Expression analysis of vitellogenins in the workers of the red imported fire ant (Solenopsis invicta)

Abstract

Vitellogenin has been proposed to regulate division of labor and social organization in social insects. The red imported fire ant (Solenopsis invicta) harbors four distinct, adjacent vitellogenin genes (Vg1, Vg2, Vg3, and Vg4). Contrary to honey bees that have a single Vg ortholog as well as potentially fertile nurses, and to other ant species that lay trophic eggs, S. invicta workers completely lack ovaries or the ability to lay eggs. This provides a unique model to investigate whether Vg duplication in S. invicta was followed by subfunctionalization to acquire non-reproductive functions and whether Vg was co-opted to regulate behavior within the worker caste. To investigate these questions, we compared the expression patterns of S. invicta Vg genes among workers from different morphological subcastes or performing different tasks. RT-qPCRs revealed higher relative expression of Vg1 in major workers compared to both medium and minor workers, and of Vg2 in major workers when compared to minor workers. Relative expression of Vg1 was also higher in carbohydrate foragers when compared to nurses and protein foragers. By contrast, the level of expression of Vg2, Vg3, and Vg4 were not significantly different among the workers performing the specific tasks. Additionally, we analyzed the relationship between the expression of the Vg genes and S-hydroprene, a juvenile hormone analog. No changes in Vg expression were recorded in workers 12 h after application of the analog. Our results suggest that in S. invicta the Vg gene underwent subfunctionalization after duplication to new functions based on the expression bias observed in these data. This may suggest an alternative and still unknown function for Vg in the workers that needs to be investigated further.

Keywords: Ants; Gene expression; Juvenile hormone; Red imported fire ant; Social insect; Task allocation; Vitellogenin; Worker caste.

Figure 1. Comparison of the head width among individuals from the worker caste of a single colony (A) Minor, (B) Medium, (C) Major.

Photography credit: Chloe Hawkings.

Figure 2. Structural domains identified in the S. invicta predicted Vg proteins, (A) Vg1, (B) Vg2, (C) Vg3.

Each protein encoded a putative cleaved signal peptide (black). These proteins possess the LPD-N (gray), DUF1943 (horizontal stripes), and VWD (vertical stripes) domains commonly found in Vg proteins. While Vg2 and Vg3 possessed one of each domain, Vg1 possessed two. Therefore, Vg1 might encode Vg1 and Vg4.

Figure 3. Expression analysis of the vitellogenin transcripts among the worker morphological subcastes.

Each bar represents the mean ± SEM (n = 5). All workers were carbohydrate foragers. (A) Vg1 transcript expression. (B) Vg2 transcript expression. (C) Vg3 transcript expression. (D) Vg4 transcript expression. Vg mRNA expression level was normalized relative to RP18 mRNA expression level. Statistical relationships between groups were assessed using one-way ANOVA with Tukey–Kramer post hoc test (p < 0.05), where different letters indicate statistical differences among the subcastes.

Figure 4. Expression analysis of the vitellogenin transcripts among medium workers performing different tasks.

Each bar represents the mean ± SEM (n = 5). (A) Vg1 transcript expression. (B) Vg2 transcript expression. (C) Vg3 transcript expression. (D) Vg4 transcript expression. Vg mRNA expression level was normalized relative to RP18 mRNA expression level. Statistical relationships between groups were assessed using one-way ANOVA with Tukey–Kramer post hoc test (p < 0.05), where different letters indicate statistical differences among the different task-allocated insects.

Figure 5. Expression analysis of the vitellogenin transcripts among major workers 12 h after topical application of S-hydroprene.

Each bar represents the mean ± SEM (n = 6). (A) Vg1 transcript expression. (B) Vg2 transcript expression. (C) Vg3 transcript expression. (D) Vg4 transcript expression. Vg mRNA expression level was normalized relative to RP18 mRNA expression level. Statistical relationships between groups were assessed using one-way ANOVA; no significant differences were found for any of the Vg transcripts (p > 0.05).

Figure 6. Evaluation of S-hydroprene application and effect in alate virgin queens 12 h after treatment.

(A) Untreated alate queens, 0% of the tested virgins dealated. (B) Alate queen treated with acetone, 0% of the tested virgin dealated. (C) Alate queens treated with S-hydroprene, 100% of the tested virgins dealated. Photography credit: Chloe Hawkings.