Source: UNIV OF WISCONSIN submitted to NRP
PREDICTION OF SUBCLINICAL KETOSIS PREDIPOSITION IN DAIRY COWS THROUGH GENOMIC RISK ANALYSIS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1006649
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Oct 1, 2015
Project End Date
Sep 30, 2019
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
UNIV OF WISCONSIN
21 N PARK ST STE 6401
MADISON,WI 53715-1218
Performing Department
Dairy Science
Non Technical Summary
The transition to lactation period in dairy cattle, defined as 3 weeks prior to and 3 weeks after calving, is known to be the most challenging period in the dairy cow life cycle, specifically in terms of metabolic disorders. Ketosis, is defined as elevated ketone bodies in the blood and is a critical challenge to transition dairy cows that has negative impacts to milk production, animal health, and profitability. Cows with sub-clinical ketosis (SCK) produce less milk, are more likely to develop a displaced abomasum, and are more likely to be culled from the herd. The overall goal of the proposed research is to develop and validate practical tools for use by producers to aid in identification and treatment of SCK. Our central hypothesis is that cowside tests, milk analysis, and genomic markers can be used for early detection and reduction of negative impacts of SCK at the individual-animal, pen, and herd levels. This research focuses on developing and validating a SCK predictor index using genetic risk scores correlated with SCK incidence.To achieve this objective, we will collect repeat blood samples from 750 early lactation dairy cows to determine if the cow is healthy or ketotic in terms of ketosis. Cows will also be genotyped and genetic information used to determine genetic markers for ketosis predisposition. Having a strong disease diagnosis and the genotype will allow for analysis of genetic risks scores and will be used to develop a SCK predictor index. This will be a valuable tool for producers, nutritionists, and geneticists as it will allow for cows at increased risk of SCK to be identified and managed or treated differently.
Animal Health Component
50%
Research Effort Categories
Basic
(N/A)
Applied
50%
Developmental
50%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
3043410108025%
3043410101025%
3113410101025%
3023410101025%
Goals / Objectives
The overall goal of the proposed research is to develop and validate practical tools for use by producers to aid in identification and treatment of sub-clinical ketosis (SCK). Our central hypothesis is that cowside tests, milk analysis, and genomic markers can be used for early detection and reduction of negative impacts of SCK at the individual-animal, pen, and herd levels. At the completion of this project, it is our expectation that we will have conducted a genome wide association study (GWAS) that identifies loci associated with SCK incidence in a comprehensive data set, allowing for use of these loci as a marker for SCK predisposition. The specific aim of the research is to develop and validate a SCK predictor index using genetic risk scores correlated with SCK incidence.
Project Methods
Repeated blood samples will be collected from 750 cows at commercial dairy farms to determine the incidence of SCK. Our collaborative group has been performing SCK testing at more than 10 dairy herds and has an ongoing relationship with these herds. At each farm, blood sampling will be conducted twice a week after morning milking, during morning feeding. Blood samples will be collected for blood BHBA concentrations at four timepoints between 4 and 17 DIM for each cow.Justification of the collection of four samples per cow is based on posthoc analysis of a dataset from the collaborative group, examining the frequency of testing required to determine a definitive diagnosis of SCK vs. healthy. Posthoc analysis of previously collected data (Oetzel, unpublished data). Blood BHBA was quantified by Precision Xtra meter from 1868 cows (across 4 herds) at 6 time points between 4 and 17 DIM to determine incidence of SCK and absolute ketosis status (positive or negative) for each cow. A positive SCK diagnosis required at least one BHBA >1.2 mmol/L between 3 and 16 DIM. Sensitivity is defined as the number of cows ketotic on that test day divided by the total number of cows determined to be ketotic during the trial, with 100% sensitivity being ideal and 90% sensitivity being a robust test. Within the whole dataset, the sensitivity of a single blood Precision Xtra test was 44%. These data highlight that accurate determination of SCK status requires repeat testing; however, the number of test required has not yet been defined. Without adequate testing timepoints, a cow can be classified as nonketotic merely because she was not tested at the correct timepoint and not because she had truly remained below the cutpoint for SCK through the entire fresh period. To determine how many timepoints are needed, the dataset was manipulated to include only 4 tests per cow between 4 and 17 DIM to determine if 4 tests per cow is adequate to determine overall SCK incidence. The sensitivity of the 4-test approach was 91% compared with the 6-test approach. This analysis supports that a 4-test approach to ketosis testing provides adequate information to determine a definitive ketosis status.Blood samples will be collected at each of the four timepoints via venipuncture of the coccygeal vein or artery into evacuated tubes (Becton Dickinson, Franklin Lakes, NJ). Determination of blood BHBA cowside will be by Precision Xtra meter (Abbott Laboratories, Abbott Park, IL) using blood (100 μL) from the evacuated tube. Use of the cowside test will allow for same-day feedback of information regarding SCK status to the producer for treatment decisions.Herd DairyComp 305 records and individual cow health and treatment records will also be collected. While there is variation in the consistency and accuracy of recorded data in DairyComp 305 across farms the data still presents a useful insight into possible risk factors and impacts of SCK. Objective data (parity, calving date, pregnancy confirmed by veterinarian, services to conception, etc.) will be considered reliable inputs into the regression models. In addition to data collected from DairyComp305 records during the testing period, outcomes data will be collected over the next six-months post-sampling. Milk production (at 30 DIM and cumulative at 30, 90, and 120 DIM, and previous lactation 305 mature equivalent milk), health incidences (specifically ketosis and displaced abomasum), culling, and reproductive efficiency data (DIM at first breeding, services to conception, and days to pregnant) will be collected to examine differences in outcomes based on SCK status. In addition to the blood samples, a hair sample will be collected for genotyping (Genex, Verona, WI). Low-density genotypes (19K SNP chip) will be analyzed against SCK status, as a continuous variable (blood BHBA concentration) and a categorical variable (healthy, sub-clinical ketosis, ketosis), to determine loci associated with SCK. These genetic risk scores can then be used to identify animals that are at increased risk for developing SCK. Output of GWAS will be used to identify individual or sets of SNP that can be used to predict an individual cow's genetic predisposition (risk) for SCK. Each cow in our study will have one of three possible genotypes at each of the 19,000 SNP loci: AA, AB, or BB, where alleles A and B refer to alternative alleles or variants at a given location on the chromosome. If the frequency of one allele (e.g., allele A) is higher among cows that are affected with SCK than among cows that are unaffected, then it is "associated" with the disease. Such an association occurs when the SNP is located on the same chromosome near an unknown functional mutation in the genome that affects a biological process related to SCK (e.g., some aspect of triglyceride synthesis). Even though the exact location and mechanism of these mutations are unknown, the SNP genotype at that location can be used as a "marker" for selection, and by changing the frequency of the SNP allele we will also change the frequency of the underlying mutant allele that increases SCK susceptibility. In the current project, our aim is not genetic selection, but rather a genetic diagnostic tool that can identify cows with high SCK susceptibility that should be targeted for a management intervention. Although the example described above refers to the use of a single SNP as a marker for SCK susceptibility, extension to multiple SNP is extremely simple, and in our study we anticipate using several dozen SNPs simultaneously when evaluating a cow's genetic predisposition for SCK. An alternative, which will also be explored in the proposed study, is to use all 19,000 SNP simultaneously. This approach has been used successfully in our lab for prediction of residual feed intake and related traits in lactating Holstein cows based on the SNP genotype, health history, or SNP genotype plus health history for each cow (Yao et al., 2014). Similar approaches have been used in humans to compute genetic risk scores for various diseases. For example, Grassman et al. (2012) developed a genetic risk prediction model for age-related macular degeneration (AMD) based on age, smoking, and 13 risk variants and reported that 87.4% of patients in the highest risk category were expected to develop AMD by 85 years of age, as compared with roughly 50% of the general population and only 2.2% of patients in the lowest risk category. One issue that can lead to false-positive associations between SNP genotypes and disease status, particularly in humans, is population stratification (e.g., people from different racial/ethnic groups may differ in disease frequency while also differing in allele frequency by chance), so we will check for population substructure prior to our GWAS analysis and make adjustments as needed. This genomic testing will allow for identification of cows that are at greater risk for SCK onset, through development of a genetic risk score. This information will allow producers to identify specific cows that need more intensive management or observation during the transition to lactation period. This tool represents something that is currently not available to producers and will improve our ability to make informed management and nutrition decisions regarding SCK on each specific farm and with each specific cow.

Progress 10/01/15 to 09/30/19

Outputs
Target Audience:The target audience was primarily the scientific community. During this period, efforts to reach this audience included presentation of abstracts at scientific meetings. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The PhD student involved learned advanced genomic analysis methods through coursework and training at UW and is implementing this on the dataset. The PhD student also worked collaboratively with PIs with GWAS expertise and a postdoc developing new GWAS methods. The graduate student has been working on writing a manuscript. How have the results been disseminated to communities of interest?The abstracts were presented as posters during scientific meetings. Additionally, the manuscript in preparation will be submitted to a manuscript appropriate for animal genomics research. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? The transition to lactation period in dairy cattle, defined as 3 weeks prior to and 3 weeks after calving, is known to be the most challenging period in the dairy cow life cycle, specifically in terms of metabolic disorders. Ketosis defined as elevated ketone bodies in the blood, is a critical challenge to transition dairy cows that negatively impacts milk production, animal health, and profitability and cost an average of $300 per case. Determining genetic risk factors for ketosis would provide a valuable tool for producers, nutritionists, and geneticists and will allow for cows at increased risk of SCK to be identified and managed or treated differently. Accomplishments during this reporting year include analysis of the dataset using newly developed genome-wide association and genome-wide interaction studies to identify novel single nucleotide polymorphisms (SNP) that may be associated with SCK onset or the interaction of parity and SCK risk. We also did downstream pathway enrichment analysis so that the metabolic pathways that these SNP are associated with could be identified to understand their role and to support future targeted research. This analysis resulted in two abstracts which were presented at scientific meetings and preparation of a manuscript that will be submitted in 2020. This year we worked to finalize our genome wide association models (GWAS) to identify SCK predisposition. We had completed preliminary analysis but because of the complexity of the metabolic disorder, and the potential interaction of parity (1st parity cows are less likely to develop SCK than cows that are 2nd or greater parity), we needed to use more advanced approaches. We collaborated with Kent Weigel's (co-PI) postdoctoral research fellow to apply brand new analysis techniques from her dissertation, to our sample set. As these are new methods, it is time consuming and rigorous to validate and run. Additionally, the sample set is very large and therefore computationally demanding to analyze. Although this took most of our year, we have completed analysis and have a manuscript draft in progress that we expect to submit in early 2020. We are very satisfied with the final model analysis as it represents not only a uniquely thorough dataset, but also a cutting-edge analysis method. Together this combination allowed us to identify several single nucleotide polymorphisms that may be linked to SCK predisposition. We were also able to push the analysis one step further and conduct pathway enrichment analysis. This analysis looks at the genes and pathways associated with the SNP to identify metabolic pathways that may be regulated during SCK development. Notably, some of the pathways flagged are pathways that we are studying in other aspects of our research. There were also some novel pathways identified that we will likely explore further in future research. The PI was honored to be invited to give a symposium review at the American Dairy Science Association where much of this work, in connect with the PI's other research, was reviewed. The invited review was published (Pralle and White, 2019).

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Pralle, R.S., K. W. Weigel, N. E. Schultz, H. M. White. 2019. Hyperketonemia genome-wide association study in Holstein cows. Annual Meeting of the European Federation of Animal Science. 70(25):538.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Pralle, R.S., K. W. Weigel, N. E. Schultz, and H. M. White. 2019. Hyperketonemia SNP by parity group genome-wide interaction study in Holstein cows. Annual Meeting of the European Federation of Animal Science. 70(25):541.
  • Type: Journal Articles Status: Accepted Year Published: 2019 Citation: Pralle, R. S. and H. M. White. 2019. Symposium Review: Big data, big predictions: Utilizing milk Fourier-transform infrared and genomics to improve hyperketonemia management. J. Dairy Sci. Accepted.


Progress 10/01/17 to 09/30/18

Outputs
Target Audience:The target audience was primarily the scientific community. During this period, efforts to reach this audience included publication of a manuscript in JDS. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The PhD student involved learned advanced genomic analysis methods through coursework and training at UW and is implementing this on the dataset. The PhD student also worked collaboratively with PIs with GWAS expertise and a postdoc developing new GWAS methods. How have the results been disseminated to communities of interest?This information is being distributed to the scientific community through a Journal of Dairy Science publication. What do you plan to do during the next reporting period to accomplish the goals?We plan to finish the GWAS analysis and present it as an abstract at a scientific meeting. We also plan to write a manuscript on the GWAS analysis.

Impacts
What was accomplished under these goals? The transition to lactation period in dairy cattle, defined as 3 weeks prior to and 3 weeks after calving, is known to be the most challenging period in the dairy cow life cycle, specifically in terms of metabolic disorders. Ketosis defined as elevated ketone bodies in the blood, is a critical challenge to transition dairy cows that negatively impacts milk production, animal health, and profitability and cost an average of $300 per case. Determining genetic risk factors for ketosis would provide a valuable tool for producers, nutritionists, and geneticists and will allow for cows at increased risk of SCK to be identified and managed or treated differently. Accomplishments during this reporting year include a publication of a predictive models utilizing advanced computational methods (Pralle et al., 2018). Using artificial neural networks, we were able to generate SCK predictive models that are more sensitive, specific, and robust than models previously generated by us and others using linear regression models. We also made significant progress on the genome wide association study (GWAS). We worked collaboratively to use the most cutting-edge analysis methods and anticipate having an abstract at a scientific meeting in 2019 and a subsequent publication.

Publications

  • Type: Journal Articles Status: Published Year Published: 2018 Citation: Pralle, R. S., K. W. Weigel, and H. M. White. 2018. Predicting blood beta-hydroxybutyrate using milk Fourier transform infrared spectrum, milk composition, and producer-reported variables with multiple linear regression, partial least squares regression, and artificial neural network. J. Dairy Sci. 101:4378-4387.


Progress 10/01/16 to 09/30/17

Outputs
Target Audience: Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The PhD student presented research at a scientific meeting. The student also learned advanced genomic analysis methods through coursework and training at UW and is implementing this on the dataset. How have the results been disseminated to communities of interest?Yes, this information is being distributed to the scientific community through presentation at the ADSA scientific meetings. Two manuscripts were publishedand one is under review. What do you plan to do during the next reporting period to accomplish the goals?We will continue genome wide association dataset analysis. Once completed, we will interpret data and prepare a research abstract.

Impacts
What was accomplished under these goals? The transition to lactation period in dairy cattle, defined as 3 weeks prior to and 3 weeks after calving, is known to be the most challenging period in the dairy cow life cycle, specifically in terms of metabolic disorders. Ketosis defined as elevated ketone bodies in the blood, is a critical challenge to transition dairy cows that negatively impacts milk production, animal health, and profitability and cost an average of $300 per case. Determining genetic risk factors for ketosis would provide a valuable tool for producers, nutritionists, and geneticists and will allow for cows at increased risk of SCK to be identified and managed or treated differently. Accomplishments during this reporting year reflect analyzing the dataset collected during the first phase of this project. We analyzed the genetic variation associated with hyperketonemia (Weigel et al., 2017) and the association between hyperketonemia, previous lactation feed efficiency, and body condition score change (Rathbun et al., 2017). We also explored genome wide association analysis of the sample set.

Publications

  • Type: Journal Articles Status: Published Year Published: 2017 Citation: Weigel, K. A., R. S. Pralle, H. Adams, K. Cho, C. Do, and H. M. White. 2017. Prediction of whole genome risk for selection and management of hyperketonemia in Holstein dairy cattle. J. Anim. Breed. Genet. 134:275-285.
  • Type: Journal Articles Status: Published Year Published: 2017 Citation: Rathbun, F. M., R. S. Pralle, S. J. Bertics, L. E. Armentano, K. Cho, C. Do, K. A. Weigel, and H. M. White. 2017. Relationships between body condition score change, prior mid-lactation phenotypic residual feed intake, and hyperketonemia onset in transition dairy cows. J. Dairy Sci. 100:3685-3696.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Pralle, R. S., K. A. Weigel, and H. M. White. 2017. Development and evaluation of hyperketonemia prediction models. J. Dairy Sci. 100, Suppl 2: 237.


Progress 10/01/15 to 09/30/16

Outputs
Target Audience:Our target audience will be two-fold. First, we will target the field specialists who would use and implement the tool together with producers. These would be nutritionists, geneticists, and consultants. We would educate these field specialists about the tool and how to use it on farm. We would also target other scientists (nutrition and genetics) as the field of genetics and nutrition interactions is an important area of research. By communicating with other scientists, we can strengthen the research and increase the impact of the work. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The PhD student involved has given two extension talks, learned new genomic analysis technqiues to apply for research, and is preparing research abstracts of the current analysis. How have the results been disseminated to communities of interest?One manuscript is currently under review.An abstract on the genome wide association study is being prepared for presentation at the summer ADSA meetings. What do you plan to do during the next reporting period to accomplish the goals?We will continue working on the data analysis and prepare the abstract for submission and presentation.

Impacts
What was accomplished under these goals? The transition to lactation period in dairy cattle, defined as 3 weeks prior to and 3 weeks after calving, is known to be the most challenging period in the dairy cow life cycle, specifically in terms of metabolic disorders. Ketosis, is defined as elevated ketone bodies in the blood and is a critical challenge to transition dairy cows that has negative impacts to milk production, animal health, and profitability and cost an average of $300 per case. Determining genetic risk factors for ketosis would provide avaluable tool for producers, nutritionists, and geneticists as it willallow for cows at increased risk of SCK to be identified and managed or treated differently. We have been analyzing geneotypes from a dataset of more than 1000 cows with four blood and milk samples postcalving, milk production data, other health data, and a ketosis diagnosis. These genotypes are being used to conduct genome wide association studies that can identify loci associated with ketosis incidence.

Publications

  • Type: Journal Articles Status: Under Review Year Published: 2017 Citation: 1. Tiberio, F. M., R. S. Pralle, S. J. Bertics, L. E. Armentano, K. Cho, C. Do, K. A. Weigel, and H. M. White. Body condition score, but not prior mid-lactation residual feed intake, influences hyperketonemia onset in transition dairy cows. J. Dairy Sci. Under review.