Categories: Immune system phene

Mendelian trait/disorder: yes

Mode of inheritance: Autosomal recessive

Disease-related: yes

Key variant known: no

Species-specific name: K88 scours

Species-specific symbol: FbcR

Species-specific description: Neonatal diarrhoea in piglets is often caused by strains of Escherichia coli bacteria that have a cell-surface antigen called K88 (renamed F4), which combines with a glycoprotein receptor on the wall of a piglet's intestine, enabling the bacteria to attach themselves to the intestine. Such E. coli are called Enterotoxigenic Escherichia coli (ETEC). Once attached, the ETEC proliferate, releasing enterotoxins and thus producing diarrhoea, which can lead to high mortality. Certain piglets lack the intestinal receptor to F4, and are therefore resistant to F4 bacteria and hence to diarrhoea caused by F4 strains. There are several different antigenic variants of F4 (ab, ac, ad), and it seems that there is a separate receptor for each. The presence or absence of at least two of these receptors (for F4ab and F4ac) is determined for each receptor by one of two closely-linked genes on chromosome 13. The immunological and population-genetic implications of segregation at these loci are very interesting: the result is selection against heterozygotes, as explained by Nicholas (2010, Introduction to Veterinary Genetics, 3rd edn, pp. 132-133).

Inheritance: Gibbons et al. (1977) provided convincing evidence that susceptibility to K88 Escherichia coli, due to the presence of a receptor for K88 E. coli on the brush-border surface of the intestine, is a single-locus autosomal dominant trait; and that resistance (due to lack of the K88 receptor) is recessive. They proposed the symbols S and s for the susceptible and resistant allele, respectively.

Mapping: Gibbons et al. (1977) reported some evidence for linkage of the S/s locus and the transferrin (Tf) locus). Bonneau et al. (1990) reported similar evidence. Vögeli et al. (1992) reported that the K88 locus is loosely linked to the L blood-group-system locus, in what was then called linkage group IV. Marklund et al. (1993) showed that the L blood-group-system locus maps to chromosome SSC4. As Marklund et al. (1993) noted, this evidence did not agree with proposed S/s linkage with Tf, because the Tf locus had by then been mapped to chromosome SSC13. Marklund et al. (1993) concluded that that the K88/blood-group-system L linkage was probably a false positive. This conclusion was reinforced by subsequent evidence supporting linkage to Tf, e.g. by Guérin et al. (1993) and Edfors-Lilja et al. (1995), the latter authors putting the SSC13 location beyond any doubt, with the K88 loci being located 7.4cM proximal to Tf. The SSc13 map location was refined by Python et al. (2002) who located the K88-receptor locus (now called F4acR) between microsatellites S0068 and Sw1030, with recombination fractions 0.04 and 0.05, respectively. Jørgensen et al. (2003) refined the location even further, to 3.8 cM between microsatellites Sw207 and S0075, and physically mapped it to SSC13q41, enabling them to identify two comparative positional regions on human chromosome HSA3, which should provide potential comparative positional candidate genes. The location was subsequently refined to 14Mb by Joller et al. (2009), to around 1.5Mb by Jacobsen et al. (2010), and to around 620kb by Rampoldi et al. (2011). A GWAS on 301 phenotyped pigs each genotyped with 62,163 SNPs via the Illumina PorcineSNP60 BeadChip by Fu et al. (2012) confirmed the location to a 2.6Mb region of SSC13q41. In the same year, Ren et al. (2012) reported a linkage analysis from a genome scan involving 194 microsatellites genotyped on each of 755 phenotyped F2 pigs from a White Duroc × Erhualian resource population, which also confirmed the candidate region. Fine mapping reduced the candidate region to 2.3MB. These same authors then conducted a GWAS and LDLA on the same 755 F2 animals genotyped with the Illumina PorcineSNP60 BeadChip, narrowing down the region to 360kb which contains 4 candidate genes, namely SLC12A8, HEG1, ITGB5 and MUC13. Goetstouwers et al. (2014) conducted a GWAS on 120 F4ab/acR phenotyped pigs (52 non-adhesive and 68 strong adhesive pigs), each genotyped with the Porcine SNP60 BeadChip (Illumina), yielding 55,289 informative SNPs. The authors reported that a "refined candidate region (chr13: 144,810,100–144,993,222) for F4ab/ac ETEC susceptibility was identified with MUC13 adjacent to the distal part of the region. This candidate region lacks annotated genes and contains a sequence gap based on the sequence of the porcine GenomeBuild 10.2."

Markers: As explained by Hu et al. (2019), "Rampoldi (2013) found that the markers CHCF1 and ALGA0106330 predicted the adhesion phenotype correctly in an experimental herd consisting of more than 400 related SLW pigs. Furthermore, 40 randomly selected SLW and six SLW pigs that were recombinant in the MUC4‐8227 marker were phenotyped and genotyped using CHCF1 and ALGA0106330 with no recombination between these two markers and the F4acR locus." Consequently, Hu et al. (2019) tested "the suitability of these markers in SLW, Duroc, Swiss Landrace and Piétrain breeding herds", concluding that "Selection of genetically F4ac‐resistant pigs [using those two marker alleles] is a sustainable and suitable alternative to decreasing animal loss and antibiotic use due to diarrhoea."

Molecular basis: By 2000, it was evident that the K88/F4 receptor was a mucin-like glycoprotein (Jin and Zhao, 2000; Van den Broeck et al., 2000). Noticing that a mucin gene called MUC4 was located in the orthologous human HSA3 regions (see Mapping section above), Peng et al. (2007) showed that the porcine MUC4 gene did, indeed, map to the candidate porcine region (SSC13q41), and, as expected with a gene in such close proximity, found strong association of a MUC4 polymorphism with susceptibility/resistance to F4 E. coli. Similar evidence was provided for another mucin gene, MUC13, by Zhang et al. (2008); for three other genes (SLC12A8, MYLK and KPNA1) by Huang et al. (2008); and for yet another mucin gene, MUC20, by Ren et al. (2009) and Ji et al. (2011). By 2011, the pig genome assembly had revealed a total of 18 annotated genes in the candidate region (Jacobsen et al., 2011). Doubt was cast on MUC13 and MUC20 as causal genes by Schroyen et al. (2012), who showed that expression of these two genes is not associated with resistance/susceptibility. However, Ren et al. (2012) provided data showing that while MUC13 is expressed in susceptible and resistant pigs, all resistant pigs across a range of breeds are homozygous for a MUC13 allele (called MUC13A) that encodes a MUC13 peptide with a specific tandem repeat structure (ASTSAPAAG). The other allele at this locus (MUC13B) results in a MUC13 peptide with a different tandem repeat structure (TPTPTTTP) to which pathogenic E coli can attach. (Text written by FN, with very helpful input from Merete Fredholm and Claus Jørgensen) From the results of their GWAS (see Mapping section above), Goetstouwers et al. (2014) hypothesised "that a porcine orphan gene or trans-acting element present in the identified candidate region has an effect on the glycosylation of F4 binding proteins and therefore determines the F4ab/ac ETEC susceptibility in pigs." Thus the causal mutation for this classic case of genetic resistance remains unknown.