AUTOSOMAL MUTATIONS IN Drosophila WHICH REDUCE ... of Drosophila melanogaster mutants was created by insertion of a P-element into autosomes at.
Neuroscience and Behavioral Physiology, Vol. 27, No. 3, 1997
A U T O S O M A L M U T A T I O N S I N Drosophila W H I C H R E D U C E OPERANT LEARNING ABILITY
N. G. Kamyshev, E. A. Kamysheva, and G. O. Ivanova
UDC 612.825 +612.821.1
A collection of Drosophila melanogaster mutants was created by insertion of a P-element into autosomes at a rate of one copy per genome. The abilities of 64 homozygous P-insertion mutants to produce a form of associative behavior were determined. Testing was based on an original paradigm of operant learning: interactions between Drosophila individuals when placed in a group situation in which the flies learned to inhibit their own activity to avoid punishment in the form of conflict with other individuals. Four lines were found in which, like the known learning mutants dunce and rutabaga, individuals did not show changes in their initial responses to each other. These lines were also studied using other paradigms of associative learning. Key words: Drosophila melanogaster, P-element, operative learning, dunce mutant, rutabaga mutant.
Studies of the mechanisms of learning have become a highly developed interdisciplinary area of research. The contribution of behavioral genetics consists of identifying and studying genes which determine animals' learning ability. This is needed to determine the extent to which the known molecular mechanisms of learning and memory are complete, and how well newly discovered macromolecules which play.important roles in these mechanisms f'tll the existing gaps. One approach to the detection of such genes consists of searching for Drosophila mutants with defective learning ability; this species is the most studied and suitable model for genetic manipulations. Existing data show significant levels of homology among genes expressed in the human brain and the Drosophila nervous system [3], suggesting that the results of this type of search will make a contribution to the description of the molecular mechanisms of learning in higher animals, including humans. Although a number of learning mutants are already known in Drosophila [5, 7], it would be naive to suppose that the entire spectrum of possible molecular defects has been exhausted. Because of the historical development of methods for artificial mutagenesis and selection of behavioral mutants, the X chromosome has been the most saturated for learning mutations in Drosophila. There is only one example of a directed search for autosomal mutations reducing the ability for associative learning in Drosophila, which resulted in the detection of two such mutations [4, 5]. The search for mutants was conducted in terms of the lack of the ability to undergo classical Pavlovian conditioning, using an electric shock as the unconditioned signal and the smell of alcohol as the conditioned and differentiating stimulus. In the present paper we report the preliminary results of a search for autosomal Drosophila mutants with reduced ability to undergo another type of associative learning - - operant learning.
METHODS A collection of mutant lines was prepared using P-insertion mutagenesis in a system using a single-copy P-element. Except for a number of details, the outline of the first series of mutagenesis experiments was basically similar to that used in [4], which we have used previously [1]. In the first generation, a short nonautonomous (unable to mobilize because of lack Laboratory of Comparative Behavioral Genetics (V. V. Ponomarenko, Director), I. P. Pavlov Institute of Physiology, Russian Academy of Sciences, St. Petersburg. Translated from Fiziologicheskii Zhurnal im. I. M. Sechenova, Vol. 81, No. 8, pp. 69-73, August, 1995. Original article submitted February 15, 1995.
254
0097-0549/97/2703-0254518.00
9
Plenum Publishing Corporation
A
B
C
c
D c
E
F
c
G c
H
I
c
J c
1 "I 4t
'~ 44~ o
211
§247'9.,l Jr,
§ 123
.. I J ~ 123
123
123
J.:,3
~:~3
,~.:3
,23
,~,3
~:,:3
Fig. 1. Movement behavior parameters in flies in a group and in control flies kept singly; normal and learning-mutant flies. The abscissa shows sequential 100-see periods of the observation time; the ordinate shows: A, C, F, 1) the number of physical contacts, initiations of activity, resting and preening periods, per see; B, E, H) total time spent in activity, resting, and preening, see; D, G, J) duration of activity, resting, and preening periods, see. 1) Wild-type Canton-S line (n = 67 and 66 for experiment and control); 2) the dunce mutant (n = 35 and 31 for experiment and control); 3) the rutabaga mutant (n = 22 and 28 for experiment and control); 4) one of the lines bearing a P-element insert in chromosome 2 (line 2v22, n = 53 for experiment and control). Black squares are experimental individuals in the presence of nine ebony mutants; black circles are control individuals kept singly. Vertical lines indicate the standard error of the mean.
of transposase) P-element (a Drosophila mobile genetic element) was combined in males, at a rate of one copy per genome, with another P-element which provided transposase but was itself unable to mobilize for other reasons. As a result, the non-autonomous P-element was able to jump from one site in the genome to another. The insertion of a P-element into one or another gene results in disruption of its normal structure and can produce alterations in normal functions. Since the non-autonomous P-element was located in the X chromosome and its sequence contained the normal allele of the white (white eyes) gene, red-eyed males produced by subsequent mating of these males with females homozygous for the white mutation could only be produced by jumping of the P-element into one of the autosomes (except for a number of significantly rarer cases, resulting from chromosomal nondisjunction, etc.). These exclusive males were used to generate lines of P-insertion mutants. Mutant lines were tested using spontaneous natural mutual training of Drosophila individuals in a group, as described previously [2]. Females of the lines were placed in a transparent Plexiglas chamber (internal diameter 25 mm, height 10 mm) with nine females bearing the ebony (black body) mutation. A DVK-3.2 computer was used to observe the behavior of the test flies for 300 sec. The amounts of time spent in activity, preening (rubbing of the legs against each other or against other parts of the body; body cleaning functions as a repellent for other individuals), and resting were measured in each of three sequential 100-sec periods, along with the frequency of initiation and duration of individual acts of each of these three types of behavior, and the number of physical contacts with other individuals (kicks, leg-leg contacts, leg-body contacts; all further designated simply as "contacts"). Similar observations were made of a control group consisting of individuals from the same line but placed in chambers singly. Statistical comparisons of behavioral parameters in the two experimental groups were made using Student's two-tailed criterion for independent sets. Paired criteria were used for comparing different time periods within a group. All differences between experimental and control groups mentioned in the text, and changes in behavior with time, were statistically significant at a 95 % confidence level.
255
RESULTS AND DISCUSSION As has been described previously for a smaller sample group [2], wild-type flies (of the Canton-S line) just hatched from the cocoon join a group and try from the very beginning to avoid encounters with other individuals (during which the flies kick each other), moving away from them and interrupting excursions which have been started. This is reflected in an increased frequency and reduced duration of excursions in the experimental group as compared with the control group (i.e., flies kept singly). However, since activity results in a greater frequency of encounters with other individuals as compared with resting (at the beginning, the frequency of activity-to-contact transitions is higher, and the frequency of rest-to-contact transitions is lower than the probability of random transitions [2]), trial and error on the part of the flies has the effect that with time they learn not to react to the vicinity of other individuals by making excursions, or to actual contacts with other individuals, and increase the duration of their resting periods. Flies in a group show a significantly lower level of activity than control flies kept singly. The frequency of contacts decreases and the frequency of activity-to-contact and rest-to-contact transitions becomes closer to the level expected from random chance. Thus, operant learning causes the flies to change their initial tactics of escaping from other individuals to a tactic of nhiding." Screening of a collection of P-insertion lines for mutants unable to undergo operant learning was carried out using wild-type controls, along with another type of control in which flies were compared with known mutants with defects in associative learning. These were the dunce mutant (dnc; the mutation affects the phosphodiesterase II structural gene [5]) and the rutabaga mutant (rut; this mutation affects the catalytic subunit of adenylate cyclase [5]). The dnc mutants showed absolutely no reduction in activity or contact frequency when flies were placed in a group as compared to flies kept singly. While in the latter situation the resting time gradually increased, flies in a group had lower frequencies of initiation of rest periods and had shorter durations of rest periods than flies kept singly. Thus, dnc flies showed no signs of adaptive changes in movement behavior which would reduce the number of contacts. The absence of a pronounced primary response to the presence of a group (an increase in the excursion frequency, with reductions in the duration of excursions) would appear to be associated with a relatively low initial excursion duration in individuals kept singly. This response was, however, seen in the third period of the observation time; the reduction in activity was slower in flies kept in a group. In rut mutants, with a higher initial excursion duration, there was a clear primary response to the presence of a group. The number of contacts did not change with time. Although a reduction in activity did occur in the third period of the observation time as compared with flies kept singly, this resulted from an increase in the preening response, which is an innate repellent behavior, rather than from an increase in the duration of resting, as occurred in wild-type flies. Studies of learning in the known mutants showed that the inability to undergo operant conditioning could be detected using a complex of parameters describing changes in the movement behavior of flies in a group. The most informative parameters were the absence of the normal reduction in the number of contacts, the absence of an increase in the total resting time, and the absence of an increase in the durations of individual resting periods. In the first P-insertion mutagenesis experiment, a total of 64 lines were obtained in which the P-insertion did not produce any recessive lethal effect, along with 29 P-insertions in which there was a recessive lethal effect. At the time of writing, tests had been conducted on lines in which insertions were in the homozygous state. Ten individuals from each line were tested, along with 10 controls. When individuals of a given line showed deviations from normal behavior (as compared with the Canton-S wild-type line), testing was repeated. Four lines were suspected to have learning defects. The results of studies of one of these (designated line 2v22), with insertion of a P-element in chromosome 2, are shown in Fig. 1. Some of the individuals (n = 23 for the experimental and control groups) were observed for 600 sec (data not shown). Line 2v22 showed a clear primary response to other individuals, thus indicating a normal ability to perceive their presence. Although the time spent in activity by control (placed singly) individuals was comparable to that in the Canton-S line, none of the six 100-sec periods showed significant differences in this parameter between the experimental and control groups. There were also no increases in the duration of individual periods of resting in flies kept in a group as compared with controls. The frequency of contacts decreased in the second period as compared with the first, but remained at a fairly high level. This reduction would appear to be associated with a reduction in the level of activity at this time in both singly-kept individuals and in flies in the group. Thus, mutant 2v22 demonstrated behavior similar to that of the dnc and rut mutants, in that they failed to show adaptive changes leading to a change in tactics from escape from other individuals to a tactic of "hiding." Similar patterns were seen in the dynamics of fly behavior when in a group as compared to the behavior of control flies kept singly in three further lines. These are currently being studied further, to assess the extent of common features in 256
the mutational defects in terms of other paradigms of operant and classical conditioning, and to exclude defects resulting from any inability of the flies to perceive physical contacts with other flies as a punishment, as well as other possible defects in sensory and motor systems which could lead to loss of learning ability in this paradigm. This work was supported by the I. P. Pavlov Institute of Physiology, Russian Academy of Sciences, and by the State Scientific-Technical Program "Priorities in Genetics."
REFERENCES I.
2. 3. 4. 5. 6. 7.
K. G. Iliadi, D. S. Veliksar, and N. G. Kamyshev, "P-insertion mutagenesis in Drosophila using a non-autonomous P-element at one copy per genome," Bull. Acad. Sti. Moldova, Sti. Biol. Chim., No. 5, 29-32 (1993). N. G. Kamyshev, E. A. Kamysheva, G. P. Smirnova, and I. V. Parafenyuk, "Mutual training of Drosophila individuals in a group situation by the method of trial and error," Zh. Obshch. Biol., 55, No. 6, 737-748 (1994). S. Benzer and C. A. Miller, "Monoclonal antibody cross-reaction between Drosophila and human brain," Proc. Natl. Acad. Sci. USA, 80, No. 24, 7641-7645 (1983). S. Boynton and T. Tully, "Latheo, a new gene involved in associative learning and memory in Drosophila melanogaster, identified from P-element mutagenesis," Genetics, 131, No. 3, 655-672 (1992). Y. Dudai, "Neurogenetic dissection of learning and short-term memory in Drosophila," Ann. Rev. Neurosci., 11, 537-563 (1988). J.-M. Dura, T. Preat, and T. Tully, "Identification of linotte, a new gene affecting learning and memory in Drosophila melanogaster," J. Neurogenetics, 9, No. 1, 1-14 (1993). M. Heisenberg, "Genetic approach to learning and memory (mnemogenetics) in Drosophila melanogaster," Fortschr. Zoologic, 37, No. 1, 3-46 (1989).
257
