INTRODUCTION

There is extensive evidence that acetylcholine (ACh) is an important neuromodulatory transmitter affecting hippocampal-dependent learning and memory (Hagan and Morris 1988; Kaneko and Thompson 1997). In addition, Alzheimer's disease (AD) and aging-related cognitive impairments have been correlated with a loss of cholinergic function (Disterhoft et al. 1999; Kasa et al. 1997; Muir 1997). Consequently, enhancement of ACh effectiveness with procholinergic drugs, especially cholinesterase inhibitors (ChEI), has been a major thrust of AD treatment. Metrifonate, which is transformed nonenzymatically to 0,0-dimethyl 2,2-dichlorovinyl phosphate producing long-lasting inhibition of both acetylcholinesterase and butyrylcholinesterase (Nordgren et al. 1978), has been shown to improve cognitive performance in AD patients (Cummings et al. 1998; Morris et al. 1998; Pettigrew et al. 1998) and, in our hands, to enhance hippocampal-dependent learning in aging rabbits (Kronforst-Collins et al. 1997).

ACh modulates conductances in hippocampal neurons that mediate spike frequency adaptation (accommodation) and the postburst afterhyperpolarization (AHP) (Halliwell 1990;Lancaster and Adams 1986; Madison et al. 1987). The cholinergic reduction of the AHP and accommodation is of particular interest since the AHP and accommodation are reduced in CA1 and CA3 hippocampal pyramidal neurons after hippocampally dependent trace eyeblink conditioning (Moyer et al. 1996, 2000; Thompson et al. 1996b). Accommodation and AHP are enhanced in CA1 neurons from animals at ages that show learning deficits (Landfield and Pitler 1984; Moyer et al. 1992,2000). Metrifonate reduces the AHP and accommodation in CA1 neurons in young and aging rabbits due to the muscarinic action of increased ambient ACh levels (Oh et al. 1999).

The postburst AHP (medium and slow AHP) and accommodation are regulated by at least four classes of outward potassium currents (I_C,I_M, sI_AHP, andI_AHP), in addition the AHP is influenced by the mixed cation current,I_h (Maccaferri et al. 1993; Stocker et al. 1999; Storm 1990). Of these currents,I_M and sI_AHP are reduced by ACh in CA1 hippocampal neurons with ACh being 10-fold more potent at reducing sI_AHP thanI_M in young rats (Madison et al. 1987). Recently we have shown an age-associated enhancement of the sI_AHP and have hypothesized that enhancement of sI_AHP is integrally involved in age-related learning deficits (Power et al. 1999). The following experiments were undertaken to determine whether metrifonate reduces the postburst AHP and accommodation by reducing the sI_AHP.

METHODS

Hippocampal slices (300 μm) were prepared from young (2–3 mo) and aging (>31 mo) female New Zealand White rabbits using procedures identical to those previously published (Oh et al. 1999). Animal use procedures were approved by Northwestern University's Animal Care and Use Committee according to the standards of the US Department of Agriculture. Whole cell patch-clamp recordings were made from the soma of visually identified CA1 pyramidal neurons (Dodt and Zieglgansberger 1990) perfused with 31°C artificial cerebral spinal fluid (ACSF, composition, in mM: 124 NaCl, 3 KCl, 1.3 MgSO₄, 1.24 NaH₂PO₄, 2.4 CaCl₂, 26 NaHCO₃, and 10d-glucose). Patch electrodes (2–4 MΩ) were filled with (in mM) 2 K-ATP, 10 HEPES, 140 KMeSO₄, and 10 KCl; pH 7.25, 280 ± 10 mOsm. The use of MeSO₄⁻ in place of Cl⁻ reduces the rundown of the slow AHP current over time (Zhang et al. 1994).

Recordings were sampled at 5–10 kHz and filtered at 2 kHz using an Axopatch 1C amplifier (Axon Instruments, Foster City, CA) and controlled by custom software. Only neurons with an action potential amplitude >90 mV, membrane resistance >60 MΩ, and a stable series resistance <0.15 MΩ were considered healthy and used in the data set. The AHP was evoked by a 100-ms voltage step to 0 mV from a holding potential of −55 mV. The −55-mV holding potential should eliminate the influence of I_h, which activates at more hyperpolarized potentials. The medium (I_AHP,I_M, andI_C) and slow (sI_AHP) AHP components were separated using a curve peeling technique (Lancaster and Adams 1986; Womble and Moises 1993) (Fig.1). The sI_AHP amplitude was also determined 1 s after pulse offset, after medium AHP currents are no longer active (Sah 1996; Stocker et al. 1999).

Voltage-clamp measurements were taken immediately before and 20 min after application of metrifonate (young 50 μM; aging 100 μM). These concentrations of metrifonate were found to reduce the AHP and accommodation in sharp electrode current-clamp recordings comparably between young and aging neurons without inducing a bursting behavior (Oh et al. 1999). Sodium and potassium channel blockers were omitted to achieve the same cholinergic activity levels present in our previous current-clamp study (Oh et al. 1999). In some neurons, current-clamp measures of the AHP were also taken. In current clamp, the postburst AHP was evoked from a −68-mV holding potential using a 100-ms depolarizing current step that elicited four action potentials.

Age-related differences were analyzed with ANOVA. Drug-related effects were analyzed using paired t-tests. The correlation between current-clamp and voltage-clamp measures was tested using Fisher'sr to z test.

Metrifonate was a gift from Bayer (West Haven, CT). KMeSO₄ was purchased from ICN (Aurora, OH). All other drugs were purchased from Sigma (St. Louis, MO).

RESULTS

Age effects

The AHP measures and passive membrane properties are given in Table 1. The current-clamp AHP tended to be enhanced in neurons from aging animals [F(18, 1) = 4.09, P = 0.0583] as had been previously observed (Moyer et al. 1992; Oh et al. 1999). In voltage-clamp mode, a 55-mV, 100-ms voltage step from a holding potential of −55 mV evoked a biphasic outward tail current in most neurons that slowly decayed back to baseline (Fig. 1). This tail current was separated into a fast (I_fast) and a slow (I_slow) decaying component representing the medium (I_AHP,I_M, andI_C) and slow (sI_AHP) AHP. The extrapolated amplitude of I_slow [F(18, 1) = 6.56, P = 0.0196] and the tail current measured 1 s after pulse offset (1s_AHP) [F(20, 1) = 15.85, P = 0.0007] showed age related enhancements. There was no age-associated change inI_fast. The 1s_AHP current was correlated with the current-clamp AHP area (r = −0.453, P = 0.0441) and duration (r = 0.527, P = 0.0155). Neurons from aging animals had a more hyperpolarized resting potential (RMP) [F(19, 1) = 5.206, P = 0.0342] and a decreased membrane resistance (R_m) [F(19, 1) = 5.40, P = 0.0313].

Metrifonate effects

Bath application of metrifonate caused a reduction of the postburst AHP (Table 1). Metrifonate also reduced the sI_AHP, measured as eitherI_slow or 1 s after pulse offset, in neurons from both young and aging animals (Table 1, Fig.2). Metrifonate did not produce a significant reduction in I_fast, nor did the effect of metrifonate on the fast component differ from that of saline controls.

DISCUSSION

An age-associated enhancement of the slow (sI_AHP), but not the fast (I_AHP,I_C, andI_M), component of the AHP tail current was observed. Furthermore metrifonate decreased the slow tail component (sI_AHP) in CA1 neurons from both young and aging animals at doses that reduce postburst AHP and accommodation in sharp-electrode current-clamp recordings (Oh et al. 1999). We did not observe a metrifonate-induced reduction in the fast component of the AHP tail current. ACh is more potent at reducing sI_AHP thanI_M (Madison et al. 1987). However, since we were unable to adequately peel the fast component of the AHP in every instance and the relative contribution of I_M to the fast tail component was not determined, it is unclear whether metrifonate reducedI_M as well as sI_AHP.

Metrifonate was shown to be less potent at reducing the postburst AHP and accommodation of CA1 neurons of aging than young rabbits in our previous current-clamp studies that evaluated dose-response relationships in the two age groups (Oh et al. 1999). Metrifonate doses that produced a comparable reduction of the postburst AHP in young and aging neurons (Oh et al. 1999), caused a similar reduction of the sI_AHP. Thus metrifonate appears to modulate neuronal excitability similarly in neurons from young and aging subjects. It is likely that the age-related potency differences are due to reduced levels of endogenous ACh release observed in aging subjects in vivo (Vannucchi et al. 1997).

In summary, metrifonate reduces the AHP, in particular, the sI_AHP. The AHP and accommodation are reduced after hippocampally dependent trace eyeblink conditioning (Moyer et al. 1996; Thompson et al. 1996b). The sI_AHP is enhanced in CA1 neurons from aging rabbits (Power et al. 1999), a group that is learning-impaired (Thompson et al. 1996a). Metrifonate facilitates learning in aging rabbits (Kronforst-Collins et al. 1997). Similar results were obtained with the M1 muscarinic agonist CI-1017, and the L-type calcium channel blocker, nimodipine. Both of these compounds reduce the AHP and accommodation in vitro and facilitate acquisition of trace eyeblink conditioning when administered to aging rabbits (Deyo et al. 1989; Moyer et al. 1992; Weiss et al. 2000). As a whole, these data support our hypothesis that the sI_AHP is integrally involved in age-related learning impairments and suggest that the functional consequences of metrifonate administration are mediated through modulation of the sI_AHP. More generally, they suggest that the ability of a drug to reduce the sI_AHP may be a good predictor of that compound's ability to facilitate learning, especially in aging subjects.

FOOTNOTES

• Address for reprint requests: J. F. Disterhoft, Dept. of Cell and Molecular Biology, Northwestern University Medical School, Searle 4-427, 303 E. Chicago Ave., Chicago, IL 60611-3008 (E-mail:[email protected]edu).