Memory reconsolidation in Hermissenda.
The consolidation of this conditioning in Hermissenda was studied by Epstein et al. (3). Consolidation of memories has been known experimentally at least since Muller and Pilzecker (4) demonstrated it in humans in 1900. It is defined as the process by which memory reaches a state in which interventions (or interferences) no longer inhibit recall of what was presented to be remembered. This does not mean, as initially inferred (5, 6), that the installed memory cannot be altered; it is just that it is stable with respect to the said interferences. Examples of such interventions include the four treatments (chemicals, sensory input) used in this study.
Reconsolidation (a much newer finding) is defined as the ending of the interference-insensitive state brought about by the recall of what was memorized. That is, after a memory has been consolidated, if it is recalled it becomes sensitive to agents such as inhibitors of mRNA synthesis and protein synthesis that did not affect it just before the recall. Since the consolidated memory is regenerated after recall, it will become important to determine whether the fact of reconsolidation has more to do with temporary memory degradation or with a weakening of the retrieval process.
Consolidation and reconsolidation are being studied in many organisms from molluscs to humans (5, 6, 7, 8). The existence of reconsolidation may allow more detailed study of many proposed aspects of consolidation. We therefore undertook a preliminary study of reconsolidation in Hermissenda. This nudibranch has proven to be a high-connectivity model for the more complex vertebrate nervous system, especially in the area of learning and memory (3, 9. 10).
Our training procedure with Hermissenda was a Pavlovian conditioning regime (2). Animals were placed in transparent acrylic plastic trays (with 16 fluid-filled lanes about 0.9 cm wide and deep and 15 cm long) in a closed 11[degrees]C incubator for 10 min of dark adaptation before training or testing. Training consists of exposing the animals to a bright white light (650-700 lux) for 6 s (the conditioned stimulus, CS), paired, after a 2-s delay, with a 4-s vigorous orbital shaking of the tray containing the animals (the unconditioned stimulus, US). The animals respond to the shaking by contracting lengthwise. This combination of the two stimuli is called a paired training event (TE) and is repeated at 1-min intervals.
Testing of recall was done by four presentations of the CS at 1-min intervals, and was recorded using a videocassette recorder whose input was registered by a camera placed below the transparent trays. Animal length was measured on the video monitor. The measure used was the percentage by which the animal's length changed between light-on and light-off times. From those percentage changes we compute a mean and the standard error of the mean.
We used four interventions after training to probe the acquisition and consolidation of memory in Hermissenda. The first intervention is the sensory input used by Epstein et al. (3). The other three are drugs dissolved in seawater, pH 8, and administered by replacement of the seawater bathing the animal.
1. Sensory input: the tray containing the animals is quickly rotated by hand 180[degrees] around its long axis and, after 5 s, quickly rotated back to its original position. This produces a vestibular input (in addition to the TE) and is termed a sensory block (SB).
2. Anisomycin (ANI) interferes with protein synthesis by inhibiting translation. It was administered at 1 [micro]g/ml. The translation-inhibition was not checked.
3. Actinomycin-D (Act-D) interferes with mRNA synthesis by inhibiting transcription. It was administered at 0.1 [micro]g/ml. The transcription inhibition was not checked.
4. RGD (arginyl-glycyl-aspartate) inhibits bond formation between cell adhesion molecules (CAMs). It was administered at 10 [micro]g/ml. The bond formation inhibition was not checked.
In the first experiment, specimens of Hermissenda were given nine TEs and tested at 4 h; this is the time at which our previous studies have shown recall to become insensitive to all four inhibitors we are using, thereby showing that the stored memory is consolidated long-term memory (CLTM). After testing to verify arrival at CLTM, ANI was added either immediately, 10 min later, or 30 min later. The data in Figure 1A show that adding anisomycin immediately or 10 min after testing for recall resulted in the loss of recall; adding the anisomycin after 30 min showed that recall has been re-established.
We next studied whether the other three inhibitors had similar effects (Fig. 1B). All three inhibitors wiped out recall if added immediately after testing at 4 h, whereas waiting for 30 min before using the inhibitors revealed that recall had been re-established, as had been found for ANI (Fig. 1A).
[FIGURE 1 OMITTED]
Thus, these results demonstrate the existence of reconsolidation in Hermissenda. Reconsolidation was triggered by testing for recall after 4 h and then probing with each of the four inhibitors. The reconsolidation was found to be reached by about 30 min after recall; thus it is a much more rapid process than the initial acquisition and consolidation phase, which has been shown to take nearly 4 h.
The sensory block (SB) was previously shown by Epstein et al. (3) to inhibit short-term memory (STM) as well as long-term (LTM) and consolidated long-term memory (CLTM). The fact that it works on both consolidation and reconsolidation raises again the question of what steps it is affecting. Molecular and physiological studies will be needed to gel at this question.
Finally, both consolidation and reconsolidation deserve study in that there may be ramifications for learning and schooling. If the first consolidation study by Muller and Pilzecker (3) is correct, teaching additional novel material within 6 to 10 min after having taught a primary point that is meant to be remembered could well weaken retention of that point. Similarly, if recalling something weakens the memory, then new information might prevent reconsolidation and thus should be avoided. These aspects need serious study by researchers in the field of education.
(1.) Abel, T., and K. M. Lattal. 2001. Curr. Opin. Neurobiol. 11: 180-187.
(2.) Lederhendler, I. I., S. Gart, and D. L. Alkon. 1986. J. Neurosci. 6: 1325-1331.
(3.) Epstein, D. A., H. T. Epstein, F. M. Child, A. M. Kuzirian, and D. L. Alkon. 2000. Biol. Bull. 199: 182-183.
(4.) Muller, G. E., and A. Pilzecker. 1900. Z. Psychol. Erganzungsband 1: 1-300.
(5.) Riccio, D. C., E. W. Moody, and P. M. Millin. 2002. Integr. Physiol. Behav. Sci. 37: 245-253.
(6.) Sara, S. J. 2000. Learn. Mem. 7: 73-84.
(7.) Nader, K. 2003. Trends Neurosci. 26: 65-72.
(8.) Pedreira, M. E., and H. M. Maldonado. 2003. Neuron 38: 863-869.
(9.) Kuzirian, A. M., H. T. Epstein, D. Buck, F. M. Child, T. Nelson, and D. L. Alkon. 2001. J. Neurocytol. 30: 993-1008.
(10.) Epstein, H. T., F. M. Child, A. M. Kuzirian, and D. L. Alkon. 2003. Neurobiol. Learn. Mem. 79: 127-131.
F. M. Child (1,*), H. T. Epstein (1), A. M. Kuzirian (1), and D. L. Alkon (2)
(1) Marine Biological Laboratory, Woods Hole, MA 02543
(2) Blanchette Rockefeller Neuroscience Institute, Rockville, MD 20850
|Printer friendly Cite/link Email Feedback|
|Author:||Child, F.M.; Epstein, H.T.; Kuzirian, A.M.; Alkon, D.L.|
|Publication:||The Biological Bulletin|
|Date:||Oct 1, 2003|
|Previous Article:||Neural recordings from the lateral line in free-swimming toadfish, Opsanus tau.|
|Next Article:||Training alone, not the tripeptide RGD, modulates calexcitin in Hermissenda.|