Fig. 1: Chemical signalling, immune gene expression and infection load of ant pupae.

a Infected worker pupae (I + , blue) showed upregulated relative abundance of cuticular hydrocarbons tritriacontadiene (C33:2) and tritriacontene (C33:1) under worker presence (W + ; dark tone) only. Their chemical profile differed from infected pupae in the absence of workers (W-; pale tone; LMM, pairwise-post hoc corrected, two-sided p-values, Supplementary Table S2: p = 0.0001), and control pupae with (I-W + ; dark red; p = 0.012) and without workers (I-W-; pale red; p = 0.002; significance groups denoted as a, b). b Queen pupae showed no difference between the four treatment groups (infection green, sham purple; LMM, overall model uncorrected, two-sided p = 0.434, n.s., Supplementary Table S5). Pupal immune gene expression (combined analysis of the three candidate genes BGBP, PGRP-SC2 and Def1), in contrast, was increased in both worker pupae (c) and queen pupae (d) only in response to infection (LM; main effects uncorrected, two-sided p-values worker pupae: p = 2.2 × 10⁻¹⁶, queen pupae: p = 0.0003, depicted as***; Supplementary Tables S3b, S8b), but not to worker presence (worker pupae: p = 0.467, queen pupae: p = 0.338). Pupal infection load (given as fold-change to the maximum caste-specific exposure dose) of (e) worker pupae increased steadily over time (LM, pairwise-post hoc corrected, two-sided p-values, early-mid: p = 0.016, mid-late: p = 0.0007, early-late: p = 7.2 × 10⁻⁹; Supplementary Table S9a), whilst in queen pupae (f) infection was reduced back to early infection levels over the course of the experiment (LM; pairwise-post hoc corrected, two-sided p-values, early-mid: p = 0.016, mid-late: p = 0.006, early-late: p = 0.180; Supplementary Table S9b, significance groups denoted as a–c). Graphs show means ± sem as dots and whiskers in shaded 95% CI box, with data log₁₀(x + 1)-transformed in (c–f). Chemical analysis based on 426 pupae, with the subset of 265 infected pupae analysed for infection load; immune gene expression based on 179 pupae. See Supplementary Fig. S2 (Supplementary Tables S1a,S4a,S6) for individual CHCs, and Supplementary Fig. S3 (Supplementary Tables S3a,S8a) for individual immune genes. Source data (1a–f) are provided as a Source Data file, raw data under https://doi.org/10.15479/AT-ISTA-20471.