Gene expression underlying differential adult brain morphogenesis in honeybee castes

Learning and memory-related skills that honeybees use for navigation, foraging and other activities are associated with a central region of the brain, the mushroom bodies, which are relatively more developed in workers than in queens. During larval period, however, the differential feeding offered to prospective queens promotes faster brain development and higher expression of several neurogenic genes (ataxin-2, cryptocephal, dachshund, Eph Receptor, fax, shot, krüppel homolog-1 and tetraspanin 5D). It seems that in some point during pupation there happens a shift in this trend. In fact, queen’s brain experiences net cell death rates while worker’s brain is favored by higher rates of cell proliferation, resulting in caste specific brains. Here we report on transcriptomic results which might represent the molecular underpinnings of the differential brain development between castes. Genome-wide expression analyses using oligonucleotide microarray approach show an opposite pattern to that observed during larval development, with workers’ brain with higher transcription levels of 324 genes (e.g., mesencephalic astrocyte derived neurotrophic factor, minibrain, signal peptide peptidase, and tumbleweed, all associated to neurogenic events or cell death prevention). This suggests that somehow the respective gene products promote differential development of honeybee brain. MANF, for example, a gene superexpressed in workers’ brain, encodes a protein with a domain homologous to SAP Ku70 C-terminal domain. Since this molecule is an inhibitor of apoptotic protein Bax, MANF is a candidate to act as an anti-apoptotic factor during worker brain development (extensive cell death events characterize queens’ pupal brain). Minibrain encodes a protein-kinase involved in regulating cell division during post-embryonic neurogenesis, and is another candidate to participate in the mechanism responsible for the reversing the brain size/body volume rate between queens and workers. Moreover, we assessed by RT-qPCR the transcription profile of the ecdysone receptor (which mediates ecdysteroid action and is probably involved in differential brain cell death/proliferation between castes) variants A and B in three time points during pupal development. Our results suggest the existence of a hormone/receptor threshold above which (hormone excess), in queens, it would be triggered more cell death than proliferation events, which through the differential participation of effector genes, would lead to the morphological differences between adult queens’ and workers’ brain. RNA samples from pools of brains from pharate-adult queens and workers were hybridized in pairs and in dye-swaps (Cy3/Cy5), twice, then, combined in silico.