Citation
Impacts of Reproductive Management Strategies and Genetic Merit on Reproductive Parameters of Dairy Heifers

Material Information

Title:
Impacts of Reproductive Management Strategies and Genetic Merit on Reproductive Parameters of Dairy Heifers
Creator:
Veronese, Anderson
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (112 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Veterinary Medical Sciences
Veterinary Medicine
Committee Chair:
CHEBEL,RICARDO C
Committee Co-Chair:
PENAGARICANO,FRANCISCO
Committee Members:
RISCO,CARLOS A

Subjects

Subjects / Keywords:
estrus -- genomic -- pgf2alfa
Veterinary Medicine -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Veterinary Medical Sciences thesis, M.S.

Notes

Abstract:
The objectives of this experiment was to evaluate the effects of PGF2alfa formulations and methods of estrus detection on physiological parameters, estrous behavior, and reproductive performance of dairy heifers. Additionally, the association between fertility traits and physiological parameters, estrous behavior, and reproductive performance of dairy heifers are described. Holstein heifers (n = 1,019) were fitted with an automated estrus detection system (AED) and enrolled in the experiment around 11 months of age. Heifers were assigned to the PGF2alfa (CLO: cloprostenol sodium or DIN: dinoprost thromethamine) and estrus detection (AED: automated estrus detection or VSI: visual detection of estrus) treatments in a 2 x 2 factorial design. At birth, heifers were genotyped and genomic daughter pregnancy rate (DPR) and heifer conception rate (HCR) were collected. Treatment with CLO increased percentage of heifers detected in estrus within 7 days after treatment and reduced progesterone concentrations at estrus but it had no effect on hazard of pregnancy. Automated estrus detection tended to improve hazard of pregnancy. Genomic daughter pregnancy rate was associated with greater ovulatory follicle size, estradiol concentrations, and estrus expression, whereas GHCR was negatively associated with estrous behavior. Selection of PGF2alfa may be according to parameters other than efficacy because reproductive performance was similar between CLO and DIN. Herds with inefficient visual estrus detection may benefit from AED. Selection of heifers for DPR is likely to improve signs of estrus and overall reproductive performance, but additional information is needed before HCR may be used extensively as a selection parameter. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2017.
Local:
Adviser: CHEBEL,RICARDO C.
Local:
Co-adviser: PENAGARICANO,FRANCISCO.
Statement of Responsibility:
by Anderson Veronese.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
LD1780 2017 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

IMPACTS OF REPRODUCTIVE MANAGEMENT STRATEGIES AND GENETIC MERIT ON REPRODUCTIVE PARAMETERS OF DAIRY HEIFERS By ANDERSON VERONESE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER IN SCIENCE UNIVERSITY OF FLORIDA 2017

PAGE 2

2017 Anderson Veronese

PAGE 3

To my family and my fianc e who supported my decisions, were comprehensive, and always there when I needed

PAGE 4

4 ACKNOWLEDGMENTS Primarily, I thank my parents Jaime and Marines for helping me to pursue my dreams and life goals and always support me, even when not agreeing with my life choices. My grandfather Felix for developing in me enthusiasm and passion for the dairy industry, and being my role model in life, and my grandmother Leonora for all love provided. My fianc e and soon to be wife Beatriz for helping me through this journey, for the immense patience, for being my partner, my supporter and my source of reasoning in difficult moments. I would not be who I am now without my family, and the y are the reason for my living. I also th ank my early mentor, Dr. Angela Veiga, for introduc ing me to and mak ing me develop an interest in science and for all the support and knowledge she provided me. My advisor Dr. Ricardo Chebel for the opportunity given and for the training and guidance provi ded during this period. Members of my committee, Dr. Carlos Risco and Dr. Francisco Peagaricano for all help provided other professors from the FARMS department, Dr. Klibs Galvao, and Dr. Rafael Bisinotto, and the lab manager Dr. Xiaojie Ma for all support with lab assays. All visiting students, interns and other members of the lab that have helped during the conduction of the studies, Anna Belli, Rafael Moreira, Kelly Flanagan, Caylen Wouters, Odinei Marques, Gustavo Soeiro, Victoria Rocha, and Jamie Horstmann, and members of other labs who have helped me, Eduardo Barros and Achilles Neto.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 6 LIST OF FIGURES ................................ ................................ ................................ ......................... 7 LIST OF ABBREVIATIONS ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................................ ... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 15 Importance of Reproductive Performance for Dairy Production ................................ ........... 15 Reproductive Management of Dairy Heifers ................................ ................................ .......... 16 2 EFFECTS OF TWO DIFFERENT PROSTAGLANDIN F FORMULATIONS AND METHOD OF ESTRUS DETECTION ON ESTR O US CHARACTERISTICS AND REPRODUCTIVE PERFORMANCE OF DAIRY HEIF ERS ................................ .............. 24 Material s and Methods ................................ ................................ ................................ ........... 26 Results ................................ ................................ ................................ ................................ ..... 35 Discussion ................................ ................................ ................................ ............................... 38 3 ASSOCIATION AMONG GENETIC MERIT FOR REPRODUCTION TRAITS AND ESTR O US CHARACTERISTICS AND FERTILITY OF HOLSTEIN HEIFERS .............. 51 Material s and Methods ................................ ................................ ................................ ........... 53 Results ................................ ................................ ................................ ................................ ..... 60 Discussion ................................ ................................ ................................ ............................... 65 4 PHYSIOLOGICAL RESPONSES OF HOLSTEIN HEIFERS WITH HIGH AND LOW GENOMIC MERIT FOR FERTILITY TRAITS ................................ ................................ ... 80 Material s and Methods ................................ ................................ ................................ ........... 81 Results ................................ ................................ ................................ ................................ ..... 87 Discussion ................................ ................................ ................................ ............................... 89 5 CONCLUSION ................................ ................................ ................................ ..................... 100 LIST OF REFERENCES ................................ ................................ ................................ ............. 102 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 111

PAGE 6

6 LIST OF TABLES Table page 2 1 Effect of PGF 2 formulation and estrus detection method on p regnancy per service and pregnancy loss ................................ ................................ ................................ ............. 44 3 1 Final logistic regression model of factors associated with characteristics of spontaneous estr o us ................................ ................................ ................................ ........... 70 3 2 Final logistic regression model of factors associated with characteristics of PGF induced estr o us ................................ ................................ ................................ ................... 71 3 3 Final logistic regression model of factors associated with the likelihood of pregnancy after the first service (75 3 d a fter service) ................................ ................................ ..... 72 3 4 Final logistic regression model of factors associated with hazard of pregnancy ............... 73 4 1 Primer reference and sequences for genes investigated by quantitative real time PCR. ... 94 4 2 Descriptive GDPR and GHCR data for the study population. ................................ ........... 94 4 3 Descriptive data for the study population. ................................ ................................ ......... 95

PAGE 7

7 LIST OF FIGURES Figure page 1 1 Activity and rumina tion data DataFlow2 ................................ ................................ ...... 23 2 1 Effect of prostaglandin ( PG ) F formulation on estrus detection by an automated estrus detection system (AED) within 7 days of first PGF treatment according to the phase of the estr o us cycle at PGF treatment.. ................................ ........................... 45 2 2 Effect of prostaglandin (PG) F formulation on interval from PGF treatment to onset of estrus only for mid diestrus heifers.. ................................ ................................ .... 45 2 3 Effect of prostaglandin (PG) F formulation on duration of estrus detected by an automated estrus detection system (AED) within 7 days of PGF treatment. ................. 46 2 4 Effect of prostaglanin (PG) F formulation on rumination nadir of estrus detected by an automated estrus detection system (AED) within 7 days of PGF tr eatment. ............ 46 2 5 Effect of prostaglandin (PG) F detected in estrus within 7 da ys of PGF treatment according to the estr o us cycle phase at PGF treatment. ................................ ................................ ................................ 47 2 6 Effect of prostaglandin (PG) F detected in estrus within 7 days of PGF treatment according to the estr o us cycle phase at PGF treatment. ................................ ................................ ................................ 47 2 7 Progesterone concentrations at the day of prostaglandin (PG) F treatment according t o PGF formulation. ................................ ................................ ................................ ....... 48 2 8. Effect of prostaglandin (PG) F formulation on progesterone concentrations ng/mL at estrus. ................................ ................................ ................................ ............................ 48 2 9 Effect of prostaglandin (PG) F formulation on estradiol concentrations at estrus. ........ 49 2 10 Effect of prostaglandin (PG) F formulation on interval from PGF to first service ..... 49 2 11 Effect of estrus detection method on interval from first to second service. ..................... 50 2 12 Effect of estrus detection method on interval from first prostaglandin (PG) F to pregnancy.. ................................ ................................ ................................ ......................... 50 3 1 Distribution of genetic merit for daughter pregnancy rate (GDPR) values in the study population.. ................................ ................................ ................................ ........................ 74 3 2 Distribution of genetic merit for heifer conception rate (GHCR) values in the study population ................................ ................................ ................................ .......................... 74

PAGE 8

8 3 3 Correlation of genetic merit for daughter pregnancy rate (GDPR) and h eifer conception rate (GHCR) .. ................................ ................................ ................................ .. 75 3 4 Duration of estrus according to genetic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR).. ................................ ................................ ................. 75 3 5 Rumination nadir according to genetic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR).. ................................ ................................ ................. 76 3 6 Activity peak according to genetic merit for daughter pregnancy rate (GDPR).. ............. 76 3 7 Heat index according to genetic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR).. ................................ ................................ ........................ 77 3 8 Interval from start of the reproductive program to first estrus detected by the AED according to GDPR quartile. ................................ ................................ .............................. 77 3 9 Interval from start of the reproductive program to pregnancy for heifers detected in estrus by an automated estrus detection device (AED) according to GDP R quartile. ....... 78 3 10 Interval from start of the reproductive program to pregnancy for heifers detected in estrus by an automated e strus detection device (AED) according to GHCR quartile ....... 78 3 11 Interval from start of the reproductive period to pregnancy for h eifers detected in estrus by visual observation (VIS) according to GDPR quartile. ................................ ...... 79 3 12 Interval from start of the reproductive period to pregnancy for heifers detected in estrus by visual observation (VIS) according to GHCR quartile. ................................ ...... 79 4 1 Genetic merit for daughter pregnancy arte (GDPR) and heifer conception rate (GHCR) breeding values in the study population.. ................................ ............................ 96 4 2 Ovulatory fo llicle size (all heifers) according to genetic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR) classes.. ................................ ................ 96 4 3 Estradiol concentrations at estrus (all heifers), according to genetic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR) classes.. ............... 97 4 4 P rogesterone concentrations at estrus, 7 and 14 days after estrus (all heifers), according to genetic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR) classes. ................................ ................................ ....................... 97 4 5 P rogesterone concentrations at estrus, 7, 14, 19 2, 28, and 35 days after estrus (only pregnant heifers 35 3 d after service), according to gene tic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR) classes. ............................... 98

PAGE 9

9 4 6 Interferon stimulated gene 15 (ISG 15) 19 2 days after estrus (only pregnant heifers 35 3 d after service), according to genetic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR) classes ................................ ......................... 98 4 7 Pregnancy specific protein B (PSPB) concentrations 19 2, 28, and 35 days after estrus (only pregnant heifers 35 3 d after service), according to genetic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR) classes. ................ 99 4 8 Insulin like growth factor 1 (IGF 1) concentrations at estrus, 7, 14, 19 2, 28 and 35 days after estrus (only pregnant heifers 35 3 days after service), according to genetic merit for daughter pregnancy rate (GDPR) and heifer conception rate (GHCR) classes. ................................ ................................ ................................ ................. 99

PAGE 10

10 LIST OF ABBREVIATIONS ACTB Beta actin AED Automated estrus detection monitoring device AI Artificial insemination CCR Cow conception rate CL Corpus luteum CLO Cloprostenol sodium CM$ Cheese merit DIN Dinoprost tromethamine DPR Daughter pregnancy rate ED Early diestrus ET Embryo transfer FM$ Fluid merit GDPR Genomic daughter pregnancy rate GHCR Genomic heifer conception rate GM$ Grazing merit HCR Heifer conception rate HH High for GDPR class / High for GHCR class HighGDPR High class for GDPR HighGHCR High class for GHCR HL High for GDPR class / Low for GHCR class IGF 1 Insulin like growth factor 1 IFN Interferon

PAGE 11

11 IOFC Income over feed cost ISG15 Interferon stimulated gene 15 LH Low for GDPR class /High for GHCR class LL Low for GDPR class/ Low for GHCR class LowGDPR Low class for GHCR LowGHCR Low class for GHCR ME Mete estrus MID Mid diestrus NM$ Lifetime net merit PBL Peripheral blood leucocytes PE Proestrus PG Prostaglandin PIE Prostaglandin induced estrus Preg/Serv Pregnancy per service PSPB Pregnancy specific protein B Q1 Quartile 1 Q2 Quartile 2 Q3 Quartile 3 Q4 Quartile 4 RIA Radioimmunoassay RPL19 Ribosomal protein L 19 SEM Standard error of the mean SNPs Single nucleotide polymorphisms

PAGE 12

12 SPE Spontaneous estrus TAI Timed artificial insemination T Ha l f life THI Temperature humidity index TMR Total mixed ration VIS Visual observation of estrus 21 d PregRate 21 d pregnancy rate 21 d ServRate 21 d service rate

PAGE 13

13 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science IMPACTS OF REPRODUCTIVE MANAGEMENT STRATEGIES AND GENE TIC MERIT ON REPRODUCTIVE PARAMETERS OF DAIRY HEIFERS By Anderson Veronese December 2017 Chair: Ricardo Carbonari Chebel Major: Veterinary Medical Sciences The objectives of this experiment was to evaluate the effects of PG F f ormulations and methods of estr us detection on physiological parameters, estr o us behavior, and reproductive performance of dairy heifers. Additionally, the association between fertility traits and physiological parameters, estr o us behavior, and reproductive performance of dairy h eifers are described. Holstein heifers (n = 1,019) were fitted with an automated estr us detection system ( AED ) and enrolled in the experiment around 11 months of age H eifers were assigned to the PGF ( CLO : cloprostenol sodium or DIN : di noprost thrometham ine) and estr us detection (AED: automated estrus detection or VSI : visual detection of estrus) treatments in a 2 x 2 factorial design. At birth, heifers were genotyped and genomic daughter pregnancy rate ( DPR ) and heifer conception rate ( HCR ) were collecte d. Treatment with CLO increased percentage of heifers detected in estrus within 7 days after treatment and reduced progesterone concentrations at estrus but it had no effect on hazard of pregnancy. Automated estrus detection tended to improve hazard of pre gnancy. Genomic d aughter pregnancy rate was associated with greater ovulatory follicle size, estradiol concentrations, and estrus expression, whereas GHCR was negatively as sociated with estr o us behavior Selection of PGF may be according to parameters other than

PAGE 14

14 efficacy because reproductive performance was similar between CLO and DIN. Herds with inefficient visual estrus detection may benefit from AED. Selection of heifers for DPR is likely to improve signs of estrus and overall reproductive performanc e, but additional information is needed before HCR may be used extensively as a selection parameter.

PAGE 15

15 CHAPTER 1 INTRODUCTION Importance of Reproductive Performance for Dairy Production The profitability of dairy herds is dependent on the efficiency of milk production, which may be simply evaluated as income over feed cost ( IOFC ). The IOFC is the difference between daily income from milk sales, which represents approximately 88% of the in come of dairy operations, and daily cost of feeding lactating cows, which represents approximately 50% of the cost of dairy operations (Santos et al., 2010) Milk production of dai ry cows is greatest during early lactation, with peak milk yield generally occurring around 5 to 8 weeks postpartum (Pollott, 2011) During early lactation, feed intake is insufficient to meet the energy and protein requirements of lactation and cows efficiently utilize body energy reserves for milk synthesis (Grummer et al., 2004) resulting in negative energy balance and maximum IOFC. The persistency of lactation, defined as the rate of decline in production after peak milk production (Cole and Null, 2009) is determined by genetics (Cole and Null, 2009) parity (Silvestre et al., 2009) use of recombinant bovine somatotropin (Van Amburgh et al., 1997) among other factors. Regardless of genetic composition of the herd, parity, or management strategies, the decline in milk yield is irrev ersible and IOFC declines sharply after approximately 100 d postpartum (Ribeiro et al., 2012) Reproductive efficiency determines the percentage of time between two calvings that cows spend in the most profitable phase of their lactation. For example, if maximum IOFC is obtained in the first 60 d postpartum, cows in a herd with an average calving in terval (interval between two consecutive calvings) of 16 months (485 d) would spend approximately 12% of this interval at maximum profitability. On the other hand, cows in a herd with an average calving interval of 12 months (364 d) would spend approximate ly 17% of this interval at maximum

PAGE 16

16 profitability. Aside from decreasing average IOFC of the herd, reproductive inefficiency results in herds having cows with a wider distribution of days postpartum and a larger proportion of cows with extended lactation. T herefore, herds with inefficient reproductive management may require changes in nutritional management to prevent losses. Cabrera and Kalantari ( 2016 ) reviewed the literature and determined that having 3 different TMR instead of 2 different TMR would increase the IOFC because it would reduce waste from feeding low producing cows energy and protein rich diets Other economic losses incurred from poor reproductive performance are increased culling because of reproductive failure (Machado et al., 2017) retention of larger number of replacement heifers (Kaniyamattam et al., 2016) reduced selection pressure on replacement heifers and, consequently, reduced genetic progress of the herd (Kaniyamattam et al., 2016) Thus, the objective of reproductive programs for lactating dairy cows is to increase 21 d pregnancy rates ( 21 d PregRate ; percentage of eligible cows that become pregnant every 21 d after the end of the voluntary waiting period), through improvements in 21 d service rate ( 21 d ServRate ; percentage of eligible cows that are serviced every 21 d after the end of the voluntary waiting period) and pregnancy per service ( P reg /Serv ; percentage of cows that conce ive after a service), and maximize annuity value per cow per year (Neves and LeBla nc, 2015) Reproductive Management of Dairy Heifers For the reasons discussed previously, reproductive performance of lactating dairy cows is extremely important for financial success of dairy operations and is generally an area in which dairy owners, managers, and consultants spend significant time and resources on. Cost of rearing replacement heifers are lower than costs of feeding and managing the lactating herd, but still represent s approximately 25% of the total cost of dairy operations (Santos et al., 2010) and is

PAGE 17

17 second only to feeding the lactating herd (Gabler et al., 2000) Despite the importance of optimum replacement heifer rearing for the profitability and future of dairy operations, managers and consultants generally dispend less resource s and time on reproductive management of heifers. Inefficient reproductive management of heifers may result in a wide range of age at first calving (Ettema and Santos, 2004) and increased rearing costs of heifers (Stevenson et al., 2000) Aside from the direct impact of reproductive ineffic iency on profitability of dairy herds by increasing age at first calving, inefficient reproductive management of heifers impact s milk production, reproductive performance, and h ealth during the first lactation and productive life ( Gabler et al., 2000; Ette ma and Santos, 2004). The goal of the reproductive management of Holstein heifers is to establish pregnancy at the appropriate size (60 to 65% of the mature body weight and 125 cm of wither height) at a reduced age (12 to 14 months of age) to shorten the i nterval from birth to the onset of the first lactation (Hoffman, 1997) Similarly to lactating cows, producers aim to increase 21 d PregRate of dairy heifers by increasing 21 d ServRate and P reg /Serv. In addition to improving heifer health and rate of growth, herds should adopt estr o us synchronization or ovulation synchroni zation protocols to assure that heifers are serviced soon after achieving the desired weight and height (Penteado and Dias, 2013) Furthermore, genetic selection for reproduction traits associated with faster establishment of pregnancy (e.g. daughter pregnancy rate DPR ) should also be a part of the long term management of replacement heifers (Jonas and de Koning, 2015) M anipulation of the E str o us Cycl e of Dairy Heifers Using Reproductive H ormones

PAGE 18

18 Reproductive management of heifers in US dairy herds is mainly based on visualization of spontaneous estrus (57.1%) and natural service (33.2%; NAHMS, 2007). Synchronization of estr o us of dairy heifers with PG F has the potential to increase 21 d ServRate compared with detection of spontaneous estrus without any detrimental effect to P reg /Serv (Stevenson et al., 2008) Prostaglandin F treatment induces luteolysis of corpus luteum causing a decrease in progesterone concentration, growth of the dominant follicle, and synchronized estr o us within 2 to 7 d after treatment (Martins et al., 2011a) Therefore, when PGF treatment is combined with accurate detection of estrus, 21 d ServRate and 21 d PregRate should be greater compared w ith visualization of spontaneous estrus. In dairies in which labor and systems for estrus detection are limiting factors, reproductive hormones (GnRH, PGF progesterone inserts) for synchronization of ovulation and fixe d time artificial insemination may be used. Ovulation synchronization protocols commonly used for lactating dairy cows (e.g. Ovsynch) tend to yield poor P reg /Serv in dairy heifers because while a large proportion of lactating dairy cows have 2 follicular waves, 44% of dairy heifers have thr ee or more follicular waves (Sartori et al., 2004) New ovulation synchronization protocols with reduced interval from fo llicular wave recruitment to induction of ovulation, however, have yielded acceptable P reg /Serv in dairy heifers (Lima et a l., 2013; Silva et al., 2015) T ools for Estrus Detection in Dairy H eifers The success of reproductive management of dairy heifers based on detected estrus is highly dependent on the efficiency and accuracy of estrus detection. The primary sign of estr us is an animal standing to be mounted (Forde et al., 2011) The duration of estrus was 14.0 0.8 h and the number of times heifers were mounted when in estrus was 50.1 6.4 events/heifer

PAGE 19

19 among beef heifers (Stevenson et al., 1994 ) and the duration of estrus of dairy heifers was 9.7 5.3 h (Yoshida et al., 2009). Therefore, unaided visualization of heifer standing to be mounted is quite difficult. Automated systems for detection of mounting ac tivity (e.g. HeatWatch) are rarely used on commercial farms because they are cumbersome and expensive. An indirect estrus detection method commonly used by dairy farms is tail painting and mounting patches (e.g. when heifers are mounted (Kamphuis et al., 2012) These systems require daily monitoring of heifers to re apply tail paint when necessary an d to diagnose which heifer are rubbed off or activated, likely a consequence of mounting activity in the previous 12 or 24 h. If such systems are used but heifers are not monitored daily inaccurate estrus detection and reduced P reg /Serv may result A mult itude of automated estrus detection monitoring ( AED ) systems are available in the USA and each one has its nuances. In general, AED systems determine the occurrence of estrus according to changes in patterns of behaviors such as steps/walking, activity, an d rumination (Chanvallon et al., 2014; Fricke et al., 2014b) Thus, most AED systems used in commercial dairy farms detect the o ccurrence of estrus based on secondary signs of estrus. Figure 1 1. depicts the activity and rumination graphs generated by the DataFlow 2 software (SCR Inc., Netanya, Israel), one of the commercially available AED systems. The AED system in question rec ords activity and rumination in 2 h intervals. Through a mathematical algorithm, the software calculates the momentary deviation of the activity/rumination from the average activity/rumination in the same time period during the previous 7 days. As seem in figure 1 1. (depicted by the cow mounting symbol). Although differences among AED systems exist, they

PAGE 20

20 generally utilize pedometers, 3D accelerometer, and microphone s to record steps, activity, and rumination, respectively and detect estrus through secondary signs Despite detecti ng estrus based on secondary si g n s of estrus, the sensitivity and specificity of AED detected estrus compared with ovulation determined by ultrasonography or visual observation of mounting activity are > 90% (V alenza et al., 2012; Dolecheck et al., 2015) Valenza et al. (2012) demonstrated a high level of agreement between an AED system based on changes in activity (SCR Engineers Ltd., Netanya, Israel) and a mounting detector (Kamar heatmount detector, Kamar Inc., Steamboat Springs, CO). Furthermore, standing to be mounted, the principal characteristic of cattle in estrus, was positively associated with duration of estrus and activity peak measured by an AED system (Silper et al., 2015b) In addition to provi ding continuous 24 h monitoring of individuals, AED systems remove human subjectivity from estrus detection (Reith and Hoy, 2017) S trategies for Selection of Dairy Heifers with Improved Reproductive P erformance A large number of genetic traits that affect overall profitability of dairy operations are available for dairy producers to select animals (Calus et al., 2013) The most common strategy used to overcome questions regarding which traits to select for is to use an index, which is a composite of the most important traits the dairy desires to select for or against (Dekkers, 2007) The USDA Animal Improvement Programs Laboratory provides a few indexes for general use, such as the lifetime net merit ( NM$ ), cheese merit ( CM$ ), fluid merit ( FM$ ), and grazing merit ( GM$ ). All these traits include production related traits (e.g. yields of milk, fat, and protein), fertility related traits (e.g. daughter pregnancy rate DPR heifer conception rate HCR and cow conception rate CCR ), somatic cell score, productive life, functional type traits, and calving ability traits (Cole, 2017)

PAGE 21

21 depen d on several farm and market specific conditions. Nonetheless, the focus on selection for Holstein cattle for milk yield and type traits with disregard for functional traits such as reproduction traits resulted in a significant decrease in reproductive eff iciency from the 50s to the early 2000s (Lucy, 20 01) Thus, modern selection for Holstein cattle has been partly focused reproduction traits such as DPR, introduced in 2004 (VanRaden et al., 2004) and CCR and HCR, introduced later (Kuhn et al., 2006 ) daughters compared with the population, whereas CCR and HCR are measures of the likelihood cows and heifers, respectively. With advancements in technology, sequencing the genome of dairy cattle has become less expensive and readily available, allowi ng producers to genotype large populations of animals (G arca R u iz et al., 2016) The large scale genotyping of cattle populations has produced high reliability genomic predicted transmitting ability ( GPTA ) values for several economically important traits (VanRaden et al., 2009) These traits started to be used first for the selection of sires, such that nearly all sires used f or semen collection in the USA today are genomically tested, and are now commonly used for selection of female cattle (Wiggans et al., 2011) Genomic testing has had a great impact on genetic selection of dairy cattle because of the improved reliability, the reduced generation interv al (faster selection of sires and dams with no need for progeny testing), and consequently faster genetic gain (G arca R uiz et al., 2016) Traits that have had historically low heritability, such as fertility traits, may benefit further from

PAGE 22

22 genomic selection because of the increased accuracy of parental information and r eliability of the test (Garca R uiz et al., 2016).

PAGE 23

23 Figure 1 1 Activity data (green bars) and deviation (brown line; panel A) and rumination data (purple bars) and deviation (brown line; panel B). DataFlow 2 (SCR Ltd., Netanya, Israel). B A

PAGE 24

24 CHAPTER 2 EFFECTS OF TWO DIFFERENT PROSTAGLANDIN F FORMULATIONS AND METHOD OF ESTR US DETECTION ON ESTR O US CHARACTERISTICS AND REPRODUCTIVE PERFORMANCE OF DAIRY HEIFERS Lifetime milk production and health of lactating cows are closely related to age and weight at first calving (Ettema and Santos, 2004) The objective of reproductive programs designed for dairy heifers is to have the majority of heifers calving in the stipulated time ( 24 months of age ) and weight (560 Kg o f live weight immediately after calving) to avoid large variations o f age at first calving (Stevenson et al., 2008) Prostaglandin ( PG ) F treatments fourteen days apart can be used to increase estrus rate, consequently increasing 21 d service rate ( 21 d ServRate ) and improve overall success of reproductive programs for dairy heifers (Steve nson et al., 2000; Lopes et al., 2013) Currently available PGF formulations include dinoprost tromethamine (DIN), a formulation composed of a molecule similar to endogenous PGF that has a relatively short ha l f life (T ~ 9 min ; Shrestha et al., 2012) and cloprostenol sodium (CLO), a formulation composed of a synthetic analogue of the PGF molecule that has a relat ively longer ha l f life (T ~ 3 h; Reeves, 1978) Different authors ( Martins et al. 2011a 2011 b; Pursley et al. 2012 ; Stevenson and Phatak, 2010 ) hypothesized that the c loprostenol sodium longer half life could induce faster and more thorough luteolysis and consequently increa se 21 d ServR ate and 21 pregnancy rate ( 21 d PregRate ) Results f rom those studies however were not consistent. Pursley et al. (2012) and Martins et al. (2011b) showed that CLO treatment reduced progesterone concentrations faster increased percentage of first lactation cows detected in estrus increased pregnancy per service ( Preg/Serv ) in cows bred 3 and 4 d after the treatment, and increased 21 d PregRate when compared with DIN treatment. Stevenson and Phatak (2010) showed that CLO treatment decreased percentage of cows with complete luteolysis compared with DIN treatment, but PGF

PAGE 25

25 formulation did not affect Preg/Serv or 21 d PregRate M ethodologies used by Pursley et al. (2012) and Martins et al. (2011b) were different from methodologies used by Stevenson and Phatak (2010) and could make comparison of their resu lts difficult I t is important to point out that both, Pursley et al. (2012) and Martins et al. (2011b) demonstrated that CLO treatment increased estrus detection and pregnancy rate among first lactation cows compared to DIN treatment The authors specula ted that reduced dry matter intake in primiparous cows compared with multiparous cows were the reasons of the different responses to CLO between primiparous and multiparous cows According to the authors, l ower dry matter intake in primiparous cows would r esult in lower hormonal clearance ( Sangsritavong et al. 2002 ; Wiltbank et al., 2006 ) and longer PGF 2 half life, greater luteolysis, and more intense behavioral estr o us. Prostaglandin F 2 however is metabolized and converted into a non active molecule (13, 14 Dihydro PGF 1 ) mainly in the lungs (Shrestha et al., 2012) To the best of our knowl edge, literature discusses possible association between high dry matter intake and high blood flow t o the liver (Sangsritavong et al., 2002; Wiltbank et al., 2006) but whether the lungs would also have high blood flow due to high dry matter intake is uncertain Furthermore, Pursley et al. (2012) and Martins et al. (2011b) did not measure dry matter intake of cows in their experiments Thus, reasons for CLO to improve percentage of primiparous c ows detected in estrus but not multiparous cows are still unknown Increasing estrus rate is important to improve reproductive performance (Lopes et al., 2013) ; howeve r, increased estrus rate must be accompanied by accurate estrus detection, otherwise it can result in reduced Preg/Serv compromising reproductive performance of dairy cows and heifers (Fricke et al., 2014b; Stevenson et al., 2014) More recently, automated estrus detection devices ( AED ), which determine estrus based on indirect signs (increased activity,

PAGE 26

26 reduced rumination, etc.), have become more efficient, accurate, and affordable and the ir use by dairy operations has increased (Denis Robichaud et al., 2016) Automated estrus detection devices have the ability to determine precisely the onset of estrus and the proper ti me of breeding, while minimizing human subjectivity during estrus detection on commercial farms (Fricke et al., 2017) Researchers have shown that AED can improve estrus detection rate and consequently 21 d Serv/Rat e in dairy cows (Fricke et al., 2014b; St evenson et al., 2014; Neves et al., 2015) Therefore, we hypothesized that treatment of dairy heifers with CLO would reduce progesterone at estrus and improve estrus detection estr o us characteristics, service rate, Preg/Serv and hazard of pregnancy co mpared with DIN treatment. Thus, our objectives were to evaluate progesterone and estradiol concentrations at estrus, percentage of heifers detected in estrus estr o us characteristics (e.g. duration, rumination nadir, and activity peak) measured by an AED, service rate, Preg/Serv, and hazard of pregnancy of heifers treated with CLO and DIN Furthermore, we hypothesized that the use of an AED for estrus detection would improve estrus detection rate, service rate, Preg/Serv and hazard of pregnancy of da iry heifers compared with detection of estrus by visual observation ( VIS ) Thus, our objectives were to evaluate service rate, Preg/Serv, and hazard of pregnancy in heifers detected in estrus by AED and VIS. Material s and Methods All procedures involving animals were approved by the animal care and use committee of the University of Florida (protocol #201609559). A nimals, Housing and M anagement This study was conduct from March 2016 to December 2016 in a commercial dairy herd with approximately 4,200 repl acement heifers, located in north central Florida. One thousand

PAGE 27

27 and nineteen heifers between 10 and 11 months of age were enrolled in the study. All heifers were genotyped within 2 months of birth using a 50k single nucleotide peptide platform commercially available (Clarifide, Zoetis, Parsippany, NJ). Data referent to genomic breeding values for daughter pregnancy rate ( DPR ) and heifer conception rate ( HCR ) recorded within 2 months of birth were used. Starting at 12 months of age heifers were weighed weekl y. Heifers for synchronization of the estr o us cycle. Heifers were housed in dry lots, with natural shade and no artificial cooling. The breeding pens had self locking head stanchions on the feed ing area. Heifers were fed twice daily (7:00 AM and 4:30 PM) a TMR formulated to meet or exceed the 800 to 1,000 g of li ve body weight per day (NRC, 2001). Weather data (daily average temperature, humidity and precipitation ) from the Gainesville airport, located approximately 40 miles east of the dairy, were used to calculate daily temper ature humidity index (THI) and prec ipitation The percentages of days during the 30 d prior to and during the 30 d after the start The cumulative precipitation during the 30 d prior to and during the 30 d after the sta rt of the reproductive program were recorded for each heifer. A utomated E strus D etection Device and Estr o us C haracteristics At enrollment, an AED (Heat Rumination Long Distance, SCR Inc., Netanya, Israel) mounted on a collar was fitted on the left, crania l area of the neck of all heifers. The device determined activity through an accelerometer and rumination based on sounds of regurgitation and mastication through a microphone. Activity and rumination data were recorded for 2 h intervals. Estrus was determ ined according to changes in patterns of activity and rumination

PAGE 28

28 within a 2 h interval compared with the average activity and rumination of the same period in the previous 5 and 7 d, respectively (DataFlow2 SCR Inc, Netanya, Israel). An internal algorith m of the DataFlow2 software produced a heat index (0 = no estrus, 100 = maximum) according to the intensity of changes in activity and rumination. Daily, study personnel evaluated the activity and rumination patterns of heifers determined to be in estrus by the DataFlow2 software. On the day heifers were moved to the breeding pen, heifers with heat index < 50, duration of estrus < 6 h, and no change in rumination time were determined to have changes in activity pattern due to pen movement and not due to e strus and were, therefore, not inseminated. Heat index, activity peak (0 = no estrus, 100 = maximum activity), and rumination nadir (maximum difference in rumination time within a 2 h period during estrus compared with the average rumination of the same pe riod in the previous 7 d) were recorded daily for all heifers in estrus. Study personnel evaluated each activity graph individually and determined the time of onset (2 h period when the activity threshold was surpassed), peak (2 h period when the activity change was maximum), and end (2 h period when the activity change was below the activity threshold) of estrus. Activity threshold was set at three fold above the average activity for the same period in the previous 5 d. Intervals from onset to peak of estr us and from onset to end of estrus were calculated. Characteristics of spontaneous estruses ( SPE ; estruses occurring before the start of the reproductive program) and PGF induced estruses ( PIE ; estruses occurring after the start of the reproductive progr am) were recorded Automated estrus detection monitor devices were removed from heifers at pregnancy diagnosis ( 28 d after service ) when heifers received a second service and when heifers were not detected in estrus within 28 d after the start of the reproductive program. Study Design and T reatments

PAGE 29

29 The study fallowed a completely randomized factorial design with 2 PGF formulations ( PGFTRT ) x 2 estrus detection methods ( EDTRT ). Before the start of the reproductive program h eifers were randomly ass igned to receive c loprostenol sodium ( CLO n = 505 ; Estrumate, Merck Animal Health, Summit NJ ) or dinoprost tromethamine ( DIN n = 490 ; Lutalyse, Zoetis, Parsippany, NJ ) and for estrus detection by an automated estrus detection system ( AED n = 530 ; Heat Rumination Long Distance, SCR Inc., Netanya, Israel ) or estrus detection by visualization of mounting activity or activation of a tail paint device ( VIS n = 465 ; Kamar, Kamar inc., Steamboat Springs, CO ) When heifers were eligible to start the reproducti ve program a list containing identification and respective treatments was available for study personnel at the dairy. Heifers were classified according to estr o us cycle phase into metestrus (d ay 0 to 3), early diestrus (day 4 to 6), mid diestrus (day 7 to 17 ) ), and no estrus observed. Heifers in metestrus were treated with the assigned PGF formulation 96 h later and heifers in early diestrus, mid diestrus and proestrus and heifers that had not had AED detected estrus were treated with the assigned PGF formulation immediately. H eifers not serviced within 14 d of the first PGF treatment received a second treatment with the same PGF formulation. H eifer s assigned to estrus detection method AED, did not receive a tail paint device at the beginning of the reproductive program, and were serviced at AED detected estrus informed by study personnel H eifers assigned to estrus detection method VIS had a tail pa int device placed by study personnel at the beginning of the reproductive program, and were serviced at estrus detected by farm personnel. According to the genetic selection program of the dairy, heifers were selected to receive artificial insemination (AI ) or to receive embryo transfer (ET). Heifers detected in estrus were AI on the same morning or received an embryo 6 to 9 days after estrus

PAGE 30

3 0 detection. As mentioned previously, a ll heifers had an A ED fitted and estrus es recorded, but estruses recorded by th e AED system were reported to farm personnel only for heifers enrolled in the AED treatment P regnancy Diagnoses and Reproductive D ata All heifers were examined for pregnancy by palpation per rectum of the uterine contents at 35 3 d after the detected e strus that resulted in AI or ET. Pregnant heifers were re examined by palpation per rectum of the uterine contents at 75 3 d of gestation. Pregnancy per service was calculated by dividing the number of heifers pregnant at 35 and 75 3 d after estrus by the number of heifers serviced. Pregnancy loss was calculated by dividing the number of heifers not pregnant at 75 3 d after service by the number of heifers pregnant 35 3 d after service. Data regarding sire of insemination, sire and dam of embryo tra nsfer, service technician, and reproductive outcomes were collected from an on farm software (PCDART; Dairy records management system, Chapel Hill, NC). B lo od S ampling In a subgroup of animals (n = 91), blood was sampled on the day of PGF treatment and on the first morn ing after estrus was detected (1 to 24 h after onset of estrus). Blood was sampled by puncture of the coccygeal vein or artery into evacuated tubes containing K2 EDTA (Vacutainer, Becton Dickinson, Franklin Lakes, NJ). Immediately upon collection, tubes were placed in ice and kept refrigerated until transported to the laboratory for processing within 2 to 3 h Blood tubes were centrifuged at 1 ,500 g for 15 min. A C until assayed. A nalysis of Pla sma S amples

PAGE 31

31 Progesterone concentrations in plasma was determined by radioimmunoassay ( RIA ) using a commercial kit (Coat a Count, MP Biomedical LLC, Solon OH). Plasma harvested from heifers on days 4 (~1 ng/mL) and 10 (~ 4 ng/mL) of the estr o us cycle were i ncorporated into each assay and used to calculate the CV. Intra and inter assay CVs were 5.8 and 10.5 % respectively. Serum concentration of estradiol 17 were quantified by RIA as described previously by Jinks et al. (2013). Intra assay coefficient of var iance for estradiol assays was 2.73%. S tatistical A nalysis Data was analyzed using SAS version 9.3 (SAS Institute Inc., Raleigh, NC). Continuous variables were analyzed by ANOVA using the MIXED procedure. Data were evaluated for normality and homogeneity of residuals after fitting the model. Data violating the assumptions of normality were transformed before analysis. Rumination nadir values were multiplied by 1 and transformed to the natural log before analysis. Thus, positive rumination nadir values wer e excluded (n = 16). Outlier detection was performed, and rumination nadir transformed values < 2 were considered outliers and removed from the analysis (n = 4) Likelihood of activity p pregnancy at 35 and 75 3 d after service and pregnancy loss between 35 and 75 3 d after service were analyzed by logistic regression using the LOGISTIC procedure of SAS. Hazard of estrus, first service, second service and pregnancy were analyzed by the Cox proportional hazard ratio using the PHREG procedure of SAS. Interval from PGF treatment that induced estrus to onset of estrus interval from PGF treatment to first service, interval from first service to second service and interval from PGF treatment to pregnancy were analyzed by the Wilcoxon test of equality using the LIFETEST

PAGE 32

32 procedure of SAS. C oncentrations of progesterone at PGF treatment and at estrus were analyzed using a non parametric procedure (Kurskal Wallis ; NPAR1WAY procedure ) Statistical models to evaluate characteris tics of PIE included PGF formulation, estr o us cycle phase at PGF treatment, number of PGF treatments prior to the first AED detected estrus, and pe and cumulative precipitation 30 days after the start of the reproductiv e program. Heifers that had been detected in estrus by the AED > 26 d prior to PGF 2 treatment (n = 10) and heifers detected in estrus > 168 h after the PGF treatment (n = 106) were not included in the analysis of PIE estr o us characteristics G enomic breeding values for DPR and HCR were also included in the model to control for a possible influence of genotype on the outcomes. For the analysis of the hazard of estrus models included PGF formulation, estr o us cycle phase at the time of PGF treatment number of PGF treatments prior to the first detected estrus, and pe and cumulative precipitation 30 days after the start of the reproductive program. G enomic breeding values for DPR and HCR were also included in the model to control for a possible influence of genotype on the evaluated outcomes. When PGF formulation and estr o us cycle phase at PGF treatment were associated with the hazard of estrus after PGF treatment, the Wilcoxon test of equality (LIFETEST procedure) was used to characterize the association between PGF formulation and estr o us cycle phase at PGF treatment and the interval from PGF treatment that induced estrus to estrus. Statistical models to evaluate pregnancy at 35 and 75 3 d after service p regnancy loss between 35 and 75 d for the first service included PGF formulation, estrus detection method, the interaction between estrus detection method and PGF formulation, estr o us cycle phase at the PGF treatment the interaction between estr o us cycle phase and PGF formulation, the

PAGE 33

33 interaction between estrus detection method and estr o us cycle phase, technician, and percentage reproductive program G enomic br eeding values for DPR and HCR were also included in the model to control for a possible influence of genotype on the evaluated outcomes. Sta tistical models to evaluate pregnancy at 35 and 75 3 d after service and pregnancy loss after ET services also inc luded embryo type (fresh in vivo produced embryo, frozen/thawed in vivo produced embryo, fresh in vitro fertilized embryo, and frozen/thawed in vitro fertilized embryo), embryo grade (excellent/good, fair, and poor), and days after estrus at embryo transfe r (6 to 9 d). For the analysis of the hazard of first service, models included PGF formulation, estrus detection method, the interaction between estrus detection method and PGF formulation, estr o us cycle phase at the PGF treatment, the interaction b etween estr o us cycle phase and PGF formulation, the interaction between estrus detection method and estr o us cycle phase and the reproductive program. G enomi c breeding values for DPR and HCR were also included in the model to control for a possible influence of genotype on the evaluated outcomes. When PGF formulation and estrus detection method were not associated with the hazard of the first service the Wi lcoxon test of equality (LIFETEST procedure) was used to characterize the association between PGF formulation on the interval from PGF treatment to first service For the analysis of the hazard of second service models included PGF formulation, estr us detection method, the interaction between estrus detection method and PGF formulation, and of the reproductive program. G enomic breeding values for DPR and HC R were also included in the model, to control for a possible influence of genotype on the evaluated outcomes. T he

PAGE 34

34 Wilcoxon test of equality (LIFETEST procedure) was used to characterize the association between PGF formulation and the interval from first service to second service For the analysis of the hazard of pregnancy, models included PGF formulation, estrus detection method, the interaction between estrus detection method and PGF formulation, estr o us cycle phase at the PG F treatment, the interaction between estr o us cycle phase and PGF formulation, the interaction between estrus detection method and estr o us cycle phase, and the reproductive program. G enomic breeding values for DPR and HCR were also included in the model to control for a possible influence of genotype on the evaluated outcomes. When PGF formulation and estrus detection method were not associated with the haz ard of pregnancy, the Wilcoxon test of equality (LIFETEST procedure) was used to characterize the association between estrus detection method and the interval from PGF treatment to pregnancy. Models for estradiol concentrations after estrus was detected included PGF formulation, estr o us cycle phase at PGF treatment, interval from onset of estrus to sample collec tion and pregnancy at 35 3 d Models for progesterone concentrations after estrus was detected only included PGF formulation. For each o f the statistical models collinearity was tested using the REG procedure of SAS considered collinear. In such cases, each variable was added to the model separately and the variable with the smallest P value was retained. A backward stepwise elimination of variables with P > 0.10 until variables that remained in the model had P < 0.10. Statistical significance was considered at P < 0.05 and a tendency was consider when 0.05 < P

PAGE 35

35 Results Age and Body Weight of S tudy P opulation Mean age at PGF treatment were CLO/AED = 376 4 d, CLO/VIS = 377 7 d, DIN/AED = 378 6 d, and DIN/VIS = 378 7 d Mean (SEM) weight at PGF treatment were CLO/AED = 388.3 26.3 Kg ; CLO/VIS = 385.6 27.5 Kg ; DIN/AED = 383.7 29.4 Kg ; DIN/VIS = 380.6 27.7 Kg E ffects of PGF 2 Formulation on Detection and Characteristics of E str o us and Concentrations of Progesterone and E stradiol The interaction between PGF formulation and estr o us cycle phase at treatment affected ( P = 0.02) the percentage of heifers detected in estrus within 7 d of PGF treatment, because a larger numerical difference between CLO and DIN was observed among heifers treated during early diestrus compared with heifers treated at mid diestrus and proestrus, respectively (Figure 2 1.). The interaction between PGF formulation and estr o us cycle phase at treatment affected ( P = 0.02 ) the hazard of estrus. T reatment with CLO reduced ( P < 0.01 ) the interval from PGF 2 treatment to estrus for mid diestrus heifers ( Figure 2 2. ), but PGF formulation did not affect the interval from PGF 2 treatment to estrus in early diestrus ( P = 0.95 ) and proestrus ( P = 0.55 ) heifers. Prostaglandin F formulation did not affect estrus duration ( P = 0.85; Figure 2 3 .) or rumination nadir ( P = 0.54; Figure 2 4 .). The interaction between PGF formulation and estr o us cycle phase affected ( P = 0.05 ) the percentage of heifers with activity peak 80 because a greater percentage of heifers in early diestrus and proestrus treated with CLO had activity peak 80 than heifers treated with DIN, whereas a greater percentage of heifers in mid diestrus treated

PAGE 36

36 with DIN had activity pe ak 80 than heifers treated with CLO (Figure 2 5 .) Similarly, the interaction between PGF formulation and estr o us cycle phase affected ( P < 0.01) the percentage of heifers with heat index 80 A greater percentage of heifers treated with CLO in early diestrus and proestrus had heat index 80 than heifers treated with DIN, whereas CLO treatment of heifers in mid diestrus resulted in slightly smaller percentage of heifers with heat index 80 than DIN treatment ( Figure 2 6 .). At PGF treatment p rogesterone concentrations were not different ( P = 0.27) between CLO a nd DIN treated heifers (Figure 2 7 .). After detection of estrus, CLO treated heifers had ( P = 0.03) lower progesterone concentrations than DIN treated heifers (Figure 2 8 .). Estradiol co ncentrations after detection of estrus were not ( P = 0.49) affected by PGF 2 formulation (Figure 2 9 .). Effects of PGF 2 Formulation and Estrus Detection Method on Reproductive P erformance Hazard of first service tended ( P = 0.06) to be greater for CLO than DIN treated heifers ( AHR = 1.14, 95% CI = 0.99 1.30 ) Estrus detection method did not ( P = 0.17) affect the hazard of first service. T he interaction between PGF formulation and estrus detection method did not ( P = 0.65) affect the hazard of first service Interval from first PGF treatment to first service was ( P = 0.04 ) shorter in CLO than in DIN treated heifers (Figure 2 10 .). Prostaglandin F formulation did not affect ( P = 0.87) the hazard of second service Automated estrus detection system tended ( P = 0.07 ) to i ncrease the hazard of second service compared with VIS ( AHR = 1.19, 95% CI = 1.00 1.43 ) T he interaction between PGF formulation and estrus detection method did not affect ( P = 0.58) the hazard of second service Interval from first service to second service was ( P = 0.04 ) shorter in heifers dete cted in estrus by the AED than in heifers detected in estrus by VIS (Figure 2 11.)

PAGE 37

37 Pregnancy at 35 3 d after first AI was not affected by PGF formulation ( P = 0.39), estrus detection method ( P = 0.95), or the interaction between PGF formulation and estrus detection method ( P = 0.47; Table 2 1. ) Pregnancy at 75 3 d after first AI was not affected by PGF formulation ( P = 0.29), estrus detection method ( P = 0.77), or the interaction between PGF formulation and estrus detection method ( P = 0.91; Table 2 1. ) Pregnancy loss from 35 to 75 3 d after first AI was not affected by PGF formulation ( P = 0.43), estrus detection method ( P = 0.39), or the intera ction between PGF formulation and estrus detection method ( P = 0.15; Table 2 1. ) Pregnancy at 35 3 d after first ET was not affected by PGF formulation ( P = 0.31), estrus detection method ( P = 0.42), or the interaction between PGF formulation and estrus detection method ( P = 0.26; Table 2 1. ). Pregnancy at 75 3 d after first ET was not affected by PGF formulation ( P = 0.76), estrus detection method ( P = 0.11), or the interaction between PGF formulation and estrus detection method ( P = 0.57; Table 2 1. ). Pregnancy loss from 35 to 75 3 d after first ET was not affected by PGF formulation ( P = 0.42), estrus detection method ( P = 0.12), or the interaction between PGF formulation and estrus detection method ( P = 0.67; Table 2 1. ). Pregnancy at 35 3 d after the second AI was not affected by PGF formulation ( P = 0.4 5 estrus detection method ( P = 0.21), or the interaction between PGF formulation and estrus detection method ( P = 0.90; Table 2 1. ) Pregnancy at 75 3 d after second AI was not affected by PGF formulation ( P = 0.7 9 ), estrus detection method ( P = 0.27), or the interaction between PGF formulation and estrus detection method ( P = 0.98; Table 2 1. ). Pregnancy loss from 35 to 75 3 d after second AI was not affected by PGF formulation ( P = 0.94), estrus

PAGE 38

38 detection method ( P = 0.78), or the interaction between PGF formulation and estrus detection method ( P > 0.99 ; Table 2 1. ). Pregnancy at 35 3 d after second ET was not affected by PGF formulation ( P = 0.76), estrus detection method ( P = 0.57), or the interaction between PGF formulation and estrus detection method ( P = 0.31; Table 2 1. ). Pregnancy at 75 3 d after second ET was not affected by PGF formulation ( P = 0.57), estrus detection method ( P = 0.72), or the interaction between PGF formulation and estrus detection method ( P = 0.18; Table 2 1. ). Pregnancy loss from 35 to 75 3 d after second ET was not affected by PGF formulation ( P = 0.15), estrus detection method ( P = 0.27), or the inter action between P GF formulation and estrus detection method ( P = 0.95; Table 2 1. ). Prostaglandin F formulation did not affect ( P = 0.59) the hazard of pregnancy. Hazard of pregnancy tended ( P = 0.07) to be greater for hei fers detected in estrus by AED than in heifers detected in estrus by VIS (AHR = 1.17 95% CI = 0.99 1.38 ) T he interaction between PGF formulation and estrus detection method ( P = 0.58) did not affected the hazard of pregnancy. Interval from first P G F treatment to pregnancy was ( P = 0.05 ) shorter for heifers detected in estrus by AED than for heifers detected in estrus by VIS ( Figure 2 12 .) Discussion T he interaction between PGF formulation and estr o us cycle phase at PGF 2 treatment affected the percentage of heifers detected in estrus within 7 d of treatment, because a large r numerical difference between CLO and DIN was observed among heifers in early diestrus than among heifers in mid diestrus and proestrus. Furthermore, CLO shortened the interval from PGF treatment to estrus among heifers in mid diestrus and reduced the progesterone concentration at estrus compared with DIN treatment Prostaglandin F 2 formulation however,

PAGE 39

39 did not affect estradiol concentration at estrus or estrus duration and rumination nadir. The interaction bet ween PGF 2 formulation and of estrous cycle phase at PGF 2 treatment was associated with the percentage of heifers with activity peak and heat index 80, because a larger numerical difference in the percentage of heifers with activity peak and heat index 80 between CLO and DIN was observed among heifers in early diestrus. G rowth of a large follicle capable to produce enough estradiol to trigger estrus and ovulation is dependent on luteal regression (Goravanahally et al., 2009) After luteal regression occur s interval from PGF treatment to onset of estrus is dependent on age and maturity of the largest follicle at the time of treatment (Martins et al., 2011b) Prostaglandin F luteolytic efficacy is highly dependent on the estr o us cycle phase when the treatment is applied (Valldecabres Torres et al., 2012; Ferraz Junior et al., 2016) Newly formed corpus luteum have concentrations of PGF receptors similar to mature corpus luteum, but the ability of exogenous PGF to induce luteolysis is reduced before day 5 or 6 of the estr o us cycle (Wenzinger and Bleul, 2012) After day 16 of the estr o us cycle, if maternal recognition of pregnancy is not established, oxytocin bind s to its receptor in the uterus which propagates secretion of endogenous PGF 2 and regression of the corpus lute um occurs spontaneously, with no nee d for exogenous PGF 2 treatment (Forde et al., 2011) In the current experiment we hypothesized that dai ry heifers would benefit from the longer half life of CLO, which would increase the percentage of heifers detected in estrus and the hazard of estrus compared with CLO than DIN treatment We used an AED to determine exact interval and characteristics of estr o us to minimize human subjective during evaluation of estr o us characteristics In the current experiment the differences in percentage of heifers detected in estrus between CLO and DIN treatments was greatest among he ifers treated at early diestrus followed by heifers treated at mid diestrus and proestrus respectively Since recently formed

PAGE 40

40 corpus luteum are not fully responsive to PGF 2 treatments (Wenzinger and Bleul, 2012) we speculate that the longer half life of CLO allowed a longer exposure of the newly formed corpus luteum to PGF 2 increasing the likelihood of luteolysis. On the other hand, heifers in proestrus benefited the least from CLO because they likely were undergo ing or had undergone spon taneous luteolysis (Forde et al., 2011) The benefits of CLO to h eifers in mid diestrus was intermediary likely because a t mid diestrus a fully functional corpus luteum is present (Forde et al., 2011) and the h alf life of the PGF 2 would not be as critical to induce complete luteolysis Cloprostenol treatment reduced the interval from PGF treatment to estrus, but only in mid diestrus heifers. Since a greater proportion of heifers treated with CLO in early dies trus were detect in estrus, we expected CLO also to reduce the interval to estrus in early diestrus, not only in mid diestrus heifers. Estr o us characteristics measured with an AED were previously associated with physiological signs of estrus such as clear vaginal mucus, uterine tone, visual mounting activity and standing to be mounted behavior (Silper et al., 2015). Because emergence of a dominant follicle capable of producing enough es tradiol concentrations to trigger estrus expression should occur within 7 d of PGF treatment (Forde et al. 2011) we only used heifers detected in estrus within 7 d of PGF treatment in the analysis of estr o us characteristics. Since CLO reduced progesterone concentrations at estrus, we expected it also to allow greater follicle growth and estradiol concen trations, and in turn produce more intense estrus compared with DIN treatment Prostaglandin F formulation however, did not affect estradiol concentrations at estrus, estrus duration, and rumination nadir Nonetheless, as discussed previously, treatment of heifers in early diestrus and proestrus with CLO resulted in greater percentage of heifers with activity peak and heat index 80. Thus, results from the current experiment suggest that although

PAGE 41

41 progesterone concentrations at estrus were lower in CLO than in DIN treated heifers, reduction in progesterone concentrations in DIN treated heifers was likely enough to allow follicle growth and a rise in estradiol concentration to trigger estrus. Treatment of dairy heifers with CLO reduced interval f rom PGF 2 treatment to first service Reduced interval from PGF treatment to first service most likely was due to the effects of CLO on percentage of heifers detected in estrus within 7 d of the first PGF treatment. Pregnancy at 35 and 75 d after estru s and pregnancy loss from 35 to 75 d after estrus, for AI and ET services were not affected by PGF formulation. These result s are in agree ment with data by Stevenson and Phatak (2010), but are not in agree ment with data by Pursley et al. (2012) and Martins et al. (2011b) who demonstrated that CLO treatment increased Preg/Serv in primiparous cows. Since progesterone concentrations at estrus were lower for CLO treated heifers, we expected it could improve Preg/Serv as previously reported by Colazo et al. ( 2017) E stradiol concentrations however, were not affected by PGF formulation and were enough to trigger estrus. Furthermore, m ean progesterone concentrations at estrus among DIN treated heif ers was only 0.11 ng/mL. Colazo et al. (2017) demonstrated that progesterone concentration > 0.5 ng/mL reduced Preg/Serv in cows. Thus, no practical benefit of the lower progesterone concentrations resulting from the CLO treatment was observed in the curre nt experiment Although CLO treated heifers had increased first service rate, hazard of pregnancy was not affected by PGF 2 formulation. Estrus detection method did not affect the hazard of first service. Automated estrus detection monitoring system however, increased the hazard of second service of non pregnant heifers and tended to increase the hazard of pregnancy. Automated estrus detection systems allow for 24 h daily estrus detection (Fricke et al., 2014 ) Giordano et al. ( 2015) showed that

PAGE 42

42 AED increased insemination of cows in estrus. Similarly, Fricke et al. ( 2014 ) showed that the interval to re insemination of cows was shortened by the use of an AED. In the current experiment we expected AED to increase hazard of first service and second service The lack of effect of AED on hazard of first service may indicate that e strus detection by farm personnel was more intense for heifers tha t had not been service d compared with heifers that had been serviced T hus heifers that did not conceive after the fi rst service benefited the most from the AED in current experiment Estrus detection using an AED did not improve Preg/Serv or pregnancy los s either on AI or ET services. N umerically however, Preg/Serv was greater for heifers detected in estrus by the AED com pared with VIS Because AED increased hazard of second service and a numerical increase in Preg/Serv was noted AED increased hazard of pregnancy in heifers in the current expeiment T reatment of dairy heifers with CLO treatment increased estrus dete ction within 7 d of treatment, tended to increase first service rate, and reduced progesterone concentrations at estrus compared with DIN treatment These responses, however, are somewhat dependent on phase of the estrous cycle when heifers were treated wi th PGF 2 E stradiol concentrations and estrus characteristics however, were minimally affect by PGF 2 formulation Furthermore, PGF 2 did not affect Preg/Serv pregnancy loss, or hazard of pregnancy. Results presented here in suggest that PGF formulation may have a small or null impact on overall reproductive performance of dairy heifers and selection of PGF 2 formulation for dairy heifers should be according to others characteristics than efficacy Use of an AED for detection of estrus in da iry heifers tended to increase hazard of second service and pregnancy in a commercial dairy farm. Although improvements in reproductive performance observed herein can potentially increase profitability of heifer operations

PAGE 43

43 economical feasibility of the u se of an AED for dairy heifers will vastly vary according to the type of reproductive program used the accuracy of estrus detection at the farm level

PAGE 44

44 Table 2 1 Effect of PGF 2 formulation and estrus detection method on p regnancy and pregnancy loss CLO DIN P value AED VIS AED VIS PGFTRT EDTRT PGFTRT x EDTRT First Service Pregnant Day 35 AI 49.6 47.1 51.6 54.2 0.39 0.95 0.47 Day 75 AI 45.9 43.3 50.0 50.0 0.29 0.77 0.91 Day 35 ET 34.0 26.9 37.8 36.5 0.31 0.42 0.26 Day 75 ET 29.3 21.3 33.3 27.0 0.76 0.11 0.57 Pregnancy loss AI 7.6 8.2 3.1 7.7 0.43 0.75 0.15 ET 14.0 20.7 11.9 26.2 0.42 0.12 0.67 Second Service Pregnant Day 35 AI 37.0 26.7 36.2 25.0 0.45 0.21 0.90 Day 75 AI 33.3 22.2 36.2 25.0 0.79 0.27 0.98 Day 35 ET 31.5 39.5 31.6 26.7 0.76 0.57 0.31 Day 75 ET 29.6 37.2 29.0 20.0 0.57 0.72 0.18 Pregnancy loss AI 10.0 16.7 0.0 0.0 0.94 0.78 > 0.99 ET 5.9 5.9 8.3 25.0 0.15 0.27 0.95 PGFTRT = PGF 2 formulation used; CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit, NJ); DIN = heifers were treated with PGF 2 formulation dinoprost tromethamine (Lutalyse, Zoetis, Parsippany, NJ) EDTRT = Es trus detection method used; AED = Automated estrus detection (Heattime, SCR Inc., Netanya, Israel); VIS = Estrus detection based on visual observation and mounting device activation (Kamar heatmount detector, Kamar Inc., Steamboat Springs, CO). AI = Artifi cial Insemination ET = Embryo Transfer

PAGE 45

45 Figure 2 2 Effect of prostaglandin ( PG ) F formulation on estrus detection by an automated estrus de tection system (AED ) within 7 days of first PGF treatment according to the phase of the estr o us cycle at PGF treatment CLO = h eifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Ani mal Health, Summit NJ); DIN = h eifers were treated with PGF 2 formulation dinoprost tromethamine (L utalyse, Zoetis, Parsippany NJ) PGF 2 formulation P < 0.01, e str o us cycle phase at PGF 2 treatment P < 0.01 PGF 2 formulation x estr o us cycle phase at PGF 2 treatment P = 0.02 Figure 2 2. Effect of prostaglandin ( PG ) F formulation on interval from PGF treatment to onset of estrus only for mid diestrus heifers. Mean SEM and median interval from PGF2 treatment to estrus: CLO = 58.3 1.6 and 48.9 h DIN = 72.8 2.4 and 55.6 h CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit NJ); DIN = heifers were treated with PGF 2 formulation dinoprost tromethamine (Lutalyse, Zoetis, Parsippany NJ) PGF treatment P < 0.01. 0 20 40 60 80 100 Early diestrus Mid diestrus Proestrus Heifers in estrus within 7 days of first PGF 2 treatment, % CLO DIN 0 10 20 30 40 50 60 70 80 90 100 Heifers not detected in estrus, % Hours since PGF 2 treatment CLO DIN

PAGE 46

46 Figure 2 3. Effect of prostaglandin ( PG ) F formulation on duration of estrus detected by an automated estrus de tection system (AED ) within 7 days of PGF treatment CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit NJ); DIN = he ifers were treated with PGF 2 formulation dinoprost tromethamine (Lutalyse, Zoetis, Parsippany NJ) PGF 2 treatment P = 0.85 Figure 2 4. Effect of p rostaglanin (P G ) F formulation on rumination nadir of estrus detected by an automated estrus de tection system (AED ) within 7 days of PGF treatment CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit NJ); DIN = heifers were treated with PGF 2 formulation dinoprost tromethamine (Lutalyse, Z oetis, Parsippany NJ) PGF 2 formulation P = 0.54 0 2 4 6 8 10 12 14 16 18 Estrus duration, h CLO DIN -40 -38 -36 -34 -32 -30 -28 -26 -24 -22 -20 Rumination nadir, min/day CLO DIN

PAGE 47

47 Figure 2 5. Effect of prostaglandin ( PG ) F on percentage of heifers with activity peak detected in estrus within 7 days of PGF treatment according to the estr o us cycle phase at PGF treatment. CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit NJ); DIN = heifers were treated with PGF 2 formulation dinoprost tromethamine (Lutalyse, Zoetis, Parsippany NJ) PGF 2 formulation P = 0.62 estr o us cycle phase at PGF 2 treatment P < 0.01 PGF 2 formulation x estr o us cycle phase at PGF 2 treatment P = 0.05 Figure 2 6. Effect of prostaglandin (PG) F on percentage of heifers with heat index detected in estrus within 7 days of PGF treatment according to the estr o us cycle phase at PGF treatment. CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit NJ); DIN = heifers were treated with PGF 2 formulation dinoprost tromethamine (Lutalyse, Zoetis, Parsippany NJ) PGF 2 formulation P = 0.02 estr o us cycle phase at PGF 2 treatment P < 0.01 PGF 2 formulation x estr o us cycle phase P < 0.01 30 40 50 60 70 80 90 Early diestrus Mid diestrus Proestrus 80, % CLO DIN 30 40 50 60 70 80 90 100 Early diestrus Mid diestrus Proestrus 80, % CLO DIN

PAGE 48

48 Figure 2 7. Progesterone concentrations at the day of prostaglandin ( PG ) F treatment according to PGF formulation CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit NJ); DIN = heifers were treated with PGF 2 formulation dinoprost tromethamine (L utalyse, Zoetis, Parsippany NJ) PGF 2 formulation P = 0.27 Figure 2 8. Effect of prostaglandin ( PG ) F formulation on progesterone concentrations at estrus. CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit NJ); DIN = heifers were treated with PGF 2 formulation dinoprost tromethamine (Lutalyse, Zoetis, Parsippany NJ) PGF 2 formulation P = 0.03 0 1 2 3 4 5 6 7 Progesterone concentrations at PGF 2 treatment, ng/mL CLO DIN 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 Progesterone concentrations at estrus ng/mL CLO DIN

PAGE 49

49 Figure 2 9. Effect of prostaglandin ( PG ) F formulation on estr adiol concentrations at estrus CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit NJ); DIN = heifers were treated with PGF 2 formulation dinoprost tromethamine (Lutalyse, Zoetis, Parsippany NJ) PGF 2 formulation P = 0.49 Figure 2 10. Effect of prostaglandin ( PG ) F formulation on interval from PGF to first service Mean ( SEM ) and median days to first service: CLO = 4.5 0.2 and 3 d DIN = 4.9 0.3 and 3 d Prostaglandin F 2 treatment: CLO = heifers were treated with PGF 2 formulation cloprostenol sodium ( Estrumate, Merck Animal Health, Summit NJ); DIN = heifers were treated with PGF 2 formulation dinoprost tromethamine (Lutalyse, Zoetis, Parsippany NJ) PGF formulation P = 0.07 0 2 4 6 Estradiol, pg/mL CLO DIN 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Heifers not serviced, % Days since first PGF 2 treatment CLO DIN

PAGE 50

50 Figure 2 11. Effect of estrus detection method on interval from first to second service Mean ( SEM ) and median days to second service : AED = 22.5 0.3 and 22 d VIS = 23.3 0.32 and 23 d Estrus detection method: AED = Automated estrus detection (Heattime, SCR Inc., Netanya, Israel ); VIS = Estrus detection based on visual observation and mounting device activation ( Kamar heatmount detector, Kamar Inc., Steambo at Springs, CO ). E strus detection method P = 0.04 Figure 2 12 Effect of estrus detection method on interval from first prostaglandin ( PG ) F to pregnancy Mean ( SEM ) and median to pregnancy : AED = 39.5 1.4 and 33 d, VIS = 43.9 1.5 and 44 d Estrus detection method: AED = Automated estrus detection (Heattime, SCR Inc., Netanya, Israel ); VIS = Estrus detection based on visual observation and mounting device activation ( Kamar heatmount detector, Kamar Inc., Steamboat Springs, CO ). Estrus detec tion method P = 0.05 0 10 20 30 40 50 60 70 80 90 100 0 3 6 9 12 15 18 21 24 27 30 33 36 39 Heifers not second serviced, % Days since first service AED VIS 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 Heifers not pregnant, % Days since first PGF 2 treatment AED VIS

PAGE 51

51 CHAPTER 3 ASSOCIATION AMONG GENETIC MERIT FOR REPRODUCTION TRAITS AND ESTR O US CHARACTERISTICS AND FERTILITY OF HOLSTEIN HEIFERS Reproductive performance is extremely important to maximize the profitability of dairy operations (Giordano et al., 2012) Factors such as reproducti ve management, nutrition, health, and genetics affect reproductive outcomes directly or indirectly. Genetic selection of da iry breeds until the early 2000 s was mainly focused on production traits, while disregarding reproducti on traits (Lucy, 2001) It is believed that such strategy contributed for th e selection of cattle with reduced estrus expression and, consequently, reduced estrus detection and reproductive performance in modern dairy operations (Lopez et al., 2005) Although recent advancements in reproductive management has allowed for the insemination of cows and heifers following ovulation synchronization protocols, even animals subjected to such protocols have greater pregnancy per service ( Preg/Serv ) when they d isplay estrus at the time of fixed time service. In a recent study, estrus expression was associated with increased fertility and decreased pregnancy losses following timed artificial insemination and fixed time embryo transfer ( TET ; Pereira et al., 2016) Automated estrus detection monitoring devices ( AED ), based on changes in walking, activity and rumination patterns, have become m ore reliable for estrus detection and are being used in a growing number of dairies (Fricke et al., 2017) This technology has allowed the recording of estrus events and estr o us characteristics (duration, intensity, etc.) from a large number of animals in a uniform manner. Burnett et al. ( 2017) demonstrated that estrous characteristics, such as duration was positively associated with pregnancy per service ( Preg/Serv ) following artificial insemination ( AI ) in dairy cows. Studies that evaluate the associations among gen etic merit, physiological parameters, and estr o us characteristics present a

PAGE 52

52 unique opportunity to understand how new strategies for ge netic selection may affect estr us behavior and Preg/Serv. In a series of experiments, Kommadath et al. ( 2011, 2013, 2017 ) and Woelders et al. ( 2014) recorded physiological estr o us behavior signs visually and assigned an estrus score to dairy cows using a score previously described by Roelofs et al. ( 2005) Among the estr o us behavior signs evaluated to assign estrus scores were mounting activity and standing to be mounted ( Kommadath et al. 2011, 2013, 2017 ; Woelders et al. 2014 ) estrus signs that were positively associated with estrus duration and activity peak measured by an AED (Silper et al., 2015b) After recording estrus scores from s everal estr o us cycles, cows we re slaughter either at mid diestrus or at estrus and had brain collected for gene expression analyses. In these studies estrus score was associated with a substantial number of genes expressed in different areas of the brain ( Kommadath et al. 2011, 2013, 2017 ; Woelders et al. 2014) suggesting a possible genetic component driving estr o us behavior in dairy cows. repr oduction traits such as daughter pregnancy rate ( DPR ), introduced in 2004 (VanRaden et al., 2004) and heifer conception rate (HCR), introduced in 2013 D aughter pregnancy rate is a measure of the hazard population (AIPL, 2013) W ith the advancement of genomic selection tools in recent years genetic gain s of selected traits in the US Holstein cattle population has been substantial (Ga rca R uiz et al., 2016) Despite improvements in the US Holstein population regarding reproduction traits such as interval from calving to first AI, 21 d pregnancy rate ( 21 d PregRate ; percentage o f eligible cows that become pregnant within a 21 d period) and calving interval there is a lack

PAGE 53

53 of information regarding the association among these genomic traits and estrus expression and estr o us characteristics. The hypothesis of the current study wa s that genomic merit values for DPR ( GDPR ) and for HCR ( GHCR ) are associated with estr o us characteristics hazard of service P reg /Serv pregnancy loss and hazard of pregnancy in Holstein heifers. Therefore, the objectives of the current study were to eva luate the association among GDPR and GHCR and estr o us characteristics hazard of service, Preg /Serv pregnancy loss and hazard of pregnancy of Holstein heifers. Material s and Methods All procedures involving animals were approved by the animal care and use committee of the University of Florida (pro tocol #201609559). A nimal, Housing, and M anagement This study was conduct from March 2016 to December 2016 in a commercial dairy herd with approximately 4,200 replacement heifers, located in north central Flo rida. One thousand and nineteen heifers between 10 and 11 months of age, were enrolled in the study. All heifers were genotyped within 2 months of birth using a 50k single nucleotide peptide platform commercially available (Clarifide, Zoetis, Parsippany, NJ). Data referent to genomic breeding values for DPR and HCR recorded within 2 months of birth were used. Starting at 12 months of age heifers were weighed weekly. Heifers with BW 340 kg were moved to a breeding pen and were treated with prostaglandin ( PG ) F for synchronization of estr o us. Heifers were housed in dry lots, with natural shade and no artificial cooling. The breeding pens had self locking head stanchions on the feed ing area Heifers were fed twice daily (7:00 AM and 4:30 PM) a TMR formulat ed to meet or exceed the nutritional requirements of Holsteins heifers weighing 340 kg

PAGE 54

54 of live body weight and gaining 800 to 1,000 g of live body weight per day (NRC, 2001). Weather data (daily average temperature, humidity and rain precipitation ) from the Gainesville airport, located approximately 40 miles east of the dairy, were used to calculate daily temperature humidity index ( THI ) T he percentages of days during the 30 d prior to and during the 30 d after the start of the reproductive program with THI 72 were recorded for each heifer. Cumulative p recipitation during the 30 d prior to and during the 30 d after the start of the reproductive program were recorded for each heifer. A utomated E strus D etection and Estr o us C haracteristics At enrollment, an AED (Heat Rumination Long Distance, SCR Inc., Netanya, Israel) mounted on a collar was fitted on the left, cranial area of the neck of all heifers. The device determined activity through an accelerometer and rumination based on sounds of regurgitation and mastication through a microphone. Activity and rumination data were recorded in 2 h intervals. Estrus was determined according to changes in patterns of activity and rumination within a 2 h interval compared with the average activity and rumination of the same period in the previous 5 and 7 d, respectively (DataFlow2 SCR Inc, Netanya, Israel). An internal algorithm of the DataFlow2 software produced a heat index (0 = no estrus, 100 = maximum) according to the intensity of changes in act ivity and rumination. Daily, study personnel evaluated the activity and rumination patterns of heifers determined to be in estrus by the DataFlow2 software. On the day heifers were moved to the breeding pen, heifers with heat index < 50, duration of estru s < 6 h, and no change in rumination time were determined to have changes in activity pattern due to pen movement and not due to estrus and were, therefore, not inseminated. Heat index, activity peak (0 = no estrus, 100 = maximum activity), and rumination nadir (maximum difference in rumination time within a 2 h period during estrus compared with the average rumination of the

PAGE 55

55 same period in the previous 7 d) were recorded daily for all heifers in estrus. Study personnel evaluated each activity graph individ ually and determined the time of onset (2 h period when the activity threshold was surpassed), peak (2 h period when the activity change was maximum), and end (2 h period when the activity change was below the activity threshold) of estrus. Activity thresh old was set at three folds above the average activity for the same period in the previous 5 d. Intervals from onset to peak of estrus and from onset to end of estrus were calcu l ated. Characteristics of spontaneous estruses ( SPE ; estruses occurring before t he start of the reproductive program) and PGF induced estruses ( PIE ; estruses occurring after the start of the reproductive program) were recorded. Automated estrus detection monitors devices were removed from heifers at pregnancy diagnosis 28 d after se rvice, when heifers received a second service and when heifers were not detected in estrus within 28 d after the star t of the reproductive program. R eproductive M anagement From enrollment to the start of the reproductive program all estruses were record ed. live body weight) they were classified according to estr o us cycle phase into metestrus ( ME ; day 0 to 3), early diestrus ( ED ; day 4 to 6), mid diestrus ( MID ; day 7 to 17 ), and proestrus ( PE ; day ), and no estrus observed. Heifers in metestrus were treated with PGF 96 h later and heifers in early diestrus, mid diestrus and proestrus and heifers that had not had AED detected estrus were treated with PGF immediately. Two PGF formulations were used (cloprostenol sodium, Estrumate, Merck Animal Health, Summit NJ; dinoprost tromethamine, Lutalyse, Zoetis, Parsippany, NJ). Fourteen days after the first PGF treatment, heifers not detected in estrus received a second treatment with the same PGF formulation. Despite all heifers being fitted

PAGE 56

56 with the AED, 537 heifers were serviced at AED detected estrus, whereas 482 heifers were serviced at estrus detected by farm personnel based on visualization of mounting activity or activation of a tail paint device (Kamar, Kamar inc., Steamboat Springs, CO ). According to the genetic selection program of the dairy, heifers were selected to receive artificial insemination ( AI ) or to receive embryo transfer ( ET ). Heifers detected in estrus were AI on the same morning or received an embryo 6 to 9 days after estrus detection. P regnancy Diagnoses and Reproductive D ata All heifers were examined for pregnancy by palpation per rect um of the uterine contents at 35 3 d after the detected estrus that resulted in AI or ET. Pregnant heifers were re examined by palpation per rectum of the uterine contents at 75 3 d of gestation. Pregnancy per service was calculated by dividing the number of heifers pregnant at 35 and 75 3 d after estrus by the number of heifers serviced. Pregnancy loss was calculated by dividing the number of heifers not pregnant at 75 3 d after service by t he number of heifers pregnant 35 3 d after service. Data regarding sire of insemination, sire and dam of embryo transfer, service technician, and reproductive outcomes were collected from an on farm software (PCDART; Dairy records manage ment system, Chapel Hill, NC). S tatistica l A nalysis Data was analyzed using SAS version 9.3 (SAS Institute Inc., Raleigh, NC). Continuous variables were analyzed by ANOVA using the MIXED procedure. Data were evaluated for normality and homogeneity of residuals after fitting the model. Data violating the assumptions of norma lity were transformed before analysis. Rumination nadir values were multiplied by 1 and transformed to the natural log before analysis. Thus, positive rumination nadir values were

PAGE 57

57 excluded (SPE = 6, PIE = 16). Outlier detection was performed, and ruminati on nadir transformed values < 2 for SPE (n = 4) and < 2.2 for PIE (n = 4) were considered outliers and removed from the analysis. Interval from onset of estrus to activity peak was square root transformed. Genetic merit for DPR and HCR are the predicted tr ansmitting ability of a trait from the parent to its offspring. The GDPR and GHCR values used in this study were referent to the individuals used in the study; therefore, GDPR and GHCR values were multiplied by 2. pregnancy loss between 35 and 75 3 d after estrus were analyzed by logistic regression using the LOGISTIC procedure. The hazard of estrus, of first service, and of pregnancy were analyzed by the Cox proportional hazard ratio using the PHREG procedure. Interval from the start of the reproductive program to the onset of first estrus and interval from the start of the reproductive program to establishment of pregnancy were analyzed by the Wilcoxon test of equality using the LIFETEST procedure. Statistical models to evaluate SPE characteristics included GDPR (linear and quadratic), GHCR (linear and quadratic), the interaction between GDPR and GHCR, and percentage of days pitation 30 days before the start of the reproductive program. Statistical models to evaluate PIE characteristics included GDPR (linear and quadratic), GHCR 72 and cumulative precipitation 30 days after the start of the reproductive program, PGF formulation, estr o us cycle phase at PGF treatment, and number of PGF treatments prior to the first detected estr o us. Heifers that had been detected in estrus by the A ED > 26 d prior to the PGF treatment (n = 10 ) and heifers detected in estrus > 168 h after the PGF treatment (n = 106 ) were not included in the analysis of PIE characteristics. Nonetheless, heifers that displayed

PAGE 58

58 estrus > 168 h after the PGF 2 and heifers that did not display estrus following PGF 2 treatment were censored for the purpose of the Cox proportional hazard ratio and Wilcoxon test of equality analyses. For the analysis of the hazard of estrus after the start of the reproductive progr am, models included GDPR (linear and quadratic), GHCR (linear and quadratic), the interaction between the start of the reproductive program, PGF formulation, and e str o us cycle phase at the time of the start of the reproductive program. When GDPR and GHCR were associated with the hazard of estrus after the start of the reproductive program, these variables were divided into quartile and the Wilcoxon test of equality (LIFETEST procedure) was used to characterize the association between GDPR and GHCR and the interval from the start of the reproductive program and first detected estrus. Statistical models to evaluate the likelihood of pregnancy and pregnancy loss includ ed GDPR (linear and quadratic), GHCR (linear and quadratic), the interaction between GDPR and GHCR, PGF formulation, estrus detection method, the interaction between estrus detection method and PGF formulation, estr o us cycle phase at the start of the reproductive program, the interaction between estr o us cycle phase and PGF formulation, the interaction between estrus detection method and estr o cumulative precipitation within 30 days after the start of the reproductive program. Statistical models to evaluate the likelihood of pregnancy and pregnancy loss after ET also included embryo type (fresh in vivo produced embryo, frozen/thawed in vivo produced embryo, fresh in vitro fertilized e mbryo, and frozen/thawed in vitro fertilized embryo), embryo grade (excellent/good, fair, and poor), and days after estrus at embryo transfer (6 to 9 d).

PAGE 59

59 For the analysis of the hazard of pregnancy after the start of the reproductive program, models includ ed GDPR (linear and quadratic), GHCR (linear and quadratic), the interaction between GDPR and GHCR; PGF formulation, the interactions between GDPR and PGF formulation and between GHCR and PGF formulation, the estr us detection method (AED vs. VIS ) and the interactions between GDPR and estrus detection method, between GHCR and estrus detection method, and between PGF formulation and estrus detection method; estr o us cycle phase at the time of the start of the reproductive program and the interactions b etween GDPR and estr o us cycle phase at the time of the start of the reproductive program, between GHCR and estr o us cycle phase at the time of the start of the reproductive program, and between PGF formulation and estr o us cycle phase at the time of the st art of the reproductive program; type of service (AI vs. ET) and the interactions between GDPR and type of service, between GHCR and type of service, between PGF formulation and type of service, and between estrus detection method and type of service, a precipitation 30 days after the start of the reproductive program. When GDPR and GHCR were associated with the hazard of pregnancy, these variables were divided into quartile and the Wilcoxon test of equal ity (LIFETEST procedure) was used to characterize the association between GDPR and GHCR and the interval from the start of the reproductive program and first detected estrus. For each of the statistical models, collinearity was tested using the REG procedu re with collinear. In such cases, each variable was added to the model separately and the variable with the smallest P value was retained. In all models, a backward stepwise elimination procedure was adopted and variables with P > 0.10 were removed until all variables that remained in the

PAGE 60

60 model had P 0.10. Statistical significance was considered at P consider when 0.05 < P Results regarding the effects of PGF fo rmulation and phase of the estr o us cycle at the start of the reproductive program on estr o us characteristics and the effects of PGF formulation, phase of the estr o us cycle at the start of the reproductive program, and estr us detection method on reproductive responses are discussed in Chapter 2. Results Characteristics of the S tudy P opulation Mean (SEM) age and body weight at the start of the reproductive period were 377 6 d and 384 28 kg respectively Mean (SEM) GDPR values in the study population were 1. 65 1.29 ( range: 1.8 to 5.9; Figure 3 1 .) and mean GHCR values for the study population were 1. 34 1.11 ( range: 2.1 to 5.5; Figure 3 2 GDPR and GHCR was 0.455 (95% CI = 0.405 0.503; P < 0.01; Figure 3 3 ). A ssociation Among Genomic Daughter Pregnancy R ate and H eif er Conception Rate and E str o us C haracteristics Duration of the SPE tended ( P = 0.08) to increase according to GDPR, but there was ( P < 0.01) a negative association between GHCR and duration of SPE (Figure 3 4.) Percentage of also was negatively associated with the duration of SPE (Table 3 1.). Interval from onset of estrus to activity peak tended ( P = 0.06) to be negat ively associated with GDPR and was ( P = 0.03) positively associated with quadratic GDPR. Interval from onset of P < 0.01) and positively associated with cumulative pre cipitation ( P = 0.03) in the last 30 d prior to

PAGE 61

61 the start of the reproductive program (Table 3 1.). Rumination nadir on the day of SPE was negatively associated with GDPR ( P = 0.03) and positively associated with GHCR ( P = 0.05; Figure 3 5), whereas cumula tive precipitation in the last 30 days before the start of the reproductive program was ( P < 0.01) negatively associated with rumination nadir (Table 3 1.). P = 0.09) to be positively associated with GDPR and wa s ( P = 0.04) positively associated with cumulative precipitation in the last 30 days before the start of the reproductive program (Table 3 1.). There was a tendency ( P = 0.06) for GDPR to GHCR was ( P = 0.03) 1.; Figure 3 7.). Percentage of days with THI > 72 in the last 30 days before the start of the reproductive program was ( P = 0.01) negatively associated with the lik 1.). No association was observed among GDPR ( P = 0.24) and GHCR ( P = 0.28) and duration of PIE. Estr o us cycle phase was associated ( P < 0.001) with duration of PIE because ED heifers had short er PIE, followed by MID and PE heifers, respectively. Duration of PIE was P < 0.01) and positively associated with cumulative precipitation ( P < 0.01) in the first 30 days after the start of the reproductive program (Table 3 2.). Interval from onset of PIE to activity peak was positively associated with GDPR ( P < 0.01) and GHCR quadratic ( P = 0.05). Conversely, the interval from onset of PIE to activity peak was ( P = 0.02) negatively associated wi th the interaction between GDPR and GHCR. Cumulative precipitation in the 30 days after the start of the reproductive program was ( P = 0.03) positively associated with the interval from onset of PIE and activity peak. Estr o us cycle phase on the day of PGF 2 treatment was ( P < 0.01) associated with the interval from onset of PIE to peak activity because ED heifers had shorter interval from onset of PIE to activity peak

PAGE 62

62 compared with MID ( P < 0.01) and PE ( P < 0.01) heifers. Rumination nadir on the day of PIE was ( P < 0.01) negatively associated with GDPR (Table 3 2.). Cumulative precipitation in the 30 days after the start of the reproductive program was ( P < 0.01) negatively associated with rumination nadir on the da y of PIE (Table 3 2.). The estr o us phase a t PGF treatment was ( P < 0.01) associated with rumination nadir on the day of PIE because ED and MID heifers had greater rumination nadir at estrus than PE heifers (Table 3 2.). There was a tendency ( P = 0.06) for GDPR to be positively associated with th interaction between GDPR and GHCR tended ( P = 0.10) to be negatively associated with the 2.). Estrous cycle phase was ( P < 0.01) associated with likelihood of activit 2.), because ED heifers were less likely to have activity 2.). There was no association between GDPR ( P = 0.74) and GHCR ( P e 3 2.) E strous cycle phase was ( P of PIE because ED heifers (Table 3 2.). Genetic merit for DPR was ( P = 0.01) positively associated with the hazard of estrus after the start of the reproductive program. Heifers in the 4 th quartile for GDPR were detected in estrus in average 93.69 6.20 h after the start of the reproductive program followed by heifers in the 3 rd quartile (109.02 h 6.66 h) and heifers in the 2 nd (128.99 7.35 h) and 1 st (124.89 7.19 h) q uartiles, respectively (Figure 3 8 ). There was no ( P = 0.93) association between GHCR and hazard of estrus after the start of the reproduc tive progra m. Phase of the estr o us cycle at the start of the reproductive program was ( P < 0.01) associated with the hazard of estrus because heifers in PE (reference ) at PGF 2 treatment were detected in estrus at faster r ate, followed by heifers in

PAGE 63

63 mid diestrus (AHR = 0.760, 95% CI = 0.6 28, 0.920) and heifers in early dietrus (AHR = 0.143, 95% CI = 0.107) and met estrus (AHR = 0.139, 95% CI = 0.103, 0.187), respectively. A ssoci ation among Genomic Daughter Pregnancy Rate and Heifer Conception Rate and Pregnancy to First S ervice Genetic merit for DPR was ( P < 0.01) positively associated wit h the likelihood of pregnancy 35 3 d after the first AI. Other factors associated wi th the likelihood of pregnancy 35 3 d after the first AI were estr o us cycle phase at PGF treatment ( P = 0.01) and service technician ( P = 0.04). The interaction between GDPR and HCR was ( P = 0.03) negatively associated wit h the likelihood of pregnancy 35 3 d after the estrus resulting in the first ET. Other factors associated wit h the likel ihood of pregnancy 35 3 d after the estrus resulting in the first ET were the type of embryo ( P < 0.01), phase of the estr o us cycle at PGF treatment ( P = 0.05), and ET technician ( P < 0.01). There was a tendency ( P = 0.09) for the percentage of days wi associated wit h the likelihood of pregnancy 35 3 d after the estrus resulting in the first ET. The interaction between GDPR and GHCR tended ( P = 0.08) to be pos itively associated with the likelihood of pregnancy 75 3 d after the first AI (Table 3 3 ). P hase of the estr o us cycle at PGF treatment was ( P = 0.05) negatively associated with the likelihood of pregnancy 75 3 d after the first AI, because ED and MI D heifers were less likely to have pregnancy 75 3 d after the first AI than PE heifers (Table 3 3.). T echnician ( P = 0.02) w as associated with the likelihood of pregnancy 75 3 d after the first AI (Table 3 3 ). The likelihood of pregnancy 75 3 d afte r the estr o us resulting in the first ET was ( P = 0.01) negatively associated with the interactio n between GDPR and GHCR (Table 3 3 ). Type of embryo ( P < 0.01) and ET technician ( P = 0.02) were associated with the likelihood of pregnancy 75 3 d after the estrus

PAGE 64

64 resulting in the first ET. Additionally, the interaction between PGF formulation and phase of the estr o us cycle at PGF treatment ( P = 0.06) and method of estrus detection ( P = 0.06) tended to be associated with the likelihood of pregnancy 75 3 d after the estrus re sulting in the first ET (Table 3 3 ). There were no associations between GDPR ( P = 0.47) and GHCR ( P = 0.84) and the like lihood of pregnancy loss from 35 3 to 75 3 d after the first AI. Similarly, GDPR ( P = 0.80) and GHCR ( P = 0.81) were not associated with the like lihood of pregnancy loss from 35 3 to 75 3 d after the estrus resulting in the first ET. There was, however, a tendency for method of estrus detection ( P = 0.07) and type of embryo ( P = 0.06) to be associated w ith the like lihood of pregnancy loss from 35 3 to 75 3 d after the estrus resu lting in the first ET. A ssociation among Genomic Daughter Pregnancy Rate and Heifer Conception Rate and Hazard of P regnancy The interaction between GDPR and estrus detectio n method tended ( P = 0.08) to be and the interaction between GHCR and estrus detection method was ( P = 0.05) associated with the hazard of pregnancy. Among heifers detected in estrus by the AED system, GDPR was ( P = 0.05) associated with the interval from onset of the reproductive program to esta blishment of pregnancy (Figure 3 9 ), but GHCR was not ( P = 0.26) associated with the interval from onset of the reproductive program to esta blishment of pregnancy (Figure 3 10 ). Among heifers detected in estrus vi sually by herd personnel, GDPR ( P = 0.97; Figure 3 11) and GHCR ( P = 0.12; Figure 3 12 ) were not associated with the interval from onset of the reproductive program to establishment of pregnancy.

PAGE 65

65 Discussion In the current study, GDPR was positively assoc iated with more intense characteristics of rumination nadir on the day of SPE. Consequently, GDPR was positively associated with the likelihood of heifers havi SPE and positively associated with rumination nadir on the day of SPE, resulting in a negative asso in characteristics of estrus according to GDPR and GHCR demonstrated herein are important because these characteristics are generally associated with mounting activi ty, vaginal mucus consistency, and uterine tone (Pahl et al., 2015; Silper et al., 2015 ) and may improve estrus detection efficiency and accuracy. Estr o us behavior is the consequence of an orchestrated sequence of events, which lead to the acceptability of the male by the female, and are regulated by a network of genes that promote mating beh avior (Woelders et al., 2014) During the growth phase there is an increase in connectivity of hypothalamic neurons controlling behavior followed by progesterone binding to its receptors amplifying estrogen induced estr o us behaviors (amplification phase), express ion of sexual receptivity by the female (preparation phase females), expression of hypothalamic driven mating behaviors (permission phase), a nd finally, synchrony of mating and ovulation to elicit fertilization ( synchronization phase; Kommadath et al., 2011, 2013) During these phases of sexual behavior, several genes are differentially expressed in the hypothalamus, amygdala, hippocampus, and pituitary of lactating dairy cows during estrus and mid diestrus (Kommadath et al., 2011, 2013) Holmberg and Andersson Eklund ( 2006) genotyped 427 Swedish Red and Swedish Holstein bulls to identify quantitative trait loci (QTL) co ntributing to the genetic variation in fertility, among which was heat intensity

PAGE 66

66 this study 5 QTL associated with heat intensity score were determined on Bos taurus autosomes 4, 7, 9, 13, and 25, of which QTL on Bos taurus autosomes 7 and 9 were significant at the genome level (Holmberg and Andersson Eklund, 2006) There is, therefore, a clear aspect of genetic control of estr o us behavior that could help explain the associations among GDPR, GHCR and estr o us behavior. In a study conducted in Ireland, cows on the top quartile in genetic merit for milk yield and on the bottom 5% for calving interval had reduced duration of and activity during estrus compared with cows on the top quartile in genetic merit for milk yield and on the top 20% for calving interval (Cummins et al., 2012) Not surprisingly, cows on the top 20% for calving interval had shorter days open and increased Preg/Serv in the first two services postpartum compared with cows on the bottom 5% fo r calving interval. The positive associations between GDPR and estr o us characteristics observed in the current study may reflect how genetic selection for this trait impacts reproductive performance of US dairy herds. Daughter pregnancy rate is a measure o f the genetic merit associated with expected differences in 21 d PregRate when comparing animals or populations (VanRaden et al., 2004) The 21 d PregRate is highly dependent on 21 d service rate ( 21 d ServRate ; percentage of eligible cows that are serviced every 21 d after the end of the voluntary waiting period or start of the reproductive program) and pregnancy per service (percentage of cows that conceive after a service). Since GDPR was marginally associated with the probability of pregnancy after AI and ET, it seems logical to speculate that the advancements in reproductive performance generally associated with the onset of genetic selection for DPR since the early 2000s may be a consequence of improved estr us expression by animals with greater GDPR and greater hazard of detection of estrus. To our surprise, GHCR was negatively associated with estr o us characteristics evaluated in this study,

PAGE 67

67 but w as not associated with hazard of PIE or pregnancy. Furthermore, the interaction between 80. Genetic merit for HCR nceive after a service compared with the population. The negative association between GHCR and estr o us characteristics should be carefully studied in order to prevent a negative effect of selection for higher GHCR on estrus expression and detection in futu re generations of Holstein animals In a companion study, we evaluated the size of ovarian follicles and concentrations of estradiol, progesterone, insulin like growth factor 1 ( IGF 1 ), pregnancy specific protein B ( PSPB ), and interferon stimulated gene 15 ( ISG15 ) of heifers with high GDPR and high GHCR, high GDPR and low GHCR, low GDPR and high GHCR, and low GDPR and low GHCR. In that study, ovulatory follicle size and estradiol concentrations were greater for high GDPR animals and were not associated with GHCR. Since estradiol is secreted from follicles in the ovary (Jinks et al., 2013) and trigger s estr o us behavior (Reith and Hoy, 2017) these results shed light on why GDPR was associated with greater estru s duration and intensity in the current study. Together, these data provides evidence that GDPR drives physiological changes that alter estr o us behavior and could have a major impact on estrous detection efficiency and accuracy by dairy herds In the curre nt study, interaction of GDPR and GHCR tended to increase likelihood of pregnancy 75 3 d after the first AI, and only GDPR increased likelihood of pregnancy 75 3 d after second service. Surprisingly though, for ET services, the interaction of GDPR and GHCR decreased likelihood of pregnancy 75 3 d after first service. In the companion study, the interaction between GDPR and GHCR was associated with ISG15 expression 19 2 d after estrus. This interaction was because LH heifers had greater expression of ISG15 than LL heifers,

PAGE 68

68 while HH heifers and HL heifers were not different from LH and or LL heifers. Although IGF 1 concentration was not statistically different according to GDPR and GHCR, IGF 1 concentration was numerically greater for LH heifers at 19 2 d after estrus, which has been associated with expression of ISG15 and pregnancy establishment and maintenance ( R ibeiro et al., 2014) Kuhn et al. ( 2006) demonstrated a significant positive association between parent average DPR and Preg/Serv of heifers of multiple breeds. Since HCR has been implemented recently in the genetic selection of dairy cattle, its true association wit h pregnancy per service is less understood. Ortega et al. ( 2016) evaluated 69 single nucleotide polymorphisms (SNPs) related to fertility trades in Holstein cattle and showed that a significant number of genes ass ociated with DPR were associated with HCR. These results suggest that genes driving fertility outcomes in cows associated with DPR may be the same driving fertility in heifers associated with HCR. The remaining different genes that compose GDPR or GHCR but do not overlap h owever, may be genes res ponsible for different functions that lead to improved reproductive performance but not necessarily by the same route. In the current study GDPR was associated with improved estrus expression, and faster onset of estrus after a PGF treatment. These res ults indicate that genomic selection for DPR has the potential to select animals with improved estrus expression, duration, and intensity, which in turn could improve reproductive performance and profitability of dairy operations. Furthermore, due to increasing concern of costumers over use of hormones for milk production, selection for GDPR can be an alternative for farmers interested in reduc ing hormonal use for estr o us cycle manipulation. Conversely selection of dairy animals based on GHCR should be carefully evaluate because in the current study it was associated with reduced estrus duration and intensity. Reduction in estrus duration and intensity can be detrimental for reproductive performance since

PAGE 69

69 it can reduce estrus detection Lastly, more studies are necessary to unravel how GDPR and GHCR drive pregnancy establishment and maintenance in dairy heifers.

PAGE 70

70 Table 3 1. Final logistic regression model of factors associated with characteristics of spontaneous estrus Variables Estimates (SE) P v alue Duration GDPR (linear) 0.08 <0.01 <0.01 Interval onset to peak of estrus GDPR 0.0028 0.0002 0.06 GDPR (quadratic) 0.0002 0.0001 0.03 0.1098 0.0176 <0.01 Precipitation 0.0005 0.0001 0.03 Rumination Nadir GDPR (linear) 0.515 0.007 0.03 GHCR (linear) 0.515 0.008 0.05 Precipitation 1.018 0.001 0.01 GDPR (linear) 0.058 0.035 0.02 Precipitation 0.073 0.03 0 0.04 GDPR (linear) 0.075 0.040 0.06 GHCR (linear) 0.098 0.047 0.03 Pct THI 72 0.644 0.260 0.01 Cum prec

PAGE 71

71 Table 3 2 Final logistic regression model of factors associated with characteristics of PGF induced estrus Variables Estimates (SE) P value Duration <0.01 0.02 <0.01 0.05 Precipitation < 0.01 Interval onset to peak of estrus GDPR 0.0045 0.0003 < 0.01 0.0009 0.0001 0.36 GDPR (quadratic) 0.0003 0.0001 0.05 0.0006 0.0001 0.02 0.1975 0.0394 < 0.01 0.0010 0.0001 0.66 0.07 Precipitation 0.0003 0.0001 0.03 Rumination Nadir GDPR (linear) 0. 52 0 0.00 5 < 0.01 1.336 0.063 < 0.01 1.101 0.033 < 0.01 Precipitation 1.014 0.004 <0.01 GDPR (linear) 0.096 0.051 0.06 GHCR (linear) 0.064 0.064 0.31 GDPR x GHCR 0.021 0.013 0.10 0.747 0.350 0.03 2.210 0.447 < 0.01 0.293 0.363 0.42 1.625 0.406 < 0.01 0.369 0.338 0.28 £ ECD: Estr o us cycle day ( ME: Meteestrus; ED: Early diestrus; MID: Mid diestrus ; PE: Proestrus) & after the start of the reproductive program *Precipitation: Cumulative precipitation 30 days after the start of the reproductive program

PAGE 72

72 Table 3 3 Final logistic regression model of factors associated with the likelihood of pregnancy after the first service ( 75 3 d after service) Variables Estimates (SE) P value First service Artificial Insemination 0.017 0.065 0.77 0.069 0.071 0.33 0.026 0.015 0.08 0.759 0.319 0.02 0.421 0.248 0.09 0.02 Embryo Transfer GDPR 0.183 0.075 0.01 0.201 0.097 0.04 0.052 0.022 0.02 0.540 0.653 0.41 0.589 0.360 0.10 <0.01 1.932 0.677 <0.01 0.212 0.941 0.82 1.201 0.560 0.03 Embryo type: 1 fresh in vivo produced embryo, 2 frozen/thawed in vivo produced embryo, 3 fresh in vitro fertilized embryo, and 4 frozen/thawed in vitro fertilized embryo.

PAGE 73

73 Table 3 4 Final logistic regression model of factors associated with hazard of pregnancy Variables Estimates (SE) P value 0.072 0.057 0.21 0.061 0.088 0.49 0.042 0.022 0.06 0.176 0.149 0.24 0.127 0.073 0.08 0.162 0.083 0.05 0.078 0.149 0.60 0.058 0.120 0.63 0.392 0.162 0.02 <0.01 0.06

PAGE 74

74 Figure 3 3 Distribution of genetic merit for daughter pregnancy rate ( GDPR ) values in the study population. Mean SD: GD PR = 1.65 1.29 (range, 1.8 5.0). Figure 3 2. Distribution of genetic merit for heifer conception rate ( G HCR ) values in the study population. Mean SD: G HCR = 1.34 1.1 (range, 2.1 5.5) 0 10 20 30 40 50 60 -2.1 -1.1 -0.1 0.9 1.9 2.9 3.9 4.9 5.9 Frequency GDPR 0 10 20 30 40 50 60 -2.1 -1.1 -0.1 0.9 1.9 2.9 3.9 4.9 5.9 Frequency GHCR

PAGE 75

75 Figure 3 3. Correlation of genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate ( GHCR ) (Confidence interval = 0.405 0.503; P < 0.01). Figure 3 4. Duration of estrus according to genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate (GHCR) GHCR = 0.1 (low ) GHCR = 1.6 ( intermediary ), GHCR = 2.9 ( high ). GDPR P = 0.08 GHCR P < 0.01. -3 -2 -1 0 1 2 3 4 5 6 -3 -2 -1 0 1 2 3 4 5 6 GHCR GDPR 15 16 17 18 19 -1.5 -0.75 0 0.75 1.5 2.25 3 3.75 4.5 5.25 Estrus duration, h GDPR GHCR = 0.1 GHCR = 1.6 GHCR = 2.9

PAGE 76

76 Figure 3 5. Rumination nadir according to genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate (GHCR). GHCR = 0.1 (low ) GHCR = 1.6 ( intermediary ), GHCR = 2.9 ( high ). GDPR P = 0.03 GHCR P = 0.05. Figure 3 6. Activity peak according to genetic merit for daughter pregnancy rate (GDPR). GDPR P = 0.02 -20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5 5.5 Rumination nadir, min/day GDPR GHCR = 0.1 GHCR = 1.6 GHCR = 2.9 50 55 60 65 70 75 80 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5 5.5 GDPR

PAGE 77

77 Figure 3 7. Heat index according to genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate (GHCR). GHCR = 0.1 (low ) GHCR = 1.6 ( intermediary ), GHCR = 2.9 ( high ). GDPR P = 0.03 GHCR P = 0.05. according to GDPR ( P = 0.06), when GHCR ( P = 0.03) is low (0.1), intermediary (1.6 ), or high (2.9). Figure 3 8. Interval from start of the reproductive program to fi rst estrus detected by the AED according to GDPR quartile: Q1 = quartile 1 (GDPR = 1.8 to 0.8); Q2 = quartile 2 (GDPR = 0.9 to 1.7); Q3 = quartile 3 (GDPR = 1.8 to 2.5 ); Q4 = quartile 4 (GDPR = 2.6 to 5.9). Mean (SEM) and median interval from the start of the reproductive program to first detected estrus: Q1 = 124.89 7.19 and 58.5 h; Q2 = 128.99 7.35 and 61 h; Q3 = 109.02 6.66 and 60.2 h; and, Q4 = 93.69 6.20 a nd 50.4 h GDPR P < 0.01 50 55 60 65 70 75 80 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5 5.5 GDPR GHCR = 0.1 GHCR = 1.6 GHCR = 2.9 0 20 40 60 80 100 0 40 80 120 160 200 240 280 320 Heifers not detected in estrus, % Hour from start of the reproductive program Q1 Q2 Q3 Q4

PAGE 78

78 Figure 3 9 Interval from start of the reproductive program to pregnancy for heifers detected in estrus by an automated estrus detection device (AED) according to GDPR quartile: Q1 = quartile 1 (GDPR = 1.8 to 0.8); Q2 = quartile 2 (GDPR = 0.9 to 1.7); Q3 = quartile 3 (GDPR = 1.8 to 2.5); Q4 = quartile 4 (GDPR = 2.6 to 5.9). Mean (SEM) and median interval from the start of the reproductive program to p regnancy : Q1 = 42.5 2.3 and 41 d; Q2 = 38.4 2.7 and 34 d ; Q3 = 35.5 3.1 and 26 d ; and, Q4 = 37.2 2.7 and 24 d GDPR P = 0.05 Figure 3 10. Interval from start of the reproductive program to pregnancy for heifers detected in estrus by an automated estrus detection device (AED) according to GHC R quartile: Q1 = quartile 1 (GHCR = 2.1 to 0.6 ); Q2 = quartile 2 ( GHCR = 0.7 to 1.4 ); Q3 = quartile 3 ( GHCR = 1.5 to 2.1 ); Q4 = quartile 4 ( GHCR = 2.2 to 5.5 ). Mean (SEM) and median interval from the start of the reproductive program to pregnancy : Q1 = 38.0 2.3 and 37 d; Q2 = 42.0 2.8 and 40 d ; Q3 = 38.7 2.8 and 29 d ; and, Q4 = 35.5 2.7 and 24 d. GHCR P = 0.26) 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 Heifers not pregnant, % Day from start of the reproductive program Q1 Q2 Q3 Q4 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 Heifers not pregnant, % Day from start of the reproductive program Q1 Q2 Q3 Q4

PAGE 79

79 Figure 3 11 Interval from start of the reproductive period to pregnancy for heifers detected in estrus by visual observation (VIS) according to GDPR quartile: Q1 = quartile 1 (GDPR = 1.8 to 0.8); Q2 = quartile 2 (GDPR = 0.9 to 1.7); Q3 = quartile 3 (GDPR = 1.8 to 2.5); Q4 = quartile 4 (GDPR = 2.6 to 5.9). Mean (SEM) and median interval from the start of the reproductive program to pregnancy : Q1 = 43.9 3.1 and 39 d; Q2 = 42.8 2.8 and 37 d ; Q3 = 45.2 2.9 and 50 d ; and, Q4 = 43.6 3.3 and 44 d. GDPR P = 0.97 Figure 3 12. Interval from start of the reproductive period to pregnancy for heifers detected in estrus by visual observation (VIS) according to GHC R quartile: Q1 = quartile 1 (GHCR = 2.1 to 0.6 ); Q2 = quartile 2 ( GHCR = 0.7 to 1.4 ); Q3 = quartile 3 ( GHCR = 1.5 to 2.1 ); Q4 = quartile 4 ( GHCR = 2.2 to 5.5 ). Mean (SEM) and median interval from the start of the reproductive program to pregnancy : Q1 = 47.6 2 .9 and 60 d; Q2 = 44.4 2.8 and 43 d ; Q3 = 43.6 3.1 and 44 d ; and, Q4 = 39.0 3.2 and 35 d. GHCR P = 0. 12 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 Heifers not pregnant, % Day from start of the reproductive program Q1 Q2 Q3 Q4 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 Heifers not pregnant, % Day from start of the reproductive program Q1 Q2 Q3 Q4

PAGE 80

80 CHAPTER 4 PHYSIOLOGICAL RESPONSES OF HOLSTEIN HEIFERS WITH HIGH AND LOW GENOMIC MERIT FOR FERTILITY TRAITS Reproductive performance of Holstein cattle has declined over the past decades, with lowest breeding values for daughter pregnancy rate ( DPR ) recorde 2005). One of the main factors negatively impacting reproductive performance is believed to be the intense genetic selection for milk yield with disregard for reproduction traits (Lucy, 2001; VanRaden et al., 2004) Between 1963 and 2003 an increment of 3,259 kg in breeding values for milk yield (AIPL, 2005) was observed; concurrently, breeding values for DPR decreased from hal t the decline in reproductive performance associated with selection for productive traits alone DPR was added to the genetic merit in 2003, allowing selection of Holstein animals with improved reproductive performance. Since the addition of DPR on genetic selection, breeding values for Because of its low heritability (Pryce et al., 2004) genetic progress for fertility traits such as DPR is low (Garca R uiz et al., 2016) With recent advances in genomic tools for prediction of breeding values and inclusion of genomic predicted transmitted ability ( GPTA ) values for DPR and other fertility traits, such as heifer conception rate ( HCR ), genetic progress for low heritable traits, such these fertility traits significantly increased (Garca R uiz et al., 2016) compared with the population and genomic daughter pregnancy rate ( GDPR ) is a genomic predicted breeding value for DPR. Hei fer conception rate is a measure of likelihood of pregnancy following compared with the population and genomic heifer conception rate ( GHCR ) is a genomic predicted breeding value for HCR.

PAGE 81

81 Genomic fertility traits are associated with actual phenotypic values observed in the Holstein population (Mikshowsky et al., 2016; Ortega et al., 2016) but still little is known about how these genetic markers affect the phenotype. In recent experiments researchers demonstrated that several genes represented by single nucleotide polymorphisms ( SNP ) known to be involved with endocrine system, cell signaling, immune function and inhibition of apoptosis, were also associated with fertility traits such GDPR and GHCR in Holstein cows (Cochran et al., 2013) Furthermore, many of th e genes Cochran et al. (2013) demonstrated to be associated with fertility traits were previously shown to be associated with s teroidogenesis in Holstein cows (Ortega et al., 2016) Although Cochran et al. (201 3) and Ortega et al. (2016) provided valuable information about the possible function s of genes compos ing genomic predicted fertility traits (e.g. GDPR and GHCR ) information about how genomic breeding values for fertility traits are associated with physio logical responses in Holstein animals is not abundant. The hypothesis of the current study was that Holstein heifers differing in GDPR and GHCR have significant differences regarding ovulatory follicle size, estradiol concentration at estrus, and progester one, insulin like growth factor 1 ( IGF 1 ), and pregnancy specific protein B ( PSPB ) concentrations after estrus, and expression of interferon stimulated gene 15 ( ISG15 ) 19 d after estrus. Therefore, the objectives of the current study were to elucidate diff erences in ovulatory follicle size, estradiol concentration at estrus, and progesterone IGF 1 and PSPB concentrations after estrus, and expression of ISG15 19 d after estrus of heifers in the extreme of GDPR and GHCR within a population of Holstein heifers Material and Methods All procedures involving animals were approved by the animal care and use committee of the University of Florida (protocol #201609559).

PAGE 82

82 A nimals, Housing, and M anagement The study was conducted from September to December 2016 in a commercial dairy herd with approximately 4,200 replacement heifers located in north central Florida. Ninety nine Holstein heifers between 10 and 11 months of age were enrolled in the study. All h eifers were genotyped within 2 months of birth using a 50k single nucleotide peptide platform commercially available (Clarifide, Zoetis, Parsippany, NJ). For the purpose of this study data referent to genomic breeding values for DPR and HCR recorded withi n 2 months of birth were used. H eifers selected for this experiment were in the top and bottom 50 percentile for GDPR or GHCR values in this population of 1,019 heifers Heifers were classified as: high GDPR (range = 1.6 to 5.3), low GDPR (range = 1.8 to 1.0), high GHCR (range = 1.5 to 5.5), and low GHCR (range = 2.1 to 1.2). The resulting combinations of GDPR and GHCR class were, respectively: HH ( n = 28), HL ( n = 20), LH ( n = 21), and LL ( n = 30). Starting at 12 months of age, heifers were weighed we ekly and heifers with 340 kg of live body weigh were moved to a breeding pen and were treated with prostaglandin ( PG ) F (cloprostenol sodium, Estrumate, Merck Animal Health, Summit NJ) for synchronization of estrus. Heifers were housed in dry lots, with natural shade and no artificial cooling. The breeding pens had self locking head stanchions on the feeding area. Heifers were fed twice daily (7:00 AM and 4:30 PM) a TMR formulated to meet or exceed the nutritional requirements of Holsteins heifers weighing 340 kg of live body weight and gaining 800 to 1,000 g of live body weight per day (NRC, 2001). A utomated Estrus M onitoring S ystem

PAGE 83

83 At enrollment, an automated estrus detection mo nitoring device ( AED ; Heat Rumination Long Distance, SCR Inc., Netanya, Israel) was fitted on the left, cranial area of the neck of all heifers. The AED determined activity through an accelerometer and rumination based on sounds of regurgitation and mastic ation through a microphone. Activity and rumination data were recorded for every 2 h periods. Estrus was determined according to changes in patterns of activity and rumination within a 2 h period compared with the average activity and rumination of the sam e period in the previous 5 and 7 d, respectively (DataFlow2 SCR Inc, Netanya, Israel). R epro ductive M anagement From enrollment to the start of the reproductive program, all estrus events were recorded. H eifer eligible to start the reproductive program ( 12 months of age and 340 kg of live body weight) were classified according to estr o us cycle phase into early metestrus ( estr o us cycle day 0 to 3), early diestrus ( estr o us cycle day 4 to 6), mid diestrus ( estr o us cycle day 7 to 17 ), proestrus ( estr o us c ycle day 1 8 ), and no estrus observed. Heifers in early diestrus, mid diestrus proestrus and heifers that had no estrus observed were treated with PGF immediately and heifers in metestrus were treated with PGF 96 h later and heifers According to the genetic selection program of the dairy, heifers were selected to be artificially inseminated ( AI ) or to receive an embryo transfer ( ET ) Heifers detected in estrus were artificially inseminated on the same morning or received an embryo 6 to 9 days after e strus detection. P regnancy Diagnoses and Reproductive D ata All heifers were examined for pregnancy by palpation per rectum of uterine contents at 35 3 d after the detected estrus that resulted in AI or ET. Pregnant heifers were re examined by palpation per rectum of the uterine contents at 75 3 days of gestation.

PAGE 84

84 Pregnancy per service was calculated by dividing the number of heifers pregnant at 35 and 75 3 d after estrus by the number of heifers serviced. Pregnancy loss was calculated by dividing the number of heifers pregnant at 75 3 d after estrus by t he number of heifers pregnant 35 3 d after estrus Data regarding sire of insemination, sire and dam of embryo transfer, service technician, and reproductive outcomes were collected from farm recor ds using dairy management software PCDART (Dairy records management system, Chapel Hill, North Carolina). B lood Sampling and Ultrasonography of the O varies Blood was sampled on the day of PGF treatment, on the first morning after detected estrus ( 2 to 26 h after onset of estrus), and at 7, 14, 19 2, 28, and 35 d after estrus Samples were not collected when heifers return ed to estrus and received a second service. Blood was sampled by puncture of the coccygeal vein or artery into evacuated tubes containing K2 EDTA (Vacutainer, Becton Dickinson, Franklin Lakes, NJ). Immediately upon collection, tubes were placed in ice and kept refrigerated until transported to the laboratory for processing within 2 to 3 h of collection Blood tubes were centrifug ed at 1,500 g for 15 min. Aliquots of plasma were frozen at 80 C until assayed. Ovaries of heifers were evaluated by transrectal ultrasonography ( MyLab TM Esaote North America, Inc., Fishers, IN ) in the first morning after detected estrus ( 1 to 24 h af ter onset of estrus) and daily until ovulation was observed (disappearance of a follicle larger than 10 mm) or 96 h after onset of estrus A nalysis of Plasma S amples Progesterone concentrations w as determined by radioimmunoassay ( RIA ) using a commercial kit (Coat a Count, MP Biomedical LLC, Solon OH). Plasma harvested from heifers

PAGE 85

85 on days 4 (~1 ng/mL) and 10 (~ 4 ng/mL) of the estr o us cycle were incorporated into each assay and used to calculate the CV. Intra and inter assay CVs were 5.8 and 10.5 % respe ctively. Serum co ncentration of estradiol quantified by RIA as described by Jinks et al. ( 2013) I ntra assay coefficient of variance for estradiol assays was 2.73%. Concentrations of PSPB were analyzed using a commercially available quantitative ELISA assay (BioPRYN; BioTracking LLC, Moscow, ID) a ccording to the method described by Humblot et al. ( 1988) Intra and inter assay CVs were 3.5 and 8.2 % respectively. Concentrations of IGF 1 were determined by a commercial ELISA kit (Quantikine ELISA Human IGF 1 Immunoassay, R&D Systems) designed for human IGF 1 but with 100% cross reactivity with bovine IGF 1, as described pre viously by Ribeiro et al. ( 2 014 ) The intra assay CV for IGF 1 was 8.0%. I solation o f Peripheral Blood L eukocytes, m RNA E xtracti on, and Quantitative Real T ime q PCR Blood sampled 19 2 d after estrus was used for isolation of peripheral blood leukocytes ( PBL ) according to Gifford et al. ( 2008) After centrifugation and harvest of plasma, buffy coat fractions were collected by pi petting and transferred to 15 mL conical tubes. A r ed cell lysis buffer was prepared (150 mM NH4Cl, 10 mM NaHCO3, and 1 mM EDTA; pH 7) and added to the buffy coat for a to tal volume of 15 mL Tubes were inverted several times and incubated at room temperature for 5 min. Samples were then centrifuged at 300 g for 10 min at 4C and the supernatant was discarded. Th e PBL pellet was mixed with 5 mL of red cell lysis buffer, i ncubated at room temperature for 5 min, and centrifuged at 300 g for 10 min at 4C, and the supernatant was discarded. The PBL pellet was washed with ice cold PBS and centrifuged at 300 g for 10 min at 4 C and the supernatant was discarded. The PBL pe llet was re suspended with 0.8 mL of Trizol (Molecular Research Center, Inc., Cincin nati, OH), transferred to 1.5 mL

PAGE 86

86 microtubes, and stored at 80 C. The time interval from blood collection and PBL sample stora ge at 80 C was no longer than 6 h. Extraction of m RNA was conducted according to the manufacturer's recommendations for the RNA extraction kit (PureLink RNA Mini Kit; Invitrogen, Carlsbad, CA). The cell ular mRNA was treated with DNase (RQ1 RNase Free DNase; Promega, Madison, WI) and was used to synthesize complementary DNA using the DyNAmo cDNA Synthesis Kit (Thermo Scientific, Waltham, MA). Complementary DNA was then used for quantitative RT PCR (ABI 75 00 Sequence Detector; Applied Biosystems Inc., Foster City, CA). Three genes were investigated : ISG15 (target gene) beta actin ( ACTB ; reference gene ) and ribosomal protein L 19 ( RPL19 ; reference gene ). P rimer reference and s equence are represented in Tab le 4 1. Each specific forward and reverse primers, SYBR Green (Applied Biosystems Inc. Foster City, CA ), and nuclease free in duplicate and comprised 40 cycles of a three step amplification protocol (30 sec at 95 C followed by 45 sec at the optimized annealing temperature [57 C 60 C] and 1 min at 72 C). Primer efficiency ranged from 81% to 85%. Melting curve analysis was also performed to ensure amplification of a single product. S tatistical A nalysis Data was analyzed using SAS version 9.3 (SAS Institute Inc., Raleigh, NC). Continues variables were analyzed by ANOVA using the MIXED procedure. Data were evaluated for normality and homogeneity of residuals after fitting the model. Data violating the assu mptions of normality were transformed before analysis. Progesterone concentration values at estrus were transformed to the square root of the real value and ISG15 relative abundance values were

PAGE 87

87 transformed to log natural of the real value to meet the assum ption of normality of residuals. Data was b ack transformed for interpretation of the results. All statistical models included GDPR class GHCR class and interaction of GDPR and GHCR class. Models for estradiol and progesterone at estr us also included pregn ancy at 35 3 d interval from onset of estrus to blood sample collection (2 to 26 hours), linear and quadratic. For ISG15 models also included bre eding code (AI vs. ET) and d ay after estrus when the sample was collected (19 2 d ). For the analysis of p rogesterone, PSPB, and IGF 1 concentrations after estrus, models also included sample, and the interaction s between GDPR and sample, GHCR and sample, and GDPR, GHCR and sample and breeding code (AI vs. ET). For each of the statistical models collinearit y was tested using the REG procedure of SAS considered collinear. In such cases, each variable was added to the model separately and the variable with the smallest P value was retained. A backward stepwise elimination of variables with P > 0.10 until variables that remained in the model had P 0.10 was performed. Statistical significance was considered at P 0.05 and a tendency was consider when 0.05 < P Results Descriptive data for GDPR and GHCR in the study population divided into classes are presented in Table 4 2 and distribution of GDPR and GHCR according to classes are presented in Figure 4 1. Descriptive d ata regarding number of heifers detected in estrus, number of heifers that ovulate d according to ultrasound and according to progesterone concentrations pregnant heifers at 35 3 d after service, and pregnancy loss from 35 to 75 are described in Table 4 3. A nalysis of Physiological Differences I ncluding All H eifers

PAGE 88

88 Ovulatory follicle size was greater ( P < 0.01) for High GDPR than Low GDPR heifers but GHCR class was not ( P = 0.12) associated with ovulatory follicle size The interaction between GDPR and GHCR class es was not ( P = 0.82 ) associated with ovulatory follicle size ( Figure 4 2.). Estradiol concentrations after heifers were detected in estrus w as greater ( P = 0.02) for High GDPR than Low GDPR heifers but GHCR class was not ( P = 0.21) associated with estradiol concentrations after heifers were detected in estrus The interaction between GDPR and GHCR class w as not ( P = 0.60) associated with estradiol concentrations after heifers were detected in estrus ( Figure 4 3.). Class of GDPR ( P = 0.88 ) and GHCR ( P = 0.78) and the interaction between GDPR and GHCR classes ( P = 0.56) were not associated with progesterone concentration within 24 h after heifers were detected in estrus ( Figure 4 4.). Class of GDPR ( P = 0.38) and GHCR ( P = 0.38) and t he interaction between GDPR and GHCR classes ( P = 0.17) w ere not associated with progesterone concentrations at 7 and 14 d after estrus ( Figure 4 5.). Classes of GDPR ( P = 0.30) and GHCR ( P = 0.71) and t he interaction between GDPR and GHCR classes were not ( P = 0.56) associated with IGF 1 concentration s after heifers were detected in estrus. A nalysis of Physiological Differences I ncludin g Only H eifer s P regnant 35 3 d After E strus Class of GDPR tended ( P = 0.08 ) to be associated with greater progesterone concentrations at estrus (Figure 4 5.). Class o f GHCR ( P = 0.43) and the interaction between GDPR and GHCR classes ( P = 0.46) were not associated with progesterone concentrations at estrus (Figure 4 5.). Class of GDPR ( P = 0.19) and GHCR ( P = 0.98 ) and t he i nteraction between GDPR and GHCR class es ( P = 0.70) were not associated with progesterone concentrations at 7, 14, 19 2, 28, and 35 d after estrus (Figure 4 5. ).

PAGE 89

89 The i nteraction between GDPR and GHCR classes tended ( P = 0.08) to be associated with relati ve expression of ISG15 19 2 d after estr us because LH heifers had greater expression of ISG15 than LL heifers, whereas the expression of ISG15 among HH heifers and HL heifers was intermediary (Figure 4 6.). Class of GDPR ( P = 0.87) and GHCR ( P = 0.58) and the i nteraction between GDPR and GHCR c lasses ( P = 0.41) w ere not associated with PSPB concentrations 19 2 d after estrus (Figure 4 7.). Concentrations of PSPB at 28 and 35 d after estrus were greater ( P = 0.03) for High GDPR than Low GDPR heifers but GHCR class ( P = 0.86) and interaction between GDPR and GHCR classes ( P = 0.63) w ere not associated with PSPB concentrations 28 and 35 d after estrus (Figure 4 7.). Class of GDPR ( P = 0.50) and GHCR ( P = 0.14 ) and t he i nteraction between GDPR and GHCR classes ( P = 0.48) were not associ ated with IGF 1 concentrations within 24 h after estrus was detected and at 7, 14, 19 2, 28, and 35 d after estrus (Figure 4 8.). Discussion In the current st ud y High GDPR heifers had greater ovulatory follicle size and estradiol concentrations which may be explained by the fact that several SNPs associated with DPR are involved in steroidogenesis or are regulated by steroids (Ortega et al., 2016) S ince a n overlap of genes that compose GDPR and GHCR exists (Cochran et al 2013 ) we expected GHCR also to be positively associated with greater ovulatory follicle size and estradiol concentrations. Proliferation of the pre ovulatory dominant follicle drives estradiol concentrations (Vasconcelos et al., 2001; Forde et al., 2011) E stradiol triggers estrus expression and is extremely import ant for accurate de tection, and breeding of animals because it increases estrus intensity and duration, and facilitates estrus detection (Reith and Hoy, 2017) In a companion study, we evaluated the association among GDPR and GHCR and estrus duration and intensity (rumination nadir,

PAGE 90

90 activity peak, and heat index) in dairy heifers In the companion study, GDPR was positively associated with estrus duration and intensity whereas GHCR was negatively associated with duration and intensity of estrus Since i n the current study, h eifers with High GDPR had greater ovulatory follicle siz e, and estradiol concentrations, we speculate that one of the mechanism s by which heifers with high GDPR had longer and more intense estruses was due to greater estradiol concentrations G enomic heifer conception rate was not associated with ovulatory follicle size or estradiol concentrations in the current study; however, numerically smaller ovulatory follicle s and estradiol concentrations were o bserved in LH heifers, which can potentially explain the negative association between GHCR and estrus duration and intensity observed in the companion study Class of GDPR and GHCR was not associated with progesterone concentrations at 7, and 14 d after estrus. Similarly, GDPR and GHCR classes were not associated with progesterone concentrations at 7, 14, 19 2, 28, and 35 d after estrus when only pregnant heifers 35 3 d after estrus were included in the analysis Progesterone is produced by luteinized granulosa and theca cells from the ovulated follicle (Forde et al., 2011) and has a crucial role on pregnancy maintenance (Stevenson and Lamb, 20 16) Ortega et al. ( 2016 ) demonstrated that GDPR was associated with Preg/Serv and days open in a selected Holstein population Genomic heifer conception rate is a measure of the likelihood of pregnancy after a service (Sun et al., 2014) Because GHCR is a newer trait however information about its association wit h actual Preg/Serv in heifers is limited Cummins et al. (2012) performed a study to evaluate ovarian follicular dynamics, reproductive hormones and estr o us behavior in lactating cows with high and low genetic merit for fertility traits. One of the main fi ndings was that progesterone concentrations were greater in cows classified as high for fertility traits than in cows classified as low for

PAGE 91

91 fertility traits. T herefore Cummins et al. (2012) suggested that greater progesterone concentrations partially explained improved reproductive performance in cows classified as high for fertility traits The lack of association among GDPR, GHCR, and progesterone concentrations in the current study however do not support the hypothesis by Cummins et al ( 2012 ). W e recognize that a small number of pregnant heifer s was evaluated in the current study and additional studies are needed to confirm our findings Nonetheless, when progesterone concentration at 7 and 14 d after estrus from all heifers was analyzed GDPR and GHCR were not associated with progesterone concentrations, leading to the speculation that GDPR and GHCR indeed may not be associated with progesterone concentrations after estrus. The interaction between GDPR and GHCR classes tended to be associated with ISG15 expression 19 2 d after estrus because LH heifers had greater expression of ISG15 than LL heifers, whereas the expression of ISG15 among HH heifers and HL heifers was intermediary. Conceptus development and maintenance are highly dependent on a series of conceptus signaling that must be recognized by the dam (Ribeiro et al., 2014) In rumi nants, IFN by the trophoblast and its responsible for the maternal recognition of pregnancy (Green et al., 2010) Interferon that block the luteolytic cascade in endometrial cells and prevent regression of the corp us luteum (Ribeiro et al., 2014) Concentrations of IFN in utero are dependent mainly of the size of the conceptus (Shirasuna et al., 2013) Interferon (e.g. leukocytes ) increasing expression of interferon stimulated genes such as ISG15 (Ribeiro et al., 2014) Mat suyama et al. ( 2012) demonstrated that interferon stimulated genes responses in utero and in peripheral blood cells were similar, suggesting that ISG15 expression in peripheral blood leukocytes may be used as an indirect measure of early embryonic devel opment. In a companion study the interaction between

PAGE 92

92 GDPR and GHCR classes were associated with Preg/Serv, and GDPR and GHCR were associated with the hazard of pregnancy. Therefore, these data combined suggest that GDPR and GHCR are associated with embryo development, maternal recognition of pregnancy, and maintenance of pregnancy Pregnancy specific protein B is secreted by binucleate trophoblastic cells and was previously described by Humblot et a l. ( 1988 ) and Green et al. ( 2005) to be associated with conceptus development and pregnancy maintenance in heifers and co ws. Ribeiro et al. (2014) d emonstrated that cows with greater expression of ISG15 19 d after insemination also had greater PSPB concentrations 21 d after insemination. Since the interaction between GDPR and GHCR classes was associated with ISG15 expression 19 2 d after estrus, we expected the interaction between GDPR and GHCR also to be associated with PSPB concentrations at 19 2, 28, and 35 d after estrus. Class of GDPR and GHCR were not associated with PSPB concentrations at 19 2 d after estrus Class of GDPR however, was as sociated with greater PSPB concentrations at 28 and 35 d after estrus but GHCR was not associated with PSPB concentrations. Greater pre ovulatory follicle size and estradiol concentrations are associated with improved endometrial environment, which favors pregnancy establishment (Madsen et al., 2015) The greater ovulatory follicle size and greater estradiol concentrations in high GDPR heifers could have led to improved uterine environment and hastened conceptus development, resulting in greater PSPB concentrations among high GDPR heifers at 28 and 35 d after estrus Reasons for GHCR class to be associated with ISG15 at 19 2 after estrus but not with PSPB concentrations on 28 and 35 d after estrus however, are unknown and require further investigation.

PAGE 93

93 There were no associations among GDPR and GHCR classes and IGF 1 concentrations at and after estrus. One of the possible mechanisms that would explain the upregulation of ISG15 expression and increased concentration of PSPB is IGF 1 induced conceptus growth which could potentially lead to increased pregnancy maintenance and Preg/Serv (Ribeiro et al., 2014). Therefore, we hypothesized that GDPR and GHCR drive n conceptus development and consequently upregulations of ISG15 and greater concentration of PSPB could result from differences in IGF 1 concentration The lack of differences in IGF 1 concentration according to GDPR and GHCR classes could be the consequence of the small sample size and insufficient power of the current study because the IGF 1 concentration of LH heifers on day 19 2 after estrus was approximately 22% greater tha n HH and HL heifers and approximately 43% greater than LL heifers G enetic merit for DPR and GHCR are predictors of reproductive performance that share some genetic markers (Ortega et al., 2016) and the current study reinforces the hypothesis that both driv e early conceptus development. P recise mechanisms by which GDPR and GHCR a ffect fertilization, embryo and conceptus development, and pregnancy maintenance however, remain unknown The greater ovulatory follicle size and estradiol concentrations obse rved among high GDPR heifers in the current study may explain why high GDPR heifers have more evident sings of estrus and suggests that continued selection for GDPR could potentially improve estrous detection efficiency and accuracy on farm The associatio n of GDPR and GHCR with ISG15 expression by PBL and the association of GDPR with concentrations of PSPB after service suggest that both genetic markers are associated with embryo/conceptus development but additional studies are necessary to further unders tand mechanisms by which GDPR and GHCR improve conceptus development.

PAGE 94

94 Table 4 1. Primer reference and sequences for genes investigated by quantitative real time PCR. Table 4 2 Descriptive GDPR and GHCR data for the study population. Target gene Gene name NCBI sequence Primer Primer sequence ISG15 Interferon stimulated gene 15 NM_174366 Forward Reverse GGTATGAGCTGAAGCAGTT ACCTCCCTGCTGTCAAGGT ACTB actin AY141970 Forward Reverse CTGGACTTCGAGCAGGAGAT GGATGTCGACGTCACACTTC Reverse Ribosomal protein L19 NM_001040516 Forward Reverse GCGTGCTTCCTTGGTCTTAG ATCGATCGCCACATGTATCA Class 28 3.54 0.69 2.5 5.3 20 2.87 0.69 1.6 4.2 21 0.30 0.70 1.8 1 30 0.50 0.60 1.8 0.5 GHCR 28 3.07 0.77 2.2 5.5 20 0.57 0.39 0.1 1.2 21 2.33 0.53 1.5 3.3 30 0.28 0.61 2.1 0.5 Class (HH = High GDPR / High GHCR; HL = High GDPR / Low GHCR; LH = Low GDPR / High GHCR; LL = Low GDPR / Low GHCR)

PAGE 95

95 Table 4 3 Descriptive data for the study population Class HH 28 28 26 28 14 1 HL 20 15 13 15 6 0 LH 21 17 14 16 7 1 LL 30 24 16 24 6 2

PAGE 96

96 Figure 4 4 Genetic merit for daughter pregnancy arte ( GDPR ) and heifer conception rate ( GHCR ) breeding values in the study population. The bars represent the division of the population into classes used in th e experiment: HH = High GDPR / High GHCR; HL = High GDPR / Low GHCR; LH = Low GDPR / High GHCR; LL = Low GDPR / Low GHCR. Figure 4 2 O vulatory follicle size (all heifers) according to genetic merit for daughter pregnancy rate (GDPR) and heifer concept ion rate (GHCR) classes HH = High GDPR / High GHCR; HL = High GDPR / Low GHCR; LH = Low GDPR / High GHCR; LL = Low GDPR / Low GHCR. GDPR P < 0.01, GHCR P = 0.12, GDPR x GHCR P = 0.82. -3 -2 -1 0 1 2 3 4 5 6 -3 -2 -1 0 1 2 3 4 5 6 GHCR GDPR LH HH HL LL 10 11 12 13 14 15 16 17 18 Ovulatory follicle size, mm HH HL LH LL

PAGE 97

97 Figure 4 3. Estradiol concentrations at estrus (all heifers), acc ording to genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate ( GHCR ) classes. HH = High GDPR / High GHCR; HL = High GDPR / Low GHCR; LH = Low GDPR / High GHCR; LL = Low GDPR / Low GHCR. GDPR P = 0.02, GHCR P = 0.21, GDPR x GHCR P = 0.60. Figure 4 4. P rogesterone concentrations at estrus, 7 and 14 days after estrus (all heifers), according to genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate ( GHCR ) classes: HH = High GDPR / High GHCR; HL = High GDPR / Low GHCR; LH = Low GDPR / High GHCR; LL = Low GDPR / Low GHCR. Day 0: GDPR P = 0.88 GHCR P = 0.78 GDPR x GHCR P = 0.56 Day 7 and 14: GDPR P = 0.38 GHCR P = 0. 38 GDPR x GHCR P = 0. 17 0.0 1.0 2.0 3.0 4.0 5.0 6.0 Estradiol, pg/mL HH HL LH LL 0 1 2 3 4 5 6 7 0 7 14 Progesterone, ng/mL Day after estrous HH HL LH LL

PAGE 98

98 Figure 4 5. P rogesterone concentrations at estrus, 7, 14, 19 2, 28, and 35 d ays after estrus (only pregnant heifers 35 3 d after service), according to genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate ( GHCR ) classes: HH = High GDPR / High GHCR; HL = High GDPR / Low GHCR; LH = Low GDPR / High GHCR; LL = Low GDPR / Low GHCR. Day 0: GDPR P = 0.08, GHCR P = 0.43, GDPR x GHCR P = 0.46. Day 7, 14, 19 2, 28, and 35: GDPR P = 0.19, GHCR P = 0.98, GDPR x GHCR P = 0.70. Figure 4 6. Interferon stimulated gene 15 ( ISG15 ) 19 2 d ays after estrus (only pregnant heifers 35 3 d after service), according to genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate ( GHCR ) classes: HH = High GDPR / High GHCR; HL = Hi gh GDPR / Low GHCR; LH = Low GDPR / High GHCR; LL = Low GDPR / Low GHCR. GDPR P = 0.87, GHCR P = 0.19, GDPR x GHCR P = 0.07. 0 2 4 6 8 10 0 7 14 19 28 35 Progesterone, ng/mL Day after estrous HH HL LH LL 0 0.5 1 1.5 2 2.5 3 3.5 ISG15, relative mRNA expression HH HL LH LL

PAGE 99

99 Figure 4 7. Pregnancy specific prote in B (PSPB) concentrations 19 2, 28, and 35 d ays after estrus (only pregnant heife rs 35 3 d after service), according to genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate ( GHCR ) classes: HH = High GDPR / High GHCR; HL = High GDPR / Low GHCR; LH = Low GDPR / High GHCR; LL = Low GDPR / Low GHCR. Day 19 2 : GDPR P = 0.87 GHCR P = 0. 58 GDPR x GHCR P = 0.4 1 Day 28 and 35: GDPR P = 0.03 GHCR P = 0. 86 GDPR x GHCR P = 0. 63 Figure 4 8. Insu lin like growth factor 1 (IGF 1) concentrations at estrus, 7, 14, 19 2, 28, and 3 5 d ays after estrus (only pregnant heifers 35 3 d ays after service), according to genetic merit for daughter pregnancy rate ( GDPR ) and heifer conception rate ( GHCR ) classes: HH = High GDPR / High GHCR; HL = High GDPR / Low GHCR; LH = Low GDPR / High GHCR; LL = Low GDPR / Low GHCR. Day 0, 7, 14, 19 2, 28, and 35: GDPR P = 0.50, GHCR P = 0.14, GDPR x GHCR P = 0.48. 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 19 28 35 PSPB, ng/mL Day after estrous HH HL LH LL 0 20 40 60 80 100 120 140 160 0 7 14 19 28 35 IGF 1, ng/mL Day after estrous HH HL LH LL

PAGE 100

100 CHAPTER 5 CONCLUSION Information about the efficacy and differences in response to PGF formulations is not abundant for dairy heifers and data available from lactating dairy cows is controversial Results presented herein provide new evidence about the differences in estr o us behavior and hazard of estrus following PGF treatments suggesting that the heifers treated with cloprostenol sodium have lower progesterone concentration at estrus and are detected in estrus faster compared with dinoprost tromethamine Despite the fact that cloprostenol sodium increased the proportion of heifers detected in estrus within 7 days of treatment and hazard of estrus it did not affect Preg/Serv or hazard of pregnancy, the most important outcome s for dairy producers Therefore, selection of PGF formulation may be according to other parameters than efficacy. Benefits of the u se of an AED for detection of estrus of dairy heifers are not definite and may be a consequence of dairy heifers having greater duration and intensity of estrus compare d with lactating dairy cows In the experiment presented herein, however, AED improv ed the hazard of pregnancy likely because it improved the accuracy of estrus detection, observed as greater Preg/Serv The feasibility of the use of AED for dairy heifers ho wever, remain s uncertain and whether a farm will benefit from adopting the system will vastly vary according to the design of the reproductive program, and especially current efficiency and accuracy of estrus detection on each specific dairy Genomic fe rtility traits such as daughter pregnancy rate (GDPR) and heifer conception rate (GHCR) although vastly used in genomic selection for dairy cattle lack information on their impact on physiological changes driv ing improvements in reproductive performance. Furthermore, the association among GDPR and GHCR and important phenotypes such as estr o us behavior have seldom been evaluated The results from the current studies contribute to the

PAGE 101

101 understand ing on how GDPR and GHCR alter estr o us behavior through physiological alterations, particularly of the ovulatory follicle and concentration of estradiol at estrus Results presented herein reinforce the strategy of selecti ng heifers and cows for GDPR, which should lead to selection of animals with greater ovula tory follicle size, estradiol concentrations and improved estrus expression, duration, and intensity. On the other hand, the data from the current study suggest that GHCR could lead to reduction in estr o us behavior and could potentially lead to reproductiv e losses in subsequent generations. Together, these studies contribute with novel information that can be used by dairy farmers, researchers and other members of the dairy industry do advance and improve reproductive performance, improve genetic selection strategies, and profitability of dairy herds

PAGE 102

102 LIST OF REFERENCES AIPL 2013. Changes to evaluation system. Accessed Oct. 01, 2017. https://www.uscdcb.com/reference/changes/eval1312.htm ARS 2017. Possibilities in an age of genomics: The future of the br eeding index. Accessed Oct. 01, 2017. https://www.ars.usda.gov/research/publications/publication/?seqNo115=342683 Van Amburgh, M.E., D.M. Galton, D.E. Bauman, and R.W. Everett. 1997. Management and economics of extended calving intervals with use of bovine somatotropin. Livest. Prod. Sci. 50:15 28. doi:10.1016/S0301 6226(97)00069 9. Burnett, T.A., A.M.L. Madureira, B.F. Silper, A.C.C. Fernandes, and R.L.A. Cerri. 2017. Integrating an automated activity monitor i nto an artificial insemination program and the associated risk factors affecting reproductive performance of dairy cows. J. Dairy Sci. 1 14. doi:10.3168/jds.2016 12246. Cabrera, V.E., and A.S. Kalantari. 2016. Economics of production efficiency: Nutrition al grouping of the lactating cow1. J. Dairy Sci. 99:825 841. doi:10.3168/jds.2015 9846. Calus, M.P.L., Y. de Haas, M. Pszczola, and R.F. Veerkamp. 2013. Predicted accuracy of and response to genomic selection for new traits in dairy cattle. Animal 7:183 191. doi:10.1017/S1751731112001450. Chanvallon, A., S. Coyral castel, J. Gatien, J. Lamy, D. Ribaud, C. Allain, P. Clment, and P. Salvetti. 2014. Theriogenology Comparison of three devices for the automated detection of estrus in dairy cows. Therio genology 82:734 741. doi:10.1016/j.theriogenology.2014.06.010. Cochran, S.D., J.B. Cole, D.J. Null, and P.J. Hansen. 2013. Single nucleotide polymorphisms in candidate genes associated with fertilizing ability of sperm and subsequent embryonic development in cattle.. Biol. Reprod. 89:69. doi:10.1095/biolreprod.113.111260. Cochran, S.D., J.B. Cole, D.J. Null, P.J. Hansen, S.D. Cochran, J.B. Cole, D.J. Null, and P.J. Hansen. 2013. Single Nucleotide Polymorphisms in Candidate Genes Associated with Fertilizin g Ability of Sperm and Subsequent Embryonic Development in Cattle Published Genes Associat 89. doi:10.1095/biolreprod.113.111260. Colazo, M.G., I. Lpez Helguera, A. B ehrouzi, D.J. Ambrose, and R.J. Mapletoft. 2017. Relationship between circulating progesterone at timed AI and fertility in dairy cows subjected to GnRH based protocols. Theriogenology 94:15 20. doi:10.1016/j.theriogenology.2017.02.004. Cole, J.B., and D. J. Null. 2009. Genetic evaluation of lactation persistency for five breeds of dairy cattle. J. Dairy Sci. 92:2248 2258. doi:10.3168/jds.2008 1825.

PAGE 103

103 Cummins, S.B., P. Lonergan, A.C.O. Evans, and S.T. Butler. 2012. Genetic merit for fertility traits in Holst hormones and estrus behavior. J. Dairy Sci. 95:3698 3710. doi:10.3168/jds.2011 4976. Dekkers, J.C.M. 2007. Prediction of response to marker assisted and genomic selection using selection index theory. J. Anim. Breed. Genet. 124:331 341. doi:10.1111/j.1439 0388.2007.00701.x. Denis Robichaud, J., R.L.A. Cerri, A. Jones Bitton, and S.J. LeBlanc. 2016. Survey of reproduction management on Canadian dairy farms. J. Dairy Sci. 99:9339 9351. doi:10.3168/jds.2016 11445. Dolecheck, K., W. Silvia, G. Heersche, C. Wood, K. McQuerry, and J. Bewley. 2015. A comparison of timed artificial insemination and automated activity monitoring with hormone intervention in 3 commercial dairy herds.. J. Dairy Sci. 99:1506 1514. doi:10.3168/jds.2015 9914. Ettema, J.F., and J.E.P. Santos. 2004. Impact of Age at Calving on Lactation, Reproduction, Health, and Income in First Parity Holsteins on Commercial Farms. J. Dairy Sci. 87:2730 2742. doi:10.3168/jds. S0022 0302(04)73400 1. Ferraz Junior, M.V.C., A. V. Pires, M.V. Biehl, M.H. Santos, D.M. Polizel, D.D. Nepomuceno, R. Sartori, J.B. Barreto Filho, J.R.S. Gonalves, and M.L. Day. 2016. Luteolysis in Bos indicus cows on Days 5 and 7 of estrous cycle with Theriogenology 86:1268 1274. doi:10.1016/j.theriogenology.2016.04.067. Forde, N., M.E. Beltman, P. Lonergan, M. Diskin, J.F. Roche, and M.A. Crowe. 2011. Oestrous cycles in Bos taurus cattle. Anim. Reprod. Sci. 124:163 169. doi:10. 1016/j.anireprosci.2010.08.025. Fricke, P., J. Giordano, A. Valenza, G. Lopes, M. Amundson, and P. Carvalho. 2014a. Reproductive performance of lactating dairy cows managed for first service using timed artificial insemination with or without detect ion of estrus using an activity monitoring system.. J. Dairy Sci. 97:2771 81. doi:10.3168/jds.2013 7366. Fricke, P.M., P.D. Carvalho, J.O. Giordano, A. Valenza, G.L. Jr, and M.C. Amundson. 2017. of new technologies 134 143. doi:10.1017/S1751731114000299. Fricke, P.M., P.D. Carvalho, J.O. Giordano, A. Valenza, G. Lopes, and M.C. Amundson. 2014b. Expression and detection of estrus in dairy cows: the role of new technologies. Animal 8:134 143. doi:1 0.1017/S1751731114000299. Gabler, M.T., P.R. Tozer, and A.J. Heinrichs. 2000. Development of a Cost Analysis Spreadsheet for Calculating the Costs to Raise a Replacement Dairy Heifer. J. Dairy Sci. 83:1104 1109. doi:10 .3168/jds.S0022 0302(00)74975 7.

PAGE 104

104 Gar ca R uiz, A., J.B. Cole, M. Paul, G.R. Wiggans, F.J. Ruiz lpez, and P. Curtis. 2016. Changes in genetic selection differentials and generation intervals in US Holstein dairy cattle as a result of genomic selection 113. doi:10.1073/pnas.1611570113. Giffor d, C. a, a M. Assiri, M.C. Satterfield, T.E. Spencer, and T.L. Ott. 2008. Receptor transporter protein 4 (RTP4) in endometrium, ovary, and peripheral blood leukocytes of pregnant and cyclic ewes. Biol. Reprod. 79:518 24. doi:10.1095/biolreprod.108.069468. Giordano, J.O., A.S. Kalantari, P.M. Fricke, M.C. Wiltbank, and V.E. Cabrera. 2012. A daily herd Markov chain model to study the reproductive and economic impact of reproductive programs combining timed artificial insemination and estrus detection. J. Dairy Sci. 95:5442 5460. doi:10.3168/jds.2011 4972. Goravanahally, M.P., M. Salem, J. Yao, E.K. Inskeep, and J. a Flores. 2009. Differential gene expression in the bovine corpus luteum during transition from early phase to midphase and its potential role in acquisition of luteolytic sensitivity to prostaglandin F2 alpha.. Biol. Reprod. 80:980 8. doi:10.1095/biolreprod.108.069518. Green, J.A., T.E. Parks, M.P. Avalle, B.P. Telugu, A.L. McLain, A.J. Peterson, W. McMillan, N. Mathialagan, R.R. Hook, S. Xie, and R.M. Roberts. 2005. The establishment of an ELISA for the detection of pregnancy associated glycoproteins (PAGs) in the serum of pregnant cows and heifer s. Theriogenology 63:1481 1503. doi:10.1016/j.theriogenology.2004.07.011. Green, J.C., C.S. Okamur a, S.E. Poock, and M.C. Lucy. 2010. Measurement of interferon tau (IFN ??) stimulated gene expression in blood leukocytes for pregnancy diagnosis within 18 20d after insemination in dairy cattle. Anim. Reprod. Sci. 121:24 33. doi:10.1016/j.anireprosci.2010 .05.010. Grummer, R.R., D.G. Mashek, and a. Hayirli. 2004. Dry matter intake and energy balance in the transition period. Vet. Clin. North Am. Food Anim. Pract. 20:447 470. doi:10.1016/j.cvfa.2004.06.013. Hoffman, P.C. 1997. Optimum Body Size of Holste in Replacement Heifers. J. Anim. Sci. 75:836 845. doi:/1997.753836x. Holmberg, M., and L. Andersson Eklund. 2006. Quantitative trait loci affecting fertility and calving traits in Swedish dairy cattle.. J. Dairy Sci. 89:3664 3671. doi:10.3168/jds.S0022 03 02(06)72406 7. Humblot, P., S. Camous, J. Martal, J. Charlery, N. Jeanguyot, M. Thibier, and R.G. Sasser. 1988. Pregnancy specific protein B, progesterone concentrations and embryonic mortality during early pregnancy in dairy cows. J Reprod Fertil 83:215 223. doi:10.1530/jrf.0.0830215.

PAGE 105

105 Jinks, E.M., M.F. Smith, J.A. Atkins, K.G. Pohler, G.A. Perry, M.D. MacNeil, A.J. Roberts, R.C. Waterman, L.J. Alexa nder, and T.W. Geary. 2013. Preovulatory estradiol and the establishment and maintenance of pregnancy in s uckled beef cows. J. Anim. Sci. 91:1176 1185. doi:10.2527/jas.2012 5611. Jonas, E., and D.J. de Koning. 2015. Genomic selection needs to be carefully assessed to meet specific requirements in livestock breeding programs. Front. Genet. 5:1 8. doi:10.3389/f gene.2015.00049. Kamphuis, C., B. DelaRue, C.R. Burke, and J. Jago. 2012. Field evaluation of 2 collar mounted activity meters for detecting cows in estrus on a large pasture grazed dairy farm.. J. Dairy Sci. 95:3045 56. doi:10.3168/jds.2011 4934. Kaniyamattam, K., M.A. Elzo, J.B. Cole, A. De Vries, M.P. Calus, P. Bijma, R.F. Veerkamp, A.P. de Roos, C. Schrooten, R.F. Veerkamp, J.A.M. van Arendonk, A. De Vries, B.J. Conlin, C.M. Dematawewa, R.E. Pearson, P.M. VanRaden, A.A. Dijkhuizen, J. Stelwagen, J.A. Renkema, L.S. Eikje, L.R. Schaeffer, T. dny, G. Klemetsdal, J. degrd, J.F. Ettema, S. stergaard, M.K. Srensen, D.S. Falconer, T.F. Mackay, R. Frankham, K.N. Galvo, P. Federico, A. De Vries, G.M. Schuenemann, M. Georges, J.M. Massey, N.G. Hosse in Zadeh, A. Nejati Javaremi, S.R. Miraei Ashtiani, H. Kohram, P.J. Hansen, J. Block, A.M. Heikkil, J. Peippo, L. Hjort, J.F. Ettema, M. Kargo, A.C. Srensen, W.D. Kelton, A.M. Law, C.L. Meyer, P.J. Berger, K.J. Koehler, J.R. Thompson, C.G. Sattler, S.V. Morant, A. Gnanasakthy, R.A. Mrode, NRC, N.J. Olynk, C.A. Wolf, L.D. Pedersen, M. Kargo, P. Berg, J. Voergaard, L.H. Buch, A.C. Srensen, P.J. Pinedo, A. De Vries, P.J. Pinedo, A. De Vries, D.W. Webb, J.E. Pryce, H.D. Daetwyler, J.E. Santos, W.W. Thatcher R.C. Chebel, R.L. Cerri, K.N. Galvao, C. Schrooten, H. Bovenhuis, J.A.M. van Arendonk, P. Bijma, G.E. Seidel, J.T. Sorensen, J.A.M. van Arendonk, P.M. VanRaden, G.R. Wiggans, K.A. Weigel, K.A. Weigel, P.C. Hoffman, W. Herring, T.J. Lawlor, K.A. Weigel, S .W. Lin, and P.D.P. Wood. 2016. Stochastic dynamic simulation modeling including multitrait genetics to estimate genetic, technical, and financial consequences of dairy farm reproduction and selection strategies.. J. Dairy Sci. 99:8187 202. doi:10.3168/jds .2016 11136. Kommadath, A., H.A. Mulder, A.A.C. De Wit, H. Woelders, M.A. Smits, B. Beerda, R.F. Veerkamp, A.C.J. Frijters, and M.F.W. Pas. 2017. Gene expression patterns in anterior pituitary associated with quantitative measure of oestrous behaviour in dairy cows 1297 1307. doi:10.1017/S1751731110000303. Kommadath, A., M.F.W. Pas, and M.A. Smits. 2013. Gene coexpression network analysis identifies genes and biological processes shared among anterior pituitary and brain areas that affect estrous behavior in dairy cows 2583 2595. Kommadath, A., H. Woelders, B. Beerda, H.A. Mulder, A.A.C. de Wit, R.F. Veerkamp, M.F.W. te Pas, and M.A. Smits. 2011. Gene expression patterns in four brain areas associate with quantitative measure of estrous behavior in dairy cows.. BMC Genomics 12:200. doi:10.1186/1471 2164 12 200.

PAGE 106

106 Kuhn, M.T., J.L. Hutchison, and G.R. Wiggans. 2006a. Characterization of Holstein heifer fertility in the United States.. J. Dairy Sci. 89:4907 4920. doi:10.3168/jds.S0022 0302(06)72541 3. Kuhn, M .T., J.L. Hutchison, and G.R. Wiggans. 2006b. Characterization of Holstein heifer fertility in the United States.. J. Dairy Sci. 89:4907 4920. doi:10.3168/jds.S0022 0302(06)72541 3. Lima, F.S., E.S. Ribeiro, R.S. Bisinotto, L.F. Greco, N. Martinez, M. Ams talden, W.W. Thatcher, and J.E.P. Santos. 2013. Hormonal manipulations in the 5 day timed artificial insemination protocol to optimize estrous cycle synchrony and fertility in dairy heifers.. J. Dairy Sci. 96:7054 65. doi:10.3168/jds.2013 7093. Lopes, G., C.R. Johnson, L.G.D. Mendona, P.R.B. Silva, J.G.N. Moraes, a Ahmadzadeh, J.C. Dalton, and R.C. Chebel. 2013. Evaluation of reproductive and economic outcomes of dairy heifers inseminated at induced estrus or at fixed time after a 5 day or 7 da y progesterone insert based ovulation synchronization protocol.. J. Dairy Sci. 96:1612 22. doi:10.3168/jds.2012 5971. Lopez, H., D.Z. Caraviello, L.D. Satter, P.M. Fricke, and M.C. Wiltbank. 2005. Relationship Between Level of Milk Production and Multiple Ovulations in Lactating Dairy Cows. J. Dairy Sci. 88:2783 2793. doi:10.3168/jds.S0022 0302(05)72958 1. Lucy, M.C. 2001. Reproductive Loss in High Producing Dairy Cattle: Where Will It End?. J. Dairy Sci. 84:1277 1293. doi:10.3168/jds.S0022 0302(01)70158 0. Machado, V.S., R. Neves, F.S. Lima, and R.C. Bicalho. 2017. The effect of Presynch Ovsynch protocol with or without estrus detection on reproductive performance by parity, and the long term effect of these different management strategies on milk produc tion, reproduction, health and survivability of dairy cows. Theriogenology 93:84 92. doi:10.1016/j.theriogenology.2017.01.041. Madsen, C.A., G.A. Perry, C.L. Mogck, R.F. Daly, M.D. MacNeil, and T.W. Geary. 2015. Effects of preovulatory estradiol on embryo survival and pregnancy establishment in beef cows. Anim. Reprod. Sci. 158:96 103. doi:10.1016/j.anireprosci.2015.05.006. Martins, J.P.N., R.K. Policelli, and J.R. Pursley. 2011a. Luteolytic effects of cloprostenol sodium in lactating dairy cows treated w ith G6G/Ovsynch. J. Dairy Sci. 94:2806 2814. doi:http://dx.doi.org/10.3168/jds.2010 3650. Martins, J.P.N., R.K. Policelli, and J.R. Pursley. 2011b. Luteolytic effects of cloprostenol sodium in lactating dairy cows treated with G6G/Ovsynch.. J. Dairy Sci. 94:2806 2814. doi:10.3168/jds.2010 3650.

PAGE 107

107 Matsuyama, S., T. Kojima, S. Kato, and K. Kimura. 2012. Relationship between quantity of IFNT estimated by IFN stimulated gene expression in peripheral blood mononuclear cells and bovine embryonic mortality after A I or ET. Reprod. Biol. Endocrinol. 10:21. doi:10.1186/1477 7827 10 21. Mikshowsky, A.A., D. Gianola, and K.A. Weigel. 2016. Assessing genomic prediction accuracy for Holstein sires using bootstrap aggregation sampling and leave one out cross validation.. J. Dairy Sci. 100:453 464. doi:10.3168/jds.2016 11496. Neves, R., S. LeBlanc, J. Giordano, M. Stangaferro, R. Wijma, W. Chandler, and R. Watters. 2015. Reproductive performance of dairy cows managed with a program aimed at increasing insemination of cows in estrus based on increased physical activity and fertility of timed artificial inseminations.. J. Dairy Sci. 98:2488 501. doi:10.3168/jds.2014 8961. Neves, R.C., and S.J. LeBlanc. 2015. Reproductive management practices and performance of Canadian dairy herds using automated activity monitoring systems. J. Dairy Sci. 98:2801 2811. doi:10.3168/jds.2014 8221. NRC (National Research Council). 2001. Nutrient Requirements of Dairy Cattle. 7th rev. ed. Natl. Acad. Sci., Washington, DC. Ortega, M.S., A.C. De nicol, J.B. Cole, D.J. Null, and P.J. Hansen. 2016. Use of single nucleotide polymorphisms in candidate genes associated with daughter pregnancy rate for prediction of genetic merit for reproduction in Holstein cows 288 297. doi:10.1111/age.12420. Pahl, C ., E. Hartung, K. Mahlkow Nerge, and A. Haeussermann. 2015. Feeding characteristics and rumination time of dairy cows around estrus.. J. Dairy Sci. 98:148 54. doi:10.3168/jds.2014 8025. Penteado, D., and M. Dias. 2013. Review Article Compte rendu 54:1041 1052. Pereira, M.H.C., M.C. Wiltbank, and J.L.M. Vasconcelos. 2016. Expression of estrus improves fertility and decreases pregnancy losses in lactating dairy cows that receive artificial insemination or embryo transfer. J. Dairy Sci. 99:2237 2247. doi:10.3168/jds.2015 9903. Pollott, G.E. 2011. Short communication: Do Holstein lactations of vaired lengths have different characteristics?. J. Dairy Sci. 94:6173 6180. doi:10.3168/jds.2011 4467. Pryce, J.E., M.D. Royal, P.C. Garnsworthy, and I.L. Mao. 2004. Fertility in the high producing dairy cow. Livest. Prod. Sci. 86:125 135. doi:10.1016/S0301 6226(03)00145 3. Pursley, J.R., J.P.N. Martins, C. Wright, and N.D. Stewart. 2012. Compared to dinoprost tromethamine, cloprostenol sodium increased rates o f estrus detection, conception and pregnancy in lactating dairy cows on a large commercial dairy. Theriogenology 78:823 829. doi:10.1016/j.theriogenology.2012.03.032.

PAGE 108

108 Reeves, P.R. 1978. Distribution, Elimination, and Residue Studies in the Cow with the Synthetic Prostaglandin Estrumate. J. Agric. Food Chem. 26:152 155. doi:10.1021/jf60215a001. Reith, S., and S. Hoy. 2017. Review: Behavioral signs of estrus and the potential of fully automated systems for detection of estrus in dairy cattle. Animal 1 10. doi:10.1017/S1751731117001975. Ribeiro, E.S., R.G.S. Bruno, A.M. Farias, J. a Hernndez Rivera, G.C. Gomes, R. Surjus, L.F. V Becker, A. Birt, T.L. Ott, J.R. Branen, R.G. Sasser, D.H. Keisler, W.W. Thatcher, T.R. Bilby, and J.E.P. Santos. 2014. Low doses of bovine somatotropin enhance conceptus development and fertility in lactating dairy cows.. Biol. Reprod. 90:10. doi:10.1095/biolreprod.113.114694. Ribeiro, E.S., K.N. Galvo, W.W. Thatcher, and J.E.P. Santos. 2012. Economic aspects of applying rep roductive technologies to dairy herds. Anim. Reprod. 9:370 387. Roelofs, J.B., F.J.C.M. Van Eerdenburg, N.M. Soede, and B. Kemp. 2005. Various behavioral signs of estrous and their relationship with time of ovulation in dairy cattle. Theriogenology 63:136 6 1377. doi:10.1016/j.theriogenology.2004.07.009. Sangsritavong, S., D.K. Combs, R. Sartori, L.E. Armentano, and M.C. Wiltbank. 2002. High Feed Intake Increases Liver Blood Flow and Metabolism of Progesterone and Estradiol in Dairy Cattle. J. Dairy S ci. 85:2831 2842. doi:10.3168/jds.S0022 0302(02)74370 1. Santos, J.E.P., R.S. Bisinotto, E.S. Ribeiro, F.S. Lima, L.F. Greco, C.R. Staples, and W.W. Thatcher. 2010. Applying nutrition and physiology to improve reproduction in dairy cattle.. Soc. Repro d. Fertil. Suppl. 67:387 403. doi:10.5661/RDR VII 387. Sartori, R., J.M. Haughian, R.D. Shaver, G.J.M. Rosa, and M.C. Wiltbank. 2004. Comparison of ovarian function and circulating steroids in estrous cycles of Holstein heifers and lactating cows.. J. Dai ry Sci. 87:905 920. doi:10.3168/jds.S0022 0302(04)73235 X. Shirasuna, K., H. Matsumoto, S. Matsuyama, K. Kimura, H. Bollwein, and A. Miyamoto. 2013. Possible role of interferon tau on the bovine corpus luteum and neutrophils during the early pregnancy. do i:10.1530/REP 15 0085. Shrestha, H.K., M.A. Beg, R.R. Burnette, and O.J. Ginther. 2012. Plasma Clearance and Half Life of Prostaglandin F2alpha: A Comparison Between Mares and Heifers1. Biol. Reprod. 87:1 6. doi:10.1095/biolreprod.112.100776. Silper, B.F ., A.M.L. Madureira, M. Kaur, T.A. Burnett, and R.L.A. Cerri. 2015a. Short monitoring systems 3158 3165.

PAGE 109

109 Silper, B.F., I. Robles, A.M.L. Madureira, T.A. Burnett, M.M. Reis, A.M. de Passill, J. Rushen, and R.L.A. Cerri. 2015b. Automated and visual measurements of estrous behavior and their sources of variation in Holstein heifers. I: Walking activity and behavior frequency. Theriogenology 84:312 320. doi:10.1016/j.theri ogenology.2014.12.029. Silva, T. V, F.S. Lima, W.W. Thatcher, and J.E.P. Santos. 2015. Synchronized ovulation for first insemination improves reproductive performance and reduces cost per pregnancy in dairy heifers.. J. Dairy Sci. 98:7810 22. doi:10.316 8/jds.2015 9704. Silvestre, A.M., A.M. Martins, V.A. Santos, M.M. Ginja, and J.A. Colao. 2009. Lactation curves for milk, fat and protein in dairy cows: A full approach. Livest. Sci. 122:308 313. doi:10.1016/j.livsci.2008.09.017. Stevenson, J., S. Hill, R. Nebel, and J. DeJarnette. 2014. Ovulation timing and conception risk after automated activity monitoring in lactating dairy cows. J. Dairy Sci. 97:4296 4308. doi:http://dx.doi.org/10.3168/jds.2013 7873. Stevenson, J.L., J.A. Rodrigues, F.A. Braga, S. Bitente, J.C. Dalton, J.E.P. Santos, and R.C. Chebel. 2008. Effect of Breeding Protocols and Reproductive Tract Score on Reproductive Performance of Dairy Heifers and Economic Outcome of Breeding Programs. J. Dairy Sci. 91:3424 3438. doi:10.3168/jds.2007 0 804. Stevenson, J.S., and G.C. Lamb. 2016. Contrasting effects of progesterone on fertility of dairy and beef cows. J. Dairy Sci. 99:5951 5964. doi:10.3168/jds.2015 10130. Stevenson, J.S., and a. P. Phatak. 2010. Rates of luteolysis and pregnancy in dai ry cows after treatment with cloprostenol or dinoprost. Theriogenology 73:1127 1138. doi:10.1016/j.theriogenology.2010.01.014. Stevenson, J.S., J.F. Smith, and D.E. Hawkins. 2000. Reproductive Outcomes for Dairy Heifers Treated with Combinations of Pr Releasing Hormone1. J. Dairy Sci. 83:2008 2015. doi:http://dx.doi.org/10.3168/jds.S0022 0302(00)75079 X. Stevenson, J.S., M.W. Smith, J.R. Jaeger, L.R. Corah, and D.G. Lefever. 1994. Detection of estrus by vi sual observation and radiotelemetry_Stevenson etal 1996.pdf 729 735. for genomic selection of dairy cattle by including dominance effects. PLoS One 9. doi:10.1371/journal.pone.0103934. Valenza, A., J.O. Giordano, G. Lopes, L. Vincenti, M.C. Amundson, and P.M. Fricke. 2012. Assessment of an accelerometer system for detection of estrus and treatment with gonadotropin releasing hormone at the time of insem ination in lactating dairy cows. J. Dairy Sci. 95:7115 7127. doi:10.3168/jds.2012 5639.

PAGE 110

110 Valldecabres Torres, X., E. Garca Rosell, A. Garca Muoz, and J. Cuervo Arango. 2012. Effects of d cloprostenol dose and corpus luteum age on ovulation, luteal func tion, and morphology in nonlactating dairy cows with early corpora lutea.. J. Dairy Sci. 95:4389 95. doi:10.3168/jds.2011 5284. VanRaden, P.M., A.H. Sanders, M.E. Tooker, R.H. Miller, H.D. Norman, M.T. Kuhn, and G.R. Wiggans. 2004. Development of a nation al genetic evaluation for cow fertility.. J. Dairy Sci. 87:2285 92. doi:10.3168/jds.S0022 0302(04)70049 1. VanRaden, P.M., C.P. Van Tassell, G.R. Wiggans, T.S. Sonstegard, R.D. Schnabel, J.F. Taylor, and F.S. Schenkel. 2009. Invited Review: Reliability of genomic predictions for North American Holstein bulls. J. Dairy Sci. 92:16 24. doi:10.3168/jds.2008 1514. Vasconcelos, J.L.M., R. Sartori, H.N. Oliveira, J.G. Guenther, and M.C. Wiltbank. 2001. Reduction in size of the ovulatory follicle reduces subseque nt luteal size and pregnancy rate. Theriogenology 56:307 314. doi:10.1016/S0093 691X(01)00565 9. luteum during its refractory period in cows.. BMC Vet. Res. 8:2 20. doi:10.1186/1746 6148 8 220. Wiggans, G.R., P.M. VanRaden, and T.A. Cooper. 2011. The genomic evaluation system in the United States: Past, present, future. J. Dairy Sci. 94:3202 3211. doi:10.3168/jds.2010 3866. Wiltbank, M., H. Lopez, R. Sartori, S. Sangsritavong, and A. Gmen. 2006. Changes in reproductive physiology of lactating dairy cows due to elevated steroid metabolism. Theriogenology 65:17 29. doi:10.1016/j.theriogenology.2005.10.003. Woelders, H., T. Van Der Lende, A. Kommadath, M.F.W. Pas, M.A. Smits, and L.M.T.E. Kaal. review. Animal 8:754 764. doi:10.1017/S1751731114000342.

PAGE 111

111 BIOGRAPHICAL SKETCH Anderson Veronese was born in Viadutos, a small agriculture based town in the southern state of Rio Grande do Sul, Brazil. He is the only child of Jaime Dionisio Veronese and Marines Bohm Veronese H is parents and grandparents Felix and Leonora Bohm owned a far m wher e Anderson grew up. Since young, Anderson started working with his family on the farm, where they milked dairy cows, and finished swine for slaughter. At age of 15, Anderson did a training to learn how artificially inseminate cows. One year later, a fter having experience on breeding cows in his family dairy, he and an older cousin decided to partner and started a small business, providing artificial insemination service to dairy farmers in the town. After graduating on high school, Anderson decided t o pursue a carrier in Veterinary Sciences, was approve on Fede ral Institute of Santa Catarina and started college in 2010. During college, Anderson started working closely with research under the supervision of Dr. Angela Veiga, an early mentor who develop his interest in science. During college breaks, Anderson did externships in a dairy production medicine and nutrition consulting in a develop several skills, had the chance to network with experienced professionals and improved his knowledge about the dairy industry, as well as consulting and dealing with dairy farmers. In 2013, Anderson received a scholarship from the Brazilian academic mobility progra unded by the federal government of Brazil, and came to US to spend one year as an exchange student at Maricopa Colleges, Phoenix AZ. During this time, he improved his English skills, and did courses related to his field. Anderson returned to Brazil in July 2014, spend one year to finis h his required classes, and returned to USA to do an externship under the supervision of Dr. Ricardo Chebel in July 2015. Fo llowing up the externship, Anderson was invite to stay at University of Florida to wor k with Dr. Chebel and pursue a Master of S cien ce. He decided to accept the invitation, returned to Brazil for graduation in January 2016, and

PAGE 112

112 immediately returned to Gainesville, where he has being working on his research and taking classes for his master degree program. He is expected to graduate in upcoming goals are to pursue a residency in production medicine and a doctoral degree at the same University.