Figures and data
Experimental protocol and differential response of activity waveform components to out-of-sync zeitgeber cycles.
(A) Schematic of all the environmental regimes. (B) Activity profiles under LDTC out-of-sync regime with TC delayed by 6 hours with respect to LD (solid trace) compared to that under LDTC in-sync regime (dashed trace). Error bands are SEM across 4 populations. For both A&B, yellow and black indicate photophase and scotophase while red and blue indicate thermophase and cryophase, respectively. Lights-ON = ZT00 (zeitgeber time 00). (C) Sum of squared differences calculated from the activity profiles under in-sync and out-of-sync zeitgeber cycles in the morning and evening window for wild-type Drosophila melanogaster populations for a range of phase delays of TC relative to LD (ΨTC-LD). For two-way repeated measures ANOVA, time window (morning/evening) F(1,18) = 14.75, p = 0.001, LDTC regime F(5,18) = 3.588, p = 0.02, LDTC regime*time window F(5,18) = 7.01, p = 0.0008 (D) Change in mean phase of activity (ΔφCoM) between in-sync and out-of-sync zeitgeber cycles in the morning and evening window for wild-type Drosophila melanogaster populations across a range of phase delays of TC relative to LD (ΨTC-LD). For two-way repeated measures ANOVA, time window (morning/evening) F(1,18) = 152.3, p < 0.0001, LDTC regime F(5,18) = 60.52, p < 0.0001, LDTC regime*time window F(5,18) = 23.72, p < 0.0001. Error bars are SEM across four populations of D. melanogaster.
Light: dark cycle and its phase relationship with temperature cycle regulates temperature-dependent timeless splicing.
Relative mRNA expression levels of timeless (A) and its splice variants (B-D) as well as splicing regulator Psi (E) across 24 hours sampled at 3 hr intervals. Subpanels 1-3 plot the expression levels of the indicated transcripts under the three environmental regimes – TC 12:12 (subpanels1), LD TC in-sync (subpanels 2) and LDTC out-of-sync (subpanels 3) with p-value for RAIN rhythmicity test mentioned inset for each. Subpanels 4 & 5 compare the peak normalised expression values between TC 12:12 versus LDTC in-sync and LDTC in-sync versus out-of-sync, respectively. Asterisks indicate significant differences at that time point between the two profiles by Sidak’s multiple comparisons test following a two-way ANOVA (with regime and time point as the factors). Colour coding for environmental cycles and ZT00 conventions are as described for Figure 1.
Differential sensitivity of evening activity bout among fly chronotype populations under environmental regimes with a range of phase relationship between the light and temperature cycles.
(A) Activity profiles of early1-4, control1-4 and late1-4 selection lines under an environmental regime with light and temperature cycles in-sync (dashed trace) and out-of-sync (solid trace). Error bands = SEM across four populations. (B) Sum of square differences calculated from the activity profiles under in-sync and out-of-sync zeitgeber cycles in the morning window for the selection lines. (C) Change in mean phase of activity between in-sync and out-of-sync zeitgeber cycles in the morning window for the selection lines. (D) Sum of square differences calculated from the activity profiles under in-sync and out-of-sync zeitgeber cycles in the evening window for the selection lines with SEM error bars. (E) Change in mean phase of activity between in-sync and out-of-sync zeitgeber cycles in the evening window for the selection lines. All error bars, unless mentioned otherwise, are 95% CIs from Tukey’s HSD for the interaction effect between the fixed factors of selection and environmental regime in a three-factor ANOVA design with block as random factor, with non-overlapping error bars indicating significant differences. Colour coding for environmental cycles and ZT00 conventions are as described for Figure 1.
Temporal expression of timeless in laboratory-selected fly chronotypes across three environmental regimes.
(A1-A3) timeless expression in one block of the selection lines early4, control4, late4 under TC 12:12 (A1), LDTC in-sync (A2), and LDTC out-of-sync (A3). Asterisks indicate significant differences at those time points among the selection lines (Tukey’s HSD, two-way ANOVA). RAIN rhythmicity p-values inset. (A4) Amplitude (left) and mean phase (right) under TC 12:12, one-way ANOVA with Tukey’s HSD indicated by asterisk. (B1-B3) Peak-normalized timeless profiles in control4, early4, and late4 under TC 12:12 (solid) and LDTC in-sync (dashed). (B4) Relative amplitude across environmental regimes in selection lines; one-way ANOVA with Tukey’s HSD indicated by asterisk. (C1-C3) Peak-normalized profiles under LDTC out-of-sync (solid) compared to LDTC in-sync (dashed). Environmental regimes for the solid and dashed traces are indicated at the top and bottom of the expression profiles, respectively. Colour coding for environmental cycles and ZT00 conventions are as described for Figure 1.
Temporal expression of cold-induced timeless splice variants in fly chronotypes across three environmental regimes.
(A1-3) tim-cold expression over 24 h in selection lines under TC 12:12, LDTC in-sync, and LDTC out-of-sync, respectively; RAIN p-values inset. (A4) Mean phase under TC 12:12 (left) and total expression during cryophase (right) under TC 12:12; one-way ANOVA with Tukey’s HSD indicated by asterisk. (B1-3) Peak-normalized tim-cold expression in early4, control4, and late4 lines under LDTC in-sync (dashed) and out-of-sync (solid) conditions. Environmental regimes for the solid and dashed traces are indicated at the top and bottom of the expression profiles, respectively. (B4) Change in expression level at ZT11 between out-of-sync and in-sync conditions; one-way ANOVA with Tukey’s HSD (*p=0.03). (C1-3) tim-short and cold expression across 24 h in selection lines under TC 12:12, LDTC in-sync, and LDTC out-of-sync, respectively, RAIN p-values inset. Colour coding for environmental cycles and ZT00 conventions as described in Figure 1. Error bars: SEM of three biological replicates.
Temporal expression of warm-induced splice variant tim-medium and of Psi, a splicing regulator of timeless, in fly chronotype populations across three environmental regimes.
(A1-3) Expression levels of tim-medium across a day in the selection lines under TC 12:12, LDTC in-sync regime and under LDTC 12:12 out-of-sync regime, respectively. RAIN p-values inset. (B1-3) Peak normalised expression profile of tim-medium in control4, early4 and late4 under TC and LDTC 12:12 in-sync conditions, solid and dashed traces, respectively. Environmental regimes for the solid and dashed traces are indicated at the top and bottom of the expression profiles, respectively. (C1-3) Expression levels of Psi across a day in the selection lines under TC 12:12, LDTC in-sync regime and under LDTC 12:12 out-of-sync regime, respectively. Colour coding for environmental cycles and ZT00 conventions as described in Figure 1. Error bars are SEM across three biological replicates.
Model for integration of light and temperature signals by evening oscillator neurons to fine tune the timing of evening activity:
The evening oscillator cluster consists of two mutually coupled neuronal groups—one predominantly light-sensitive (EL) and the other predominantly temperature-sensitive (ET). The phasing of evening activity results from their combined influence in response to environmental cues. Photic input affects the light-sensitive group via its coupling with morning cells and presence of CRY, while temperature input influences the temperature-sensitive group through its connection with temperature-sensitive dorsal neurons and absence of CRY. Additionally, the overall evening activity phase depends on the coupling between EL and ET. This model assumes symmetric sensitivity to zeitgebers and mutual coupling between EL and ET, consistent with theoretical models for emergence rhythms under similar experimental conditions.
Expression pattern of Psi along with the warm and cold-induced transcripts in control4 under three environmental regimes.
(A1&2) Peak normalised expression profile of Psi with the cold-induced transcripts and tim-medium under TC 12:12. (B1&2) Peak normalised expression profile of Psi with the cold-induced transcripts and tim-medium under LDTC in-sync. (C1&2) Peak normalised expression profile of Psi with the cold-induced transcripts and tim-medium under LDTC out-of-sync. Lights-ON = ZT00 (zeitgeber time 00). Colour coding for environmental cycles and ZT00 conventions as described in Figure 1. All error bars are SEM across three biological replicates.
Model for correlated evolution of enhanced plasticity of evening bout in late populations in response to temperature cycles:
Any one or more of the above mechanisms could be involved in the enhanced plasticity of evening activity in late in response to temperature cycles. Late flies show reduced levels of the cry transcript (Nikhil et al., 2016), lower CRY can facilitate a stronger response to temperature cues (Harper et al., 2016). This is also possible if the evening oscillator receives weaker input from the morning cells and stronger input from the temperature-sensitive dorsal cells in late populations compared to control. Finally, asymmetry in coupling between the EL and ET in the above indicated direction can also cause stronger responses to temperature cues because ET could then dictate the rhythms in EL.
Experimental protocol.
(left) schematic of all the environmental regimes used in this study and (right) an example batch actogram, activity was first recorded under conditions with light (70 lux) and temperature cycles (19 °C and 28 °C) being in-sync and then transferred to one of the out-of-sync regimes. In this example flies were transferred on the 4th day of recording to an environmental regime wherein the temperature cycle was delayed by 8 hours with respect to the light cycle. Green hue = light + low temperature, orange hue = light + high temperature, red hue = darkness + high temperature and grey hue = darkness + low temperature.
Summary of experimental protocol:
early4, control4 and late4 flies (3-5 day old virgin males) after one generation of common rearing (no selection) were entrained to three different environmental regimes for 5 days. Sample collection at the above indicated time points were conducted on the 5th cycle. Three biological replicates (with 50 flies each) were flash frozen using liquid nitrogen at each of the time point under all three regimes. Previously reported primers have been used in this study (see Table S6). Primers amplifying the exons 5&6 are used to quantify all timeless mRNA since these exons are common to all timeless variants. Amplification of the introns that are unique to each of the variants are used to quantify tim-medium, tim-cold and tim-short and cold.
Activity profiles of divergent chronotype fly populations under environmental regimes with a range of phase relationship between the light and temperature cycles.
(A1-6) Activity profiles of wildtype Drosophila melanogaster populations (control1-4) under environmental regimes with specific phase relationships between light and temperature cycles. (B1-6) Activity profiles of wildtype Drosophila melanogaster populations selected for morning fly emergence (early1-4) under environmental regimes with specific phase relationships between light and temperature cycles. (C1-6) Activity profiles of wildtype Drosophila melanogaster populations selected for evening fly emergence (late1-4) under environmental regimes with specific phase relationships between light and temperature cycles. Activity profiles under in-sync regime have been plotted as dashed traces while those under out-of-sync regimes have been plotted as solid traces for each of the experiments. Error bands are SEM across the four populations. The bars on top indicate the out-of-sync environmental regime with yellow and black indicating photophase and scotophase while red and blue indicate thermophase and cryophase, respectively.
(A) Amplitude of timeless expression (maximum – minimum) compared between TC 12:12 and LDTC 12:12 in-sync regimes. The asterisks indicate that the amplitude was significantly lower under TC compared to LDTC in-sync conditions by an unpaired t-test (p = 0.0005). (B) Amplitude of tim-cold expression (maximum – minimum) compared between LDTC 12:12 out-of-sync and LDTC 12:12 in-sync regimes. The asterisks indicate that the amplitude was significantly lower under LDTC out-of-sync compared to LDTC in-sync conditions by an unpaired t-test (p = 0.0004). Error bars are SEM. (C) Amplitude of tim-short and cold expression (maximum – minimum) compared between TC 12:12 and LDTC 12:12 in-sync regimes. The asterisks indicate that the amplitude was significantly lower under TC compared to LDTC in-sync conditions by an unpaired t-test (p = 0.0003). Error bars are SEM.
(A1-3) Peak normalised expression profile of tim-cold in control, early and late under TC 12:12 and under LDTC 12:12 in-sync conditions, solid and dashed traces, respectively. (B1-3) Peak normalised expression profile of tim-sc in control, early and late under TC 12:12 and under LDTC 12:12 in-sync conditions, solid and dashed traces, respectively. (C1-3) Peak normalised expression profile of tim-sc in control, early and late under LDTC out-of-sync and LDTC 12:12 in-sync conditions, solid and dashed traces, respectively. Environmental regimes for the solid and dashed traces are indicated at the top and bottom of the expression profiles, respectively. The bars indicating environmental regimes have solid or dashed outlines, corresponding to the solid or dashed traces for the respective expression profiles. Yellow and black bars for light and dark, respectively, while red and blue bars are for warm and cool phases, respectively. Error bars are SEM across three biological replicates.
(A1-3) Peak normalised expression profile of tim-medium in control, early and late under TC 12:12 and under LDTC 12:12 in-sync conditions, solid and dashed traces, respectively. (B1-3) Peak normalised expression profile of PSI in control, early and late under TC 12:12 and under LDTC 12:12 in-sync conditions, solid and dashed traces, respectively. (C1-3) Peak normalised expression profile of PSI in control, early and late under LDTC out-of-sync and LDTC 12:12 in-sync conditions, solid and dashed traces, respectively. Environmental regimes for the solid and dashed traces are indicated at the top and bottom of the expression profiles, respectively. The bars indicating environmental regimes have solid or dashed outlines, corresponding to the solid or dashed traces for the respective expression profiles. Yellow and black bars for light and dark, respectively, while red and blue bars are for warm and cool phases, respectively. Error bars are SEM across three biological replicates.
Two-way ANOVA tables for comparison of peak normalised expression values between two regimes for a given transcript.
Regime and time point were the two factors, and the tables list their main effects as well as the interaction effect between the two factors. The transcript and the two environmental regimes between which its expression profile is compared are mentioned in the top cell of the first column for each table. For all the tables, statistically significant effects have been marked using red font.
Three-way ANOVA tables with LDTC regime (phase relationship between light and temperature cycles) and selection as the main factors and blocks (4 populations that have undergone selection) as random factors.
The measure (SSD or mean phase of activity) and the time window (morning or evening activity) is mentioned in the top cell of the first column.
Expression profiles of total timeless in selection lines across three environmental regimes.
(A) (top) Two-way ANOVA table comparing expression profiles of the transcript among the selection lines under TC 12:12. (bottom) Significant differences among the selection lines at indicated time points by Tukey’s multiple comparisons following the above two-way ANOVA. (B) (top) Two-way ANOVA table comparing expression profiles of the transcript among the selection lines under LDTC in-sync. (bottom) Significant differences among the selection lines at indicated time points by Tukey’s multiple comparisons following the above two-way ANOVA. (C) (top) Two-way ANOVA table comparing expression profiles of the transcript among the selection lines under LDTC out-of-sync. (D) Three-way ANOVA table for comparing expression profiles of the transcript across TC 12:12 and LDTC in-sync regimes among the selection lines. The table lists the three factors and their interaction effects. (D) Three-way ANOVA table for comparing expression profiles of the transcript across LDTC in-sync and LDTC out-of-sync regimes among the selection lines. The table lists the three factors and their interaction effects. For all the tables, statistically significant effects and differences have been marked by using red font.
Expression profiles of cold-induced timeless transcripts in selection lines across three environmental regimes.
(A) Two-way ANOVA table comparing expression profiles of tim-cold among the selection lines under TC 12:12. (B) Two-way ANOVA table comparing expression profiles of tim-cold among the selection lines under LDTC in-sync. (C) Two-way ANOVA table comparing expression profiles of tim-cold among the selection lines under LDTC out-of-sync. (D) Three-way ANOVA table for comparing expression profiles of tim-cold across TC 12:12 and LDTC in-sync regimes among the selection lines. The table lists the three factors and their interaction effects. (E) Three-way ANOVA table for comparing expression profiles of tim-cold across LDTC in-sync and LDTC out-of-sync regimes among the selection lines. The table lists the three factors and their interaction effects. (F) Two-way ANOVA table comparing expression profiles of tim-short and cold among the selection lines under TC 12:12. (G) Two-way ANOVA table comparing expression profiles of tim-short and cold among the selection lines under LDTC in-sync. (H) Two-way ANOVA table comparing expression profiles of tim-short and cold among the selection lines under LDTC out-of-sync. (I) Three-way ANOVA table for comparing expression profiles of tim-short and cold across TC 12:12 and LDTC in-sync regimes among the selection lines. The table lists the three factors and their interaction effects. (J) Three-way ANOVA table for comparing expression profiles of tim-short and cold across LDTC in-sync and LDTC out-of-sync regimes among the selection lines. The table lists the three factors and their interaction effects. For all the tables, statistically significant effects and differences have been marked by using red font.
Expression profiles of warm-induced timeless transcript tim-medium and timeless splicing regulator Psi in selection lines across three environmental regimes.
(A) Two-way ANOVA table comparing expression profiles of tim-medium among the selection lines under TC 12:12. (B) Two-way ANOVA table comparing expression profiles of tim-medium among the selection lines under LDTC in-sync. (bottom) Significant differences among the selection lines at the indicated time point by Tukey’s multiple comparisons following the above two-way ANOVA. (C) Two-way ANOVA table comparing expression profiles of tim-medium among the selection lines under LDTC out-of-sync. (D) Three-way ANOVA table for comparing expression profiles of tim-medium across TC 12:12 and LDTC in-sync regimes among the selection lines. The table lists the three factors and their interaction effects. (bottom) Significant differences from Tukey’s HSD following the above ANOVA for the interaction term between selection and regime. (E) Three-way ANOVA table for comparing expression profiles of tim-medium across LDTC in-sync and LDTC out-of-sync regimes among the selection lines. The table lists the three factors and their interaction effects. (F) Two-way ANOVA table comparing expression profiles of Psi among the selection lines under TC 12:12. (G) Two-way ANOVA table comparing expression profiles of Psi among the selection lines under LDTC in-sync. (H) Two-way ANOVA table comparing expression profiles of Psi among the selection lines under LDTC out-of-sync. (I) Three-way ANOVA table for comparing expression profiles of Psi across TC 12:12 and LDTC in-sync regimes among the selection lines. The table lists the three factors and their interaction effects. (J) Three-way ANOVA table for comparing expression profiles of Psi across LDTC in-sync and LDTC out-of-sync regimes among the selection lines. The table lists the three factors and their interaction effects. For all the tables, statistically significant effects and differences have been marked using red font.
