Pollen-deprived bumblebees induce leaf damage, which activates phytohormone signaling pathway(s) and response(s) within tomato plants, but this does not accelerate or induce flowering

Pollinators and plants share complex and mutually beneficial relationships critical to ecological sustainability. The timing of some interactions is crucial; if pollinator populations are not synchronized with the blooming periods of flowers, plants may fail to reproduce, or pollinators may face starvation. Environmental challenges such as pollen scarcity can disrupt this delicate balance, prompting adaptive behaviors in pollinators. A recent study shows that pollen-deprived bumblebees (Bombus terrestris) bite tomato plants, and this accelerates flowering. We have replicated this bee behaviour, but are unable to replicate the plant response at a macro-level. As insect saliva facilitates digestion, maintains mouthparts, provides antimicrobial protection, and is able to bypass plant defences we are interested to determine whether pollinator saliva may play a role in mediating plant interactions and whether this can be used to identify molecular mechanisms that drive adaptive behaviors associated with plant--pollinator co-evolution. To unravel the association, we detail the salivary gland proteome of B. terrestris, analyze plant miRNAs and protein markers in the bee gut, use Florigen to determine whether the plants were already committed to flowering, analyze phytohormones in bee-bitten leaves, and establish a Raman microscopy system. Our data suggest that the plants are responding to these bee bites at a micro-level. To aid further research, we highlight the utility of Attenuated Total Reflectance–Fourier Transform Infrared (ATR-FTIR) spectroscopy integrated with multivariate analyses to understand insect secretions through the differences in chemical composition of tissues (salivary gland, proboscis, and pretarsus) from two species of bumble bee (B. terrestris and B. hortorum). This analytical method required minimal sample volume and provided detailed spectral fingerprints for analysis. Unsupervised Principal Component Analysis revealed natural clustering in the dataset, while supervised Partial Least Squares Discriminant Analysis further classified the data based on known classes. Notably, chemical profiles remained conserved under nutritional stress.