 3), sustained for the duration of antagonist delivery (Fig. 3), and concentration-dependent (Fig. 4). These data encouraged efforts to identify the mAChR subtype modulating cortical ACh release. The Fig. 5 results show minimum ACh-releasing concentrations of 3 nM for scopolamine and AF-DX116 and 300 nM for pirenzepine.

Functional Implications of M2 Autoreceptors as Modulators of ACh Release in Prefrontal Cortex. The present ACh release data from anesthetized mouse cannot directly address autonomic or behavioral functions regulated by prefrontal cortex (Uylings et al., 2000). Evidence that ACh release in prefrontal cortex is modulated by M2 receptors, however, does have clear functional implications. ACh is understood to be an essential regulator of cortical excitability (reviewed in Sarter and Bruno, 2000). The prefrontal cortex contributes to the regulation of attention and arousal (Groenewegen and Uylings, 2000). Deactivation of prefrontal cortex has been interpreted as "the single most salient feature common to both nonrapid eye movement and rapid eye movement sleep and may be a defining characteristic of sleep per se" (Braun et al., 1997). Behavioral (Horne, 1993; Harrison and Horne, 1997) and functional MRI (Drummond et al., 1999) data show that sleep deprivation interferes with consolidation of working memory. The foregoing data are consistent with the view that prefrontal cortical function is particularly vulnerable to sleep deprivation (Horne, 1993; Harrison and Horne, 1997) and anesthesia (Andrade, 1996; Casele-Rondi, 1996). Autoreceptor loss has been suggested as a factor contributing to Alzheimer's disease (Mesulam, 1998) characterized by cognitive decline and sleep disruptions. Age-related decrements in mAChR density recently have been demonstrated in human frontal cortex by PET imaging studies (Zubieta et al., 2001). Thus, the present results are consistent with functional evidence that cell excitability in prefrontal cortex is altered by mAChRs (Haj-Dahmane and Andrade, 1998).

Conclusions. The conclusion that ACh release is modulated by autoreceptors of the M2 subtype may be limited to B6 mouse. The results encourage future studies aiming to determine whether the M2 subtype modulates ACh release in different mouse strains. Such comparative studies may contribute to a better understanding of autoreceptors as a protein phenotype expressed by genetically distinct strains of mice. The relative value of using ACh release as a phenotype versus other pharmacological and behavioral methods remains to be demonstrated. The present data provide an essential first step in the evaluation of ACh as a phenotype versus other measures. Elucidating presynaptic mechanisms modulating cholinergic neurotransmission may lead to pharmacological strategies that can offset disease-related reductions in cortical ACh release (Buccafusco and Terry, 2000).