Final Report Summary - INNATELYBETTERCOWS (A genomics approach to increasing disease resistance in dairy cows through improvements in innate immunity)
Background:
Production of high quality milk by disease resistant cattle through selective breeding is an important goal in the dairy industry. Multiple approaches are needed to explain the genetic mechanisms underlying quantitative trait loci (QTL) in dairy cows. Identifying single nucleotide polymorphisms (SNP) in candidates genes for immune traits and mastitis should help to elucidate the significance of genotype in determining the efficiency of the cow’s immune system and her ability to resist mastitis. This in turn can lead to the production of superior quality milk together with a reduced reliance on antibiotics. We proposed that the situation could be improved by identifying key genes in signaling pathways involving innate immune function which are activated by a wide range of pathogens. It would then be feasible to increase disease resistance by selectively breeding cows using marker assisted selection to improve their innate immune response.
Research Objectives:
➢ Validate the relevance of these genes in association studies using disease traits in Chinese dairy cows.
➢ Perform a functional validation study on peripheral blood cell responses to selected bacterial pathogens in cows with differing genotypes for innate immune genes.
Work performed and main results to date
1. SNP Identification and association study:
(1.1) TREM 1:
Two SNPs were identified in the TREM1 gene in the Chinese Holstein population (rs109937179 and rs208224995) (Table1). These were both associated with somatic cell score (SCS), the major indicators of mastitis, an indicator of inflammatory disease of the mammary gland in cattle. We found that the GG genotype of the rs109937179 polymorphism and CA genotype of rs208224995 were both associated with an increased risk of mastitis (Table 2).
(1.2) MAP4K4:
A total of 19 SNPs (Table1) were detected in the MAP4K4 gene, of which four were located in the 5’ upstream region, eight were located in exons and seven in introns. Bioinformatic analysis showed that seven SNPs were synonymous substitutions and one was a non-synonymous substitution (c.1562C/T) that changed the amino acid Ser into Phe. For the two SNPs investigated in exon 18 (rs209280849 and rs208343085) the SCS was significantly higher (P＜0.01) in cows with the TT genotype. The SNP in exon 18 (rs209280849) was also associated with protein percentage and milk yield (Table2). A further two SNPs in Intron 21 were associated with 305 d yield, one of which（rs208279998）was also associated with fat%. We also used http://www.urogene.org/methprimer/ and http://www.ebi.ac.uk/Tools/emboss/cpgplot to analyze the data for CpG prediction and possible transcription start sites in MA4K4. Analysis of the 3000 bp upstream sequence revealed no predicted CpG islands in the MAP4K4 gene of domestic cattle Bos taurus, whereas one was predicted in the buffalo MAP4K4 gene Bubalus bubalis. Using http://www.fruitfly.org/seq_tools/promoter.html we predicted possible promoter sequence and used Megablast (NCBI) to compare the mRNA sequence of MAP4K4 between cow and buffalo. This revealed differences in 58 positions (including 3’ UTR), four of which were responsible for an amino acid change. Analysis of interactions of cow and buffalo MAP4K4 proteins with other proteins revealed significant differences in both interaction and domain structure. The results suggest that the identified polymorphisms in MAP4K4 could be useful genetic markers for selection of dairy cattle against mastitis.
(1.3) MBL1, FABP4 and FABP9:
A total of 31 SNPs (Table1) were found in the exons and flanking regions of MBL1, FABP4, FABP5 and FABP9. There were seven mis-sense SNPs: FABP4 Exon2 rs110757796 and Exon3 rs110652478; FABP5 Exon3 rs479404894; FABP9 Exon2 rs137165793 and rs110571965 and FABP9 Exon4 rs136262565 and rs137228156. There was also one splicing mutation in FABP4 at Exon3 rs110383592 and two synonymous mutations in MBL1 at Exon2 rs110326717 and FABP4 at Exon4 rs110370032. Additionally, six SNPs were found in 3’-, 5’- flanking sequences, six SNPs in 3’ non-translated regions and seven SNPs in introns.
Twelve sites were selected to conduct genotyping and linkage disequilibrium analysis from these 28 SNPs. Eight SNPs in FABP4 and FABP9 were in high linkage disequilibrium (Figure 1) (at rs110757796, rs110383592, rs110370032 and rs109077068 in FABP4 and rs137165793, rs137228156, rs110807318 and rs109240772 in FABP9). There were haplotypes (CCCCCAGA, TAGGTGCG, TAGGTGCA) found in the eight highly linked sites. All the SNPs in FABP4 and FABP9 were associated with SCS. Four of the five SNPs in FABP4 were also associated with 305 d milk yield and fat% and two with protein%. Three of the SNPs in FABP9 were also related to the 305 day yield (P<0.05) (Table 2). The two SNP in MBL1 were associated with milk production traits but not SCS. For rs208491630, AC was the superior genotype and was related to fat and protein %. The locus in MBL1 at Intron 2 rs110326717 was related to milk fat and protein % in the second lactation.
1.2 Bioinformatics and functional validation studies
The two SNP found in TREM1 were both at important sites, a splice region and an miRNA binding domain (Figure 2), which may regulate gene expression and produce splice variants.The up-regulation of TREM1 induced by lipopolysaccharide (LPS) in RAW 264.7 cells (a murine macrophage cell line) stimulated the over production of the pro-inflammatory cytokines IL-6 and TNF-α mRNA whereas there was no change in the anti-inflammatory cytokine IL-10 compared with the control cells (Figure 3). Combining LPS with TREM1 mRNA down-regulation using siRNA transfected cells decreased expression of TLR4, IL6 and TNF-α but increased expression of IL10 mRNA. Further addition of Resveratrol, a natural phenol produced by some plants, further decreased expression TLR4, IL6 and TNF-α but increased expression of IL10.
Work in the UK (in collaboration with Dr Stefan Neuenschwander, Institute of Animal Sciences at the ETH Zürich, Switzerland), continued to explore the functional relevance of two other genes, TLR2 and INOS. These were selected as work reported previously under Project 328205 Objective 8 suggested that these genes might be important in the greater resistance to mastitis found in Brown Swiss compared to Holstein cows. Macrophages from Brown Swiss cows produced significantly more reactive oxygen species and their reactive nitrogen species production was less variable following in vitro challenges representative of gram negative and positive bacteria and fungi (Gibson et al. 2015). To investigate this further, these genes were sequenced in 450 pedigree Holstein Friesian and 115 pedigree Brown Swiss cattle. Eight SNP with low to moderate effects on the gene product were identified in INOS and four with a moderate effect in TLR2. Further work is in progress to determine if these polymorphisms translate into differences in protein production.
Additional bioinformatics analysis was performed to compare the MAPK4 gene between domestic cattle Bos taurus and buffalo Bubalus bubalis. Analysis of the 3000 bp upstream sequence revealed no predicted CpG islands in the MAP4K4 gene of domestic cattle Bos taurus, whereas one was predicted in the buffalo MAP4K4 gene. Using http://www.fruitfly.org/seq_tools/promoter.html we predicted possible promoter sequence and used Megablast (NCBI) to compare the mRNA sequence of MAP4K4 between cow and buffalo. This revealed differences in 58 positions (including 3’ UTR), four of which were responsible for an amino acid change. Analysis of interactions of cow and buffalo MAP4K4 proteins with other proteins revealed significant differences in both interaction and domain structure. In conclusion, these results suggest that the expression and function of MAP4K4 may differ between cow and buffalo leading to differences in expression of inflammatory mediators, apoptosis or cell proliferation. Future work will perform in vitro analysis and epigenetic studies to follow up these possibilities.
Summary and Impact
The data obtained from the SNP identification and association study meets the hypothesis that differences in genotype for candidate genes involved in the regulation of innate immunity play important roles in susceptibility to mastitis and also influence milk production of Holstein dairy cows. Two of the genes investigated were MAP4K4 and TREM1. MAP4K4 plays an important role in inflammation as it mediates the TNF-alpha signaling pathway, whereas TREM1 encodes a receptor belonging to the Ig superfamily which amplifies neutrophil and monocyte-mediated inflammatory responses. Bioinformatic comparison of the MAP4K4 gene between cow and buffalo revealed differences in protein structure which may potentially alter inflammatory responses. Up-regulation of pro-inflammatory cytokines in vitro with the bacterial product lipopolysaccharide (LPS) was inhibited by down-regulation of TREM1 mRNA using siRNA transfected cells. This supports the role of TREM1 as another potential candidate gene for immune function. An improved understanding of which key genes modulate innate immunity has the potential to benefit future generations of dairy cows through improved health while at the same time reducing antibiotic use in the dairy industry. This in turn will help to minimize the numbers of animals required to maintain milk production, so also reducing the impact of dairy farming on the environment.
Production of high quality milk by disease resistant cattle through selective breeding is an important goal in the dairy industry. Multiple approaches are needed to explain the genetic mechanisms underlying quantitative trait loci (QTL) in dairy cows. Identifying single nucleotide polymorphisms (SNP) in candidates genes for immune traits and mastitis should help to elucidate the significance of genotype in determining the efficiency of the cow’s immune system and her ability to resist mastitis. This in turn can lead to the production of superior quality milk together with a reduced reliance on antibiotics. We proposed that the situation could be improved by identifying key genes in signaling pathways involving innate immune function which are activated by a wide range of pathogens. It would then be feasible to increase disease resistance by selectively breeding cows using marker assisted selection to improve their innate immune response.
Research Objectives:
➢ Validate the relevance of these genes in association studies using disease traits in Chinese dairy cows.
➢ Perform a functional validation study on peripheral blood cell responses to selected bacterial pathogens in cows with differing genotypes for innate immune genes.
Work performed and main results to date
1. SNP Identification and association study:
(1.1) TREM 1:
Two SNPs were identified in the TREM1 gene in the Chinese Holstein population (rs109937179 and rs208224995) (Table1). These were both associated with somatic cell score (SCS), the major indicators of mastitis, an indicator of inflammatory disease of the mammary gland in cattle. We found that the GG genotype of the rs109937179 polymorphism and CA genotype of rs208224995 were both associated with an increased risk of mastitis (Table 2).
(1.2) MAP4K4:
A total of 19 SNPs (Table1) were detected in the MAP4K4 gene, of which four were located in the 5’ upstream region, eight were located in exons and seven in introns. Bioinformatic analysis showed that seven SNPs were synonymous substitutions and one was a non-synonymous substitution (c.1562C/T) that changed the amino acid Ser into Phe. For the two SNPs investigated in exon 18 (rs209280849 and rs208343085) the SCS was significantly higher (P＜0.01) in cows with the TT genotype. The SNP in exon 18 (rs209280849) was also associated with protein percentage and milk yield (Table2). A further two SNPs in Intron 21 were associated with 305 d yield, one of which（rs208279998）was also associated with fat%. We also used http://www.urogene.org/methprimer/ and http://www.ebi.ac.uk/Tools/emboss/cpgplot to analyze the data for CpG prediction and possible transcription start sites in MA4K4. Analysis of the 3000 bp upstream sequence revealed no predicted CpG islands in the MAP4K4 gene of domestic cattle Bos taurus, whereas one was predicted in the buffalo MAP4K4 gene Bubalus bubalis. Using http://www.fruitfly.org/seq_tools/promoter.html we predicted possible promoter sequence and used Megablast (NCBI) to compare the mRNA sequence of MAP4K4 between cow and buffalo. This revealed differences in 58 positions (including 3’ UTR), four of which were responsible for an amino acid change. Analysis of interactions of cow and buffalo MAP4K4 proteins with other proteins revealed significant differences in both interaction and domain structure. The results suggest that the identified polymorphisms in MAP4K4 could be useful genetic markers for selection of dairy cattle against mastitis.
(1.3) MBL1, FABP4 and FABP9:
A total of 31 SNPs (Table1) were found in the exons and flanking regions of MBL1, FABP4, FABP5 and FABP9. There were seven mis-sense SNPs: FABP4 Exon2 rs110757796 and Exon3 rs110652478; FABP5 Exon3 rs479404894; FABP9 Exon2 rs137165793 and rs110571965 and FABP9 Exon4 rs136262565 and rs137228156. There was also one splicing mutation in FABP4 at Exon3 rs110383592 and two synonymous mutations in MBL1 at Exon2 rs110326717 and FABP4 at Exon4 rs110370032. Additionally, six SNPs were found in 3’-, 5’- flanking sequences, six SNPs in 3’ non-translated regions and seven SNPs in introns.
Twelve sites were selected to conduct genotyping and linkage disequilibrium analysis from these 28 SNPs. Eight SNPs in FABP4 and FABP9 were in high linkage disequilibrium (Figure 1) (at rs110757796, rs110383592, rs110370032 and rs109077068 in FABP4 and rs137165793, rs137228156, rs110807318 and rs109240772 in FABP9). There were haplotypes (CCCCCAGA, TAGGTGCG, TAGGTGCA) found in the eight highly linked sites. All the SNPs in FABP4 and FABP9 were associated with SCS. Four of the five SNPs in FABP4 were also associated with 305 d milk yield and fat% and two with protein%. Three of the SNPs in FABP9 were also related to the 305 day yield (P<0.05) (Table 2). The two SNP in MBL1 were associated with milk production traits but not SCS. For rs208491630, AC was the superior genotype and was related to fat and protein %. The locus in MBL1 at Intron 2 rs110326717 was related to milk fat and protein % in the second lactation.
1.2 Bioinformatics and functional validation studies
The two SNP found in TREM1 were both at important sites, a splice region and an miRNA binding domain (Figure 2), which may regulate gene expression and produce splice variants.The up-regulation of TREM1 induced by lipopolysaccharide (LPS) in RAW 264.7 cells (a murine macrophage cell line) stimulated the over production of the pro-inflammatory cytokines IL-6 and TNF-α mRNA whereas there was no change in the anti-inflammatory cytokine IL-10 compared with the control cells (Figure 3). Combining LPS with TREM1 mRNA down-regulation using siRNA transfected cells decreased expression of TLR4, IL6 and TNF-α but increased expression of IL10 mRNA. Further addition of Resveratrol, a natural phenol produced by some plants, further decreased expression TLR4, IL6 and TNF-α but increased expression of IL10.
Work in the UK (in collaboration with Dr Stefan Neuenschwander, Institute of Animal Sciences at the ETH Zürich, Switzerland), continued to explore the functional relevance of two other genes, TLR2 and INOS. These were selected as work reported previously under Project 328205 Objective 8 suggested that these genes might be important in the greater resistance to mastitis found in Brown Swiss compared to Holstein cows. Macrophages from Brown Swiss cows produced significantly more reactive oxygen species and their reactive nitrogen species production was less variable following in vitro challenges representative of gram negative and positive bacteria and fungi (Gibson et al. 2015). To investigate this further, these genes were sequenced in 450 pedigree Holstein Friesian and 115 pedigree Brown Swiss cattle. Eight SNP with low to moderate effects on the gene product were identified in INOS and four with a moderate effect in TLR2. Further work is in progress to determine if these polymorphisms translate into differences in protein production.
Additional bioinformatics analysis was performed to compare the MAPK4 gene between domestic cattle Bos taurus and buffalo Bubalus bubalis. Analysis of the 3000 bp upstream sequence revealed no predicted CpG islands in the MAP4K4 gene of domestic cattle Bos taurus, whereas one was predicted in the buffalo MAP4K4 gene. Using http://www.fruitfly.org/seq_tools/promoter.html we predicted possible promoter sequence and used Megablast (NCBI) to compare the mRNA sequence of MAP4K4 between cow and buffalo. This revealed differences in 58 positions (including 3’ UTR), four of which were responsible for an amino acid change. Analysis of interactions of cow and buffalo MAP4K4 proteins with other proteins revealed significant differences in both interaction and domain structure. In conclusion, these results suggest that the expression and function of MAP4K4 may differ between cow and buffalo leading to differences in expression of inflammatory mediators, apoptosis or cell proliferation. Future work will perform in vitro analysis and epigenetic studies to follow up these possibilities.
Summary and Impact
The data obtained from the SNP identification and association study meets the hypothesis that differences in genotype for candidate genes involved in the regulation of innate immunity play important roles in susceptibility to mastitis and also influence milk production of Holstein dairy cows. Two of the genes investigated were MAP4K4 and TREM1. MAP4K4 plays an important role in inflammation as it mediates the TNF-alpha signaling pathway, whereas TREM1 encodes a receptor belonging to the Ig superfamily which amplifies neutrophil and monocyte-mediated inflammatory responses. Bioinformatic comparison of the MAP4K4 gene between cow and buffalo revealed differences in protein structure which may potentially alter inflammatory responses. Up-regulation of pro-inflammatory cytokines in vitro with the bacterial product lipopolysaccharide (LPS) was inhibited by down-regulation of TREM1 mRNA using siRNA transfected cells. This supports the role of TREM1 as another potential candidate gene for immune function. An improved understanding of which key genes modulate innate immunity has the potential to benefit future generations of dairy cows through improved health while at the same time reducing antibiotic use in the dairy industry. This in turn will help to minimize the numbers of animals required to maintain milk production, so also reducing the impact of dairy farming on the environment.