Strain-related variations of AMPA receptor modulation by calcium-dependent mechanisms in the hippocampus: contribution of lipoxygenase metabolites of arachidonic acid

Abstract

Several studies have demonstrated that C57 and DBA mice exhibit behavioural differences in diverse learning tasks as well as variations in the expression of long-term potentiation (LTP) in the hippocampus. In the present investigation, we tested the possibility that these differences between the two strains might be attributable to differential regulation of hippocampal α-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptors by calcium-dependent mechanisms. Using in vitro receptor autoradiography, we found that calcium treatment of C57 mice sections resulted in a marked increase of ³H-AMPA binding in areas CA₃ and CA₁ of the hippocampus and in the dentate gyrus. However, we discovered that the ability of calcium to upregulate ³H-AMPA binding in the DBA strain was much lower than in corresponding regions from the C57 strain. Western blot and immunohistochemical experiments indicated that truncation of AMPA receptor subunits by calcium-dependent mechanisms was possibly not responsible for the binding differences, as no significant variations in glutamate receptor subunit 1 (GluR1) and GluR2/3 immunoreactivity were observed between the two strains after calcium treatment. Interestingly, we found that strain-related variations in the regulation of ³H-AMPA binding by calcium were totally eliminated when brain sections were preincubated with preferential inhibitors of lipoxygenase (LO) pathways of arachidonic acid (AA) metabolism. Taken together, these results suggest that calcium-induced regulation of AMPA receptors varies between the two strains and that this variation might be linked to the production of specific AA metabolites.

Introduction

Recent literature has documented profound behavioural and electrophysiological differences between C57BL/6 (C57) and DBA/2 (DBA) mice. From a behavioural perspective, these mice exhibit marked differences in hippocampal learning tasks, while both strains seem to perform well in non-hippocampal tasks [2], [18], [19], [26], [36], [40], [42], [45], [46], [47]. For instance, when tested for spatial learning performance tasks, such as the Morris water maze and the radial arm maze [32], DBA mice show poor learning capabilities in comparison to C57 [2], [13], [27], [31]. Studies on the spatial learning abilities of C57 and DBA have demonstrated that strain-related differences in behaviour might be linked to hippocampal anatomy [35], more specifically to the size of the hippocampal intra- and infrapyramidal mossy fiber terminals [5], [13], [14], as well as the density of pyramidal cells in the dorsal hippocampus [49]. At the electrophysiological level, hippocampal long-term potentiation (LTP), an electrophysiological model of memory formation, is also found to be more robust in C57 than in DBA, a phenomenon that is frequently correlated with observed differences in behaviour [22], [31], [33], [34].

It is widely assumed that change in the α-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) subtype of glutamate receptors (GluRs) is an important component of LTP expression in the hippocampus. For instance, increased ³H-AMPA binding in the hippocampus correlates well with heightened synaptic responses after LTP formation [4], [29], [43]. The exact biochemical mechanisms underlying alterations in AMPA receptor properties are still a matter of debate, but a large number of experiments support the notion that activation, by calcium ions, of protein kinases and proteases could be critical for LTP expression [3], [28]. Several arguments have been advanced to support the hypothesis that activation of calcium-dependent lipases may be part of the molecular mechanisms involved in LTP [30]. N-methyl-d-aspartate (NMDA) receptor activation has been found to generate long-lasting enhancement of endogenous phospholipase A₂ (PLA₂) activity [23], and various PLA₂ inhibitors have been shown to block LTP formation in area CA₁ of the hippocampus (for review, see Ref. [30]). On the other hand, incubation of hippocampal tissues with exogenous PLA₂ has been observed to augment, similarly to LTP, AMPA receptor binding [15], [16], [25] without altering the binding of ligands to NMDA receptors [30]. However, we recently discovered that PLA₂ could elicit, depending on its activity, either increased or decreased agonist binding of AMPA receptors, suggesting a possible role of PLA₂ in bidirectional control of synaptic efficacy at glutamatergic neurons [12]. At the molecular level, generation of lipoxygenase (LO) by-products of arachidonic acid (AA) metabolism appears to play an important role in mediating PLA₂-induced bidirectional modulation of AMPA receptor binding [12], [25]. In agreement with this notion, our recent patch clamp experiments revealed that LO by-products of AA are capable of selectively influencing AMPA receptor-mediated synaptic transmission in CA₁ pyramidal cells [41].

In this study, we have hypothesized that behavioural and electrophysiological (i.e., LTP) variations observed among mouse strains might be associated with differences in calcium-induced changes in AMPA receptors. This concept was examined by qualitative as well as quantitative analyses of ³H-AMPA binding to investigate how treatment of previously frozen brain sections with calcium is capable of differentially modulating AMPA receptor properties in the hippocampus of C57 and DBA mice. As PLA₂ is known to play a central role in the regulation of AMPA receptor binding properties, particular attention was also given to comparing the effects of LO and cyclooxygenase (COX) inhibitors of AA metabolism on the calcium-induced changes of AMPA binding in C57 and DBA mouse brain sections. Previous reports have indicated that AMPA receptor subunits are also very sensitive in their C-terminal domain to truncation by calcium-dependent proteases [6], [7], [8]. Hence, we examined, by Western blot analyses and immunohistochemistry, whether calcium-induced truncation might be altered differentially in these two mouse strains.