Fig. 1. Timeline of experimental manipulations used for studies I, II and III of Haplochromis burtoni males. Three experiments (I, II and III) examined how social opportunity changes behaviors and gonadotropin-releasing hormone (GnRH) expression. These are shown schematically by the three types of aquaria used and the corresponding time courses of behavioral observations and physiological and molecular measures. T, territorial male; NT, non-territorial male; F, female. (A) Study I: undisturbed aquaria were used to identify males that maintained a stable social status of either T or NT for more than 2 weeks. Six NT males and five T males fulfilled this behavioral criterion. After 10 weeks of observation, fish were captured, and blood was sampled for measurement of serum cortisol levels. Gonad and body masses were obtained for calculating the gonosomatic index (GSI; see Materials and methods), used to confirm the behaviorally defined social state. Brains were processed for the detection of GnRH gene expression. DI, index of dominance (see Materials and methods). (B) Study II: to determine how fast GnRH gene expression changes during social ascent (NTT), half-tanks were prepared with distinct communities on either side of a Plexiglas barrier. A pair of NT males, matched for age and size, were housed in one half of an aquarium containing two T males and a few females. After 2 weeks of baseline behavioral observations, blood samples were obtained, and one member of each pair (NTT) was then moved to the opposite half of the tank, where there were only females. The other (control NT) was returned to the first side. Behavioral observations continued for 3 or 7 days, when blood was again sampled and NTT and control NT males were killed for analysis of GSI and GnRH mRNAs. (C) Study III: to examine changes in GnRH expression during both social ascent (NTT) and descent (TNT), males were observed for 3 weeks to ascertain their social state. Control males were killed, and experimental males were moved to new tanks containing communities designed to induce a change in their social status. Only males for whom behavioral observations confirmed that the intended social change had taken place were selected for measurement of GSI, GnRH mRNAs and preoptic immunoreactive GnRH neuronal soma sizes at the time points indicated. Asterisks indicate when serum cortisol levels were measured.

Return to article