Comparative study of resistance to heat in two species of leaf-cutting ants
- First Online:
- Cite this article as:
- Bouchebti, S., Jost, C., Caldato, N. et al. Insect. Soc. (2015) 62: 97. doi:10.1007/s00040-014-0378-y
- 182 Downloads
Atta laevigata and Atta capiguara are two species of leaf-cutting ants that are found in the pastures of Central Brazil and build huge underground nests linked to the outdoor environment by underground tunnels, which can reach several tens of meters and further extend through foraging trails to distant foraging grounds. The tunnels built by mature colonies of A. capiguara are usually longer and deeper than those built by mature colonies of A. laevigata. The physical trails are also shorter on average. We hypothesized that these differences could be related to differences in thermotolerance between the two species. To test this we collected ants on foraging trails and placed them individually in waterproof test tubes plunged in a thermostatic bath at 25 °C (control), 37 and 39 °C (test temperatures). The results showed that at both 37 and 39 °C, the survival time of A. laevigata was much more extended than that of A. capiguara. A possible explanation for the longer and deeper underground foraging tunnels, as well as the shorter foraging trails, built by A. capiguara may thus be their lower resistance to heat stress. The longer tunnels built by A. capiguara colonies may reduce the exposure to heat of the foraging workers that commute between their nest and the foraging grounds or act as a thermal refuge in which the workers can find temporary protection against high outdoor temperatures.
KeywordsLeaf-cutting ants Atta laevigata Atta capiguara Longevity Thermotolerance
Atta capiguara and Atta laevigata are two species of leaf-cutting ants that are found in the pastures of Central Brazil and build huge underground nests which consist for mature colonies of more than thousand chambers containing brood and fungus (Moreira et al., 2004; Bollazzi et al., 2012). These chambers are linked to the outdoor environment by underground tunnels that radiate from the nest and further extend through physical trails to distant foraging grounds (Moreira et al., 2004). Cement cast and subsequent excavation of these tunnels show that the tunnels built by mature colonies of A. capiguara are usually longer and deeper under the soil surface than those built by mature colonies of A. laevigata (Moreira et al., 2004; Bollazzi et al., 2012). The physical trails built by A. capiguara are also on average shorter than those built by A. laevigata. We hypothesized that the differences in tunnel architecture and physical trail length could be related to differences in thermotolerance between the two species. Long foraging tunnels and short foraging trails may reduce the duration of exposure of foraging workers to stressful outdoor temperatures when collecting vegetation far away from their nests, especially for workers that travel all the way from their nest to distant foraging grounds. The tunnels may also act as a thermal refuge in which foraging workers can retreat regularly to reduce exposure to high outdoor temperatures. To test this hypothesis, we compared the resistance to heat of A. capiguara and A. laevigata workers by examining their mortality rate when exposed to high temperatures in a hot water bath.
We worked on three colonies of each species located in a pasture near Botucatu, state of São Paulo, Brazil (22°50′46″S, 48°26′02″W). Ants were collected from the main trails of the nests on the same day between 7 am and 9 am, between October 2012 and March 2013. Back in the laboratory, their fresh weight was determined to the nearest 0.01 mg and each ant was immediately placed individually in an Eppendorf tube which had a piece of wet cotton plugged into its lid to saturate the air with humidity. After an acclimation period of 30 min at room temperature (25 °C), the tubes were placed in a hot water bath that was set at three different temperatures: 25 °C (control temperature), 37 or 39 °C (test temperatures), which were just above the maximal temperature recorded in Botucatu in 2012 (36 °C). The state of the ants was noted every 5 min during a maximum time of 5 h. Dead ants were immediately removed from the bath and placed for 3 days in a dry oven at 50 °C. After this period, they were removed from the tube and their dry weight was measured. Ants of the two species were tested at the same time and in the same hot water bath as groups of 20 individuals of each species. In total, 120 ants were tested for each temperature and each species, 40 for each colony.
To study the effect of worker size on longevity, ants were categorized into four size classes: <3, 3–6, 6–12 and >12 mg. We used the R package coxme (Therneau, 2012) to fit the longevity data for 37 and 39 °C with two separate mixed effects Cox proportional hazards models. Ants that did not die at the end of the 5-h heat exposure were considered as censored. The effects of species, ant size class, as well as the interaction between these two variables were tested in the models. The variable colony was entered as a random factor. Each explanatory variable was tested by comparing the change in likelihood between the models with and without this variable with a likelihood ratio test based on the χ2 statistic. All statistical tests were conducted with R 2.13.1 (R Development Core Team, 2011).
Except for the largest individuals, i.e., those weighing more than 12 mg, workers of A. capiguara had generally a lower proportion of water [calculated using the formula: (1 − dry weight/fresh weight)*100] in their bodies than those of A. laevigata (mean ± SE, for A. capiguara and A. laevigata, respectively, and results of Wilcoxon test: <3 mg: 60.44 ± 0.96 and 65.64 ± 0.72, W = 3296.0, p < 0.001; 3–6 mg: 61.80 ± 0.76 and 65.45 ± 0.51, W = 1908.0, P < 0.001; 6–12 mg: 63.14 ± 0.30 and 64.32 ± 0.29, W = 1545.0, p < 0.001; >12 mg: 60.95 ± 1.26 and 60.19 ± 1.32, W = 1094.5, p = 0.258). The most likely explanation for the higher resistance of A. laevigata workers is therefore their higher body water content compared to A. capiguara workers. However, other factors that are known to increase resistance to dehydration stress, such as the thickness of the cuticle and the amount and composition of cuticular hydrocarbons (Hood and Tschinkel, 1990), or the presence of particular metabolic compounds such as heat shock proteins (Gehring and Wehner, 1995; Chown et al., 2011) cannot be excluded.
Our results show therefore that the longer and deeper underground foraging tunnels and the shorter foraging trails built by A. capiguara may allow reducing the exposure of its foraging workers to high outdoor temperatures. Note also that A.laevigata is characterized by a more extended size polymorphism than A.capiguara (range of body dry mass in our samples: 0.50–41.70 and 0.70–92.30 mg, for A. capigurara and A. laevigata, respectively). Given the fact that bigger workers are more resistant to heat stress than smaller workers, this should also allow A. laevigata to forage at higher temperatures than A. capiguara.
This work was supported by a French–Brazilian CAPES-COFECUB grant (No. 633/09). S. B was financed by a doctoral grant from the French Ministry of Education. We thank J. B. Ferdy for statistical advice.