Task-specific odorant receptor expression in worker antennae indicates that sensory filters regulate division of labor in ants

Description

Division of labor (DOL) is a characteristic trait of insect societies, where tasks are generally performed by specialized individuals. Inside workers focus on brood or nest care, while others take risks by foraging outside. Theory proposes that workers have different thresholds to perform certain tasks when confronted with task-related stimuli, leading to specialization and consequently DOL. Workers are presumed to vary in their response to task-related cues rather than in how they perceive such information. Here, we test the hypothesis that DOL instead stems from workers varying in their efficiency to detect stimuli of specific tasks. We use transcriptomics to measure mRNA expression levels in the antennae and brain of nurses and foragers of the ant Temnothorax longispinosus. We find seven times as many genes to be differentially expressed between behavioral phenotypes in the antennae compared to the brain. Moreover, nearly half of all odorant receptors are differentially expressed, with an overrepresentation of the 9-exon gene family upregulated in the antennae of nurses. Nurses and foragers thus apparently differ in the perception of their olfactory environment and task-related signals. Our study supports the hypothesis that antennal sensory filters predispose workers to specialize in specific tasks.

Steps to reproduce

A total of seven colonies of the ant Temnothorax longispinosus were selected with an average colony size of 110 ± 31.5 workers. Upon collection, we housed each colony in a plaster-floored nesting box divided into three chambers containing a single slide nest, in which the colony relocated. We performed behavioral observations and noted down how many times an individual performed brood care and foraging behavior and the position in the nest. Based on these behavioral observations, the marked individuals found outside the nest, exploring the surroundings for food or water, were identified as foragers. We identified nurses as workers that remain inside the nest in direct contact with brood and were unlikely to leave the nest. After all observations were completed, the marked nurses and foragers were collected. For RNA extraction, we dissected brain and antennae tissues from 48 nurses and 49 foragers. We pooled the brains and antennae from seven workers from each behavioral state and colony. Brain and antennae tissues were homogenized with a pestle. Sample brains were transferred separately to a 1.5 ml Eppendorf tube containing 50μl of TRIzol. We added 50μl chloroform to each brain and antenna samples, gently inverted for 30 s and then centrifuged samples at 12,000 g for 15 min at 4°C. We collected the resulting supernatant and precipitated RNA with 25μl 70% ethanol. We conducted the subsequent RNA extraction with the RNeasy Mini Kit (Qiagen), following the manufacturer’s instruction. The resulting 28 samples (14 brains and 14 antennae) were stored at -80°C until library preparation. RNA-seq libraries were prepared by Novogene Company Limited, Cambridge, UK, using the NEBNext Ultra RNA Library Prep Kit for Illumina . After amplification and purification, 28 libraries were sequenced on an Illumina NovaSeq 6000 S4 flow cell platform using a paired-end 150 bp. Raw data obtained from Novogene were checked using FastQC v.0.11.9, and Illumina adapters were removed using Trimmomatic v.0.36. For gene expression analysis, reads were mapped to our T. longispinosus genome assembly, and the read counts table was generated using STAR 2.7.0 with default settings. We used the deseq2 v1.16.1 package for R to identify differentially expressed genes. Then, we conducted a differential gene expression analysis with DESeq2. We began with comparisons between nurses and foragers using the ~Colony+Task model, followed by a likelihood ratio test (LRT) approach, with colony ID as a fixed factor. Genes were considered differentially expressed if the false discovery rate (FDR), using Benjamini-Hochberg procedure, had an adjusted p-value of ≤ 0.05. The resulting lists of DEG refer to genes that are overexpressed and underexpressed in foragers compared to nurses.

Institutions

Categories

Funding

Licence