Thermal preferences of infected lizards and abundance of vectors might explain a high blood parasite prevalence
Description
Ectotherm physiology operates optimally within a thermal range. However, when ectotherms are infected, they can respond with subtle, yet significant, adjustments in their thermal behavior. This behavioral plasticity acts at the expense of reducing the pathogen’s fitness. A previous study on males of a Mediterranean lizard, Psammodromus algirus, found that lizards responded with behavioral hypothermia to the inoculation of a bacterial antigen. We experimentally assessed whether males of the same lizard species can have a similar reaction norm against natural infections. To this end, half of a 37-adult male sample was assigned to a deparasitization treatment that significantly reduced the number of hematophagous mites compared to a control group and quantified the parasitemia of Karyolysus and Schellackia, two genera of blood parasites transmitted by hematophagous mites. To test for a reaction norm in response to these parasites, we compared the thermal behavior of the lizards before and after the deparasitization. We found a significant interaction between the parasitemia of Karyolysus and the deparasitization treatment that suggested lizards in the deparasitized group selected higher temperatures in response to greater blood parasite load (behavioral fever), whereas in the control group, individuals with high blood parasite loads tended to select lower temperatures (behavioral hypothermia). The relatively cooler thermal preference of lizards with both high mite loads and high Karyolysus parasitemia combined with the fact that lizards are subject of heat stress during seasonal peaks of mite abundance can explain the high prevalence of Karyolysus in the studied population.
Files
Steps to reproduce
2.8.1 Experimental manipulation of mite load To assess the efficacy of the experimental ectoparasite treatment, a repeated-measures ANOVA was performed using Statistica 12.0 (StatSoft Inc., Tulsa, OK, USA) comparing mite counts before and after treatment (log initial_mites and log final_mites). Treatment type (two levels: Control and “Deparasitized”) and sampling batch with four levels were included as factors. Mite counts were log-transformed (Megía-Palma et al., 2025). A Bonferroni post-hoc test was used to determine whether significant differences in mite abundance existed between the resulting groups. 2.8.2 Change in lizard thermal behavior Using four repeated-measures ANOVAs, we compared initial (day 1, before the deparasitization treatment) and final (day 10, after the deparasitization experiment) values of the central tendency (Median_temp) and precision (SD_temp) of selected temperatures, as well as the maximum (temp_max) and minimum (temp_min) temperature for each trial. All models included date (four levels, corresponding to the capture batch), treatment (two levels: Control and “Deparasitized”), and the covariates representing the percentage of erythrocytes infected by the two genera of blood parasites (perc_Karyolysus and perc_Schellackia). Two-way interactions between treatment (Control vs Deparasitized) and the parasitemia of the two blood parasites (perc_Karyolysus and perc_Schellackia) were also included.
Institutions
- Universidad Complutense de MadridComunidad de Madrid, Madrid