Comparative transcriptome analysis of the antenna and proboscis reveals feeding state-dependent chemosensory genes in Eupeodes corollae

Abstract

The physiological state of an insect can affect its olfactory system. However, the molecular mechanism underlying the effect of nutrition-dependent states on odour-guided behaviours in hoverflies remains unclear. In this study, comparative transcriptome analysis of the antenna and proboscis from Eupeodes corollae under different feeding states was conducted. Compared with the previously published antennal transcriptome, a total of 32 novel chemosensory genes were identified, including 4 ionotropic receptors, 17 gustatory receptors, 9 odorant binding proteins and 2 chemosensory proteins. Analysis of differences in gene expression between different feeding states in male and female antennae and proboscises revealed that the expression levels of chemosensory genes were impacted by feeding state. For instance, the expression levels of EcorOBP19 in female antennae, EcorOBP6 in female proboscis, and EcorOR6, EcorOR14, EcorIR5 and EcorIR84a in male antennae were significantly upregulated after feeding. On the other hand, the expression levels of EcorCSP7 in male proboscis and EcorOR40 in male antennae were significantly downregulated. These findings suggest that nutritional state plays a role in the adaptation of hoverflies' olfactory system to food availability. Overall, our study provides important insights into the plasticity and adaptation of chemosensory systems in hoverflies.

Keywords: Eupeodes corollae; chemosensory-related genes; feeding state; mRNA sequencing; olfactory plasticity.

Figure 1.

Phylogenetic tree and heatmap of ORs. (a) Phylogenetic tree constructed using OR protein sequences from E. corollae (blue), E. balteatus (black), D. melanogaster (red), B. dorsalis (green), C. stygia (grey) and M. domestica (brown). The phylogenetic tree was rooted using Orco orthologues, and bootstrap values are shown. (b) Heatmap of OR gene expression in the antenna and proboscis. The expression values are represented as mean values, and the data were normalized as follows: log₁₀(FPKM + 0.001). FAC, female antennae from control; FAT, female antennae from treatment; FPC, female proboscis from control; FPT, female proboscis from treatment; MAC, male antennae from control; MAT, male antennae from treatment; MPC, male proboscis from control; MPT, male proboscis from treatment. Control is E. corollae fed with ddH₂O, and treatment is E. corollae fed with 10% honey and pollen.

Figure 2.

Volcano plots of the chemosensory DEGs in the antenna and proboscis under different feeding states. (a–d) DEGs between feeding and starvation treatments in female antennae (a), female proboscises (b), male antennae (c) and male proboscises (d). Grey dots represent genes that did not meet the FDR threshold of 0.05, and red and green dots represent genes that were significantly up- and downregulated, respectively, in the different treatment groups.

Figure 3.

Phylogenetic tree of GRs and heatmap of GR expression. (a) Phylogenetic tree constructed using GR protein sequences from E. corollae (blue), E. balteatus (black), D. melanogaster (red), B. dorsalis (green), C. stygia (grey) and M. domestica (brown). The phylogenetic tree was rooted using carbon dioxide GR orthologues, and bootstrap values are shown on the left. (b) Heatmap of GR gene expression levels in the antenna and proboscis. The expression values are represented as mean values, and the data were normalized as follows: log₁₀(FPKM + 0.001).

Figure 4.

Phylogenetic tree of IRs and heatmap of IR expression. (a) Phylogenetic tree constructed using IR protein sequences from E. corollae (blue), E. balteatus (black), D. melanogaster (red), C. stygia (grey) and A. gambiae (green). The IR phylogenetic tree was rooted using IR8a/IR25a orthologues, and bootstrap values are shown on the left. (b) Heatmap of IR gene expression levels in the antenna and proboscis. The expression values are represented as mean values, and the data were normalized as follows: log₁₀(FPKM + 0.001).

Figure 5.

Phylogenetic tree of OBPs and heatmap of OBP gene expression. (a) Phylogenetic tree of OBPs constructed using proteins from E. corollae (blue), E. balteatus (black), D. melanogaster (red), B. dorsalis (green), C. stygia (grey) and M. domestica (brown). The tree was rooted using Lush-OBP orthologues, and bootstrap values are shown on the left. (b) Heatmap of OBP gene expression levels in the antenna and proboscis. The expression values are represented as mean values, and the data were normalized as follows: log₁₀(FPKM + 0.01).

Figure 6.

Phylogenetic tree of CSPs and heatmap of CSP expression. (a) Phylogenetic tree constructed using CSPs from E. corollae (blue), E. balteatus (black), D. melanogaster (red), C. stygia (grey) and A. gambiae (green). Bootstrap values are shown on the left. (b) Heatmap of CSP gene expression levels in the antenna and proboscis. The expression values are represented as mean values, and the data were normalized as follows: log₁₀(FPKM + 0.01).

Figure 7.

Analysis of differentially expressed chemosensory genes under different feeding states in different tissues of E. corollae. (a–d) The expression levels of chemosensory-related genes in female antennae (a), female proboscises (b), male antennae (c) and male proboscises (d) between the feeding and starvation groups. Data are shown as mean ± s.e.m. (n = 3). The asterisks indicate statistically significant differences (independent samples t-test, *p < 0.05; **p < 0.01, ***p < 0.001, ^#p < 0.0001), and n.s. indicates no significant difference.