Heat tolerance inDrosophila subobscuraalong a latitudinal gradient: Contrasting patterns between plastic and genetic responses

Abstract

Susceptibility to global warming relies on how thermal tolerances respond to increasing temperatures through plasticity or evolution. Climatic adaptation can be assessed by examining the geographic variation in thermal-related traits. We studied latitudinal patterns in heat tolerance in Drosophila subobscura reared at two temperatures. We used four static stressful temperatures to estimate the thermal death time (TDT) curves, and two ramping assays with fast and slow heating rates. Thermal death time curves allow estimation of the critical thermal maximum (CT(max)), by extrapolating to the temperature that would knock down the flies almost "instantaneously," and the thermal sensitivity to increasing stressful temperatures. We found a positive latitudinal cline for CT(max), but no clinal pattern for knockdown temperatures estimated from the ramping assays. Although high-latitude populations were more tolerant to an acute heat stress, they were also more sensitive to prolonged exposure to less stressful temperatures, supporting a trade-off between acute and chronic heat tolerances. Conversely, developmental plasticity did not affect CT(max) but increased the tolerance to chronic heat exposition. The patterns observed from the TDT curves help to understand why the relationship between heat tolerance and latitude depends on the methodology used and, therefore, these curves provide a more complete and reliable measurement of heat tolerance.

How to assess Drosophila heat tolerance: Unifying static and dynamic tolerance assays to predict heat distribution limits

“…Previous studies have reported a positive correlation between z and sCT max (1 min) which was interpreted as a trade‐off between chronic and acute heat stress (Castañeda et al, ; Rezende et al, ; Truebano et al, ). When analysing this relationship within the 11 closely related drosophilids, we did not find any positive correlation, instead linear regression on z and the extrapolated sCT max (1 min) showed a non‐significant and slightly negative correlation (slope = −0.062, F = 0.003, df = 1,9, p = 0.955, see Supporting Information Figure S3).…”

“…In previous literature, this has been illustrated as the intercept with the x ‐axis, sCT max (1 min) (because log 10 (1 min) = 0, Figure b). Additionally, the temperature that would cause knockdown after 1 hr (log 10 (60 min) = 1.778) was calculated as well as Q 10 ( Q 10 = 10 10/ z , Castañeda et al ()), to report the thermal sensitivity of knockdown time.…”

“…Likewise, it is clear that the critical temperature measured by dynamic methods depends on the rate at which temperature is increased (Overgaard, Kristensen, & Sørensen, ; Sørensen, Loeschcke, & Kristensen, ; Terblanche, Deere, Clusella‐Trullas, Janion, & Chown, ). It has also been discussed if static and dynamic assays impose the same form of heat stress on the animal (Castañeda, Rezende, & Santos, ; Cooper, Williams, & Angilletta, ; Kilgour & McCauley, ; Mitchell & Hoffmann, ; Rezende et al, ; Santos, Castañeda, & Rezende, ; Sgrò et al, ; Terblanche et al, ), and several studies have therefore highlighted the apparent problem of comparing results across the two assays. Finally, both assays have been questioned for their aptness in describing thermal tolerance in ecologically relevant settings (Chown, Jumbam, Sørensen, & Terblanche, ; Lutterschmidt & Hutchison, ; Mitchell & Hoffmann, ; Ribeiro, Camacho, & Navas, ; Terblanche et al, ).…”

“…Finally, both assays have been questioned for their aptness in describing thermal tolerance in ecologically relevant settings (Chown, Jumbam, Sørensen, & Terblanche, ; Lutterschmidt & Hutchison, ; Mitchell & Hoffmann, ; Ribeiro, Camacho, & Navas, ; Terblanche et al, ). In an attempt to solve some of these problems, Rezende et al () called renewed attention to the interaction between the intensity of heat stress and the exposure duration, and through their introduction of thermal tolerance landscapes (TTLs), they discussed how parameterisation of this interaction could give novel insights that provided more information than a single static or dynamic assay (see also Castañeda et al () and Truebano, Fenner, Tills, Rundle, and Rezende ()). Thermal tolerance landscapes are based on knockdown time curves (thermal death time curves in Rezende et al ()), which use multiple static experiments to describe heat tolerance across a temperature range.…”

“…Several static measurements are needed to estimate z and sCT max but once these parameters are estimated it is possible to calculate knockdown time to a range of assay temperatures. Furthermore, it was suggested that z and sCT max together describe thermal tolerance strategies of ectotherms, such that a high sCT max describes high tolerance to acute heat stress while a low z (i.e., steep slope) indicates high tolerance to moderately lower (chronic) temperatures (see discussions in Castañeda et al () and Rezende et al ()). In this respect, it was suggested that TTLs from different species indicated a general trade‐off between acute and chronic heat tolerance (positive correlation between sCT max and z ) in ectotherms in general (Rezende et al, ), in Drosophila subobscura from a latitudinal cline (Castañeda et al, ), and in the ontogeny of an intertidal gastropod (Truebano et al, ) (Figure b, insert).…”