Studies on the circadian rhyt hm of eclosion in Musca domestica

Transcription

1 No. 10] Proc. Japan Acad., 77, Ser. B (2001) 191 Studies on the circadian rhyt hm of eclosion in Musca domestica By Makoto AIZAWA*)and Hiroshi YosHINO**) (Communicated by Koichi HIWATasxl, M.J.A., Dec. 12, 2001) Abstract: The circadian of eclosion in the house fly, Musca domestica, was measured using new apparatus, designed for large populations. The of eclosion was found to depend upon separate biological clocks in the pupal and larval stages. A comparison of the wild type and triple mutant wbp (white eyes, brown body and pointed wings) showed the mutant has a short periodicity of eclosion. Cross breeding analysis suggested a single gene controls the short periodicity of the eclosion in the triple mutant (wbp). Key words: Circadian ; eclosion; mutant; house fly. Introduction. The circadian of various activities, including eclosion, has been extensively studied in Drosophila, and genes controlling the biological clock in this genus have been reported.' 8) In Lucilia (Diptera: Calliphoridae), an aric mutant of eclosion and activity was reported.9~ However, it is still unclear whether genetic control of eclosion is common to other insects. In the house fly, Musca domestica, the circadian of locomotory activity and adult emergence patterns has been described,lo)" but the genetic control of eclosion is unknown. In this study, we measured the circadian of eclosion using newly devised automatic recording apparatus is convenient for measuring the eclosion of individuals in a large population. The eclosion s of two strains of M. domestica, a wild type and triple mutant, were compared by cross breeding analyses. Our results suggest a major gene controls eclosion in the house fly and is responsible for the difference in eclosion s of the strains. The present study may improve our general understanding of the genetic control mechanisms of circadian s in insects. Materials and methods. Strains. The wild type strain, P-9, and a triple mutant strain, wbp, were used in this study. The latter strain is a mutant of three genes on chromosome III, coding for white eyes, brown *> Kogota Agriculture and Forestry High School, 30, Ushikai, Aza Isedoura, Kogota-cho, Toda-gun, Miyagi , Japan. **> Shiogama Women's High School, 7-1, Isumigaoka, Shiogama, Miyagi , Japan. t) Correspondence to: M. Aizawa. body and pointed wings. Both strains were obtained from Dr. T. Hiroyoshi of Osaka University. l Methods of breeding. Adult flies were bred in a wooden box, 20 x 20 x 20 cm3, with windows for observation and feeding. The flies were fed on dried milk in a Petri dish. Water was supplied from an Ehrenmeyer flask with a cotton plug covered with cheesecloth, the opposite end of which was dipped in water. An egg-laying plate was made with a 6 cm-diameter Petri dish contained a mixture of fish meal, wheatgluten bread and water. The plate was placed in the breeding box for three hours from 9 a.m. After oviposition, the egg-laying plate was put in a 1 liter beaker and moist wheat-gluten bread was added into the beaker daily for 5-6 days. The eggs usually hatched in the same day of oviposition. Two days before pupation, the plate of larvae was transferred into a large glass bottle, which was covered on top with a piece of linen. The number of larvae was adjusted to about 2,000 individuals. Two days before eclosion, the eclosion-measurement apparatus was placed on the top of the bottle. Measurement of eclosion. Eclosion was measured by an infrared photodiode (950 nm) and its sensor (Fig. 1). As house flies habitually migrate upward immediately after eclosion, the number of eclosed flies was determined, by the infrared sensors, as the number of interruptions of the infrared beam. One count was defined as a set of interruptions detected by both the lower and higher sensors. The number of interruptions (eclosions) was recorded hourly in a personal computer. The apparatus was designed to accommodate four routes of flying within each recorder.

2 192 M. AIZAWA and H. YOSHINO [Vol. 77(B), The eclosion was recorded from a population of mixed ages, reared in a breeding bottle. Eclosion was observed every day during several days in a single breeding bottle, and the distribution pattern of eclosion was recorded as a ic pattern. Eggs of different females were collected for three hours from 9 a.m. for 4 days, and cultured at 23 C under 12 hours of light and dark (LD:12:12). From the seventh day after the first oviposition, to the end of the experiment, larvae were grown under constant darkness (DD). Two days before eclosion, when the flies were pupae, the eclosion measurement apparatus was set on the breeding bottle. Eclosion occurred for 4-5 days. The times of peak eclosion were recorded and the mean distance between the neighboring peaks was used to identify the eclosion. To check the temperature compensation of the eclosion, breeding temperatures were set from 19 C to 30 C in increments of 1 C. We found the development of house flies is temperature dependent. No eclosion occurred below 18 C, the period from oviposition to eclosion was reduced from 18 to 11 days between 19 C and 30 C, and no reduction in eclosion time was observed over 30 C. The eclosion of the wild type (P-9) was as anticipated, as eclosion always began before the lights were turned on. This suggests time of eclosion reflects the biological clocks of both larvae and pupae. For each breeding temperature, the eclosion time of individuals of the same age was measured. These experiments were done under LD12:12. However, before the start of pupation, larvae migrated into the wheat-gluten bread, where it was dark. Therefore, house flies seemed to stay under DD conditions from pupation to eclosion. For the genetic analysis of the FI population, sex identification of individuals of the wild type (P-9) were made immediately after eclosion and mixed with the complementary sex of the mutant (wbp) a few days later. To obtain the F2 generation, siblings of the F1 population were crossed. For the backcross analysis, F1 males were crossed with wild-type (P-9) females, because the mutant (wbp) was a little weak and the ordinary backcross to the recessive parent was difficult. Statistical analyses. The time of the peak of eclosion was determined by least square spectrum analysis. Eclosion s were compared among the wild type, triple mutant and their progeny with t-tests. Regression lines of the eclosion time were determined by least square analysis. The distribution patterns of eclosion and the segregation of phenotypes in the F2 population were analysed using normal probability plots. Results. Eclosion under LD12:1 2. As shown in Fig. 2a, eclosion started on 13 days after collecting the eggs and continued for 7 days. During these 7 days, five clear peaks of eclosion were observed. The mean distance between the peaks was calculated as 26.3 h by least square spectrum analysis. In the same way, the eclosion of 21 populations (n = 57,175 individuals in total) was examined and the mean distances between neighboring peaks were counted. The mean eclosion was 25.7±0.8 h (Mean ± SEM; n = 2,157;1,175 individuals in total). Eclosion under constant light. Since one of the characteristics of circadian is its absence under constant light (LL), the eclosion of wild-type strain P-9 was tested under LL. As shown in Fig. 2b, eclosion occurred continuously over 2 to 3 days and no clear peak of eclosion was observed (n = 1,702 individuals). This result, together with those in the previous section, indicates eclosion is controlled by circadian in the house fly, and demonstrates the apparatus we devised produces almost no erroneous data for determining circadian. Eclosion of the triple mutant (wbp) under LD12:1 2. The eclosion s of 16 populations (n = 15,124 individuals in total) of various ages of the triple mutant (wbp) were measured using the same

3 No. 10] Circadian of eclosion 193 Fig. 2. a: The number of emerged flies in wild type populations of heterogeneous ages. A typical pattern of eclosion is shown by single curve, calculated using least square spectrum analysis. Numbers in parentheses indicate days after egg collection (n = 13,296 individuals). b: The number of emerged flies in a population, with individuals of the same age, under constant light (LL). Numbers in parentheses, same as Fig. 2a (n = 1,702 individuals). analysis as for the wild type (P-9). The mean period of eclosion s, estimated by least square spectrum analysis, was 22.7± 0.6 h (n = 16 ; 15,124 individuals in total). Effect of breeding temperatures upon the eclosion. As described in the Materials and Methods, the effect of breeding temperature, from 19 C to 30 C in increments of 1 C, was examined on the wildtype (P-9), the triple mutant (wbp) and their F1 generation under LD 12:12. Individuals from all populations were the same age in this experiment. As shown in Fig. 3, the regression lines of the times of eclosion against the days to eclosion (different in different breeding temperatures) were almost linear. The mean eclosion of p-9 was 26.7±2.4 h (n =125 experiments, 301,136 individuals in total), wbp was 23.9± 1.9 h (n =106;188,560 individuals) and F1 was 24.5± 1.5 h (n = 50 ; 87,825 individuals). These data indicate eclosion is independent of breeding temperature for the strains examined. Genetic difference of eclosion between wild-type (P-9) and the triple mutant (wbp). When F1 males were backcrossed to the P-9 female parent, the progeny showed two peaks of eclosion time (Fig. 4a). This result was almost the same as the eclosion times estimated from F1 and P-9 in Fig. 3. The eclosion time of F2, obtained by crossing siblings of F1, is shown in Fig. 4b. The distribution of eclosion s in the F2 population was examined for normality by normal probability plot analysis. The linearity of the plot profile did not fit the pattern expected from a single normal distribution, suggesting the distribution of eclosion s in the F2 generation was composed of multiple populations, rather than a single population. These results suggest the difference of eclosion between the wild type and triple mutant is controlled by a single locus.

4 194 M. Alz.awa a,nd H. YoSHiNo [Vol. 77(B), Fig. 3. Distribution of eclosion s in populations with individuals of the same age for the wild type (P-9), triple mutant (wbp) and Fl populations, bred at different temperatures. Closed circles, wild type (P-9, n populations; 301,136 individuals); red circles, triple mutant (wbp, n = 106; 188,560 individuals) and blue triangles, Fl (n = 50; 87,825 individuals). Open bars in the bottom, lightness; closed bars, darkness. Discussion. A comparison was made of eclosion s between populations of both different ages the same age. When eggs laid for 4 days were mixed and the resultant larvae were bred, three to five peaks of eclosion were observed. The mean distances of the peaks were 25.7± 0.8 h (SEM) in the wild type P-9 and 22.7± 0.6 h in the triple mutant wbp. A t-test indicated a significant difference between the wild type and triple mutant (P < 0.01). Conversely, when populations of the same age were bred at different temperatures, the mean length of eclosion s estimated from Fig. 3, were 26.7±2.4 h in P-9, 23.9± 1.9 h in wbp and 24.5± 1.5 h in F, (Table I). Statistical analyses among those strains indicated a significant difference between the wild type and triple mutant (P<0.01) and between the triple mutant and F1 Table I. Circadian s of house Eclosion fly eclosion Strain Heterogeneous age Homogeneous population P ±0.8*lh 26.7±2.4 F, 22.7±0.6 *) Mean±SE bred using populations in LD12:12 until larvae pupated, where it was presumably (P<0.05). As seen in Table I, the period of eclosion in populations of different ages and of the same age were significantly different in the wild type (P < 0.02) and in the triple mutant (P < 0.01). In the experiments using populations of different ages, larvae were bred in LD12:12, and transferred into constant darkness 7 days before pupation. In the pupal stage they is affected stage found same into If we assume biological of the mechanisms age, ages and same are their clock larvae when the food eclosion in the larval by in the age, working it appears in the two stages. of the is h However, to migrate of different clock developmental of the pupation. dark. the in populations different One 23.9±1.5 tended by in populations h. experiments h 24.5 ± 1.9 h wbp were age population the important characteristics it is independent period of eclosion of circadian of temperature. in a strain We did

5 No. 10] Circadian of eclosion 195 Fig. 4. a: The number of emerged flies, in the progeny of a backcross between F, and wild type, in populations of the same age (n = 510 individuals). Arrows 1 and 2 indicate the estimated eclosion time of F1 and P-9, respectively, b: The number of emerged flies in F2 populations of the same age (n = 9,385 individuals). Arrows 1, 2 and 3 indicate the estimated eclosion time of wbp, F1 and P-9, respectively. not change as the breeding temperature increased from 19 C to 30 C. This indicates the eclosion in the house fly is typical circadian. The triple mutant (wbp) used in this study showed a shorter eclosion than the wild type. The F1 generation, produced from the cross between the wild type and the triple mutant, exhibited an intermediate distribution of eclosion s (Table I). This suggests the shorter eclosion of the mutant is an incomplete recessive against the wild type. The backcross of the F1 to wild-type showed a nearly 1:1 segregation of the wild-type and mutant, and the F2 showed an intermediate distribution of eclosion, with a single peak between them. Although it is difficult to determine the segregation ratio of phenotypes in the F2 by the distribution pattern of eclosion, normal probability plot analysis indicates the F2 population is composed of individuals with different lengths of eclosion s. Neumann13~ used a similar analysis in his study of eclosion in the sea midge, Clunio, and concluded the was controlled by multiple genes. Conversely, a single gene was found to control the eclosion circadian in Drosophila rnelanogaster,5~ but study did not report whether the eclosion in the F1 generation was intermediate. Molecular studies on circadian s, especially the clock mutant genes pers and their expressions, have advanced rapidly in recent times.l4) 17) The functional homology of per genes between Drosophila and the house fly was demonstrated by introducing the

6 196 M. AIZAwA and H. Yo SHINO [Vol. 77(B), Musca per gene into aric Drosophila per mutants.18) It will be interesting to know the genetic relationship between pers and the genes controlling eclosion s. Acknowledgments. We thank Dr. Koichi Hiwatashi, M. J. A., and Dr. Nobuyuki Haga for their critical discussions and comments. 1) 2) 3) 4) 5) 6) References Allada, R., White, N. E., So, W. V., Hall, J. C., and Rosbash, M. (1998) Cell 93, Bargiello, T. A., and Young, M. W. (1984) Proc. Natl. Acad. Sci. U.S.A. 81, Darlington, T. K., Wager-Smith, K., Ceriani, M. F., Staknis, D., Gekakis, N., Streeves, T. D. L., Welts, C. J., Takahashi, J. S., and Kay, S. A. (1998) Science 280, Ikeda, H., and Saito, M. (1983) Zoological Magazine 92, Konopka, R. J., and Benzer, S. (1971) Proc. Natl. Acad. Sci. U.S.A. 68, Reddy, P., Zehring, W. A., Wheeler, D. A., Pirrotta, V., Hadfield, C., Hall, J. C., and Rosbash, M. (1984) Cell 38, ) 8) 9) 10) 11) 12) 13) 14) 15) 16) 17) 18) Roberts, S. K. de F. (1956) Science 124, 172. Sehgal, A., Price, J. L., Man, B., and Young, M. W. (1994) Science 263, Smith, P. H. (1987) Physiol. Entomol. 12, Engelmann, W., Hellrung, W., and Johnsson, A. (1996) Bioelectromagnetics 17, Kim, Y., and Krafsur, E. S. (1993) J. Medical Entomol. 30, Hiroyoshi, H. (1977) Japan J. Genetics 52, Neumann, D. (1967) Helgolaander Wiss. Meer-Unters 15, Gekakis, N., Saez, L., Delahaye-Brown, A. M., Myers, M. P., Sehgal, A., Young, M. W., and Weitz, C. J. (1995) Science 270, Lee, C., Bae, K., and Edery, I. (1998) Nature 4, Rutila, J. E., Sun, V., Le, M., So, W. V., Rosbash, M., and Hall, J. C. (1998) Cell 93, Young, M. W. (1998) Annu. Rev. Biochem. 67, Piccin, A., Couchman, M., Clayton, J. D., Chalmers, D., Costa, R., and Kyriacou, C. P. (2000) Genetics 154,