Expression of insulin pathway genes during the period of caste determination in the honey bee, Apis mellifera

Abstract

Female honeybees have two castes, queens and workers. Developmental fate is determined by larval diet. Coding sequences made available through the Honey Bee Genome Sequencing Consortium allow for a pathway-based approach to understanding caste determination. We examined the expression of several genes of the insulin signalling pathway, which is central to regulation of growth based on nutrition. We found one insulin-like peptide expressed at very high levels in queen but not worker larvae. Also, the gene for an insulin receptor was expressed at higher levels in queen larvae during the 2nd larval instar. These results demonstrate that the insulin pathway is a compelling candidate for pursing the relationship between diet and downstream signals involved in caste determination and differentiation.

Figure 1

Experimental design for manipulation of larval diets. Larvae 16 hrs post hatching were either grafted into queen cells or remained in worker cells. At 40 and 64 hours, larvae that had spent 1 or 2 days in queen or worker cells, respectively, were transferred to the alternate food type. The design yielded 12 treatments that varied by age and diet history. QQQ and WWW are normally developing queen and workers.

Figure 2

Diets by day for female larval bees in worker and queens cells. Nurse workers produce different larval diets through different mixtures of glandular secretions, crop contents and, in older worker larvae, pollen. A white component is produced by the mandibular glands and a clear component is a mixture of hypopharyngeal gland secretions and crop contents. The royal jelly (RJ) placed in queen cells is made up of equal amounts of the white and clear components. Nurse workers provision developing workers during their first 2 days with a mixture of 20–40% white secretion and the remainder the clear substance. Therefore, young worker jelly (YWJ) can be nearly as rich as royal jelly. On the third day, worker food becomes almost entirely clear and pollen is added, making worker jelly (WJ) (Jung-Hoffman, 1966 in Michener, 1974).

Figure 3

Expression of AmILP-1, an insulin-like peptide that is highly expressed in queen larvae. Standard errors are shown only for queen and worker larvae for ease of viewing.

Figure 4

Expression of AmILP-2, a second insulin-like peptide. Standard errors are shown only for queen and worker larvae for ease of viewing.

Figure 5

Expression of an Am insulin/IGF receptor protein, AmInR-2. Expression is high in young grafting age larvae and reaches even higher levels in queen-destined larvae at 40 hours. Standard errors are shown only for queen and worker larvae for ease of viewing.

Figure 6

Expression of Am insulin receptor substrate, AmIRS. Differences among treatments were not significant. Standard errors are shown for WWQ, which showed a statistically non-significant increase, in addition to those for queen and worker larvae.

Figure 7

Expression levels of AmPTEN. Expression in queen and worker larvae was the same. Only the transfer of queen larvae to worker cells during the 3rd instar caused a significant change in expression. Standard errors are shown for QQW, which showed a statistically significant increase, in addition to those for queen and worker larvae.