Fitness variation in response to artificial selection for reduced cell area, cell number and wing area in natural populations of Drosophila melanogaster

Abstract

Background: Genetically based body size differences are naturally occurring in populations of Drosophila melanogaster, with bigger flies in the cold. Despite the cosmopolitan nature of body size clines in more than one Drosophila species, the actual selective mechanisms controlling the genetic basis of body size variation are not fully understood. In particular, it is not clear what the selective value of cell size and cell area variation exactly is. In the present work we determined variation in viability, developmental time and larval competitive ability in response to crowding at two temperatures after artificial selection for reduced cell area, cell number and wing area in four different natural populations of D. melanogaster.

Results: No correlated effect of selection on viability or developmental time was observed among all selected populations. An increase in competitive ability in one thermal environment (18 degrees C) under high larval crowding was observed as a correlated response to artificial selection for cell size.

Conclusion: Viability and developmental time are not affected by selection for the cellular component of body size, suggesting that these traits only depend on the contingent genetic makeup of a population. The higher larval competitive ability shown by populations selected for reduced cell area seems to confirm the hypothesis that cell area mediated changes have a relationship with fitness, and might be the preferential way to change body size under specific circumstances.

Figure 1

Response to selection. Mean values of wing area, cell area and cell number of females and males (± standard errors) of the four populations during nine generations of different selection regime.

Figure 2

Standardised size differences. Standardised differences between the experimental lines and their respective base population (± standard errors) after one generation of mass-breeding for the three selection regimes. The values of males and females are pooled. Circles indicate selection lines and squares inbred control lines. Belém population is represented in blue, Madagascar in red, Paris in black and USA in violet.

Figure 3

Relative viability. Standardised mean viability (± standard error) of the selected lines at 25°C (a) and 18°C (b).

Figure 4

Relative developmental time. Standardised mean developmental time (± standard error) of the selected lines at 25°C (a) and 18°C (b).

Figure 5

Relative competitive ability. Standardised mean competitive ability (± standard error) of selected lines. a) Relative percentage of wild-type flies from total emerging flies at 25°C and 15 wt/45 y, w density. b) Relative percentage of wild-type flies from total emerging flies at 18°C and 15 wt/45 y, w density. c) Relative percentage of wild-type flies from total emerging flies at 25°C and 30 wt/90 y, w density. d) Relative percentage of wild-type flies from total emerging flies at 18°C and 30 wt/90 y, w density.

Figure 6

Drosophila melanogaster wing. The black outline superimposed on the wing joins the six points (1–6) used to determine wing area. The box (A) indicates the standard region used for trichome counting to estimate average cell area. On wings of different size, the region was chosen corresponding to the equivalent location with respect to veins and wing margin.