Fast weight recovery, metabolic rate adjustment and gene-expression regulation define responses of cold-stressed honey bee brood

In temperate climates, low ambient temperatures in late winter and in spring can result in cold stress conditions in brood areas of weakened honey bee colonies, leading to increased levels of developmental interruptions and death of the brood. Very little is known about the physiological and molecular mechanisms that regulate honey bee brood responses to acute cold-stress. Here, we hypothesized that central regulatory pathways mediated by insulin/insulin-like peptide signalling (IIS) and adipokinetic hormone (AKH) are linked to metabolic changes in cold-stressed honey bee brood. A. mellifera brood reared at suboptimal temperatures showed diminished growth rate and arrested development progress. Notably, cold-stressed brood rapidly recovers the growth in the first 24 h after returning at control rearing temperature, sustained by the induction of compensatory mechanisms. We determined fast changes in the expression of components of IIS and AKH pathways in cold-stressed brood supporting their participation in metabolic events, growth and stress responses. We also showed that metabolic rate keeps high in brood exposed to stress suggesting a role in energy supply for growth and cell repair. Additionally, transcript levels of the uncoupling protein MUP2 were elevated in cold-stressed brood, suggesting a role for heat generation through mitochondrial decoupling mechanisms and/or ROS attenuation. Physiological, metabolic and molecular mechanisms that shape the responses to cold-stress in honey bee brood are addressed and discussed.


INTRODUCTION
Temperature is an environmental factor that has a dramatic effect on survival, growth and 59 development in ectotherms like honey bee larvae, if the heating capacity of the colony is

68
Honey bee brood rearing starts in winter and peaks in spring, decreases through summer, and 69 ceases in early fall in temperate climates (7). Cold snaps or periods of extreme temperatures 70 for three or more consecutive days frequently occur during winter and early spring in 71 temperate climates. Therefore, under these conditions, a deficient colony thermoregulation is 72 a significant source of cold stress for honey bee brood. This is a subject to take into account 73 due to climate change is predicted to cause increasing temperature variability, including cold

166
The primer sequences are listed in S2 Table. To verify the purity of the PCR products, a melting curve was produced after each run. LinRegPCR was the program employed for the inside RStudio (24). The "rearing temperature" (34°C and 25°C) and "brood age" (days after 182 larva hatching) were included in the model as fixed effects and "the independent experiment" 183 was included as random effects. All statistical tests used =0.05 to establish significance.

186
The growth curve of honey bee brood exposed to cold-stress (25°C) for 3 days strongly 187 differed compared to brood reared continuously at control temperature (34°C) (Fig 1A). During  Honey bee brood grown at 34°C gained body mass until day 6 after larva hatching while in 191 cold-stressed brood the gaining weight and larval period was extended until day 9 of 192 development, 2 days after they had been transferred back to control temperature (Fig 1A and 193 B). The maximum body mass observed in 9-days-old cold-stressed brood was similar to that

206
In cold-stressed brood, the food accumulated in rearing wells, indicating that food intake 207 decreased in these larvae (S3 Fig). This fact correlates with the diminished growth rate of 208 brood exposed to cold-stress.

209
To understand the physiological processes associated with cold-stress response in honey bee 210 brood, we measured the metabolic rate (MR). Changes occured in the MR mediated by the 211 developmental stage and the current rearing temperature (ANOVA, interaction effect between 212 rearing temperature and brood age, F 5,160 =68.6, P<0.0001; Fig 2). When honey bee brood 213 was continuously reared at 34°C, the MR gradually decreased from day 4 to day 9 after larva 214 hatching (t 11 =8.5, P=0.0001; Fig 2). The MR in the cold-stressed larvae decreased during the 215 first 48 h of low temperature (4-days-old vs 6-days-old: t 161 =13.9, P<0.0001 ; Fig 2), at the 216 same amount as the control larvae reared at 34°C (6-days-old cold-stressed vs control brood:   Fig 3B), whereas their levels were kept elevated in cold-244 stressed brood (t 175 =2.3, P=0.56; Fig 3B), even one day after returning to 34°C (t 175 =3.1,

245
P=0.1229; Fig 3B). MUP2 expression followed a similar pattern to that determined for the MR,

246
suggesting a relationship between these two parameters.  Fig 4C). Two days after returning to control temperature, the 9-days-old 271 cold-stressed brood reached the same levels of ILP2 transcript than non-stressed brood (t 215 =-272 0.6, P=0.99; Fig 4C). Coincidently, at that time, cold-stressed brood reached the maximum 273 body mass (Fig 1A).

294
We also analysed the transcript levels of AKH and its receptor AKHR. Almost no differences 295 were observed for the AKH hormone transcript between control and cold-stressed brood (10-296 days-old: t 224 =3.9, P=0.0072; Fig 5A). On the contrary, while AKHR expression levels steadily 297 decreased from day 4 to day 8 in brood reared at 34°C (4-days-old vs 8-days-old: t 186 =7.3,

317
Here, we examined physiological and molecular responses in A. mellifera brood exposed to a 318 short period of three days at low temperature. Firstly, we determined that growth rate is 319 strongly diminished when larvae are incubated at 25°C (Fig 1A), and development progress 320 is arrested (Fig 1B)  we showed that during the period of cold-stress this pattern is severely affected (Fig 2). In

338
insects under adverse environmental conditions, growth and development suppression is 339 accompanied by a decrease in MR (33). We determined that 5-and 6-day-old cold-stressed 340 brood display the same MR than brood growing at 34°C, indicating that probably a certain lag 341 of time is required until MR is adjusted. Even more, in 7 and 8-days-old cold-stressed brood

342
MR augmented compared to those reared at control temperature (Fig 2). In insects exposed 343 to low temperature an increase in MR has been suggested to provide energy for cellular 344 repairing processes (34). Thus, high levels of MR could be a necessary response contributing 345 to the recovery process in cold-stressed honey bee brood.

346
We determined that ATP accumulates in honey bee brood reared continuously at 34°C from remain at low levels in cold-stressed brood until day 8 ( Fig 4C). Interestingly, the high ILP2 388 transcript levels coincide with the growth points at which the brood reached the maximum 389 body mass, both in brood reared at control temperature and in brood exposed to cold-stress.

390
When D. melanogaster reaches a critical weight this determines the duration of the growth 391 period, and therefore the onset of metamorphosis (27). Therefore, ILP2 could be a key signal 392 in honey bee brood mediating the developmental switch with the initiation of metamorphosis 393 when optimal body mass is achieved. FoxO transcript levels remained low in honey bee brood 394 independent of the rearing temperature but with a sharp peak in 7-days-old cold-stressed

402
The AKH pathway can be involved in the regulation of carbohydrate metabolism in adult honey 403 bee workers (44). Transcript levels of AKH were not altered during cold-stress in honey bee 404 brood, while its receptor AKHR showed increased transcript accumulation all along the cold-405 stress period and even after returning to control temperature for one day (Fig 5). These results 406 may indicate that the AKH pathway is turned on in honey bee brood during the period of cold 407 exposure. We determined that glucose and trehalose levels in the haemolymph, which fuel 408 MR, maintained stable in cold-stressed brood even though larvae reduced drastically food 409 intake in that condition (S5 Fig). This result indicate that a mechanism would be activated to 410 maintain glucose and trehalose homeostasis in cold stressed brood. In the antarctic midge 411 Belgica antarctica, upregulation of genes supporting rapid glucose mobilization occurs during 412 both heat and cold-stress (45). Our results support the hypothesis that the AKH pathway