Sex and tissue differences in virus susceptibility across species 1

Abstract


32
Emerging pathogens often arise from a host shift event -where a pathogen jumps into and 33 establishes in a novel host species. Species vary in their susceptibility to pathogens, but little 34 is known about the factors underlying these differences, and whether differences between and females, as typically only one sex is utilized to remove between sex differences [13].

48
Sexual dimorphism is seen across most animal systems in a range of life history traits from 49 body size, growth rate, reproductive effort and immunity [14][15][16]. In mammals, males and 50 females often differ in their pathogen burdens and mortality rates [17]. For example, in 51 SARS-COV-2 infection in humans, women have a lower risk of morbidity and mortality than 52 men [18]. In HIV infected individuals, women have up to 40% lower HIV viral RNA in 53 circulation but a greater likelihood of developing AIDS than men with matched viral loads 54 [19]. Sex biases in parasitism rates n mammals have been suggested to be due to males  89 and show greater rates of shedding, lower clearance and higher transmission potential of 90 . CC-BY 4.0 International license perpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in

100
Thirty one species of Drosophilidae were used to examine sex differences in viral infection.

101
Stock populations were reared in the laboratory in multi generation populations, in 102 Drosophila stock bottles (Fisherbrand) on 50 ml of their respective food medium (Table S1) 103 at 22˚C and 70% relative humidity with a 12-hour light-dark cycle. Flies were then collected 104 twice a day in order to try and control for age of maturity in an effort to minimize the 105 chances that flies would have reached sexual maturity and mated before the sexes were 106 separated out. Although no effect of mating status on DCV viral load was previously

178
Inferring the host phylogeny 179 . CC-BY 4.0 International license perpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in

219
The models took the form of:

250
The proportion of between strain variance that can be explained by the phylogeny was 251 calculated from model (1) using the equation CC-BY 4.0 International license perpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in

281
The mean change in viral load across all species was similar between the sexes (females =   . CC-BY 4.0 International license perpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for this this version posted November 2, 2022. ; https://doi.org/10.1101/2022.11.01.514663 doi: bioRxiv preprint Roberts and Longdon Sex and tissues differences in virus susceptibility 8 The full model including the species-specific random effect independent of the host 292 phylogeny ( (:!# ) allowed us to calculate the proportion of the variation between the 293 species that can be explained by the phylogeny ( ), equivalent to phylogenetic 294 heritability or Pagel's lambda [36,37]. The host phylogeny explains a large proportion of the 295 inter-specific variation for both males and females (females = 0.68, 95% CIs: 0.06, 0.99; 296 males = 0.66, 95% CIs: 0.04, 0.99) consistent with previous findings for males [3,5,9,25].

297
However, we note these estimates have broad confidence intervals, due to the model 298 struggling to separate out the phylogenetic and non-phylogenetic components. The

313
. CC-BY 4.0 International license perpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for this this version posted November 2, 2022. ; https://doi.org/10.1101/2022.11.01.514663 doi: bioRxiv preprint

322
We found that viral susceptibility between females and males of 31 host species showed a 323 strong positive correlation with a close to 1:1 relationship, suggesting that susceptibility 324 across species is not sex specific. We also found differences in viral load between tissues of . CC-BY 4.0 International license perpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in

343
In Drosophila melanogaster, some previous studies have reported males have higher DCV 344 viral loads than females [26]. However, others found no effect of sex on viral load, but did 345 find effects of sex on viral shedding, clearance, and transmission potential, with these traits

346
showing interactions with host genotype [27]. Sexual dimorphism in infection avoidance 347 behaviour has also been reported, when female flies previously exposed to DCV were found

353
The tissue tropism results here show that susceptibility in a given host is general across 354 tissue types -for example D. sturtevanti has a high viral load across all tissues whereas D.

358
This does not exclude the possibility that pathology is due to high viral loads in a given 359 tissue, but does suggest that the mechanism restricting viral load is general across tissues.

360
This may be linked to the ability of the virus to bind to or enter hosts cells, utilise the hosts 361 cellular components for replication or to avoid or supress the host immune response [

372
In summary, our results demonstrate that in this system there is little evidence for sexual 373 dimorphism in susceptibility to viral infection across species. As such susceptibility in one sex 374 is predictive of that in the other. We find that susceptibility of a species is general across 375 . CC-BY 4.0 International license perpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for this this version posted November 2, 2022. ; https://doi.org/10.1101/2022.11.01.514663 doi: bioRxiv preprint a p tissue types, suggesting virulence is not due to species specific differences in viral tropism.

376
Further work is needed to explore how sex differences can vary with the environment and 377 pathogen type, and the underlying mechanisms as to why species vary in their susceptibility.

382
Thanks to group members for discussions of these data. For the purpose of Open Access, the 383 author has applied a CCBY public copyright licence to any Author Accepted Manuscript

423
. CC-BY 4.0 International license perpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in

473
. CC-BY 4.0 International license perpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for this this version posted November 2, 2022. ; https://doi.org/10.1101/2022.11.01.514663 doi: bioRxiv preprint