Ca2+/calmodulin protein kinase II and memory: learning-related changes in a localized region of the domestic chick brain
1 Institute of Physiology, Georgian Academy of Sciences, 14 L. Gotua Str., Tbilisi 380060, Republic of Georgia
2 Sub-Department of Animal Behaviour, University of Cambridge, Madingley, Cambridge CB3 8AA, UK
Abstract
The role of calcium/calmodulin-dependent protein kinase II (CaMKII) in the recognition memory of visual imprinting was investigated. Domestic chicks were exposed to a training stimulus and learning strength measured. Trained chicks, together with untrained chicks, were killed either 1 h or 24 h after training. The intermediate and medial hyperstriatum ventrale/mesopallium (IMHV/IMM), a forebrain memory storage site, was removed together with a control brain region, the posterior pole of the neostriatum/nidopallium (PPN). Amounts of membrane total CaMKII (tCaMKII) and Thr286-autophosphorylated CaMKII (apCAMKII) were measured. For the IMHV/IMM 1 h group, apCaMKII amount and apCAMKII/tCaMKII increased as chicks learned. The magnitude of the molecular changes were positively correlated with learning strength. No learning-related effects were observed in PPN, or in either region at 24 h. These results suggest that CaMKII is involved in the formation of memory but not in its maintenance.
, 百拇医药
Introduction
It has often been suggested that learning and memory depend on synaptic modification (e.g. Tanzi, 1893; Cajal, 1911; Konorski, 1948; Hebb, 1949). This suggestion has stimulated many studies of synaptic plasticity, and of proteins associated with it. Prominent among such proteins has been Ca2+/calmodulin-dependent protein kinase II (CaMKII), the main protein of the postsynaptic density (PSD) in the vertebrate central nervous system (Grab et al. 1981; Kennedy et al. 1983; Kelly et al. 1984; Kennedy, 1998). Smilowitz et al. (1981) demonstrated that the phosphorylation of acetylcholine receptors was regulated by a calcium/calmodulin-dependent kinase. More recently CaMKII in the mammalian hippocampus has been shown to phosphorylate and promote the delivery of -amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptors into the postsynaptic membrane (Barria et al. 1997; Mammen et al. 1997; Chen et al. 2000; Hayashi et al. 2000; Passafaro et al. 2001; Zhu et al. 2002). Presynaptically, CaMKII may increase neurotransmitter release by the phosphorylation of synapsin I and the subsequent dissociation of synapsin I from synaptic vesicles (Greengard et al. 1993). CaMKII plays a critical role in hippocampal long-term potentiation (Fukunaga et al. 1993; Otmakhov et al. 1997; Song & Huganir, 2002; Lisman, 2003) and it has been suggested that the kinase is involved in synaptic changes underlying memory (Horn, 1985, pp. 219–221; Lisman et al. 2002).
, 百拇医药
Transient exposure of CaMKII to Ca2+/calmodulin activates the kinase, which, through autophosphorylation at Thr286, remains active after Ca2+ is withdrawn (Miller & Kennedy, 1986). This property of the kinase is one of the reasons for proposing that it could serve as a molecular switch, rather than a trigger, that is capable of long-term memory storage (see Lisman et al. 2002 for review). A prediction of this hypothesis is that the enzyme should remain in the autophosphorylated state so long as memory persists. To test this hypothesis it is necessary to study the effect of learning on CaMKII in a brain region in which memories are formed. One type of learning that offers an opportunity for such a study is visual imprinting, whereby the young of certain species learn the features of, and subsequently recognize, an object as a result of being exposed to it (Bateson, 1966; Sluckin, 1972; Horn, 1985; Bolhuis, 1991). Training conditions may so be contrived that, before learning, no visual experience has made its mark on the brain. This property of visual imprinting has made it possible to identify a brain region in which information is stored. The region is the intermediate and medial hyperstriatum ventrale (IMHV; intermediate and medial mesopallium (IMM) in the new terminology, Reiner et al. 2004) of the domestic chick forebrain. The evidence for a role of this region in memory is reviewed in detail elsewhere (Horn, 1985, 1998, 2000, 2004). Some of the cellular and synaptic changes occurring in the region after imprinting may be linked with CaMKII activation (see for example, Bradley et al. 1981; Horn et al. 1985; Sheu et al. 1993; McCabe et al. 2001; Meredith et al. 2004). We have therefore inquired whether imprinting leads to changes in the amount of the -subunit, CaMKII, and its autophosphorylation state at Thr286, in a membrane fraction of IMHV/IMM. The posterior pole of the neostriatum (PPN; now nidopallium), a region where learning-related changes following imprinting have not been observed (Horn et al. 1979; Solomonia et al. 1997, 1998, 2003b), was studied as a control region.
, 百拇医药
A preliminary account of these results has been published (Solomonia et al. 2003a).
Methods
Behavioural procedures
Chicks were hatched, trained and reared as previously described (McCabe et al. 1982; Solomonia et al. 1997, 2003b). Twenty-two batches of chicks (Ross 308, Grampian Country Chickens) were used. Chicks were hatched and reared in darkness. When 22–28 h old, some of the chicks in each batch were trained by exposure for 1 h in individual running wheels to an imprinting stimulus (a rotating, internally illuminated red box (RB)). As a chick attempted to approach the RB the chick rotated the running wheel. The number of revolutions of the wheel were counted to provide a measure of approach activity. Chicks that have learned the characteristics of the training stimulus prefer it to an alternative stimulus. Ten minutes after training, chicks' preferences were measured by showing each chick sequentially the RB and an alternative stimulus, a rotating, internally illuminated blue cylinder (BC). A preference score was then calculated, providing a measure of preference and hence of learning strength (Sluckin, 1972; Bolhuis et al. 2000). The preference score was the chick's approach to the RB expressed as a percentage of approach to RB plus approach to BC. If RB and BC were approached equally, the preference score would be 50 (no choice; indicative of no learning (Bolhuis et al. 2000)). If the chick approached only the training object, the preference score would be 100 indicating a strong preference for the training object. Preference scores were used to relate the strength of learning with changes in CaMKII. This was achieved (i) by correlation analysis and (ii) by subdividing chicks into three groups based on their preference scores. This procedure makes it possible to tease apart effects of learning from non-specific effects of training (see Discussion). Chicks with preference scores 46–60 were classified as poor learners, chicks with preference scores 80% as good learners and those with preference scores between 61 and 79 were classified as intermediate learners (Solomonia et al. 2003b). One group of chicks in each batch served as untrained controls.
, 百拇医药
Chicks from 11 batches were killed 1 h after the end of training, and from the other 11 batches 24 h after training. Four pieces of tissue were removed from each brain, from the left and right IMHV/IMM and the left and right PPN, and frozen immediately on dry ice. Details of the method for removal of tissue from the IMHV/IMM and PPN, respectively, are described elsewhere (Horn, 1991; Solomonia et al. 1998). Tissue from each region and side (e.g. right IMHV/IMM) and for a particular training condition (e.g. intermediate learners) was obtained from three chicks having similar preference scores (i.e. all three chicks being either good learners, or intermediate learners, or poor learners) and pooled to form a single sample. This pooling was necessary to provide sufficient membrane protein for analysis. Similar samples were obtained from untrained chicks. Thus, for example, the right IMHV/IMMs from three chicks that were intermediate learners constituted a single sample. Each sample from the trained chicks was assigned a preference score that was the mean for the three chicks. Each set of experimental conditions (good learners, intermediate learners, poor learners, untrained) yielded 16 samples (left and right IMHV/IMM, left and right PPN x 4 experimental conditions). These conditions were replicated 11 times yielding 176 samples (11 x 16) for the 1 h group: that is, 88 IMHV/IMM and 88 PPN samples. These samples were obtained from 132 chicks (11 replications x 4 experimental conditions x 3 chicks per sample). The same number was used for the 24 h group. Samples were coded after collection and all further procedures conducted blind, that is without knowledge of the treatment of the chicks which contributed to a sample.
, http://www.100md.com
Chicks experienced no suffering prior to death and were covered by UK Home Office licence. The number of animals was estimated to be the minimum required for adequate statistical analysis.
Electrophoresis and immunoblotting
Pooled samples were rapidly homogenized in 20 mM Tris-HCl (pH 7.4), 0.32 M sucrose, 1 mM ethylenediamine-tetraacetic acid, 1 mM sodium orthovanadate, 10 mM sodium pyrophosphate, 0.5 mM ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid, 1 mM phenylmethanesulphonyl fluoride and 5 μM okadaic acid (to inhibit phosphatases 1 and 2A; see Strack et al. 1997) and centrifuged at 1000 g for 10 min. The supernatant was further centrifuged at 15 000 g for 20 min. The pellet was washed once and is referred to as the P2 membrane fraction (Cotman & Taylor, 1972). The P2 membrane fraction was dissolved in 5% SDS solution. Protein concentration was determined in quadruplicate using a micro bicinchoninic acid protein assay kit (Pierce Biotechnology, Inc., Rockford, IL, USA). Aliquots containing 30 μg of protein and of equal volume were applied to the gels. SDS gel electrophoresis and Western blotting were carried out as previously described (Solomonia et al. 1998, 2000). After protein had been transferred onto nitrocellulose membranes, the membranes were stained with Ponceau S solution to confirm transfer and the uniform loading of the gels. The membranes were then washed with phosphate-buffered saline + 0.05% Tween 20. In each batch, samples were run in duplicate and one filter was stained with monoclonal antibody (Chemicon, Temecula, CA, USA) against CaMKII; the antibody reacts with both phosphorylated and non-phosphorylated forms of the enzyme. The second filter was stained with a monoclonal antibody (Sigma) against phosphorylated CaMKII, which reacts only with Thr286-autophosphorylated CaMKII. Standard immunochemical procedures using peroxidase-labelled secondary antibodies and SuperSignal West Pico Chemiluminescent substrate (Pierce) were carried out. The blots were then exposed with intensifying screens to X-ray films preflashed with Sensitize (Amersham). The optical densities of bands corresponding to total CaMKII (tCaMKII) and Thr286-autophosphorylated CaMKII (apCAMKII) were measured. The autoradiographs were calibrated by including in each gel four standards comprising membrane fraction from the IMHV/IMMs of a group of untrained chicks. Each standard contained one of a range of amounts of total protein (10–60 μg). One set of standards was used for the 1 h experiments and another set for the 24 h experiments. Optical density was proportional to the amount of tCaMKII and apCAMKII. To obtain the data given in Figs 3, 4 and 5 the optical density of each band from an experimental sample (e.g. intermediate learners) was divided by the optical density which, from the calibration of the same autoradiograph, corresponded to 30 μg of total protein in the standard (Solomonia et al. 1997, 2003b). The data expressed in this way will be referred to as ‘relative amount’ of apCaMKII and tCaMKII. A measure of the proportion of CaMKII in a sample that was autophosphorylated (referred to as apCaMKII/tCaMKII) was obtained as the expression (relative amount of apCaMKII/relative amount of tCaMKII) for that sample. The two relative amounts of protein in this expression are ratios (optical density of sample/optical density of corresponding 30 μg standard) and depend on the level of autophosphorylation in the corresponding standards. As a result, the ratio apCaMKII/tCaMKII is not the actual proportion of CaMKII autophosphorylated in a sample but is linearly related to that value.
, http://www.100md.com
Error bars represent ± 1 standard error of the mean. A, IMHV/IMM. There was a significant effect of Training Condition: F3,26= 5.74, P= 0.004. *t= 2.39, 26 d.f., P= 0.024; **t= 3.21, 26 d.f., P= 0.004; ***t= 3.55, 26 d.f., P= 0.002. B, PPN. The ANOVA showed no significant difference between the four means: F3,26= 0.88, P= 0.46. The black bar indicates the mean value for all trained chicks combined. This combined mean is not significantly different from the mean of the untrained chicks: t= 1.90, 27 d.f., P= 0.07.
, 百拇医药
Error bars represent ± 1 standard error of the mean. A, IMHV/IMM. The amount of tCaMKII in the IMHV/IMM was significantly affected by Training Condition: F3,26= 4.71, P= 0.009. *t= 2.11, 26 d.f., P= 0.044; **t= 2.34, 26 d.f., P= 0.027; ***t= 3.68, 26 d.f., P= 0.001. n, number of samples. B, PPN. There was a significant effect of Training Condition: F3,26= 8.53; P < 0.001. The means of all three groups of trained chicks were statistically homogeneous: F2,16= 3.14; P= 0.07. However, each of these measures was significantly greater than the mean of the untrained chicks: **t= 2.86, 26 d.f.; P= 0.008. ***(intermediate learners versus untrained) t= 3.71, 26 d.f.; P < 0.001; ***(good learners versus untrained) t= 4.85, 26 d.f.; P < 0.001. The combined mean for the three trained groups was 1.38 ± 0.03 and was significantly greater (by 27%) than that (1.09 ± 0.06) of the untrained chicks: t= 4.83, 24 d.f., P < 0.001.
, http://www.100md.com
Statistical analysis
Because of the functional differences between the IMHV/IMM and the PPN (see Introduction), data from the two regions and for the different time groups (1 h and 24 h) were analysed separately. Data were subjected to a split-plot analysis of variance (ANOVA) with the fixed factors Training Condition (untrained, poor learners, intermediate learners, good learners) and Side (left, right), replicated in Blocks. In consequence, comparisons within chicks (factor Side) were made using different standard errors from those used for comparisons between groups of chicks (factor Training Condition). Preference score was subjected to an arcsin square root transformation (Mosteller & Youtz, 1961) before analysis, in order to normalize the data. The least squares regression of each biochemical measure on preference score, approach during training and approach during the preference test were conducted using a general linear model. Values of apCaMKII amount and standard errors in Figs 3, 4 and 5 are given after subtraction of Block mean for trained and untrained chicks and then adding back the overall mean for all chicks. This procedure eliminates variation between blocks, and the significance of correlation coefficients was tested using correspondingly fewer degrees of freedom for error, 10 degrees of freedom being removed for the 11 blocks (see, e.g. Snedecor & Cochran, 1989).
, 百拇医药
Results
Behaviour
Preference scores at both 1 h and 24 h were significantly greater in good learners than in intermediate or poor learners, and significantly greater in intermediate learners than in poor learners (Fig. 1). The full statistical analysis is given in the legend to Fig. 1. Approach activity during training did not differ significantly between the three trained groups of chicks at either time.
, http://www.100md.com
The dashed line indicates the no-preference score of 50. Error bars represent ± 1 standard error of the mean. Preference scores differed significantly among experimental groups both at 1 h (A; F2,17= 176.1, P < 0.001) and 24 h (B; F2,19= 195.6, P < 0.001). The mean preference scores of good learners were significantly greater than those of intermediate learners and poor learners (1 h, t= 10.6 and 18.4, respectively, 17 d.f., P < 0.001; 24 h, t= 11.2 and 19.7, respectively, 19 d.f., P < 0.001). The mean preference scores of intermediate learners were significantly greater than those of poor learners (2 h, t= 8.1, 17 d.f., P < 0.001; 24 h, t= 8.5, 19 d.f., P < 0.01). ***Preference score significantly greater than 50, t > 20.9, P < 0.001. n, number of samples.
, http://www.100md.com
Immunostaining of apCaMKII and tCAMKII
The antibodies against apCaMKII and tCaMKII reacted with the same single 52 kDa band, corresponding to the CaMKII of chick brain (see Fig. 2 and Solomonia et al. 2000).
A and B each contain an experimental replicate (right and left IMHV/IMM and right and left PPN of good learners, intermediate learners, poor learners and untrained chicks). All samples in an experimental replicate were analysed using a single gel. The discontinuity between lanes 4 and 5 represents irrelevant intercalated lanes, which have been excluded. The optical density of a band was linearly related to the amount of protein in the band. Each lane was derived from a single sample. Lanes 1–4 are from good learners, 5–8 from intermediate learners, 9–12 from poor learners and 13–16 from untrained chicks. Odd numbered lanes are from the left side, even lanes from the right. All procedures after training were conducted ‘blind’. Measurements for individual samples (cf. Fig. 4) are not of course the same as their respective experimental group means (cf. Figs 3 and 5).
, 百拇医药
Changes in CaMKII
One hour after the end of training. IMHV/IMM. The amounts of apCaMKII and tCaMKII did not differ significantly between the left and right sides: apCaMKII, F1,36= 0.01, P= 0.93; tCaMKII, F1,35= 0.28, P= 0.60. Mean values ±S.E.M. were: apCaMKII, left 0.92 ± 0.03, right 0.93 ± 0.03; tCaMKII, left 1.17 ± 0.02, right 1.19 ± 0.04. Furthermore, the interactions between Side and Training Condition were not significant. Accordingly, data from the left and right sides were combined.
, http://www.100md.com
There was a highly significant effect of Training Condition on the mean amounts of apCaMKII in IMHV/IMM: F3,26= 5.74, P= 0.004. There was significantly more apCaMKII in good learners than in intermediate learners (33%), poor learners (22%) and untrained chicks (35%) (see Fig. 3A and Table 1). Amongst the trained chicks, the mean values of poor and intermediate learners were statistically homogeneous (t= 1.16, 26 d.f., P= 0.26), the combined mean value being 0.88 ± 0.02. The mean amount of apCAMKII in the good learners was significantly higher (by 27%; t= 5.39, 26 d.f., P < 0.001) than this combined mean. Together, these results imply that the effects of training were expressed amongst chicks with the higher preference scores. Consistent with this implication (i) for intermediate and good learners there was a significant linear fit between the amount of apCaMKII and preference score (Fig. 4A) and (ii) for all groups of trained chicks, the relationship between the amounts of apCAMKII and preference score was fitted significantly, F2,15= 8.09, P= 0.004, by an exponential function (Fig. 4B). The fitted line in Fig. 4B was interpolated to yield a value of apCaMKII corresponding to a preference score of 50. The score of 50 indicates an absence of preference for either stimulus and is characteristic of chicks that have not become imprinted. The interpolated amount of apCaMKII was close to the mean value for the untrained chicks. These findings imply that exposure to the training stimulus alone did not lead to an increase in the amount of apCaMKII; apCAMKII levels only increase as the chicks learn. The separate variances of good, intermediate and poor learners, and untrained chicks, did not differ significantly from one another.
, 百拇医药
A and B refer to the relationship between preference score and relative amount of autophosphorylated CaMKII (apCaMKII); for details of how relative amount was calculated, see Methods. Lines were fitted by the method of least squares. A, intermediate learners and good learners. These data were fitted by the expression y=a+bx; r= 0.72, 10 d.f., P= 0.01. B, all trained chicks. These data are fitted by the exponential function y=a+bcx, F2,15= 8.09, P= 0.004. The arrow indicates the mean amount of apCaMKII +S.E.M. in untrained chicks. The interpolated value of the amount of apCaMKII for a ‘no-preference’ score of 50 in trained chicks is estimated from the dashed line. The black bar on the ordinate represents ± 1 standard error for the fitted line at this preference score. The interpolated value is very close to the mean for the untrained chicks. C, relationship between the ratio apCaMKII/tCaMKII for IMHV/IMM and preference score in intermediate and good learners. Data were fitted by a linear function; r= 0.70, 9 d.f., P= 0.02, fitted by the method of least squares.
, 百拇医药
The amount of tCaMKII was significantly higher in good learners than in both poor learners and untrained chicks and significantly higher in intermediate learners than in untrained chicks (see Fig. 5A and Table 1). In spite of the systematic increase in the mean values from poor through to good learners, the correlation between tCaMKII and preference score was not significant: r= 0.32, 16 d.f., P= 0.20.
The ratio apCaMKII/tCaMKII increased significantly with preference score for intermediate and good learners (see Fig. 4C). This finding suggests that for these groups of chicks the amount of apCAMKII rose faster with preference score than did the amount of tCaMKII: as the strength of learning increased there was a selective increase in the level of autophosphorylation.
, 百拇医药
There were no significant correlations between approach activity during training and approach activity during the preference test on the one hand and the amounts of apCaMKII and tCaMKII on the other (see Table 2).
One hour after the end of training. PPN. There was significantly more apCaMKII, but not tCaMKII, in the left PPN than in the right: apCaMKII, F1,35= 4.97, P= 0.032; tCaMKII, F1,32= 0.41, P= 0.53. The mean values ±S.E.M. were: apCamKII, left 1.12 ± 0.03, right 1.04 ± 0.04; tCaMKII, left 1.30 ± 0.04, right 1.30 ± 0.03.
, 百拇医药
For apCaMKII there were no significant effects of training on either the left or the right sides: the Training Condition–Side interaction was not significant: F3,35= 1.05, P= 0.38. To examine the effects of training on apCaMKII level the mean values for right and left sides were therefore combined. The combined means are given according to training condition in Fig. 3B. There were no significant differences between the means for the three groups of trained chicks (Fig. 3B). The mean value for all trained chicks combined was 13% greater than that of untrained chicks but the difference was not significant: t= 1.90, 27 d.f., P= 0.07 (Fig. 3B).
, 百拇医药
For tCaMKII, data from left and right PPN were combined. The means of the three groups of trained chicks were statistically homogeneous and each was significantly greater than that of the untrained chicks (Fig. 5B). The mean value for all trained chicks combined (see Fig. 5B) was significantly higher, by 27%, than the mean of the untrained chicks: t= 5.41, 24 d.f., P < 0.001. There were no significant correlations between amount of apCaMKII and tCaMKII on the one hand and preference score, approach activity during training or approach activity during the preference test on the other (Table 2). The ratio apCaMKII/tCaMKII was not significantly correlated with preference score, either for all trained chicks (r= 0.32; P= 0.14), or for intermediate and good learners (r= 0.07; P= 0.79). The exponential fit for all trained chicks was also not significant: F2,15= 1.98, P= 0.60.
, http://www.100md.com
Twenty-four hours after the end of training. IMHV/IMM and PPN. For both the IMHV/IMM and the PPN, no significant effects of Training Condition (Table 2) were found and there were no significant correlations between any of the measures of CaMKII protein on the one hand, and preference score, approach activity during training or approach activity during the preference test on the other (Table 2).
Discussion
If CaMKII is essential for the storage of information acquired through learning, then, in a brain region storing that information, the activity of the kinase should correlate with the strength of learning. The present study has addressed this issue and further inquired whether such changes are time dependent. We have demonstrated that (i) the amount of CaMKII autophosphorylated at Thr286 (apCaMKII) and (ii) the ratio of apCaMKII/tCaMKII are changed in a learning-specific way (see below) in a brain structure storing information acquired through imprinting. The effects are regionally specific, since they occur in the IMHV/IMM but not in the PPN; they are time dependent since they occur 1 h after the end of training, not at 24 h.
, 百拇医药
The evidence that the changes measured 1 h after training in the amount of membrane apCAMKII in IMHV/IMM are a function of learning is as follows. (i) There was a significant, positive correlation between preference score and the amount of apCaMKII. (ii) The mean amount of apCAMKII present in good learners was significantly higher than in intermediate learners, poor learners or untrained chicks. (iii) The mean amount of apCAMKII present in the untrained controls did not differ significantly from the amount in the poor learners or extrapolated from the regression line for a no-choice preference score of 50. These findings taken together imply that sensory stimulation, movement, arousal, motivational state and non-specific factors (collectively referred to as ‘side-effects of training’) did not influence the amount of apCaMKII in IMHV/IMM. Were they to have done so, the mean amount of apCaMKII in poor learners, which had been exposed to the whole training procedure, would be expected to be higher than that of the untrained controls. This is not the case. Furthermore, changes in the amount of apCaMKII in IMHV/IMM could not be attributed to the preference test, since poor learners were given this test and untrained chicks were not: the mean amount of apCaMKII in these two groups was not significantly different. (iv) The pattern of apCaMKII means for IMHV/IMM in Fig. 3A contrasts with that for the PPN (Fig. 3B). In the PPN the mean amounts of apCaMKII within the trained group were statistically homogeneous. (v) The amount of apCaMKII was not affected by approach activity during training or during the preference test. (vi) The changes in apCaMKII were not an effect of ontogenetic selection whereby chicks hatched with, for example, high levels of apCaMKII become good learners, and those with low levels of apCaMKII become poor learners. According to this ontogenetic selection hypothesis, therefore, good and poor learners are subclasses of untrained chicks with ontogenetically high or low levels of apCaMKII respectively (see Horn, 2004). If this were so, the variance in levels of apCaMKII in untrained chicks, being the ‘parent’ population, should be higher than those of the selected subgroups of poor, intermediate and good learners. This was not the case. There were no significant differences between the variances of the four groups of chicks. These findings are not therefore consistent with the ontogenetic hypothesis and are further evidence that the observed effects are brought about through learning.
, 百拇医药
The increase in apCaMKII is not simply a reflection of an increase in the total amount of CaMKII. Learning had a differential effect on these two states of the kinase: the ratio of membrane apCaMKII/tCaMKII in intermediate and good learners increased with preference score (Fig. 4C).
The amount of apCaMKII in IMHV/IMM increased as the chicks learned, the effects being most clear in the chicks with the higher preference scores, and hence (Bolhuis et al. 2000; see Fig. 1) the most robust learning. The non-linear increase in the amount of apCaMKII resembles that of NMDA receptor binding that occurs in the left IMHV/IMM after imprinting (McCabe & Horn, 1988). This non-linearity suggests that there is an accelerating level of autophosphorylation as chicks learn more about the imprinting stimulus. Levels of apCaMKII are effectively stationary for preference scores below 60 and thereafter change rapidly. This relationship may have a significant bearing on the magnitude of the retained preference. Thus, Bolhuis et al. (2000) trained chicks and divided them into two groups, with respective mean preference scores less than 65 and greater than 65. The chicks were given a second preference test 24 h later. The chicks with low preference scores at the first preference test performed at chance in the second preference test. The chicks with high preference score continued strongly to prefer the training stimulus. These findings raise the possibility that a high level of apCaMKII in the IMHV/IMM is associated with the persistence of a strong preference for the training stimulus, that is, with a higher level of memory retention.
, 百拇医药
There was no significant hemispheric asymmetry for any of the learning-specific effects that were found (Figs 3A and 4). Some learning-specific biochemical changes are predominantly expressed in the left IMHV/IMM (McCabe & Horn, 1988; Sheu et al. 1993; Solomonia et al. 1997, 1998, 2003b; McCabe et al. 2001; Meredith et al. 2004), some in both (McCabe & Horn, 1994; Meredith et al. 2004), and some of them in the right IMHV/IMM (Solomonia et al. 2000). Mechanisms that may underlie this variation of left and right IMHV/IMM involvement are discussed elsewhere (Horn, 2004). The learning-related changes we have described are restricted to the IMHV/IMM; they do not occur in the PPN. In this region the mean amounts of tCaMKII in the three groups of trained chicks combined was significantly higher, by 13%, than in untrained chicks. These results suggest that the amount of tCaMKII in PPN is affected by training, for example by handling and locomotor activity, and is not affected by learning.
, 百拇医药
The significant increase of tCaMKII in the P2 membranes of IMHV/IMM could be accounted for by at least two mechanisms: (i) translocation of the enzyme from the cytosol to the PSD and (ii) de novo synthesis of CaMKII. De novo synthesis of CaMKII could take place by two possible scenarios: (i) by the increased expression of the CaMKII gene in the nucleus and its cytoplasmic translation; and (ii) by local synthesis from the mRNA that is present at the base of dendritic spines (Steward & Schuman, 2001). Synthesis via this route could be very fast; there is an increase in CaMKII via dendritic protein synthesis in hippocampal neurones within 5 min of tetanic stimulation (Quyang et al. 1999).
, 百拇医药
The increased amounts of apCaMKII levels in IMHV/IMM could have diverse effects. CaMKII phosphorylates the AMPA receptor subunit GluR1 and enhances channel function (Barria et al. 1997; Mammen et al. 1997; Derkach et al. 1999). CaMKII binds NR2B and NR1 subunits of NMDA receptors and certain other proteins within the PSD; this binding is enhanced when the enzyme is autophosphorylated (Gardoni et al. 1998; Strack et al. 2000; Bayer et al. 2001; Walikonis et al. 2001). The enhanced binding leads to the formation of new anchoring assembles for additional AMPA receptors and stimulates the delivery of AMPA receptors to the membrane (Wu et al. 1996; Liao et al. 2001; Lisman & Zhabotinsky, 2001; Lisman et al. 2002). Thus the learning-related increase of apCaMKII in imprinting might increase the efficiency of fast excitatory transmission across IMHV/IMM synapses, or a subset of them, as chicks begin to learn. Excitatory transmission may be further facilitated by the up-regulation of NMDA receptors 8 h after training (McCabe & Horn, 1988). On the presynaptic side, activated CaMKII may phosphorylate synapsin I, promote its dissociation from synaptic vesicles and increase their availability for neurotransmitter release (Benfenati et al. 1992; Ceccaldi et al. 1995). Such enhancement of transmitter release has been demonstrated after training in explants of IMHV/IMM (McCabe et al. 2002; Meredith et al. 2004). CaMKII in membranes may also affect the transcription processes in the nucleus. Activated CaMKII phosphorylates the transcription factor NF-B, which is translocated from membranes to the nucleus (Meffert et al. 2003).
, 百拇医药
A number of studies using genetically modified mice have implicated CaMKII autophosphorylation in spatial learning and fear conditioning. (Giese et al. 1998; for review see Lisman et al. 2002). More recently Irvine et al. (2005) used autophosphorylation-deficient CaMKII-T286A mutant mice to study contextual and cued fear conditioning. They found that, whereas learning was impaired in mutants compared with wild type, memory, measured 24–48 h after training, was not impaired.
, 百拇医药
The approach we have adopted is a correlative one and not an interventionalist one. Nevertheless, by controlling the effects of ontogenetic selection, brain regions, time after training and side-effects of training including differences in visual experience and movement, the evidence strongly implies a crucial link between the changes in apCaMKII that we have observed and memory. In particular, we have provided evidence that the autophosphorylation of CaMKII is involved in memory formation at 1 h but not at 24 h after training. Although this finding cannot exclude the possibility that the autophosphorylation of CaMKII is involved in long-term memory, our findings suggest that, for imprinting, once a memory has been formed its stability is not dependent on the autophosphorylation of the kinase.
, 百拇医药
References
Barria A, Derkach V & Soderling TR (1997). Identification of the Ca2+/calmodulin-dependent protein kinase II regulatory phosphorylation site in the -amino-3-hydroxyl-5-methyl-izoxazole-propionate-type glutamate receptor. J Biol Chem 272, 32727–32730.
Bateson PPG (1966). The characteristics and context of imprinting. Biol Rev 41, 177–220.
Bayer KU, De Koninck P, Leonard AS, Hell JW & Schulman H (2001). Interaction with the NMDA receptor locks CaMKII in an active conformation. Nature 411, 801–805.
, http://www.100md.com
Benfenati F, Valtorta F, Rubenstein JL, Gorelick FS, Greengard P & Czernik AJ (1992). Synaptic vesicle-associated Ca2+/calmodulin-dependent protein kinase II is a binding protein for synapsin I. Nature 359, 417–420.
Bolhuis JJ (1991). Mechanisms of avian imprinting: a review. Biol Rev 66, 303–345.
Bolhuis JJ, Cook S & Horn G (2000). Getting better all the time: improving preference scores reflect increases in the strength of filial imprinting. Anim Behav 59, 1153–1159.
, 百拇医药
Bradley P, Horn G & Bateson P (1981). Imprinting: an electron microscopic study of chick hyperstriatum ventrale. Exp Brain Res 41, 115–120.
Cajal SR (1911). Histologie Du Systeme Nerveux de L'homme et Des Vertebres, vol. 2. Maloine, Paris.
Ceccaldi PE, Grohovaz F, Benfenati F, Chieregatti E, Greengard P & Valtorta F (1995). Dephosphorylated synapsin-I anchors synaptic vesicles to actin cytoskeleton – An analysis by videomicroscopy. J Cell Biol 128, 905–912.
, 百拇医药
Chen GQ, Chen KS, Knox J, Inglis J, Bernard A, Martin SJ, Justice A, McConlogue L, Games D, Freedman SB & Morris RGM (2000). A learning deficit related to age and beta-amyloid plaques in a mouse model of Alzheimer's disease. Nature 408, 975–979.
Cotman CW & Taylor D (1972). Isolation and structural studies on synaptic complexes from rat brain. J Cell Biol 55, 696–711.
Derkach V, Barria A & Soderling TR (1999). Ca2+/calmodulin-kinase II enhances channel conductance of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate type glutamate receptors. Proc Natl Acad Sci U S A 96, 3269–3274.
, 百拇医药
Fukunaga K, Stoppini L, Miyamoto E & Muller D (1993). Long-term potentiation is associated with an increased activity of Ca/calmodulin-dependent protein kinase II. J Biol Chem 268, 7863–7867.
Gardoni F, Caputi A, Cimino M, Pastorino L, Cattabeni F & Di Luca M (1998). Calcium/calmodulin-dependent protein kinase II is associated with NR2A/B subunits of NMDA receptor in postsynaptic densities. J Neurochem 71, 1733–1741.
Giese KP, Fedorov NB, Filipkowski RK & Silva AJ (1998). Autophosphorylation at Thr286 of the alpha calcium-calmodulin kinase II in LTP and learning. Science 279, 870–873.
, http://www.100md.com
Grab DJ, Carlin RK & Siekevitz P (1981). Function of calmodulin in postsynaptic densities. II. Presence of a calmodulin-activatable protein kinase activity. J Cell Biol 89, 440–448.
Greengard P, Valtorta F, Czernik AJ & Benfenat F (1993). Synaptic vesicle phosphoproteins and regulation of synaptic function. Science 259, 780–785.
Hayashi Y, Shi SH, Esteban JA, Piccini A, Poncer JC & Malinow R (2000). Driving AMPA receptors into synapses by LTP and CaMKII: Requirement for GluR1 and PDZ domain interaction. Science 287, 2262–2267.
, 百拇医药
Hebb DO (1949). The Organization of Behaviour. John Wiley, New York.
Horn G (1985). Memory, Imprinting and the Brain. Oxford University Press, Oxford.
Horn G (1991). Technique for removing IMHV from the chick brain. In Neural and Behavioural Plasticity: the Use of the Domestic Chick as a Model, ed. Andrew RJ, pp. 44–48. Oxford University Press, Oxford.
Horn G (1998). Visual imprinting and the neural mechanisms of recognition memory. Trends Neurosci 21, 300–306.
, 百拇医药
Horn G (2000). In memory. In Brain, Perception, Memory, ed. Bolhuis JJ, pp. 329–363. Oxford University Press, Oxford.
Horn G (2004). Pathways of the past: the imprint of memory. Nature Rev Neurosci 5, 108–120.
Horn G, Bradley P & McCabe BJ (1985). Changes in the structure of synapses associated with learning. J Neurosci 5, 3161–3168.
Horn G, McCabe BJ & Bateson PPG (1979). An autoradiographic study of the chick brain after imprinting. Brain Res 168, 36l–373.
, 百拇医药
Irvine EE, Vernon J & Giese KP (2005). CaMKII autophosphorylation contributes to rapid learning but is not necessary for memory. Nature Neurosci 8, 411–412.
Kelly PT, McGuinness TL & Greengard P (1984). Evidence that the major postsynaptic density protein is a component of a Ca2+/calmodulin-dependent protein kinase. Proc Natl Acad Sci U S A 81, 945–949.
Kennedy MB (1998). Signal transduction molecules at the glutamatergic postsynaptic membrane. Brain Res Rev 26, 243–257.
, http://www.100md.com
Kennedy MB, Bennet MK & Erondu NE (1983). Biochemical and immunochemical evidence that the ‘major postsynaptic density protein’ is a subunit of a calmodulin-dependent protein kinase. Proc Natl Acad Sci U S A 80, 7357–7361.
Konorski J (1948). Conditioned Reflexes and Neuron Organisation. Cambridge University Press, Cambridge.
Liao DZ, Scannevin RH & Huganir R (2001). Activation of silent synapses by rapid activity-dependent synaptic recruitment of AMPA receptors. J Neurosci 21, 6008–6017.
, http://www.100md.com
Lisman J (2003). Long-term potentiation: outstanding questions and attempted synthesis. Phil Trans Roy Soc Lond B 358, 829–842.
Lisman J, Schulman H & Cline H (2002). The molecular basis of CaMKII function in synaptic and behavioural memory. Nature Rev Neurosci 3, 175–190.
Lisman JE & Zhabotinsky AM (2001). A model of synaptic memory: a CaMKII/PP1 switch that potentiates transmission by organizing an AMPA receptor anchoring assembly. Neuron 31, 191–201.
, 百拇医药
Mammen A, Kameyama K, Roche KW & Huganir RL (1997). Phosphorylation of the -amino-3-hydroxy-5-methylisoxazole4-proprionic acid receptor GluR1 subunit by calcium/calmodulin-dependent kinase II. J Biol Chem 272, 32528–32523.
McCabe BJ, Cipolla-Neto J, Horn G & Bateson P (1982). Amnesic effects of bilateral lesions in the hyperstriatum ventrale of the chick after imprinting. Exp Brain Res 48, 13–21.
McCabe BJ & Horn G (1988). Learning and memory: regional changes in N-methyl-D-aspartate receptors in the chick brain after imprinting. Proc Natl Acad Sci U S A 85, 2849–2853.
, http://www.100md.com
McCabe BJ & Horn G (1994). Learning-related changes in Fos-like immunoreactivity in the chick forebrain after imprinting. Proc Natl Acad Sci U S A 91, 11417–11421.
McCabe BJ, Kendrick KM & Horn G (2001). -Aminobutyric acid, taurine and learning: release of amino acids from slices of chick brain following filial imprinting. Neuroscience 105, 317–324.
Meffert MK, Chang JM, Wiltgen BJ, Fanselow MS & Baltimore D (2003). NF-kappa B functions in synaptic signaling and behavior. Nature Neurosci 6, 1072–1078.
, 百拇医药
Meredith RM, McCabe BJ, Kendrick KM & Horn G (2004). Amino acid neurotransmitter release and learning: a study of visual imprinting. Neuroscience 126, 249–256.
Miller SG & Kennedy MB (1986). Regulation of brain type II Ca2+/calmodulin-dependent protein kinase by autophosphorylation: a Ca2+-triggered molecular switch. Cell 44, 861–870.
Mosteller F & Youtz C (1961). Tables of the Freeman-Tukey transformations for the binomial and Poisson distributions. Biometrika 48, 433–440.
, 百拇医药
Otmakhov N, Griffith LC & Lisman JE (1997). Postsynaptic inhibitors of calcium/calmodulin-dependent protein kinase type II block induction but not maintenance of pairing- induced long-term potentiation. J Neurosci 17, 5357–5365.
Passafaro M, Piech V & Sheng M (2001). Subunit-specific temporal and spatial patterns of AMPA receptor exocytosis in hippocampal neurons. Nature Neurosci 4, 917–926.
Quyang Y, Rosenstein A, Kreiman G, Schuman EM & Kennedy MB (1999). Tetanic stimulation leads to increased accumulation of Ca2+/calmodulin-dependent protein kinase II via dendritic protein synthesis in hippocampal neurons. J Neurosci 19, 7823–7833.
, 百拇医药
Reiner A, Bruce L, Butler A, Csillag A, Kuenzel W, Medina L, Paxinos G, Perkel D, Shimizu T, Striedter G, Wild M, Ball G, Durand S, Gunturkun O, Lee D, Mello CV, Powers A, White S, Hough G, Kubikova L, Smulders TV, Wada K, Dugas-Ford J, Husband S, Yamamoto K, Yu J, Siang C & Jarvis ED (2004). Revised nomenclature for avian telencephalon and some related brainstem nuclei. J Comp Neurol 473, 377–414.
Sheu FS, McCabe BJ, Horn G & Routtenberg A (1993). Learning selectively increases protein kinase C substrate phorphorylation in specific regions of the chick brain. Proc Natl Acad Sci U S A 90, 2705–2709.
, 百拇医药
Sluckin W (1972). Imprinting and Early Learning. Methuen, London.
Smilowitz H, Hijian RA, Dwyer J & Feinstein MB (1981). Regulation of acetylcholine receptor phosphorylation by calcium and calmodulin. Proc Natl Acad Sci U S A 78, 4708–4712.
Snedecor GW & Cochran WG (1989). Statistical Methods. Iowa State University Press, Ames.
Solomonia R, Kiguradze T, McCabe BJ & Horn G (2000). Neural cell adhesion molecules, CaM kinase II and long-term memory in the chick. Neuroreport 11, 3139–3143.
, 百拇医药
Solomonia RO, Kotorashvili A, Kiguradze T, McCabe BJ & Horn G (2003a). Calcium/calmodulin-dependent protein kinase II and memory. Act Neurobiol Exper 63, S76.
Solomonia RO, McCabe BJ & Horn G (1998). Neural cell adhesion molecules, learning and memory. Behav Neurosci 112, 646–655.
Solomonia RO, McCabe BJ, Jackson AP & Horn G (1997). Clathrin proteins and recognition memory. Neuroscience 80, 59–67.
, 百拇医药
Solomonia RO, Morgan K, Kotorashvili A, McCabe BJ, Jackson AP & Horn G (2003b). Analysis of differential gene expression supports a role for amyloid precursor protein and a protein kinase C substrate (MARCKS) in long-term memory. Eur J Neurosci 17, 1073–1081.
Song I & Huganir RL (2002). Regulation of AMPA receptors during synaptic plasticity. Trends Neurosci 25, 578–588.
Steward O & Schuman EM (2001). Protein synthesis at synaptic sites on dendrites. Ann Rev Neurosci 24, 299–325.
, 百拇医药
Strack S, Barban MA, Wadzinski B & Colbran RJ (1997). Differential inactivation of postsynaptic density-associated and soluble Ca2+/calmodulin-dependent protein kinase II by protein phosphatases 1 and 2A. J Neurochem 68, 2119–2128.
Strack S, Robison A, Bass MA & Colbran RJ (2000). Association of calcium/calmodulin dependent kinase II with developmentally regulated splice variants of the postsynaptic density protein densin-180. J Biol Chem 275, 25061–25064.
, 百拇医药
Tanzi E (1893). I fatti e le induzioni nell' odierna istologia del sistema nervoso. Riv Sper Freniat Med Leg Alien Ment 19, 419–472.
Walikonis RS, Oguni A, Khorosheva EM, Jeng CJ, Asuncion FJ, Kennedy MB (2001). Densin 180 forms a ternary complex with the a-subunit of Ca2+/calmodulin-dependent protein kinase II and -actinin. J Neurosci 21, 423–433.
Wu G, Malinow R & Cline HT (1996). Maturation of a central glutamatergic synapse. Science 274, 972–976.
Zhu JJ, Qin Y, Zhao M, Van Aelst L & Malinow R (2002). Ras and Rap control AMPA receptor trafficking during synaptic plasticity. Cell 110, 443–455., 百拇医药(Revaz O. Solomonia, Adam )
2 Sub-Department of Animal Behaviour, University of Cambridge, Madingley, Cambridge CB3 8AA, UK
Abstract
The role of calcium/calmodulin-dependent protein kinase II (CaMKII) in the recognition memory of visual imprinting was investigated. Domestic chicks were exposed to a training stimulus and learning strength measured. Trained chicks, together with untrained chicks, were killed either 1 h or 24 h after training. The intermediate and medial hyperstriatum ventrale/mesopallium (IMHV/IMM), a forebrain memory storage site, was removed together with a control brain region, the posterior pole of the neostriatum/nidopallium (PPN). Amounts of membrane total CaMKII (tCaMKII) and Thr286-autophosphorylated CaMKII (apCAMKII) were measured. For the IMHV/IMM 1 h group, apCaMKII amount and apCAMKII/tCaMKII increased as chicks learned. The magnitude of the molecular changes were positively correlated with learning strength. No learning-related effects were observed in PPN, or in either region at 24 h. These results suggest that CaMKII is involved in the formation of memory but not in its maintenance.
, 百拇医药
Introduction
It has often been suggested that learning and memory depend on synaptic modification (e.g. Tanzi, 1893; Cajal, 1911; Konorski, 1948; Hebb, 1949). This suggestion has stimulated many studies of synaptic plasticity, and of proteins associated with it. Prominent among such proteins has been Ca2+/calmodulin-dependent protein kinase II (CaMKII), the main protein of the postsynaptic density (PSD) in the vertebrate central nervous system (Grab et al. 1981; Kennedy et al. 1983; Kelly et al. 1984; Kennedy, 1998). Smilowitz et al. (1981) demonstrated that the phosphorylation of acetylcholine receptors was regulated by a calcium/calmodulin-dependent kinase. More recently CaMKII in the mammalian hippocampus has been shown to phosphorylate and promote the delivery of -amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptors into the postsynaptic membrane (Barria et al. 1997; Mammen et al. 1997; Chen et al. 2000; Hayashi et al. 2000; Passafaro et al. 2001; Zhu et al. 2002). Presynaptically, CaMKII may increase neurotransmitter release by the phosphorylation of synapsin I and the subsequent dissociation of synapsin I from synaptic vesicles (Greengard et al. 1993). CaMKII plays a critical role in hippocampal long-term potentiation (Fukunaga et al. 1993; Otmakhov et al. 1997; Song & Huganir, 2002; Lisman, 2003) and it has been suggested that the kinase is involved in synaptic changes underlying memory (Horn, 1985, pp. 219–221; Lisman et al. 2002).
, 百拇医药
Transient exposure of CaMKII to Ca2+/calmodulin activates the kinase, which, through autophosphorylation at Thr286, remains active after Ca2+ is withdrawn (Miller & Kennedy, 1986). This property of the kinase is one of the reasons for proposing that it could serve as a molecular switch, rather than a trigger, that is capable of long-term memory storage (see Lisman et al. 2002 for review). A prediction of this hypothesis is that the enzyme should remain in the autophosphorylated state so long as memory persists. To test this hypothesis it is necessary to study the effect of learning on CaMKII in a brain region in which memories are formed. One type of learning that offers an opportunity for such a study is visual imprinting, whereby the young of certain species learn the features of, and subsequently recognize, an object as a result of being exposed to it (Bateson, 1966; Sluckin, 1972; Horn, 1985; Bolhuis, 1991). Training conditions may so be contrived that, before learning, no visual experience has made its mark on the brain. This property of visual imprinting has made it possible to identify a brain region in which information is stored. The region is the intermediate and medial hyperstriatum ventrale (IMHV; intermediate and medial mesopallium (IMM) in the new terminology, Reiner et al. 2004) of the domestic chick forebrain. The evidence for a role of this region in memory is reviewed in detail elsewhere (Horn, 1985, 1998, 2000, 2004). Some of the cellular and synaptic changes occurring in the region after imprinting may be linked with CaMKII activation (see for example, Bradley et al. 1981; Horn et al. 1985; Sheu et al. 1993; McCabe et al. 2001; Meredith et al. 2004). We have therefore inquired whether imprinting leads to changes in the amount of the -subunit, CaMKII, and its autophosphorylation state at Thr286, in a membrane fraction of IMHV/IMM. The posterior pole of the neostriatum (PPN; now nidopallium), a region where learning-related changes following imprinting have not been observed (Horn et al. 1979; Solomonia et al. 1997, 1998, 2003b), was studied as a control region.
, 百拇医药
A preliminary account of these results has been published (Solomonia et al. 2003a).
Methods
Behavioural procedures
Chicks were hatched, trained and reared as previously described (McCabe et al. 1982; Solomonia et al. 1997, 2003b). Twenty-two batches of chicks (Ross 308, Grampian Country Chickens) were used. Chicks were hatched and reared in darkness. When 22–28 h old, some of the chicks in each batch were trained by exposure for 1 h in individual running wheels to an imprinting stimulus (a rotating, internally illuminated red box (RB)). As a chick attempted to approach the RB the chick rotated the running wheel. The number of revolutions of the wheel were counted to provide a measure of approach activity. Chicks that have learned the characteristics of the training stimulus prefer it to an alternative stimulus. Ten minutes after training, chicks' preferences were measured by showing each chick sequentially the RB and an alternative stimulus, a rotating, internally illuminated blue cylinder (BC). A preference score was then calculated, providing a measure of preference and hence of learning strength (Sluckin, 1972; Bolhuis et al. 2000). The preference score was the chick's approach to the RB expressed as a percentage of approach to RB plus approach to BC. If RB and BC were approached equally, the preference score would be 50 (no choice; indicative of no learning (Bolhuis et al. 2000)). If the chick approached only the training object, the preference score would be 100 indicating a strong preference for the training object. Preference scores were used to relate the strength of learning with changes in CaMKII. This was achieved (i) by correlation analysis and (ii) by subdividing chicks into three groups based on their preference scores. This procedure makes it possible to tease apart effects of learning from non-specific effects of training (see Discussion). Chicks with preference scores 46–60 were classified as poor learners, chicks with preference scores 80% as good learners and those with preference scores between 61 and 79 were classified as intermediate learners (Solomonia et al. 2003b). One group of chicks in each batch served as untrained controls.
, 百拇医药
Chicks from 11 batches were killed 1 h after the end of training, and from the other 11 batches 24 h after training. Four pieces of tissue were removed from each brain, from the left and right IMHV/IMM and the left and right PPN, and frozen immediately on dry ice. Details of the method for removal of tissue from the IMHV/IMM and PPN, respectively, are described elsewhere (Horn, 1991; Solomonia et al. 1998). Tissue from each region and side (e.g. right IMHV/IMM) and for a particular training condition (e.g. intermediate learners) was obtained from three chicks having similar preference scores (i.e. all three chicks being either good learners, or intermediate learners, or poor learners) and pooled to form a single sample. This pooling was necessary to provide sufficient membrane protein for analysis. Similar samples were obtained from untrained chicks. Thus, for example, the right IMHV/IMMs from three chicks that were intermediate learners constituted a single sample. Each sample from the trained chicks was assigned a preference score that was the mean for the three chicks. Each set of experimental conditions (good learners, intermediate learners, poor learners, untrained) yielded 16 samples (left and right IMHV/IMM, left and right PPN x 4 experimental conditions). These conditions were replicated 11 times yielding 176 samples (11 x 16) for the 1 h group: that is, 88 IMHV/IMM and 88 PPN samples. These samples were obtained from 132 chicks (11 replications x 4 experimental conditions x 3 chicks per sample). The same number was used for the 24 h group. Samples were coded after collection and all further procedures conducted blind, that is without knowledge of the treatment of the chicks which contributed to a sample.
, http://www.100md.com
Chicks experienced no suffering prior to death and were covered by UK Home Office licence. The number of animals was estimated to be the minimum required for adequate statistical analysis.
Electrophoresis and immunoblotting
Pooled samples were rapidly homogenized in 20 mM Tris-HCl (pH 7.4), 0.32 M sucrose, 1 mM ethylenediamine-tetraacetic acid, 1 mM sodium orthovanadate, 10 mM sodium pyrophosphate, 0.5 mM ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid, 1 mM phenylmethanesulphonyl fluoride and 5 μM okadaic acid (to inhibit phosphatases 1 and 2A; see Strack et al. 1997) and centrifuged at 1000 g for 10 min. The supernatant was further centrifuged at 15 000 g for 20 min. The pellet was washed once and is referred to as the P2 membrane fraction (Cotman & Taylor, 1972). The P2 membrane fraction was dissolved in 5% SDS solution. Protein concentration was determined in quadruplicate using a micro bicinchoninic acid protein assay kit (Pierce Biotechnology, Inc., Rockford, IL, USA). Aliquots containing 30 μg of protein and of equal volume were applied to the gels. SDS gel electrophoresis and Western blotting were carried out as previously described (Solomonia et al. 1998, 2000). After protein had been transferred onto nitrocellulose membranes, the membranes were stained with Ponceau S solution to confirm transfer and the uniform loading of the gels. The membranes were then washed with phosphate-buffered saline + 0.05% Tween 20. In each batch, samples were run in duplicate and one filter was stained with monoclonal antibody (Chemicon, Temecula, CA, USA) against CaMKII; the antibody reacts with both phosphorylated and non-phosphorylated forms of the enzyme. The second filter was stained with a monoclonal antibody (Sigma) against phosphorylated CaMKII, which reacts only with Thr286-autophosphorylated CaMKII. Standard immunochemical procedures using peroxidase-labelled secondary antibodies and SuperSignal West Pico Chemiluminescent substrate (Pierce) were carried out. The blots were then exposed with intensifying screens to X-ray films preflashed with Sensitize (Amersham). The optical densities of bands corresponding to total CaMKII (tCaMKII) and Thr286-autophosphorylated CaMKII (apCAMKII) were measured. The autoradiographs were calibrated by including in each gel four standards comprising membrane fraction from the IMHV/IMMs of a group of untrained chicks. Each standard contained one of a range of amounts of total protein (10–60 μg). One set of standards was used for the 1 h experiments and another set for the 24 h experiments. Optical density was proportional to the amount of tCaMKII and apCAMKII. To obtain the data given in Figs 3, 4 and 5 the optical density of each band from an experimental sample (e.g. intermediate learners) was divided by the optical density which, from the calibration of the same autoradiograph, corresponded to 30 μg of total protein in the standard (Solomonia et al. 1997, 2003b). The data expressed in this way will be referred to as ‘relative amount’ of apCaMKII and tCaMKII. A measure of the proportion of CaMKII in a sample that was autophosphorylated (referred to as apCaMKII/tCaMKII) was obtained as the expression (relative amount of apCaMKII/relative amount of tCaMKII) for that sample. The two relative amounts of protein in this expression are ratios (optical density of sample/optical density of corresponding 30 μg standard) and depend on the level of autophosphorylation in the corresponding standards. As a result, the ratio apCaMKII/tCaMKII is not the actual proportion of CaMKII autophosphorylated in a sample but is linearly related to that value.
, http://www.100md.com
Error bars represent ± 1 standard error of the mean. A, IMHV/IMM. There was a significant effect of Training Condition: F3,26= 5.74, P= 0.004. *t= 2.39, 26 d.f., P= 0.024; **t= 3.21, 26 d.f., P= 0.004; ***t= 3.55, 26 d.f., P= 0.002. B, PPN. The ANOVA showed no significant difference between the four means: F3,26= 0.88, P= 0.46. The black bar indicates the mean value for all trained chicks combined. This combined mean is not significantly different from the mean of the untrained chicks: t= 1.90, 27 d.f., P= 0.07.
, 百拇医药
Error bars represent ± 1 standard error of the mean. A, IMHV/IMM. The amount of tCaMKII in the IMHV/IMM was significantly affected by Training Condition: F3,26= 4.71, P= 0.009. *t= 2.11, 26 d.f., P= 0.044; **t= 2.34, 26 d.f., P= 0.027; ***t= 3.68, 26 d.f., P= 0.001. n, number of samples. B, PPN. There was a significant effect of Training Condition: F3,26= 8.53; P < 0.001. The means of all three groups of trained chicks were statistically homogeneous: F2,16= 3.14; P= 0.07. However, each of these measures was significantly greater than the mean of the untrained chicks: **t= 2.86, 26 d.f.; P= 0.008. ***(intermediate learners versus untrained) t= 3.71, 26 d.f.; P < 0.001; ***(good learners versus untrained) t= 4.85, 26 d.f.; P < 0.001. The combined mean for the three trained groups was 1.38 ± 0.03 and was significantly greater (by 27%) than that (1.09 ± 0.06) of the untrained chicks: t= 4.83, 24 d.f., P < 0.001.
, http://www.100md.com
Statistical analysis
Because of the functional differences between the IMHV/IMM and the PPN (see Introduction), data from the two regions and for the different time groups (1 h and 24 h) were analysed separately. Data were subjected to a split-plot analysis of variance (ANOVA) with the fixed factors Training Condition (untrained, poor learners, intermediate learners, good learners) and Side (left, right), replicated in Blocks. In consequence, comparisons within chicks (factor Side) were made using different standard errors from those used for comparisons between groups of chicks (factor Training Condition). Preference score was subjected to an arcsin square root transformation (Mosteller & Youtz, 1961) before analysis, in order to normalize the data. The least squares regression of each biochemical measure on preference score, approach during training and approach during the preference test were conducted using a general linear model. Values of apCaMKII amount and standard errors in Figs 3, 4 and 5 are given after subtraction of Block mean for trained and untrained chicks and then adding back the overall mean for all chicks. This procedure eliminates variation between blocks, and the significance of correlation coefficients was tested using correspondingly fewer degrees of freedom for error, 10 degrees of freedom being removed for the 11 blocks (see, e.g. Snedecor & Cochran, 1989).
, 百拇医药
Results
Behaviour
Preference scores at both 1 h and 24 h were significantly greater in good learners than in intermediate or poor learners, and significantly greater in intermediate learners than in poor learners (Fig. 1). The full statistical analysis is given in the legend to Fig. 1. Approach activity during training did not differ significantly between the three trained groups of chicks at either time.
, http://www.100md.com
The dashed line indicates the no-preference score of 50. Error bars represent ± 1 standard error of the mean. Preference scores differed significantly among experimental groups both at 1 h (A; F2,17= 176.1, P < 0.001) and 24 h (B; F2,19= 195.6, P < 0.001). The mean preference scores of good learners were significantly greater than those of intermediate learners and poor learners (1 h, t= 10.6 and 18.4, respectively, 17 d.f., P < 0.001; 24 h, t= 11.2 and 19.7, respectively, 19 d.f., P < 0.001). The mean preference scores of intermediate learners were significantly greater than those of poor learners (2 h, t= 8.1, 17 d.f., P < 0.001; 24 h, t= 8.5, 19 d.f., P < 0.01). ***Preference score significantly greater than 50, t > 20.9, P < 0.001. n, number of samples.
, http://www.100md.com
Immunostaining of apCaMKII and tCAMKII
The antibodies against apCaMKII and tCaMKII reacted with the same single 52 kDa band, corresponding to the CaMKII of chick brain (see Fig. 2 and Solomonia et al. 2000).
A and B each contain an experimental replicate (right and left IMHV/IMM and right and left PPN of good learners, intermediate learners, poor learners and untrained chicks). All samples in an experimental replicate were analysed using a single gel. The discontinuity between lanes 4 and 5 represents irrelevant intercalated lanes, which have been excluded. The optical density of a band was linearly related to the amount of protein in the band. Each lane was derived from a single sample. Lanes 1–4 are from good learners, 5–8 from intermediate learners, 9–12 from poor learners and 13–16 from untrained chicks. Odd numbered lanes are from the left side, even lanes from the right. All procedures after training were conducted ‘blind’. Measurements for individual samples (cf. Fig. 4) are not of course the same as their respective experimental group means (cf. Figs 3 and 5).
, 百拇医药
Changes in CaMKII
One hour after the end of training. IMHV/IMM. The amounts of apCaMKII and tCaMKII did not differ significantly between the left and right sides: apCaMKII, F1,36= 0.01, P= 0.93; tCaMKII, F1,35= 0.28, P= 0.60. Mean values ±S.E.M. were: apCaMKII, left 0.92 ± 0.03, right 0.93 ± 0.03; tCaMKII, left 1.17 ± 0.02, right 1.19 ± 0.04. Furthermore, the interactions between Side and Training Condition were not significant. Accordingly, data from the left and right sides were combined.
, http://www.100md.com
There was a highly significant effect of Training Condition on the mean amounts of apCaMKII in IMHV/IMM: F3,26= 5.74, P= 0.004. There was significantly more apCaMKII in good learners than in intermediate learners (33%), poor learners (22%) and untrained chicks (35%) (see Fig. 3A and Table 1). Amongst the trained chicks, the mean values of poor and intermediate learners were statistically homogeneous (t= 1.16, 26 d.f., P= 0.26), the combined mean value being 0.88 ± 0.02. The mean amount of apCAMKII in the good learners was significantly higher (by 27%; t= 5.39, 26 d.f., P < 0.001) than this combined mean. Together, these results imply that the effects of training were expressed amongst chicks with the higher preference scores. Consistent with this implication (i) for intermediate and good learners there was a significant linear fit between the amount of apCaMKII and preference score (Fig. 4A) and (ii) for all groups of trained chicks, the relationship between the amounts of apCAMKII and preference score was fitted significantly, F2,15= 8.09, P= 0.004, by an exponential function (Fig. 4B). The fitted line in Fig. 4B was interpolated to yield a value of apCaMKII corresponding to a preference score of 50. The score of 50 indicates an absence of preference for either stimulus and is characteristic of chicks that have not become imprinted. The interpolated amount of apCaMKII was close to the mean value for the untrained chicks. These findings imply that exposure to the training stimulus alone did not lead to an increase in the amount of apCaMKII; apCAMKII levels only increase as the chicks learn. The separate variances of good, intermediate and poor learners, and untrained chicks, did not differ significantly from one another.
, 百拇医药
A and B refer to the relationship between preference score and relative amount of autophosphorylated CaMKII (apCaMKII); for details of how relative amount was calculated, see Methods. Lines were fitted by the method of least squares. A, intermediate learners and good learners. These data were fitted by the expression y=a+bx; r= 0.72, 10 d.f., P= 0.01. B, all trained chicks. These data are fitted by the exponential function y=a+bcx, F2,15= 8.09, P= 0.004. The arrow indicates the mean amount of apCaMKII +S.E.M. in untrained chicks. The interpolated value of the amount of apCaMKII for a ‘no-preference’ score of 50 in trained chicks is estimated from the dashed line. The black bar on the ordinate represents ± 1 standard error for the fitted line at this preference score. The interpolated value is very close to the mean for the untrained chicks. C, relationship between the ratio apCaMKII/tCaMKII for IMHV/IMM and preference score in intermediate and good learners. Data were fitted by a linear function; r= 0.70, 9 d.f., P= 0.02, fitted by the method of least squares.
, 百拇医药
The amount of tCaMKII was significantly higher in good learners than in both poor learners and untrained chicks and significantly higher in intermediate learners than in untrained chicks (see Fig. 5A and Table 1). In spite of the systematic increase in the mean values from poor through to good learners, the correlation between tCaMKII and preference score was not significant: r= 0.32, 16 d.f., P= 0.20.
The ratio apCaMKII/tCaMKII increased significantly with preference score for intermediate and good learners (see Fig. 4C). This finding suggests that for these groups of chicks the amount of apCAMKII rose faster with preference score than did the amount of tCaMKII: as the strength of learning increased there was a selective increase in the level of autophosphorylation.
, 百拇医药
There were no significant correlations between approach activity during training and approach activity during the preference test on the one hand and the amounts of apCaMKII and tCaMKII on the other (see Table 2).
One hour after the end of training. PPN. There was significantly more apCaMKII, but not tCaMKII, in the left PPN than in the right: apCaMKII, F1,35= 4.97, P= 0.032; tCaMKII, F1,32= 0.41, P= 0.53. The mean values ±S.E.M. were: apCamKII, left 1.12 ± 0.03, right 1.04 ± 0.04; tCaMKII, left 1.30 ± 0.04, right 1.30 ± 0.03.
, 百拇医药
For apCaMKII there were no significant effects of training on either the left or the right sides: the Training Condition–Side interaction was not significant: F3,35= 1.05, P= 0.38. To examine the effects of training on apCaMKII level the mean values for right and left sides were therefore combined. The combined means are given according to training condition in Fig. 3B. There were no significant differences between the means for the three groups of trained chicks (Fig. 3B). The mean value for all trained chicks combined was 13% greater than that of untrained chicks but the difference was not significant: t= 1.90, 27 d.f., P= 0.07 (Fig. 3B).
, 百拇医药
For tCaMKII, data from left and right PPN were combined. The means of the three groups of trained chicks were statistically homogeneous and each was significantly greater than that of the untrained chicks (Fig. 5B). The mean value for all trained chicks combined (see Fig. 5B) was significantly higher, by 27%, than the mean of the untrained chicks: t= 5.41, 24 d.f., P < 0.001. There were no significant correlations between amount of apCaMKII and tCaMKII on the one hand and preference score, approach activity during training or approach activity during the preference test on the other (Table 2). The ratio apCaMKII/tCaMKII was not significantly correlated with preference score, either for all trained chicks (r= 0.32; P= 0.14), or for intermediate and good learners (r= 0.07; P= 0.79). The exponential fit for all trained chicks was also not significant: F2,15= 1.98, P= 0.60.
, http://www.100md.com
Twenty-four hours after the end of training. IMHV/IMM and PPN. For both the IMHV/IMM and the PPN, no significant effects of Training Condition (Table 2) were found and there were no significant correlations between any of the measures of CaMKII protein on the one hand, and preference score, approach activity during training or approach activity during the preference test on the other (Table 2).
Discussion
If CaMKII is essential for the storage of information acquired through learning, then, in a brain region storing that information, the activity of the kinase should correlate with the strength of learning. The present study has addressed this issue and further inquired whether such changes are time dependent. We have demonstrated that (i) the amount of CaMKII autophosphorylated at Thr286 (apCaMKII) and (ii) the ratio of apCaMKII/tCaMKII are changed in a learning-specific way (see below) in a brain structure storing information acquired through imprinting. The effects are regionally specific, since they occur in the IMHV/IMM but not in the PPN; they are time dependent since they occur 1 h after the end of training, not at 24 h.
, 百拇医药
The evidence that the changes measured 1 h after training in the amount of membrane apCAMKII in IMHV/IMM are a function of learning is as follows. (i) There was a significant, positive correlation between preference score and the amount of apCaMKII. (ii) The mean amount of apCAMKII present in good learners was significantly higher than in intermediate learners, poor learners or untrained chicks. (iii) The mean amount of apCAMKII present in the untrained controls did not differ significantly from the amount in the poor learners or extrapolated from the regression line for a no-choice preference score of 50. These findings taken together imply that sensory stimulation, movement, arousal, motivational state and non-specific factors (collectively referred to as ‘side-effects of training’) did not influence the amount of apCaMKII in IMHV/IMM. Were they to have done so, the mean amount of apCaMKII in poor learners, which had been exposed to the whole training procedure, would be expected to be higher than that of the untrained controls. This is not the case. Furthermore, changes in the amount of apCaMKII in IMHV/IMM could not be attributed to the preference test, since poor learners were given this test and untrained chicks were not: the mean amount of apCaMKII in these two groups was not significantly different. (iv) The pattern of apCaMKII means for IMHV/IMM in Fig. 3A contrasts with that for the PPN (Fig. 3B). In the PPN the mean amounts of apCaMKII within the trained group were statistically homogeneous. (v) The amount of apCaMKII was not affected by approach activity during training or during the preference test. (vi) The changes in apCaMKII were not an effect of ontogenetic selection whereby chicks hatched with, for example, high levels of apCaMKII become good learners, and those with low levels of apCaMKII become poor learners. According to this ontogenetic selection hypothesis, therefore, good and poor learners are subclasses of untrained chicks with ontogenetically high or low levels of apCaMKII respectively (see Horn, 2004). If this were so, the variance in levels of apCaMKII in untrained chicks, being the ‘parent’ population, should be higher than those of the selected subgroups of poor, intermediate and good learners. This was not the case. There were no significant differences between the variances of the four groups of chicks. These findings are not therefore consistent with the ontogenetic hypothesis and are further evidence that the observed effects are brought about through learning.
, 百拇医药
The increase in apCaMKII is not simply a reflection of an increase in the total amount of CaMKII. Learning had a differential effect on these two states of the kinase: the ratio of membrane apCaMKII/tCaMKII in intermediate and good learners increased with preference score (Fig. 4C).
The amount of apCaMKII in IMHV/IMM increased as the chicks learned, the effects being most clear in the chicks with the higher preference scores, and hence (Bolhuis et al. 2000; see Fig. 1) the most robust learning. The non-linear increase in the amount of apCaMKII resembles that of NMDA receptor binding that occurs in the left IMHV/IMM after imprinting (McCabe & Horn, 1988). This non-linearity suggests that there is an accelerating level of autophosphorylation as chicks learn more about the imprinting stimulus. Levels of apCaMKII are effectively stationary for preference scores below 60 and thereafter change rapidly. This relationship may have a significant bearing on the magnitude of the retained preference. Thus, Bolhuis et al. (2000) trained chicks and divided them into two groups, with respective mean preference scores less than 65 and greater than 65. The chicks were given a second preference test 24 h later. The chicks with low preference scores at the first preference test performed at chance in the second preference test. The chicks with high preference score continued strongly to prefer the training stimulus. These findings raise the possibility that a high level of apCaMKII in the IMHV/IMM is associated with the persistence of a strong preference for the training stimulus, that is, with a higher level of memory retention.
, 百拇医药
There was no significant hemispheric asymmetry for any of the learning-specific effects that were found (Figs 3A and 4). Some learning-specific biochemical changes are predominantly expressed in the left IMHV/IMM (McCabe & Horn, 1988; Sheu et al. 1993; Solomonia et al. 1997, 1998, 2003b; McCabe et al. 2001; Meredith et al. 2004), some in both (McCabe & Horn, 1994; Meredith et al. 2004), and some of them in the right IMHV/IMM (Solomonia et al. 2000). Mechanisms that may underlie this variation of left and right IMHV/IMM involvement are discussed elsewhere (Horn, 2004). The learning-related changes we have described are restricted to the IMHV/IMM; they do not occur in the PPN. In this region the mean amounts of tCaMKII in the three groups of trained chicks combined was significantly higher, by 13%, than in untrained chicks. These results suggest that the amount of tCaMKII in PPN is affected by training, for example by handling and locomotor activity, and is not affected by learning.
, 百拇医药
The significant increase of tCaMKII in the P2 membranes of IMHV/IMM could be accounted for by at least two mechanisms: (i) translocation of the enzyme from the cytosol to the PSD and (ii) de novo synthesis of CaMKII. De novo synthesis of CaMKII could take place by two possible scenarios: (i) by the increased expression of the CaMKII gene in the nucleus and its cytoplasmic translation; and (ii) by local synthesis from the mRNA that is present at the base of dendritic spines (Steward & Schuman, 2001). Synthesis via this route could be very fast; there is an increase in CaMKII via dendritic protein synthesis in hippocampal neurones within 5 min of tetanic stimulation (Quyang et al. 1999).
, 百拇医药
The increased amounts of apCaMKII levels in IMHV/IMM could have diverse effects. CaMKII phosphorylates the AMPA receptor subunit GluR1 and enhances channel function (Barria et al. 1997; Mammen et al. 1997; Derkach et al. 1999). CaMKII binds NR2B and NR1 subunits of NMDA receptors and certain other proteins within the PSD; this binding is enhanced when the enzyme is autophosphorylated (Gardoni et al. 1998; Strack et al. 2000; Bayer et al. 2001; Walikonis et al. 2001). The enhanced binding leads to the formation of new anchoring assembles for additional AMPA receptors and stimulates the delivery of AMPA receptors to the membrane (Wu et al. 1996; Liao et al. 2001; Lisman & Zhabotinsky, 2001; Lisman et al. 2002). Thus the learning-related increase of apCaMKII in imprinting might increase the efficiency of fast excitatory transmission across IMHV/IMM synapses, or a subset of them, as chicks begin to learn. Excitatory transmission may be further facilitated by the up-regulation of NMDA receptors 8 h after training (McCabe & Horn, 1988). On the presynaptic side, activated CaMKII may phosphorylate synapsin I, promote its dissociation from synaptic vesicles and increase their availability for neurotransmitter release (Benfenati et al. 1992; Ceccaldi et al. 1995). Such enhancement of transmitter release has been demonstrated after training in explants of IMHV/IMM (McCabe et al. 2002; Meredith et al. 2004). CaMKII in membranes may also affect the transcription processes in the nucleus. Activated CaMKII phosphorylates the transcription factor NF-B, which is translocated from membranes to the nucleus (Meffert et al. 2003).
, 百拇医药
A number of studies using genetically modified mice have implicated CaMKII autophosphorylation in spatial learning and fear conditioning. (Giese et al. 1998; for review see Lisman et al. 2002). More recently Irvine et al. (2005) used autophosphorylation-deficient CaMKII-T286A mutant mice to study contextual and cued fear conditioning. They found that, whereas learning was impaired in mutants compared with wild type, memory, measured 24–48 h after training, was not impaired.
, 百拇医药
The approach we have adopted is a correlative one and not an interventionalist one. Nevertheless, by controlling the effects of ontogenetic selection, brain regions, time after training and side-effects of training including differences in visual experience and movement, the evidence strongly implies a crucial link between the changes in apCaMKII that we have observed and memory. In particular, we have provided evidence that the autophosphorylation of CaMKII is involved in memory formation at 1 h but not at 24 h after training. Although this finding cannot exclude the possibility that the autophosphorylation of CaMKII is involved in long-term memory, our findings suggest that, for imprinting, once a memory has been formed its stability is not dependent on the autophosphorylation of the kinase.
, 百拇医药
References
Barria A, Derkach V & Soderling TR (1997). Identification of the Ca2+/calmodulin-dependent protein kinase II regulatory phosphorylation site in the -amino-3-hydroxyl-5-methyl-izoxazole-propionate-type glutamate receptor. J Biol Chem 272, 32727–32730.
Bateson PPG (1966). The characteristics and context of imprinting. Biol Rev 41, 177–220.
Bayer KU, De Koninck P, Leonard AS, Hell JW & Schulman H (2001). Interaction with the NMDA receptor locks CaMKII in an active conformation. Nature 411, 801–805.
, http://www.100md.com
Benfenati F, Valtorta F, Rubenstein JL, Gorelick FS, Greengard P & Czernik AJ (1992). Synaptic vesicle-associated Ca2+/calmodulin-dependent protein kinase II is a binding protein for synapsin I. Nature 359, 417–420.
Bolhuis JJ (1991). Mechanisms of avian imprinting: a review. Biol Rev 66, 303–345.
Bolhuis JJ, Cook S & Horn G (2000). Getting better all the time: improving preference scores reflect increases in the strength of filial imprinting. Anim Behav 59, 1153–1159.
, 百拇医药
Bradley P, Horn G & Bateson P (1981). Imprinting: an electron microscopic study of chick hyperstriatum ventrale. Exp Brain Res 41, 115–120.
Cajal SR (1911). Histologie Du Systeme Nerveux de L'homme et Des Vertebres, vol. 2. Maloine, Paris.
Ceccaldi PE, Grohovaz F, Benfenati F, Chieregatti E, Greengard P & Valtorta F (1995). Dephosphorylated synapsin-I anchors synaptic vesicles to actin cytoskeleton – An analysis by videomicroscopy. J Cell Biol 128, 905–912.
, 百拇医药
Chen GQ, Chen KS, Knox J, Inglis J, Bernard A, Martin SJ, Justice A, McConlogue L, Games D, Freedman SB & Morris RGM (2000). A learning deficit related to age and beta-amyloid plaques in a mouse model of Alzheimer's disease. Nature 408, 975–979.
Cotman CW & Taylor D (1972). Isolation and structural studies on synaptic complexes from rat brain. J Cell Biol 55, 696–711.
Derkach V, Barria A & Soderling TR (1999). Ca2+/calmodulin-kinase II enhances channel conductance of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate type glutamate receptors. Proc Natl Acad Sci U S A 96, 3269–3274.
, 百拇医药
Fukunaga K, Stoppini L, Miyamoto E & Muller D (1993). Long-term potentiation is associated with an increased activity of Ca/calmodulin-dependent protein kinase II. J Biol Chem 268, 7863–7867.
Gardoni F, Caputi A, Cimino M, Pastorino L, Cattabeni F & Di Luca M (1998). Calcium/calmodulin-dependent protein kinase II is associated with NR2A/B subunits of NMDA receptor in postsynaptic densities. J Neurochem 71, 1733–1741.
Giese KP, Fedorov NB, Filipkowski RK & Silva AJ (1998). Autophosphorylation at Thr286 of the alpha calcium-calmodulin kinase II in LTP and learning. Science 279, 870–873.
, http://www.100md.com
Grab DJ, Carlin RK & Siekevitz P (1981). Function of calmodulin in postsynaptic densities. II. Presence of a calmodulin-activatable protein kinase activity. J Cell Biol 89, 440–448.
Greengard P, Valtorta F, Czernik AJ & Benfenat F (1993). Synaptic vesicle phosphoproteins and regulation of synaptic function. Science 259, 780–785.
Hayashi Y, Shi SH, Esteban JA, Piccini A, Poncer JC & Malinow R (2000). Driving AMPA receptors into synapses by LTP and CaMKII: Requirement for GluR1 and PDZ domain interaction. Science 287, 2262–2267.
, 百拇医药
Hebb DO (1949). The Organization of Behaviour. John Wiley, New York.
Horn G (1985). Memory, Imprinting and the Brain. Oxford University Press, Oxford.
Horn G (1991). Technique for removing IMHV from the chick brain. In Neural and Behavioural Plasticity: the Use of the Domestic Chick as a Model, ed. Andrew RJ, pp. 44–48. Oxford University Press, Oxford.
Horn G (1998). Visual imprinting and the neural mechanisms of recognition memory. Trends Neurosci 21, 300–306.
, 百拇医药
Horn G (2000). In memory. In Brain, Perception, Memory, ed. Bolhuis JJ, pp. 329–363. Oxford University Press, Oxford.
Horn G (2004). Pathways of the past: the imprint of memory. Nature Rev Neurosci 5, 108–120.
Horn G, Bradley P & McCabe BJ (1985). Changes in the structure of synapses associated with learning. J Neurosci 5, 3161–3168.
Horn G, McCabe BJ & Bateson PPG (1979). An autoradiographic study of the chick brain after imprinting. Brain Res 168, 36l–373.
, 百拇医药
Irvine EE, Vernon J & Giese KP (2005). CaMKII autophosphorylation contributes to rapid learning but is not necessary for memory. Nature Neurosci 8, 411–412.
Kelly PT, McGuinness TL & Greengard P (1984). Evidence that the major postsynaptic density protein is a component of a Ca2+/calmodulin-dependent protein kinase. Proc Natl Acad Sci U S A 81, 945–949.
Kennedy MB (1998). Signal transduction molecules at the glutamatergic postsynaptic membrane. Brain Res Rev 26, 243–257.
, http://www.100md.com
Kennedy MB, Bennet MK & Erondu NE (1983). Biochemical and immunochemical evidence that the ‘major postsynaptic density protein’ is a subunit of a calmodulin-dependent protein kinase. Proc Natl Acad Sci U S A 80, 7357–7361.
Konorski J (1948). Conditioned Reflexes and Neuron Organisation. Cambridge University Press, Cambridge.
Liao DZ, Scannevin RH & Huganir R (2001). Activation of silent synapses by rapid activity-dependent synaptic recruitment of AMPA receptors. J Neurosci 21, 6008–6017.
, http://www.100md.com
Lisman J (2003). Long-term potentiation: outstanding questions and attempted synthesis. Phil Trans Roy Soc Lond B 358, 829–842.
Lisman J, Schulman H & Cline H (2002). The molecular basis of CaMKII function in synaptic and behavioural memory. Nature Rev Neurosci 3, 175–190.
Lisman JE & Zhabotinsky AM (2001). A model of synaptic memory: a CaMKII/PP1 switch that potentiates transmission by organizing an AMPA receptor anchoring assembly. Neuron 31, 191–201.
, 百拇医药
Mammen A, Kameyama K, Roche KW & Huganir RL (1997). Phosphorylation of the -amino-3-hydroxy-5-methylisoxazole4-proprionic acid receptor GluR1 subunit by calcium/calmodulin-dependent kinase II. J Biol Chem 272, 32528–32523.
McCabe BJ, Cipolla-Neto J, Horn G & Bateson P (1982). Amnesic effects of bilateral lesions in the hyperstriatum ventrale of the chick after imprinting. Exp Brain Res 48, 13–21.
McCabe BJ & Horn G (1988). Learning and memory: regional changes in N-methyl-D-aspartate receptors in the chick brain after imprinting. Proc Natl Acad Sci U S A 85, 2849–2853.
, http://www.100md.com
McCabe BJ & Horn G (1994). Learning-related changes in Fos-like immunoreactivity in the chick forebrain after imprinting. Proc Natl Acad Sci U S A 91, 11417–11421.
McCabe BJ, Kendrick KM & Horn G (2001). -Aminobutyric acid, taurine and learning: release of amino acids from slices of chick brain following filial imprinting. Neuroscience 105, 317–324.
Meffert MK, Chang JM, Wiltgen BJ, Fanselow MS & Baltimore D (2003). NF-kappa B functions in synaptic signaling and behavior. Nature Neurosci 6, 1072–1078.
, 百拇医药
Meredith RM, McCabe BJ, Kendrick KM & Horn G (2004). Amino acid neurotransmitter release and learning: a study of visual imprinting. Neuroscience 126, 249–256.
Miller SG & Kennedy MB (1986). Regulation of brain type II Ca2+/calmodulin-dependent protein kinase by autophosphorylation: a Ca2+-triggered molecular switch. Cell 44, 861–870.
Mosteller F & Youtz C (1961). Tables of the Freeman-Tukey transformations for the binomial and Poisson distributions. Biometrika 48, 433–440.
, 百拇医药
Otmakhov N, Griffith LC & Lisman JE (1997). Postsynaptic inhibitors of calcium/calmodulin-dependent protein kinase type II block induction but not maintenance of pairing- induced long-term potentiation. J Neurosci 17, 5357–5365.
Passafaro M, Piech V & Sheng M (2001). Subunit-specific temporal and spatial patterns of AMPA receptor exocytosis in hippocampal neurons. Nature Neurosci 4, 917–926.
Quyang Y, Rosenstein A, Kreiman G, Schuman EM & Kennedy MB (1999). Tetanic stimulation leads to increased accumulation of Ca2+/calmodulin-dependent protein kinase II via dendritic protein synthesis in hippocampal neurons. J Neurosci 19, 7823–7833.
, 百拇医药
Reiner A, Bruce L, Butler A, Csillag A, Kuenzel W, Medina L, Paxinos G, Perkel D, Shimizu T, Striedter G, Wild M, Ball G, Durand S, Gunturkun O, Lee D, Mello CV, Powers A, White S, Hough G, Kubikova L, Smulders TV, Wada K, Dugas-Ford J, Husband S, Yamamoto K, Yu J, Siang C & Jarvis ED (2004). Revised nomenclature for avian telencephalon and some related brainstem nuclei. J Comp Neurol 473, 377–414.
Sheu FS, McCabe BJ, Horn G & Routtenberg A (1993). Learning selectively increases protein kinase C substrate phorphorylation in specific regions of the chick brain. Proc Natl Acad Sci U S A 90, 2705–2709.
, 百拇医药
Sluckin W (1972). Imprinting and Early Learning. Methuen, London.
Smilowitz H, Hijian RA, Dwyer J & Feinstein MB (1981). Regulation of acetylcholine receptor phosphorylation by calcium and calmodulin. Proc Natl Acad Sci U S A 78, 4708–4712.
Snedecor GW & Cochran WG (1989). Statistical Methods. Iowa State University Press, Ames.
Solomonia R, Kiguradze T, McCabe BJ & Horn G (2000). Neural cell adhesion molecules, CaM kinase II and long-term memory in the chick. Neuroreport 11, 3139–3143.
, 百拇医药
Solomonia RO, Kotorashvili A, Kiguradze T, McCabe BJ & Horn G (2003a). Calcium/calmodulin-dependent protein kinase II and memory. Act Neurobiol Exper 63, S76.
Solomonia RO, McCabe BJ & Horn G (1998). Neural cell adhesion molecules, learning and memory. Behav Neurosci 112, 646–655.
Solomonia RO, McCabe BJ, Jackson AP & Horn G (1997). Clathrin proteins and recognition memory. Neuroscience 80, 59–67.
, 百拇医药
Solomonia RO, Morgan K, Kotorashvili A, McCabe BJ, Jackson AP & Horn G (2003b). Analysis of differential gene expression supports a role for amyloid precursor protein and a protein kinase C substrate (MARCKS) in long-term memory. Eur J Neurosci 17, 1073–1081.
Song I & Huganir RL (2002). Regulation of AMPA receptors during synaptic plasticity. Trends Neurosci 25, 578–588.
Steward O & Schuman EM (2001). Protein synthesis at synaptic sites on dendrites. Ann Rev Neurosci 24, 299–325.
, 百拇医药
Strack S, Barban MA, Wadzinski B & Colbran RJ (1997). Differential inactivation of postsynaptic density-associated and soluble Ca2+/calmodulin-dependent protein kinase II by protein phosphatases 1 and 2A. J Neurochem 68, 2119–2128.
Strack S, Robison A, Bass MA & Colbran RJ (2000). Association of calcium/calmodulin dependent kinase II with developmentally regulated splice variants of the postsynaptic density protein densin-180. J Biol Chem 275, 25061–25064.
, 百拇医药
Tanzi E (1893). I fatti e le induzioni nell' odierna istologia del sistema nervoso. Riv Sper Freniat Med Leg Alien Ment 19, 419–472.
Walikonis RS, Oguni A, Khorosheva EM, Jeng CJ, Asuncion FJ, Kennedy MB (2001). Densin 180 forms a ternary complex with the a-subunit of Ca2+/calmodulin-dependent protein kinase II and -actinin. J Neurosci 21, 423–433.
Wu G, Malinow R & Cline HT (1996). Maturation of a central glutamatergic synapse. Science 274, 972–976.
Zhu JJ, Qin Y, Zhao M, Van Aelst L & Malinow R (2002). Ras and Rap control AMPA receptor trafficking during synaptic plasticity. Cell 110, 443–455., 百拇医药(Revaz O. Solomonia, Adam )