| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Marine Biological Laboratory, Woods Hole, MA
2 University of New Hampshire, Durham, NH
3 Carnegie Mellon University, Pittsburgh, PA
4 Blanchette Rockefeller Neuroscience Institute, Johns Hopkins University, Rockville, MD
* Corresponding author: akuziria{at}mbl.edu
In the nudibranch mollusc Hermissenda crassicornis, the intensity of immunostaining for the Ca2+/GTP-binding protein calexcitin (1) correlates positively with the degree of learning obtained and the level of memory expressed after Pavlovian conditioning (2,3). When inhibitors of transcription (actinomycin-D; Act-D) and translation (anisomycin; ANI) are applied after training, they affect the animal’s ability to recall the learned behavior. Epstein et al. (4) have recently described the time windows for these effects. They noted that there are two phases of sensitivity to the inhibitors: an early phase, immediately after training (0–13 min), and a later phase (70–160 min for Act-D, and 70–220 min for ANI). Subsequently, Epstein et al. (5) reported that long-term memory also has two distinct biological configurations: long-term memory (LTM), which lasts about 24 h, and consolidated long-term memory (CLTM), which persists up to 6 days. The amount of memory retained by conditioned animals appears to depend, in part, on how much consolidation has been completed; and the transition from LTM to CLTM is sensitive to a cell adhesion molecule (CAM) inhibitor, the tripeptide arginyl-glycyl-aspartate (RGD). One working hypothesis predicts that increased calexcitin levels are needed to establish LTM, and that RGD-sensitive CAMs are subsequently involved in the transition from LTM to CLTM. What was unknown was the possible relationship between calexcitin levels and the effects of RGD during this critical transition period. Thus, as part of a continuing study to describe and define the memory stages demonstrated in Hermissenda (6,4,5), we undertook an immunocytochemical study to investigate possible correlations between calexcitin levels and the observed effects of RGD during the transition from LTM to CLTM.
Hermissenda were purchased from Sea Life Supply (Sand City, California), and the tripeptide CAM inhibitor, arginyl-glycyl-aspartate (RGD), was obtained from Calbiochem (San Diego, CA). Animals were acclimated to laboratory conditions for a minimum of 3 days. The training procedure used was adapted from a Pavlovian conditioning regime developed by Crow and Alkon (7), elaborated later by Lederhendler, Gart, and Alkon (8), and modified by Kuzirian et al. (9) and Epstein (10). Before training or testing, animals were dark-adapted for 10 min in a transparent acrylic plastic tray with 16 lanes (15 cm long and 0.9 cm by 0.9 cm in cross section) in an 11 °C incubator. In paired training, animals were exposed to a bright, white light (650–700 lux) for 6 s (the conditioned stimulus, CS) coincident with, after a 2-s delay, 4 s of vigorous agitation (the unconditioned stimulus, US). Animals respond to the agitation by contracting lengthwise (unconditioned response, UR). The two stimuli are designated a paired training event (TE), which was repeated at 1-min intervals for nine repetitions. Recall of training was tested by four presentations of the CS alone at 1-min intervals. Behavioral recall was assessed by calculating the percentage by which the animal’s length changed between light-on and light-off. Untrained (naive) animals show no recall and typically lengthen as a normal response to light, as do animals given light and agitation in an unpaired or random fashion (10).
Experimental conditions shown by Epstein et al. (5) to demonstrate the RGD-sensitive transition between LTM and CLTM were repeated to test for possible correlations between different times of RGD application after training and changes in calexcitin intensities. RGD was diluted to an effective working dosage of 10 µg/ml (29 µM) with Tris-buffered (20 mM) natural seawater (NSW-Tris) (pH 8.0). All solutions were administered post-training, at designated times, by bath application using a training tray with a modified injection-port cover (4). After training, the RGD-exposed animals were tested for retention of recall at 4 and 24 h, then rapidly decapitated. The central nervous systems (CNS) were immediately fixed in 4% paraformaldehyde/NSW-Tris to preserve, as accurately as possible, the expressed calexcitin levels.
To discern the temporal effects of training alone on the immunostaining intensities of calexcitin, and to test for possible effects from testing with the CS alone, animals were similarly conditioned with 9 TEs in natural seawater, but they were not exposed to RGD or tested for recall. They were decapitated and fixed at time points similar to those for the RGD-exposed animals. Naive, untrained animals were used as overall experimental controls and were tested after 24 h.
All fixed CNSs were then processed for embedding in polyethylene glycol-400-distearate and were sectioned (6 µm). Calexcitin was immunolabeled with a primary polyclonal antibody (25U2; cloned from squid optic lobe and raised in rabbit; 1:1000 dil) (2,3); subsequently, color was developed with a biotinylated secondary antibody, and avidin-bound microperoxidase reacted with the chromogen 3-amino-9-ethylcarbazole (AEC). Gray-scale intensity was measured (using NIH Image software) from digital photomicrographs of serial sections of the B-photoreceptors in the eyes of each animal. Intensity differences between conditions were statistically analyzed using ANOVA and Student’s t tests.
The data indicate that the levels of calexcitin remained steady from 40 to 200 min post-training, whether the animals were exposed to RGD or not (Fig. 1). Levels then fell to a baseline about 240 min after training. ANOVA and pairwise Student’s t tests among and between intensities measured during the first phase (40–200 min post-training) indicated no significant differences between all intensity values (F = 0.67, ts < 2.0; P = >0.7; n = 4–10 eyes measured), whether animals were treated with RGD and tested at 4 and 24 h, or simply trained and fixed at similar time points. The same was true for the second phase (240 min to 24 h post-training). Again, the intensities were statistically identical, whether animals were tested or not, at 240 min or at 24 h (F = 0.07, ts < 2.0; P = >0.9; n = 4–10 eyes measured). However, the overall ANOVA between all conditions did show significant differences between the first and second phases (F = 7.49; P = <0.001).
|
The calexcitin immunostaining intensities observed during this study were similar to levels reported previously for LTM (3). The initial study by Kuzirian et al. (3) showed that the rise in calexcitin intensity occurred by 90 min post-training, the earliest time point sampled. However, this study indicates that the intensity rise begins earlier, by 40 min or even sooner. The results of this study further demonstrated that the treatment of Hermissenda with the CAM inhibitor RGD is independent of the expressed levels of calexcitin. Therefore, the behavioral effects related to RGD exposure reported by Epstein et al. (5) were not mediated directly through changes in calexcitin levels. The similarity of intensity levels in tested and nontested animals also suggests that calexcitin levels may be insensitive to the immediate perturbation of recall caused solely by testing; a phenomenon known as reconsolidation (11) (also described as extinction). Memory generated by an associative conditioning regime can be weakened by testing with the conditioned stimulus (CS) alone and must be reconsolidated to remain fixed. The data indicated quite clearly that there were no differences in calexcitin intensity related to testing alone.
Previous reports (3,12) show that calexcitin is involved in establishing LTM. This was accomplished by exposing Hermissenda to a sensory block (an additional vestibular input) at time points before LTM is established (at 60 min post-training) (5) and afterward. The sensory block suppressed this rise in calexcitin levels until LTM was fixed. However, the current data indicate that calexcitin levels remain high through to the period coinciding with the establishment of CLTM (4,5), and thus calexcitin may also be involved in some way with the consolidation of LTM to CLTM. Calexcitin was also elevated during the time period designated by Epstein et al. (4) as potentially being an intermediate-term memory phase, a phenomenon known to occur in Hermissenda and Aplysia, among other animals (13,14). Since calexcitin, once activated by PKC, is known to regulate the release of internal calcium stores by binding to ryanodine receptors on the ER, and is responsible for early gene activation, all of which are involved in memory acquisition, storage, and recall (15), it would be logical to broaden this study by measuring the effects of transcription and translation inhibitors on the immuno-expression levels of calexcitin under experimental conditions similar to those just described. Indeed, such studies are underway.
Drs. Daniel Alkon and Thomas Nelson provided Hermissenda and calexcitin antibodies for this study. MEM and CEO acknowledge the assistance of Beth Linnon, program coordinator, and the student internship program sponsored by the Marine Resources Center, Marine Biological Laboratory.
Literature Cited
This article has been cited by other articles:
|
|
 |
|
  A. M. Kuzirian, H. T. Epstein, C. J. Gagliardi, T. J. Nelson, M. Sakakibara, C. Taylor, A. B. Scioletti, and D. L. Alkon Bryostatin Enhancement of Memory in Hermissenda Biol. Bull., June 1, 2006; 210(3): 201 - 214. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
