Identification of genomic regions implicated in susceptibility to Schistosoma mansoni infection in a murine genetic model (backcross)

High levels of infection and severe liver fibrosis in schistosomiasis appear only in a few cases of infected people with high susceptibility. Tissue damage is caused by the inflammatory response to eggs trapped in the liver. The genetic background influences susceptibility to schistosome infection. To assess the genetic basis of susceptibility to schistosomiasis and identify the chromosomic regions involved, we used a backcross strategy to generate a mouse cohort with high variation in schistosomiasis susceptibility. Thus, we crossed the resistant C57BL/6 mouse strain with the susceptible CBA one; and later; the F1 females (C57 x CBA) were backcrossed with CBA males generating the F1BX cohort. The spectrum of phenotypes in the F1BX mice showed gradation from mild to severe disease, lacking a fully resistant group. We differentiated four levels of infection intensity using cluster and principal component analyses and K-means based on parasitological, pathological and immunological trait measurements. Mice were massively genotyped with 961 informative SNPs. We identify 19 new quantitative trait loci (QTL) associated with phenotype indicators of parasite burden, liver lesion, white blood cell populations, and antibody responses evaluated in the backcross cohort. Two QTLs on chromosomes 15 and 18 were simultaneously linked to the number of granulomas, grade of liver lesion and IgM levels. The human syntenic regions are located in chromosomes 8 and 18. None of the significant QTL identified coincided with previously reported mice association or were syntenic with human chromosomes. Author summary High number of cases of infection and high levels of infection and liver fibrosis in schistosomiasis appear only in a few cases of people with high susceptibility, because the genetic background influences this susceptibility. To assess this, we used a backcross strategy crossing a resistant strain of mouse C57BL/6 with another susceptible CBA, and later F1 females were backcrossed with males susceptible, generating the F1BX cohort, which showed gradation from mild to severe disease. We differentiated four levels of infection according to cluster, principal component analysis and K-means, and based on parasitological, pathological and immunological measurements. Mice were massively genotyped and 19 quantitative trait loci (QTL) were identified, associated with phenotype indicators evaluated in backcross cohort. Two QTLs on chromosomes 15 and 18 were simultaneously linked to the number of granulomas, grade of liver lesion and IgM levels. None of the significant QTL identified coincided with previously reported mice association or were syntenic with human chromosomes.


58
Schistosomiasis is a widespread parasitic disease caused by trematodes of the genus 59 Schistosoma in tropical areas. It is estimated that more than 200 million people worldwide are 60 infected with Schistosoma spp., which caused more than 1.4 million DALYs (Disability and Senegal identified the SM1 (S. mansoni 1) locus linked to high prevalence and high 68 infection burden in particular family groups [4,5]. The SM1 susceptibility locus is localized in 69 the long arm of human chromosome 5 (5q31-33). Another locus associated with severe 70 schistosomiasis is called SM2 (S. mansoni 2), related to the development of liver fibrosis and 71 located on the long arm of human chromosome 6 (6q22-23) [6,7]. Despite the reports 72 suggesting genomic regions associated with high intensity of the infection and severe liver 73 damage no compressive information about the genomic regions involved in these processes. 74 Knowledge of molecular markers linked to susceptibility could be helpful to develop 75 protocols for diagnosis and treatment against severe schistosomiasis and enable efficient 76 control measures to prevent infection in more susceptible populations. 77 Identifying genetic markers of complex diseases, such as schistosomiasis is difficult in 78 the human population. It requires population studies with many cases and controls and a great 79 expense of time and resources without guaranteeing success. However, syngeneic mouse 80 strains are genetically and phenotypically homogenous in schistosomiasis susceptibility and 81 other traits [8].Also, crosses between syngeneic mouse strains generate cohorts of individuals 82 with a tremendous variation in infection susceptibility, which allows the identification of 83 genetic regions, such as QTLs (Quantitative Trait Loci) and expression QTLs (eQTLs), 84 associated with any quantitative trait variation; thus, involved in the resistance or 85 susceptibility to schistosomiasis [9]. Identifying these genetic loci linked to the disease 86 susceptibility will help elucidate the molecular basis of this complex disease [10,11]. Indeed, 87 linkage studies with SNPs (single nucleotide polymorphisms) have identified chromosomal 88 regions associated with the genetic influence of schistosomiasis and its wide variability in 89 severity. It has been possible to delimit genetic regions involved both in the intensity of the 90 schistosome infection and in the degree of liver fibrosis induced in an individual [12][13][14]. In 91 mice, laboratory infections with S. mansoni lead to granulomatous lesions in the intestine and 92 liver. We carried out, a preliminary study with five syngeneic mice strains (C57BL/6J, 93 FVB/NH, DBA/2J, BALB/c and CBA/2J) and we identified the two most divergent strains in 94 terms of susceptibility to infection, which were CBA/2J more susceptible and C57BL/6J less. 95 The susceptibility to infection in the F1B6CBA hybrid mice was between these two parental 96 strains yielding intermediate susceptibility [15]. The strain differences in the response to 97 S. mansoni also offer the chance to study the genetic factors influencing whether an animal 98 can develop intense or mild worm recovery or lesions to this infection. The evidence of 99 differences between more and less susceptible strains led us to adopt a strategy of 100 backcrossing of C57BL/6J onto the CBA background. This approach should generate mice 101 with high phenotypic variability between the two strains.

102
The general objective of this work was to identify chromosomal regions that carry low 103 penetrance or modifiers genes associated with the susceptibility to infection by S. mansoni, by 104 a backcross murine genetic model, with the following specific objectives: characterize the 105 susceptibility to infection by S. mansoni defining different disease patterns; to study the 106 association between the pathophenotypes and the related parasitological, pathological and 107 immunological traits and to identify the genomic regions (QTL) linked to the intermediate 108 phenotypes defined after infection with S. mansoni.

110
There was a different susceptibility to schistosomiasis in mice related to the genetic 111 background but not to mouse sex 112 We had previously studied the susceptibility/resistance to Schistosoma mansoni four levels from mild (Group 1) to severe (Group 4).

190
Pathophenotypic variation in the F1BX cohort in the clusters identified 191 We evaluated the distribution of each schistosomiasis pathophenotypes and intermediate Tukey test was made between every two groups.

213
Further, we studied medians in a heat chart where the maximum median of each 214 variable within the four groups was taken as a reference ( Figure 4). Data showed that Group 1 215 gathered mice with the lowest worm recovery, eggs in tissues and liver damage, thus,  group has respect to the maximum, which indicates the variation experienced throughout 231 the groups. The color represented when the difference between medium is low, it is red. 232 Conversely, when the difference between medians is greater, the color represented is 233 bluer.
For cell populations, no significant differences were observed between the four groups 235 studied for the CD4 and CD8 populations, both in splenocytes or peripheral blood 236 (supplementary Figure 6 A and B). Group 4 with the highest liver damage and eggs trapped in 237 tissues, showed the lowest percentages in B220 and CD45 splenocytes (Supplementary Figure   238 6A). These differences were statistically significant in comparison with Group 1 and 2, and 239 median analysis is concordance (Figure 4). In contrast, peripheral blood B220 and CD45 had 240 no significant differences (Supplementary Figure 6B). Finally, we only found differences in 241 total IgG antibodies, Group 3 and 4 showed more absorbance than Group 1 (Supplementary 242 Figure 6C) with the highest medians ( Figure 4). In the other hand, we did not find significant 243 differences in IgG1, IgG2a or IgM levels between groups.     in the four groups of mice, according to the severity of the disease.

313
Only a minority of chronically exposed patients develops the most severe clinical form 314 of schistosomiasis, severe hepatosplenic disease, periportal fibrosis and portal hypertension. with chromosome 18 by only one SNP with a high LOD score linked to granulomas and liver 378 damage. The QTL3 and QTL4 were only associated with affected liver surface and IgM 379 production and was located in chromosome 2 and 17. The QTL5 and QTL6 linked to 380 chromosomes 6 and 8 were related to worm recovery and eggs in tissues and IgG2a.

381
Next, we looked for the syntenic regions of the QTLs identified by the Ensembl to study 382 the homology between mice and humans. We chose for this study the QTL1 related to liver fine mapping to propose reliable candidate genes influencing the defence response. 409 We can conclude that a F1BX cohort has been generated through backcross, with significant  Mean and standard error of the mean were determined in each variable and normal 507 distribution was checked with the Kolmogorov-Smirnov test. ANOVA followed by post-hoc 508 Tukey' test or Student t-test were employed to identify differences between groups. The

509
Pearson correlation coefficient (r) was used to study the dependence of two variables and the 510 statistical significance was performed using the Student t-test.

511
Multivariant models were generated to study mice's similarities and consider sex influence in 512 worm recovery, eggs in the liver and intestine, fecundity, granulomas and the affected liver