Survival for Immunity: The Price of Immune System Activation for Bumblebee Workers

Sheldon B. C., Verhulst S., Trends Ecol. Evol. 11, 317 (1996);

Kraaijeveld A. R., Godfray H. J. C., Nature 389, 278 (1997);

Boots M., Begon M., Funct. Ecol. 7, 528 (1993);

Yan G., Severson D. W., Christensen B. M., Evolution 51, 441 (1997);

Lenski R. E., Evolution 42, 425 (1988).

Nordling D., Andersson M., Zohari S., Gustafsson L., Proc. R. Soc. London Ser. B 265, 1291 (1998);

König C., Schmid-Hempel P., Proc. R. Soc. London Ser. B 260, 225 (1995);

Demas G. E., Chefer V., Talan M. C., Nelson R. J., Am. J. Physiol. 273, R1631 (1997).

To measure fitness costs, starvation protocols were also used by G. S. Wilkinson [Nature 308, 181 (1984)].

Cartar R. V., Dill L. M., Can. Entomol. 123, 283 (1991).

Schmid-Hempel P., Müller C., Schmid-Hempel R., Shykoff J. A., Insect Soc. 37, 14 (1990);

Shykoff J. A., Schmid-Hempel P., Apidologie 22, 117 (1991);

; P. Schmid-Hempel, Parasites in Social Insects (Monographs in Behavior and Ecology, Princeton Univ. Press, Princeton, NJ, 1998).

B. Heinrich, Bumblebee Economics (Harvard Univ. Press, Cambridge, MA, 1979).

N. A. Rattclife, A. F. Rowley, in Insect Hemocytes, A. P. Gupta, Ed. (Cambridge Univ. Press, London, 1979), pp. 331–414.

Rattclife N. A., Rowley A. F., Fitzgerald S. W., Rhodes C. P., Int. Rev. Cytol. 97, 183 (1985);

Rees J. A., Moniatte M., Bulet P., Insect Biochem. Mol. Biol. 27, 413 (1997).

Hultmark D., Trends Genet. 9, 178 (1993).

Söderhäll K., Dev. Comp. Immunol. 6, 601 (1982);

; M. Ashida, R. Iwama, H. Iwahana, H. Yoshida, in Proceedings of the Third International Colloquium on Invertebrate Pathology, C. C. Payne, H. B. Burges, Eds. (Univ. of Sussex, Brighton, UK, 1982), pp. 81–86;

Azumi K., Ozeki S., Yokosawa H., Ishii S., Dev. Comp. Immunol. 15, 9 (1991);

Jomori T., Kubo T., Natori S., Eur. J. Biochem. 190, 201 (1990).

Kato Y., et al., Insect Biochem. Mol. Biol. 24, 539 (1994).

Kato Y., et al., Insect Biochem. Mol. Biol. 24, 547 (1994).

Groups of 30 bumblebee workers were exposed to different doses of LPS (0, 0.1, 0.5, 1, 2, 3, 4, and 5 mg per milliliter) or latex beads (0, 10², 10³, 10⁴, 10⁵, 10⁶, and 10⁷ beads per milliliter), dissolved in Ringer, and injected as a total volume of 5 μl per insect. Controls only received Ringer but otherwise were treated identically. Before injection, bumblebee workers were anesthetized on ice and immobilized. The injections occurred into the hemocoel through the pleural membrane between the second and the third tergite, using a sterilized glass capillary that had been pulled out to a fine point. The inoculated bumblebees were kept under standard conditions in a climate chamber and fed with sugar water ad libitum.

For the assay, thoraces were separated from the rest of the body and homogenized in an Eppendorf tube containing 300 μl of Ringer solution. The homogenate was centrifuged (1300 g, 10 min, 4°C), and the supernatant were collected. The resulting solution was kept in the freezer (−80°C) for later tests. Test plates (diameter: 9 cm; Sterlin) were prepared by adding 0.05 ml of live Arthrobacter globiformis bacteria suspension (10⁷ cells per milliliter) to 5 ml of sterile broth medium (10 g of bactotryptone, 5 g of yeast extract, 10 g of NaCl, 1000 ml of distilled water, pH 7.5), with 1% of bacto-agar at 45°C. Plates were swirled to disperse the bacteria and left to settle at room temperature. Ten test holes (diameter: 2 mm) per plate were made, and 2 μl of the test solution was added per hole. Plates were then incubated overnight at 28°C. During this time, the antibacterial substances in the hemolymph inhibited bacterial growth, leading to a circular, clear zone around each hole with a diameter proportional to the strength of the inhibition. The mean of the minimum and maximum diameters of each zone was used as data point.

For this experiment, random samples of 30 to 48 workers were removed from each of nine colonies kept in a climate chamber (constant 28°C, 60% relative humidity) and first fed ad libitum with sugar water and pollen. Each colony sample of workers was divided into six groups and assigned to one of the treatments, i.e., control, low-LPS (0.1 mg/ml), or high-LPS (0.5 mg/ml). Because no dose effect was observed for beads, only one concentration for beads was used (10⁵ beads/ml); combinations were low-LPS + beads (0.1 mg/ml; 10⁵ beads/ml) and high-LPS + beads (0.5 mg/ml; 10⁵ beads/ml). After injection, workers were fed with sugar water for 1 hour before the starvation protocol; i.e., all feeding stopped after this period. For each individual, the time from starvation to death was recorded. Animals that died during the initial 1-hour feeding period were discarded from the data set, on the assumption that this was due to non–treatment-related manipulation of the insects.

Cox regression allows the comparison of entire survival curves against an assigned reference survival function for different variables simultaneously. The best statistical model was found by a backward stepwise procedure. Treatment, colony (i.e., where the worker came from), and the interactions between these factors were entered into the full model. The treatments were coded as categorical variables (0, 1, 2) for LPS absent, low-LPS, and high-LPS, respectively, and with a second variable (0, 1) for beads being absent or present, respectively.

Casteels-Josson K., Zhang W., Capaci T., Casteels P., J. Biol. Chem. 269, 28569 (1994).

Supplemental Web material is available at Science Online at www.sciencemag.org/feature/data/1052093.shl.

Supported by a grant from the Swiss National Science Foundation (number 3100-049040.95 to P.S.H.). We thank M. Brown, J. Jokela, P. Mutikainen, C. Reber, M. Rigby, R. Schmid-Hempel, and J. Wiehn for comments.