Limits...
Evidence that natural selection maintains genetic variation for sleep in Drosophila melanogaster.

Svetec N, Zhao L, Saelao P, Chiu JC, Begun DJ - BMC Evol. Biol. (2015)

Bottom Line: We also found evidence for geographic variation for sunrise anticipation.Furthermore, the present study suggests that natural selection plays a major role in generating transcriptomic variation associated with circadian behaviors.Finally, we identified genomic variants plausibly causally associated with the observed behavioral and transcriptomic variation.

View Article: PubMed Central - PubMed

Affiliation: Department of Evolution and Ecology, University of California, 3352 Storer Hall, One Shields Ave., Davis, CA, 95618, USA. nhsvetec@ucdavis.edu.

ABSTRACT

Background: Drosophila melanogaster often shows correlations between latitude and phenotypic or genetic variation on different continents, which suggests local adaptation with respect to a heterogeneous environment. Previous phenotypic analyses of latitudinal clines have investigated mainly physiological, morphological, or life-history traits. Here, we studied latitudinal variation in sleep in D. melanogaster populations from North and Central America. In parallel, we used RNA-seq to identify interpopulation gene expression differences.

Results: We found that in D. melanogaster the average nighttime sleep bout duration exhibits a latitudinal cline such that sleep bouts of equatorial populations are roughly twice as long as those of temperate populations. Interestingly, this pattern of latitudinal variation is not observed for any daytime measure of activity or sleep. We also found evidence for geographic variation for sunrise anticipation. Our RNA-seq experiment carried out on heads from a low and high latitude population identified a large number of gene expression differences, most of which were time dependent. Differentially expressed genes were enriched in circadian regulated genes and enriched in genes potentially under spatially varying selection.

Conclusion: Our results are consistent with a mechanistic and selective decoupling of nighttime and daytime activity. Furthermore, the present study suggests that natural selection plays a major role in generating transcriptomic variation associated with circadian behaviors. Finally, we identified genomic variants plausibly causally associated with the observed behavioral and transcriptomic variation.

Show MeSH
Geographic variation in locomotor activity among five AmericanD. melanogasterpopulations. The graph was obtained from the Eduction analysis of FaasX software. The white bar underneath the graph represents the photophase, the black bar the scotophase. The populations are Maine (ME; latitude: 44°37′N), Rhode Island (RI; 41°49′N), Virginia (VA; 37°32′N), Florida (FL, 30°20′N) and Panama City (PC; 8°58′N). The raw locomotor activity profiles are shown on Additional file 1: Figure S2.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4374177&req=5

Fig1: Geographic variation in locomotor activity among five AmericanD. melanogasterpopulations. The graph was obtained from the Eduction analysis of FaasX software. The white bar underneath the graph represents the photophase, the black bar the scotophase. The populations are Maine (ME; latitude: 44°37′N), Rhode Island (RI; 41°49′N), Virginia (VA; 37°32′N), Florida (FL, 30°20′N) and Panama City (PC; 8°58′N). The raw locomotor activity profiles are shown on Additional file 1: Figure S2.

Mentions: D. melanogaster males sampled from populations collected along a latitudinal gradient ranging from Maine (USA; 44°N) to Panama City (Panama; 8°N) were entrained under semi-natural conditions (i.e. oscillating light and temperature; for more details see Methods and Additional file 1: Figure S1) prior to measurement of their locomotor activity. Nighttime locomotor activity profiles from higher and the lower latitudes differ substantially (Figure 1, Additional file 1: Figure S2). The regressions over latitude of the average locomotor activity during the photophase and the scotophase (Figure 2A and B respectively) show that nighttime locomotor activity is more strongly correlated with latitude (R-square = 0.62) than daytime locomotor activity (R-square = 0.09) suggesting a contrast between nighttime vs. daytime patterns. To further investigate population differences, locomotor activity was parsed into two main components: sleep (i.e. average sleep bout duration; Figure 2C and D) and walking speed (Figure 2E and F). The former showed a very strong relationship with latitude, which explained 80% of the observed phenotypic variation (p = 0.03; Figure 2D); the difference in average sleep bout duration between temperate and equatorial populations was about two-fold. Both sleep duration and sleep bout number contribute to the observed pattern (see Additional file 1: Figure S3). The observation that nighttime walking speed shows no evidence of latitudinal variation (Figure 2E and F) supports the idea that sleep (i.e. the bouts of inactivity), rather than walking speed (i.e. the absolute number of infrared beam crosses), constitutes the key behavioral difference in nighttime activity levels in higher vs. lower latitude populations. To our knowledge this is the first demonstration of genetically determined geographic differentiation for sleep behavior in Drosophila. Importantly, the different patterns of geographic variation of sleep do not result from sharp differences occurring over short time periods (Figure 3), but rather, from a general night versus day pattern. These observations support the idea that nighttime and daytime sleep are mechanistically distinct [17-20] and may evolve independently.Figure 1


Evidence that natural selection maintains genetic variation for sleep in Drosophila melanogaster.

Svetec N, Zhao L, Saelao P, Chiu JC, Begun DJ - BMC Evol. Biol. (2015)

Geographic variation in locomotor activity among five AmericanD. melanogasterpopulations. The graph was obtained from the Eduction analysis of FaasX software. The white bar underneath the graph represents the photophase, the black bar the scotophase. The populations are Maine (ME; latitude: 44°37′N), Rhode Island (RI; 41°49′N), Virginia (VA; 37°32′N), Florida (FL, 30°20′N) and Panama City (PC; 8°58′N). The raw locomotor activity profiles are shown on Additional file 1: Figure S2.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4374177&req=5

Fig1: Geographic variation in locomotor activity among five AmericanD. melanogasterpopulations. The graph was obtained from the Eduction analysis of FaasX software. The white bar underneath the graph represents the photophase, the black bar the scotophase. The populations are Maine (ME; latitude: 44°37′N), Rhode Island (RI; 41°49′N), Virginia (VA; 37°32′N), Florida (FL, 30°20′N) and Panama City (PC; 8°58′N). The raw locomotor activity profiles are shown on Additional file 1: Figure S2.
Mentions: D. melanogaster males sampled from populations collected along a latitudinal gradient ranging from Maine (USA; 44°N) to Panama City (Panama; 8°N) were entrained under semi-natural conditions (i.e. oscillating light and temperature; for more details see Methods and Additional file 1: Figure S1) prior to measurement of their locomotor activity. Nighttime locomotor activity profiles from higher and the lower latitudes differ substantially (Figure 1, Additional file 1: Figure S2). The regressions over latitude of the average locomotor activity during the photophase and the scotophase (Figure 2A and B respectively) show that nighttime locomotor activity is more strongly correlated with latitude (R-square = 0.62) than daytime locomotor activity (R-square = 0.09) suggesting a contrast between nighttime vs. daytime patterns. To further investigate population differences, locomotor activity was parsed into two main components: sleep (i.e. average sleep bout duration; Figure 2C and D) and walking speed (Figure 2E and F). The former showed a very strong relationship with latitude, which explained 80% of the observed phenotypic variation (p = 0.03; Figure 2D); the difference in average sleep bout duration between temperate and equatorial populations was about two-fold. Both sleep duration and sleep bout number contribute to the observed pattern (see Additional file 1: Figure S3). The observation that nighttime walking speed shows no evidence of latitudinal variation (Figure 2E and F) supports the idea that sleep (i.e. the bouts of inactivity), rather than walking speed (i.e. the absolute number of infrared beam crosses), constitutes the key behavioral difference in nighttime activity levels in higher vs. lower latitude populations. To our knowledge this is the first demonstration of genetically determined geographic differentiation for sleep behavior in Drosophila. Importantly, the different patterns of geographic variation of sleep do not result from sharp differences occurring over short time periods (Figure 3), but rather, from a general night versus day pattern. These observations support the idea that nighttime and daytime sleep are mechanistically distinct [17-20] and may evolve independently.Figure 1

Bottom Line: We also found evidence for geographic variation for sunrise anticipation.Furthermore, the present study suggests that natural selection plays a major role in generating transcriptomic variation associated with circadian behaviors.Finally, we identified genomic variants plausibly causally associated with the observed behavioral and transcriptomic variation.

View Article: PubMed Central - PubMed

Affiliation: Department of Evolution and Ecology, University of California, 3352 Storer Hall, One Shields Ave., Davis, CA, 95618, USA. nhsvetec@ucdavis.edu.

ABSTRACT

Background: Drosophila melanogaster often shows correlations between latitude and phenotypic or genetic variation on different continents, which suggests local adaptation with respect to a heterogeneous environment. Previous phenotypic analyses of latitudinal clines have investigated mainly physiological, morphological, or life-history traits. Here, we studied latitudinal variation in sleep in D. melanogaster populations from North and Central America. In parallel, we used RNA-seq to identify interpopulation gene expression differences.

Results: We found that in D. melanogaster the average nighttime sleep bout duration exhibits a latitudinal cline such that sleep bouts of equatorial populations are roughly twice as long as those of temperate populations. Interestingly, this pattern of latitudinal variation is not observed for any daytime measure of activity or sleep. We also found evidence for geographic variation for sunrise anticipation. Our RNA-seq experiment carried out on heads from a low and high latitude population identified a large number of gene expression differences, most of which were time dependent. Differentially expressed genes were enriched in circadian regulated genes and enriched in genes potentially under spatially varying selection.

Conclusion: Our results are consistent with a mechanistic and selective decoupling of nighttime and daytime activity. Furthermore, the present study suggests that natural selection plays a major role in generating transcriptomic variation associated with circadian behaviors. Finally, we identified genomic variants plausibly causally associated with the observed behavioral and transcriptomic variation.

Show MeSH