- Email: [email protected]
GAP-43 mRNA in chick brain: effects of imprinting
MOLECULAR BRAIN RESEARCH ELSEVIER
Molecular Brain Research 35 (1996) 149-156
Research report
MARCKS and protein F1/GAP-43 mRNA in chick brain: effects of imprinting Peter J. Meberg a,1, Brian J. McCabe b, Aryeh Routtenberg a.* a Cresap Neuroscience Laboratory, 2021 Sheridan Road, Northwestern University, Ecanston, IL 60208, USA b Department of Zoology, Unit,ersity of Cambridge, Cambridge CB2 3EJ, UK Accepted 1 August 1995
Abstract
The phosphorylation of MARCKS, but not protein F1/GAP-43, is increased in the intermediate and medial portion of the hyperstriatum ventrale (IMHV) after chick imprinting. Here we investigated if MARCKS, but not F1/GAP-43, gene expression would also be altered after imprinting. We first investigated the constitutive mRNA distribution of MARCKS and FI/GAP-43 in chick brain. MARCKS mRNA was expressed in most cells and exhibited a relatively homogeneous distribution. In contrast, FI/GAP-43 mRNA levels were elevated in discrete brain regions, as we had observed in mammals. The highest F1/GAP-43 mRNA levels in the chick brain were in sensory and associational structures such as the hyperstriatal complex and neostriatum, and lower levels were in structures involved in motor control, such as paleostriatum. These results in chick are consistent with the previously drawn generalization that F1/GAP-43 mRNA is expressed in those brain regions which exhibit synaptic plasticity. After imprinting, MARCKS mRNA levels in IMHV were higher in good learners than poor learners. In contrast, analysis of FI/GAP-43 mRNA levels revealed no differences related to training in any brain region sampled. These selective results for MARCKS but not F1/GAP-43 parallel the prior findings on their phosphorylation, and are consistent with our hypothesis that the very same proteins that are post-translationally modified in association with learning and memory also undergo alterations in their gene expression. Keywords: MARCKS; Imprinting; Protein FI; GAP-43; mRNA; IMHV; Learning
1. Introduction
The phosphorylation of both MARCKS [16] and protein F1 [5,16] (also known as GAP-43, B-50, neuromodulin), two protein kinase C (PKC) substrates [6,20,21,24], is increased after long-term potentiation of synaptic transmission (LTP), a type of synaptic plasticity that may be important for learning and memory in mammals (see [3] for review). In addition, in monkey cerebral cortex the phosphorylation of F1/GAP-43 and an 81 kDa PKC substrate, probably MARCKS, is greater in temporal lobe areas associated with visual information storage than in occipital regions which are less important for such memory [15]. MARCKS is present in synaptic terminals and axonal growth cones [16,17,24], and crosslinks actin in a
• Corresponding author. Fax: (I) (303) 491-3557. i Present address: Deparment of Biochemistry and Molecular Biology, Colorado State University, Fort Collins, CO 80523-1870, USA. 0169-328X/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved SSDI O1 6 9 - 3 2 8 X ( 9 5 ) 0 0 2 0 0 - 6
PKC-regulated manner [8]. Protein F1/GAP-43 is highly expressed during periods of developmental or regenerative axonal growth and is enriched in presynaptic terminals and growth cones (see [6,20] for review). In the chick, the left intermediate and medial portion of the hyperstriatum ventrale (IMHV) is important for recognition memory [9,10,18]. To determine if MARCKS and F1/GAP-43 might play a broader role in the synaptic modifications that underlie learning and memory, our laboratories previously studied the phosphorylation of MARCKS and F1/GAP-43 in chick IMHV after imprinting [19]. The phosphorylation of MARCKS, but not F 1 / G A P 43, was found to be increased in the IMHV after imprinting [19]. Both the phosphorylation of F1/GAP-43 and its gene expression are altered after LTP [14]. We therefore wished to determine if, in addition to its phosphorylation, the gene expression of MARCKS might also be altered after imprinting. Conversely, if F1/GAP-43 is not involved in the imprinting process, then its gene expression should not
/'..I. Meberg et al. /.~4oh,cular Brain Research 35 (19961 149-156
150
change. Since evidence on the cellular distribution of MARCKS and F 1 / G A P - 4 3 gene expression in chick brain was not available, we first studied their constitutive levels using in situ hybridization. We then determined if imprinting differentially alters the expression of the MARCKS and F 1 / G A P - 4 3 genes.
corresponding to bases 5 6 6 - 1 0 0 0 of the chick GAP-43 cDNA clone [2]. DNA probes made from each of the subclones hybridized to single, appropriate-sized transcripts on northern blots of rat brain total RNA (data not shown).
2.2. Imprinting of chicks 2. Materials and methods
2. l. In situ hybridization In situ hybridization was performed as previously described [13], using an acetic anhydride pretreatment and stringent washing at 60°C in 0.1 × SSC. Two subclones of the MARCKS cDNA were constructed that corresponded to either bases 1-547 or 8 7 6 - 1 5 2 6 of the full-length chick MARCKS cDNA [7]. Hybridization patterns were qualitatively equivalent using riboprobes transcribed from either subelone, but a stronger signal was obtained from the 3' subclone, so that was selected for further studies. The F 1 / G A P - 4 3 riboprobe was transcribed from a subclone
F1/GAP-43
O
The imprinting of chicks (Gallus gallus domesticus) was performed as previously described [19]. Briefly, 15-30 h after hatching chicks were exposed to white light for 3(1 rain and 30 min after the exposure were placed in running wheels with exposure to an illuminated, rotating red box for 60 rain. Twenty minutes after the end of training chicks were exposed to the red box and a novel object, and the percent of approach to the training stimulus (the red box) was taken as an index of imprinting. This preference score would be 50% if by chance, and 100% if all approach activity was directed to the red box. Chicks were killed 3 0 - 4 5 rain after training. There was a significant effect of experimental treatment on preference score ( F [ I , 5 ] = 58.61, P = 0.001), but not on training approach.
¢-.,
MARCKS
Fig. 1. In situ hybridizationon frontal sections through chick telencephalonusing probes complementaryto FI/GAP-43 (A, C) and MARCKS(B, D). Followinghybridizationsectionswere exposed to X-ray film and the resultantfilm was used as a photographic negative. Hyperstriatumaccessorium(HA), hyperstriatum ventrale(HV), and neostriatum(N) exhibit strong FI/GAP-43 hybridization,in marked contrast to the paucity of labeling in paleostriatum augmentatum (PA). Intense FI/GAP-43 labeling is apparent over the archistriatum (A). Although paleostriatum augrnentatumexhibits less MARCKS hybridization than hyperstriatum,the difference in labeling intensityis much less than seen for FI/GAP-43 (compare also Fig. 2A and 2B). In general, MARCKS mRNA expression is more widespread and homogeneousthan F1/GAP-43. A notablc exception is the intense hybridizationalong the ventral portions of the lateral ventricles, indicatedby arrows in D. The sections in A and B are anterior to those in C and D. Sectionsare from imprintedchicks which were poor (A) or g~w,d learners (C. D), or which were dark-reared (B). but the general hybridizationpatterns are representative of all treatment groups.
151
P J . Meberg et al. / Molecular Brain Research 35 (1996) 1 4 9 - 1 5 6
After preference scores were recorded, chicks were assigned three to a group and coded so that in situ hybridization and subsequent quantitation were performed
blind of the training history. Each blocked group w a s from a single experiment and consisted of one good learner, one poor learner and one chick w h i c h w a s not exposed to light.
MARCKS
F1/GAP-43
..•
,
-
:a =,-
'~,~j
-
,_ee--
:
.;~=
'..
#
x?
i"
, 'r.
'~
". %",.:; "~.~:.
,'.7,t~,-'J. q~'~,'.~ '+7'~"~ "
*'
• .~. • "
19 •
, (.
z~t
.
.-
.
"It: ¢
Fig. 2. Comparison of F1/GAP-43 and MARCKS mRNA distributions in specific chick brain structures using in situ hybridization. Dark-field photomicrographs of emulsion-coated slides are shown. F1/GAP-43 hybridization is shown in the photomicrographs on the left (A, C, E, G) and MARCKS is shown on the right (B, D, F, H). The marked difference in hybridization levels between neostriatum (N) and paleostriatum (PA) for FI/GAP--43 (A) is not obtained for MARCKS (B). Both F1/GAP-43 and MARCKS mRNAs are moderately to highly expressed in hippocampus (Hp) and 1MHV (C, D). Hybridization is also shown for the cerebellum (E, F) and the rectum (G, H). The top two pairs of photomicrographs (A-D) are frontal sections, the bottom two pairs are horizontal sections (E-H). gc, granule cell layer of the cerebellum. Scale bars = 200 p.m.
152
P..I. Meberg et al. / Molecular Brain Research 35 (1996) 149-150
The average preference score of the good learners was 97.6 + 3.8% (approach + SD), and that of the poor learners was 52.9 + 12.8%. Brain sections from each group were processed simultaneously and exposed to the same piece of film.
2.3. Quantitatite analysis Approximately 10 sections were analyzed per animal. Measurements of optical density were made over IMHV, hyperstriatum accessorium (HA), and neostriatum on both hemispheres. The background optical density of the film directly adjacent to each section was subtracted from that on the section. Six groups of chicks were analyzed (n = 18 total); for analysis the data were blocked as the matched sets of 3 chicks. To minimize the variation among the different blocked groups, the optical density data were standardized for comparisons as previously described [ 19]: briefly, the group mean was subtracted from each value, and the difference was divided by the standard deviation of the group values. This caused the data to be distributed about a mean of zero for each block. In cases where population variances could not be assumed to be equal, tests of statistical significance were made using a FisherBehrens test, as indicated.
3. Results
3.1. Differential mRNA distribution in chick brain of MARCKS and FI / GAP-43 In situ hybridization was used to determine the cellular distribution of MARCKS and F 1 / G A P - 4 3 mRNA expression in chick brain. The identification and nomenclature of chick brain structures and their mammalian homologues,
discussed below, are based on the atlas of Kuenzel and Masson [ 12]. The cellular selectivity of F 1 / G A P - 4 3 gene expression was readily apparent in coronal sections: hybridization levels in the hyperstriatal complex (which includes IMHV; Figs. 1 and 2C), the neostriatum and a ventromedial telencephalic region were quite high, while hybridization levels in the paleostriatal complex ( Figs. 1 and 2A) and ectostriatum were rather low. Hybridization levels were also high in the hippocampus (Fig. 2C), the tectum (Fig. 2G), and the mitral cell layer of the olfactory bulb (Fig. 3A) and moderate in cerebellar granule cells (Fig. 2E). Although at low magnification hybridization appears to be high in cerebellar granule cells, this impression is distorted by the high cell density. On an individual cell basis, hybridization in granule cells appears at higher magnification to be lower than hybridization in cells of the hyperstriatal complex. The selectivity and distribution of F I / G A P - 4 3 mRNA expression in chick appears to parallel that reported [13] tor homologous structures in the rat. In both chick and rat, F I / G A P - 4 3 mRNA expression is highest in sensory and associational structures (e.g., hyperstriatum/neostriatum and cerebral cortex). In contrast, expression tends to be low in structures involved in motor control (e.g., paleostriatum and caudate/putamen). MARCKS m R N A exhibited a widespread and fairly homogeneous distribution in chick brain, while F I / G A P 43 was more restricted and elevated in particular loci. MARCKS hybridization levels were elevated in HA (Fig. 1B), IMHV ( Figs. ! and 2D) and neostriatum ( Figs. 1 and 2B), with medial structures more highly labeled than lateral ones (Fig. 1B,D), while hybridization levels were lower in the paleostriatal complex ( Figs. 1 and 2B). MARCKS hybridization levels were relatively moderate in the tectum (Fig. 2H) and cerebellar granule cells (Fig. 2F) compared to other brain regions. o .
Fig. 3. F1/GAP-43 mRNA expression in olfactory bulb and pirifi~rmcortex. In A, strong hybridization is apparent in the mitral cell layer of the olfactory bulb (ml) and in B the piriform cortex (Cp) of the chick. Relative F1/GAP-43 gene expression is comparable in the homologous brain structures in the rat [13]. Frontal sections arc shown. Scale bar = 2(IO p_m.
P.J. Meberg et al. / Molecular Brain Research 35 (1996) 149-156
153
~t
"~-
.
Fig. 4. Cells adjacent to the lateral ventricles (Iv) exhibit intense MARCKS mRNA hybridization. Bright-field photomicrographs are shown of frontal sections dipped in nuclear emulsion and stained with cresyl violet. The photomicrograph on the left is a normal autoradiographic exposure, the one on the fight is a shorter exposure in the same region. The cells lining the ventricle show little labeling, but the small cells adjacent to these, and spread slightly farther out, are almost completely covered by silver grains (single cells completely covered with grains are indicated by arrows). These cells exhibit the highest levels of MARCKS hybridization in the chick brain. Scale bar = 20/zm.
The highest level of MARCKS hybridization was over small, densely packed cells immediately adjacent to the ependyma of the lateral ventricles (Fig. 1D, Fig. 4). These are not the ependymai cells. The layer of cells directly lining the ventricle, presumably ependyma, do not exhibit hybridization. This intense labeling was specific, as neither F1/GAP-43 nor MARCKS sense probes nor F1/GAP-43 antisense probes labeled these cells (compare Fig. 1C with Fig. 1D). Because there is no report on the cellular distribution of MARCKS mRNA expression in the rat brain, we are unable to compare chick MARCKS mRNA expression
directly with that of the rat. However, immunohistochemical studies indicate that MARCKS protein is widely expressed in the majority of cell types in the rat brain [17], which is consistent with the widespread mRNA expression found in chick brain.
3.2. Effects of imprinting on MARCKS and F1/GAP-43 gene expression MARCKS mRNA levels differed among the good learners, poor learners and the dark-reared chicks both in left IMHV (F[2,15] = 3.75, P < 0.05) and in right IMHV
Table 1 Average F1/GAP-43 and MARCKS mRNA hybridization in the chick after imprinting (raw OD, standard deviation) Region
Hybridization dark-reared
poor learners
good learners
MARCKS mRNA IMHV (L) IMHV (R) HA (L) HA (R) NEO (R) NEO (L)
48.2 48.5 39.4 41.7 37.1 37.7
(41.4) (41.1) (23.4) (26.0) (31.5) (35.3)
19.0 (11.2) 20.0 (10.1) 22.5 (13.2) 20.1 (13.8) 13.6 (8.0) 12.8 (5.4)
31.3 32.8 32.0 28.9 23.9 24.9
(17.7) (17.7) (15.4) (12.0) (13.0) (12.4)
FI / GAP-43 mRNA IMHV (L) IMHV (R) HA (L) HA (R) NEO (L) NEO (R)
49.5 50.5 44.6 52.1 40.7 40.6
(15.7) (16.0) (20.8) (19.4) (12.0) (12.5)
48.4 48.7 52.4 50.2 39.7 41.3
43.1 43.1 50.9 50.6 36.3 37.7
(15.4) (14.7) (19.4) (17.6) (14.4) (16.9)
Abbreviations: L, left side; R, fight side; NEO, neostriatum.
(12.9) (11.9) (13.9) (17.9) (13.1) (9.6)
P..I. Meberg et al. / Molecular Brain Research 35 (19961 149-156
154
Table 2 Standardized scores for average F 1 / G A P - 4 3 and M A R C K S m R N A hybridization in the chick after imprinting (standardized OD; see Methods) Region
Hybridization dark-reared
ANOVA poor learners
g o o d learners
F[2,151
P
MARCKS mRNA IMItV (L) I M H V (R) H A (13 IIA (R) N E O (I.) N E O (R)
11.43 11.41 11.33 0.39 11.37 0.36
- 11.66 - 0.65 - 11.46 - 11.51 - 0.49 -- 11.59
0.23 [I.24 11.13 Ikl2 1~.12 11.23
3.75 3.67 1.54 2.114 1.80 2.66
11.048 0.050 11.25 0.17 11.20 11.10
0.37 0.38 - [I.22 11.22 11.39 [I.29
I).114 0.11 11.29 11.112 - 0.19 0.18
- 1).41 -0.48 • [I.07 - 0.29 - 0.2(1 - 11.48
1.38 1.80 0.48 0.49 0.97 1.58
0.28 0.20 0.63 0.62 0.40 11.24
FI / GAP-43 mRNA IMHV (L) I M H V (R) HA (L) HA (RI N E O (R) N E O (L)
Abbreviations: L, left side; R, right side; NEO, n e o s t r i a t u m
( F [ 2 , 1 5 ] = 3.67, P = 0 . 0 5 0 ) . In the other brain regions analyzed no differences in MARCKS mRNA were observed among the three groups (see Tables 1 and 2). There was greater MARCKS mRNA expression in good learners (standardized mean + SD = 0.225 + 0.735) than in poor learners ( - 0 . 6 5 6 + 0.41t0) in the left IMHV (t = 2.58, 10 dr, P < 0.05). Surprisingly, greater MARCKS mRNA expression was detected in dark-reared chicks (0.431 + 0.95) than in poor learners (Fisher-Behrens test, 6 df, P < 0.05; see Materials and methods), and there was no difference between good learners and dark-reared chicks (t = 0.42, 10 df, P = 11.68) in left IMHV. Similar results were obtained for the right IMHV, with greater MARCKS mRNA expression in good learners than in poor (t = 2.59, 10 df, P < 0.05). There was also a significant correlation between standardized MARCKS expression and preference scores in left IMHV ( r = +11.59, P < 0 . 0 5 ) and right IMHV (r = + 0.58, P < 0.05), but no correlation between preference score and MARCKS mRNA expression in neostriatum or HA (e.g., in left HA: r = + 0.35). F I / G A P - 4 3 gene expression did not differ among good learners, poor learners, and dark-reared chicks in any of the brain regions analyzed (see Tables 1 and 2).
4. Discussion
4.1. Distribution of MARCKS and F 1 / G A P - 4 3 mRNAs in chick brain MARCKS mRNA was widely expressed throughout the chick brain, with some variation in hybridization levels among different brain regions. In contrast to MARCKS, F 1 / G A P - 4 3 mRNA hybridization demonstrated a high degree of selective expression on a regional and cellular
basis. In particular, sensory neurons and associational structures such as the hyperstriatal complex and the neostriatum exhibited high levels of hybridization, while structures involved in motor control, such as the paleostriatum, exhibited lower levels. The similarity in relative levels of F I / G A P - 4 3 m R N A expression between homologous chick and rat brain structures suggests that there is not only an evolutionarily conserved primary protein sequence [2], but also conservation of gene regulation on a brain regional basis. The highest expression of MARCKS mRNA was found in cells along the lateral ventricles. Although this is a proliferative zone for neurons, neurogenesis and migration of neurons are complete days before the chicks hatch [23]. These cells are likely to be glia, based on location and the small size of the cell bodies, lmmunocytochemistry has indicated that the MARCKS protein is highly expressed in many glial cells in the rat brain [17].
4.2. Effects of imprinting on gene expression After imprinting MARCKS gene expression in IMHV was greater in good learners than in poor learners. In contrast to MARCKS, imprinting effects on F I / G A P - 4 3 mRNA expression were not observed. Since we previously found that the phosphorylation of MARCKS, but not F I / G A P - 4 3 , is increased after imprinting [19], these results are consistent with the hypothesis [14] that activitydependent alterations in gene expression occur for the same proteins that are initially post-translationally modified. However, in-the present study alterations were observed in right IMHV for MARCKS mRNA, but such a difference was not statistically significant in right IMHV for MARCKS phosphorylation [19]. It should be noted, however that the right IMHV showed a 26% increase in
P.J. Meberg et al. / Molecular Brain Research 35 (1996) 149-156
phosphorylation after training and a correlation with training of +0.50. In the left IMHV the increase, though significant, was just 25.5% while the correlation was + 0.76. Thus a non-significant trend was observed in the right IMHV. Although these data indicate a relation between the level of MARCKS mRNA expression and imprinting in the chick, it is possible that chicks having greater MARCKS mRNA expression in IMHV before imprinting were more capable of being strongly imprinted. Consistent with this view was the finding that the expression in good learners did not differ from that in dark-reared chicks. But then one would also expect the mean hybridization for the dark-reared group to be the average of both poor and good learners. However, the mean value for the dark-reared animals was significantly greater than the poor learners, and non-significantly greater than the good learners. It is therefore possible that light exposure itself, or some other aspect of the imprinting procedure, affects MARCKS gene expression. However, when chicks received equivalent light exposure, those which were good learners had greater MARCKS expression than those which were not. This strongly suggests that the level of MARCKS gene expression and imprinting may be mechanistically related. The synaptic modifications known to occur after imprinting [11] could directly involve MARCKS. MARCKS is present in dendrites and axon terminals [17,24], is closely associated with microtubules [17], and binds actin [8]. Since phosphorylation by PKC inhibits the ability of MARCKS to cross-link actin filaments [8] and associate with the plasma membrane [22], the increased MARCKS phosphorylation after imprinting [19] may be part of a dynamic mechanism for altering synaptic morphology. The greater MARCKS mRNA expression in good learners may contribute to a coordinated cytoskeletal mechanism whereby MARCKS phosphorylation, on the one hand, allows for transient synaptic growth through the re-organization of actin filaments, and the increased MARCKS synthesis, on the other hand, results in a greater stabilization, re-crosslinking the actin filaments and solidifying the imprinting-modified synapse. Such a stabilization process has also been suggested to occur when decreased F1/GAP-43 mRNA expression occurs after LTP [14]. Thus, for long-term storage of new memories, both synaptic plasticity and subsequent stabilization are required. It is curious that MARCKS, but not F1/GAP-43, is altered in relation to imprinting since F1/GAP-43 has been implicated in the process of learning and its underlying synaptic plasticity in mammals [6]. Even though the learning is in immature birds, the absence of a change in F1/GAP-43 is puzzling. It is also surprising since F1/GAP-43 phosphorylation is increased in IMHV after avoidance learning in chicks [1,4]. Perhaps learning in a more mature bird would require the involvement of F1/GAP-43, or perhaps F1/GAP-43 serves a role only in specific types of learning-related synaptic modifications. It
155
is possible that in both chick and rat, the common mechanism of learning and synaptic change is the activation of protein kinase C, and the difference lies in the target substrates that are both post-translationally modified and altered in gene expression.
Acknowledgements We would like to thank L. Baizer for generously providing the chick F1/GAP-43 cDNA and P.J. Blackshear for generously providing the chick MARCKS cDNA. Thanks also to S. Hicks for assisting with quantitation. This work supported by NIMH grant MH25281-21 to A.R.
References [1] Ali, S.M., Bullock, S. and Rose, S.P.R., Phosphorylation of synaptic proteins in chick forebrain; changes with development and passive avoidance training, J. Neurochem., 50 (1988) 1579-1587. [2] Baizer, L., Alkan, S., Stocker, K. and Ciment, G, Chicken growthassociated protein (GAP)-43: primary structure and regulated expression of mRNA during embryogenesis, Mol. Brain Res., 7 (1990) 61-68. [3] Bliss, T.V.P. and Collingridge, G.L., A synaptic model of memory: long-term potentiation in the hippocampus, Nature, 361 (1993) 31-39. [4] Bullock, S., De Graan, P.N.E., Oestreicher, A.B., Gispen, W.-H. and Rose, S.P.R., Identification of a 52 kDa chick brain membrane protein showing changed phosphorylation after passive avoidance training as B-50 (GAP 43), Neurosci. Res. Comman., 6 (1990) 181-186. [5] Gianotti, C., Nunzi, M.F., Gispen, W.H. and Corradetti, R., Phosphorylation of the presynaptic protein B-50 (GAP-43) is increased during electrically induced long-term potentiation, Neuron, 8 (1992) 843-848. [6] Gispen, W.H. and Routtenberg, A. (Eds.), Protein Kinase C and its Brain Substrates: Role in Neuronal Growth and Plasticity, Elsevier, Amsterdam, 1991. [7] Graft, J.M., Stumpo, D.J. and Blackshear, P.J., Molecular cloning, sequence and expression of a cDNA encoding the chicken myristoylated alanine-rich C kinase substrate (MARCKS), Mol. Endocrinol., 3 (1989) 1903-1906. [8] Hartwig, J.H., Thelen, M., Rosen, A., Janmey, P.A., Nairn, A.C. and Aderem, A., MARCKS is an actin filament crosslinking protein regulated by protein kinase C and calcium-calmodulin, Nature, 356 (1992) 618-622. [9] Horn, G., Neural mechanisms of learning: an analysis of imprinting in the domestic chick, Proc. R. Soc. Lond. B, 213 (1981) 101-137. [10] Horn, G., Memory, Imprinting and the Brain: An Inquiry into Mechanisms, Clarendon Press, Oxford, 1985. [11] Horn, G., Bradley, P. and McCabe, B.J., Changes in the structure of synapses associated with learning, J. Neurosci., 5 (1985)3161-3168. [12] Kuenzel, W.J. and Masson, M., A Stereotaxic Atlas of the Brain of the Chick (Gallus domesticus), Johns Hopkins University Press, Baltimore, MD, 1988. [13] Meberg, P.J. and Routtenberg, A., Selective expression of protein FI mRNA in pyramidal but not granule cells of the hippocampus, Neuroscience, 45 (1991) 721-733. [14] Meberg, P.J., Barnes, C.A., McNaughton, B.L. and Routtenberg, A., Protein kinase C and FI gene expression in hippocampus inversely related to synaptic enhancement (LTE/LTP) lasting three days, Proc. Natl. Acad. Sci. USA, 90 (1993) 12050-12054.
156
P.J. Meberg et ak /Molecular Brain Research 35 (1996) 149-156
[15] Nelson, R.B., Friedman, D.P., O'Neill, J.B., Mishkin, M. and Routtenberg, A., Gradients of protein kinase C substrate phosphorylation in primate visual system peak in visual memory storage areas, Brahl Res., 416 (1987) 387-392. [16] Nelson, R.B., Linden, D.J., Hyman, C., Pfenninger, K.H. and Routtenberg, A., The two major phosphoproteins in growth cones are probably identical to two protein kinase C substrates correlated with persistence of long-term potentiation, d. Neurosci.. 9 (1989) 381389. [17] Ouimet, C.C., Wang, J.K.T., Walaas, S.I., Albert, K.A. and Greengard, P., Localization of the MARCKS (87 kDa) protein, a major specific substrate for protein kinase C, in rat brain..I. Neurosci., 10 (1990) 1683-1698. [18] Rose, S.P.R., How chicks make memories: the cellular cascade from c-fos to dendritic modelling, Trends Neuroxci., 14 (1991) 390-397. [19] Sheu, F.-S., McCabe, B.J., Horn, G. and Routtenberg, A., Learning selectively increases protein kinase C substrate phosphorylation in specific regions of the chick brain. Proc. Natl. Acad. Sci. USA. oh) (1993) 2705-271)9.
[20] Skene, J.H.P., Axonal growth-associated proteins, Annu. Rec. Neurosci., 12 (1989) 127-156. [21] Stumpo, D.J., Graft, J.M., Albert, K.A., Greengard, P. and Blackshear, P.J.. Molecular cloning, characterization and expression of a eDNA encoding the '80- to 87-kDa' myristoylated alanine-rich C kinase substrate: a major cellular substrate for protein kinase C, Proc. Natl. Acad. Sci. USA, 86 (1989) 4012-4016. [22] Thelen, M., Rosen, A., Nairn. A.C. and Aderem, A., Regulation by phosphorylation of reversible association of a myristoylated protein kinase C substrate with the plasma membrane, Nature, 351 (1991) 320-322. [23] Tsai, H.M, Garber, B.B. and Larramendi, L.M.H., 3H-thymidine autoradiographic analysis of telencephalic histogenesis in the chick embryo: I. Neuronal birthdates of telencephalic compartments in situ, J. Comp. Neurol., 198 (1981) 275-292. [24] Wu, W.C.-S., Walaas, S.I., Nairn, A.C. and Grcengard, P., Calcium/phospholipid regulates phosphorylation of a M, '87k' substrate protein in brain synaptosomes, Proc. Natl. Acad. Sci. USA, 79 (1982) 5249-5253.
Molecular Brain Research 35 (1996) 149-156
Research report
MARCKS and protein F1/GAP-43 mRNA in chick brain: effects of imprinting Peter J. Meberg a,1, Brian J. McCabe b, Aryeh Routtenberg a.* a Cresap Neuroscience Laboratory, 2021 Sheridan Road, Northwestern University, Ecanston, IL 60208, USA b Department of Zoology, Unit,ersity of Cambridge, Cambridge CB2 3EJ, UK Accepted 1 August 1995
Abstract
The phosphorylation of MARCKS, but not protein F1/GAP-43, is increased in the intermediate and medial portion of the hyperstriatum ventrale (IMHV) after chick imprinting. Here we investigated if MARCKS, but not F1/GAP-43, gene expression would also be altered after imprinting. We first investigated the constitutive mRNA distribution of MARCKS and FI/GAP-43 in chick brain. MARCKS mRNA was expressed in most cells and exhibited a relatively homogeneous distribution. In contrast, FI/GAP-43 mRNA levels were elevated in discrete brain regions, as we had observed in mammals. The highest F1/GAP-43 mRNA levels in the chick brain were in sensory and associational structures such as the hyperstriatal complex and neostriatum, and lower levels were in structures involved in motor control, such as paleostriatum. These results in chick are consistent with the previously drawn generalization that F1/GAP-43 mRNA is expressed in those brain regions which exhibit synaptic plasticity. After imprinting, MARCKS mRNA levels in IMHV were higher in good learners than poor learners. In contrast, analysis of FI/GAP-43 mRNA levels revealed no differences related to training in any brain region sampled. These selective results for MARCKS but not F1/GAP-43 parallel the prior findings on their phosphorylation, and are consistent with our hypothesis that the very same proteins that are post-translationally modified in association with learning and memory also undergo alterations in their gene expression. Keywords: MARCKS; Imprinting; Protein FI; GAP-43; mRNA; IMHV; Learning
1. Introduction
The phosphorylation of both MARCKS [16] and protein F1 [5,16] (also known as GAP-43, B-50, neuromodulin), two protein kinase C (PKC) substrates [6,20,21,24], is increased after long-term potentiation of synaptic transmission (LTP), a type of synaptic plasticity that may be important for learning and memory in mammals (see [3] for review). In addition, in monkey cerebral cortex the phosphorylation of F1/GAP-43 and an 81 kDa PKC substrate, probably MARCKS, is greater in temporal lobe areas associated with visual information storage than in occipital regions which are less important for such memory [15]. MARCKS is present in synaptic terminals and axonal growth cones [16,17,24], and crosslinks actin in a
• Corresponding author. Fax: (I) (303) 491-3557. i Present address: Deparment of Biochemistry and Molecular Biology, Colorado State University, Fort Collins, CO 80523-1870, USA. 0169-328X/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved SSDI O1 6 9 - 3 2 8 X ( 9 5 ) 0 0 2 0 0 - 6
PKC-regulated manner [8]. Protein F1/GAP-43 is highly expressed during periods of developmental or regenerative axonal growth and is enriched in presynaptic terminals and growth cones (see [6,20] for review). In the chick, the left intermediate and medial portion of the hyperstriatum ventrale (IMHV) is important for recognition memory [9,10,18]. To determine if MARCKS and F1/GAP-43 might play a broader role in the synaptic modifications that underlie learning and memory, our laboratories previously studied the phosphorylation of MARCKS and F1/GAP-43 in chick IMHV after imprinting [19]. The phosphorylation of MARCKS, but not F 1 / G A P 43, was found to be increased in the IMHV after imprinting [19]. Both the phosphorylation of F1/GAP-43 and its gene expression are altered after LTP [14]. We therefore wished to determine if, in addition to its phosphorylation, the gene expression of MARCKS might also be altered after imprinting. Conversely, if F1/GAP-43 is not involved in the imprinting process, then its gene expression should not
/'..I. Meberg et al. /.~4oh,cular Brain Research 35 (19961 149-156
150
change. Since evidence on the cellular distribution of MARCKS and F 1 / G A P - 4 3 gene expression in chick brain was not available, we first studied their constitutive levels using in situ hybridization. We then determined if imprinting differentially alters the expression of the MARCKS and F 1 / G A P - 4 3 genes.
corresponding to bases 5 6 6 - 1 0 0 0 of the chick GAP-43 cDNA clone [2]. DNA probes made from each of the subclones hybridized to single, appropriate-sized transcripts on northern blots of rat brain total RNA (data not shown).
2.2. Imprinting of chicks 2. Materials and methods
2. l. In situ hybridization In situ hybridization was performed as previously described [13], using an acetic anhydride pretreatment and stringent washing at 60°C in 0.1 × SSC. Two subclones of the MARCKS cDNA were constructed that corresponded to either bases 1-547 or 8 7 6 - 1 5 2 6 of the full-length chick MARCKS cDNA [7]. Hybridization patterns were qualitatively equivalent using riboprobes transcribed from either subelone, but a stronger signal was obtained from the 3' subclone, so that was selected for further studies. The F 1 / G A P - 4 3 riboprobe was transcribed from a subclone
F1/GAP-43
O
The imprinting of chicks (Gallus gallus domesticus) was performed as previously described [19]. Briefly, 15-30 h after hatching chicks were exposed to white light for 3(1 rain and 30 min after the exposure were placed in running wheels with exposure to an illuminated, rotating red box for 60 rain. Twenty minutes after the end of training chicks were exposed to the red box and a novel object, and the percent of approach to the training stimulus (the red box) was taken as an index of imprinting. This preference score would be 50% if by chance, and 100% if all approach activity was directed to the red box. Chicks were killed 3 0 - 4 5 rain after training. There was a significant effect of experimental treatment on preference score ( F [ I , 5 ] = 58.61, P = 0.001), but not on training approach.
¢-.,
MARCKS
Fig. 1. In situ hybridizationon frontal sections through chick telencephalonusing probes complementaryto FI/GAP-43 (A, C) and MARCKS(B, D). Followinghybridizationsectionswere exposed to X-ray film and the resultantfilm was used as a photographic negative. Hyperstriatumaccessorium(HA), hyperstriatum ventrale(HV), and neostriatum(N) exhibit strong FI/GAP-43 hybridization,in marked contrast to the paucity of labeling in paleostriatum augmentatum (PA). Intense FI/GAP-43 labeling is apparent over the archistriatum (A). Although paleostriatum augrnentatumexhibits less MARCKS hybridization than hyperstriatum,the difference in labeling intensityis much less than seen for FI/GAP-43 (compare also Fig. 2A and 2B). In general, MARCKS mRNA expression is more widespread and homogeneousthan F1/GAP-43. A notablc exception is the intense hybridizationalong the ventral portions of the lateral ventricles, indicatedby arrows in D. The sections in A and B are anterior to those in C and D. Sectionsare from imprintedchicks which were poor (A) or g~w,d learners (C. D), or which were dark-reared (B). but the general hybridizationpatterns are representative of all treatment groups.
151
P J . Meberg et al. / Molecular Brain Research 35 (1996) 1 4 9 - 1 5 6
After preference scores were recorded, chicks were assigned three to a group and coded so that in situ hybridization and subsequent quantitation were performed
blind of the training history. Each blocked group w a s from a single experiment and consisted of one good learner, one poor learner and one chick w h i c h w a s not exposed to light.
MARCKS
F1/GAP-43
..•
,
-
:a =,-
'~,~j
-
,_ee--
:
.;~=
'..
#
x?
i"
, 'r.
'~
". %",.:; "~.~:.
,'.7,t~,-'J. q~'~,'.~ '+7'~"~ "
*'
• .~. • "
19 •
, (.
z~t
.
.-
.
"It: ¢
Fig. 2. Comparison of F1/GAP-43 and MARCKS mRNA distributions in specific chick brain structures using in situ hybridization. Dark-field photomicrographs of emulsion-coated slides are shown. F1/GAP-43 hybridization is shown in the photomicrographs on the left (A, C, E, G) and MARCKS is shown on the right (B, D, F, H). The marked difference in hybridization levels between neostriatum (N) and paleostriatum (PA) for FI/GAP--43 (A) is not obtained for MARCKS (B). Both F1/GAP-43 and MARCKS mRNAs are moderately to highly expressed in hippocampus (Hp) and 1MHV (C, D). Hybridization is also shown for the cerebellum (E, F) and the rectum (G, H). The top two pairs of photomicrographs (A-D) are frontal sections, the bottom two pairs are horizontal sections (E-H). gc, granule cell layer of the cerebellum. Scale bars = 200 p.m.
152
P..I. Meberg et al. / Molecular Brain Research 35 (1996) 149-150
The average preference score of the good learners was 97.6 + 3.8% (approach + SD), and that of the poor learners was 52.9 + 12.8%. Brain sections from each group were processed simultaneously and exposed to the same piece of film.
2.3. Quantitatite analysis Approximately 10 sections were analyzed per animal. Measurements of optical density were made over IMHV, hyperstriatum accessorium (HA), and neostriatum on both hemispheres. The background optical density of the film directly adjacent to each section was subtracted from that on the section. Six groups of chicks were analyzed (n = 18 total); for analysis the data were blocked as the matched sets of 3 chicks. To minimize the variation among the different blocked groups, the optical density data were standardized for comparisons as previously described [ 19]: briefly, the group mean was subtracted from each value, and the difference was divided by the standard deviation of the group values. This caused the data to be distributed about a mean of zero for each block. In cases where population variances could not be assumed to be equal, tests of statistical significance were made using a FisherBehrens test, as indicated.
3. Results
3.1. Differential mRNA distribution in chick brain of MARCKS and FI / GAP-43 In situ hybridization was used to determine the cellular distribution of MARCKS and F 1 / G A P - 4 3 mRNA expression in chick brain. The identification and nomenclature of chick brain structures and their mammalian homologues,
discussed below, are based on the atlas of Kuenzel and Masson [ 12]. The cellular selectivity of F 1 / G A P - 4 3 gene expression was readily apparent in coronal sections: hybridization levels in the hyperstriatal complex (which includes IMHV; Figs. 1 and 2C), the neostriatum and a ventromedial telencephalic region were quite high, while hybridization levels in the paleostriatal complex ( Figs. 1 and 2A) and ectostriatum were rather low. Hybridization levels were also high in the hippocampus (Fig. 2C), the tectum (Fig. 2G), and the mitral cell layer of the olfactory bulb (Fig. 3A) and moderate in cerebellar granule cells (Fig. 2E). Although at low magnification hybridization appears to be high in cerebellar granule cells, this impression is distorted by the high cell density. On an individual cell basis, hybridization in granule cells appears at higher magnification to be lower than hybridization in cells of the hyperstriatal complex. The selectivity and distribution of F I / G A P - 4 3 mRNA expression in chick appears to parallel that reported [13] tor homologous structures in the rat. In both chick and rat, F I / G A P - 4 3 mRNA expression is highest in sensory and associational structures (e.g., hyperstriatum/neostriatum and cerebral cortex). In contrast, expression tends to be low in structures involved in motor control (e.g., paleostriatum and caudate/putamen). MARCKS m R N A exhibited a widespread and fairly homogeneous distribution in chick brain, while F I / G A P 43 was more restricted and elevated in particular loci. MARCKS hybridization levels were elevated in HA (Fig. 1B), IMHV ( Figs. ! and 2D) and neostriatum ( Figs. 1 and 2B), with medial structures more highly labeled than lateral ones (Fig. 1B,D), while hybridization levels were lower in the paleostriatal complex ( Figs. 1 and 2B). MARCKS hybridization levels were relatively moderate in the tectum (Fig. 2H) and cerebellar granule cells (Fig. 2F) compared to other brain regions. o .
Fig. 3. F1/GAP-43 mRNA expression in olfactory bulb and pirifi~rmcortex. In A, strong hybridization is apparent in the mitral cell layer of the olfactory bulb (ml) and in B the piriform cortex (Cp) of the chick. Relative F1/GAP-43 gene expression is comparable in the homologous brain structures in the rat [13]. Frontal sections arc shown. Scale bar = 2(IO p_m.
P.J. Meberg et al. / Molecular Brain Research 35 (1996) 149-156
153
~t
"~-
.
Fig. 4. Cells adjacent to the lateral ventricles (Iv) exhibit intense MARCKS mRNA hybridization. Bright-field photomicrographs are shown of frontal sections dipped in nuclear emulsion and stained with cresyl violet. The photomicrograph on the left is a normal autoradiographic exposure, the one on the fight is a shorter exposure in the same region. The cells lining the ventricle show little labeling, but the small cells adjacent to these, and spread slightly farther out, are almost completely covered by silver grains (single cells completely covered with grains are indicated by arrows). These cells exhibit the highest levels of MARCKS hybridization in the chick brain. Scale bar = 20/zm.
The highest level of MARCKS hybridization was over small, densely packed cells immediately adjacent to the ependyma of the lateral ventricles (Fig. 1D, Fig. 4). These are not the ependymai cells. The layer of cells directly lining the ventricle, presumably ependyma, do not exhibit hybridization. This intense labeling was specific, as neither F1/GAP-43 nor MARCKS sense probes nor F1/GAP-43 antisense probes labeled these cells (compare Fig. 1C with Fig. 1D). Because there is no report on the cellular distribution of MARCKS mRNA expression in the rat brain, we are unable to compare chick MARCKS mRNA expression
directly with that of the rat. However, immunohistochemical studies indicate that MARCKS protein is widely expressed in the majority of cell types in the rat brain [17], which is consistent with the widespread mRNA expression found in chick brain.
3.2. Effects of imprinting on MARCKS and F1/GAP-43 gene expression MARCKS mRNA levels differed among the good learners, poor learners and the dark-reared chicks both in left IMHV (F[2,15] = 3.75, P < 0.05) and in right IMHV
Table 1 Average F1/GAP-43 and MARCKS mRNA hybridization in the chick after imprinting (raw OD, standard deviation) Region
Hybridization dark-reared
poor learners
good learners
MARCKS mRNA IMHV (L) IMHV (R) HA (L) HA (R) NEO (R) NEO (L)
48.2 48.5 39.4 41.7 37.1 37.7
(41.4) (41.1) (23.4) (26.0) (31.5) (35.3)
19.0 (11.2) 20.0 (10.1) 22.5 (13.2) 20.1 (13.8) 13.6 (8.0) 12.8 (5.4)
31.3 32.8 32.0 28.9 23.9 24.9
(17.7) (17.7) (15.4) (12.0) (13.0) (12.4)
FI / GAP-43 mRNA IMHV (L) IMHV (R) HA (L) HA (R) NEO (L) NEO (R)
49.5 50.5 44.6 52.1 40.7 40.6
(15.7) (16.0) (20.8) (19.4) (12.0) (12.5)
48.4 48.7 52.4 50.2 39.7 41.3
43.1 43.1 50.9 50.6 36.3 37.7
(15.4) (14.7) (19.4) (17.6) (14.4) (16.9)
Abbreviations: L, left side; R, fight side; NEO, neostriatum.
(12.9) (11.9) (13.9) (17.9) (13.1) (9.6)
P..I. Meberg et al. / Molecular Brain Research 35 (19961 149-156
154
Table 2 Standardized scores for average F 1 / G A P - 4 3 and M A R C K S m R N A hybridization in the chick after imprinting (standardized OD; see Methods) Region
Hybridization dark-reared
ANOVA poor learners
g o o d learners
F[2,151
P
MARCKS mRNA IMItV (L) I M H V (R) H A (13 IIA (R) N E O (I.) N E O (R)
11.43 11.41 11.33 0.39 11.37 0.36
- 11.66 - 0.65 - 11.46 - 11.51 - 0.49 -- 11.59
0.23 [I.24 11.13 Ikl2 1~.12 11.23
3.75 3.67 1.54 2.114 1.80 2.66
11.048 0.050 11.25 0.17 11.20 11.10
0.37 0.38 - [I.22 11.22 11.39 [I.29
I).114 0.11 11.29 11.112 - 0.19 0.18
- 1).41 -0.48 • [I.07 - 0.29 - 0.2(1 - 11.48
1.38 1.80 0.48 0.49 0.97 1.58
0.28 0.20 0.63 0.62 0.40 11.24
FI / GAP-43 mRNA IMHV (L) I M H V (R) HA (L) HA (RI N E O (R) N E O (L)
Abbreviations: L, left side; R, right side; NEO, n e o s t r i a t u m
( F [ 2 , 1 5 ] = 3.67, P = 0 . 0 5 0 ) . In the other brain regions analyzed no differences in MARCKS mRNA were observed among the three groups (see Tables 1 and 2). There was greater MARCKS mRNA expression in good learners (standardized mean + SD = 0.225 + 0.735) than in poor learners ( - 0 . 6 5 6 + 0.41t0) in the left IMHV (t = 2.58, 10 dr, P < 0.05). Surprisingly, greater MARCKS mRNA expression was detected in dark-reared chicks (0.431 + 0.95) than in poor learners (Fisher-Behrens test, 6 df, P < 0.05; see Materials and methods), and there was no difference between good learners and dark-reared chicks (t = 0.42, 10 df, P = 11.68) in left IMHV. Similar results were obtained for the right IMHV, with greater MARCKS mRNA expression in good learners than in poor (t = 2.59, 10 df, P < 0.05). There was also a significant correlation between standardized MARCKS expression and preference scores in left IMHV ( r = +11.59, P < 0 . 0 5 ) and right IMHV (r = + 0.58, P < 0.05), but no correlation between preference score and MARCKS mRNA expression in neostriatum or HA (e.g., in left HA: r = + 0.35). F I / G A P - 4 3 gene expression did not differ among good learners, poor learners, and dark-reared chicks in any of the brain regions analyzed (see Tables 1 and 2).
4. Discussion
4.1. Distribution of MARCKS and F 1 / G A P - 4 3 mRNAs in chick brain MARCKS mRNA was widely expressed throughout the chick brain, with some variation in hybridization levels among different brain regions. In contrast to MARCKS, F 1 / G A P - 4 3 mRNA hybridization demonstrated a high degree of selective expression on a regional and cellular
basis. In particular, sensory neurons and associational structures such as the hyperstriatal complex and the neostriatum exhibited high levels of hybridization, while structures involved in motor control, such as the paleostriatum, exhibited lower levels. The similarity in relative levels of F I / G A P - 4 3 m R N A expression between homologous chick and rat brain structures suggests that there is not only an evolutionarily conserved primary protein sequence [2], but also conservation of gene regulation on a brain regional basis. The highest expression of MARCKS mRNA was found in cells along the lateral ventricles. Although this is a proliferative zone for neurons, neurogenesis and migration of neurons are complete days before the chicks hatch [23]. These cells are likely to be glia, based on location and the small size of the cell bodies, lmmunocytochemistry has indicated that the MARCKS protein is highly expressed in many glial cells in the rat brain [17].
4.2. Effects of imprinting on gene expression After imprinting MARCKS gene expression in IMHV was greater in good learners than in poor learners. In contrast to MARCKS, imprinting effects on F I / G A P - 4 3 mRNA expression were not observed. Since we previously found that the phosphorylation of MARCKS, but not F I / G A P - 4 3 , is increased after imprinting [19], these results are consistent with the hypothesis [14] that activitydependent alterations in gene expression occur for the same proteins that are initially post-translationally modified. However, in-the present study alterations were observed in right IMHV for MARCKS mRNA, but such a difference was not statistically significant in right IMHV for MARCKS phosphorylation [19]. It should be noted, however that the right IMHV showed a 26% increase in
P.J. Meberg et al. / Molecular Brain Research 35 (1996) 149-156
phosphorylation after training and a correlation with training of +0.50. In the left IMHV the increase, though significant, was just 25.5% while the correlation was + 0.76. Thus a non-significant trend was observed in the right IMHV. Although these data indicate a relation between the level of MARCKS mRNA expression and imprinting in the chick, it is possible that chicks having greater MARCKS mRNA expression in IMHV before imprinting were more capable of being strongly imprinted. Consistent with this view was the finding that the expression in good learners did not differ from that in dark-reared chicks. But then one would also expect the mean hybridization for the dark-reared group to be the average of both poor and good learners. However, the mean value for the dark-reared animals was significantly greater than the poor learners, and non-significantly greater than the good learners. It is therefore possible that light exposure itself, or some other aspect of the imprinting procedure, affects MARCKS gene expression. However, when chicks received equivalent light exposure, those which were good learners had greater MARCKS expression than those which were not. This strongly suggests that the level of MARCKS gene expression and imprinting may be mechanistically related. The synaptic modifications known to occur after imprinting [11] could directly involve MARCKS. MARCKS is present in dendrites and axon terminals [17,24], is closely associated with microtubules [17], and binds actin [8]. Since phosphorylation by PKC inhibits the ability of MARCKS to cross-link actin filaments [8] and associate with the plasma membrane [22], the increased MARCKS phosphorylation after imprinting [19] may be part of a dynamic mechanism for altering synaptic morphology. The greater MARCKS mRNA expression in good learners may contribute to a coordinated cytoskeletal mechanism whereby MARCKS phosphorylation, on the one hand, allows for transient synaptic growth through the re-organization of actin filaments, and the increased MARCKS synthesis, on the other hand, results in a greater stabilization, re-crosslinking the actin filaments and solidifying the imprinting-modified synapse. Such a stabilization process has also been suggested to occur when decreased F1/GAP-43 mRNA expression occurs after LTP [14]. Thus, for long-term storage of new memories, both synaptic plasticity and subsequent stabilization are required. It is curious that MARCKS, but not F1/GAP-43, is altered in relation to imprinting since F1/GAP-43 has been implicated in the process of learning and its underlying synaptic plasticity in mammals [6]. Even though the learning is in immature birds, the absence of a change in F1/GAP-43 is puzzling. It is also surprising since F1/GAP-43 phosphorylation is increased in IMHV after avoidance learning in chicks [1,4]. Perhaps learning in a more mature bird would require the involvement of F1/GAP-43, or perhaps F1/GAP-43 serves a role only in specific types of learning-related synaptic modifications. It
155
is possible that in both chick and rat, the common mechanism of learning and synaptic change is the activation of protein kinase C, and the difference lies in the target substrates that are both post-translationally modified and altered in gene expression.
Acknowledgements We would like to thank L. Baizer for generously providing the chick F1/GAP-43 cDNA and P.J. Blackshear for generously providing the chick MARCKS cDNA. Thanks also to S. Hicks for assisting with quantitation. This work supported by NIMH grant MH25281-21 to A.R.
References [1] Ali, S.M., Bullock, S. and Rose, S.P.R., Phosphorylation of synaptic proteins in chick forebrain; changes with development and passive avoidance training, J. Neurochem., 50 (1988) 1579-1587. [2] Baizer, L., Alkan, S., Stocker, K. and Ciment, G, Chicken growthassociated protein (GAP)-43: primary structure and regulated expression of mRNA during embryogenesis, Mol. Brain Res., 7 (1990) 61-68. [3] Bliss, T.V.P. and Collingridge, G.L., A synaptic model of memory: long-term potentiation in the hippocampus, Nature, 361 (1993) 31-39. [4] Bullock, S., De Graan, P.N.E., Oestreicher, A.B., Gispen, W.-H. and Rose, S.P.R., Identification of a 52 kDa chick brain membrane protein showing changed phosphorylation after passive avoidance training as B-50 (GAP 43), Neurosci. Res. Comman., 6 (1990) 181-186. [5] Gianotti, C., Nunzi, M.F., Gispen, W.H. and Corradetti, R., Phosphorylation of the presynaptic protein B-50 (GAP-43) is increased during electrically induced long-term potentiation, Neuron, 8 (1992) 843-848. [6] Gispen, W.H. and Routtenberg, A. (Eds.), Protein Kinase C and its Brain Substrates: Role in Neuronal Growth and Plasticity, Elsevier, Amsterdam, 1991. [7] Graft, J.M., Stumpo, D.J. and Blackshear, P.J., Molecular cloning, sequence and expression of a cDNA encoding the chicken myristoylated alanine-rich C kinase substrate (MARCKS), Mol. Endocrinol., 3 (1989) 1903-1906. [8] Hartwig, J.H., Thelen, M., Rosen, A., Janmey, P.A., Nairn, A.C. and Aderem, A., MARCKS is an actin filament crosslinking protein regulated by protein kinase C and calcium-calmodulin, Nature, 356 (1992) 618-622. [9] Horn, G., Neural mechanisms of learning: an analysis of imprinting in the domestic chick, Proc. R. Soc. Lond. B, 213 (1981) 101-137. [10] Horn, G., Memory, Imprinting and the Brain: An Inquiry into Mechanisms, Clarendon Press, Oxford, 1985. [11] Horn, G., Bradley, P. and McCabe, B.J., Changes in the structure of synapses associated with learning, J. Neurosci., 5 (1985)3161-3168. [12] Kuenzel, W.J. and Masson, M., A Stereotaxic Atlas of the Brain of the Chick (Gallus domesticus), Johns Hopkins University Press, Baltimore, MD, 1988. [13] Meberg, P.J. and Routtenberg, A., Selective expression of protein FI mRNA in pyramidal but not granule cells of the hippocampus, Neuroscience, 45 (1991) 721-733. [14] Meberg, P.J., Barnes, C.A., McNaughton, B.L. and Routtenberg, A., Protein kinase C and FI gene expression in hippocampus inversely related to synaptic enhancement (LTE/LTP) lasting three days, Proc. Natl. Acad. Sci. USA, 90 (1993) 12050-12054.
156
P.J. Meberg et ak /Molecular Brain Research 35 (1996) 149-156
[15] Nelson, R.B., Friedman, D.P., O'Neill, J.B., Mishkin, M. and Routtenberg, A., Gradients of protein kinase C substrate phosphorylation in primate visual system peak in visual memory storage areas, Brahl Res., 416 (1987) 387-392. [16] Nelson, R.B., Linden, D.J., Hyman, C., Pfenninger, K.H. and Routtenberg, A., The two major phosphoproteins in growth cones are probably identical to two protein kinase C substrates correlated with persistence of long-term potentiation, d. Neurosci.. 9 (1989) 381389. [17] Ouimet, C.C., Wang, J.K.T., Walaas, S.I., Albert, K.A. and Greengard, P., Localization of the MARCKS (87 kDa) protein, a major specific substrate for protein kinase C, in rat brain..I. Neurosci., 10 (1990) 1683-1698. [18] Rose, S.P.R., How chicks make memories: the cellular cascade from c-fos to dendritic modelling, Trends Neuroxci., 14 (1991) 390-397. [19] Sheu, F.-S., McCabe, B.J., Horn, G. and Routtenberg, A., Learning selectively increases protein kinase C substrate phosphorylation in specific regions of the chick brain. Proc. Natl. Acad. Sci. USA. oh) (1993) 2705-271)9.
[20] Skene, J.H.P., Axonal growth-associated proteins, Annu. Rec. Neurosci., 12 (1989) 127-156. [21] Stumpo, D.J., Graft, J.M., Albert, K.A., Greengard, P. and Blackshear, P.J.. Molecular cloning, characterization and expression of a eDNA encoding the '80- to 87-kDa' myristoylated alanine-rich C kinase substrate: a major cellular substrate for protein kinase C, Proc. Natl. Acad. Sci. USA, 86 (1989) 4012-4016. [22] Thelen, M., Rosen, A., Nairn. A.C. and Aderem, A., Regulation by phosphorylation of reversible association of a myristoylated protein kinase C substrate with the plasma membrane, Nature, 351 (1991) 320-322. [23] Tsai, H.M, Garber, B.B. and Larramendi, L.M.H., 3H-thymidine autoradiographic analysis of telencephalic histogenesis in the chick embryo: I. Neuronal birthdates of telencephalic compartments in situ, J. Comp. Neurol., 198 (1981) 275-292. [24] Wu, W.C.-S., Walaas, S.I., Nairn, A.C. and Grcengard, P., Calcium/phospholipid regulates phosphorylation of a M, '87k' substrate protein in brain synaptosomes, Proc. Natl. Acad. Sci. USA, 79 (1982) 5249-5253.