Exploring gene expression as a sublethal endpoint in gammarids exposed to pesticides. insights from next-generation sequencing

Abstract

Pesticide residues are frequently detected in surface waters, with several compounds known to adversely affect aquatic organisms. Gammarids are particularly suitable indicator organisms for assessing the sublethal effects of such contaminants due to their high sensitivity and their central ecological role in freshwater ecosystems. While behavioral endpoints and feeding rates have been commonly used to evaluate sublethal pesticide effects, gene expression changes have received comparatively little attention, despite their proven value in other ecotoxicological contexts. This study investigates the potential of gene expression as a sensitive sublethal endpoint in gammarids collected from natural populations. A laboratory exposure experiment was conducted using the model pesticides azoxystrobin and acetamiprid, both of which are regularly detected in surface waters. Gammarids collected from the wild were exposed under controlled conditions to sublethal concentrations of the test substances. Subsequently, RNA sequencing (RNA-seq) was performed to characterize genome-wide transcriptional responses. Two independent exposure and sequencing experiments were carried out, resulting in the identification of 145 and 326 differentially expressed transcripts per experiment when comparing exposed animals to controls. Gene ontology (GO) term enrichment analyses revealed significant effects on metabolic processes, cell proliferation, and cell differentiation. Notably, the two experimental runs yielded distinct transcriptional profiles, with minimal overlap in differentially expressed transcripts despite the use of gammarids from the same population and the short interval (12 days) between experiments. The study demonstrates the applicability of transcriptomic analyses for detecting sublethal pesticide effects in field-collected gammarids and provides a practical workflow for the evaluation of RNA-seq data in non-model organisms. At the same time, it highlights important limitations, including high genetic variability within wild populations and incomplete transcriptome annotation, which together contribute to inconsistencies across repeated experiments.