Dissecting genetic and sex-specific host heterogeneity in pathogen transmission potential

Jonathon A. Siva-Jothy; Pedro F. Vale

The size of non-zero DCV loads reflects quantitative variation and was affected by 312 similar interactions between mating and sex with genetic background (Figure 2b ; 313 Table 4 ). While <1% of variance was explained by sex and mating, much more was 314 explained by genetic background (7.94% and 11%) alongside its interactions with 315 sex (19.2%) and mating (4.38%) ( Figure 5 ; Table 4 ). Table 5 ). Sex however, explained <1% of the deviance, while genetic 341 background and its interaction with sex explained 2.2% and 3.07% ( Figure 5 ). Mating 342 did not affect qualitative DCV shedding (Figure 3a ; Table 5 ) and explained <1% of 343 the deviance ( Figure 5 ; Table 5 ). In flies where DCV was detected in shedding, 344 quantitative DCV shedding was affected by genetic background and the extent of 345 this variation was determined by female mating status, but not sex ( Figure 3b ; Table 346 6). The amount of variance explained by sex and in our models was <1%, in 347 comparison with genetic background (9.48% and 5.82%) and its interactions with sex 348 (8.87%) or mating (6.53%) ( Figure 5 ; Table 6 ). Qualitative and quantitative DCV 349 shedding peaked at day 2 (Figures 3a; Tables 5 & 6, pairwise comparisons, 350 p<0.0001). Across all treatment groups, there was no significant relationship between 351 viral load and shedding ( Figure S1 ; Table 6 ). The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/733915 doi: bioRxiv preprint Table 7 ). Sex (0.28%), genetic 374 background (2.3%) and the interaction between the two (2.83%) explained relatively 375 little deviance in our models ( Figure 5 ; Table 7 ). In quantitative variation in V, sex 376 explained <1%, while genetic background and its interaction with sex explained 377 4.13% and 11.4% of variance respectively ( Figure 5 ; Table 8) . We also observed clear qualitative and quantitative differences in V between males and 473 females, which is suggestive of sex-specific variation in disease transmission. While the 474 extent of any difference between males and females is also determined by genetic 475 background, a greater proportion of males tend to transmit DCV than females across 476 these backgrounds. In DCV shedding, a greater proportion of males from several genetic 477 backgrounds (RAL-379, RAL-738 and RAL-818) shed DCV than females (Figure 3a) . Sex-specific variation in qualitative differences in shedding exerts a significant influence 489 over shedding (Figure 3a) . It is important to note however, that in isolation, sex accounts 490 for a miniscule 0.64% of the deviance in qualitative variation in shedding. Sex appears to 491 play a more important role in conjunction with genetic background, the interaction 492 between the two explaining 3.07% of deviance ( Figure 5 ). While significant, sex-specific 493 variation may play a relatively minor role in shedding. A variety of factors appear to 494 underlie sex-differences in shedding across host-pathogen systems. For example, male-495 biased infection is common to many mammal hosts but generally absent from arthropod 496 hosts (Sheridan et al., 2000) . In the water flea, Daphnia magna, parasite spores are 497 released into the environment upon death and females have been shown to release 498 significantly more than males (Thompson et al., 2017) . In the vole, Microtus gryalis, the 499 faeces of females contains significantly more parasite eggs than that of males (Sanchez 500