<%BANNER%>

Manipulating Ovarian Function and Uterine Health with the Aim of Improving Fertility in Dairy Cattle

MISSING IMAGE

Material Information

Title:
Manipulating Ovarian Function and Uterine Health with the Aim of Improving Fertility in Dairy Cattle
Physical Description:
1 online resource (301 p.)
Language:
english
Creator:
Lima, Fabio Soares De
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Animal Sciences
Committee Chair:
Santos, Jose Eduardo
Committee Members:
Thatcher, William W
Hansen, Peter J
Risco, Carlos A

Subjects

Subjects / Keywords:
cattle -- fertility -- ovarian -- uterine
Animal Sciences -- Dissertations, Academic -- UF
Genre:
Animal Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
The objectives of this dissertation were to determine reproductive management strategies and hormonal manipulations of ovarian responses to optimize fertility of dairy cattle and to investigate the efficacy of treatments of uterine diseases and its subsequent effects on fertility of dairy cows. Anadditional objective was to determine mechanisms of induced infusion with. T. pyogenes on length of the estrouscycle. Chapter 3 describes effects of 1 (1TAI) or 3 timed AI (3TAI)before natural service in lactating dairy cows not detected in estrus on pregnancy,and revealed that 3TAI had greater hazard of pregnancy and 21-day cycle pregnancy rate than 1TAI. Chapter 4 evaluated the effects of the first gonadotropin-releasing hormone (GnRH) injection of the 5-d timed AI program on ovarian responses and pregnancy per AI (P/AI) and the effect of timing of the final GnRH to induce ovulation relative to AI on P/AI in dairy heifers. Use of GnRH on the 1stday of the 5-d timed AI program resulted in low ovulation and no improvement inP/AI when a single prostaglandin (PG) F2a injection on d 5 was used.Extending proestrus by delaying the final GnRH from 56 to 72 h concurrent withAI benefited P/AI of heifers not in estrus at AI. Chapter 5 investigated the effects of GnRH at the initiation of the 5-d timed AI program combined with two injections of PGF2aon ovarian responses and P/AI in dairy heifers, and the role of progesterone on LH release and ovulation in response to GnRH. Increased ovulation with GnRH given at initiation of 5-d timed AI program combined with two doses of PGF2aon d 5 and 6 optimized P/AI. High progesterone at GnRH administration suppressed LH release and impaired ovulation. Chapter 6 investigated the efficacy of PGF2a asa therapy to reduce the prevalence of subclinical endometritis and to increase P/AIin cows subjected to a timed AI program. The results of this study revealed that treatments with PGF2a before initiation of the timed AI program were unable to improve uterine health and P/AI. Chapter 7 investigated if intrauterine infusion of T. pyogenes influence uterine expression of genes related to inflammation, luteolytic cascade and luteal lifespan.Although T. pyogenes reduced luteal lifespan in half of the cows, only minor changes in uterine gene expression were identified. Chapter 8 investigated the efficacy of ampicillin trihydratedfor treatment of metritis and subsequent effects on fertility. Ampicillin was an efficacious therapy for metritis but had no differential effects on P/AI when compared with ceftiofur hydrochloride.
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.
Statement of Responsibility:
by Fabio Soares De Lima.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Santos, Jose Eduardo.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045956:00001

MISSING IMAGE

Material Information

Title:
Manipulating Ovarian Function and Uterine Health with the Aim of Improving Fertility in Dairy Cattle
Physical Description:
1 online resource (301 p.)
Language:
english
Creator:
Lima, Fabio Soares De
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Animal Sciences
Committee Chair:
Santos, Jose Eduardo
Committee Members:
Thatcher, William W
Hansen, Peter J
Risco, Carlos A

Subjects

Subjects / Keywords:
cattle -- fertility -- ovarian -- uterine
Animal Sciences -- Dissertations, Academic -- UF
Genre:
Animal Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
The objectives of this dissertation were to determine reproductive management strategies and hormonal manipulations of ovarian responses to optimize fertility of dairy cattle and to investigate the efficacy of treatments of uterine diseases and its subsequent effects on fertility of dairy cows. Anadditional objective was to determine mechanisms of induced infusion with. T. pyogenes on length of the estrouscycle. Chapter 3 describes effects of 1 (1TAI) or 3 timed AI (3TAI)before natural service in lactating dairy cows not detected in estrus on pregnancy,and revealed that 3TAI had greater hazard of pregnancy and 21-day cycle pregnancy rate than 1TAI. Chapter 4 evaluated the effects of the first gonadotropin-releasing hormone (GnRH) injection of the 5-d timed AI program on ovarian responses and pregnancy per AI (P/AI) and the effect of timing of the final GnRH to induce ovulation relative to AI on P/AI in dairy heifers. Use of GnRH on the 1stday of the 5-d timed AI program resulted in low ovulation and no improvement inP/AI when a single prostaglandin (PG) F2a injection on d 5 was used.Extending proestrus by delaying the final GnRH from 56 to 72 h concurrent withAI benefited P/AI of heifers not in estrus at AI. Chapter 5 investigated the effects of GnRH at the initiation of the 5-d timed AI program combined with two injections of PGF2aon ovarian responses and P/AI in dairy heifers, and the role of progesterone on LH release and ovulation in response to GnRH. Increased ovulation with GnRH given at initiation of 5-d timed AI program combined with two doses of PGF2aon d 5 and 6 optimized P/AI. High progesterone at GnRH administration suppressed LH release and impaired ovulation. Chapter 6 investigated the efficacy of PGF2a asa therapy to reduce the prevalence of subclinical endometritis and to increase P/AIin cows subjected to a timed AI program. The results of this study revealed that treatments with PGF2a before initiation of the timed AI program were unable to improve uterine health and P/AI. Chapter 7 investigated if intrauterine infusion of T. pyogenes influence uterine expression of genes related to inflammation, luteolytic cascade and luteal lifespan.Although T. pyogenes reduced luteal lifespan in half of the cows, only minor changes in uterine gene expression were identified. Chapter 8 investigated the efficacy of ampicillin trihydratedfor treatment of metritis and subsequent effects on fertility. Ampicillin was an efficacious therapy for metritis but had no differential effects on P/AI when compared with ceftiofur hydrochloride.
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.
Statement of Responsibility:
by Fabio Soares De Lima.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Santos, Jose Eduardo.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045956:00001


This item has the following downloads:


Full Text

PAGE 1

1 MANIPULATING OVARIAN FUNCTION AND UTERINE HEALTH WITH THE AIM OF IMPROVING FERTILITY IN DAIRY CATTLE By F BIO SOARES DE LIMA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLME NT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 F bio Soares de Lima

PAGE 3

3 To my wife Bianca, my parents Gaspar and Dolores, my siblings, and God that uncondition ally supported me with love and comprehension under all circumstances providing me strength and focus necessary to surpass all challenges and successfully accomplish this important goal of my career

PAGE 4

4 ACKNOWLEDGMENTS I would like to express my sincere gr atitude to my advisor Dr. Jos Eduardo Portela Santos, for the opportunity of pursuing a PhD degree at the University of Florida, for his unconditional support, encouragement, inspiration and enthusiasm during the undertaking of this achievement. I vastly admire his outstanding knowledge, work ethics, dedication and passion for research, which has been a role model for young scientists I extended my appreciation to my committee members Dr. Carlos. A. Risco, Dr. Klibs. N. Galvo, Dr. Peter J. Hansen, and Dr. William W. Thatcher, for their insightful contribution throughout the 4 year s of my program helping me to develop as scientist and person. I feel extremely honored for having some of most prestigious reproductive biologists and theriogenologists in th e world as part of committee. I am genuinely thankful to each of you for the countless hour s spent guiding me through design of studies, interpretation and discussion of results and overall development of critical thinking skills to become a better scienti st. I have been always passionate about research a nd the interaction s with this superb group of scientists magnif y my admiration for science and how it can enrich lives and how it advance s knowledge. I would like to extend my appreciation to Dr. Marcel Ams tald en from Texas A&M University for his collaboration helping me with sample analysis for luteinizing hormone I also would like to thanks Dr. Xiaoling Wang and Dr. Mary Reinhard from the College of Veterinary Medicine at the University of Florida for the ir help with laborator y procedures and analyse s. I owe a special thanks to all of my lab mates Rafael S. Bisinotto, Eduardo S. Ribeiro, Leandro F. Greco, Natalia P. Martinez, Letcia D. P. Sinedino and Gabriel C.

PAGE 5

5 Gomes. Their innumerable hours helping de signing studies, discussing ideas, solving problems and conducting research in the lab, in the farm and in our offices were essential to conclude successfully all the work presented in this dissertation. I extend my appreciation to the Animal Sciences gra duate students Guilherme Marquezini, Izabella Thompson, Paula Morelli, Sha Tao, Davi Arajo, Dan Wang, Jo o Henrique Bittar, Vitor Mercadante and Milerky Perdomo ; and visiting students Maurcio Favoreto, Henderson Ayres, Rafael Marsola, Mariana Carvalho, Ana Lcia Sevarolli, Ricardo Surjus, Alana Calaa, Andressa Ranieri, Bianca Libanori, Erika Ganda Luis Fernando Vilela, Vinicius Rezende, Wedson Costa, Jos Tiago Neves Neto, Thiago Villar, Anglica Pedrico, Achilles Vieira Neto, Guilherme Vasconcellos, R odolfo Mingoti, Tony Bruinje Eduard Sole, Javier Juarez, Pedro Monteiro Andr Dias, Raylon Maciel, Jssica Felice and Alberto Zerlotini for their valuable help with the experiments. I would like to extend my appr graduate program at the University of Florida for all support and for the alumni fellowship that funded my PhD program. I also would like to thank the faculty and staff of the Department of Animal Science s at the University of Florida. In particular, Dr. Alan Ealy and Dr. Joel Yelich for allowing me to use their laboratories, Sergji Sennikov and Joyce Hayen for their assistance with sample processing and general lab oratory procedures. I would like to thank Joann M. F ischer for all her help with the paperwork throug hout my PhD program Finally, I would li ke to thank Eric Diepersloot, Jay Lemmermen and Sher r y Hay at the University of Florida Dairy Unit for their assistance with animal handling. I also would like to thank Dr. Carlos Risco and Dr. Klibs Galvo at the C ollege of Veterinary Medicine at the University of Florida, for sharing their laboratory.

PAGE 6

6 I am very grateful to the owner of Alliance Dairies, Ronald St. John, and staff for use of their cows and facilities. Special thanks to Dr. Nilo Francisco, Pete Hethe rington, Guy Wayne, Antonio, Franklin, Geraldo, Ricardo, Felino, Tino and Amadeo de Pa z for assistance with the experiments. I am also very grateful to the owners of Dairy Production System, Dave Samural and Michael Pedreiro, and staff for use of their cow s and faci lities in some of my experiments. I thank all my friends in Gainesville that throughout the last 4 years were like family to me and helped me overcome t he distance from home. I especially thank Rafael Bisinotto, Leandro Greco, Eduardo Ribeiro an d Natalia Martinez whom, in addition to the uncountable days of hard work with the design and conduct of all experiments, also made me feel at home in Gainesville and were by my side through all challenges of life. I really appreciate their friendship I o we special recognition to my entire family in Brazil. To my parents, Gaspar and Dolores, for their sacrifice, prayers and unconditional support. I am so grateful to have a model of humbleness and moral that centered my principles and built my character th roughout the years. They are the foundation of who I am and I can only imagine how much sacrifice and pain they have been through to prov ide me an opportunity to pursue my dream and acquire a PhD degree in another country far away from home. I would like t o especially thank my brother and sisters, Eliton Elaine and Andressa for their friendship and great support during my journey. Final and special thanks go to my wife Bianca Marti ns that support ed me with love, patience, complicity and comprehension dur ing this very important stage of my career.

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ ........ 13 LIST OF TABLES ................................ ................................ ................................ .......... 11 LIST OF ABBREVIATIONS ................................ ................................ ........................... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 20 2 LITERATURE REVIEW ................................ ................................ .......................... 34 Estrous Cycle in Dairy Cattle ................................ ................................ .................. 34 Follicle Development and Function ................................ ................................ ......... 40 Corpus Luteum Formation, Function and Regressio n ................................ ............. 49 Reproductive Management, Efficiency and Timed Artificial Insemination ............... 61 Uterine Diseases Relevance and Characterization ................................ ................. 73 Uterine Disease Etiology: Immunological and Microbiological Aspects .................. 78 Treatments of Uterine Diseases ................................ ................................ ............. 92 3 EFFECT OF ONE OR THR EE TIMED ARTIFICIAL INSEMINATIONS BEFORE NATURAL SERVICE ON R EPRODUCTIVE PERFORMA NCE OF LACTATING DAIRY COWS NOT OBSER VED FOR DETECTION OF ESTRUS ...................... 103 Introductory Remarks ................................ ................................ ............................ 104 Materials and Methods ................................ ................................ .......................... 106 Cows, Housing and Diets ................................ ................................ .............. 106 Treatments, Exclusion Criteria and Reproductive Programs .......................... 107 Bull Management ................................ ................................ ........................... 109 Body Condition Scor i ng ................................ ................................ .................. 110 Seasonality ................................ ................................ ................................ ..... 110 Experimental Design and Statistical Analy sis ................................ ................. 111 Results ................................ ................................ ................................ .................. 113 Pregnancy per AI to the First Timed AI ................................ .......................... 113 Reproductive Performance During the Entire Lactation ................................ 114 Discussion ................................ ................................ ................................ ............ 116 Conclusion ................................ ................................ ................................ ............ 121 4 EFFECTS OF GNRH AT INITIATION OF THE 5 D TIMED A I PROGRAM AND TIMING OF INDUCTION OF OVULATION RELATIVE TO AI ON OVARIAN DYNAMICS AND FERTILITY OF DAIRY HEIFERS ................................ ............. 127

PAGE 8

8 Introductory Remarks ................................ ................................ ............................ 128 Materials and Methods ................................ ................................ .......................... 130 Experiment 1 ................................ ................................ ................................ .. 130 Heifers, Diets, and Housing ................................ ................................ ............ 130 Experimental Design and Treatments ................................ ............................ 131 Ultrasonography of Ovaries ................................ ................................ ............ 132 Blood Sampling and Analysis of Progester one in Plasma .............................. 132 Evaluation of Ovulation and Progesterone at AI ................................ ............. 132 Pregnancy Diagnoses and Evaluation of P/AI and Pregnancy Loss .............. 133 Experiment 2 ................................ ................................ ................................ .. 133 Heifers, Diets and Housing ................................ ................................ ............ 133 Experimental Design and Treatments ................................ ............................ 134 Pregnancy Diagnoses and Evaluation of Pregnancy Outcomes .................... 134 Statistical Analys i s ................................ ................................ .......................... 134 Results ................................ ................................ ................................ .................. 136 Experiment 1 ................................ ................................ ................................ .. 136 Experiment 2 ................................ ................................ ................................ .. 136 Discussion ................................ ................................ ................................ ............ 137 Conclusion ................................ ................................ ................................ ............ 141 5 HORMONAL MANIPULATIONS IN THE 5 D TIMED AI PROTOCOL TO OPTIMIZE ESTROUS CY CLE SYNCHRONY AND FERTILITY IN DAIRY HEIFERS ................................ ................................ ................................ .............. 148 Introductory Remarks ................................ ................................ ............................ 149 Materials and Methods ................................ ................................ .......................... 153 Study 1 ................................ ................................ ................................ ........... 153 Heifers, Diets, and Housing ................................ ................................ ............ 153 Experimental Design and Treatments ................................ ............................ 154 Pregnancy Diagnosis and Calculation of Reproductive Outcomes ................. 154 Study 2 ................................ ................................ ................................ ........... 155 Heif ers, Diets, and Housing ................................ ................................ ............ 155 Experimental Design and Treatments ................................ ............................ 155 Ultrasonography of Ovaries and Evaluation of Ovulatory Response s ............ 156 Blood Sampling and Analysis of Progesterone Concentrations ...................... 156 Pregnancy Diagnosis and Calculation of Reproductive Outcomes ................. 157 Study 3 ................................ ................................ ................................ ........... 157 Experimental Design and Treatments ................................ ............................ 157 Blood Sampli ng and Analyses of LH and Progesterone Concentrations ........ 158 Statistical Analys i s ................................ ................................ .......................... 159 Results ................................ ................................ ................................ .................. 161 Study 1 ................................ ................................ ................................ ........... 161 Study 2 ................................ ................................ ................................ ........... 162 Study 3 ................................ ................................ ................................ ........... 1 63 Dis cussion ................................ ................................ ................................ ............ 164 Conclusion ................................ ................................ ................................ ............ 167

PAGE 9

9 6 EFFECTS OF ONE OR TWO TREATMENTS WITH PROSTAGLANDIN F ON SUBCLINICAL ENDOMETRITIS AND FERTILITY IN LACTATING DAIRY COWS INSEMINATED BY TIMED AI ................................ ................................ ... 177 Introductory Remarks ................................ ................................ ............................ 178 Materials and Methods ................................ ................................ .......................... 181 Animals, Housing, and Feeding ................................ ................................ ...... 181 Reproductive Management ................................ ................................ ............ 182 Treatments and Body Condition Scori ng ................................ ........................ 182 Evaluation of Uterine Health ................................ ................................ ........... 182 Statistical Analys i s ................................ ................................ .......................... 183 Re sults ................................ ................................ ................................ .................. 186 Effects of PGF Treatments on the Prevalence of Subclinical Endometritis .. 186 Effects of PGF on Pregnancy per AI and Pregnancy Loss .......................... 187 Associ ations Among PVD and/or Subclinical Endometritis with Measures of Fertility ................................ ................................ ................................ ......... 187 Discussion ................................ ................................ ................................ ............ 188 Conclusions ................................ ................................ ................................ .......... 194 7 EFFECTS OF INTRAUTER INE INFUSION OF TRUE PERELLA PYOGENES ON ENDOMETRIAL m RNA EXPRESSION OF GE NES ASSOCIATED WITH LUTEOLYSIS AND CORPU S LUTEUM LIFESPAN IN DAIRY COWS ................ 200 Introductory Remarks ................................ ................................ ............................ 201 Materials and Methods ................................ ................................ .......................... 203 Animals, Housing and Diets ................................ ................................ .......... 203 Study Design and Treatments ................................ ................................ ........ 204 Experiment 1 ................................ ................................ ................................ .. 206 Uterine Tissue Sample for Biopsy ................................ ................................ .. 206 RNA Extraction and Quantitative, Real Time, Reverse Transcription PCR .... 207 Experiement 2 ................................ ................................ ................................ 208 Blood Sampling, Analysis of Progesterone and PGFM Concentrations and Ultrasound Scanning ................................ ................................ ................... 209 Statistical Analys i s ................................ ................................ .......................... 210 Results ................................ ................................ ................................ .................. 211 Experiment 1 ................................ ................................ ................................ .. 211 Experiment 2 ................................ ................................ ................................ .. 212 Discussion ................................ ................................ ................................ ............ 213 Conclusion ................................ ................................ ................................ ............ 218 8 EFFICACY OF AMPICILL IN TRIHYDRATE FOR TR EATMENT OF METRITIS AND SUBSEQUENT FERTI LITY IN DAIRY COWS ................................ ............. 232 Introductory Remarks ................................ ................................ ............................ 233 Materials and Methods ................................ ................................ .......................... 235 Cows, Housing, and Diets ................................ ................................ .............. 235 Experimental Design, Treatments and Body Condition Scor ing .................... 236

PAGE 10

10 Rectal temperatures, Vaginal Discharge and Cure Definitions ....................... 237 Evaluation of Purulent Vaginal Discharge, Subclinical Endometritis and Estrous Cyclicity ................................ ................................ .......................... 238 Reproductive Management ................................ ................................ ............ 239 Statistical Analysis ................................ ................................ .......................... 240 Results ................................ ................................ ................................ .................. 241 Rectal Temperatures and Incidence of Fever ................................ ................. 242 Metritis Cure Based on Vaginal Discharge Score < 5 ................................ ..... 242 Metritis Cure Based on Vaginal Discharge Score < 5 and Rectal Tem perature < 39.5 o C ................................ ................................ ................ 243 Metritis Cure Based on Vaginal Discharge Score < 5, Rectal Temperature < 39.5 o C, and no Additional Antimicrobial Therapy ................................ ....... 244 Purulent Vaginal Discharge and Subclinical Endometritis .............................. 244 Estrous Cyclicity, Pregnancy per AI and Pregnancy Loss .............................. 245 Discussion ................................ ................................ ................................ ............ 245 Conclusion ................................ ................................ ................................ ............ 250 9 CONCLUSION S AND FUTURE DIRECTION S ................................ .................... 261 LIST OF REFERENCES ................................ ................................ ............................. 265 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 301

PAGE 11

11 L IST OF TABLES Table page 2 1 Characteristics of the estrous cycle of heifers ................................ .................... 38 3 1 Effect of number of timed AI before exposure to natural service on reproduction of dairy cows not observed for detection of estrus ....................... 123 3 2 Factors affecting the hazard of pregnancy of dairy cows not observed for detection of estrus and subjected to 1 or 3 timed AI before exposure to natural service ................................ ................................ ................................ .. 124 4 1 Ovarian responses of heifers treated with or without GnRH at the initiation of the 5 d timed AI protocol ................................ ................................ .................. 143 4 2 Effect of the first GnRH injection o f the 5 d timed AI protocol on fertility responses of dairy heifers Experiment 1 ................................ ........................ 144 4 3 Effect of time of administration of the final GnRH of the 5 d timed AI protocol relative to inseminati on on fertility responses of dairy heifers Experiment 2 .. 145 5 1 Effect of the initial GnRH and two doses of PGF on fertility responses of dairy heifers subjected to the 5 d timed AI program (Study 1) .......................... 169 5 2 Effect of the initial GnRH and two doses of PGF on ovarian responses in dairy heifers subj ected to the 5 d timed AI program (Study 2) .......................... 170 5 3 Effect of the initial GnRH and two doses of PGF on fertility responses in dairy heifers subjected to the 5 d timed AI program (Study 2) .......................... 171 5 4 Effects of concentration of progesterone in plasma on LH release and ovulation in response to GnRH in dairy heifers (Study 3) ................................ 172 6 1 Effect of one or two treatments of PGF on pregnancy per AI and pregnancy loss of dairy cows subjected to a timed AI program ................................ .......... 195 6 2 Association between purulent vaginal discharge (PVD) and/or subclinical endometritis (SCE) with f ertility of dairy cows at first postpartum insemination 196 6 3 Association between subclinical endometritis (SCE) with fertility of dairy cows at first postpartum insemination postpartum inse mination ................................ 197 7 1 Primer reference and sequences for genes investigated by quantitative real time PCR ................................ ................................ ................................ .......... 219

PAGE 12

12 8 1 Effect of antibiotic treatment on mean rectal temperature (RT) and incidence of fever in cows with only metritis or cows with puerperal metritis (metritis and o C) ................................ ................................ ................................ .... 251 8 2 Effect of treatment on estrous c yclicity, pregnancy per AI and pregnancy loss following the 1st postpartum insemination. ................................ ....................... 252

PAGE 13

13 L IST OF FIGURES Figure page 3 1 Diagram of hormonal tr eatments and insemination for the double Ovsynch program used to synchronize ovulation for first insemination in all cows at the first service and timeline of reproductive activities for 1 and 3 timed AI treatments. ................................ ................................ ................................ ........ 125 3 2 Kaplan Meyer survival curves for proportion of nonpregnant cows according to treatment. ................................ ................................ ................................ ..... 126 4 1 Diagram of activities in experiment 1. ................................ ............................... 146 4 2 Diagram of activities in experiment 2. ................................ ............................... 147 5 1 Diagram of activities in study 1. ................................ ................................ ........ 173 5 2 Diagram of activities in study 2. ................................ ................................ ........ 174 5 3 Diagram of activities in study 3. ................................ ................................ ........ 175 5 4 Effect of progesterone concentration on the LH release in response to GnRH. 176 6 1 Diagram of treatments according to d in milk ( 3). ................................ .......... 198 6 2 Effect of one or two treatmen ts of PGF on the prevalence of subclinical endometritis on d 32 and 46 postpartum in dairy cows. ................................ .... 199 7 1 Diagram of experimetal activities for experiments 1 and 2.. ............................. 220 7 2 time. ................................ ................................ ................................ .................. 221 7 3 Relative endometrial mRNA expression of IL6 according to tre atments and time. ................................ ................................ ................................ .................. 222 7 4 Relative endometrial mRNA expression of IL8 according to treatments and time. ................................ ................................ ................................ .................. 223 7 5 Relative endometri al mRNA expre ssion of TNF according to treatments and time. ................................ ................................ ................................ .................. 224 7 6 Relative endometrial mRNA expression of PGES according to treatments and time. ................................ ................................ ................................ ........... 225 7 7 Relative endometrial mRNA expression of PGFS according to treatments and time. ................................ ................................ ................................ ........... 226

PAGE 14

14 7 8 Relative endometrial mRNA expression of OXR according to treatments and time. ................................ ................................ ................................ .................. 227 7 9 Uterine tissue pictures stained with hematoxylin and eosin. ............................. 228 7 10 Concentrations of progesterone in plasma according to da y of the estrous cycle. ................................ ................................ ................................ ................ 229 7 11 Size of the corpus luteum accord ing to day of the estrous cycle. ..................... 230 7 12 Concentrations of prost aglandin F metabolite (PGFM) in plasma according to day of the estrous cycle. ................................ ................................ ................... 231 8 1 Diagram of treatments for metritis, monitoring of cure, and evaluation of uterine health. ................................ ................................ ................................ ... 253 8 2 Diagram of reproductive program used for first insemination and pregnancy diagnosis. ................................ ................................ ................................ ......... 254 8 3 Rectal temperature on d 2, 3, 4, 5, 6, 7 and 1 2 after the diagnosis of metritis according to treatment. ................................ ................................ ..................... 255 8 4 Adjusted proportions ( SEM) of cows with metritis cured on d 5, 7 and 12 after initiation of treatments with ampicillin or cef tiofur. ................................ .... 256 8 5 Adjusted proportions ( SEM) of cows with metritis cured on d 5, 7 and 12 after initiation of treatments according to initial diagnosis of metritis only or puerperal metritis. ................................ ................................ ............................. 257 8 6 Adjusted proportions ( SEM) of cows with metritis cured on d 5, 7 and 12 after initiation of treatments according to parity. ................................ ............... 258 8 7 Adjusted proportions ( SEM) of cows with metritis cured on d 5, 7 and 12 after initiation of treatments according to diagnosis of calving related disorders (dystocia, stillbirth, twins, and retained placenta). ............................. 259 8 8 Adjusted proportions ( SEM) of cows with purulent vaginal discharge on d 32 3 postpartum (panel A) or with cytological subclinical endometris on d 39 3 postpartum (panel B). ................................ ................................ ............ 260

PAGE 15

15 LIST OF ABBREVIATIONS HSD 3 beta AI artificial insemination AMH anti Mullerian hormone A ng a ngiotensin A NOVA analysis of variance A HR adjusted hazard ratio AOR adjusted odds ratio ATP adenos ine tri phosphate BCS body condition score BHBA hydroxybuty rate BMP bone morphogenetic protein C a calcium cbp collagen binding protein CARTPT cocaine and amphetamine regulated transcript C AMP cyclic adenosine monophosphate CD cluster differentiation CCL2 0 chemokine C C motif ligand 20 CIDR control internal drug release CL corpus luteum COS72 cosynch timed artificial insemination protocol with 72h of proestrus CV coefficient of variation CXCL1 chemokine C X C motif ligand 1 DAG 1,2 diacylglycerol DIM days in milk

PAGE 16

16 DMI dry matter intake EDTA ethylenediaminetetraacetic acid E ND endot helin E RK extracellular signal regulated kinase FGF fibroblast growth factor fim t ype I fimbrae FSH follicle stimulating hormone fyu ferric yersiniabactin uptake GDF growth diff erentiation factor GH growth hormone G n RH g ona dotropin releas ing hormone IFN interferon tau IGF insulin like growth factor IGFBP insulin like growth factor binding protein IL interleukin ISG interferon stimulant genes LH luteinizing hormone LPS lipopolysa ccharide MAPK mitogen activated protein kinase M D2 myeloid differentiation factor 2 MMP matrix metalloproteinase m RNA messenger ribonucleic acid MYD88 myeloid differentiation factor 88 NEFA nonesterified fatty acid nuclear factor kappa B

PAGE 17

17 NO nitric ox ide OVS56 Ovsynch time artificial insemination program with 56h of proestrus P/AI pregnancy per artificial insemination PAPP A plasma associated pregnancy protein A PG prostaglandin PKA protein kinase A PKC protein kinase C PLC phospholipase C plo pyolysin PVD purulent vaginal discharge RIA radioimmunoassay S t AR steroidogenic acute regulatory protein SCE subclinical endometritis TGF transforming growth factor TIMP tissue inhibitor metalloproteinase TLR4 toll like receptor 4 TMR total mixed ration TNF tumor necrosis factor US United States VEGF vascular endothelial growth factor VF virulence factors

PAGE 18

18 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MANIPULATING OVARIAN FUNCTION AND UTERINE HEALTH WITH THE AIM OF IMPROVING FERTILITY IN DAIRY CATTLE By Fbio Soares de Lima August 2013 Chair: Jos Eduardo Portela Santos Major: Animal Sciences The objectives of this dissertation were to d etermine reproductive management strategies and hormonal manipulations of ovarian responses to opti mize fertility of dairy cattle and to investigate the efficacy of treatments of uterine diseases and its subsequent effects on fertility o f dairy cows. An ad ditional objective was to determine mechanisms of induced infusion with. T. pyogenes on length of the estrous cycle. C hapter 3 describes effects of 1 (1TAI) or 3 ti med AI (3TAI) before natural service in lactating dairy cows not detected in estrus on pregn ancy and revealed that 3TAI had greater hazard of pregnancy and 21 day cycle pregnancy rate than 1TAI. Chapter 4 evaluated the effects of the first gonadotropin releasing hormone ( GnRH ) injection of the 5 d timed AI program on ova rian responses and pregn ancy per AI (P/AI) and the effect of timing of the final GnRH to induce ovulation relative to AI on P/AI in dairy heifers Use of GnRH on the 1 st day of the 5 d timed AI program resulted in low ovulation and no improvement in P/AI when a single prostagland in (PG) F injection on d 5 was used Extend ing proestrus by delaying the final GnRH from 56 to 72 h concurrent with AI benefited P/AI of heifers not in estrus at AI

PAGE 19

19 C hapter 5 investigated the effects of GnRH at the initiation of the 5 d timed AI progra m combined with two injections of PGF on ovarian responses and P/AI in dairy heifers, and the role of progesterone on LH release and ovulation in response to GnRH. I ncreased ovulation with GnRH given at initiation of 5 d timed AI program combined with tw o doses of PGF on d 5 and 6 optimize d P/AI High progesterone at GnRH administration suppressed LH release and impaired ovulation. Chapter 6 investigated the efficacy of PGF as a therapy to reduce the prevalence of subclinical endometritis and to incre ase P/AI in cows subjected to a timed AI program. The results of this study revealed that treatments with PGF before initiation of the timed AI program were un able to improve uterine health and P/AI. Chapter 7 investigated if intrauterine infusion of T. pyogenes influence uter ine expression of genes related to inflam mation, luteolytic cascade and luteal lifespan. Although T. pyogenes reduced luteal lifespan in half of the cows only minor changes in uterine gene expression were identified. Chapter 8 inve stigated the efficacy of ampicillin trihydrated for treatment of metritis and subsequent effect s on fertility. Ampicillin was an efficacious therapy for metritis but had no differential effects o n P/AI when compared with ceftiofur hydrochloride

PAGE 20

20 CHAPTE R 1 INTRODUCTION The evolution of the dairy industry in the United States with d evelopment and implementation of genetic selection of dairy cattle refinement s in housing, physiological and nutritional aspects and coordinated management strategies improve d productivity per animal and overall efficiency in dairy operation s However the historical trend for fertility of lactating dairy cows followed a different path with worsening reproductive responses and a negligible genetic selection for reproductive an d heath traits (Lucy et al., 2001; Norman et al., 2009). Some issues related with this decline in reproductive performance include aspects of c ow physiology, reproductive management and animal health and actions in these areas are expected to reverse the decline and improve cow fertility ( Lucy et al. 2001 Royal et al., 2008). One of the most important advancements made by reproductive physiologist s was the development of timed AI allowing the insemination of cows without the need for detection of estrus with acceptable fertility ( Pursley et al., 1996; Schmitt et al. 1996 ). The first study presented in this dissertation discusses how strategic incorp oration of timed AI to breeding program s using natural service impact s reproductive performance. The studie s included in C hapters 4 and 5 were conducted to understand mechanism s underlining reproductive physiology of dairy heifers and how to improve hormonal manip ulation of the estrous cycle to maximize fertility when 5 d timed AI program is used The studies presented in C hapter s 6 and 8 contain a n additional focus of this dissertation with evaluation of the efficacy of therapies for subclinical endometritis and metritis with the aim of mitigating the negative impact of these uterine dis eases on fertility. Fin ally C hapter 7 of the dissertation discuss es the

PAGE 21

21 effects of intrauterine infusion of T. pyogenes on uterine gene expression of inflammatory mediators luteolytic factors and luteal lifespan. Inadequate or inaccurate detection of estrus are critical factor s responsible for poor reproductive performance in dairy cows (Senger, 1994; Roel ofsa et al., 2010). Timed AI and natural service (NS) are common methods to manage reproduction in dairy herds in the United States (Champagne et al. 2002, NAHMS, 2002; Smith et al. 2004; De Vries et al., 2005; Caraviello et al., 2006; Lima et al., 2009) and can be used successfully without detection of estrus (Lima et al., 2009). Although AI hastens genetic progress, controls venereal diseases, and provides a safer environment for cows and farm personnel, breeding programs relying primarily on NS are still widely used by dairy producers. Several studies and surveys conducted in different regions of the US report ed that 43% to 84% of the dairy farms use NS either alone or combin ed with AI in the United States (Champagne et al. 2002, NAHMS, 2002; Smith et al. 2004; De Vries et al., 2005; Caraviello et al., 2006). In many cases, NS is incorporated into breeding programs after cows have received one or more AI, which is commonly nam ed as a detection of estrus and still incorporate AI, continuous synchronization of ovulation for insemination at fixed time is an option (Lima et al., 2009). Timed AI has been shown to be an economical option to manage reproduction in high producing dairy cows that experience a reduction in estrous intensity (Lima et al., 2010; Risco et al., 1998). The reproductive performance (Lima et al., 2009) and costs (Lima et al., 2010) of cows bred by timed AI or NS have been compared directly and, although cows exposed to NS had similar 21 d cycle pregnancy rate compared with cows receiving only timed

PAGE 22

22 AI (Lima et al., 2009), the economic evaluation favored those receiving AI (Lima et al., 2010). The economic advantage of timed AI was even greater when genetic progress was considered, and when marginal and feed cost and increased. Interestingly, the 21 d cycle pregnancy rate, a common metric used to evaluate reproduction in dai ry herds, was similar between cows bred by timed AI or NS (25.0 and 25.7%, respectively), and they were both superior when compared with average values for high producing dairy herds that ranges from 15.0 to 17.9% (De Vries et al., 2005; LeBlanc, 2010). Th e improvement in hazard of pregnancy for NS was attributed to a greater number of breeding opportunities, as NS cows were exposed continuously to bulls. In contrast, timed AI cows could only be inseminated after diagnosed nonpregnant, which created inter A I intervals of 35 d (Lima et al., 2009). It is possible that a combination of timed AI and NS might benefit reproductive performance of dairy cows not observed for detection of estrus, as timed AI allows for all cows to be inseminated on the first day past the voluntary waiting period and exposure to NS after that will likely shorten the interval between breedings. In many dairy farms using a combination of AI and NS, cows initially are inseminated one or more times and then moved to bull breeding groups ( Overton and Sischo, 2005); however, it is unclear how many inseminations cows should receive before exposed to bulls to maximize pregnancy rate. This is particularly important in herds managing reproduction without the aid of estrous detection, as the inte rval between inseminations is determined by when a cow can be resynchronized for AI. In a previous study, the benefit of NS over TAI was only observed after 150 d postpartum, when timed AI cows had already received three inseminations (Lima et al., 2009).

PAGE 23

23 Therefore, it is plausible to suggest that three timed AI may result in a similar reproductive performance when compared with one timed AI, despite the long inter insemination interval. In fact, Overton and Sischo (2005) concluded that in herds using both AI and NS, allowing cows more opportunities for AI may benef it reproduction and that was the goal of experiment 3 The use of timed AI programs in dairy heifers is low compared with that for lactating dairy cows (NAHMS, 2009). Programs to synchronize ovul ation of dairy heifers based on GnRH and PGF resulted in low pregnancy per AI (P/AI) compared with insemination performed after detection of estrus (Schmitt et al., 1996, Pursley et al., 1997 and Rivera et al., 2004). The depressed P/AI for most timed AI programs based on GnRH and PGF and the perception by dairy producers that heifers become pregnant easily without the need for intervention explains the low use of ovulation synchronization protocols for management of reproduction in heifers. Recently, a 5 d timed AI protocol investigated by Rabaglino et al. (2010a) resulted in P/AI ranging from 52.2 to 61 .0 % in dairy heifers in the first two inseminations, which resembled the reproductive performance obtained when heifers are artificially inseminated af ter detect ed estrus (Kuhn et al., 2006). In fact, additional work by the same investigators evaluating anti luteolytic strategies with 325 heifers synchronized with the 5 d timed AI program observed P/AI of 59.5% on d 45 after insemination (Rabaglino et al ., 2010b). Therefore, it is possible to achieve acceptable P/AI in dairy heifers following synchronized ovulation with the 5 d timed AI protocol. The 5 d timed AI program is comprised of an injection of GnRH and insertion of a controlled internal drug rel ease (CIDR) intravaginal device containing progesterone,

PAGE 24

24 followed 5 d later by CIDR removal and an injection of PGF and AI concurrent with a second GnRH injection 72 h after PGF (Rabaglino et al., 2010a). Only 23% of the heifers had multiple corpora lutea (CL) 5 d after the injection of the GnRH (Rabaglino et al., 2010a), suggesting that ovulation to the initial G nRH was probably low. In fact, heifers receiving a single injection of PGF 5 d after GnRH had similar luteolysis and P/AI to those receiving two injections given 12 h apart (Rabaglino et al., 2010a). The same was not true when lactating dairy cows were s ubjected to a similar program with a 5 d interval between GnRH and PGF (Santos et al., 2010 a ). Therefore, the low incidence of ovulation induced by the first GnRH combined with more rapid turnover of follicles in heifers (Sirois and Fortune, 1988) might result in little benefit from the initial GnRH in the 5 d ti med AI program in dairy heifers and the need for the initial injection of GnHR was exp lored in study one of C hapter 4. In the Ovsynch protocol a ltering the timing of the final GnRH to induce ovula tion relative to AI influences P/AI in lactating dairy cows. Brusveen et al. (2008) reported that GnRH administered 56 h after PGF increased P/AI compared with GnRH given concurrent with timed AI at 72 h. In a series of experiments with beef cows subjected to the 5 d timed AI program, extending the proestrus from 60 to 72 h was beneficial to fertility (Bridges et al., 2008). In dai ry cows subjected to the 5 d timed AI program, P/AI did not differ when the final GnRH was administered either 16 h before or concurrent with AI at 72 h after PGF (Bisinotto et al., 2010). Although inducing ovulation 16 h before AI benefits fertility of dairy cows in the standard 7 d timed AI Ovsynch program, it is unclear if a similar benefit would occur in dairy heifers when follicle dominance is redu ced such as in the 5 d timed AI protocol. Therefore, E xperiment 2 of C hapter 4 was

PAGE 25

25 de signed to investiga te which would be the ideal interval between GnRH injection to induce ovulation and insemination. Reproductive efficiency in dairy heifers directly affects age at first calving and therefore, has major impacts o n rearing costs and subsequent productive l ife (Gabler et al., 2000; Ettema and Santos, 2004). Most dairy operations in the United States use AI after observed estrus to manage reproduction in heifers (NAHMS, 2009), which requires daily labor and adequate detection of estrus. Nevertheless, advances in protocols for synchronization of the estrous cycle have supported the use of timed AI as an alternative method to control reproduction in dairy heifers and improve economics when detection of estrus is less than ideal (Ribeiro et al. 2012a). Recent stu dies have consistently reported P/AI ranging from 50 to 60% in dairy heifers subjected to the 5 d timed AI program (Rabaglino et al., 2010; Lima et al., 2011). These results are comparable to those observed in heifers inseminated at detected estrus (Kuhn e t al., 2006). Further optimization of such programs to either simplify or result in improved fertility will likely increas e acceptance by dairy producers. Despite the acceptable P/AI in heifers subjected to the 5 d timed AI protocol, important physiologica l aspects of ovarian responses and respective impacts on fertility have not been elucidated completely. Ovulation in response to the initial GnRH injection in timed AI programs enhances synchrony of estrous cycle, shortens follicle dominance, and improves embryo quality and P/AI (Vasconcelos et al., 1999; Chebel et al., 2006; Cerri et al., 2009a). Nevertheless, only 15 to 35% of heifers ovulate when treated with GnRH at random stages of the estrous cycle (Stevenson et al., 2008; Lima et al., 2011). In addit ion, heifers that ovulate in response to the initial GnRH will hav e a newly formed

PAGE 26

26 CL 5 d later, which is generally refractory to a single treatment with PGF (Rowson et al., 1972; Henricks et al., 1974). Eliminating the first GnRH in the 5 d timed AI pro gram reduced ovulation at the beginning of the synchronization protocol, but increased the proportion of heifers that underwent luteolysis at AI when a single PGF injection was used (Lima et al. 2011). Because the benefits associated with follicle turnov er were offset by a less effective CL regression, P/AI did not differ between heifers that received or not GnRH at the initiation of the timed AI program (Lima et al. 2011). These results indicate that the initial GnRH is not necessary when a single PGF is used, which simplifies and reduce costs associated with the synchronization protocol. H eifers that receive GnRH in the beginning of 5 d timed AI program may require multiple PGF injection s to optimize luteolysis and P/AI. Results from lactating dairy cows subjected to the 5 d timed AI program indicate that the use of two injections of PGF administered 24 h apart improved CL regression and P/AI (Santos et al., 2010 a ), particularly when ovulation to initial GnRH is high and more cows present a newly fo rmed CL (Ribeiro et al., 2012b). Shorter intervals between PGF treatments, ranging from 7 to 12 h, have been shown to increase P/AI compared with a single injection in beef cows (Kasimanickam et al., 2009), although preliminary results in dairy heifers d id not confirm such a benefit (Rabaglino et al., 2010). Therefore, in E xperiment s 1 and 2 of C hapter 5 we hyp o th e sized that the combination of the initial GnRH and the administration of PGF o n d 5 and 6 of the protocol would improve follicle turnover and luteal regression, which is expected to result in increased P/AI in dairy heifers. Presumably because of less catabolism of steroid hormones in heifers than lactating dairy cows as a result of differences in splanchnic blood flow (Sangsritavong et

PAGE 27

27 al., 2 002), progesterone concentrations in the plasma of dairy heifers are nearly 1.5 ng/mL greater than those in lactating cows during mid diestrus (Sartori et al., 2004). Because of reduced catabolism, the increase in progesterone concentrations with a CIDR is expected to be gre ater in nonlactating cows (Zuluaga and Williams, 2008) than in lactating cows (Cerri et al., 2009b). Progesterone affects LH secretion, which might compromise ovulatory response to GnRH treatment, and partially explain the low ovulatory response to GnRH in dairy heifers. In fact, results from beef heifers support the idea that progesterone compromise s LH release and impair s ovulation following an injection of GnRH E xperim ent 3 of C hapter 5 explored the concept that LH release might be hi ndered by high plasmatic concentration of progesterone (Colazo et al., 2008; Dias et al., 2010). Another focus of the dissertation was uterine health. U terine diseases are prevalent in dairy cows and they are associated with reduced reproductive performanc e, which ultimately affects herd profitability (Gilbert et al., 2005; LeBlanc, 2008). Uterine diseases are often classified according to clinical presentation and defined based on their impacts on P/AI or time to pregnancy (Sheldon et al., 2006). Among the m, clinical endometritis is defined as presence of inflammation in the reproductive tract visible by the type of vaginal discharge that typically contains pus and persists after 21 DIM (Leblanc et al., 2002a; Sheldon et al., 2006). More recently, clinical endometritis as diagnosed by presence of pus in the vagina was classified as purulent vaginal discharge (PVD) because of the large proportion of cows without concurrent neutrophil infiltration in the endometrium (Dubuc et al., 2010). On the other hand, a l arge proportion of cows not diagnosed with any clinical signs of uterine disease have

PAGE 28

28 presence of inflammatory cells in the endometrium, and usually more than 5% PMNL in endometrial cytology reduce s P/AI and extends the interval postpartum to pregnancy (Gi lbert et al., 2005; Galvo et al., 2009a). In the United States, no particular treatment is labeled for use in cows that have either PVD or subclinical endometritis, although use of intrauterine infusion of 500 mg of cephapirin as benzathine has demonstra ted efficacy in reducing interval to pregnancy in cows wit h PVD (Leblanc et al., 2002b) and in improving pregnancy at first AI in cows with subclinical endometritis (Kasimanickam et al., 2005). In those studies, cows were not subjected to standardized time d AI programs for first postpartum AI and many were inseminated following detection of estrus. When cows were subjected to a presynchronized timed AI program, use of intrauterine antibiotics did not benefit P/AI of dairy cows (Galvo et al., 2009b), even i n those with previous diagnosis of PVD. An alternative therapy is the use of PG F in an attempt to induce estrus and eliminate bacterial contamination that might be causing the inflammatory process in the endometrium. Use of PGF in cows during diestrus results in luteolysis and induces cows to return to estrus, which has been suggested to enhance uterine immunity by removal of immunosuppressive effects of progesterone (Lewis, 2004). Kasimanickam et al. (2005) suggested that the improvements in P/AI cause d by PGF in postpartum cows were caused by inducing estrus and concurrent opening of the cervix and myometrium contractions that might enhance mechanical cleansing of the endometrium. When PGF is administered in early lactation, it is possible that the benefi ts to fertility might not be related to enhancing uterine health, but confounded with effects of presynchronizing the estrous cycle before timed AI programs (Moreira et al., 2001;

PAGE 29

29 Galvo et al., 2007). It is known that the stage of the estrous cycle when c ows initiate timed AI protocols based on GnRH is critical for fertility (Vasconcelos et al., 1999), and treatment with PGF 11 to 12 d before the initiation of the timed AI increased P/AI (Moreira et al., 2001; Galvo et al., 2007). In fact, in most studies evaluating PGF as therapy for treatment of uterine diseases and subsequent impacts on fertility, uterine health was n ot evaluated after treatment to justify the increase in P/AI (Leblanc et al., 2002b; Kasimanickam et al., 2005). In some cases, when uterine health was evaluated after PGF treatment, P/AI at first AI improved, but the benefits were not linked to a reduct ion in the prevalence of subclinical endometritis in treated cows (Galvo et al., 2009a). Timed AI programs are used commonly for reproductive management of dairy herds for first and resynchronized inseminations to mitigate the negative impacts of poor est rous detection in lactating dairy cows (Caraviello et al., 2006). An alternative presynchronization treatment, in which stage of the estrous cycle is synchronized in cyclic and anovular cows, is called Double Ovsynch (Souza et al., 2008). When PGF is administered before the Double Ovsynch protocol, the effects on uterine health or measures of fertility are not expected to be mediated by altering the stage of the estrous cycle when cows are subjected to the timed AI protocol. The goal of the study in C hapter 6 was to demonstrate an improvement in P/AI in dairy cows with the systematic use of PGF by enhancing uterine health based on the reduction in the prevalence of subclinical endometritis. Uterine diseases affects nearly half of the dairy cows after parturition leading to disruption of uterine and ovarian function which frequently results in hindered fertility,

PAGE 30

30 culling and remarkable economic losses for dairy producers (Sheldon et al., 2009). The economic losses caused by metritis alone are stri king and calculated to be $380 per affected cow due to reduced milk production, delayed conception, treatment and increased culling (Dri l lich et al., 2001). Thus, if we consider a conservative incidence rate of 20% for the 8.5 million dairy cows in US the annual cost of metritis alone is $650 million. E ndometritis is another manifestation of uterine diseases with remarkable detrimental effects on fertility ( Dubuc et al., 2011 ). Therefore, understanding the mechanism by which microbes subvert host innate im munity to disrupt ovarian and uterine function is fundamental to develop preventatives to mitigate the negative impacts of uterine diseases. Trueperella pyogenes is considered one of the most relevant pathogens involved in uterine diseases, especially end ometritis. This is due to its relative high prevalence in the environment, persistence in the uterus, severity of lesions on the endometrium, resistance to treatment, and synergistic action with gram negative anaerobes (Ruder et al., 1981; Huszenicza et al ., 1999; Mateus et al., 2002a, b; Williams et al., 2005,). However, the mechanism by which T. pyogenes affects the endometrium and reproductive events in dairy cows such as length of the estrous cycle and concentration of ovarian steroids remain elusive (W illiams et al., 2007; Kaneko and Kawakami, 2009; Kaneko et al., 2013). Several studies reported that intrauterine infusion of live T. pyogenes disrupts luteal development leading to early demise of the CL and ovulation of a first wave dominant follicle (K aneko and Kawakami, 2007; Kaneko and Kawakami, 2007; Kaneko et al., 2013). Cows receiving an intrauterine infusion of T. pyogenes on d 3 after

PAGE 31

31 ovulation had a peak of prostaglandin F metabolite (PGFM) 3 d later, followed by the regression of the newly form ed CL and ovulation of the dominant follicle of the first follicular wave in approximately 50% of the time (Kaneko and Kawakami, 2008; Kaneko and Kawakami, 2009; Kaneko et al., 2013). The explanation of how T. pyogenes induces short estrous cycle to disrup t normal ovarian and luteal function is unclear. Culture of endometrial cells with bacteria free filtrate of T. pyogenes induce d synthesis of PGF (Miller et al., 2007). This bacterium possesses a number of virulence factors that may contribute to its pathogenic potential. One of the most important is a cholesterol dependent cy tolysin, pyolysin, which is a h emolysin cytolytic for macrophages (Jost and Bilington, 2005). A second important virulence factor is peptidoglycan, which is a pathogen associated molecular pattern molecule that induces pro inflammatory cytokines such as tumor necrosis factor al., 1993; Stewart et al., 2003; Bromfield and Sheldon, 2011), which can stimulate endometrial synthesis of PGF (Davidson et al., 1995; Hansen et al., 2004; Skarzynski et al., 2000). However, a possible stimulation of inflammatory mediators and their dir ect relationship s with luteolytic cascade factors has never been investigated with an in vivo model of intrauterine induced infection with T. pyogenes Therefore, the possible molecular mechanisms by which intrauterine infusion with live T. pyogenes lead t o shortening of luteal phase in dairy cows remains to be elucidated The hypothesis investigated in C hapter 7 is that intrauterine inoculation of live T. pyogenes in cows with newly formed CL would increase endometrial expression

PAGE 32

32 of genes affecting the inf lammation and the luteolytic cascade leading to an acute endometrial production of PGF and early demise of the newly formed CL. Metritis is a prevalent postpartum disease in lactating dairy cows characterized by abnormally enlarged uterus and a fetid, w atery red brown fluid to viscous off white purulent uterine discharge that can be accompanied or not with fever within 21 days postpartum, but more frequently diagnosed in the first week postpartum (Sheldon et al., 2006). The incidence rates of dairy cows developing metritis ranges from 10 to 36% (Goshen and Shpigel, 2006; Santos et al., 2010 b ; Chapinal et al., 2011). The economic losses caused by metritis are striking ranging from to $328 to $380 per affected cow due to reduced milk production, delayed con ception, cost of treatment and increased culling and death ( Drillich et al., 2001; Overton and Fetrow, 2008). Additionally, cows diagnosed with metritis have an increased risk to develop both clinical and subclinical endometritis (Galvo et al., 2009; Mart inez et al., 2012). The main bacteria isolated from cases of uterine infection include Escherichia coli T pyogenes f ormerly known as Arcanobacterium pyogenes and anaerobic bacteria such as Prevotella spp., formerly known as Bacteroides spp. and Fusoba cterium necrophorum (Griffin et al., 1974; Noakes et al., 1989; Sheldon et al., 2002). Recently, the expression of some specific virulence factors by these bacteria wer e associated with increase d risk s for development of uterine diseases (Bicalho et al., 2 012). Escherichia coli expressing the adhesin type I fimbriae fimH identified in the uterus of cows in the first 3 days postpartum was associated significantly with development of metritis and endometritis. Fusobacterium necrophorum expressing the luekoto xin/hemolysin lktA in the first 3 days or between days 8 and 12 postpartum was associated with endometritis.

PAGE 33

33 Trueperella pyogenes expressing the type I fimbriae adhesin fimA and the pyolysin plo between 8 and 10 d or between 34 and 36 d postpartum was asso ciated with endometritis (Bicalho et al., 2012). Therefore, it has been suggested that the presence of E. coli expressing virulence factor fimH in the uterus of cows in the first few days postpartum paves the way for the other bacterial infection s coordina ting the initial process of tissue damage and development of uterine diseases. Thus, it is reasonable to suggest that a reduction on the extent of E. coli load in the uterus of metritic cows might mitigate the negative impact of the disease and minimize th e risk of subsequent chronic uterine infections such as clinical and subclinical endometritis. Ampicillin is a beta lactam antibiotic that acts as an irreversible inhibitor of dd transpeptidase, an essential enzyme that bacteria use to make their cell wall s. Therefore, ampicillin generally inhibits the third and final stage of bacterial cell wall synthesis in binary fission, which ultimately leads to bacterial cell lysis. Ampicillin has received FDA approval for its mechanism of action and it has been shown effective against E. coli (Burrows, 1993; Lehtolainen et al., 2003). However, to date, no published study has evaluated efficacy of ampicillin treatment of metritis in dairy cows. Thus, it was hypothesized in C hapter 8 that ampicillin would be an effectiv e therapy for metritis having si milar efficacy t o ceftiofur, one of the major antibiotics used to treat metri tis in the United States and the objectives were to evaluate the efficacy of ampicillin trihydrate for treatment of metritis in dairy cows compared with ceftiofur hydrochloride and subsequent effect on P/AI to the first service.

PAGE 34

34 CHAPTER 2 LITERATURE REVIEW Estrous C ycle in Dairy Cattle The estr o us cycle in dairy cattle is characterized by a rhythmic pattern of cyclic ovarian activity initiated with attainment of puberty by heifers with first ovulation enabling sexual receptivity and repeated opportunities for mating, and establishment of pregnancy. The activities within the estrous cycle in dairy cattle are controlled mostly by the interplay among ov arian steroid hormones (progesterone and estradiol), hormones of the hypothalamus (gonadot rop in releasing hormone; GnRH), the anterior pituitary (follicle stimulating hormone; FSH and luteinizing hormone; LH), a nd the uterus (prostaglandin F ; PGF ), which through a mechanism of negative and positive feedback, modulate ovarian activity (Roche, 1996). The normal length of the estr o us cycle in dairy cattle ranges from 18 to 24 days ( Peter et al., 2009 ). The average inter ovulatory period was calculated based on series of studies as 20.8 days for dairy heifers and 23.0 days for dairy cows (Sartori et al., 2004). C hanges in ovarian dynamics are dictated to a large extent by patterns of release and concen trations of regulatory hormones. T he bovi ne estrous cycle is divided in to two phases the follicular and luteal phases ; and four periods proestrus, estrus, metestrus and diestrus (Peter et al., 2009). The f ollicular phase encompasses the proestrus and estrus periods going from regression of the corp us luteum (CL) to ovulation with a shift from progesterone to estradiol dominance lasting 4 to 6 days. The proestrus phase is the period in which structural and functional CL regression initiates and a substantial increase in the synthesis of estradiol by pre ovulatory dominant follicles occur s leading to an onset of sexual receptivity that defines inception of estrus (Wattemann et al.,

PAGE 35

35 1972). The period of e strus comprises the initiation of sexual receptivity and ovulation of the dominant follicle. Fro m the day of occurrence of luteolysis to the day preceding ovulation the maximal serum estradiol concentration s increases about 3 fold (~2.5 pg/mL to 7.9 p g/mL) and the maximal serum estradiol concentration preceding ovulation is greater for dairy hei fers than for dairy cows (11.3 pg/mL vs. 7.9 p g/mL, Sartori et al., 2004). The concurrent increase in estradiol concentration s in blood and reduced concentrations of progesterone leads to stimulation of arcuate nucleus, ventromedial nucleus, and the preoptic ar ea of hypothalamus allowing the display of behavioral estrus (Frandson et al., 2003, Molenda Figueira et al., 2006). The duration of estrus in modern dairy cows have been reported to be approximately 7 hours with 9.1 standing ev ents or mounts recorded dur ing standing estrus (Drasnfield et al., 1998; Lopez et al., 2004, Diskin et al., 2008). A f ew decades earlier the duration of estrus was twice as long with 14.9 hours (Esslemont and Bryant, 1976). Milk production, the major trait used for genetic selection of dairy cattle, ha s been negatively associated with duration of estrus. Cows producing 55 kg/day had 2.8 hours of estrus duration, whereas cows producing 25 kg/day had 14.7 hours of estrus duration. The duration of estrus duration in heifers range d from 1 2 to 14 hours (Diskin et al., 2008). A recent study reported heritability estimates for estrus duration and intensity to be low (2% to 8%; Lovendahl and Chagunda, 2009), but genetic control of estr o us behavior remain s elusive. It has been suggested that ch anges in the underlying molecular mechanisms that regulate estrous behavior could be manifested by altered gene expression patterns, although very little is currently known in dairy cattle (Boer et al., 2012) Duration of estrus is likely to also be influe nced by environmental factor other

PAGE 36

36 than milk production. Cow footing, heat stress, and lameness all influence the ability of cows to display and maintain normal duration of estrous expression (Lucy, 2001) The luteal phase begins with ovulation and ends wit h CL regression comprising the metestrus and diestrus periods with duration of 14 to 18 days. The period of metestrus initiates after ovulation and typically lasts 3 to 4 days being characterized by the formation of the CL from the collapsed ovulated folli cle (corpus haemorragicum) and metestrus bleeding in some cows. After ovulation, progesterone concentrations begin to increase because of the formation of the CL with luteinization of granulosa and theca cells forming large and small luteal cells respecti vely. Increased progesterone concentrations in plasma prepare the uter us for the establishment and maintenance of pregnancy or in case of lack of pregnancy establishment re occurrence of the estrous cycle (Peter et al., 2009). Diestrus is the period i n wh ich the CL matures, becomes responsive to PGF 2 and sustains secretion of progesterone, ending with its demise during luteolysis. Transition from metestrus to diestrus is marked by a sharp increase of progesterone on day 4 after ovulation (Stabenfeldt et al., 1969). During diestrus, progesterone conc entrations in the plasma increase more than 3 fold (1.5 to 5.5 ng/mL) from day s 4 to 8 after ovulation. Subsequently, the rate of increase is reduced and reaches the maximal value on day 16 of estrous cycle at a pproximately 7 ng/mL (Stabenfeldt et al., 196 9). Dairy heifers have greater maximal serum progesterone concentration than dairy cows (7.3 ng/mL vs. 5.6 ng/mL, Sartori et al., 2004) and this difference has been attributed to the increase hepatic catabolism of steroids caused by increased hepatic cata bolism of ovarian steroids as splanchnic blood flow is markedly increased by increased dry matter and caloric intake in dairy cows (Sangsritavong et al.,

PAGE 37

37 2002). During the diestrus progesterone concentrations remain elevated and recurrent follicular waves occur due to periodic release of FSH from the anterior pituitary gland. Nevertheless, the dominant follicles that develop throughout the luteal phase do not ovulate because progesterone induces a negative feedback on LH, thus only allowing the secretion o f greater amplitude but lesser frequency LH pulses that are not sufficient for ovulation and a block of the preovulatory surge of LH (Rahe et al., 1980). If pregnancy does not occur in dairy cattle, pulsatile release of PGF by the endometrium will lead t o spontaneous luteolysis between days 17 and 19 of the estrous cycle ( Ginther et al., 2010 ). The development of ultrasonograph y and its application to bovine reproduction lead to major advance ments in t he understanding and characterization of ovarian dynam ics during the estrous cycle. I t is well established that dairy cattle have between 2 to 3 waves of follicular development within each estrous cycle with only the last wave of the cycle leading to ovulation ( Sirois and Fortune, 1988; Sartori et al., 2004 ). In dairy heifers occurrence of 3 foll icular waves ranges from 33 to 84 % of the cycles (Savio et al., 1988; Sirois and Fortune, 1988; Sartori et al., 2004) A pproximately 50% of the estrous cycles are 2 wave cycles and 50% are 3 wave cycles ( Table 2 1). On the other hand, 79% of cycles of high producing dairy cows have 2 follicular wave s (Sartori et al., 2004). Each wave of follicular development consists of emergence, selection and dominance followed by either atresia or ovulation of the dominant follicle.

PAGE 38

38 Table 2 1. Characteristics of the estrous cycle of heifers Follicle wave Emergence of 2 nd wave Estrous cycle length 2 wave 3 wave 2 wave 3 wave 2 wave 3 wave Progesterone peak Reference % (n) Day of the cycle Day ng/mL Beef heifers Savi o et al. (1988) 16 (4) 84 (21) 10 ~10 20.5 1.3 21.3 1.5 N/A Bong et al. (1993) 25 (3) 75 (9) N/A N/A 20.8 0.9 21.3 0.5 N/A Dairy heifers Sirois and Fortune (1988) 20 (2) 80 (8) 11 9.4 0.5 20 1 20.70.4 ~7.5 Ginther et al. (1989) 82 ( 18) 18 (4) 9.6 0.2 9.0 0.0 20.4 0.3 22.8 0.6 N/A Knopt et al. (1989) 90 (9) 10 (1) 10 0.4 10 21 23 N/A Ko et al. (1991) 75 (9) 25 (3) ~10 ~8 ~20 ~23 N/A Kulick et al. (2001) 57 (13) 43 (10) ~10 ~9 ~19.5 ~23.0 N/A Sartori et al. (2004) 56 (15 ) 44 (12) 9.7 0.2 8.2 0.4 20.7 0.3 23.1 0.7 7.3 0.4 Wolfenson et al. (2004) 70 (14) 30 (6) ~10 ~10 ~22 ~22 ~10 Overall 54 (87) 46 (74) 10.1 0.2 9.3 0.3 20.6 0.3 22.2 0.3 8.3 0.9 N/A = not applicable

PAGE 39

39 The beginning of gonadotropin depe ndent follicle development is characterized associated with a transient increase in FSH concentrations peaking at 0.7 days before follicle emergence (Ginther et al ., 1989; Adams et al., 1992; Sunderland et al., 1994). Although follicle emergence has been describe d as when the cohort of follicles attain 4 mm in diameter detailed ultrasonographic examination demonstrated that even smaller follicles ranging from 1 to 3 mm develop in wave pattern associated with FSH surge (Jaiswal et al., 2004). The dominant follicle emerges in humans from the cohort of recruited follicles based initially on its morphology as the largest developing follicle (Gougeon and Lefevre, 1983 ; G inther et al., 1996 ). This increase in size is paralleled by an increase in follicular fluid estradiol and inhibin concentrations, which suppresses FSH concentrations from the anterior pituitary gland via negative feedback (Sunderland et al., 1994; Ginther et al., 2000). The next event in follicular dynamics is called deviation. This is characterized by the differentiation in growth rates between the upcoming dominant follicle a nd its largest subordinate follicle which in dairy cattle occurs when the large st follicle achieves approximately 8.5 mm of diameter and the second largest follicle is approximately 7.2 mm of diameter (Ginther et al., 1996). The hallmarks of deviation are inability of subordinate follicle to develop if the dominant follicle is ablate d and nadir concentrations of FSH (Adams et al., 1992; Mihm and Evans; 2008). Before deviation, the elimination of the largest follicle leads to development of the second largest as the dominant follicle (Ginther et al., 1997). Moreover, follicular deviati on has been associated temporally with an increase in the expression of LH receptors in granulosa cells and in LH responsiveness in some studies

PAGE 40

40 (Sartori et al., 2001; Mihm et al., 2006; Nogueira et al., 2007), but not in all studies (Evans and Fortune, 19 97). In fact, treatment of cows with an ovulatory dose of LH will only cause ovulation in follicles that have grown past the time of deviation (Sartori et al., 2001), suggesting that the development of LH responsiveness associated with deviation is require d for ovulatory capacity. After deviation, dominance is established and the dominant follicle becomes LH dependent and it can either develop and ovulate or enter in atresia if progesterone concentrations are high. Therefore, only second or third follicular wave dominant follicles undergo ovulation in cows having normal estrous cycle and no exogenous hormonal manipulation. The emergence of the second follicle wave occurs at around day s 9 to 11 in cows with a two wave pattern, and around day s 8 or 9 in those a with three wave pattern, whereas the third follicle wave arises around day 15 to 16 of estr ous cycle (Adams et al., 2008). F ollicle Development and Function A pre determined number of primordial follicles are established during fetal development with ova rian follicle growth taking a period of approximately 3 to 5 months with distinct gonadotropin independent and dependent stages (Webb et al., 2004). Gonadotropin dependent follicle growth in cattle occurs in wave pattern with occurrence of 2 to 3 follicula r waves in each estrous cycle (Savio et al., 1988). Temporal histological evaluation of the bovine ovary revealed 6 distinct types of structural and developmental follicles (Braw Tal and Yossefi, 1997). Folli cles were classified in primordial, transitory, primary, small pre antral, large pre antral and antral follicles. Primordial follicles are composed of the oocyte surrounded single layer of flattened granulosa cells generally Transitory follicles resemble the primordial fo llicles with

PAGE 41

41 the only difference being the granulosa cells are cuboidal and multiplying. Primary follicles have the oocyte bordered for 1 or 1.5 layer of cuboidal granulosa cells with a size ranging from 40 t antral follicles have 2 to 3 layer s of cuboidal gr anulosa cells around the oocyte and ra n ges in size It is the earliest follicle type to present the zona pel l ucida. Large pre antral follicles have 4 to 6 layers of cuboidal granulosa cells surrounding the oocyte with size ranging from 131 to l ucida that for the first time forms a complete ring around the oocyte. Lastly, the antral follicle characterized by the oocyte bordered by more than 6 layers of cuboidal granulosa cells with a size great presence of an antrum. Cattle are born with a finite number of primordial follicles (Fortune et al., 2010) H igh variation in ovarian size, ovarian reserve, an d number of follicles recruited in each follicular wave in the bovine is reported to be positive ly associated with fertility (Ireland et al., 2011). The interval from the activation of primordial follicles to the format ion of pre ovulatory follicle have been estimated to be approximately 180 days with 138 days spent in the p r e antral stages and the remaining 42 days in the antral stages (Lussier et al., 1987). The delay between the appearance of the first primordial and the first primary follicles is 50 days in cattle and is associated with progression through to meiotic prop hase I and arrest at diplotene (McNatty et al., 2007). Although there have been many advances on the understanding about the molecular and biochemical factors regulating follicle development, the pivotal factors con trolling the activation of primordial fo llicles remains elusive. Primordial follicles have transforming growth factor (TGF) other growth factor ligands and receptors

PAGE 42

42 expressed in oocytes and granulosa cells. Activin like kinase (ALK) 3, 5 and 6, betaglycan, bone mor p hogenic protein (BM P) 6, bone morphogenic protein receptor (BMPR) II, growth differentiation factor (GDF) 9, connexin 37, kit ligand c kit, and primordial follicles that pote ntially are part of the interplay regulating activation of primordial follicles (McNatty et al., 2007). It is likely that several of these growth promoting factors are involved, both negatively and positively, with the initiation of follicular growth. The results of in vi tro studies using the cortical portion of the bovine ovary revealed that the majority of primordial follicles are activated within a few days of isolation from the ovary, suggesting that inhibitory factors may play a key role on regulation of exit from the pool of primordial follicles. It has been suggested that a paracrine communication with the ovarian stroma is required to prevent activation of primordial follicles and control the pace of follicular development (Fortune et al., 1998). Addi tionally, Fort une (2003) reported that anti ovaries. It is important to notice that activation of primordial follicles is not dependent on gonadotropic stimulus (Braw Tal and Yoss efi, 1997; Fortune et al., 1998). Once follicles transition from the pool of primordial follicles, they are still committed to a gonadotropin independent growth. As the follicle grows to the primary stage, the granulosa cells increase in number and change shape, becoming uniformly cuboidal. The oocyte also enlarges, with 3 to 10 fold increases in the volume of smooth endoplasmic reticulum, mitochondria, ribosomes and lipid droplets, and the zona pellucida, absent in primordial follicles, is deposited (Lun dy et al., 1999).

PAGE 43

43 Analysis of expressed sequence tags from primary follicles shows that several hundred genes not found in primordial follicles are activated in primary follicles, including some involved in mitochondrial function, cell signaling and commu nication, and the synthesis of the zona pellucida. With the advance to the primary stage, the follicle becomes responsive to progestins, estrogens, androgens, and gonadotropins; however, gonadotropin dependent regulation is not observed until the antral fo l licle stage. In addition, several factors related to regulation of the follicular growth begin to be synthesized by the oocyte (BMP 6, BMP 15, GDF 9), granulosa (inhibin, follistatin), and theca cells (TGF The current view is that GDF9 and BMP 15 secreted by the oocyte act in a concentration dependent paracrine manner on adjacent cumulus and granulosa cells. Furthermore, reduced concentrations of BMP15 and GDF9 alter only slightly the responsiveness of granulosa cells to gonadotropins, whereas the absence of BMP15 or GDF9 arrest follicular growth at primordial or primary stages (McNatty et al., 2007). Interaction between GDF9 and BMP 15 produced by the oocyte induces cell proliferation and controls responsiveness to gonadotropins in a spatial dependent manner because cells that diverge to become the cumulus and mural granulosa cells are affected differently (McNatty et al., 2007). GDF9 and BMP15 are produced as pre pro proteins consisting of a signal peptide, a lar ge pro r egion and a mature region (Shimasaki et al., 2004), and they signal to granulosa and cumulus cells through the TGF receptor BMP RII (Juengel and McNatty 2005). When GDF9 binds to BMP RII, ALK5 is activated and phosphorylates SMAD2/3. Thereafter, S MAD2/3 associates with the common SMAD4 and t his complex translocates to the nucleus where it interacts with

PAGE 44

44 specific DNA motifs and transcriptional regulators leading to the transcription of target genes. When BMP15 binds to BMP RII, it causes recruitment and activation of ALK6, leading to signaling through the alternative BMP pathway mediated by SMADs 1, 5 and 8 (Juengel and McNatty 2005). Although FSH receptors are present in granulosa cells since the primordial follicles stage in many species (Cortvrindt et al., 1997; Findlay and Drummond, 1999), the transition from primary to small antral follicles seems to remain independent of gonadotropins (Knight and Glister, 2001), but influenced by many factors such as: testosterone, fibroblast growth factor (FGF) 7 and vascular endothelium growth factor (V EGF) that direct ly or indirectly influ ence granulosa cells and oocyte development (Fortune et al., 2010; Buratini and Price, 2011). Once follicles transit from pre antral to antral stage they become gonadotropin dependent. At this stage follicles undergo cyclic recruitment and develop either to the pre ovulatory state or undergo atresia. The transition from pre antral to antral stages is characterized by a reduction in somatic cell proliferation and improved differentiation into cells responsive to gonadot ropins that can produce increased amounts of steroids (Webb and Campbell, 2007). The gonadotropins FSH and LH secreted by gonadotroph cells in the anterior pituitary act on granulosa cells and theca cells dictating the fate of the antral follicles. The on set of antral follicle growth is stimulated by FSH that binds to granulosa cells and initiates the recruitment of a new follicular wave. Luteinizing hormone binds to granulosa and theca cells inducing the development and final maturation of the selected do minant

PAGE 45

45 antral follicle. The two gonadotropins act in concert promoting steroidogenesis in follicular somatic cells The secretion of FSH and LH is regulated by GnRH and the release of GnRH is modulated by ovarian steroids. It has been reported that GnRH n eurons have no receptors for progesterone and estradiol (Skinner et al., 2001; Herbison and Pape, 2001) ; therefore steroid modulation of GnRH secretion occurs indirectly through intermediary neurons that possess the relevant steroid receptors and respond to their stimulation (Clarke and Pompolo, 2005). Neurons with receptors to ovarian steroids and communication with GnRH neurons are classified as stimulatory and inhibitory neurons. Kisspeptin, dopamine and glutamine are neurotransmitters that stimulate r elease of GnRH while aminobutyric acid, nitric oxide and opioids are neurotransmitters that inhibit GnRH secretion (Clarke and Pompolo, 2005). Kisspeptin neu rons express ER progesterone receptor and, therefore, have the potential to relay feedback ef fects on the GnRH neuron. Evidence now suggests that reduced activity of kisspeptin neurons in the arcuate nucleus of sheep is responsible for translating estrogen negative feedback to GnRH neurons. Ovariectomized sheep have an increased level of KiSS 1 mR NA in the neurons compared to controls. Also, if estrogen replacement is given to ovariectomized female sheep, then KiSS 1 mRNA levels are reduced to control levels (Roa and Tena Sempere, 2007). This suggests that steroids are negatively regulating KiSS 1 mRNA in the arcuate nucleus, hence reducing stimulation of GnRH neurons. The interplay between ovarian steroids and GnRH is defined as a bimodal system. Thus, GnRH is released as either tonic mode (low frequency and low amplitude pulses) at luteal phase or surge mode (high frequency and high amplitude) at the follicular phase.

PAGE 46

46 The intra ovarian regulation of both gonadotropin responsive and gonadotropin dependent follicles has been largely attributed to the IGF system. Studies have shown that IGF I increa ses the sensitivity of small follicles (5 mm in cattle) to gonadotropin stimulation and simulates their transition from the gonadotropin responsive to the gonadotropin dependent stages (Mazerbourg et al., 2003). In ovarian tissue in vitro IGF I stimulated steroidogenesis by thecal cells and both proliferation and differentiation of granulosa cells (Mazerbourg et al., 2003). The secretion of estradiol by granulosa cells in culture was also stimulated by IGF I (Campbell et al., 1996). Therefore, it seems tha t IGF I can stimulate either the proliferation or the differentiation and differentiated functions of granulosa cells, depending on the stage of development of the follicle. The IGF binding protein (IGFBPs) are important modulator of the IGF system. The su pply of IGF I to the follicle is outside the control of the reproductive axis, therefore intra follicular IGF activity is regulated locally, primarily by the IGFBPs. The low molecular weight IGFBPs (BP 2, 4 and 5) are inhibitory to IGF actions because th ey bind IGF, preventing it from binding to its receptor. Therefore, what dictates changes during folliculogenesis is the bioavailability of IGF I and not the total concentration. In the cow, intrafollicular concentrations of IGFBP 2 and IGFBP 4 decrease as follicles grow to pre ovulatory size and of IGFBP 5 increase in follicles, as they become atretic (Mazerbourg et al. 2003). The pregnancy associated plasma protein A (PAPP A) is a pivotal factor modulating the concentration of IGFBPs. The low concentrat ions of IGFBPs in healthy growing follicles are caused by increased rates of proteolytic degradation of IGFBP 2, 4 and 5 by PAPP A and by low rates of gene transcription for IGFBP 2 (Mazerbourg et

PAGE 47

47 al., 2003). In bovine granulosa cells the expression of the mRNA for PAPP A is the highest in ovulatory follicles and is positively correlated with aromatase and LH receptor (Mazerbourg et al., 2001). Furthermore, PAPP A activity was increased in dominant follicles in comparison with subordinate follicles on da ys 2 and 3 of the first follicular wave of the estrous cycle and the levels of activity correlated positively with estradiol and negatively with low molecular weight IGFBPs in follicular fluid (Rivera et al. 2001). Cattle were treated with low doses of re combinant bovine FSH for 2 days shortly after wave emergence, two co dominant follicles were selected, both with higher PAPP A activities and estradiol concentrations and lower amounts of IGFBP 4 in their follicular fluid than subordinate follicles (Rivera and Fortune 2001). These results suggest that the action of FSH on dominant follicles is to increase PAPP A activity. Careful analyses of various characteristics of follicles just before and after follicle emergence showed that an increase in PAPP A acti vity in one follicle of the wave was detected before any detectable difference in diameter or in the concentrations of estradiol or IGFBP 4 or 5 in follicular fluid (Rivera and Fortune 2003). Recently, a new postulated theory suggested that acquisition o f LH receptors on granulosa cells occurs prior to changes in PAPP A and the subsequent increased bioavailability of IGF 1 (Luo et al., 2011). Dominant follicles expressed LH receptors in the granulosa cells of 12 h before follicle deviation, but an increas e in PAPP A expression was only observed at follicle deviation (Luo et al., 2011). Authors speculated that LH receptors in the granulosa cells w ere induced by LH secretion itself. Additionally, LH signaling also increased expression of steroidogenic enzyme s in theca (steroidogenic acute regulatory protein StAR, cholesterol side chain cleavage

PAGE 48

48 P450scc) and granulosa cells (a romatase P450arom). Before deviation, LH receptors are only expressed in theca cells through which LH stimulates expression of ste roidogenic enzymes essential for production of androgens that are then used by granulosa cells to produce estradiol. After deviation, expression of LH receptors in the granulosa cells allow s the follicle to respond to LH and undergo further development (Lu o et al., 2011). Acquisition of LH receptors by the granulosa cells is the hallmark for acquisition of ovulatory capacity in response to LH surge, which it is supported by occurrence of an inducible ovulation when using exogenous GnRH or LH only after foll icle deviation. Another important factor recently reported to be involved with health status of follicles is the cocaine and amphetamine regulated transcript (CARTPT) and its peptide CART (Lv et al., 2009). Results revealed that granulosa cell CARPT mRNA expression and follicular fluid CART concentration are higher in estrogen inactive atretic follicles than in estrogen active healthy follicles collected at pre deviation stage and immediately after selection (early dominance stage) and relatively low in th e remaining stages of the follicular wave (Lv et al., 2009). Additionally, CART negatively regulates FSH and IGF 1 actions on granulosa cells in vitro, reducing CYP19A1 expression and inhibit s estradiol production in vivo (Lv et al., 2009). Moreover, CART concentrations in healthy follicles decrease after dominance (Lv et al., 2009). Follicular dominance is largely acknowledged as a n evolutionary mechanism to control number of offspring per pregnancy in mono cotous species. Interestingly, in poly c o t ous spec ies such as pig and mice that have a less strict control of follicle dominance, CARTPT is not expressed in their ovaries suggesting that CART could be a

PAGE 49

49 pivotal functional modulator of selection of a single dominant follicle in monocotous species (Smith et al., 2010). Throughout the diestrus stage of the estrus cycle cattle will have dominant follicle s that will not undergo ovulation because progesterone block s pulsatile LH release. Once the CL regresses and progesterone concentrations fall LH secretion r ises and the dominant follicle is capable of undergo ing ovulation. However, if subluteal or low concentrations of progesterone are present, dominant follicles may continue to develop in size and remain dominant longer without occurrence of ovulation due to progesterone blocks of the LH surge (Mihm et al., 1994). Regression or ovulation of the dominant follicle lead to reduced concentrations of estradiol and inhibin allowing plasmatic concentrations of FSH to increase inducing the emergence of a new follicul ar wave. Corpus Luteum Formation, Function and Regression The CL is an ephemeral endocrine gland that throughout its lifespan undergoes a period of extremely rapid growth that involves hypertrophy, proliferation and differentiation of the steroidogenic ce lls, as well as extensive angiogenesis. The major function of the CL is to produce progesterone that is essential for the maintenance of pregnancy. The CL lifespan in normal estrous cycle non pregnant cows is approximately 17 days (Ginther et al 2010, Sart ori et al., 2004). After ovulation, the remaining cells in the ruptured follicle undergo a developmental phase within 8 to 10 days that is marked initially by a period of tissue remodeling with intense sprouting of endothelial cells up to the first third of cycle, with the mature CL being characterized by a dense network of vessels critical to support the differentiation of follicular cells. The alterations toward the development of a functional

PAGE 50

50 CL are named luteinization and involve cellular structural c hanges, such as an increase in the cytoplasmic to nuclear area ratio and an accumulation of lipid droplets (Brnnstrm and Fridn, 1997). After achieving maturity the bovine CL is composed of vascular endothelial cells (50%), large and small steroidogenic lut eal cells (30%) and the remaining 20% of cells are constituted by pericytes, fibrocytes, nerves, immune and smooth muscle cells t al., 1989; Lei et al., 1991). The population of immune cells in the CL is diverse with presen ce of macrophages, neutrophils, eosinophils, CD4 + T lymphocytes, and CD8 + T lymphocytes that play important role on CL formation, protection and demise (Penne et a l., 1999; Bauer et al., 2001). The mature CL engages in an unmatched production of steroids, resulting in extremely high metabolic activity within the tissue, with the greatest blood flow per unit of tissues, and oxygen consumption on a cell basis estimated to be 2 to 6 times greater than that for the liver, kidney, or heart (Niswender et al., 200 0; Robinson et al., 2008). Once CL formation is initiated, one of the most important event s that take place is the shift from estradiol to progesterone production by large and small steroidogenic luteal cells that are formed from granulosa and theca cells respectively. For small luteal cells LH is the major stimulator of progesterone production through increase of cAMP tha t activates protein kinase A (PKA) leading to up regulation of P450scc, StAR, and 3 beta HSD). Large luteal cells are unresponsive to LH relying on constitutive activated PKA and perhaps other luteotropic horm ones such as growth hormone (GH), IGF, FGF and oxytocin to stimu l ate synthesis of progesterone (Miyamoto et al., 2010).

PAGE 51

51 Addition ally, it seems that progesterone self stimulates its own synthesis through progesterone nuclear receptors a nd through non genomic pathways, which additionally protects luteal cells from apoptosis (Okuda et al., 2004; Liszewska et al., 2005, Rekawiecki et a l., 2008). During the peri ovulatory period, follicular steroidogenesis by theca and granulosa is dramatically down regulated, but the transport of cholesterol across the outer mitochondrial membrane remains unaltered during the late stages of development of the ovulatory follicle based on StAR mRNA expression (Nimz et al., 2009). Granulos HSD, P450arom and theca cells HSD in the theca cells were down regulated within 18 hours preceding the pre ov ulatory surge of LH (Voss and Fortune, 1993a; Voss and Fortune, 1993b). Consequently, androstenedione, testosterone and estradiol concentrations in the follicular fluid were reduced between 4 and 10 hours after the pre ovulatory surge of LH (Komar et al., 2001). Accordingly, estradiol concentrations declined 50% within the firs t 5 hours of the pre ovulatory surge of LH, and returned to basal values within the next 9 hours (C henault et al., 1974). After 72 hours of culture, HSD, but not P450c17 and P450arom, was reestablished in follicular cells (Voss and Fortune, 1993a; Voss and Fortune, 1993b). These results are consistent with increasing pro duction of progesterone in vitro by granulosa and theca cells (i.e.: luteinization) between 24 and 72 hours of culture not associated with an increase in androgen production (Voss and Fortune, 1993a; Voss and Fortune, 1993b). The development of the CL is d ependent largely on the vascular system, which has two main types of functional luteal blood vessels: arteriolovenous vessels (ie,

PAGE 52

52 muscle layer; and capillary vessels, which do not have a s mooth muscle layer. In the bovine CL, the number of arteriolovenous and microcapillary vessels drastically increases from the early to midl uteal phase (Bauer et al., 2003; Hojo et al., 2009; Shirasuna et al., 2010). Angiogenesis in the bovine CL is control led mostly by vascular endothelial growth factor A (VEGFA) and FGF 2 and its respective receptors VEGFR 2 and FGF receptor (Ferrara et al., 1997; Connolly, 1991). The expression s of VEGFA, VEGFR 2, FGF2 and FGF receptor are up regulated in early luteal phas e, but remarkably down regulated during midluteal to the regression phase s in the bovine (Schams et al., 1994; Berisha et al., 2000). Robinson et al. (2008) using a system of luteal cell culture to stimulate angiogenesis induced endothelial cell network fo rmation with FGF2 and VEGFA at physiological dose of 1 ng/mL. Furthermore, addition of VEGFR2 inhibitor to this culture system reduced endothelial cell networks formation by 60%. W hen a FGFR1 inhibitor was added to the medium even at the presence of VEGFA endothelial cell network formation was reduced by 90% These results suggest that FGF2 is more important than VEGFA for the formation of luteal vascular networks (Woad et al., 2009). The effects of VEGFA and FGF2 o n the bovine CL to influence progestero ne production go beyond angiogenesis. Studies using microdialysis system revealed VEGFA and FGF2 stimulate progesterone secretion in the bovine CL (Miyamoto et al., 1992; Kobayashi et al., 2001). Moreover, Yamashita et al. (2008) reported that the injectio n of specific antibodies for VEGFA or FGF2 antibodies directly into the cow CL reduced progest erone secretion, CL volume, and down regulated luteal mRNA

PAGE 53

53 expression of VEGFA, FGF2 and up regulated expression of angiopoietins confirming the pivotal importanc e of VEGFA and FGF2 on CL development. The immune system is also critically involved with CL development. In cows, the developing CL has an increased number of macrophages and monocytes (Penny et al., 1999). Macrophages are the most abundant immune cells i n the bovine ovary deriving from differentiation of monocytes arriving at the ovaries as a response to local cytokin es such as granulocyte macrophage colony stimulating factor, macrophage colony mma (Zhao et al., 1995; Townson et al., 2003; Zhang et al., 2008). Macrophage elimination was associate d positively with increased amounts of ovarian hemorrhage, which lead to a remarkable depletion of endothelial cells and an increase in the number of ery throcytes in the bovine ovary (Turner et al., 2011). Removal of macrophages prompted the disruption of the critical pericyte endothelium interaction, compromising dramatically endothelial function (Turner et al., 2011). It seems that macrophages play a cr itical role in maintaining the integrity of ovarian vasculature. Polymorphonuclear leukocytes (PMNL) such as eosinophils and neutrophils are detected in the CL during the estrous cycle. Eosinophils infiltrate into the CL soon after occurrence of ovulation in the bovine CL (Reibiger and Spanel Borowski, 20 00). Studies using human ovaries suggested that P selectin expressed on endothelial cells are responsible to recruit eosinophils into the developing CL (Aust et al., 2000). A large number of neutrophils an d elevated concentrat ion of the neutrophil specific chemokine interleukin 8 (IL 8) are present in the CL of cows during early luteal phase (Jiemtaweeboon et al., 2011). The formation of early CL lead to PMNL migration in vitro

PAGE 54

54 when IL 8 and the supernatant of activated PMNL were used; and IL 8 stimulated the formation of capillary like structures of CL derived endothelial cells suggesting that PMNL and IL 8 may be involved in regulation of angiogenesis in the developing CL (Jiemtaweeboon et al., 2011). Th e process of CL maintenance when preg nancy occurs involves endocrine vascular and immunological factors that orchestrate a response critical to the fate of the CL. The abrogation of luteolysis involves blocking of the pulsatile release of PGF in the uterus by molecules secreted by developing conceptus. A body of evidence gathered throughout the last few decades suggest that th the primary embryonic signal in ruminants (Bazer et al., 2008) involved with matern al recognition of the pregnancy by suppressing pulsatile uterine release of PGF The mechanism of suppressing the pulsatile release of PGF is mediated by the su ppression of the expression of estrogen receptor (OXR) mRNA i occurs on day 17 of gestation at the same time of maternal recognition of pregnancy (Bazer et al. 1997). Interferon tau mediates its effects in bovine endometrium by binding to endometrial type I IFN receptors (Li and Roberts 1994) activating the janus kinase signal transducer and activator of transcription (JAK STAT) pathway (Binelli et al. 2001). endometrium of cattl e or sheep, such as IFN stimulated gene 15 (ISG15), 20 50 oligoadenylate synthetase bovine ubiquitin activating E1 like enzyme members of the 1 8 family mixovirys resitance protein 1 (MX1), granulocyte macrophage colony stimulating factor 1, IFN regulat ory factors 1 and 2 and signal transducer and activator

PAGE 55

55 of transcriptions 1 and 2 (Thatcher et al. 2001, Wolf et al. 2003). The conceptus secretes more than one type I IFN (Cochet et al., 2009) and a number of interferon responsive genes are upregulated on the endometrium of cows in response to IFNs, which may induce other important factors related to endometrial function. Mor eover, responsive genes in circulating immune cells, which could possibly be involved with rescuing the CL for maintenance of pregnancy (Gifford et al., 2007; Ott and Gifford 2010). Inter feron stimulated gene 15 and MX1 are up regula ted in CL of pregnant cows (Yang et al., released into the uterine vein and may itself exert endocrine effects on the CL (Spencer et al., 1999; Hansen et al., 201 0 ). Yang et al. (2010) was unable to identify IF induced stimulation of ISG15 in cultured bovine luteal cells, suggesting that the effects o n the CL may be mediated by immune or endothelial cells not present in the culture media. However, previou s the stimulated prostaglandin production and expression of class II major histocompatibility molecules in cultured luteal steroidogenic cells, i ndicating that these cells can respond to type I IFN (Pate, 1995). The am ount of blood vessels remain s unchanged from late luteal phase to early pregnancy, but number of pericytes and smooth muscle cells undergo es reduction in CL of early pregnant cows (Beindorff et al., 2010). Furthermore, in early pr egnancy VEGFA luteal expression remain unchanged and the vasculature becomes stable with the decreased angiopoietin 2/angiopoietin 1 mRNA abundance ratio (an index of instability of vasculature) in the CL compared with the CL in the late luteal phase of a nonpregnant

PAGE 56

56 cycle (Beindorff et al., 2010). Another system apparently involved with CL maintenance is the lymphatic system. Recent studies demonstrated up regulated expression of lymphatic endothelial hya luronan receptor 1, a lymph vessel marker, in luteal tissues at d 40 of pregnancy (Nita et al., 2011). This study also show ed that lymphatic endothelial in vitro (Nitta et al., 2011). Moreover, lymphangiogenic factors such as VEGFC and lymphatic endothelial hyaluronan receptor 1 were also up regulated luteal tissue at d 16 of pregnancy (Nitta et al., 2011). Taken together, these results suggest that perhaps it is the lymphatic system, and not the blood vascular system of the bovine CL, that is rebuilt during early pregnancy. During early pregnancy the CL is more resi lient to the luteolytic effects of PGF than the CL on the same day of the normal estrous cycle (Pratt et al., 1977; Silvia and Niswender 1986). This suggests that perhaps intraluteal factors can influence what occurs with in the bovine CL. The sensitivity of the early developing CL ( fewer th an 5 days) also has been exploited. Tsai and Wiltbank (1998) suggested that the early CL, in spite of presenting functional PGF receptors, is unable to stimulat ing intra luteal synthesis of PGF via prostaglandin endoperoxidase synthase 2, increasing ex pression of monocyte chemoatractant protein 1, and inhibiting progesterone production through StAR Miyamoto et al. (2009) suggested a site restricted action of PGF depending on the stage of the estrous cycle. In the mid cycle CL (d 8 to 12), PGF induc es an acute increase in blood flow in the periphery of the CL concurrent with expression of endothelial nitric oxide synthase, but the same phenomenon is not observed in the early cycle CL (d 4). Moreover, Atli et al. (2012) reported that although the init ial pulse

PAGE 57

57 of PGF upregulates mRNA expression of many pathways related to luteolysis, the second and later pulses of PGF are actually responsible for a distinct pattern of gene expression that result in luteolysis. L uteolysis is a phenomenon coordinated by increased pu lsatile release of PGF in the bovine uterus, which reaches the ovaries through a countercurrent exchange mechanism between the utero ovarian vein and ovarian artery (Hixon and Hansel, 1974). The synthesis of PGF occurs mostly in epithelial endometrial cells that have the pool of phospholipids replenished by exposure to progesterone during diestrus. The exposure to progesterone during the diestrus cycle downregulates its own receptor in the reproductive tract and hypothalamus and upregulates estrogen and oxytocin receptors previously unresponsive to progesterone (Spencer and Bazer, 1995; Wathes et al., 1996, McCracken et al., 1999). Simultaneously, estradiol production by the dominant follicle of the last follicular wave stimulates pulsatile release of ox ytocin that binds its receptor in the endometrial cells and activate s phospholipase A 2 (PLA 2 ). The synthesis of PGF in endometrial cells occurs through hydrolysis of ester bonds between phospholipids and arachidonic acid by PLA 2 The released arachidonic acid is converted by prostaglandin G/H synthase to PGH 2; and then PGH 2 is converted to PGF through the action of prostaglandin F synthase. The blood vascular system of the CL also plays a key role on luteal regression. Treatment with PGF leads to acut e in crease in blood flow at the periphery of the CL, progesterone increase s and LH changes followed by a gradual decrease in luteal blood flow in the d 10 midcycle CL. However, the same events are not observed in the d 4 early CL in cattle (Acosta et al., 2002; Ginther et al., 2007). Moreover, each peak of

PAGE 58

58 uterine PGF during spontaneous luteolysis lead s to increased luteal blood flow in cattle (Miyamoto et al., 2005; Shirasuna et al., 2008). Nitric oxide (NO) is a potent vasodilator and has been reported to directly inhibit progesterone secretion and induce apoptosis in bovine luteal cells (Skarzynski et al., 2000; Skarzynski et al., 2003). In the bovine, treatments with PGF up regulated expression of endothelial NO synthase (NOS) and increased luteal blood flow in mature CL 30 min after PGF injection (Shirasuna e t al., 2008). Treatment with the NO donor S nitrolo N acetyl D,L penicillamine into CL led to an acute increase in luteal blood flow and shortened the estrous cycle, whereas injection of the NOS inhibitor L NG nitroarginine methyl ester into CL completely blocked the acute increase in luteal blood flow induced by PGF and delayed the beginning of luteolysis (Shirasuna et al., 2008). Therefore, it seems that the increased blood flow in the mature CL is modulated by NO, indicating that an acute elevation in peripheral blood flow to CL is likely one of the first physiol ogical indicators of NO action in response to PGF Endothelial cells are the primary type of cells to undergo apoptosis. It has been suggested that some capillaries disappear earlier than larger vessels with smooth muscle in the CL during luteolysis (Ho jo et al., 20 09). The strong vasoconstrictive factors endothelin 1 (EDN1) and angiotensin II (Ang II) are involved in the process of luteal regression in ruminants (Miyamoto et al., 1997; Hayashi and Miyamoto, 1999). Indeed, EDN1, Ang II, and NO i nhibited progesterone secretion and PGF up regulated expression of EDN1 in bovine CL in vitro (Miayamoto et al., 1997). Therefore, luteolytic PGF is likely responsible for the regulation of endothelial and vascular functions

PAGE 59

59 through induction of angiolysis and vasoconstriction to limit the o xygen and nutrient supply during luteolysis. During luteolysis, leukocytes, particularly macrophages and T lymphocytes, significantly increase in number, and 70% of proliferating cells in the bovine CL are CD14 + macrophages (Bauer et al., 2001). In cows, l arge numbers of CD4 + and CD8 + T cells were found in regressing CL (Bauer et al., 2001), indicating an active role for t hese immune cell types in luteolysis. Furthermore, inflammatory cytokines, such as TNF IL involved in luteal regression (Pate and Keyes, 2001; Okuda and Sakumoto, 2003; Neuvi a ns et al ., 2004). These immunomodulatory factors may stimulate the accumulation of T lymphocytes and macrophages within the CL to support the luteolytic cascades. Afterwards, the CL regresses mostly through the loss of cells by apoptosis (Juengel et al., 1 993), and apoptotic luteal cells are phagocytized by macrophages (Kato et al., 2005). L uteal cells express both class I and class II major histocompatibility complex (MHC) molecules These MHC molecules are essential for the recognition of cells by T lymp hocytes as either self or nonself (Fairchild and Pate, 1989; Khoury and Marshall, 1990). The expression of MHC class II on luteal cells increased when luteal regression was induced by PGF suggesting that the demise of the CL might be involved in local au toimmune response mechanisms facilitated by increased expression of class II MHC molecules at the time of luteolysis (Benyo et al., 1991). Furthermore, bovine luteal cells can stimulate T cell proliferation and this response is increased in the presence of cells from regressing CL as compared to responses in cells from midcycle CL in cows This

PAGE 60

60 results suggests that luteal cells may act as antigen presenting cells, initiating a transient autoimmune response during luteolysis (Petroff et al., 1997). Macroph ages and T lymphocytes can produce several cytokines, such as TNF IFN connect with peripheral resident cells depending on the stimulatory conditions. The cytokine TNF in large and small luteal cells, endothelial cells, and immune cells in the bovine CL and inhibits progesterone secretion and induces IFN mediated apoptotic cell death in bovine luteal and endothelial cells by inc reasing caspase 3 activity (Tani guchi et al., 2002; Pru et al., 2003). Interferon gamma inhibits LH stimulated progesterone production, increases prostaglandin synthesis, upregulates MHC class I and II molecules, and induces cell death (Fairchild and Pate, 1989, Fairchild and Pate, 1995) The interaction between leukocytes and endothelial cells for leukocyte recruitment implies an overlapping succession of adhesive events encompassing leukocyte induction, rolling, and firm adhesion onto endothelial cells. P selectin and E selectin on endo thelial cells can interact with leukocytes to promote leukocyte rolling and transient adhesion (Sako et al., 2003) The bovine luteolytic cascade can be compared to an acute process with massive immune infiltration and vascular alteration with increased ang iogenesis and blood flow. Initially, there is vasoconstriction likely induced by PGF and simultaneously the cell adhesion molecule P selectin is acutely translocated to endothelial cells membranes, on which P selectin strongly interact with its receptor present in neutrophils promoting rolling and transit adhesion of neutrophils (Shiras una et al., 2012). In fact, it was demonstrated that after 5 min of the administration of PGF P selectin expression

PAGE 61

61 and neutrophil adhesion increased in luteal endothelial before any chemoattractants of neutrophils were increased. This suggests that P s electin actively contribute to infiltration of neutrophils in CL during PGF induced luteolysis. At 10 to 30 min after the beginning of acute inflammation, vascular dilatation starts in local arterioles, and the velocity and volume of blood flow increase s resulting in the induction of high levels of neutrophil adhesion. This acute increase in blood flow peaks several hours after the start of acute inflammation, and blood flow disappears approximately 1 d later (Shirasuna et al., 2012). Reproductive Manage ment, Efficiency and Timed Artificial Insemination Reproductive efficiency has been acknowledged widely as a major contributor for herd profitability in dairy operations, which makes the adoption of a dequate reproductive management practices a critical com ponent for the success of dairy farms (Britt, 1985; Giordano et al., 2011; Ribeiro et al., 2012a; Galvo et al., 2013). Failure to obtain proper reproductive efficiency results in reduced percentage of cows at the early stages of lactation, increased costs with AI, and delay ed genetic progress (Santos et al., 2010b). Furthermore, low reproductive performance leads to reduced total milk sales, increased number of culled cows, and reduced number of replacement heifers born (Britt, 1985; De Vries, 2004; Meadow s et al., 2005 ; Ribeiro et al., 2012a ). Other factors influenced by the selection of a specific reproductive management program and its efficiency include feed cost, labor cost, and veterinary expenses that are generally associated with lactation length, d ry period length, and number of services to conceive (Lima et al., 2010 ; Giordano et al., 2011 ; Ribeiro et al., 2012a ). Therefore, the selection a long term reproductive management strategy that improves reproductive efficiency is a critical decision for d airy operations sustainability.

PAGE 62

62 Recently, a body of evidence in the literature suggests that incorporation of timed AI in reproductive programs either as sole strategy or in combination with AI at estrous detection can lead to economic benefits for lactati ng dairy cows and heifers (Lima et al., 2010 ; Giordano et al., 2011 ; Ribeiro et al., 2012 a; Galvo et al., 2013). Lima et al. (2010) compared the cost of timed AI and natural service using as input s reproductive results and economical information from a fi eld study that evaluate d these two reproductive programs (Lima et al., 2009), and reported that the cost of a pregnancy was $9.73/cow less for timed AI than natural service. Furthermore, sensitivity analysis using increased cost of feeding or increased P/A I resulted in an even greater economic advantage of timed AI compared with natural service (Lima et al., 2010). In another study, the profitability of three different reproductive program s (100% estr o us detection vs. two programs with 100% timed AI) on a specific dairy farm were estimated using Markov chain simulation with partial budgeting and sensitivity analysis performed to assess the effect of varying specific reproductive parameters on profitability (Giordano et al., 2011). The results of this study revealed two reproductive program s w ith 100% timed AI, the Double Ovsynch and d 32 resynch at first and second or subsequent service or Double Ovsynch and Double Ovsynch resynch at first and second or subsequent service were superior to the 100% estr o us detection program For programs with 100% timed AI Double Ovsynch for resynch resulted in increased P/AI for resynchronized services and it was economically superior despite having higher costs and a longer interbreeding interval. A 4% increase in P/AI fo r resynchronized AI was sufficient for the Double Ovsynch resynch to outperform the program with d ay 32 resynch for subsequent services. Adding estr o us detection to the

PAGE 63

63 100% timed AI programs was only beneficial for the program with the lower P/AI The impr ovement in service rate required for the 100% estr o us program to have the same profitability as the superior 10 0% timed AI program was 12%, which ultimately suggests that well managed timed AI is superior to estr o us detection. A third study compared the ec onomic outcome of reproductive programs using estr o us detection, timed AI, or a combination of both timed AI and estrous detection using a stochastic dynamic Monte Carlo simulation model (Galvo et al., 2013). The programs evaluated were 100% estrous detec tion ; timed AI with Presynch Ovsynch for first AI, and Ovsynch for resynchronization of open cows at 32 d after AI; or timed AI estrous detection with Presynch Ovsynch for first AI, but estrous detection and AI after first AI, and cows diagnosed open 32 d after AI were resynchronized using Ovsynch. Sensitivity analysis was performed using estrous detection rate s of 40 vs. 60%, accuracy of estrus of 85 vs. 95%, and compliance with timed AI of 85 vs. 95%. For programs with an estrous detection rate of 40%, ti med AI program with 95% of compliance was more profitable than the 100% estrous detection or the timed AI estrous detection programs. For programs with estrous detection rate of 60%, the program timed AI estrous detection resulted in the greatest profit fo llowed by 100% estrous detection programs and 100% timed AI program, respectively. Combining timed AI and estrous detection increased profits within each level of accuracy or compliance Adding timed AI to estrous detection would increase overall profit/co w per year by $46.8 to $74.7 with 40% estrous detection rate, and by $8.9 to $30.5 with 60% estrous detection rate. Therefore, the combination of timed AI with estrous detection program

PAGE 64

64 was the reproductive management program that maximized profit for dair y producers (Galvo et al., 2013). An economic analysis of reproductive programs for dairy heifers with timed AI and estrous detection was performed using a simulation to calculate pregnancy rates, average time to pregnancy, total costs per AI, and pregna ncy for four reproductive management strategies with breeding period of 84 days allowing approximately four estrous cycle s (Ribeiro et al., 2012a). The four breeding programs used were: 1) 100% timed AI; 2) 100 % detection of estrus; 3) timed AI for first breeding and detection of estrus for the remaining services; and 4) timed AI for first breeding followed by insemination upon detected estrus or resynchronized insemination after nonpregnancy diagnosis. Sensitivity analyses were performed for four estrous detection rates of 50, 60, 70 and 80%. The results of this study revealed that incorporation of timed AI for first service lowered the cost per pregnancy compared with estrous detection alone, although the benefits were less as estrous detection rates incr eased. Likewise, programs with 100% timed AI were less expensive than programs with exclusive use of insemination at detected estrus only if detection of estrus were below 70%. When additional timed AI were incorporated in to the breeding program to resynch ronize nonpregnant heifers that had not been detected in estrus, it further benefited the program with low estrous detection rates of 50 and 60%. H owever for higher estrous detection rates of 70 and 80% the benefits were minor or inexistent. Incorporati on of detection of estrus after one timed AI was superior to timed AI alone only when estrous detection rate was 60% or more. Most of the changes in costs per pregnancy resulted from feed costs associated with heifers becoming pregnant later in the breedin g period.

PAGE 65

65 In summary, combination of timed AI program for first insemination with estrous detection and additional timed AI for nonpregnant heifers maximized the economic success of a reproductive program; however when estrous detection rates were equal or greater than 70% benefits of additional timed AI for nonpregnant heifers was negligible (Ribeiro et al., 2012a). Although reproductive efficiency is critical for economical sustainability of dairy operations, there is substantial evidence that measuremen ts of reproductive performance declined in the last four decades in the US until the mid (Washburn et al., 2002; Butler, 2003; VanRaden et al., 2004, De Vries and Risco et al., 2005; Norman et al., 2009). Several factors have contributed to the impa irment of reproductive efficiency including cow physiology, nutritional and reproductive management and genetics (Lucy et al., 2001; Weigel et al., 2006). The evolvement of the dairy industry in the US have been marked by intensive genetic selection for pr oduction traits and consolidation of farms resulting in increased herd size, housing of cows on concrete floor with smaller area per cow and reduced time detecting estrus at individual cow basis (Senger, 1994; Roelofsa et al., 2010). Two major problems aro used with the modernization of dairy industry: 1) milk production, the major trait used for genetic selection, has an antagonistic relationship with reproduction in dairy cows and estrous behavior (Laben et al., 1982; Pryce et al., 1997, Butler, 2003; VanR aden et al., 2004 ; Cochran et al.2013 ), and 2) the new management practices and housing of dairy farms are associated with impaired expression of estrous behavior and reduced estrous detection rates (Senger, 1994; Lopez et al., 2004; Roelofsa et al., 2010)

PAGE 66

66 Although some studies evaluating the association of milk production and fertility within a herd and among many herds did not reveal the same trend of a negative relationship between production and fertility (Santos et al., 2009; LeBlanc, 2010), it is a general consensus that milk production modifies behavior, metabolic, nutritional and health demands in lactating dairy cows (Wiltbank et al., 2006). In fact, Lopez et al., (2004) reported high producing cows (averaging 46.4 kg/day) in comparison with low p roducers (averaging 33.5 kg/day) had shorter duration of estrus (10.9 h vs.6.2 h), reduced number of total standing events (6.3 vs. 8.8) and reduced standing time (21.7 s vs. 28.2 s). These results clearly exemplif y how behavior al estrus a key component f or proper detection of estrus is hindered by milk production. In the last 20 years, research efforts were devoted to the development of knowledge, techniques and strategies to reverse this trend of declining fertility in dairy cows (Lucy et al., 2001). On e of the most important reproductive techniques developed in the recent years was timed AI, which is a program to synchronize ovulation characterized by a time sensitive sequential use of hormonal treatments to induce occurrence of follicle turnover, luteo lysis and synchronous ovulation allowing cows to inseminated in pre determined optimum time. Timed AI was instrumental and added value to the classic estrous synchronization programs based on PGF to control luteal lifespan because it overcomes issues rel ated to cow factors (anovulation, lameness, low intensity and duration of estr o us behavior), environmental factors (heat stress, poor flooring) and human errors that limit estrous detection rates and subsequent reproductive performance (Wiltbank et al., 2 006).

PAGE 67

67 The first stride towards development of timed AI occurred when GnRH was given seven days prior PGF injections leading to increased control of follicle development and acceptable fertility (Thatcher et al., 1989). The use of GnRH resulted in occurrence of LH surge within two hours and ovulation 28 hours later in dairy cattle (Bodensteiner et al., 1996 ). The decisive stride towards the development of timed AI program was the addition of second GnRH injection 48 hours after the injection of PGF followed by insemination 24 to 32 hours later (Pursley et al., 1995; Burke et al., 1996). This program was la ter named Ovsynch (Pursley et al., 1997) and it is now part of the reproductive programs in most dairy herds in the US. The proportion of synchronized breeding in the US was estimated to be between 10 and 19% depending upon region and it increased from 2% in 1996 to as much 35% in 2005 for all cows inseminated for their first service (Miller et al., 2007). After the development of timed AI several studies exploited key aspects of reproductive physiology aiming to optimize the results of timed AI program for dairy cows. Pursley et al., (1998) investigated the optimal interval between second GnRH injection and AI for the Ovsynch program inseminating at 0, 8, 16, 24, or 32 h after the second GnRH treatment. This study found a quadratic effect of time of AI o n P/AI, which increas ed from 0 to 16 h with subsequent decreases from 24 to 32 h when the expected time of ovulation occurs However, only insemination performed after the expected time of ovulation , resulted in a statistically decrease in both P/AI and percentage calving per AI (Pursley et al., 1998). Analysis of resulted in similar rates of calving. Therefore, the optimal time for AI was determined to

PAGE 68

68 be approximately timed AI before ovulation (between 0 to 24 hours), on which pregnancy outcomes should be not ideal, bu t acceptable (Pursley et al., 1998 ). The physiological basis for an optimal time for insemination at 16 h after GnRH or 12 h before ovulation is an ideal synchrony between sufficient time for optimal sperm capacitation and the presence of oocyte that is no t overly aged in the reproductive tract (Saacke et al., 2000). Another important aspect of reproductive physiology related to Ovsynch protocol is that synchronization of ovulation and fertility are optimized when cows receive the first GnRH of the protoco l in early diestrus between days 5 to days 9 of the estrous cycle (Vasconcelos et al., 1999). Indeed, cows receiving the first GnRH during early diestrus are unlikely to have spontaneous luteolysis in the middle of the program, have limited follicle domina nce and are benefited by the presence of a CL and optimal concentrations of progesterone, which in turn is critical for follicle development (Bleach et al., 2004, Cerri et al., 2009a, Bisinotto et al., 2013). Using the concept of ideal timing to initiate a timed AI program Moreira et al. (2001) developed a program named P resynch, which aimed to increase the chance of obtaining cows on early diestrus at initiation of Ovsynch protocol. The Presynch Ovsynch program developed by Moreira et al. (2001) was comp osed of two injections of PGF given 14 days apart, with the second injection given 12 days prior to the first GnRH of the Ovsynch protocol W hen compared with Ovsynch alone the Presynch Ovsynch remarkabl y increased P/ AI in cyclic lactating c ows (from 25% to 43%). Although the prog ram developed by Moreira et al., (2001) was successful and other studies confirmed the benefits of this program (El Zarkouny et al., 2004; Navanukraw et

PAGE 69

69 al., 2004), it was based solely on PGF 2 which has limited to no efficacy in anovular cows and cows no t bearing a CL at time of the PGF 2 injection s Therefore, other programs were successfully developed to presynchronize the estrous cycle in dairy cows using GnRH before the PGF 2 injection increasing the chance of obtaining cows in early diestrus at initi ation of the timed AI programs independent of cyclic status (Bello et al., 2006, Galvo et al., 2007; Souza et al., 2008). When programs based on the incorporation of one treatment of GnRH before PGF were compared with the presynchronization programs ba sed on PGF 2 alone they did not benefit fertility of dairy cows (Galvo et al., 2007; Ribeiro et al., 2011). The Double Ovsynch protocol that incorporates an entire Ovsynch program as presynchronization improved P/AI in comparison with the presynchronizat ion based on PGF 2 (Souza et al., 2008). Other alternative presynchronization programs used progesterone supplementation aiming to induce estrous cyclicity in of anov ular dairy cows before timed AI. D espite the fact that these programs induced cyclicity in about 50% of the anovular cows no improvements in fertility were observed (Chebel et al., 2006; Bicalho et al., 2007; Stevenson, 2011). Generally, presynchronization is only feasible to the first insemination postpartum to avoid delays in re inseminatio n for subsequent services. However, a recent study that compared resynchronization programs using the conventional Ovsynch to Double Ovsynch the latter used as a presynchronization and synchronization program all i n one, revealed that cows rebre d using th e Double Ovsynch program had increased synchronization rate (72 vs. 51%) and P/AI (39 vs. 30%) compar ed with regular Ovsynch (Giordano et al., 2012). Although this program had additional costs with

PAGE 70

70 hormonal treatments and increased interval between insemin ations an economic analysis indicated that the incorporation of the Double Ovsynch for resynchronization within a reproductive program resulted in a more profitable breeding program than use of Ovsynch alone for resynchronization (Giordano et al., 2011). Nevertheless, this benefit was observed because cows began the resychronization before p regnancy diagnosis in an attemp t to offset the lengthy resynchronization protocol with Double Ovsynch. Another aspect related to timed AI exploited by some recent stud ies was the reduction in the period of follicular dominance (Santos et al., 2010a), which previously was shown to be important to improve embryo quality (Cerri et al., 2009a). Santos et al., (2010a) shortened the interval between the initial GnRH and the i njection of PGF from 7 to 5 days, which lead to an increased P/AI in lactating dairy cows (37.9 vs. 30.9%). It is important to mention that for 5 d program an additional injection of PGF was given on d 6 of the program to ensure that a newly formed CL in response to the initial GnRH was fully regressed (Santos et al., 2010a). Additionally, on the 5 d program the administration of the second GnRH 56 hour after the PGF did not improve P/AI (Bisinotto et al., 2010a) such as reported in the 7 d program (Pursley et al. 1998; Brusveen et al., 2008) The 16 hours prolonged proestrus in a 5 day program likely resulted in additional growth of the ovulatory follicle, greater concentrations of estradiol, which explain ed the increased percentage of cows in estru s (Bisinotto et al., 2010a). Therefore, the potential benefits of having an improved synchrony between oocyte availability and numbers of viable spermatozoa for fertilization (Saacke, 2008) encountered in programs of 7 days with GnRH given 56

PAGE 71

71 hours after t he PGF is offset by prolonged proestrus in programs of 5 days (Bisinotto et al., 2010a). Anovulation or in some cases, the absence of a CL due to stage of the estrous cycle when a timed AI is initiated reduces the risk for pregnancy by 30 % suggesting t hat the concentration of progesterone on which a future dominant follicle grow s is critical for fertility (Bisinotto et al., 2010b). A recent study targeting supplementation of progesterone in cows lacking CL at the initiation of the timed AI progra m was a ble to reverse the negative impact of lack of CL on P/AI (Bisinotto et al., 2013). Holstein s cows were evaluated for the presence of CL and cows not bearing a CL at initiation of 5 d timed AI program were allocated randomly to either remain as untreated c o ntrol or receive two controlled internal drug release (CIDR) inserts containing progesterone for 5 days. Cow s without the CL that received two CIDR inserts as a treatment had similar P/AI than cows in diestrus (46.8 vs. 49.9 %) and both had greater P/AI th an cows without CL that remained untreated (30.8 %). The use of two CIDR devices was successful because elevated plasma concentration of progesterone to 2.65 ng/mL, which is greater than the 0.8 to 1.0 ng/mL identified in cows supplemented with only one CI DR insert (Cerri et al., 2009b; Lima et al., 2009b). These results suggest that one CIDR insert likely released an insufficient concentration of progesterone to optimize follicle or oocyte maturation during the final stages of development before AI. In fac t, supplementation of lactating dairy cows with a single CIDR insert during timed AI programs improves fertility of those bearing a CL, but not in cows without a CL (Bisinotto and Santos, 2012).

PAGE 72

72 Timed AI programs for dairy heifers have had a different hist orical trend than timed AI programs for lactating dairy cows. The first research projects that exploited timed AI for dairy heifers failed to obtain acceptable P/AI rang ing from 25.8% to 45.5%, which w ere consistently inferior to results obtained for AI af ter detected estrus within the same studies (Schmitt et al., 1996; Pursley et al., 1997, Stevenson et al., 2000; Rivera et al., 2004; Rivera et al., 2005; Rivera et al; 2006). The first studies investigating timed AI for dairy heifers used Ovsynch or simil ar protocols with a 7 days interval between the first GnRH and PGF 2 (Schmitt et al., 1996: Pursley et al., 1997; Stevenson et al., 2000). This approach likely lead to a period of follicle dominance that potentially is too long for the 50% dairy heifers th at have 3 follicular waves (Table 3 1) The next series of studies reduced the period of follicle dominance using a 6 day program (Rivera., et al., 2004) and added a CIDR insert between first GnRH and PGF to limit follicle turnover in the middle of the p rotocol (Rivera et al., 2005; Rivera et al., 2006), but the results for these studies were still not comparable to AI performed after detected estrus The first stride toward the development of a timed AI program in heifers that result ed in acceptable fert ility occurred when Rabaglino et al. (2010b) reduced the period of follicle dominance even further using a 5 day program and a CIDR insert between the 1 st GnRH and PGF injection followed 72 hours later by concurrent GnRH and AI. The results of P/AI for this study were the first comparable to insemination performed after estrous detection ranging from 53.1% to 59.5% (Rabaglino et al., 2010b; Kuhn et al., 2010). Another study subsequent ly investigated the effects of the addition of a second PGF 2 12 hours after the first treatment in the 5 day timed AI program for dairy heifers finding no improvements in P/AI and luteolysis (Rabaglino et

PAGE 73

73 al., 2010a). However, an interest ing finding from this study was that only 23% of the heifers had more than one CL 5 d after the injection of GnRH suggesting that ovulation to the initial GnRH was probably low (Rabaglino et al., 2010a). The importance of the proestrus length and follicle dominance period was investigated in recent study that compared a modified 5 day timed AI program with GnRH injection and insemination performed concurrently from 53 to 60 hours after the PGF with a 7 day timed program using also 53 to 60 hour interval from PGF (Lopes Jr. et al., 2013). The P/AI for the modified 5 day timed AI with shorter than usual proestrus was only 44.8%, but was still greater than the P/AI for the 7 day timed AI prog ram (35.7%) suggesting the shorter proestrus might be detrimental to P/AI and a longer period of follicle dominance associated may aggravated even more the detrimental effect on fertility of dairy heifers. In contrast with most of the literature, two rece nt studies compared 5 d timed AI program with 7 d timed AI in dairy heifers reporting no differen ces in P/AI (Colazo and Ambrose 2011; Mellieon Jr. et al., 2012). Although, these studies were well designed caution is required on their interpretation, bec ause Colazo and Ambrose (2011) used only 64 heifers which makes it difficult to have sufficient power to identify difference in P/AI. Mellieon Jr. et al. (2012) used sexed semen and had very low P/AI approximately 30%, which also can mask the ability to i dentify a statistical difference among treatments. Uterine Diseases Relevance and Characterization Uterine diseases affect half of the dairy cows after parturition leading to disruption of uterine and ovarian fu nction which frequently hinders fertility, in creases culling and contributes to remarkable economic losses for dairy producers (Sheldon et

PAGE 74

74 al., 2009). The most common forms of uterine diseases includes retained fetal membranes (RFM), metritis and endometritis. Retained fetal membranes are character ized as failure to release the placenta within 12 (Van Werven et al., 1992) or 24 h post calving (Kelton et al., 1998). Ninety five % of the cows that released the placenta within 24 hour s did so within 12 hours; therefore, this distinction between 12 or 2 4 hours seems to be of little, if any relevance (Van Werven et al., 1992). Once RFM occurred, fetal membranes tend to be retained for 7 days on average (Eiler, 1997). Retained fetal membranes have detrimental impact s on fertility and are a maj or risk facto r for metritis, clinical endometritis, and other periparturient diseases that ultimately compromise survival, production, and reproductive performance (Grohn et al., 1990; Oltenacu et al., 1990; Laven and Peters, 1996). Retained fetal membranes affect 7.8% of the US dairy cow population (NAHMS, 2009). There are a number of risk factors associated with RFM, including induced parturition, shortened gestation, abortion, dystocia, fetotomy, cesarean section surgery amin E and selenium, and immunosuppression (Muller and Owens, 1974; Terblanche et al.,1976; Julien and Conrad, 1986; Jo osto n et al., 1987; Rajala and Grohn, 1998; Laven and Peters, 1996). The impact of RFM ranges from impaired reproductive performance to de v elopment to severe metritis with loss of milk production and reproductive performance and increase risk of culling. A meta analysis revealed that the daily rate of pregnancy decreased by 16% in cows diagnosed with RFM relative to unaffected cows (Fouric hon et al., 2000), which extend ed interval to pregnancy by up to 51 days (Borsberry and Dobson, 1989). The estimated cost of each case of RFM ranges from

PAGE 75

75 US $355 to $464 (Kossaibati and Esslemont, 1997; Laven, 2006) and the major cause of this economic lo ss is reduced milk yield and consequent reduction of milk sales. In fact, losses of milk production are associated with progression to metritis. It has been estimated that cows affected by RFM have a 3 kg/d reduction in milk yield for the first 5 d after c alving, and cows with a RFM have 2 to 4 fold increase relative risk of developing metritis (Santos et al., 201 0 b). Cows affected by RFM that develop metritis have a cumulative loss of milk of approximately 110 kg during the first 7 weeks of lactation (Delu yker et al., 2012). When left untreated, RFM alone reduced milk yield in primiparous and multiparous cows by 412 and 537 kg/lactation, respectively. Similarly, when left untreated, metritis alone reduced milk yield in primiparous and multiparous cows by 33 8 and 498 kg/lactation, respectively. Metritis is characterized by an abnormally enlarged uterus and a fetid, watery red brown fluid to viscous off white purulent uterine discharge within 21 days postpartum, but more frequently diagnosed in the first week postpartum (Sheldon et al., 2006). Metritis can be classified according to the severity of the disease as: grade 1 if they have an abnormally enlarged uterus and a purulent uterine discharge without any systemic signs of disease; grade 2 when additional si gns of systemic illness such as decreased milk yield, dullness and fever >39.5 C are present; and grade 3 when signs of toxemia such as inappetence, cold extremities, depression and/or collapse are present, which generally is associated to a poor prognosi s (Sheldon et al., 2006). Grades 2 and 3 can also be called acute puerperal metritis or toxic puerperal metritis (Drillich et al., 2011). The incidence of metritis in dairy cows range from 10 to 36% (Goshen and Shpigel, 2006; Santos et al., 2010b; Chapinal et al., 2011). In many

PAGE 76

76 studies, the reported incidence of metritis was low, likely because only cows that develop fever concurrent with metritis were classified as having metritis. However, approximately 50 to 60% of the cows diagnosed wit h metritis do no t develop fever, based on o C, which results in underestimation of the incidence of the disease (Benzaquen et al., 2007, Martinez et al., 2012). The economic losses caused by metritis are striking and it has been calculated at $38 0 per affected cow due reduced milk production, delayed conception, treatment and increased culling ( Drillich et al., 2001). Thus, if a conservative incidence rate of 20% for the 8.5 million dairy cows in US the annual cost of metritis alone is $650 millio n. Many factors have been associated with increased risk of developing metritis. behavior and the subseque nt development of metritis. C ows with severe presentation of metriti s consumed 2 to 6 kg of less of dry matter than healthy cows at 2 to 3 weeks before the diagnosis of metritis. Increased concentrations of nonesterified fatty acids ( NEFA ) hydr o xybutyric acid ( BHBA ) early postpartum and neutrophils with less intracellul ar glycogen were associated with cows diagnosed with metritis (Galvo et al., 2010). Other factors associated with metritis were plasma concentrations of hap toglobin in the first week postpartum (Dubuc et al., 2010b). Additionally, cows presenting subclinical hypocalcemia based on serum total 8.59 mg/dL in one or more of the first 3 days post calving had 3.2 and 11.5 times greater risk of developing metritis and puerperal metritis, res pectively (Martinez et al., 2012). Although dystocia, twinning stillbirth, and male offspring are all risk factor for the development of metritis RFM is considered the

PAGE 77

77 most important risk factor for metritis (Grhn et al., 1990; Correa et al., 1993). Any condition that may impair feed intake and immune function increase s the risk of metritis (Leblanc et al., 2007). Endometritis is an inflammatory disease of the pelvic tissue diagnosed after 21 days postpartum caused by the inability of the host to elimin ate microbial infection (Sheldon et al., 2006). Endometritis is classified as clinical or subclinical. Clinical endometritis is defined by presence of purulent vaginal discharge detectable 21 days or more after parturition, or mucopurulent discharge detect able in the vagina after 26 days postpartum. Subclinical endometritis is characterized by inflammation of the endometrium defined by presence of PMNL exceeding between 5.5% (Santos et al., 2009) and 10% of cells (Kasimanickam et al., 2004) in samples colle cted by flushing the uterine lumen or by endometrial cytobrush, in the absence of clinical signs at approximately 5 weeks postpartum (Sheldon et al., 2006). The inflammation is presumably associated with recovery of the tissues after clinical endometritis, trauma or other non microbial disease. T he nomenclature for clinical endometritis may not be the most appropriate because a large proportion of cows presenting pus in the vaginal discharge do not have concurrent neutrophil infiltration in the endometrium (Dubuc et al., 2010). Thus, purulent vaginal discharge was suggested as an alternative terminology for clinical endometritis to resemble properly what have been diagnosed in cases of clinical endometritis (Dubuc et al., 2010). The incidence of subclinical endometritis is dependent on the cut off for diagnosis and the time after parturition but is in the order of 37 to 74% of animals (Gilbert et al., 2005). Most importantly subclinical

PAGE 78

78 endometritis affects 40% cows classified healthy at the breeding time (G ilbert et al., 2005). Clinical and subclinical endometritis have been associated with reduced P/AI at first service, increased pregnancy loss, and prolonged time to pregnancy (Kasimanickam et al., 2005; Galvo et al., 2009a; Dubuc et al., 2011). Cows diag nosed with both clinical and subclinical endometritis had the longest interval from calving to pregnancy compared with those diagnosed with only one of the two diseases or with cows having no diagnosis of uterine diseases ( Dubuc et al. 201 1) Subclinical endometritis, a persistent uterine disease leads to reduced fertilization, and compromises early embryo development and survival (Cerri et al., 2009; Hill and Gil bert, 2008; Galvo et al., 2009 ). Additionally, cows diagnosed with subclinical endometritis have altered endometrial and embryonic gene expression (Hoelker et al., 2012). Endometrium from cows diagnosed with subclinical endometritis had altered patterns of expression of genes involved in cell adhesion and immune modulation and embryos from cows w ith subclinical endometritis had altered pattern of gene expression involving pathways in cell cycle and apoptosis. Uterine Disease Etiology: Immunological and Microbiological Aspects Throughout the last few decades, major advances have been made to better understand how host uterine innate immunity is subverted by microbes leading to development of uterine diseases. Although, microbes are inextricably linked to uterine diseases development, a dysregulation of uterine immunity alone may also lead to develop ment of uterine disorders that impair fertility of dairy cows. Research conducted in the last 3 decades identified three general causes of RFM in cattle: factors around parturition maintaining blood pressure within the chorionic

PAGE 79

79 villi, occurrence of uteri ne atony, and dysregulation at a molecular and cellular level (McNaughton and Murray, 2009). The first cause is associated with pathological changes that interfere with the detachment of fetal and maternal epithelial components within the placentome such a s occurrence of villous edema as result of cesarean section surgery or uterine torsion (Grunert, 1986). Uterine atony is a second cause and it is conside red very unusual occurring in 1 to 2% of the cases and it is associated with multiple pregnancies in wh ich the myometrium had been stretched excessively and muscle tonicity is compromised (Grunert, 1986). The third and most important possible cause of RFM is dysregulated inflammation and immune function. Histological investigation of the placentome of cows with RFM revealed small areas of necrosis between the villous trophoblast and the crypt epithelium, suggesting that there may have been a more generalized disease condition, which was associated with the presence of microorganisms (Al Sadi et al., 1994). R ecently, many other immune dysfunctional related factors have been associated with RFM including collagenase, hyaluronidase, matrix metalloproteinases (MMPs), reactive oxygen species (ROS), major histocompatibility complex (MHC), and feto maternal steroido genesis. Factors related to extracellular matrix components in the placentome are suggested to be involved with the cause of RFM. After a normal parturition, collagen fibers in the caruncular connective tissue have a swollen appearance with indistinct cont ours and a linear arrangement, whose breakdown is regulated by the enzymes collagenase and hyaluronidase. R etained fetal membranes increase the activity of these enzymes in both the maternal and fetal compartments of the placentome (Kankofer et al., 1998). M atrix metalloproteinases are a family of zinc

PAGE 80

80 and calcium dependent endopeptidases that degrade extracellular matrix proteins and are also involved in the breakdown of extracellular matrix components such as collagen (Woessner, 1998). They are synthesiz ed by some inflammatory and epithelial cells and secreted as inactive precursors that are activated in extracellular spaces by enzymes such as plasmin. The activities of MMP 2 and MMP 9 are regulated at the cellular level by proenzyme activation and by tis sue inhibitors metalloproteinase TIMP 1 and TIMP 2. ProMMP 9 activity is found only in the maternal compartment and has no influence on placental separation after calving. However, several active isoforms of MMP 2 are present within both compartments of t he feto maternal unit during normal stage 3 of labor, but only one, the 68 kDa form, has been detected in cows with a retained placenta. Maj and Kankofer (1997) considered that higher activities of proMMP 2 and the absence of other MMP 2 isoforms affected the hydrolysis of collagen so that an increase in rigidity of the intravillous extracellular matrix caused placental retention. Immunological antioxidant mechanisms against ROS may also play an important role in the release of fetal membranes. R eactive ox ygen species produced during normal cellular metabolism are harmful if not removed causing peroxidative damage to cell membranes and other cellular components. Alternatively, they may react with other cellular reducing equivalents, such as NADPH, which dis rupt cellular biochemical processes such as glucose metabolism. Some of their negative influences include changes to the steroidogenic and arachidonic acid cascades (Staats et al., 1988). Enzymes such as glutathione peroxidase, catalase and superoxide dism utase remove ROS. The antioxidant status of placentomes, in terms of their enzyme activity for glutathione transferase, catalase and superoxide dismutase, was remarkably reduced

PAGE 81

81 for up to two weeks before calving in cows that subsequently were diagnosed wi th RFM (Kankofer, 2001). In case these mechanisms fail, chain breaking antioxidants are also present, such as vitamin E, which inhibits the chain reactions initiated by free ROS (Miller and Brzezinska Slebodzinska, 1993). Oxidative DNA damage increase s co deoxyguanosine (8 OH dG), which is associated with damage to DNA was identified in increased concentrations in both maternal and fetal compartments have been recorded in cows with RFM (Kankofer and Schmerold, 2002). I n ad dition to having an antioxidant role, vitamin E also regulates the activity of PL A 2 which is important in cleaving arachidonic acid from cell membranes during the tocopherol binds to and inhibits P LA specifically and effectively. A dysregulation of the metabolism of prostaglandin was also revealed in cows with RFM having an increased plasmatic concentration of PGF and reduced concentration of PGE 2 in comparison to unaffected cows (Chandra et al., 2002). Another possible factor involved with RFM was a downregulation of maternal antigen recognition. Cows suffering from RFM had reduced chemotactic activity of maternal leucocytes and periparturient cows in which there was no leucocyte activity, the in cidence of RFM was 100 % (M i yoshi et a l ., 20 02) MHC I compatibility of pregnant cows and their calves was another relevant factor linked to RFM. Cows that were more homologous with their fetuses had a greater risk of RFM than cows that were more dissimilar (Jooston et al., 1991). Therefore, from an immunological perspective, the development of RFM could be associated with a reduction in the variability of MHC I

PAGE 82

82 expression between dam and fetus that reduces the production of appropriate lymphokines within th e feto maternal unit as it develops towards its mature state. The synthesis of prostaglandins by the uterus is another possible dysregulated pathway in the bovin e leading to RFM. Takagi et al. (2002) reported that the ratio of caruncular PGE 2 to PGF at parturition and six hours later was lower in cows with RFM, suggesting a hampered role for these two prostaglandins in placental separation. Additionally, there might be an imbalance in arachidonic acid metabolism in the endometrium RFM cows, possibl y involving vasoactive related peptide systems such as endothelin 1 (Takagi et al., 2008). In vitro studies of the synthesis of prostaglandins in uninucleate and binucleate cells obtained from cows with a RFM produced predominantly PGE 2 on the other hand, cells from healthy cows synthesized more PGF (Gross and Williams, 1988). Uninucleate cells synthesize more prostaglandin than binucleate cells, but binucleate cells readily convert PGF into PGE 2 There are fewer viable binucleate cells in placentas that have been released normally than in placen tas that have been retained for one hour and then removed. The enzyme PGE 2 9 keto reductase has been suggested to reverse the metabolism of PGF into PGE 2 and regulat ing the ratio of these two hormones. Its activity is significantly higher in both the fet al and maternal compartments of placentomes obtained from cows diagnosed with RFM than in normal healthy cows (Kankofer and Schmerold, 2002). The bovine placentome is a target organ for the steroid hormones progesterone and estrogen, and receptors for the se hormones have been identified. Placental steroid hormones help to regulate placental growth and differentiation (Hoffman and Schuller, 2002). Their biosynthesis in trophectodermal cells requires the transfer of cholesterol

PAGE 83

83 fro m the outer to the inner mi tochondrial membrane, where pregnenolone is synthesized. Under the influence of fetal cortisol, the concentration of which increases hydroxylase progresses to convert placen tal progesterone to estrogen The resulting increase in the ratio of estrogen to progesterone is responsible for increasing the rate of synthesis and/or release of utero stimulatory hormones such as prostaglandin and oxytocin, which act through cell mediated pathways to increase intr acellular calcium and activate myometrial contractile fibers (Takagi et al., 2002). Boos et al. (2000) showed tha t the immunoreactivity of both estrogen and progesterone receptors in placentomes tended to be l ess in cows that expelled the placental membran es spontaneously than in cows that retained them. These authors suggested that a lower steroid hormone receptor status in the placentome 40 to 50 hours before parturition was crucial for normal placental separation to occur. Higher immunoreactivity scores in cows with retained fetal membranes, particularly in the maternal crypt epithelium, suggested a degree of immaturity in tissue that correlated with an increase in the number of epithelial cells undergoing apoptosis after calving (Boos et al., 2000) In a mature feto maternal unit, this process should have been completed at or immediately after calving. Binucleate cells express TIMP 2 and, with a reduction in the number of binucleate cells as parturition approaches, less is produced at the end of a normal pregnancy (Walter and Boos, 2001). Takagi et al. (2008) suggested that TIMP 2 activity and MMP 9 expression may be influenced by placental progesterone. During pregnancy, both are necessary to maintain the structure and function of a healthy feto maternal unit by preventing epithelial separation within the placentome, a role similar to

PAGE 84

84 their role in the cervix and corpus luteum of other mammals. Furthermore, enzyme activity controlling placental steroidogenesis continues after parturition, and results in si gnificantly increased expression of MMP 9 mRNA in the cotyledons of cows with RFM. In summary, although many factors may be involved with occurrence of RFM, a clear predominant mechanism remains unidentified. The etiology of metritis and endometritis remai ns partially elusive, but it has been always highly linked with microbial infections occurring primarily at time of parturition. The major pathogens commonly present and associated with uterine diseases are E coli and T pyogenes followed by a range of an aerobic bacteria such as F necrophorum, Prevotella melaninogenicus, and opportunistic bacteria such as Pseudomonas spp., Streptococcus spp., and Staphylococcus spp. that were identified in a variety of combinations using conventional cultures methods (She ldon et al., 2002; Williams et al., 2005; Santos et al., 2010c; Santos et al., 2010d). Many studies suggest that early postpartum uterine infection with E. coli paves the way for subsequent infection with other bacteria or viruses (Dohmen et al., 2000; Do nofrio et al., 2008; Bicalho et al., 2011). Moreover, E. coli infection during the first week postpartum is associated with negative effects on the ovary, hypothalamic pituitary axis, general health, as well as uterine disease (Williams et al., 2007). Rece ntly many studies have exploited the molecular and epidemiological characterization of bovine uterine E. coli (Silva et al., 2009, Sheldon et al., 2010, Bicalho et al., 2010). Silva et al. (2009) report ed the genomic and phenotypic characteristics of 72 E. coli isolates recovered from the uterus of dairy cows with normal puerperium or metritis and evaluated 15 E. coli virulence factor genes identifying none associated with uterine

PAGE 85

85 disease. Sheldon et al. (2010) identified 114 uterine E. coli isolates in dai ry cows with or without metritis in the first 4 weeks postpartum and investigated the presence of 17 virulence factor genes and only the virulence factor ferric yersiniabactin uptake gene (fyuA) was found to be associated with uterine disease. Sheldon et a l. (2010) also suggest ed that endometrial pathogenic E. coli were expressing the type I fimbriae (fimH) gene, because mannose treatment of E. coli decreased their ability to adhere to endometrial cells. Bicallho et al. (2010) used multiplex PCR protocols t o screen the isolates for the presence of 32 virulence factor genes in cows with and without uterine diseases. Six virulence factors, common to extra intestinal and entero aggregative E. coli were found to be associated with metritis and endometritis: fimH hemolysin (hlyA), cytolethal distending toxin (cdt), capsule K2 and K5 (kpsMII), invasion brain endothelium A (ibeA), and arginine succinyltransferase (astA). The virulence factor gene fimH was the most prevalent and the most significant. The virulence factor fimH has been described as critical factor for bacteria adhesion on epithelial cells in models of urinary tract infection in mouse and humans and fimH immunization in a murine model prevented in vivo colonization of the bladder mucosa by 99% (Lange rmann et al., 1997). In bovine metritis, recent research indicates that the infected uterus is predominated by E. coli in the first week postpartum, which alters the intrauterine environment to support future infection by other opportunistic anaerobic bact eria starting in the second week postpartum (Dohmen et al., 2000). In another study, Bicalho et al. (2012) reported that E. coli fimH was associated significantly with metritis and endometritis when detected at 1 to 3 days postpartum.

PAGE 86

86 Trueperella pyogenes is considered one of the most relevant pathogens involved in uterine diseases, especially endometritis. This is due to its relative high prevalence in the environment, persistence in the uterus, severity of lesions on the endometrium, resistance to treatm ent, and synergistic action with gram negative anaerobes (Ruder et al., 1981; Bonett et al., 1991; Huszenicza et al., 1999; Mateus et al., 2002a, b; Williams et al., 2005). The strains of T. pyogenes isolated from the uter i of cows all expressed the virule nce factor pyolysin (plo), which encodes a cholesterol dependent cytotoxin (Jost and Billington, 2005; Silva et al., 2008). Cholesterol dependent cytotoxin molecules are attracted to cholesterol rich domains in cell membranes, where they aggregate to form a pore leading to osmotic death of the cell, and pyolysin readily kills endometrial epithelial and stromal cells in vitro (Miller, 2009). Intrauterine infusion of live T. pyogenes on day 3 after ovulation induced a peak of PGF metabolite 3 days later, f ollowed by the regression of the newly formed CL and ovulation of the dominant follicle of the first follicular wave in 50% of the cows (Kaneko and Kawakami, 2008, Kaneko and Kawakami, 2009). A later study infused T. pyogenes 6 times, every 3 days from day s 3 to 18 after ovulation inducing early demise of the CL 5 of 12 cows with a sharp rise of PGF metabolite on day 6 after ovulation (Kaneko et al., 2013). In the same way, culture of endometrial cells with a bacteria free filtrate of T. pyogenes induced synthesis of PGF (Miller et al., 2007). It has been suggested that inflammation induced by pe ptidoglycan gram positive cell wall, such as the one present in T. pyogenes might induce release of pro inflammatory cytokines such as tumor necrosis fa c tor IL

PAGE 87

87 which could stimulate endome trial synthesis of PGF (Davidson et al., 1995; Hansen et al., 2004; Skarzynski et al., 2000). A genomic characterization of T. pyogenes was conducted to characterize field isolates recovered from the uterus of cows with and without clinical metritis, in an attempt to identify factors that might be associated with the establishment and persistence of the disease (Silva et al., 2009). Eight virulence factor genes plo, neuraminidase P and H (nanP, nanH), collagen binding protein A (cbpA), Type I fimbriae A, C, E and G (fimA, fimC, fimE, fimG) wer e used in this characterization. H owever none of them were related with development of metritis, suggesting that the type of T. pyogenes may not be a determinant factor in the occurrence of the disease. It was suggest ed that host intrinsic factors, the synergism between T. pyogenes and other bacteria, and the differential gene expression of virulence factor genes may play a more relevant role in the establishment of puerperal uterine infections (Silva et al., 2009). I n recent years major advancements were made on understanding the pathogenesis of E. coli lipopolysaccharides (LPS) in uterine innate immunity of dairy cows using endometrial cells explants and granulosa cells and the severity of metritis has been linked to its mechanism of pathogenicity (Sheldon et a., 2010, Cronin et al., 2011, Sheldon and Bromfield, 2011). Studies showed that LPS are major component s of the outer membrane of gram negative bacteria that can lead to endotoxic shock, sepsis and death (Bryant et al., 2010). The endotoxin LPS is one of the major pathogen associated molecular pattern (PAMPs) molecules that is identified by the pathogen recognition receptor (PRR) toll like receptor 4 (TLR4) in bovine epithelial and stromal endometrial cells, gran ulosa cells and professional immune cells (Cronin et al., 2011;

PAGE 88

88 Sheldon and Bromfield, 2011). When LPS binds to TLR4 it forms a heterodimer complex with co receptors cluster of differentiation 14 (CD14) and myeloid differentiation factor 2 (MD2). LPS bindi ng activates the signaling molecule myeloid differentiation factor 88 (MYD88) inducing the phosphor ylation of extracellular signal regulated kinase 1 and 2 (ERK1 and ERK2), p38 mitogen activated protein kinase (p38 MAPK) and nuclear translocation of nuclea binds DNA motifs and transcriptional regulators leading t o production of X C mot if) ligand 1 (CXCL1), chemokine (C C motif) ligand 20 (CCL20), and IL 8 (Cronin et al., 2011; S heldon and Bromfield, 2011). Although LPS TLR4 inflammatory signaling pathway has been shown consistently in bovine endometrial cells (Sheldon et al., 2010; Cron in et al., 2011), hitherto there is lack of investigation of molecular LPS mediate d mechanisms in vivo on dairy cows showing how metritis really develops. Infection of the postpartum uterus is common in dairy cows and becomes a burden to uterine health by damaging the endometrium and causing cows to become systemically ill. The prevalence of E. coli in the first week postpartum is high and reported in healthy cows as well (Bicallho et al., 2010). Thus, in spite of presence of sufficient LPS to induce an inf lammatory response, uterine inflammation, endotoxic shock, and sepsis does not necessarily occur in dairy cows. Another potential component of E. coli pathogenicity is the virulence factor fimH that was recently shown to be highly prevalent in cows in th e first week postpartum, and it was associated with increased prevalence of clinical metritis and endometritis (Bicalho

PAGE 89

89 et al., 2010; Bicalho et al., 2012). The virulence factor fimH, the adhesion portion of type 1 fimbriae produced by most uropathogenic E coli is a conserved protein involved in bacterial attachment to mucosal epithelial cells (Connel et al., 1996). The fimH protein generally binds to mannose receptor on epithelial cells leading to colonization of mucosa by E. coli (Langermann et al., 199 7). Moreover, fimH can activate the innate immune system through TLR4 and its signaling molecule MYD88 in the genital mucosa leading to increased influx of PMNL cells in the urinary tract (Ashkar et al., 2008). The role fimH may play on the pathogenicity o f E. coli contaminating the uterus of postpa rtum dairy cows remains unknown. H owever, the high prevalence of E. coli carrying fimH associated with uterine diseases suggest that it likely plays a role in the development of metritis and endometritis and dese rves further investigation. The general structure of bacterial LPS consists of a hydrophobic lipid A domain, Withfield, 2002). The lipid A moiety alone is sufficient to activa te the innate immune response; adaptive (antibody) responses are generated to the O antigen polysaccharide later in the course of an infection. Lipid A consists of a diglucosamine diphosphate headgroup that is substituted with a variable number of acyl cha ins, ranging from 4 to 8. E. coli lipid A contains a diglucosamine diphosphate headgroup and six acyl chains. In general, such hexa acyl lipid A molecules are powerful immuno stimulants. Changes in the number of acyl chains and in the phosphorylation status of the headgroup can have a profound influence on the biological activity of lipid A. The synthetic compound of lipid A eritoran (also known as E5564) has four acyl chains and act an antagonist of TLR4 in all species investigated so far (Christ et al; 199 5; Figueiredo et al., 2008). Eritoran

PAGE 90

90 acting as TLR4 antagonist can block the excessive immune reaction triggered by this receptor. Therefore, if LPS and fimH are involved in the pathogenicity of uterine disease through TLR4, blocking the receptor might be come an alternative method to minimize the risk of metritis. An interesting study revealed that granulosa cells responded acutely to LPS with rapid phosphorylation of TLR signaling components, p38 and ERK, and increased expression of IL6 and IL8 mRNA, alt hough nuclear translocation of p65 was not evident (Bromfield and Sheldon, 2011) Additionally, TLR4 was targete d with small interfering RNA leading to attenuated granulosa cell accumulation of IL 6 in response to LPS. Moreover, LPS stimulated IL 6 secreti on and expansion by cumulus oocyte complexes and increased rates of meiotic arrest and germinal vesicle breakdown failure. In summary, LPS elicits innate immune response via TLR4 pathway, which ultimately compromise meiotic competence (Bromfield and Sheldo n, 2011). Another recent study investigated and tested the hypothesis that LPS perturbs the development of primordial ovarian follicles revealing that of bovine ovarian cortex ex vivo exposure to LPS reduced the primordial follicle pool associated with inc reased p rimordial follicle activation. Key intracellular regulators of follicle activation were modulated by LPS exposure with loss of the primordial follicle phosphatase and tensin homolog (PTEN) and cytoplasmic translocation of forkhead family transcript ion factor (FOXO3). In the same study acute exposure of mice in vivo to LPS also reduced the primordial follicle pool associated with increased follicle atresia and the increased follicle atresia was TLR4 dependent. In conclusion, LPS reduced the primordia l ovarian follicle

PAGE 91

91 pool in the bovine ovarian cortex ex vivo and in the murine ovary in vivo which ultimately may hinder later fertility (Bromfield and Sheldon, 2013) Although the understanding of how pathogens subvert host defenses and lead to developmen t of uterine diseases evolved considerably in the last few year s a thorough mechanism remains elusive. One of the major factors hampering a better understanding the etiology of uterine diseases is a limited knowledge of the uterine microbiome. C urrent kno wledge is based on the understanding of microbial communities involved in the pathogenesis of metritis and endometritis identified by traditional methods of culture, which was seriously hindered by the fact that more than 99% of the environmental microbes are not amenable to culture under standard laboratory conditions (Aman et al., 1995; Handelsman, 2004). Recently, results using DNA high throughput pyrosequencing of uteri ne fluid samples of healthy, metritic and endometritic cows at 3 day intervals after calving revealed that the core bacterial community was different in healthy cows, when compared to cows suffering from uterine diseases. Furthermore, the phylogenetic diversity in all the combined samples changed gradually ov er time, particularly at the 3 4 to 36 days postpartum, and the core community seemed to be specific for each health status (Santos and Bicalho, 2012). Although these results were enlightening, many limitations such as small sample size, cows from one single farm, sample collection in s hort period of time and use of antibiotics for treatment of metriti s require careful consideration before any generalization are made. Thus, further investigation comparing environmental microbiota present in different farms, immunological status of the co ws and the individual immune response to pathogenic and opportunistic bacteria is critical to characterize molecular pathogenesis of persistent

PAGE 92

92 uterine diseases and identify entry points to development of new preventatives, prognostic tools and effective t reatments. T herapy for Uterine Diseases Retained fetal membrane was recognized as a clinical problem occurred the first time almost two centuries ago (Knowlson, 1834). The first description discussed the manual removal of fetal membranes as an issue for ca ttle. Knowlson (1834) described sighted man nowlson 1834). Therefore, it has been health. Knowlson (1834) reported that the first treatment for retained fetal membranes included 1 oz of spe rmaceti, 1 oz of gum myrrh, 2 oz of juniper berries, 2 oz of bay berries, 1 oz of round birthwort root and 1 oz of galangal ground up with a pestle and mortar and administered to the cow with three pints of cold ale. Alternatively, the same basic formula could be given with a quart of warm gruel and a wine glass of gin or brandy (Clater 1839). Throughout the last 200 years, much has been learned about the etiology and risk factors for RFM, and several therapies and prevention were attempted most of them with relative ly no success. Although manual removal remains a common practice, many studies failed to show any benefit of this approach on reproductive performance or milk production (Bolinder et al., 1988; Kulasekar et al., 2004; Drillich et al., 2006a; Drillich et al., 2007). In fa ct, manual removal can result in more frequent and severe uterine infections, when compared with more conservative treatment (Bolinder et al.,1988). Manual removal prolonged the interval from calving to cyclicity by 20 days and presence

PAGE 93

93 of intrauterine pat hogenic bacteria was 100% in cows with manually removed RFM versus only 37% of untreated cows at 3 weeks postpartum, and further 37% of treated versus 12% of untreated cows at 5 weeks postpartum (Bolinder et al.,1988). The possible explanation is that re mo val of the placenta can cause damag e to the endometrium, suppress uterine leukocyte phagocytosis and leave behind necrotic po rtions, which together encourage bacterial invasion (Peters and Laven 1996). While research does not support manual removal as an effective treatment for RFM, it is still commonly used both because of aesthetic benefits, including parlor hygiene and removal of offensive odors, and perceived but not realistic benefits that removing the placenta eliminates a potential source of infecti on and reduces likelihood of uterine diseases. Results of use of antimicrobial to treat RFM are controversial (Peters and Laven, 1996). Cows diagnosed with RFM are more likely to develop metritis and the claim by the same researchers behind the use of ant ibiotics to treat RFM is to prevent or mitigate possible severity of metritis and its subsequent negative effects on fertility. Intrauterine antimicrobials, given as infusions or boluses, were unable to reduce the incidence of metritis or improve fertility (Peters and Laven, 1996). Drillich et al. (2007) used t w o strategies to treat RFM. I n the first cows with RFM and fever received 1 mg/kg of ceftiofur systemically for 3 to 5 consecutive days, whereas cows with RFM and no fever remained untreated. In seco nd treatment strategy, all cows with RFM were treat ed with 6 g of tetracycline administered into the uterus for 3 days and RFM cows with fever received additional 10 mg/kg of amoxicillin systemically. Although treatment w ith intrauterine antibiotics lower ed the incidence of postpartum fever, no differences were

PAGE 94

94 found among treatment groups in terms of milk yield or reproductive performance (Drillich et al., 2007). Goshen and Shpigel (2006) assessed if treating cows with RFM and metritis with intrauterine chlortetracycline would influence reproductive performance and milk production. Benefits in reproductive performance and milk production only occurred in cows for metritis, and no difference in either milk yield or reproductive performance were found betwe en treated and untreated RFM cows. These results indicate that although intrauterine antibiotics can benefit cows with metritis they are unlikely to cause earlier release of membranes or prevent metritis in cows with RFM. Another possibility explored by researchers was that intrauterine antibiotics might control local bacterial growth and interfe re with the necrotizing process that is responsible for the eventual release of RFM (Roberts, 1986). Oxytetracycline, which is often used for intrauterine treatme nt in cattle with RFM and metritis, inhibit MMPs important for endometrial repair in other species, which potentially could interfere with the normal placental detachment mechanisms (Eiler and Hopkins, 1993). Systemic antibiotics are believed to be benefic ial in RFM cases with concurrent metritis (Risco and Herna n dez, 2003; Drillich et al., 2006a; LeBlanc et al., 2008). Combining systemic with intrauterine antimicrobials to treat RFM d id not improve efficacy of the trea tment (Drillich et al., 2006a). Curren tly there are no studies on which cows with concurrent RFM and fever were left untreated, therefore, it is unclear whether the resolution of fever is caused by

PAGE 95

9 5 Treating all RFM cows with systemic ceftiofur irrespective of temperature was not beneficial to reduce occurrence of fever, increase shedding of RFM, or enhance subsequent reproductive performance, in comparison with selective antibiotic treatment of cows only presenting fever (Dri llich et al., 2006b). The results of this study suggest that treating only cows with RFM and fever can substantially reduce unnecessary antimicrobial use (Drillich et al., 2006b). Treatment of cows with RFM for 5 days with 2.2 mg/kg of ceftiofur hydrochlor ide systemically was beneficial in preventing metritis when compared with estradiol cypionate or no treatment; however, no significant subsequent improvements in reproductive performance were identified (Risco and Hernandez, 2003). Another controversial to pic regarding RFM is the use of horm ones to aid on release of membranes Oxytocin and PGF are the most commonly used hormones in cases of RFM. Although these ho rmones are powerful inducers of uterine contraction, it is thought that uterine atony accounts for a very small percentage of cases of retained placenta cases (Laven and Peters, 1996) and several studies have not supported their use as a general treatment for RFM (Stevens and Dissimore, 1997, Drillich et al., 2005). A study suggested that use of c ollagenase, an enzyme capable of breakdown collagen, might aid detachment of the caruncle cotyledon bond in cows with RFM (Eiler and Hopkins, 1993). The umbilical arteries of cows with retained placenta were injected with 200,000 IU of bacterial collagenas e leading to earlier placental release than untreated herd mates. When applied within 24 to 72 hours after calving, collagenase le d to the release of membranes in 85% of the cases within 36 hours, whereas none of the 24 control cows released their membrane s within this time period. Although collagenase

PAGE 96

96 therapy shows promise as an option of treating retained placenta the cost is high (~ $75.00). Unfortunately, no s tudies have evaluated dosage regimens and long term production and reproduction outcomes of co llagenase treatments in cows with RFM to determine if collagenase treatments of RFM would be an alternative for therapy of RFM Cows diagnosed with metritis develop moderate to severe illness, therefore there is a consensus that most of metritis cases req uire systemic antibiotic treatment (LeBlanc, 2008). Sheldon et al., (2004) reported that 1 mg/kg of ceftiofur maintains therapeutic conce ntrations in uterine tissues against E. coli (Sheldon et al., 2004), but the same results were not observed for all cow s in other studies (Okker et al., 2002; Drillich et al., 2006c). Currently, the most common treatments of choice for metritis are ceftiofur, given at dosage of 2.2 mg/kg IM once a day, or procaine penicillin, given at the dosage of 21,000 IU/kg IM once or twice a day for a period of 3 to 5 days (Smith et al., 1998; Drillich et al., 2001, 2006a; Chenault et al., 2004). Other alternative treatments reported with efficacy similar to ceftiofur or penicillin to treat metritis include systemic use of tetracyclin e at a dosage of 10 mg/kg, systemic use of ampicillin at a dosage of 11 mg/kg and intrauterine treatments with oxytetracycline and ampicillin (Schmitt et al., 2001, Smith et al., 199 8; Drillich et al., 2003, 2006b) Although tetracycline at a dosage of 10 mg/kg was an effective treatment for metritis (Schmitt et al., 2001), it did not achieve therapeutic concentrations in uterine tissues (Bretzlaff et al., 1983). Uterine concentration of ampicillin in dairy cows diagnosed or not with metritis has not been reported in the literature. Addition of one dose of flunixin meglumine does not improve outcomes over the use of systemic antibiotics alone (Drillich et al., 2007).

PAGE 97

97 A recent a large field study investigate d the efficacy of a 2 do se regimen treatment 72 ho urs a part of ceftiofur crystalline free acid (CCFA) sterile suspension given at 6.6 mg/kg s.c. in the base of the ear, a long acting ceftiofur formulation, to treat puerperal metritis in dairy cows (McLaughlin et al., 2012). C linical cure was greater for c ows treated with CCFA than untreated control cows (74.3 vs. 55.3%) and average rectal temperatures were lower for CCFA than control suggesting that CCFA was an effective treatment for acute metritis in dairy cows (McLaughlin et al. 2012). A second study with CCFA investigate d the efficacy of a single treatment to prevent or reduce incidence of metritis in high risk dairy cows those having dystocia, twins, stillbirth, or RFM (McLaughlin et al., 2013). The use of CCFA decreased the incidence of subsequent metritis and lowered rectal temperature for the first 2 days after treatment, but had no improvements in reproductive performance or milk produc tion (McLaughlin et al., 2013). von Krueger et al. (2013) investigated concentrations of ceftiofur derivatives i n serum, endometrial tissue, and lochia of cows with fever postpartum or metritis from 4 to 6 d after treatment with a single dose of 6.6 mg of CCFA. The results of this study revealed that mean concentrations of desfuroylceftiofuracetamide, an act i ve meta bolite of ceftiofur, detected were above the reported minimum inhibitory concentrations required to inhibit relevant pathogens such as E coli and T. pyogenes in serum on days 4, 5 and 6 and in endometrial tissue and lochia only on d 4 in CCFA treated cows These results support the concept that one single treatment with 6.6 mg/kg of CCFA is no t sufficient to efficaciously treat metritis if the disease is not resolved within 4 days (von Krueger et al., 2013).

PAGE 98

98 Lately the focus of some studies ha s been on t he development of effective preventatives. Machado et al. (2012) evaluated the effects of intrauterine administration of mannose or a bacteriophage cocktail and the presence of E. coli and T. pyogenes in the uterine lumen on uterine health and reproductive performance of lactating dairy cows Their results revealed no effects on uterine health, reproduction performance, or responses in cultures for E. coli and T. pyogenes The rationale behind this study was th at mannose is a fimH antagonist that potentiall y could lead to elimination of pathogenic mechanism of fimH and consequently prevent endometrial colonization by pathogenic E. coli Indeed, King et al. (1998) reported that mannose might be effective in reducing bacterial infection in the equine endometri um. Secondly, bacteriophages are viruses that infect bacteria being obligate intracellular parasites without their own metabolis m and could potentially parasitize E. coli and other bacteria mitigating their potential impact on development of metritis. An i mportant factor related to treatment of metritis is the cost benefit to producers. A case can be ma d e that treatment of metritis is justified to improve cow welfare and reduc e the probability of death in severe cases. However, the criteria to measure effic acy of treatments are not consistent. In an ideal world, the aim would be to return cows to their normal level of production without any further issues. However, what recent studies demonstrated was an expected reduction of body temperature and remission o f fever in approximately 70% with an improvement but not complete resolution of fetid uterine discharge (Chenault et al., 2004 ; McLaughlin et al., 2012 ). LeBlanc et al. (2008) discuss ed the limited data on the efficacy of treatment of metritis for preven tion of subsequent related diseases; or for improvement of milk production, or

PAGE 99

99 improvement of reproductive per formance. Moreover, it is not known if aggressive programs aiming for early diagnosis and treatment of metritis can prevent progression to severe disease and losses in performance, or whether they result in treatment that is not medically or economically beneficial. The new trend suggests that incorporation of e screening tests to select cows for further examination. Cows diagnosed with RFM needs to be observed daily for potential progression to metritis until the placenta is released. Clearly, further research using large scale field studies are necessary to es tablished better criteria for early treatment of metritis. Endometritis treatments reported include systemic or intrauterine administered antibiotics, intrauterine substances and systemic use of PGF A plethora of studies evaluated use of intrauterine su bstances to treat endometritis including tetracycline (Thurmond et al., 1993; Sheldon and Noakes, 1998), penicillin (Thurmond et al., 1993), chloramphenicol (Steffan et al., 1984), cephapirin (Dohmen et al., 1995; McDougall, 2001; LeBlanc et al., 2002b), g entamycin, spectinomycin, sulfonamides, nitrofurasone (Gustafsson, 1984; Gilbert and Schwark, 1992) and non antimicrobial substances such Schwark, 1992), enzymes (Drillich et al., 2005) and hypertonic dextrose (Brick et al., 2012). With exception of the cephapirin in two different studies (McDougall, 2001; LeBlanc et al., 2002b) and hypertonic d extrose (Brick et al., 2012), all other studies had negligible to no benefits on reproductive performance and suffered f rom many issues such as lack of negative controls and statistical power, diagnostic criteria for endometritis that were not validated as having an impact on r eproductive performance,

PAGE 100

100 and no label approv al for intraut erine use with no published information on withdrawal times. LeBlanc et al., (2002b) compared the effect of intrauterine administration of 500 mg of cephapirin benzathine or intramuscular administration of PGF as cloprost enol sodium on time to p regnancy in dairy cows diagnosed with clinical endometritis between 20 and 33 postpartum. No benefits of treatment between 20 and 26 days postpartum were observed, but cows treated with cephapirin between 27 and 33 days postpartum had a 60% increase in th e hazard of pregnancy than untreated controls. No benefits were observed for cows treated with PGF (LeBlanc et al., 2002b). McDougall (2001) randomly allocated cows with risk factors for uterine diseases to receive cephapirin intrauterine or remain untreated at 41 14 days postpartum. Cows that delivered a dead calf, had RFM, or had purulent dischar ge from the vulva observed after 13 DIM and then treated with cephapirin were approximately 2 to 3 times more likely to become pregnant by 56 days into the breeding season when compared to untreated cows (McDougall, 2001). Brick et al. (2012) assessed the c ure rates and P/AI in cows with clinical endometritis in cows treated with intrauterine infusion of a hypertonic solution of 50% dextrose or subcutaneous CCFA. Cows diagnosed with clinical endometritis treated with dextrose tend to have increased P/AI (29. 8%) than untreated control cows (21.1%) and cows rece iving subcutaneous CCFA (19.7%). In a different approach, using a non antibiotic substance, Drillich et al. (2005) evaluated the efficacy of proteolytic enzymes to treat chronic endometritis in comparis on with PGF The product used in the study contained the enzymes chymotrypsin (16 mg), trypsin (16 mg), and papain (8 mg), and additionally 200,000 IU

PAGE 101

101 of retinol palmitat e (vitamin A) and 240 mg of a tocopherol acetat e (vitamin E). Although no differences in cure r ates were identified between the proteolytic enzymes and PGF conception rate to all services for cows with endometritis was higher in cows treated with PGF than in cows treated with proteolytic enzymes. One of the most common treatments investigated f or endometritis to date is PGF R esults regarding the effects of PGF on clinical and subclinical endometritis and subsequent reproductive outcomes are controversial. LeBlanc et al. (2002a) reported that administration of PGF between 20 and 26 days po stpartum to cows with endometritis without a CL was associated with a significant reduction in pregnancy rate, and no differences in pregnancy rates were observed in cows treated with PGF or not between 27 and 33 days postpartum. Kasimanickam et al. (200 5) reported that a single treatment with PGF between 20 to 33 days postpartum in cows diagnosed with subclinical endometritis or not improved P/AI to the first service and media n days open. Moreover, cows with subclinical endometritis treated with PGF had a significantly increased hazard rate to pregnancy com pared to control. Galvo et al. (2009a) reported that cows treated with PGF on days 21, 35 and 49 postpartum had no benefits on prevalence of subclinical endometritis and time to first AI, but inc reased P/AI to the first service. Kaufmann et al., (2010) compared effects of PGF with ceftiofur to treat cows diagnosed with clinical endometritis revealing no differences in AI submission rate, days to first service, first service conception rate, days open and proportion of cows pregnant. Dubuc et al. (2010) reported that treatment with PGF at 35 and 49 days postpartum did not affect the probability of cure of clinical and subclinical endometritis irrespective

PAGE 102

102 of cyclic status and did not mitigate th e negative effects of clinical and subclinical endometritis on reprod uctive performance.

PAGE 103

103 CHAPTER 3 EFFECT OF ONE OR THR EE TIMED ARTIFICIAL INSEMINATIONS BEFORE NATURAL SERVICE ON R EPRODUCTIVE PERFORMA NCE OF LACTATING DAI RY COWS NOT OBSERVED FO R DETECTION OF ESTRUS The objectives of this study were to determine the effects of one or three timed AI before natural service (NS) in lactating dairy cows not observed for detection of estrus on hazard of pregnancy, days nonpregnant, and 21 d cycle pregnancy rate A total of 1,050 lactating Holstein cows were subjected to a double Ovsynch program for first postpartum AI. On the day of first AI (78 3 d in milk), cows were blocked by parity and randomly assigned to receive either one timed AI (1TAI, n = 533) or th ree timed AI (3TAI, n = 517) before being exposed to NS. Cows assigned to 1TAI were exposed to bulls 7 d after the first AI. Nonpregnant cows in 3TAI were resynchronized with the Ovsynch protocol twice, with intervals between AI of 42 d, before being expos ed to NS 7 d after the third AI. Cows were evaluated for pregnancy 32 d after each timed AI or every 28 d after being exposed to NS. Pregnant cows were re examined for pregnancy 28 d later (i.e., 60 d gestation). Exposure to heat stress was categorized bas ed on the first AI being performed during the hot or cool season according to the temperature and humidity index. Body condition was scored at first AI. All cows were allowed a period of 231 d of breeding, after which nonpregnant cows were censored. Pregna ncy to the first AI did not differ between 1TAI and 3TAI on Day 60 after insemination (30.8 vs. 33.5%). Cows receiving 3TAI had a 15% greater hazard of pregnancy and a 17% greater 21 d pregnancy rate than 1TAI and these benefits originated from the first 8 4 d of breeding. These changes in rate of pregnancy reduced the median and mean days nonpregnant by 9 and 10 d, respectively. Despite the long inter AI interval in cows subjected to 3TAI, reproductive performance was improved compared with a single timed A I and

PAGE 104

104 subsequent exposure to NS. In dairy herds that use a combination of AI and NS, allowing cows additional opportunities to AI before onset of breeding with bulls is expected to improve reproductive performance. Introductory Remarks Inadequate or inacc urate detection of estrus are critical factors responsible for poor reproductive performance in dairy cows (Senger, 1994; Roellofsa et al., 2010). Timed artificial insemination (AI) and natural service (NS) are common methods to manage reproduction in dair y herds in the United States (Champagne et al. 2002, NAHMS, 2002; Smith et al. 2004; De Vries et al., 2005; Caraviello et al., 2006; Lima et al., 2009) and can be successfully used without detection of estrus (Lima et al., 2009). Although AI hastens geneti c progress, controls venereal diseases, and provides a safer environment for cows and farm personnel, breeding programs relying primarily on NS are still widely used by dairy producers. Several studies and surveys conducted in different regions of the US s howed that 43% to 84% of the dairy farms use NS either alone or combined with AI States (Champagne et al. 2002, NAHMS, 2002; Smith et al. 2004; De Vries et al., 2005; Caraviello et al., 2006). In many cases, NS is incorporated into breeding programs after cows have received one or more AI, which is commonly not to use detection of estrus and still incorporate AI, continuous synchronization of ovulation for insemination at fixed time is an option (Lima et al., 2009). Timed AI has been shown to be an economical option to manage reproduction in high producing dairy cows that experience a reduction in estrous intensity (Lima et al., 2010; Risco et al., 1998).

PAGE 105

105 The reproducti ve efficiency (Lima et al., 2009) and costs (Lima et al., 2010) of cows bred by timed AI or NS have been directly compared and, although cows exposed to NS only showed minor improvements in reproductive performance (Lima et al., 2009), the economic evaluat ion favored those receiving timed AI (Lima et al., 2010). The economic advantage of timed AI was even greater when genetic progress was considered, and when marginal feed cost and milk price increased. Interestingly, the 21 d cycle pregnancy rate, a common metric used to evaluate reproduction in dairy herds, was similar between cows bred by timed AI or NS (25.0 and 25.7%, respectively), and they were both superior when compared with average values for high producing dairy herds that ranges from 15.0 to 17.9 % (De Vries et al., 2005; LeBlanc, 2010). The improvement in hazard of pregnancy for NS was attributed to a greater number of breeding opportunities, as NS cows were exposed continuously to bulls. In contrast, timed AI cows could only be inseminated after diagnosed nonpregnant, which created inter AI intervals of 35 d (Lima et al., 2009). It is possible that a combination of timed AI and NS might benefit reproductive performance of dairy cows not observed for detection of estrus, as timed AI allows for all cows to be inseminated on the first day past the voluntary waiting period and exposure to NS after that will likely shorte n the interval between breedings In many dairy farms using a combination of AI and NS, cows initially are inseminated one or more ti mes and then moved to bull breeding groups (Overton and Sischo, 2005); however, it is unclear how many inseminations cows should receive before exposed to bulls to maximize pregnancy rate. This is particularly important in herds managing reproduction witho ut the aid of estrous detection, as the interval

PAGE 106

106 between inseminations is determined by when a cow can be resynchronized for AI. In a previous study, the benefit of NS over TAI was only observed after 150 d postpartum, when timed AI cows had already receiv ed three inseminations (Lima et al., 2009). Therefore, it is plausible to suggest that three timed AI may result in a similar reproductive performance when compared with one timed AI, despite the long inter insemination interval. In fact, Overton and Sisch o (2005) concluded that in herds using both AI and NS, allowing cows more opportunities for AI may benefit reproduction. The hypothesis of the current study was that cows subjected to three sequential timed AI would have similar hazard and time to pregnan cy, and 21 d cycle pregnancy rate compared with cows exposed to NS 7 d after the first postpartum timed AI. Therefore, the objective of this study was to determine the effect of one or three timed AI followed by NS on reproductive performance of lactating dairy cows not observed for detection of estrus. Materials and Methods Cows, Housing and Diets All procedures performed during this study were approved by the University of Florida Institutional Animal Care and Use Committee. The study was conducted betw een July of 2009 and October of 2010 in a commercial dairy farm milking approximately 2,000 Holstein cows during the study period and located in north central Florida, USA. Cows were housed in free stall barns equipped with fans and sprinklers for forced e vaporative cooling during the hot season. Primiparous and multiparous cows were housed separately. Lactating cow diets were formulated to meet or exceed the nutrient requirements established by NRC (2001) for lactating Holstein cows weighing 650 kg, consum ing 24 kg/d of dry matter, and producing 45 kg/d of milk containing 3.5%

PAGE 107

107 fat and 3.1% true protein. Diet was fed as totally mixed ration and composed of corn silage, annual ryegrass ( Lolium multiflorum Lam.) silage, Tifton 85 Bermudagrass silage, ground co rn, citrus pulp, cottonseed hulls, expeller soybean meal, solvent extracted soybean meal, and a mineral vitamin premix. Cows were milked at least three but no more than four times daily according to the farm milking throughput system that operated continuo usly throughout a 24 h period. Treatments, Exclusion Criteria and Reproductive Programs Cows with uterine adhesions or abscesses displaced abomasum, cesarean section or fetotomy at calving and cows that missed any part of their experimental protocol fo r the first service were not included in the study. After enrollment, cows sold dead or that missed any part of their program were censored on the respective days All cows were enrolled in a double Ovsynch program at 51 3 d postpartum, which was desig nat ed as Day 27 of the study (Figure 3 1). All cows received 100 g of Ltd., Iselin, NJ, USA) on Day 27, followed 7 d later by 25 mg of PGF i.m. (5 mL Lutalyse sterile solution; 5 mg/mL of dinoprost tromethamine; Pfizer Animal Health In c., Madison, NJ, USA) and a second GnRH injection on Day 17. On Day 10, the breeding Ovsynch was initiated with administration of GnRH, followed by PGF on Day 3, and a final injection of GnRH on Day 1, approximately 56 h after the PGF of the breedi ng Ovsynch. On Day 0, 16 h after the final injection of GnRH, all cows received timed AI. All timed AI were performed by 5 technicians using 13 different sires. On Day 0, weekly cohorts of cows were blocked by parity and, within each block, randomly assig ned to one of two treatments: one (1TA I) or three timed AI (3TAI; Figure

PAGE 108

108 3 2). Cows assigned to 1TAI were moved to pens with bulls for NS 7 d after the first insemination. Cows enrolled in the 3TAI treatment remained in the same group and were re synchroni zed up to 2 times following a non pregnancy diagnosis. Resynchronization of ovulation of 3TAI cows was initiated when diagnosed not pregnant with the Ovsynch protocol including a controlled internal drug release (CIDR) insert containing progesterone (Eazi Breed CIDR Cattle Insert; Pfizer Animal Health Inc.) that was present on the days between injections of GnRH and PGF (Figure 3 1). Because of the reproductive program selected for 3TAI cows, the interval between inseminations was 42 d. Seven days after the third AI, cows in the 3TAI treatment were moved to NS pens together with 1TAI cows. Pregnancy was diagnosed by t ransrectal ultrasonography of the uterus and its contents 32 d after the first AI in all cows by visualization of an embryo with heartbeat. Cows with a CL and fluid in the ipsilateral uterine horn, but without a visible embryo or with an embryo without hea rtbeat were considered as not pregnant. Cows diagnosed pregnant were re examined by transrectal palpation of the uterus and its contents 28 d later (i.e., 60 d of gestation) to reconfirm pregnancy and to identify pregnancy loss. The same procedure was used for pregnancy diagnoses after the second and third AI in cows in the 3TAI treatment. For nonpregnant cows exposed to NS, pregnancy was diagnosed by transrectal ultrasonography every 28 d (Figure 3 1) until detected pregnant, or culled, or dead, or until 2 31 d after the first AI (309 3 d postpartum), which was selected as the end of the experiment. Because cows could have been bred by NS on study Day 231, a final pregnancy diagnosis was performed 28 d later, on Day 259 (337 3 d postpartum) and pregnant cows were reconfirmed 28 d later. Two

PAGE 109

109 hundred and thirty one days after the first AI was chosen as the criterion to end the study because it allowed all cows to have eleven 11 complete 21 d estrous cycles. Based on farm records, it was anticipated that mor e than 85% of the cows would be pregnant by 309 d postpartum; therefore, extending the study beyond those days would have little impact on the results of the experiment. The criterion of repeated pregnancy diagnosis every 28 d was chosen to allow an accura te determination of gestation age in pregnant cows exposed to NS when gestation is between 28 and 55 d. Age of pregnancy for cows bred by NS was estimated according to the diameter of the amniotic vesicle (Zemjanis, 1970; Ginther, 1998). Pregnant cows by N S were re examined for pregnancy 28 d after the initial diagnosis. Bull Management Bulls assigned to the study were at least 18 months of age and negative for persistent infection caused by bovine viral diarrhea virus examined by immunohistochemistry of s kin using an ear notch sample. Every bull underwent a breeding soundness evaluation according to the guidelines of the Society for Theriogenology (Chenoweth, 1992) and only those classified as satisfactory breeders were used in the study. In addition, bull s were tested for Tritrichomonas foetus using a smegma sample cultured in a modified diamond media (InPouchTM TF, Biomed Diagnostics, White City, OR, USA). The breeding soundness evaluation and T. foetus test were repeated every 6 months in each bull. All bulls were vaccinated once a year against respiratory diseases and leptospirosis (Bovi Shield GOLD 5 L5, Pfizer Animal Health Inc.), clostridiosis (Ultrabac 8, Pfizer Animal Health Inc.), and campylobacteriosis (Vibrin, Pfizer Animal Health Inc.). The ratio of bulls per nonpregnant cows was maintained at one to twenty. Bulls were rotated every 14 d such

PAGE 110

110 that they remained with cows for 14 d and were allowed to rest for 14 d. When resting, bulls were placed in Tifton 85 Bermudagrass ( Cynodon spp.) past ure, with portable shades and trees for heat abatement. Resting bulls also received orts from lactating cows. Breeding bulls were fed the lactating cow diet ad libitum while in the breeding pens, and received an average of 16.9 kg of dry matter per day. Bo dy Condition Scor ing All cows enrolled had their body condition (BCS) scored at 110 3 d postpartum concurrently with the evaluation of pregnancy diagnosis for the first timed AI. Cows were scored in a one to five scale (1 = emaciated, 5 = obese; Ferguson et al., 1994). Seasonality The temperature (C) and relative humidity (%) data were obtained from the Florida automated Weather Network (http://fawn.ifas.ufl.e du/scripts/reportrequest.asp). The data were collected from July of 2009 to October of 2010. The weather station is located in Alachua, Florida, approximately 48.3 kilometers from the experimental location. Average daily temperature humidity index (THI) wa s calculated as described by the NOAA (1996). Ambient temperature was converted from o C to o F [Temperature in o F = (temperature o C x 1.8) + 32], and the following formula was used to compute the o F 58). The THI was the criterion used to determine effect of season (hot or cool) on reproductive performance. The average daily THI was categorized as cool when THI < 72, or hot

PAGE 111

111 those in which the average daily THI in the week of first AI was < 72 were considered to have not been exposed to heat stress (cool season) and this categorization was used for statistical analyses. Experiment al Design and Statistical Analysis The study followed a randomized complete block design. Weekly cohort of cows were blocked according to parity (primiparous or multiparous) on the day of the first AI and, within each block, randomly assigned to either 1TAI or 3TAI. The sample size was percentage unit diffe rence in the proportion of pregnant cows was observed at any time in the study. It was assumed that 35% of the cows would be pregnant at the first AI in both treatments. Starting at 35% proportion of pregnancy, the sample size was calculated at 5 percentag e unit intervals (35 vs. 41%; 40 vs. 46%; 45 vs. 51%; 50 vs. 56%; etc) to determine the maximum number of cows needed. The maximum number of cows needed per treatment was 543 (when the proportions of cows pregnant were 50 vs. 56%) and the minimum was 246 ( when the proportions of cows pregnant were 85 vs. 91%). The reproductive responses of interest for analyses were the proportion of cows pregnant within the first 21, 42 and 84 d of breeding, the proportion of pregnant cows at the end of the study (Day 231 ), the hazard of pregnancy, median d to pregnancy, and the 21 d cycle pregnancy rate. A cow was considered pregnant only when the reconfirmation of pregnancy 28 d after the initial diagnosis was positive (pregnant on Day 60 of gestation). The pregnancy per AI after the first postpartum insemination was also evaluated to determine if equal proportions of cows became pregnant when assigned to treatments. The 21 d cycle pregnancy rate was calculated in both

PAGE 112

112 treatments based on the number of cows that became pr egnant in an interval of 21 d divided by the number of eligible nonpregnant cows in that 21 d period. Binary responses were analyzed by multivariate logistic regression using the logistic procedure of SAS version 9.2 (SAS Institute Inc., Cary, NC). Binary data with repeated measurements such as 21 d cycle pregnancy rate were analyzed by the GLIMMIX procedure of SAS with cow within treatment as a random effect. The adjusted odds ratios (AOR) and respective 95% confidence intervals (CI) were calculated for b inary responses. Additional analyses for binary data were performed with modified Poisson regression model with the GENMOD procedure of SAS using a log link function and correction for data dispersion (Spiegelman et al., 2005; Fang, 2011). These analyses w ere performed to estimate the adjusted risk ratios (ARR). For the analyses of 21 d cycle pregnancy rates, the repeated statement with an exchangeable correlation matrix was used to cluster cows within treatment to indicate a random effect of cow within tre atment. hazard model using the PHREG procedure of SAS. Cows that left the study either because they were sold or died before study Day 231 were censored. The adjusted hazard ratios (AHR) and respective 95% CI were calculated for time dependent categorical data. Proportionality was assessed by evaluating the Kaplan Meier curves using the LIFETEST procedure of SAS and by including an interaction between treatment and days postpa then interval to pregnancy was partitioned into two periods to accommodate

PAGE 113

113 proportionality. Median and mean days to pregnancy were generated by the Kaplan Meyer method using the LIFETEST procedure of SAS. For all responses analyzed, multivariate models were built and included the effects of treatment, covariates (sire, technician, parity, BCS, and season), and interactions between treatment and covariates. Covariates were sequentially removed from statistical models in a stepwise backward fashion if P > 0.10. Treatment was forced in all final statistical models. Treatment differences with P P 0.10 were considered as a tendency. Results The mean (1TAI = 80.7 0.4 vs. 3TAI = 80.3 0.4 d; P = 0.30) and median (1TAI = 78 vs. 3TAI = 79 d; P = 0.61) days postpartum at first AI did not differ betwe en treatments. Similarly, the mean (1TAI = 3.05 0.02 vs. 3TAI = 3.03 0.02; P = 0.26) and median (1TAI = 3.00 vs. 3TAI = 3.00; P = 0.23) BCS did not differ between treatments. The proportions of 1TAI and 3TAI cows exposed to heat stress, based on receiv ing their first AI during the hot season, were similar (P = 0.62) and were 39.6 and 38.1%, respectively. Pregnancy per AI to the First Timed AI As expected, there were no differences in pregnancy per AI on Days 32 and 60 and on pregnancy loss after the fi rst AI in cows receiving 1TAI and 3TAI (Table 3 1). Primiparous cows were more likely to become pregnant than multiparous at the first AI on Days 32 (primiparous = 63.2% vs. multiparous = 35.2%, P < 0.001) and 60 (primiparous = 55.2% vs. multiparous = 29.0% P = 0.002) after insemination. Pregnancy loss did not differ (P = 0.62) between parities (primiparous = 12.7 % vs.

PAGE 114

114 2.75 = 22.3%) after AI, and were also less likely (P = 0.02) to lose their pregna the hot season were less likely (P < 0.001) to become pregnant at the first AI evaluated on Days 32 (hot = 24.3% vs. cool = 47.7%) and 60 (hot = 19.6% vs. cool = 40.0%) after insemination, but pregnancy loss did not differ with season (hot = 19.2% vs. cool = 16.0%, P = 0.32). Reproductive Performance During the Entire Lactation Of the 331 1TAI cows not pregnant to the first AI, only 9 (2.7%) became pregnant by N S within the next 21 d after the first insemination. This low pregnancy did not improve the proportion of pregnant cows by study Day 21 (Table 3 1) compared with cows in 3TAI that did not have the opportunity to be re inseminated until study Day 42. In fac t, after two potential estrous cycles, by study Day 42, the proportion of pregnant cows was greater (P < 0.01) for 3TAI than 1TAI (ARR = 1.24; 95% CI = 1.09 1.42). A similar response was observed on study Day 84, when cows in 3TAI had the opportunity to re ceive their third AI and had a 20% greater (P < 0.01) risk of being pregnant than cows in 1TAI (ARR = 1.20; 95% CI = 1.09 1.33). When the entire breeding period was analyzed, the rate of pregnancy was greater (P = 0.04) for 3TAI than 1TAI (Figure 3 2). In fact, 3TAI cows had fewer median and mean days to pregnancy (Table 3 2). No interactions between treatment and parity, BCS, or season were observed for interval to pregnancy. Because of lack of parallelism of survival curves between 1TAI and 3TAI in the f irst 84 d of breeding (averag e of 162 d postpartum; Figure 3 2) and interaction between

PAGE 115

115 treatment and day postpartum, therefore, lack of proportional hazard, data also were analyzed separately for before and after study Day 84. For the analyses before stud y Day 84, nonpregnant cows by study Day 84 were censored. For analysis after study Day 84, only cows that became pregnant or censored after that were included. For the 725 cows (336 in 1TAI and 389 in 3TAI) that became pregnant or were censored before stud y Day 84, the hazard of pregnancy was 26% greater (P < 0.01) for 3TAI than 1TAI (AHR = 1.26; 95% CI = 1.07 1.47). However, for the remaining 325 cows (197 in 1TAI and 128 in 3TAI) that became pregnant or were censored after study Day 84, the hazard of preg nancy did not differ (P = 0.30) between 3TAI and 1TAI (AHR = 0.87; 95% CI = 0.66 1.14). Therefore, the increased hazard of pregnancy for 3TAI cows was primarily the result of faster pregnancy attained after during the period of timed AI, before study Day 8 4. Pregnancy rate was also greater (P = 0.01) for 3TAI than 1TAI when analyzed based on 21 d cycles (Table 3 1). Interestingly, the 17% increased relative risk (ARR = 1.17; 95% CI = 1.04 1.31) of becoming pregnant over time for 3TAI compared with 1TAI was similar to the 15% increased hazard observed from the survival analysis. Despite improvements in the rate of pregnancy with 3TAI, the proportion of cows diagnosed pregnant on Day 60 of gestation was not different between 1TAI and 3 TAI at the end of the s tudy (Table 3 1). Treatment altered the proportion of cows becoming pregnant to AI and NS. At the end of the study 30.8% of 1TAI cows and 64.8% of 3TAI cows became pregnant to AI, whereas 49.9 and 16.3% of the 1TAI and 3TAI cows became pregnant to NS, resp ectively.

PAGE 116

116 In addition to treatment, parity, BCS and season also affected the rate of pregnancy. Primiparous cows had greater (P < 0.01) rate of pregnancy than multiparous cow, which resulted in fewer median and mean days to pregnancy (Table 3 2). Similarl season had faster (P < 0.01) rate of pregnancy and fewer median and mean days nonpregnant (Table 3 2). Similarly, the 21 d cycle pregnancy rate was greater (P < 0.01) in primiparo us than multiparous (ARR = 1.27; 95% CI = 1.05 1.54), in cows with 1.71), and for cows in the cool than hot season (ARR = 1.75; 95% CI = 1.55 1.99). Discussion A major objective of this study was to determine if the incorporation of timed AI in cows not observed for detection of estrus after breeding should be restricted to one or more services because of the long re insemination interval resulting from this program (Lima et al., 2009). In the current study, increasing the number of timed AI before introduction to NS improved reproductive performance of dairy cows by increasing the rate of pregnancy by 15% and by reducing median and mean days nonpregnant in 9 and 10 d, respectively. Furthermore, when considering a 21 d cycle, the pregnancy rate of cows in the 3TAI treatment was 17% greater than that of 1TAI. The modernization of the dairy industry in many countries has been marked by consolidation of farms resulting in increased herd size, housin g of cows on concrete floor with smaller area per cow, increased milk production per cow, and less time allotted to individual observation of cows for estrus (Senger, 1994; Roellofsa et al., 2010). These changes in management and the increase in production have often been cited as impediments for estrous expression and high estrous detection (Senger, 1994;

PAGE 117

117 Lopez et al., 2004; Roellofsa et al., 2010). Because of low detection of estrus in high producing dairy cows, timed AI has become an integral component o f the reproductive management of many herds that use AI (Caraviello et al., 2006; NAHMS, 2009). Likewise, low detection of estrus has often been described as a reason to use NS either as the sole breeding program or for breeding of cows with advanced lacta tion (NAHMS, 2002; NAHMS, 2009). When the sole use of timed AI was compared with NS, minor differences were observed in reproductive performance of dairy cows in the first 7 months postpartum (Lima et al., 2009). Because of the long re insemination interv al in cows subjected only to timed AI, and more opportunities for breeding in NS cows, NS cows had a slight decrease in time to pregnancy of 5 fewer days (Lima et al., 2009). Despite this minor difference, pregnancies from cows exposed only to NS were gen erally more expensive than that of cows exposed only to timed AI (Lima et al., 2010). Nonetheless, retrospective analysis of herds that used both AI and NS observed greater pregnancy rates for cows in herds that kept cows in the AI groups longer than those that moved them sooner to NS groups (Overton and Sischo, 2005). In fact, the authors suggested that in herds that practice a combination of AI and NS, reproductive performance might be improved by allowing cows more opportunities for AI before moving them into clean up bullpens (Overton and Sischo, 2005). The current study clearly demonstrated that in herds using both AI and NS, increasing the opportunities for AI improved pregnancy rate. This is noteworthy as cows subjected to AI in the current study had a 42 d interval between inseminations. Nevertheless, despite this long re insemination interval, the 21 d cycle pregnancy rate was greater for 3TAI than 1TAI. Furthermore, the results of 21 cycle pregnancy rate obtained for cows in 3TAI were far

PAGE 118

118 superior t han those often cited from observational studies across many herds in North America (De Vries et al., 2005; Leblanc, 2010). As expected, the proportion of pregnant cows at the first AI and the pregnancy loss between 32 and 60 d of gestation did not differ between treatments. Approximately 16.8% of the cows lost their pregnancy. Santos et al. (2009) reviewed data on pregnancy losses in dairy cattle and concluded that approximately 12.8% of the pregnant cows lose their pregnancy between 30 and 45 d of gestat ion, with studies reporting from 3.2% in grazing cows in Ireland to as much as 42.7% in high producing cows exposed to heat stress. This extensive pregnancy loss in high producing dairy cows is multifactorial, but the exact underlying mechanisms remain unk nown (Santos et al., 2009). Exposure to NS starting 7 d after the first AI in 1TAI cows was expected to increase the proportion of pregnant cows in the first 21 and 42 d of breeding because of the delay in rebreeding nonpregnant cows in 3TAI. Nonetheless, cows in 3TAI had similar probability of pregnancy on Day 21, but a 24 and 20% increased risk of becoming pregnant by Days 42 and 84 of breeding, respectively. Only an additional 9 cows became pregnant in the first 21 d after the first AI in 1TAI. Also, onl y 57 cows in 1TAI compared with 93 cows in 3TAI became pregnant in the first 42 d after the initial insemination The low number of cows becoming pregnant between Days 7 and 21 in the 1TAI treatment may be attributed to several factors, such as a reduced pe rcentage of cows returning to estrus before Day 21, low estrous expression of cows housed on concrete, inability of bulls to detect cows in estrus when expression of estrus is compromised by lactation or footing, and low fertility of the bulls. Similarly, the lower

PAGE 119

119 number of pregnancies between Days 7 and 42 for 1TAI compared with 3TAI might be the result of a combination of low expression/detection of estrus and low fertility of bulls. Bulls in the current study underwent a breeding soundness evaluation be fore being used for breeding and were managed with alternate periods of 2 weeks of breeding and 2 weeks of rest. This approach to bull management was chosen to minimize the risk of subfertile bulls impairing reproductive performance of cows exposed to NS i n both treatments. In general, the management used for bulls in the current study and by others (Lima et al., 2009) is more intensive than that of most dairy producers relying on NS as a component of the breeding program (Champagne et al., 2002). Despite the advantages of increasing the number of AI in 3TAI compared with 1TAI, the proportion of pregnant cows at the end of the study did not differ between treatments. This was expected as management changes that impact rate of pregnancy might not necessaril y alter the final proportion of pregnant cows when the period of observation is long enough such that cows have multiple opportunities for rebreeding. The experimental period in the current study was 231 d of breeding, which allowed cows in 1TAI a total of 11 opportunities for breeding based on the standard 21 d duration of the estrous cycle. Nevertheless, the benefits of 3TAI in increasing the rate of pregnancy and reducing days to pregnancy were observed during the first 84 d in the study, when 3TAI cows received their third insemination. The two additional pre determined AI performed in the 3TAI treatment resulted in 64.8% of cows pregnant to AI, whereas only 30.8% of the 1TAI cows became pregnant to AI. The enhanced reproductive performance for the 3TAI treatment may be attributed to a greater submission to breeding in the first 84 d after the first AI, but also

PAGE 120

120 to a potential benefit to increased fertility to a breeding. High producing dairy cows have reduced expression of estrus associated with high mi lk production (Lopez et al., 2004), although production has not been associated with reduced pregnancy per AI or risk of pregnancy loss (Santos et al., 2009). Furthermore, exposure of cows and bulls to concrete might limit mounting activity and further imp air estrous expression (Senger, 1994, Roelofsa et al., 2010). The resynchronization program used starting on Day 32 after the previous AI and incorporating supplemental progesterone likely optimized fertility when cows are subjected to timed AI (Dewey et a l. 2010; Bisinotto et al., 2010 a ). Incorporating a progesterone insert during resynchronization with the Ovsynch protocol improved pregnancy per AI similar to presynchronizing the estrous cycle with GnRH 7 d before resynchronization (Dewey et al., 2010). W hen cows are subjected to NS, no hormonal manipulation is possible as it is unknown if a cow without a detectable pregnancy has been recently bred by a bull. Furthermore, cows with ovarian problems such as follicular cysts or low concentrations of progeste rone during the final stages of follicle development would be less likely to manifest estrus. This would likely reduce submission to bull breeding and potentially compromise fertility if bred. In this study parity, BCS, and season influenced all measures of reproductive performance evaluated and their impacts were similar for cows in 1TAI and 3TAI. Primiparous cows had greater pregnancy per AI at the first service, greater rate of pregnancy and fewer median days to pregnancy than multiparous, corroborating findings from previous studies (Lima et al., 2009, Santos et al., 2009). In lactating dairy cows, a BCS usually above 3.00 (one to five scale) at first AI is a critical indicator of fertility and those with improved degree of fatness usually have marked i ncreases in

PAGE 121

121 pregnancy per AI and reduced risk of pregnancy loss (Santos et al., 2009). Finally, season when cows received the first AI influenced reproduction during the entire lactation. Depression in fertility of dairy cows exposed to heat stress has be en shown in cows exposed to AI following spontaneous estrus, timed AI, or NS (Lima et al., 2009; De Vries et al., 2005). Because heat stress not only depresses pregnancy per AI, but also reduces estrous behavior of cows (Roelofsa et al., 2010) and affects semen quality (Kastelic et al., 1997) and libido in bulls collectively, these detrimental effects would alternatively extend the benefit of increased number of timed AI to favorably influence fertility during the warm season. The lack of interaction betwee n treatment and season suggests that the improved reproduction in 3TAI was observed in cows exposed to the hot and cool seasons. Others had also observed a lack of interaction between method of breeding (NS or AI) and season of year (Lima et al., 2009; De Vries et al., 2005). Conclusion In spite of the long re insemination interval for second and third AI, cows receiving 3TAI became pregnant at a faster rate than cows receiving a single timed AI before introduction to natural service. The improved reproduct ive performance of 3TAI cows resulted in 15% greater hazard of pregnancy, 17% greater risk of pregnancy, and 9 fewer days nonpregnant than 1TAI cows. The faster pregnancy rate was likely a combination of increased breeding of nonpregnant cows associated wi th improved probability of pregnancy to a breeding, which improved reproduction during the first 84 d in the study. Therefore, in herds in which detection of estrus is not carried out, a combination of AI and natural service is used, bulls are managed to o ptimize their fertility, and the resynchronized timed AI is implemented using the Ovsynch protocol

PAGE 122

122 with a CIDR insert, it is advantageous to allow cows multiple inseminations before bull exposure for natural service to optimize pregnancy rate. Results fro m this study indicate that in herds in which detection of estrus is not carried out, and a combination of AI and natural service is used, cows should receive at least three timed AI before bull exposure for natural service.

PAGE 123

123 Table 3 1 Effect of number of timed AI before exposure to natural service on reproduction of dairy cows not observed for detection of estrus 1 1TAI = cows received timed AI for first insemination and were subjected to breeding by natural service 7 d later; 3TAI = cows received up to t hree timed AI and were subjected to breeding by natural service 7 d after the third insemination. 2 AOR = adjusted odds ratio (1TAI was the reference for comparison); CI = confidence interval. 3 Based on pregnancy eval uation on Day 60 after breeding. Treatment 1 1TAI 3TAI AOR (95% CI) 2 P % (n/n) First AI Pregnant Day 32 37.9 (202/533) 39.3 (2 03/517) 1.07 (0.82 1.39) 0.60 Pregnant Day 60 30.8 (164/533) 33.5 (173/517) 1.15 (0.87 1.51) 0.30 Pregnancy loss 18.8 (38/202) 14.8 (30/203) 0.75 (0.44 1.27) 0.27 Pregnant 3 Study Day 21 32.5 (173/533) 33.5 (173/517) 1.06 (0.81 1.39) 0.67 Study Da y 42 41.5 (221/533) 51.5 (266/517) 1.58 (1.22 2.05) < 0.01 Study Day 84 54.0 (288/533) 64.8 (335/517) 1.63 (1.27 2.15) < 0.01 Study Day 231 80.7 (430/533) 81.1 (419/517) 1.05 (0.77 1.43) 0.75 21 d cycle pregnancy rate Pregnant Day 32 22.2 (486/2190) 26.6 (483/1818) 1.31 (1.09 1.59) < 0.01 Pregnant Day 60 19.6 (429/2190) 23.1 (419/1818) 1.27 (1.06 1.53) 0.01

PAGE 124

124 Table 3 2 Factors affecting the hazard of pregnancy of dairy cows not observed for detection of estrus and subjected to 1 or 3 timed AI before exposure to natural service Days to pregnancy Item Median (95% CI) 1 Mean SEM Adjusted HR 2 (9 5% CI) P Treatment 3 1TAI 142 (130 150) 165.3 3.9 --3TAI 123 (121 144) 155.6 3.7 1.15 (1.00 1.31) 0.04 Parity Multiparous 145 (130 160) 166.3 2.9 --Primiparous 81 (80 89) 115.7 5.1 1.44 (1.16 1.78) < 0.01 Body condition 161 (148 172) 180.6 4.6 --121 (118 123) 147.6 3.2 1.59 (1.38 1.84) < 0.01 Season Hot 186 (169 194) 193.2 4.2 --Cool 117 (111 120) 137.5 3.1 1.77 (1.53 2.05) < 0.01 1 CI = confidence interval. 2 HR = hazard ratio. 3 1TAI = cows received timed AI for first insemination and were subjected to breeding by natural service 7 d later; 3TAI = cows received up to three timed AI and were subjected to breeding by natural service 7 d after the third insemination.

PAGE 125

125 Figure 3 1 Diagram of hormonal treatments and insemination for the double Ovsynch program used to synchronize ovulation for first insemination in all cows at the first service and timeline of reproductive activities for 1 and 3 timed AI treatme nts. CIDR = controlled internal drug release containing progesterone; DIM = days in milk; US = ultrasonography for pregnancy diagnosis.

PAGE 126

126 Figure 3 2. Kaplan Meyer survival curves for proportion of nonpregnant cows according to treatment. 1TAI (dashed lin e) = cows received timed AI for first insemination and were subjected to breeding by natural service 7 d later; 3TAI (continuous line) = cows received up to three timed AI and were subjected to breeding by natural service 7 d after the third insemination.

PAGE 127

127 CHAPTER 4 EFFECTS OF GNRH AT INITIATION OF THE 5 D TIMED AI PROGRAM AND TIMING OF INDUCTION OF OVULATION RELATIVE TO AI ON OVARIAN DYNAMICS AND FERTILITY OF DAIRY HEIFERS Two experiments evaluated the effects of the first GnRH injection of the 5 d t imed AI program on ovarian responses and pregnancy per artificial insemination (P/AI), and the effect of timing of the final GnRH to induce ovulation relative to AI on P/AI. In experiment 1, 605 Holstein heifers were synchronized for their second inseminat ion and assigned randomly to receive GnRH on study d 0 (n = 298) or to remain as untreated controls (n = 307). Ovaries were scanned on study d 0 and 5. All heifers received a controlled internal drug release (CIDR) insert containing progesterone on d 0, a single injection of PGF and removal of the CIDR on d 5, and GnRH concurrent with timed AI on d 8. Blood was analyzed for progesterone at AI. Pregnancy was diagnosed on d 32 and 60 after AI. Ovulation on study d 0 was greater for GnRH than control (35.4 vs. 10.6%). Presence of new corpus luteum (CL) at PGF injection was greater for GnRH than control (43.1 vs. 20.8%), although the proportion of heifers with a CL at PGF did not differ between treatments and averaged 87.1%. Progesterone on the day of AI was greater for GnRH tha n control (0.50 0.07 vs. 0.28 0.07 ng/mL). The proportion of heifers at AI with progesterone < 0.5 ng/mL was less for GnRH than control (73.8 vs. 88.2%). Proportion of heifers in estrus at AI did not differ between treatments and averaged 66.8%. Pregna ncy per AI was not affected by treatment at days 32 or 60 (GnRH = 52.5 and 49.8% vs. control = 54.1 and 50.0%), and pregnancy loss averaged 6.0%. Responses to GnRH were not influenced by ovarian status on study d 0. In experiment 2, 1,295 heifers were sync hronized for their first insemination and assigned randomly to receive a CIDR on d 0, PGF and removal of the CIDR on d 5, and either

PAGE 128

128 GnRH 56 h after PGF and AI 16 h later (OVS56, n = 644) or GnRH concurrent with AI 72 h after PGF (COS72; n = 651). Es trus at AI was greater for COS72 than OVS56 (61.4 vs. 47.5). Treatment did not affect P/AI on d 32 in heifers displaying signs of estrus at AI, but COS72 improved (P = 0.05) P/AI compared with OVS56 (55.0 vs. 47.6%) in those not in estrus at AI. Similarly, P/AI on d 60 did not differ between treatments for heifers displaying estrus, but COS72 tended (P = 0.07) to improve P/AI compared with OVS56 (53.0 vs. 44.7%) in those not in estrus at AI. Administration of GnRH on the first day of the 5 d timed AI progra m resulted in low ovulation rate and no improvement in P/AI when heifers received a single PGF injection 5 d later. Moreover, extending the proestrus by delaying the final GnRH from 56 to 72 h concurrent with AI benefited fertility of dairy heifers that did not display signs of estrus at insemination following the 5 d timed AI protocol. Introductory Remarks The use of timed AI programs in dairy heifers is low compared with that for lactating dairy cows (NAHMS, 2009). Programs to synchronize ovulation of dairy heifers based on GnRH and PGF resulted in low pregnancy per AI (P/AI) compared with insemination performed after detection of estrus (Schmitt et al., 1996, Pursley et al., 1997 and Rivera et al., 2004). The depressed P/AI for most timed AI programs based on GnRH and PGF and the pe rception by dairy producers that heifers become pregnant easily without the need for intervention justifies the low use of ovulation synchronization protocols for management of reproduction in heifers. Recently, a 5 d timed AI protocol investigated by Rab aglino et al. (2010a) resulted in P/AI ranging from 52.2 to 61% in dairy heifers in the first two inseminations, which resembled the reproductive performance obtained to AI after detection of estrus

PAGE 129

129 (Kuhn et al., 2006). In fact, additional work by the same investigators evaluating anti luteolytic strategies with 325 heifers synchronized with the 5 d timed AI program observed P/AI of 59.5% on d 45 after insemination (Rabaglino et al., 2010b). Therefore, it is possible to achieve acceptable P/AI in dairy heif ers following synchronized ovulation with the 5 d timed AI protocol. The program is comprised of an injection of GnRH and insertion of a controlled internal drug release ( CIDR ) intravaginal device containing progesterone, followed 5 d later by CIDR remova l and an injection of PGF and AI concurrent with a second GnRH injection 72 h after PGF (Rabaglino et al., 2010a). Only 23% of the heifers had multiple corpora lutea ( CL ) 5 d after the injection of the GnRH (Rabaglino et al., 2010a), suggesting that o vulation to the initial GnRH was probably low. In fact, heifers receiving a single injection of PGF 5 d after GnRH had similar luteolysis and P/AI to those receiving 2 injections given 12 h apart (Rabaglino et al., 2010a). The same was not true when lact ating dairy cows were subjected to a similar program with a 5 d interval between GnRH and PGF (Santos et al., 2010 a ). Therefore, the low incidence of ovulation induced by the first GnRH combined with more rapid turnover of follicles in heifers (Sirois an d Fortune, 1988) might result in little benefit from the initial GnRH in the 5 d timed AI program in dairy heifers. Altering the timing of the final GnRH to induce ovulation relative to AI in the Ovsynch protocol influences P/AI in lactating dairy cows. Br usveen et al. (2008) reported that GnRH administered 56 h after PGF increased P/AI compared with GnRH given concurrent with timed AI at 72 h. In a series of experiments with beef cows subjected to the 5 d timed AI program, extending the proestrus from 60 to 72 h was

PAGE 130

130 beneficial to fertility (Bridges et al., 2008). In dai ry cows subjected to the 5 d timed AI program, P/AI did not differ when the final GnRH was administered either 16 h before or concurrent with AI at 72 h after PGF (Bisinotto et al., 2010 a ). Although inducing ovulation 16 h before AI benefits fertility of dairy cows in the standard 7 d timed AI Ovsynch program, it is unclear if a similar benefit would occur in dairy heifers when follicle dominance is reduced such as in the 5 d timed AI. The hypotheses of the current study were that the first GnRH would res ult in low ovulation rate, thereby having little or no impact on fertility of dairy heifers subjected to the 5 d timed AI protocol. A second hypothesis was that administration of the final GnRH concurrent with AI at 72 after PGF would result in similar P/AI as that when GnRH is administered 16 h before AI and AI is performed 72 h after PGF Two experiments with heifers inseminated following the 5 d timed AI protocol were designed to test our hypotheses. The first experiment e valuated the effect of the first GnRH injection on ovarian responses and P/AI, whereas the second experiment evaluated the effect of timing of the final GnRH to induce ovulation relative to AI on P/AI. Materials and Methods The University of Florida Instit ute of Food and Agricultural Sciences Animal Research Committee approved all procedures in this study. Experiment 1 Heifers, Diets, and Housing Six hundred and five nulliparous nonpregnant Holstein heifers on d 32 after the first insemination were synchro nized to receive second AI. Heifers were from a commercial dairy farm in north central Florida. Heifers averaged 15.3 1.7 mo of age, and were enrolled in the study in the months of December of 2009 and March of 2010.

PAGE 131

131 Heifers were managed on pasture, with access to portable shades and trees, and fed a TMR once daily that met or exceeded the nutritional requirements of Holstein heifers weighing 360 kg and gaining 0.8 kg/d (NRC, 2001). The diet was based on a mixture of lactating cow ration orts, Bermuda gra vitamin supplement. For implementation of synchronization protocols, insemination, blood collection, and pregnancy examination, heifers were handled in an open sided barn with self locking stanchions. Exper imental Design and Treatments Nonpregnant heifers on d 32 after the first AI were blocked according to age and, within each block, allocated randomly to receive 100 g of GnRH (gonadorelin hydrochloride; Factrel, Pfizer Animal Health, New York, NY) adminis tered i.m. on study d 0 ( GnRH = 298) or to remain as untreated controls (control = 307). All heifers received a CIDR (Eazi Breed CIDR Cattle Insert, Pfizer Animal Health) containing 1.38 g of progesterone on study d 0. On study d 5, the CIDR was removed an d heifers received an i.m. injection of 25 mg of PGF (dinoprost tromethamine; Lutalyse sterile solution, Pfizer Animal Health). On study d 8, an injection of GnRH was administered concurrently with timed AI (Figure 4 1). Beginning on the day of PGF administration, tailheads were painted daily with chalk and removal of chalk was used as an indication of estrus. Heifers were inseminated by 5 technicians and semen from 5 Holstein and 6 Jersey sires were used. Technicians and sires were balanced between treatments and later used in the statistical analyses. Heifers were classified according to their age as <

PAGE 132

132 Ultrasonography of Ovaries Ovaries of all heifers were scanned using a 5 MHz ultrasound unit (Easi Scan, BCF Systems, Livingston, UK) on study d 0 and ovarian maps were drawn wit h the heifers were scanned and presence and location of CL were recorded. Blood Sampling and Analysis of Progesterone in Plasma Blood was sampled from 312 of the 473 heife rs evaluated for ovulation to the first GnRH. Blood was sampled on study d 8 by puncture of the median coccygeal vein or artery using evacuated tubes (Becton Dickinson, Franklin Lakes, NJ) containing K 2 EDTA for plasma separation. Samples were placed immed iately in ice and kept refrigerated until transported to the laboratory. Blood tubes were centrifuged at 2,000 x g for 15 min, and plasma frozen at 20 C until analysis. Concentration of progesterone in plasma was analyzed in all samples by RIA using a co mmercial kit (Coat a Count, Siemens Healthcare Diagnostics, Los Angeles, CA). The sensitivity of the assay was 0.05 ng/mL calculated at 2 SD below the mean counts per min at maximum binding. Samples were analyzed in a single assay. Two known plasma samples containing 1.5 ng/mL and 2.5 ng/mL of progesterone were included in the assay several times to calculate the intra assay CV, and they were 2.5% for the sample with 1.5 ng/mL and 2.9% for the sample with 2.5 ng/mL. Evaluation of Ovulation and Progesterone at AI O d 0 and a new CL was observed on study d 5. Heifers with follicles < 10 mm on study d 0, but with a new CL on study d 5 were considered to have a new CL, but ovulated bef ore study d 0. The proportion of heifers with a visible CL by ultrasound on the day of

PAGE 133

133 PGF that had low progesterone at AI was calculated. Three different cut off values for plasma concentrations of progesterone were used, < 1.0 ng/mL, < 0.50 ng/mL, and < 0.25 ng/mL. These values were chosen because traditionally 1 ng/mL has been used to indicate CL regression, and values < 0.50 ng/mL have been implicated as cut off values that best predict P/AI (Rabaglino et al., 2010a; Santos et al., 2010 a ). Pregnancy D iagnoses and Evaluation of P/AI and Pregnancy Loss Pregnancy was diagnosed 32 d after AI by transrectal ultrasound. The presence of an embryo with a heartbeat was the criterion used to determine pregnancy. Heifers diagnosed pregnant were re examined by tra nsrectal palpation of uterine contents 28 d later, at 60 d of gestation to reconfirm pregnancy and to identify pregnancy loss. Pregnancy per AI was calculated by dividing the number of heifers diagnosed pregnant at 32 or 60 d after AI by the number of heif ers receiving AI. Proportion of pregnancy loss was calculated as the number of heifers that lost a pregnancy between 32 and 60 d of gestation divided by the number of heifers diagnosed pregnant on d 32 after AI. Experiment 2 Heifers, Diets and Housing A t otal of 1,295 nulliparous Holstein, Jersey and crossbreed Holstein Jersey heifers (15.5 2.6 mo of age) located in two farms in north central Florida were enrolled in the study between January and March of 2010. Heifers in both locations were managed on p astures and fed as described in experiment 1. Heifers were moved to an open sided barn with self locking stations in farm 1 or to a palpation rail in farm 2 for hormonal treatments, insemination, and pregnancy diagnoses.

PAGE 134

134 Experimental Design and Treatments Within farm, nulliparous heifers were blocked by breed and age and, within each block, allocated randomly to one of 2 treatments for the first AI. All heifers received a CIDR on study d 0. On study d 5, the CIDR was removed and heifers received an i.m. in jection of PGF Heifers in the 5 d timed AI program denominated OVS56 (n = 644) received an injection of GnRH at 56 h after the PGF and timed AI was performed 16 h later. Heifers in the 5 d timed AI program denominated COS72 (n = 651) received an injection of GnRH a t 72 h after the PGF concurrent with AI. Therefore, in both treatments heifers were inseminated at 72 h after CIDR removal and PGF but in OVS56, induction of ovulation was 16 h before AI (Figure 4 2). Beginning on the day of PGF administration, tai lheads were painted daily with chalk, and removal of chalk was used as an indication of estrus. The same nine technicians inseminated heifers in both farms, and 3 Holstein and 4 Jersey sires were used. Technicians and sires were balanced between treatme nts and later used in the statistical analyses. Heifers were classified according to age, e.g., < 13 mo Pregnancy Diagnoses and Evaluation of Pregnancy Outcomes Pregnancy diagnoses and calculation of P/AI and pregnancy loss were exactly as described for experiment 1. Statistical Analys i s Sample sizes were calculated for both studies to allow for sufficient experimental units to detect a difference of 8 percentage units in experiment 1 and 6 percentage units P/AI of 53% for the second i nsemination at experiment 1 and 58% for experiment 2. We anticipated that P/AI for the second AI (experiment 1) would range from 45% to 60%.

PAGE 135

135 Similarly, it was anticipated that P/AI for the first AI (experiment 2) would range from 50 to 62% based on previou s studies with the 5 d timed AI protocol (Rabaglino et al., 2010a; 2010b) and pregnancy results at the study farms. Under these assumptions, a total of 300 experimental units per treatment would be necessary for experiment 1 and 550 experimental units per treatment in experiment 2. In both experiments, binary responses were analyzed by logistic regression using the LOGISTIC procedure of SAS version 9.2 (SAS/STAT, SAS Institute Inc., Cary, NC, USA). Backward stepwise logistic regression models were used and variables were continuously removed from the models by the Wald statistic criterion when P > 0.10. In experiment 1, the models for ovarian responses to treatments, proportion of heifers with low progesterone at AI, and estrus at AI included the effects of treatment (GnRH vs. control), age of the heifer, ovarian status on study d 0 (presence or absence of CL), and interaction between treatment and ovarian status. The models for P/AI and pregnancy loss included the effects of treatment, age of the heifer, ov arian status on study d 0, sire, technician, and interaction between treatment and ovarian status. In experiment 2, the model for detection of estrus included the effects of treatment (OVS56 vs. COS72), farm, breed of the heifer, and age of the heifer. The models for P/AI and pregnancy loss included the effects of treatment, farm, breed of the heifer, age of the heifer, sire, technician, and display of signs of estrus at AI. In all analyses in both experiments, treatment was forced in the final model. In ex periment 1, concentration of progesterone at AI was analyzed by ANOVA using the GLM procedure of SAS, and the model included the effects of treatment, age of the heifer, and ovarian status on study d 0.

PAGE 136

136 Results Experiment 1 A CL visible by ultrasound was o bserved in 88.5% of the heifers on study d 0, indicating that the majority of the heifers were cycling. Ovulation on study d 0 and presence of a new CL at the injection of PGF were both greater (P < 0.01) for GnRH than control heifers (Table 4 1). Ovulation on study d 0 was greater (P < 0.01) for heifers without a CL than those with CL, and this was observed in both, GnRH (59.7 vs. 26.9%) and control (32.5 vs. 6.1%) heifers. Although ovulation rate increased with GnRH, the proportion of heifers with a visible CL on study d 5 did not differ between treatments and averaged 87.1%. Treatment with GnRH influenced (P < 0.03) the proportion of heifers with low progesterone at AI, and the effect was observed when the progesterone cut off was either 0.50 or 0.25 ng/mL. This difference resulted in GnRH heifers having greater (P < 0.01) concentration of progesterone at AI than control heifers. Detection of estrus at AI did not differ betw een treatments and averaged 67.4% (Table 4 2). Treatment with GnRH on study d 0 did not affect P/AI on either d 32 or 60 after insemination. The response to treatment was not influenced by the ovarian status on study d 0. For instance, P/AI on d 32 for hei fers with CL on study d 0 were 53.3 and 54.5% for GnRH and control, respectively. For heifers without a CL on study d 0, P/AI were 48.6 and 42.3% for GnRH and control, respectively. Similarly, pregnancy loss between 32 and 60 d of gestation did not differ between treatments and averaged 6.0%. Experiment 2 As expected, the detection of heifers in estrus on the day before timed AI was similar between treatments (Table 4 3). However, when GnRH was given 16 h before AI

PAGE 137

137 in OVS56, it decreased (P < 0.001) the pro portion of heifers in estrus on the day of timed AI. Heifers in estrus at timed AI had greater (P < 0.001) P/AI than those not detected in estrus (62.0 vs. 50.8%). An interaction (P = 0.05) between treatment and detection of signs of estrus at AI was obser ved for pregnancy on d 32. For heifers in estrus, treatment did not affect P/AI (COS72 = 60.5% vs. OVS56 = 64.1%), but for those not displaying estrus at AI, COS72 tended (P = 0.07) to increase P/AI than OVS56 (55.0 vs. 47.6%). On d 60 after AI, heifers re ceiving COS72 had greater (P = 0.05) P/AI than those receiving OVS56, and this effect was observed primarily because for heifers not detected in estrus, those in the COS72 group had greater (P < 0.05) P/AI than heifers in the OVS56 (53.0 vs. 44.7%). For he ifers detected in estrus, P/AI was not influenced by treatment (COS72 = 57.3 vs. OVS56 = 59.2%). Pregnancy loss between 32 and 60 d of gestation did not differ between treatments and averaged 5.8%. Discussion Optimization of the 5 d timed AI program to syn chronize ovulation of dairy heifers allows producers to incorporate timed insemination when needed with acceptable fertility. Earlier work with the Ovsynch program in dairy heifers resulted in low P/AI (Pursley et al., 1997), and it was suggested that time d AI should not be used in dairy heifers. The 5 d timed AI program initially described by Bridges et al. (2008) and investigated for dairy heifers by Rabaglino et al. (2010a; 2010b) resulted in P/AI that resemble those of heifers inseminated following estr us and usually better than results previously obtained with heifers subjected to the standard Ovsynch protocol and some of its variations (Pursley et al., 1997; Rivera et al., 2004). In fact, results from Rabaglino et al. (2010a; 2010b) and those from the current experiments are close to the 57% P/AI reported for Holstein heifers in the US (Kuhn et al., 2006).

PAGE 138

138 An interesting aspect of the 5 d timed AI program evaluated by Rabaglino et al. (2010a) was the low incidence of heifers with multiple CL 5 d after the injection of GnRH, thereby suggesting poor ovulatory response. In lactating dairy cows subjected to ovulation synchronization programs such as Ovsynch and Cosynch, ovulation to the first GnRH is variable according to day of the cycle when it is adminis tered (Vasconcelos et al., 1999). It is optimized when GnRH it administered on day 6 of the estrous cycle (Bello et al., 2006). In dairy heifers, ovulation to the first GnRH usually is less than that observed for lactating dairy cows, even when the estrous cycle is presynchronized (Stevenson et al., 2008). When given at random stages of the estrous cycle, GnRH resulted in only 35.4% ovulation in experiment 1, and only 26.9% of the heifers with a CL ovulated in response to administration of GnRH. Because a l arge proportion of heifers had CL on study d 0, the low ovulatory response to GnRH did not influence the proportion of heifers with visible luteal tissue by ultrasound on study d 5. Nevertheless, treatment with GnRH reduced the proportion of heifers with p rogesterone concentrations < 0.50 ng/mL. The traditional cut off for luteolysis commonly cited in the literature has been 1 ng/mL, but on the day of insemination, P/AI is optimized when progesterone concentrations are usually < 0.50 ng/mL. In fact, in many cases with both dairy heifers and lactating dairy cows, the optimized cut off value for progesterone to predict pregnancy was < 0.30 ng/mL (Rabaglino et al., 2010a; Santos et al. 2010). Therefore, although ovulation rate was low, it was sufficient to comp romise the proportion of heifers with low progesterone at AI when a single injection of PGF is administered 5 d later. In lactating dairy and beef cows, compromised luteolysis can reduce P/AI in the 5 d timed AI program, thereby requiring 2 sequential treatments with

PAGE 139

139 PGF (Kasimanickam et al., 2009; Santos et al., 2010 a ). However, in dairy he ifers, an additional treatment with PGF did not further improve P/AI (Rabaglino et al., 2010a). In the current study, administration of GnRH on study d 0 did not benefit P/AI of dairy heifers, and this response was observed regardless of the presence or absence of a CL on study d 0. In lactating dairy cows, ovulation to the initial GnRH of the timed AI protocol is critical to improve P/AI (Vasconcelos et al., 1999; Bello et al., 2006; Santos et al., 2010 a ), but the same has not been observed in dairy hei fers (Stevenson et al., 2008). Stevenson et al. (2008) administered GnRH 6 d before the initiation of the timed AI protocol to presynchronize the estrous cycle of dairy heifers. Although they were able to increase ovulation to the first GnRH of the timed A I program in presynchronized heifers, ovulation did not influence P/AI or pregnancy loss. Therefore, it is possible that in dairy heifers, typically having 3 waves of follicle development (Sirois and Fortune, 1988), ovulation and recruitment of a new wave has less impact on fertility in timed AI protocols because of the typically shorter period of ovulatory follicle dominance than in lactating cows. Also, it is possible that the benefit of GnRH inducing ovulation was mitigated by the single injection of PGF that resulted in a smaller proportion of cows with progesterone < 0.5 ng/mL at AI. Nevertheless, when a single injection of PGF is used, the results of this study indicate that the initial GnRH is not required to optimize P/AI in dairy heifers subject ed to the 5 d timed AI protocol. Because of the lack of benefit from GnRH on fertility of dairy heifers, experiment 2 was designed to evaluate whether induction of ovulation 16 h before AI would improve P/AI of dairy heifers in the 5 d timed AI protocol wi thout the GnRH on study d 0. Administering GnRH to induce ovulation concurrent with AI, in general, was beneficial to

PAGE 140

140 P/AI of dairy heifers compared with GnRH 16 h before AI. However, the benefit was observed only in heifers that did not display signs of e strus on the day of AI. Bisinotto et al. (2010) observed that for lactating dairy cows subjected to the 5 d timed AI program, induction of ovulation before AI was not beneficial to fertility. When cows undergo timed AI protocols with 7 d between the first GnRH and PGF administration of the final GnRH 16 h before AI benefits P/AI (Brusveen et al., 2008). This is thought to be mediated by improved synchrony between sufficient numbers of spermatozoa capable of fertilization in the oviduct and the presence of a viable o ocyte (Saacke, 2008). In fact, results from Brusveen et al. (2008) agree with those of Dransfield et al. (1998) in which the highest P/AI was achieved when insemination was performed 4 to 16 h after the onset of estrus. When cows are subjected to the 5 d t imed AI program, the period of ovulatory follicle development is reduced by approximately 2 d compared with conventional programs (Santos et al., 2010 a ). This reduction results in ovulatory follicles of smaller diameter, reduced concentration of estradiol in plasma, and a smaller proportion of cows in estrus at AI compared with cows in the conventional 7 d program (Santos et al., 2010 a ). In dairy heifers, these parameters have not been fully characterized. In experiment 2, delaying the administration of the GnRH to 72 h increased the proportion of heifers in estrus at AI, which is a sign of increased exposure to endogenous estradiol. In a series of experiments with beef cows subjected to the 5 d program, extending proestrus from 60 to 72 h was beneficial to fertility (Bridges et al., 2008). It would be expected that induction of ovulation 16 h before AI might be beneficial to fertility of dairy heifers because of the potentially better synchrony between the moment of ovulation

PAGE 141

141 and availability of capacitated spermatozoa in the oviduct for fertilization of the oocyte (Saacke, 2008). Nevertheless, 45% of the heifers were in estrus in the afternoon before the scheduled AI. Furthermore, when heifers were subjected to COS72, more than 61% were in estrus at the mome nt of timed AI. Heifers in estrus receiving COS72 likely already had a spontaneous LH surge when insemination was performed, which would diminish the benefit of administering GnRH 16 h before AI. The fact that delaying the administration of GnRH to the mom ent of AI improved P/AI of heifers not in estrus suggests that the additional period of proestrus was beneficial to fertility. This might have been mediated by additional exposure to estradiol and additional growth of the ovulatory follicle. The additional proestrus might be particularly important in a program of reduced period of follicle dominance to allow for sufficient pre ovulatory follicle growth and production of estradiol as suggested by Bisinotto et al. (2010). Thus, these results suggest that the prolonged proestrus in COS72 benefit fertility of dairy heifers in the 5 d program counterbalancing the potentially better synchrony between ovulation and insemination obtained by the OVS56. Conclusion Fertility of dairy heifers subjected to a 5 d timed A I protocol was not affected by administration of the first GnRH on study d 0. The lack of benefit is attributed to the low ovulation rate to the initial GnRH and the reduced proportion of heifers with low progesterone at AI when receiving a single PGF tr eatment 5 d later. Timing of induction of ovulation with GnRH relative to AI influenced P/AI of heifers not displaying estrus at AI, and it was usually better when heifers received the ovulatory stimulus concurrent with AI at 72 h after PGF than 16 h bef ore timed AI. Therefore, when heifers are subjected to the 5 d timed AI program with a single treatment of PGF it is

PAGE 142

142 suggested that the initial GnRH is not necessary and the period of proestrus should be 72 h with administration of GnRH to induce ovulat ion concurrent with timed AI.

PAGE 143

143 Table 4 1. Ovarian responses of heifers treated with or without GnRH at the initiation of the 5 d timed AI protocol 1 GnRH = d 0 GnRH and CIDR, d 5 PGF and removal of CIDR, d 8 GnRH and timed AI; Control = d 0 CIDR, d 5 PGF and removal of CIDR, d 8 GnRH and timed AI. 2 AOR = adjusted odds ratio. Control is the reference for comparison. Treatment 1 GnRH Control AOR 2 (95% CI) P % (n/n) Study d 0 Heifers with foll 90.4 (216/239) 88.6 (210/237) 1.21 (0.67 2.17) 0.53 CL on study d 0 74.1 (177/239) 83.1 (197/237) 0.57 (0.36 0.89) 0.01 Ovulation on study d 0 All heifers 35.4 (84/237) 10.6 (25/236) 4.49 (2.68 7.51) < 0.01 mm 39.3 (84/214) 12.0 (25/209) 4.82 (2.80 8.30) <0.01 New CL at PGF 43.1 (102/237) 20.8 (49/236) 3.13(1.91 5.13) < 0.01 CL at PGF 88.2 (209/237) 86.0 (203/236) 1.30 (0.75 2.24) 0.35 Proportion of heifers according to progesterone at AI < 1 ng/mL 90.1 (127/141) 94.9 (129/136) 0.51 (0.20 1.30) 0.16 < 0.50 ng/mL 73.8 (104/141) 88.2 (120/136) 0.34 (0.18 0.66) < 0.01 < 0.25 ng/mL 41.1 (58/141) 52.9 (72/136) 0.57 (0.35 2.93) 0.03 Progesterone at AI, ng/mL 0.50 0.07 0.28 0.07 --< 0.01

PAGE 144

144 Table 4 2. Effect of the first GnRH injection of the 5 d tim ed AI protocol on fertility responses of dairy heifers Experiment 1 Treatment 1 GnRH Control AOR 2 (95% CI) P % (n/n) Estrus at timed AI 3 69.2 (206/298) 64.5 (198/307) 1.23 (0.88 1.73) 0.23 Pregnant 4 Day 32 52.5 (155/295) 54.1 (165/305) 0.9 4 (0.68 1.30) 0.70 Day 60 49.8 (147/295) 50.0 (150/300) 0.99 (0.72 1.37) 0.97 Pregnancy loss 5 5.8 (9/155) 6.2 (10/160) 0.93 (0.37 2.34) 0.87 1 GnRH = d 0 GnRH and CIDR, d 5 PGF and removal of the CIDR, d 8 GnRH and timed AI; Control = d 0 CIDR, d 5 PGF and removal of CIDR, d 8 GnRH and timed AI. 2 AOR = adjusted odds ratio. Control is the reference for comparison. 3 Evaluated based on removal of tail chalk on the day of time d AI. 4 Three GnRH and 2 control heifers left the study bef ore pregnancy diagnosis on d 32. 5 Five control heifers left the study before reconfirmation of pregnancy on d 60 after AI.

PAGE 145

145 Table 4 3. Effect of time of administration of the final GnRH of the 5 d timed AI protocol relative to insemination on fertility responses of dairy heifers Experiment 2 1 COS72 = d 0 CIDR, d 5 PGF and removal of CIDR, d 8 GnRH and timed AI; OVS56 = d 0 GnRH and CIDR, d 5 PGF and removal of CIDR, d 7.3 GnRH, d 8 timed AI. 2 AOR = adjusted odds ratio. COS72 is the reference for comparison. 3 Evaluated based on removal of tail chalk on the day bef ore and day of timed AI. Treatment 1 OVS56 COS72 AOR 2 (95% CI) P % (n/n) Estrus 3 Day before timed AI 46.7 (301/644) 43.2 (281/651) 1.16 (0.93 1.45) 0.19 Da y of timed AI 47.5 (306/644) 61.4 (400/651) 0.52 (0.45 0.70) < 0.001 Pregnant Day 32 55.4 (357/644) 58.4 (380/651) 0.75 (0.42 1.08) 0.08 Day 60 51.6 (332/644) 55.6 (362/651) 0.72 (0.39 1.04) 0.05 Pregnancy loss 7.0 (25/357) 4.7 (18/380) 1.52 (0.81 2.83) 0.19

PAGE 146

146 Figure 4 1. Diagram of activities in experiment 1. BS = blood samples for analysis of progesterone concentration; CIDR = controlled internal drug release relin hydrochloride; PGF = injection of 25 mg of dinoprost as tromethamine salt; US = ultrasonography of the ovaries.

PAGE 147

147 Figure 4 2. Diagram of activities in experiment 2. CIDR = controlled internal drug release containing 1.38 g of progesterone; GnRH gonadorelin hydrochloride; PGF = injection of 25 mg of dinoprost as tromethamine salt.

PAGE 148

148 CHAPTER 5 HORMONAL MANIPULATIONS IN THE 5 D TIMED AI PROTOCOL TO OPTIMIZE ESTROUS CYCLE SYNCHRONY AND FERTILITY IN DAIRY HEIFERS Obje ctives were to determine the effects of GnRH at the initiation of the 5 d timed artificial insemination (AI) program combined with two injections of PGF on ovarian responses and pregnancy per AI (P/AI) in dairy heifers, and the role of progesterone conce ntrations on LH release and ovulation in response to GnRH. In study 1, heifers received a controlled internal drug release (CIDR) insert containing 1.38 g of progesterone on d 0, an injection of 25 mg of PGF and CIDR removal on d 5, and an injection of 1 no additional treatment (control, n = 559) or an injection of GnRH on d 0 and a second injection of PGF on d 6 (G2P, n = 547). In study 2, all heifers were treated as described for control in study 1, and were allocated to receive no additional treatment (control = 723), an injection of PGF on d 6 (NG2P = 703), or an injection of GnRH on d 0 and an injection of PGF on d 6 (G2P = 718). In study 3, heifers received a CIDR on d 7 af ter ovulation and were assigned randomly to a low progesterone (LP; n = 6) treatment in which two injections of 25 mg of PGF each were administered 12 h apart, on d 7 and 7.5 after ovulation, or to a high progesterone (HP, n = 12) in which no PGF was a every 15 min from 30 to 180 min relative to the GnRH for assessment of LH concentrations. Additionally, 94 heifers were assigned to LP or HP and ovulation in response to GnRH wa s evaluated. In study 1, P/AI was greater for G2P than for control on d 32 (59.4 vs. 53.5%) and 60 after AI (56.6 vs. 51.3%). In study 2, administration of GnRH on d 0 increased the proportion of heifers with a new corpus luteum on d 5 (control = 21.9 vs. NG2P = 20.1 vs. G2P = 34.4%). Administration of a second PGF

PAGE 149

149 increased the proportion of heifers with progesterone < 0.5 ng/mL at AI (control = 83.1 vs. NG2P = 93.0 and G2P = 87.2%). Pregnancy per AI was greater for G2P than for control and NG2P on d 32 (control = 52.9 vs. NG2P = 55.0 vs. G2P = 61.7%) and 60 (con trol = 49.0 vs. NG2P = 51.6 vs. G2P = 59.1%). In study 3, HP attenuated LH release and reduced ovulation (19.0 vs. 48.4%) in response to GnRH compared with LP. Combining GnRH and two doses of PGF in the 5 d timed AI protocol improved follicle turnover, l uteolysis, and P/AI in heifers. Elevated concentrations of progesterone suppressed LH release and are linked with the low ovulatory response to the initial GnRH treatment of the protocol. Introductory Remarks Reproductive efficiency in dairy heifers affe cts age at first calving which has major impacts on rearing costs and subsequent productive life (Gabler et al., 2000; Ettema and Santos, 2004). Most dairy operations in the United States use AI after observed estrus to manage reproduction in heifers (NAHM S, 2009). Nevertheless, advances in protocols for synchronization of the estrous cycle have supported the use of timed AI as an alternative method to control reproduction and improve economics when detection of estrus is less than 70% (Ribeiro et al. 2012a ). Recent studies have consistently reported pregnancy per AI ( P/AI ) ranging from 50 to 60% in dairy heifers subjected to the 5 d timed AI program (Rabaglino et al., 2010; Lima et al., 2011), which are comparable to those observed in heifers inseminated at detected estrus (Kuhn et al., 2006). Further optimization of such programs to either simplify or improve fertility will likely increase acceptance by dairy producers. Ovulation in response to the initial GnRH injection in timed AI programs enhances synchr ony of estrous cycle, shortens follicle dominance, and improves

PAGE 150

150 embryo quality and P/AI (Vasconcelos et al., 1999; Chebel et al., 2006; Cerri et al., 2009a). Nevertheless, only 15 to 35% of heifers ovulate when treated with GnRH at random stages of the est rous cycle (Stevenson et al., 2008; Lima et al., 2011). In addition, heifers that ovulate in response to the initial GnRH will have a newly formed corpus luteum ( CL ), which is generally refractory to a single treatment with PGF on d 5 of the cycle (Rowso n et al., 1972; Henricks et al., 1974). Eliminating the first GnRH reduced ovulation at the beginning of the synchronization protocol, but increased the proportion of heifers that underwent luteolysis at AI when a single PGF injection was used (Lima et a l. 2011). Because the benefits associated with follicle turnover were offset by a less effective CL regression, P/AI did not differ between heifers that received or not GnRH at the initiation of the timed AI program (Lima et al. 2011). These results indica te that the initial GnRH is not necessary when a single PGF is used, which simplifies and reduce costs associated with the synchronization protocol. Results from lactating dairy cows subjected to the 5 d timed AI program indicate that the use of two injections of PGF administered 24 h apart improved CL regress ion and P/AI (Santos et al., 2010 a ), particularly when ovulation to initial GnRH was high (Ribeiro et al., 2012b). Shorter intervals between PGF treatments, ranging from 7 to 8 h, have been shown to increase P/AI compared with a single injection in beef cows (Kasimanickam et al., 2009), although preliminary results in dairy heifers did not confirm such benefit (Rabaglino et al., 2010). Therefore, it is reasonable to speculate that the combination of the initial GnRH and the administration of PGF on d 5 and 6 of the protocol will improve follicle turnover and luteal regression, which are expected to increase P/AI.

PAGE 151

151 The differences in catabolism of steroid hormones (Sangsritavong et al., 2002) explain the nearly 1.5 ng/mL greater progesterone concentration in heifers than lactating cows during mid diestrus (Sartori et al., 2004). In fact, the increase in progesterone concentrations with a controlled internal drug release ( CIDR ) is expected to be greater in nonlactating (Zuluaga and Williams, 2008) than in la ctating cows (Cerri et al., 2009b). Progesterone affects LH secretion, which might compromise ovulatory response to GnRH treatment, which might partially explain the low ovulatory response to GnRH in dairy heifers. Results from beef heifers support this id ea (Colazo et al., 2008; Dias et al., 2010). It was hypothesized that a combination of GnRH at the initiation of the 5 d timed AI and injections of PGF on d 5 and 6 of the protocol improves the synchrony of the estrous cycle and fertility in dairy heifers. Furthermore, it was hypothesized that elevated concentrations of progesterone compromise the release of LH and ovulation in response to a GnRH injec tion in dairy heifers. Study 1 was designed to compare a simplified 5 d timed AI protocol with a protocol that is expected to optimize P/AI by inducing ovulation and optimizing regression of newly formed CL. Study 2 was designed to evaluate the effects of GnRH at the initiation of 5 d timed AI program combined with two injections of PGF on ovarian responses and fertility. Finally, the objectives of study 3 were to assess LH release and ovulation in response to GnRH in dairy heifers with low or high concentrations of progesterone in plasma. In heifers that receive GnRH in the beginning of 5 d timed AI program might require multiple PGF to optimize luteolysis and P/AI. Results from lactating dairy cows subjected to the 5 d timed AI program indicate that the use of two injections of PGF

PAGE 152

152 administered 24 h apart improved CL regression an d P/AI (Santos et al., 2010a), particularly when ovulation to initial GnRH is high and more cows present a newly formed CL (Ribeiro et al., 2012b). Shorter intervals between PGF treatments, ranging from 7 to 8 h, have been shown to increase P/AI compared with a single injection in beef cows (Kasimanickam et al., 2009), although preliminary results in dairy heifers did not confirm such benefit (Rabaglino et al., 2010). Therefore, it is reasonable to speculate that the combination of the initial GnRH and th e administration of PGF on d 5 and 6 of the protocol will improve follicle turnover and luteal regression, which is expected to result in increased P/AI in dairy heifers. Presumably because of less catabolism of steroid hormones in heifers than lactating cows because of differ ences in splanchnic blood flow (Sangsritavong et al., 2002) ; progesterone concentrations in the plasma of dairy heifers are nearly 1.5 ng/mL greater than those in lactating cows during mid diestrus (Sartori et al., 2004). Because of reduced catabolism, the increase in progesterone concentrations with a CIDR is expected to be greater in nonlactating (Zuluaga and Williams, 2008) than in lactating cows (Cerri et al., 2009b). Progesterone affects LH secretion, which might compromise ovulatory response to GnRH t reatment, which might partially explain the low ovulatory response to GnRH in dairy heifers. In fact, results from beef heifers support the idea that progesterone compromise LH release and impair ovulation following an injection of GnRH (Colazo et al., 200 8; Dias et al., 2010). It was hypothesized that a combination of GnRH at the initiation of the 5 d timed AI and injections of PGF on d 5 and 6 of the protocol improve the synchrony of the estrous cycle and fertility in dairy heifers. Furthermore, it was hypothesized that

PAGE 153

153 elevated concentrations of progesterone compromise the release of LH and ovulation in response to a GnRH inject ion in dairy heifers. Study 1 was designed to compare a simplified 5 d timed AI protocol with a protocol that is expected to optimize P/AI by inducing ovulation and optimizing regression of newly formed CL. Study 2 was designed to evaluate the effects of G nRH at the initiation of 5 d timed AI program combined with study 3 were to assess LH release and ovulation in response to GnRH in dairy heifers with low or high concent rations of progesterone in plasma. Materials and Methods The University of Florida Institute of Food and Agricultural Sciences Animal Research Committee approved all procedures in the 3 studies reported. Study 1 Heifers, Diets, and Housing A total of 1,106 nulliparous Holstein and crossbred Holstein Jersey heifers at an average ( SD) of 14.0 2.1 mo of age from two farms in north central Florida were enrolled in the study between March and June, 2010. Crossbred heifers (n = 231) were located only in farm 2. Four hundred and fifty seven heifers received their first insemination, whereas the remaining 649 heifers were diagnosed non pregnant on d 32 after insemination and resynchronized to receive their second AI. Heifers in both locations were managed on pasture with access to portable shades and trees. Heifers were fed a TMR once daily formulated to meet or exceed the nutritional requirements of Holstein heifers weighing 360 kg and gaining 0.8 kg/d (NRC, 2001). The diet consisted of a mixture of lactat minerals and vitamins supplement. For administration of hormonal treatments,

PAGE 154

154 insemination, and pregnancy examination, heifers were moved to an open sided barn with self locking stations in farm 1 or to a palpation rail in farm 2. Experimental Design and Treatments All heifers received a CIDR containing 1.38 g of progesterone (Eazi Breed CIDR Cattle Insert, Zoetis, Madison, NJ) on d 0, an i.m. injection of 25 mg of PGF (dinoprost tromethamine; Lutalyse sterile solution, Zoetis, Madison, NJ) and CIDR removal on d 5, and an i.m. injection of 100 g of GnRH (gonadorelin hydrochloride; Factrel, Zoetis, Madison, NJ) concurrently with AI on d 8. On d 0, heifers were blocked according to number of AI (first or second) and then age and, within each block, they were allocated randomly to receive no additional treatment ( control, n = 559) or an injection of GnRH on d 0 and a second injection of PGF on d 6 ( G2P, n = 547; Figure 4 1). On d 5 and 6, heifers had their tailheads painted using paintsticks (All Weather Paintstick; LA CO Industries, Chicago, IL). Expression of estrus at AI was evaluated based on removal of the tail paint on d 8. Inseminations were performed by 11 techn icians with semen from six Holstein and six Jersey sires. Pregnancy Diagnosis and Calculation of Reproductive Outcomes Pregnancy was diagnosed 32 d after AI by transrectal ultrasonography of the uterus using a portable ultrasound equipped with a 7.5 MHz transrectal probe (Easi Scan, BCF Technologies, Rochester, MN). The visualization of an amniotic vesicle containing an embryo with heartbeat was used as the determinant of pregnancy. Pregnant heifers on d 32 were re examined by transrectal palpation of ute rine contents on d 60 after AI. Pregnancy per AI was calculated by dividing the number of heifers diagnosed pregnant on d 32 or 60 after AI by the number of heifers receiving AI. Pregnancy loss was calculated as the number of heifers that lost their pregna ncy

PAGE 155

155 between d 32 and 60 after AI divided by the number of heifers diagnosed pregnant on d 32 after AI. Study 2 Heifers, Diets, and Housing A total of 2,144 nulliparous Holstein heifers with an average ( SD) of 13.5 1.0 mo of age from a single farm in n orth central Florida were enrolled in the study between July 2010 and April, 2011. A total of 1,723 heifers received their first insemination, whereas the remaining 421 heifers were diagnosed non pregnant on d 32 after their first AI and resynchronized to receive the second AI. Heifers were managed and fed as described in study 1. For administration of hormonal treatments, ultrasonography examination, blood sampling, insemination, and pregnancy diagnosis, heifers were moved to an open sided barn with self l ocking stations. Experimental Design and Treatments All heifers received a CIDR containing 1.38 g of progesterone on d 0, an i.m. injection of 25 mg of PGF and CIDR removal on d 5, and an i.m. injection of 100 g of GnRH concurrently with AI on d 8. On d 0, heifers were blocked by number of AI (first or second) and then age and, within each block, they w ere allocated randomly to receive no additional treat ment (control, n = 723), an additional injection of PGF on d 6 (NG2P, n = 703), or an injection of GnRH on d 0 and an additional injection of PGF on d 6 (G2P, n = 718; Figure 5 2). Estrus at AI was dete cted as described in study 1. Inseminations were p erformed by 10 technicians with semen from 10 Holstein and 3 Jersey sires.

PAGE 156

156 Ultrasonography of Ovaries and Evaluation of Ovulatory Responses Ovaries from a subset of 623 heifers were scanned on study d 0 and 5 and the presence and location of CL > 15 mm a Ovulation at the beginning of the timed AI program was considered when the heifer had 5. Heifers with follicles < 10 mm on s tudy d 0 but with a new CL on study d 5 were considered to have a new CL, but ovulated before study d 0. Blood Sampling and Analysis of Progesterone Concentrations Blood was sampled from 610 of the same subset of 623 heifers evaluated for ovulation describ ed previously. Blood samples were collected on d 8 by puncture of the median coccygeal vein or artery using evacuated tubes (Becton Dickinson, Franklin Lakes, NJ) containing K 2 EDTA for plasma separation. Samples were placed immediately on ice and kept ref rigerated until arrival to the laboratory. Blood tubes were centrifuged at 2,000 x g for 15 min, and an aliquot of 2 mL of plasma was frozen at 20 C until analysis. Concentration of progesterone in plasma was analyzed in all samples by RIA using a commer cial kit (Coat a Count, Siemens Healthcare Diagnostics, Los Angeles, CA). The sensitivity of the assay was 0.05 ng/mL calculated at 2 SD below the mean counts per min at maximum binding. All samples were analyzed in a single assay. Two plasma samples with progesterone concentrations of 1.5 and 2.5 ng/mL were included throughout the sequence of samples in the assay for quality control. The intra assay CV were 2.5 and 2.9% for the samples containing 1.5 and 2.5 ng/mL, respectively. Three different cut off val ues for plasma concentrations of progesterone at AI were used to determine luteolysis, progesterone < 1.0, < 0.50, and < 0.30 ng/mL. These

PAGE 157

157 three values were selected based on the traditional threshold used to indicate CL regression (1 ng/mL), or based on c oncentration of progesterone at AI that have been used as cut off values that best predicted P/AI (Rabaglino et al., 2010; Santos et al., 2010 a ). Pregnancy Diagnosis and Calculation of Reproductive Outcomes Pregnancy was diagnosed 32 and 60 d after AI as described in study 1. Similarly, P/AI and pregnancy loss were calculated as described for study 1. Of all 2,144 heifers, 26 did not have a pregnancy diagnosis performed (control = 12; NG2P = 7; G2P = 7) because they were moved to another farm before d 32 a fter AI. Study 3 Experimental Design and Treatments Holstein nulliparous heifers from the University of Florida Dairy Unit had their diameter received an injection of PGF Heifers had their tailheads painted and were observed daily for signs of estrus based on removal of the tail chalk. Heifers detected in estrus between 48 and 72 h after the injection of PGF had their ovaries scanned to map the ovarian follicles. Ovarie s were scanned again 24 and 48 h after estrus and the assumed to have occurred in the preceding day to the scanning when the dominant follicle was no longer visible by ultra sonography. Only heifers that ovulated within 48 h of detected estrus were included in the study. The day of ovulation was considered study d 0. Heifers were assigned randomly to either a low (LP; n = 6) or high progesterone (HP; n = 12) treatment (Figure 5 3). All heifers received a CIDR on d 7 after ovulation and those assigned to LP received tw o injections of PGF 12 h apart

PAGE 158

158 beginning at CIDR insertion. On d 8, all heifers received an i.m. injection of 100 g of GnRH (gonadorelin diacetate tetrahydrate; Cystorelin, Merial, Iselin, NJ). An additional group of 94 nulliparous Holstein heifers were randomly assigned to the same treatments administered on d 8 and ovaries were scanned on d 7, 8 and 12 to characterize ovulation in response to GnRH. Blood Sampling and Analyses of LH and Progest erone Concentrations 18 heifers subjected to LP and HP treatments. A 14 gauge x 14 cm indwelling catheter (Abbocath T; Hospira Inc., Lake Forest, IL) was placed in the left jug ular vein for the duration of blood sampling. Samples were collected at 30, 15, 0, 15, 30, 45, 60, 75, 90, 105, 120, 135, 150, 165, 180, 240, and 300 min relative to the injection of GnRH. Before each sampling, approximately 8 mL of blood were draw and d iscarded for cleansing of the catheter. Samples were then collected using a 10 mL syringe and transferred into evacuated tubes (Becton Dickinson, Franklin Lakes, NJ) for subsequent serum separation. After sampling, the catheters were flushed with hepariniz ed solution (30 USP heparin sodium; Sigma Aldrich, Saint Louis, MO) to avoid clogging. Samples were immediately place on ice remaining there for 30 min and then placed in room temperature for 30 min before centrifugation for serum separation. Tubes were ce ntrifuged at 2,200 x g at 4 o C for 15 min for serum separation. Serum samples were frozen at 20 o C until later analysis. Concentrations of LH in serum were determined by RIA as previously reported (McVey and Williams, 1991). Highly purified ovine LH (AFP 8614B, NIDDK oLH I 4; National Hormone and Pituitary Program, Harbor UCLA Medical Center, Torrance, CA)

PAGE 159

159 was used as both the reference preparation and as iodinated tracer. The primary antiserum used was produced in rabbits immunized against ovine LH (AFP 192279, NIDK anti oLH 1; National Hormone and Pituitary Program, Harbor UCLA Medical Center, Torrance, CA). This antiserum displays similar cross reactivity between highly purified preparations of ovine (NIDDK I 2) and bovine (AFP11743B) LH and does not cross react with other pituitary hormones. Sensitivity of the assay was 0.1 ng/mL and the intra and inter assay CV averaged 11.9 and 10.4%, respectively. An additional blood sample was collected immediately before the injection of GnRH and transferred to evacuated tubes (Becton Dickinson, Franklin Lakes, NJ) containing K 2 EDTA for plasma separation. Plasma was separated, stored, and assayed for progesterone concentration as described in study 1. Concentrations of progesterone in plasma were analyzed in the same assay described in study 2 with intra assay CV of 2.5 and 2.9% for known samples containing 1.5 and 2.5 ng/mL, respectively. Statistical Analys i s Power analyses were performed to calculate sample sizes in all three studies using Minitab 16 (Minita b Inc., State College PA, USA). Sample sizes were calculated for studies 1 and 2 to allow sufficient experimental units t o detect a difference of 6 tailed test). The expected P/AI for first and second AI combined was of 58% for G2P and 52% for the remaining treatments based on previous studies (Rabaglino et al., 2010; Lima et al., 2011). Under these assumptions, a minimum of 540 and 636 experimental units per treatment were deemed necessary in studies 1, which had two treatments, and 2, which had three treatments, respectively. Because of potential attrition additional heifers were added to all treatments in both studies. In study 3, the sample size was calculated to

PAGE 160

160 allow sufficient experimental units to detect a difference of 30 percentage units in tory response was anticipated to increase from 20% in HP to 50% in LP based on previous studies with the 5 d timed AI protocol (Lima et al., 2011). Categorical data were analyzed by logistic regression using the GLIMMIX procedure of SAS version 9.3 (SAS/S TAT, SAS Institute Inc., Cary, NC) fitting a binary distribution. Treatment was forced in the final models, but covariates and the interaction between treatment and covariates were sequentially removed from the model if P > 0.10. In study 1, the model for detection of estrus included the effects of treatment, breed, farm, number of AI, age of heifer (< 13 mo vs. 13 to 15 mo vs. > 15 mo), and the interactions of treatment and breed, treatment and farm, and treatment and number of AI. The models for P/AI and pregnancy loss included the effects of treatment, breed, farm, number of AI, age of heifer, sire, AI technician, and the interactions of treatment and breed, treatment and farm, and treatment and number of AI. In study 2, the models for analyses of ovaria n responses to treatments, proportion of heifers with progesterone at AI below one of three cut off points (< 1.0, < 0.50, and < 0.30 ng/mL), and proportion of heifers in estrus at AI included the effects of treatment, age of heifer (< 13 mo vs. 13 to 14 m o vs. > 14 mo), presence of CL on d 0, and interaction between treatment and presence of CL on d 0. The models for P/AI and pregnancy loss included the effects of treatment, number of AI, age of heifer, sire, AI technician, and interaction between treatmen t and AI number.

PAGE 161

161 Concentrations of progesterone at AI in study 2 were analyzed by ANOVA using the GLM procedure of SAS. The model included the effects of treatment, age group, and presence of CL on study d 0. In study 3, concentrations of LH were analyzed by ANOVA for repeated measurements using the MIXED procedure of SAS. The values for LH concentration at 30, 15 and 0 min was averaged for each individual heifer and used as a covariate. The model included the effects of treatment, time, and the interact ion between treatment and time, with heifer nested within treatment as the random term for test of effects of treatment. The covariance structure that resulted in Differences with P 0.05 were considered significant and those with 0.05 < P 0.10 were considered tendencies. Results Study 1 T The proportion of heifers detected in estrus on the day of timed AI did not differ between treatments and averaged 76.8% (Table 5 1). A greater ( P < 0.001) proportion of heifers were detected in estrus at AI in farm 1 compared with farm 2 (81.5 vs. 62.1%). There were no interactions between treatment and other independent variables for detection of estrus at AI. Pregnancy per AI on d 32 was great er ( P = 0.04) for G2P than for control heifers (Table 5 1). Similarly, the proportion of pregnant heifers on d 60 tended ( P = 0.06) to be greater in G2P than in control. Although P/AI was greater ( P = 0.02) in farm 1 compared with farm 2 on d 32 (57.3 vs. 54.0%) and 60 after AI (54.4 vs. 52.7%), no interaction between treatment and farm was observed for P/AI. Pregnancy loss was not affected by treatment or by the interactions between treatment and other independent variables.

PAGE 162

162 Study 2 A CL visible by ultras onography was observed in 80.4% of the heifers on d 0, indicating that the majority were cyclic. As anticipated, ovulation on d 0 and the presence of a new CL on d 5 were greater ( P heifers (Table 5 2). Ovulation on study d 0 was greater ( P < 0.01) for heifers without a CL (n = 122) than in those with a CL (n = 501) on d 0 (45.9 vs. 10.4%). Although ovulation at the initiation of the timed AI program was increased by the administration of GnRH, the proportion of heifers with a visible CL on study d 5 did not differ between treatments. The proportion of heifers with progesterone concentration at AI below the cutoffs of 0.30 and 0.50 ng/mL was le ss for control than NG2P and G2P (Table 5 2). A tendency (P = 0.09) for interaction between treatment and new CL was observed for luteolysis based on progesterone < 0.50 ng/mL on study d 8. This interaction was because treatments with 2 PGF increased luteolysis in heifers with a new CL compared with control, but no difference was observed for those without a new CL. For controls, the proportions of heifers with progesterone < 0.50 ng/mL were 85.7 and 73.9% for those without and with a new CL; for NG2P, the same proportions were 93.7 and 90.2%, and for G2P they were 85.0 and 91.4%. There were not interactions between treatments and any other independent variables for any of the ovarian responses in study 2. Treatment affected ( P = 0.02) the proportion of heifers in estrus at AI (Table 5 3). Detection of estrus was greater (P < 0.01) for G2P than control, and tended (P = 0.06) to be greater for G2P than NG2P. No difference was observed between control and NG2P. Pregnancies per AI on d 32 and 6 0 after insemination were greater (P < 0.01) for G2P than control and NG2P. Heifers detected in estrus (n = 1,612) on the day of AI had

PAGE 163

163 greater (P < 0.001) P/AI on d 32 (58.6 vs. 50.7%) and on d 60 (56.7 vs. 48.8%) than those not in estrus (n = 532). Heife rs receiving their second insemination had greater ( P = 0.05) P/AI on d 60 than heifers receiving their first insemination (57.6 vs. 52.2). Pregnancy loss tended (P = 0.06) to be less for G2P than controls (Table 5 3). Heifers receiving their first AI had greater ( P = 0.03) pregnancy loss than heifers receiving their second AI (6.7 vs. 2.9%). When P/AI were analyzed in the subset of 623 heifers with ovarian ultrasound and progesterone concentration at AI, an interaction (P = 0.03) between number of PGF a nd new CL was observed for P/AI on d 32 and 60 after insemination. For heifers without a new CL, number of PGF (control vs. NG2P + G2P) did not ( P = 0.36) influence P/AI on d 32 (59.3 vs. 54.9%), but for those heifers with a new CL on d 5 of the timed AI protocol, administration of 2 doses of PGF (NG2P + G2P) increased ( P < 0.05) P/AI compared with control (62.8 vs. 45.7%). The same response was observed on d 60 after AI, in heifers without a new CL, number of PGF did not (P = 0.53) influence P/AI (co ntrol = 54.9 vs. NG2P + G2P = 51.9%); however, in those with a new CL, administration of 2 doses of PGF increased ( P < 0.05) P/AI compared with control (61.1 vs. 43.5%). Study 3 As expected, the concentration of progesterone on d 8 was less ( P = 0.01) fo r LP than HP heifers (Table 5 4). The elevated concentration of progesterone in HP treatment reduced ( P = 0.04) the mean concentration of LH in serum compared with LP heifers. A tendency (P = 0.07) of interaction between treatment and time relative GnRH in jection was observed for LH concentrations. Serum concentrations of LH were less (P < 0.05) from 45 to 135 min after the injection of GnRH in HP compared with LP

PAGE 164

164 treatment (Figure 5 4). Furthermore, HP treatment tended to reduce ( P = 0.09) the peak concent ration of LH in plasma compared with LP treatment. The interval from GnRH injection to peak of LH concentration did not differ between treatments. Finally, HP heifers had reduced (P = 0.01) incidence of ovulation in response to GnRH compared with LP heifer s. Discussion The results of current study clearly support the concept that increased ovulation with administration of GnRH at the initiation of the protocol combined with improved luteolysis at insemination, by using two doses of PGF are necessary to optimize P/AI in dairy heifers subjected to the 5 d timed AI program. These results are consistent with our hypothesis and clarify unanswered questions from findings of a previous study (Lima et al., 2011). Previously, administration o f GnRH on d 0 of the 5 d timed AI protocol increased ovulation compared with no GnRH administration, but did not improve P/AI in heifers receiving a single dose of PGF (Lima et al., 2011). This lack of benefit was suggested to be caused by the inability of a single injection of PGF to fully regress newly formed CL to optimize fertility. Study 1 was designed as a practical study to evaluate two breeding strategies for dairy heifers, one with minimum intervention without GnRH on d 0 and with a single PGF 2 on d 5, and another treatment expected to optimize P/AI by improving follicle turnover and CL regression. Although the study design does not permit identification of the mechanisms for improved fertility in heifers receiving G2P compared with controls, i t is clear that a simplified version of the 5 d timed AI protocol, as implemented in control heifers, does not maximize P/AI. Based on results of study 1, it was unclear if benefits in P/AI in heifers were derived from the combined improved follicle turnov er and luteolysis

PAGE 165

165 or if just an improved luteolysis would be sufficient to improve P/AI. Therefore, study 2 was designed to investigate if a combined improved follicle turnover and luteolysis was necessary to maximize P/AI in dairy heifers. Dairy cows tha t ovulate to the initial GnRH of the timed AI program have enhanced overall synchrony of the estrous cycle (Vasconcelos et al., 1999; Rutigliano et al., 2008), which shortens the period of follicle dominance and improves embryo quality (Cerri et al., 2009a ). This follicle response is beneficial to fertility in dairy cows (Vasconcelos et al., 1999; Rutigliano et al., 2008; Santos et al., 2010 a ). For instance, the results of a previous study in dairy heifers suggested that follicle turnover was not as critica l as it appears to be in dairy cows to improve fertility, when only one PGF was used in the 5 d timed AI program (Lima et al., 2011). However, it is possible that the potential benefits of improved follicle turnover were offset by reduced luteolysis obta ined with a single treatment with PGF In fact, the 5 d interval between GnRH and PGF might limit adequate luteolysis when a newly formed CL is present. The refractoriness of the early CL with fewer than 5 d of development to a single PGF treatment has been reported (Rowson et al., 1972; Henricks et al. 1974), and one solution to overcome this problem is the use of multiple PGF treatments (Santos et al., 2010 a ; Ribeiro et al., 2012b). Although multiple doses of PGF have been shown to successfully induce luteolysis even in cows with CL having less than 5 d (Beal et al., 1980), the use of two PGF injections did not always improve fertility (Cruppe et al., 2010; Rabaglino et al., 2010). One possibility to explain the lack of benefit to two doses of PGF in the 5 d timed AI program is a poor ovulatory response to GnRH when the protocol is initiated, so only a small proportion of animals have newly formed CL. The

PAGE 166

166 mechanism of refractoriness of the early CL to PGF has not been elucidated completely. Tsai and Wiltbank (1998) suggested that early CL, in spite of having functional PGF receptors, is incapable of inducing intra luteal synthesis of PGF via prostaglandin endoperoxidase synthase 2, increasing expression of monocyte chemoatractant protein 1, and inhibiting progesterone production through StAR Additionally, Miyamoto et al. (2009) suggested a site restricted action of PGF depending on the stage of the estrous cycle. In the mid cycle CL (d 8 to 12), PGF induces an acute increase in blood flow in the periphery of the CL concurrent with expression of endothelial nitric oxide synthase, but the same phenomenon is not observed in the early cycle CL (d 4). Moreover, Atli et al. (2012) reported that although the initial pulse of PGF upregulate s mRNA expression of many pathways related to luteolysis, the second and later pulses of PGF are actually responsible for a distinct pattern of gene expression that results in luteolysis. Therefore, either multiple doses of PGF are needed to fully regr ess the newly formed CL, or the fact that the second dose of PGF was administered on d 6 after the GnRH might have allowed some CL to become more mature and responsive to the luteolytic effects of prostaglandin. It is important to indicate that when no G nRH was administered, as in control and NG2P, the second dose of PGF did not improve P/AI despite the increase in luteolysis. These results reinforce the need to combine follicle turnover with adequate CL regression to optimize fertility in the 5 d timed AI protocol for dairy heifers. Low ovulatory response to GnRH in dairy heifers has been demonstrated by others (Stevenson et al., 2008; Lima et al., 2011). One of the factors that might affect ovulation to GnRH is the diameter of the dominant follicle (Sa rtori et al., 2001), which is

PAGE 167

167 known to be related to the expression of LH receptors on granulosa cells (Xu et al., 1995). Granulosa cells acquired LH receptors at the time of follicle deviation, approximately 3 d after the emergence of a follicular wave, w hen the dominant follicle achieves 8.5 mm in diameter. However, most heifers in previous studies had follicles of at least 10 mm in diameter when GnRH was administered (Martinez et al., 1999; Lima et al., 2011). Another possibility is that approximately 50 % of the heifers have 3 waves of follicle development (Bisinotto and Santos, 2012), which limits the period of follicle dominance and the opportunity to have a dominant follicle responsive to a GnRH/LH surge to induce ovulation. Interestingly, dairy heifer s without a CL had greater ovulation to GnRH than heifers with a CL in a previous study (Lima et al., 2011). This finding suggests that high progesterone concentration might be another impediment to ovulation, particularly when heifers are in mid luteal ph ase and receive a CIDR as in the 5 d timed AI protocol. The negative effects of progesterone on LH release in response to GnRH have been demonstrated in beef heifers, mature beef cows, and dairy cows (Colazo et at., 2008; Dias et al., 2010; Giordano et al. 2012). The results of the current study confirm the hypothesis of negative effects of progesterone on LH release and ovulation in dairy heifers treated with GnRH. Therefore, the present study clarifies that induction of follicle turnover with GnRH in dai ry heifers subjected to the 5 d timed AI protocol is inhibited by elevated concentrations of progesterone, which attenuates LH release. Conclu sion Increased follicle turnover at initiation of 5 d timed AI program by using GnRH combined with two doses of P GF administered on d 5 and 6 to optimize luteolysis was a successful strategy to optimize P/AI in dairy heifers. Results of the current study

PAGE 168

168 demonstrate similar concepts of previous work with lactating dairy cows reinforcing the need for synchronization protocols to incorporate physiological principles to optimize fertility in dairy heifers. Follicle turnover by inducing ovulation with GnRH, although low in dairy heifers, was beneficial to fertility. However, the benefit of GnRH to optimize fertility req uires two doses of PGF administered 24 h apart to increase regression of a newly formed CL. The P/AI of approximately 60% obtained in the current study supports the use of the 5 d timed AI protocol as an alternative breeding program for reproductive management of heifers when detection of estrus is not used. Finally, it was demonstrated that high concentrations of progesterone when GnRH was administered suppressed the LH release and impaired ovulation. Further research is needed to determine if additional increase in ovulation to the initial GnRH of the 5 d timed AI protocol can further improve fertility in dairy heifers.

PAGE 169

169 Table 5 1. Effect of the initial GnRH and two doses of PGF on fertility responses of dairy heifers subjected to the 5 d timed AI program (Study 1) Treatment 1 Control G2P AOR (95% CI) 2 P ----------------% (n/n) ----------------Estrus at AI 3 76.7 (429/559) 76.9 (421/547) 1.03 (0.77 1.38) 0.83 Preg nant Day 32 53.5 (299/559) 59.4 (325/547) 1.28 (1.01 1.63) 0.04 Day 60 51.3 (287/559) 56.6 (309/546) 1.24 (0.98 1.58) 0.07 Pregnancy loss 4 4.0 (12/299) 4.6 (15/324) 1.15 (0.53 2.52) 0.72 1 Control = d 0 CIDR insertion, d 5 PGF and removal of CIDR d 8 GnRH and timed AI. G2P = d 0 GnRH and CIDR insertion, d 5 PGF and removal of CIDR, d 6 PGF d 8 GnRH and timed AI. 2 AOR = adjusted odds ratio, CI = confidence interval. Control is the reference for comparison. 3 Evaluated based on removal of ta il paint on the d of AI. 4 Calculated as the number of heifers that lost their pregnancies between d 32 and 60 after AI divided by the number of heifers pregnant on d 32. One pregnant heifer from G2P left the study before reconfirmation of pregnancy on d 6 0.

PAGE 170

170 Table 5 2. Effect of the initial GnRH and two doses of PGF on ovarian responses in dairy heifers subjected to the 5 d timed AI program (Study 2) 1 Control = d 0 CIDR insertion, d 5 PGF and removal of CIDR, d 8 GnRH and AI; NG2P = d 0 CIDR insertion, d 5 PGF and removal of CIDR, d 6 PGF d 8 GnRH and AI ; G2P = d 0 GnRH and CIDR insertion, d 5 PGF and removal of CIDR, d 6 PGF d 8 GnRH and AI. 2 3 Proportion of heifers with a new CL on d 5 independent of th a,b Different superscripts within the same row differ ( P c,d Different superscripts within the same row tend to differ ( P = 0.07). Treatment 1 Control NG2P G2P P Study d 0 ------------------------% (n/n) -------------------------95.2 (200/210) 95.1 (194/204) 91.0 (190/209) 0.76 Presence of CL 81.0 (170/210) 82.3 (168/204) 78.0 (163/209) 0.83 Ovulation 2 13.8 (29/210) b 11.8 (24/204) b 26.3 (55/209) a 0.001 Study d 5 Presence of CL 88.6 (186/210) 91.2 (186/20 4) 88.5 (185/209) 0.60 Presence of a new CL 3 21.9 (46/210) b 20.1 (41/204) b 34.4 (72/209) a 0.01 Progesterone on study d 8 < 1.0 ng/mL 4 95.1 (196/206) 97.0 (192/198) 97.5 (197/202) 0.39 < 0.5 ng/mL 83.0 (171/206) b 92.9 (184/198) ac 87.1 (176/202) d 0.0 1 < 0.3 ng/mL 62.6 (129/206) bd 74.7 (148/198) a 70.8 (143/202) c 0.02

PAGE 171

171 Table 5 3. Effect of the initial GnRH and two doses of PGF on fertil ity responses in dairy heifers subjected to the 5 d timed AI program (Study 2) Treatment 1 Control NG2P G2P P ----------------------------% (n/n) ----------------------------Estrus at AI 2 72.1 (513/711) b 74.3 (517/696) b 78.6 (559/711) a 0.01 Pre gnant 3 Day 32 52.8 (376/711) b 55.0 (383/696) b 61.7 (439/711) a 0.002 Day 60 48.9 (348/711) b 51.6 (359/696) b 59.1 (420/711) a 0.001 Pregnancy loss 4 7.4 (28/376) B 6.3 (24/383) AB 4.3 (19/711) A 0.15 1 Control = d 0 CIDR insertion, d 5 PGF and removal of CIDR, d 8 GnRH and AI; NG2P = d 0 CIDR insertion, d 5 PGF and removal of CIDR, d 6 PGF d 8 GnRH and AI; G2P = d 0 GnRH and CIDR insertion, d 5 PGF and removal of CIDR, d 6 PGF d 8 GnRH and AI. 2 Evaluated based on removal of tai l paint on the d of AI. 3 Twelve control heifers, 7 NG2P heifers, and 7 G2P heifers did not have a pregnancy diagnosis performed. 4 Calculated as the number of heifers that lost their pregnancies between d 32 and 60 after AI divided by the number of heife rs pregnant on d 32. a,b Different superscripts within the same row differ ( P A,B Different superscripts within the same row tended to differ ( P

PAGE 172

172 Table 5 4. Effects of concentration of progesterone in plasma on LH release and ovulation in response to GnRH in dairy heife rs (Study 3) 1 LP = low progesterone heifers received a CIDR insert and two PGF injections 12 h apart on d 7 after ovulation and were challenged with GnRH on d 8; HP = high progesterone heifers received a CIDR insert and were challenged with GnRH on d 8. 2 Progesterone and LH concentrations were evaluated in 18 heifers (6 LP and 1 2 HP). 3 Ovulation was evaluated in 94 heifers. Treatment 1 LP HP P Progesterone, ng/mL 2 3.14 0.88 7.35 0.67 0.001 LH response to GnRH, ng/mL Mean 16.3 2.3 10.1 1.8 0.04 Peak 31.3 2.3 19.2 4.1 0. 09 Minutes after GnRH to peak 81.4 14.4 76.3 11.1 0.78 Ovulation response to GnRH, % (n/n) 3 48.4 (15/31) 19.0 (12/63) 0.001

PAGE 173

173 Figure 5 1. Diagram of activities in study 1. All heifers received a CIDR on d 0, an injection of PGF and CIDR removal on d 5, and an injection of GnRH concurrently with AI on d 8. Control = no additi onal treatment (n = 559); G2P = additional injection of GnRH on d 0 and a second injection of PGF on d 6 (n = 547). AI = artificial insemination; CIDR = controlled internal drug release device containing 1.38 g of progesterone; PD = pregnancy diagnosis.

PAGE 174

174 Figure 5 2. Diagram of activities in study 2. All heifers received a CIDR on d 0, an injection of PGF and CIDR removal on d 5, and an injection of GnRH concurrently with timed AI on d 8. Control = no additional treatment (control = 711); NG2P = a second injection of PGF on d 6 (n = 696); G2P = an injection of GnRH on d 0 and a second injection of PGF on d 6 (n = 711). AI = artificial insemination; BS = blood sample for analysis of progesterone; CIDR = controlled internal drug release device co ntaining 1.38 g of progesterone; PD = pregnancy diagnosis.

PAGE 175

175 Figure 5 3. Diagram of activities in study 3. Heifers had their estrous cycle synchronized with PGF and those in estrus were evaluated for ovulation (study d 0). Heifers received a CIDR in sert on d 7 and were assigned to low progesterone (LP, n = 6) in which two injections of PGF were administered 12 h apart on d 7.0 and 7.5, or high progesterone (HP, n = 12) in which no PGF 8 a nd blood was sampled every 15 min from 30 to 180 min and at 240 and 300 min relative to the GnRH injection for assessment of LH concentrations. BS = blood sample; CIDR = controlled internal drug release containing 1.38 g of progesterone; GnRH = injection = injection of 25 mg of dinoprost as tromethamine salt.

PAGE 176

176 Figure 5 4. Effect of progesterone concentration on the LH release in response to relin hydrochloride; LP (Low progesterone) = d 7 after ovulation CIDR and PGF d 7.5 after ovulation PGF d 8 after ovulation GnRH. HP (High progesterone) = d 7 after ovulation CIDR, d 8 after ovulation GnRH. Blood samples for LH were collected each 15 minutes from 30 before to 180 minutes after GnRH injection a nd at 4 and 5 hours after GnRH.

PAGE 177

177 CHAPTER 6 EFFECTS OF ONE OR TWO TREATMENTS WITH PROSTAGLANDIN F 2 ON SUBCLINICAL ENDOMETRITIS AND FERTILITY IN LACTATING DAIRY COWS INSEMINATED BY TIMED AI The objectives of the current study were to investigate the efficacy of PGF as a therapy to reduce the prevalence of subclinical endometritis and improve pregnancy per artificial insemination (P/AI) in cows subjected to a timed AI program. A total of 1 ,342 lactating Holstein dairy cows were allocated randomly at 25 3 d in milk (DIM) to remain as untreated controls (control, n = 454), receive a single PGF treatment at 39 3 DIM (1PGF, n = 474), or two treatments with PGF at 25 3 and 39 3 DIM ( 2PGF, n = 414). All cows were enrolled in the double Ovsynch program at 48 3 DIM, and were inseminated at 75 3 DIM. A subset of 357 cows had uterine samples collected for cytological examination at 25 3, 32 3, and 46 3 DIM to determine the percen tage of polymorphonuclear leukocytes (PMNL). Subclinical endometritis w as 3 DIM and used to define the prevalence of purulent vaginal discharge. Body condition score was assessed at 25 3 DIM. Pregnancy was diagnosed 32 d after AI and reco nfirmed 28 d later. At 32 3 DIM, the prevalence of subclinical endometritis was reduced by treatment with PGF at 25 3 DIM in 2PGF (control = 23.5% vs. 1PGF = 28.3% vs. 2PGF = 16.7%); however, this benefit disappeared at 46 3 DIM and 14% of the cows remained with subclinical endometritis. One or two treatments with PGF did not influence P/AI on d 32 or 60 after timed AI, and they averaged 39.9 and 35.2%. Similarly, treatment with PGF had no effect on pregnancy loss between 32 and 60 d of gestatio n (11.9%). Cows diagnosed with both purulent vaginal discharge and subclinical endometritis had the lowest P/AI and the highest pregnancy loss compared

PAGE 178

178 with those diagnosed with only one of the two diseases or with cows having no diagnosis of uterine dise ases. Interestingly, subclinical endometritis depressed P/AI and increased pregnancy loss only when it persisted until 46 d in milk (DIM). On d 32 after AI, cows not diagnosed with subclinical endometritis and those that resolved subclinical endometritis b y 46 DIM had greater P/AI than those that remained with subclinical endometritis at 46 DIM (45.4 vs. 40.0 vs. 25.0%). Similar to P/AI, cows not diagnosed with subclinical endometritis and those that resolved subclinical endometritis by 46 DIM had less preg nancy loss than those with subclinical endometritis at 46 DIM (9.6 vs. 13.5 vs. 43.9%). One or two treatments with PGF before initiation of the timed AI program were unable to improve uterine health, P/AI, and maintenance of pregnancy in lactating dairy cows. Cows diagnosed with both purulent vaginal discharge and subclinical endometritis had the greatest depressions in measures of fertility at first AI, particularly when subclinical endometritis persisted in the early postpartum period. Introductory Remarks Uterine diseases are prevalent in dairy cows and they have been associated with reduced reproductive performance which ultimately affects herd profitability (Gilbert et al., 2005; LeBlanc, 2008). Uterine diseases are often classified according to clinical presentation and defined based on their impacts on pregnancy per AI ( P/AI ) or time to pregnancy (Sheldon et al. 2006). Among them, clinical endometritis is defined as presence of inflammation in the reproductive tract visible by the type of vaginal discharge that typically contains pus and persists after 21 DIM (Leblanc et al., 2002a; Sheldon et al., 2006). More r ecently, clinical endometritis as diagnosed by presence of pus in the vagina was classified as purulent vaginal discharge ( PVD ) because of the

PAGE 179

179 large proportion of cows without concurrent neutrophil infiltration in the endometrium (Dubuc et al., 2010). On t he other hand, a large proportion of cows not diagnosed with any clinical signs of uterine disease have presence of inflammatory cells in the endometrium, and usually more than 5% PMNL in endometrial cytology reduce P/AI and extends the interval postpartum to pregnancy (Gilbert et al., 2005; Galvo et al., 2009a). In the United States, no particular treatment is labeled for use in cows that have either PVD or subclinical endometritis, although use of intrauterine infusion of 500 mg of cephapirin as benzath ine has demonstrated efficacy in reducing interval to pregnancy in cows with PVD (Leblanc et al., 2002b) or improving pregnancy at first AI in cows with subclinical endometritis (Kasimanickam et al ., 2005). In those studies, cows were not subjected to stan dardized programs for first postpartum AI and many were inseminated following detection of estrus. When cows were subjected to a presynchronized timed AI program, use of intrauterine antibiotics did not benefit P/AI of dairy cows (Galvo et al., 2009b), ev en in those with previous diagnosis of PVD. An alternative therapy is the use of prostaglandin (PG) F in an attempt to induce estrus and eliminate bacterial contamination that might be causing the inflammatory process in the endometrium. Use of PGF in cows during diestrus results in lu teolysis and induces cows to return to estrus, which has been suggested to enhance uterine immunity by removal of immunosuppressive effects of progesterone (Lewis, 2004). Kasimanickam et al. (2005) suggested that the impro vements in P/AI caused by PGF in postpartum cows were caused by inducing estrus and concurrent opening of the cervix and myometrium contractions that might enhance mechanical cleansing of the endometrium.

PAGE 180

180 When PGF is administered in early lactation, it is possible that the benefi ts to fertility might not be related to enhancing uterine health, but confounded with effects of presynchronizing the estrous cycle before timed AI programs (Moreira et al., 2001; Galvo et al., 2007). It is known that the stage of the estrous cycle when c ows initiate timed AI protocols based on GnRH is critical for fertility (Vasconcelos et al., 1999), and treatment with PGF 11 to 12 d before the initiation of the timed AI increased P/AI (Moreira et al., 2001; Galvo et al., 2007). In fact, in most studies evaluating PGF as therapy for treatment of uterine diseases and subsequent impacts on fertility, uterine health was n ot evaluated after treatment to justify the increase in P/AI (Leblanc et al., 2002b; Kasimanickam et al., 2005). In some cases, when uterine health was evaluated after PGF treatment, P/AI at first AI improved, but the benefits were not linked to a reduct ion in the prevalence of subclinical endometritis in treated cows (Galvo et al., 2009a). Timed AI programs are commonly used for reproductive management of dairy herds for first and resynchronized inseminations to mitigate the negative impacts of poor est rous detection in lactating dairy cows (Caraviello et al., 2006). An alternative presynchronization treatment, in which stage of the estrous cycle is synchronized in cyclic and anovular cows, is called Double Ovsynch (Souza et al., 2008). When PGF is administered before the Double Ovsynch protocol, the effects on uterine health or measures of fertility are not expected to be mediated by altering the stage of the estrous cycle when cows are subjected to the timed AI protocol. The goal of the curre nt study was to demonstrate an improvement in P/AI in dairy cows with the systematic use of

PAGE 181

181 PGF by enhancing uterine health based on the reduction in the prevalence of subclinical endometritis. The hypotheses of the current study was that treatment with PGF would reduce the prevalence of subclinical endometritis and improve first service P/AI in cows subjected to a presynchronized timed AI program. Therefore, the objectives were to investigate the efficacy of systematic use of one or two treatments wit h PGF preceding a presynchronized timed AI protocol on the prevalence of subclinical endometritis and P/AI in lactating dairy cows. Materials and Methods The University of Florida Institute of Food and Agricultural Sciences Animal Research Committee appr oved all procedures in this study. Animals, Housing, and Feeding A total of 1,342 lactating Holstein cows from a commercial dairy farm located in north central Florida were used in this study. Cows enrolled in the study calved from August 2009 to July 201 0. Cows were housed in free stall barns equipped with fans and sprinklers for forced evaporative cooling. Cows from all treatments were kept together in the same pens throughout the entire period of the study. Lactating cow diets were formulated using the CPM Dairy cattle ration analyzer (Cornell Pen Miner Ver. 3.0.8) to meet or exceed the nutrient requirements established by NRC (2001) for lactating Holstein cows weighing 650 kg, consuming 24 kg of DM, and producing 45 kg/d of milk containing 3.5% fat and 3.1% true protein during the first 80 d of lactation. The first insemination for cows in the study occurred between November 2009 and October 2010.

PAGE 182

1 82 Reproductive Management All cows in the study had ovulation synchronized for first postpartum AI with the d ouble Ovsynch program starting on 48 3 DIM as depicted in Figure 3 1 (Souza et al., 2008). A total of 4 technicians and 18 sires were distributed randomly for all treatments. At 32 d after the first postpartum timed AI cows were diagnosed for pregnancy by ultrasonographic examination of the uterus and its contents. The presence of an embryo with a heartbeat was the criterion used to determine pregnancy. Cows diagnosed pregnant were re examined by palpation per rectum of the uterus and its contents 28 d late r, at 60 d of gestation, to reconfirm pregnancy status and to identify pregnancy loss. Treatments and Body Condition Scoring Weekly cohorts of cows at 25 3 DIM were blocked by parity and, within each block, allocated randomly to remain untreated (contro l, n = 454), or receive a single i.m. injection of 25 mg PGF (dinoprost tromethamine; Lutalyse sterile solution, Zoetis, Madison, NJ, USA) treatment at 39 3 DIM (1PGF, n = 474), or two treatments with PGF at 25 3 and 39 3 DIM (2PGF, n = 414), as depicted in Figure 6 1. The body condition of all cows was a ssessed at 25 3 DIM using a 1 (emaciated) to 5 (obese) scale according to Ferguson et al. (1994) as depicted in the Elanco BCS chart (Elanco, 2009). Evaluation of Uterine Health Samples of vaginal discharge and uterine endometrial cytology were collecte d from a subset of 357 cows (control = 115; 1PGF = 125; and 2PGF =117). All samples were collected by the investigators who were blinded to treatments. Vaginal discharge retrieved using the Metricheck (Metricheck, Simcro, New Zealand) at 25 3 DIM was use d as a criterion to determine PVD (Dubuc et al., 2010), formerly known and

PAGE 183

183 classified as clinical endometritis (Sheldon et al., 2006). Briefly, vaginal discharge was scored as: 1 = clear or translucent mucus; 2 = mucus containing flecks of white or off whi te pus; 3 = discharge containing 50% or less white or off white mucopurulent material; 4 = discharge containing more than 50% purulent material, usually white or yellow; and 5 = bloody, purulent and fetid discharge. Cows with score > 2 were classified as h aving PVD. Uterine cytology samples were collected on d 25 3, 32 3 and 46 3 postpartum using the cytobrush technique (Kasimanickam et al., 2005) with the stainless steel gun protected by a one way plastic tube protector (Continental plastics, Delav al, WI). The DIM at sampling were selected to be able to evaluate the effects of treatments with PGF before cows were enrolled in the Double Ovsynch protocol at 48 DIM. The evaluations on d 32 and 46 postpartum were to maintain the same interval of 7 d between each PGF treatment and the endometrial cytology. After collecting the endometrial cytology the cytobrush was rolled onto a slide and air dried immediately. The slides were transported to the laboratory and stained using diff quick stain kit (IMEB, San Marcos, CA). Three technicians not aware of treatments read the slides. Two hundred cells wer e counted in each slide using a microscope at 400 x magni fications to determine the proportion of PMNL relative to the total of PMNL, mononuclear, and endometrial cells counted. Cows with a proportion of Statistical Analys i s Th e sample size was calculated using Minitab 15 (Minitab Inc., State College, PA) for a two

PAGE 184

184 endometritis would affect 30% of the cows and that treatment with PGF would reduce the prevalence of subclinical endometritis by 12 percentage units. Under those assumptions, 120 cows/treatment were needed to evaluate the effects of treatment on the prevalence of subclinical endometritis. For P/AI in all cows, the sample size was calculated based on a n increase in at least 6 percentage units. Others have demonstrated that administration of PGF to postpartum cows inseminated on estrus or following a combination of detected estrus and timed AI had increments in pregnancy at first postpartum AI of 12 to 15 percentage units (Kasimanickam et al., 2005; Galvo et al. 2009a). A maximum of 433 cows/t reatment was calculated to allow for detection of statistical effect when the difference between treatments was of at least 6 percent units in P/AI. Binary data such as prevalence of subclinical endometritis and P/AI were analyzed by logistic regression us ing the GLIMMIX procedure of SAS version 9.3 (SAS/STAT, SAS Institute Inc., Cary, NC, USA) and fitting a binary distribution. The models included the effects of treatment (control vs. 1PGF vs. 2PGF), parity (primiparous vs. multiparous), BCS categorized as breeding classified as cool, when AI occurred from October 1 to May 14 or hot, when AI occurred from May 15 to September 30. For subclinical endometritis, the prevalence on d 25 postpartum was used as covariate. For P/AI, technician and sire were also included in the statistical models. Treatment was forced in the final models. Covariates and interactions between treatment and covariates were maintained in the statistical models if P < 0.10. Orthogonal contrasts were perfor med to determine the impact of PGF (control vs. 1PGF + 2PGF), and number of PGF treatments (1PGF vs. 2PGF).

PAGE 185

185 Two additional multivariable analyses were performed with the subset of cows in which uterine health was evaluated to model P/AI and pregnancy loss. In the first model, cows wer e classified based on uterine health as no uterine disease, when no PVD or subclinical endometritis was diagnosed any time in the first 46 DIM, or as having PVD only, subclinical endometritis only, or both PVD and subclinical endometritis. A cow was consid ered positive for subclinical endometritis if it was present at least once in the evaluations at 25, 32, and 46 DIM. The models included the effects of treatment (control vs. 1PGF vs. 2PGF) and uterine health (no uterine disease vs. PVD only vs. subclinica l endometritis only vs. PVD and subclinical endometritis). Orthogonal comparisons were performed to evaluate the effect of uterine health (no uterine disease vs. all others), the differential effect of PVD compared with subclinical endometritis (PVD only v s. subclinical endometritis only), and the additive effect of PVD and subclinical endometritis (PVD only + subclinical endometritis only vs. both PVD and subclinical endometritis). In the second model, cows were classified only based on subclinical endomet ritis as never being diagnosed with subclinical endometritis, having resolved subclinical endometritis when diagnosed on d 25 and/or 32, but negative on d 46 postpartum, and those that persisted with subclinical endometritis based on diagnosis on d 46 post partum. The models for P/AI and pregnancy loss included the effects of treatment (control vs. 1PGF vs. 2PGF) and subclinical endometritis (no subclinical endometritis vs. subclinical endometritis that resolved by d 46 vs. subclinical endometritis on d 46). Orthogonal comparisons were performed to evaluate the effect of subclinical endometritis (no subclinical endometritis vs. resolved + persistent), and the effect of persistent subclinical endometritis (resolved vs. persistent).

PAGE 186

186 Adjusted proportions for bin ary data were generated by back transforming the estimates using the ilink function of SAS. Proportions are displayed for binary data, tendency to differ. Results Cows in the three treatments had similar lactation nu mber (2.93 0.05), DIM at enrolment in the study (24.7 0.05), DIM at AI (81.0 0.2), vaginal discharge score on d 25 postpartum (2.08 0.06), and percentage of PMNL in endometrial cytology on d 25 postpartum (6.98 0.67); however, BCS at enrollment w as greater ( P = 0.01) for control and 2PGF than 1PGF cows (control = 2.96 0.02 vs. 1PGF = 2.91 0.02 vs. 2PGF = 2.98 0.02). Effects of PGF Treatments on the Prevalence of Subclinical Endometritis The prevalence of PVD on d 25 postpartum tended ( P = 0.10) to be greater for cows in the 2PGF than those receiving 1PGF, whereas control cows had intermediate prevalence that did not differ from t he other two groups (Figure 6 2). Nevertheless, the prevalence of subclinical endometritis in cows on d 25 postpartum, before treatments were applied, did not differ among treatments and averaged 29.5%. Treatment with PGF at 25 3 DIM in 2PGF reduced ( P < 0.05) the prevalence of subclinical endometritis on d 32 postpartum compared with control and 1PGF cows (Figure 6 2). However, there was no difference in the prevalence of subclinical endometritis at 46 DIM. Cows with PVD on d 25 postpartum had greater ( P < 0.001) prevalence of subclinical endometritis than those without PVD on d 25 (47.3 vs. 17.8%) and 32 (40.3 vs. 16.8%) postpartum, but the same association was not observed on d 46 postpartum

PAGE 187

187 (PVD = 17.4 vs. no PVD = 12.1%; P = 0.20). Additionally, no association was observed between parity, BCS and season of AI with the prevalence of subclinical endometritis. Effects of PGF on Pregnancy per AI and Pregnancy Loss Treatments with 1PGF or 2PGF failed to increase P/AI on d 32 and 60 after insemination in cows bred exclusively after subjected to the double Ovsynch timed AI program (Table 6 1). Overall, 39.9 and 35.2% of the cows were pregnant on d 32 and 60 after AI, respectively. Pregnancy loss between 32 and 60 d after AI affected 11.9% of the pregnant c ows and treatment with either 1PGF or 2PGF had no influence on maintenance of pregnancy in the first 60 d of gestation. Parity and season of AI affected ( P < 0.01) P/AI on d 32 and 60 after timed insemination. Primiparous cows had greater ( P = 0.01) P/AI than multiparous cows on d 32 (42.5 vs. 35.0%) and 60 after AI (37.1 vs. 30.5%). Cows inseminated during the cool season had greater ( P < 0.001) P/AI than those inseminated during the hot season on d 32 (47.1 vs. 30.8%) and 60 after insemination (41.7 vs. 26.6%). None of the other covariates evaluated influenced pregnancy loss between 32 and 60 d of gestation. Associations Among PVD and/or Subclinical Endometritis with Measures of Fertility The negative impacts of uterine diseases on P/AI and maintenance o f pregnancy were only observed when cows were diagnosed with both PVD and subclinical endometritis (Table 6 2 ). On d 32 and 60 after insemination, cows not diagnosed with uterine diseases had greater ( P < 0.05) P/AI than those diagnosed with both PVD and s ubclinical endometritis. However, P/AI did not differ statistically among cows not diagnosed with uterine diseases and those diagnosed with only PVD or subclinical endometritis. Interestingly, cows diagnosed with both diseases had lower ( P = 0.03)

PAGE 188

188 P/AI on d 60 than those diagnosed with only either PVD or subclinical endometritis. Similar to P/AI, pregnancy loss increased ( P = 0.05) only when cows were diagnosed with both PVD and subclinical endometritis. Subclinical endometritis depressed P/AI and increase d pregnancy loss in dairy cows, but these negative effects were only observed when the disease persisted until 46 DIM (Table 6 3). Of the 50 cows categorized as persistent subclinical endometritis remaining 31 had a previous diagnosis. Cows with no diagnosis of subclinical endometritis had similar P/AI on d 32 or 60 after insemination compared with cows diagnosed with subclinical endometritis that resolved by 46 DIM. Nevertheless, those cows in which subclinical endometritis persisted until 46 DIM, immediately before enrollment on the Double Ovsynch protocol tended ( P = 0.08) to have lower P/AI on d 32 and had lower P/AI ( P = 0.01) on d 60 after insemination because of greater ( P = 0.04) pregnancy loss than cows that resolved subclinical endometritis by 46 DIM. Discussion Treatments with one or two doses of PGF before enrollment in a timed AI protocol had minor impacts on the prevalence of subclinical endometritis and did not improve P/AI or reduce pregnancy loss in lactating Holstein cows. The design of the current experiment allowed the evaluatio n of the effects of 1 or 2 doses of PGF on measures of fertility in dairy cows while excluding the confounding effect of presynchronization of the estrous cycle when cows are subjected to timed AI protocols (Moreira et al., 2001; Galvo et al., 2007), o r inducing earlier insemination because of estrus. The current study was designed to evaluate the effects of systematic use of PGF on the prevalence of subclinical endometritis and P/AI, but it was not our aim to

PAGE 189

189 assess PGF as a direct therapy for cows with PVD or only those diagnosed with subclinical endometritis. The strategic use of PGF early postpartum, when PVD and subclinical endometritis are prevalent (Gilbert et al., 2005), was initially thought to reduce the prevalence of subclinical endometr itis, which could improve P/AI in cows inseminated exclusively by timed AI. The suggested mechanism of PGF action in cyclic cows involves induction of luteolysis and return to estrus leading to opening of the cervix and myometrium contractions that might improve mechanical cleansing of the uterus by eliminating bacteria and the products that attract PMNL. However, PGF may have effects on the uterus beyond induction of luteolysis and estrus in cyclic cows. In human uterine tissue in vitro PGF has been shown to induce myometrium contractions (Senior et al., 1992). Ulug et al. (2001) demonstrated that PGF stimulated the release of pro matrix metalloproteinase 2 and pro matrix metalloproteinase 9 from uterine tissue explants that are involved with break down of extracellular matrix mainly by degrading collagen type IV, which might aid myometrium contraction and uterine involution independent of cyclic status. Administration of exogenous PGF early postpartum promoted uterine involution in cows (Lindell a nd Kindahl, 1983). In the current study, reduction in the prevalence of subclinical endometritis only occurred at 32 DIM in cows receiving PGF on d 25 postpartum, when the proportion of estrous cyclic cows and luteolytic response to PGF are typically l ow (Galvo et al., 2010). The positive effect of PGF on subclinical endometritis was no longer apparent by 46 DIM, probably because of the observed high spontaneous cure documented in control cows. In fact, a single treatment with PGF at 39 DIM resulte d in no benefit in reducing the prevalence of

PAGE 190

190 subclinical endometritis compared with control cows. Therefore, the initial benefits of PGF on uterine health of earlier postpartum cows were offset by spontaneous resolution of uterine inflammatory process 2 weeks later. Interestingly, PGF as administered was unable to reduce the prevalence of subclinical endometritis at 46 DIM, which was observed to have marked effects on P/AI and maintenance of pregnancy at first postpartum AI. One of the potential limita tions of the study is that estrous cyclic status and presence of CL on the days when PGF was administered was unknown, so it is possible that a high prevalence of anovular cows would have limited the benefit of PGF on improving uterine health through i nduction of luteolysis. Nevertheless, Dubuc et al. (2011) observed that administration of PGF on d 35 and 49 postpartum did not improve P/AI at first service or reduced time to pregnancy in estrous cyclic cows. The effects of PGF on uterine health and fertility have been evaluated by others in cows subjected to varying reproductive management at first AI (Kasimanickam et al., 2005; Galvo et al., 2009a; Dubuc et al., 2011). In agreement with our findings, administration of PGF given 14 d apart with th e last treatment at 49 DIM (Galvo et al. 2009a; Dubuc et al. 2011) did not reduce the prevalence of subclinical endometritis in dairy cows. Kasimanickam et al. (2005) observed that cows with subclinical endometritis treated with PGF had increased P/AI at first AI and pregnancy rate compared with untreated controls. The benefits of PGF on fertility were similar between treatment with PGF or with intrauterine administration of cephapirin suggesting an effect on the uterine microbiota and improved uter ine health, although these responses were not verified. The lack of reduction in the prevalence of subclinical endometritis at 46 DIM

PAGE 191

191 with PGF treatments supports the fact that PGF did not improve P/AI or reduce pregnancy loss in cows bred exclusively by timed AI. The double Ovsynch timed AI protocol presynchronizes the estrous cycle of dairy cows (Ribeiro et al., 2012 c ), thereby eliminating a potential effect of prior presynchronization with PGF at improving response to the timed AI program. On the o ther hand, one cannot completely exclude the possibility that, by imposing the double Ovsynch protocol for first AI, the benefits of PGF might have been reduced as cows receive additional hormonal interventions for synchronization of ovulation that might have effects on the uterus through induction sequential estruses and ovulations. Cows with subclinical endometritis had reduced P/AI which corroborates with previous reports (Kasimanickam et al., 2005; Galvo et al., 2009a; Dubuc et al., 2011) that iden tified a negative association between PVD and/or subclinical endometritis and fertility. In the current study cows that persisted with subclinical endometritis at 46 DIM had a remarkable reduction in P/AI in comparison with healthy cows or cows that resolv ed subclinical endometritis by 46 DIM. It is clear that the prevalence of subclinical endometritis decreases with day postpartum, but those that persisted until enrollment in the timed AI protocol suffered reductions in P/AI and had increased risk of pregn ancy loss. It is unknown if the prevalence of subclinical endometritis reduced even further after 46 DIM. With the exception of Galvo et al. (2009a) that observed a tendency for increased pregnancy loss in cows with PVD, none of the previous studies repor ted an association between PVD or subclinical endometritis and increased risk of pregnancy loss (Kasimanickam et al., 2005; Gilbert et al., 2005; Dubuc et al., 2011). In some studies, pregnancy was evaluated only once, so data for pregnancy loss were not

PAGE 192

192 a vailable (Kasimanickam et al., 2005; Gilbert et al., 2005). It is unclear the exact mechanism by which subclinical endometritis decreases P/AI and increases pregnancy loss. Products of endometrial inflammation compromised early embryo development in vitro (Hill and Gilbert, 2008). Soto et al. (2003) suggested that mediators of the inflammatory cascade, including cytokines can impair early embryo development and might be part of the mechanism by which fertility is depressed in cows suffering from inflammator y diseases in early lactation. There is mounting evidence that cows with subclinical endometritis have altered embryo quality and endometrial function. Inflammation in the endometrium has been shown to reduce fertilization in single ovulating postpartum da iry cows (Cerri et al., 2009 c ). Dairy cows with no detectable PMNL in endometrial cytology had increased number of transferable embryos when subjected to superstimulation compared with cows with presence of PMNL in endometrial cytology (Drillich et al., 20 12). Cows diagnosed with subclinical endometritis have altered endometrial and embryonic gene expression that might explain the reduced fertility (Hoelker et al., 2012). Endometrium from cows diagnosed with subclinical endometritis had an altered pattern o f expression of genes involved in cell adhesion and immune modulation, which was then linked to changes in d 7 embryo gene expression. The changes in endometrial gene expression might be induced by altered number of immune cells present in the tissue. Neve rtheless, embryos from cows with subclinical endometritis had altered pattern of gene expression involving pathways in cell cycle and apoptosis, which might explain a reduction in P/AI or even increased risk of pregnancy loss (Hoelker et al., 2012). Whethe r subclinical endometritis per se is the causative agent of changes in

PAGE 193

193 endometrial and embryonic gene expression or that cows that develop subclinical endometritis have underlying factors that also cause changes in the transcriptome remain to be elucidated It is noteworthy that the combination of PVD and subclinical endometritis led to additive negative effects on P/AI and pregnancy loss in dairy cows when compared with PVD and subclinical endometritis alone. Dubuc et al. (2011) reported a decline in first service P/AI in cows suffering from both PVD and subclinical endometritis when compared with cows diagnosed with only subclinical endometritis. In the same study, cows with both PVD and subclinical endometritis had longer interval to pregnancy compared wi th counterparts diagnosed with only one of the two problems. Similar to our findings, Dubuc et al. (2011) also showed that cows that persist with uterine disease before the end of the voluntary waiting period are those that suffer the greatest negative con sequences to interval to pregnancy. In the current study, it was observed that 14% of the cows had subclinical endometritis by 46 DIM, and these cows suffered the most losses in fertility at first postpartum AI. Similar to subclinical endometritis, Dubuc e t al. (2011) observed that PVD can also persist in some cows. According to their data, 42% of the cows diagnosed with PVD and subclinical endometritis on d 35 postpartum persisted with PVD at 56 DIM. The persistence of PVD was not evaluated in the current study. The mechanisms by which some cows are unable to eliminate inflammation from the uterus are not completely elucidated; however, previous studies suggest that endometrial inflammation is regulated by immune response rather than pathogen load (Herath e t al., 2009). It is possible that cows with inadequate immune function are those with longer duration of the endometrial inflammatory process that compromises fertility.

PAGE 194

194 Conclusions Treatment with one or two injections of PGF in early lactation before cows were subjected to a presynchronized timed AI protocol was unable to improve uterine health and measures of fertility in lactating dairy cows. Subclinical endometritis impaired P/AI and maintenance of pregnancy in lactati ng dairy cows, particularly when associated with PVD, and the negative effect of subclinical endometritis was observed when the inflammatory process persisted until 46 DIM. Interestingly, when both PVD and subclinical endometritis were associated or when s ubclinical endometritis persisted by 46 DIM, pregnancy loss increased.

PAGE 195

195 Table 6 1. Effect of one or two treatments of PGF on pregnancy per AI and pregnancy loss of dairy cows subjected to a timed AI program Treatment 1 P 2 Item Control 1PGF 2PGF TRT C1 C2 Pregnant d 32 38.1 (173/454) 40.7 (193/474) 41.1 (170/414) 0.58 0.32 0.72 d 60 33.7 (153/454) 36.7 (174/474) 35.0 (145/414) 0.70 0.43 0.78 Loss 11.6 (20/173) 9.8 (19/193) 14.7 (25/170) 0.36 0.87 0.16 1 Control = no treatment with PGF before enrollment in the timed AI protocol; 1PGF = a single treatment with PGF on d 39 postpartum; 2PGF = treatment with PGF on d 25 and 39 postpartum. 2 TRT = effect of treatment; C1 = contrast for the effect of treatment with PGF (control vs. 1PGF + 2PGF); C2 = contrast for the effect of number of treatments with PGF (1PGF vs. 2PGF).

PAGE 196

196 Table 6 2. Association between purulent vaginal discharge (PVD) and/or subclinical endometritis (SCE) with fertility of dairy cows at first postpartum insemination a,b,c Different superscripts within a row differ ( P < 0.05). A,B Different superscripts within a row tend to differ (P < 0.10). 1 No disease = no diagnosis of PVD or SCE; PVD = vaginal discharge score > 2 on d 25 3 postpartum. SCE = cows with e days in which diagnosis was performed (25 or 32 or 46 3 d postpartum). 2 C1 = effect of uterine disease (no uterine disease vs. PVD only + SCE only + PVD and SCE); C2 = effect of PVD compared with SCE (PVD only vs. SCE only); C3 = additive effect of PV D and SCE (PVD only + SCE only vs. PVD and SCE). Category 1 Contrasts 2 No disease PVD only SCE only PVD and SCE C1 C2 C3 Cows, n 156 22 105 74 ------Pregnant, % d 32 48.0 a 49.1 a,b 39.8 a,b 33.4 b 0.21 0.43 0.17 d 60 43.3 a 44.9 a 34.0 A 22.8 b, B 0.08 0.34 0.03 Loss, % 9.1 b 7.1 b 13.7 B 30.3 a, A 0.33 0.52 0.05

PAGE 197

197 T able 6 3. Association between subclinical endometritis (SCE) with fertility of dairy cows at first postpartum insemination postpartum insemination 1 Diagnosis of subclinical endometritis was based on uterine cyto diagnosed with SCE on d 25 and/or 32 postpartum, but negative on d 46 postpartum. Persistent SCE = cows with presence of SCE on d 46 postpartum. 2 C1 = effect of subclinical endometritis (No SCE vs. Resolved S CE + Persistent SCE); C2 = Effect of persistent SCE (Resolved SCE vs. Persistent SCE). Category 1 Contrasts 2 No SCE Resolved SCE Persis tent SCE C1 C2 Cows, n 178 129 50 Pregnant, % d 32 45.4 40.0 25.0 0.03 0.08 d 60 40.5 34.3 13.7 < 0.01 0.01 Loss, % 9.6 13.5 43.9 0.03 0.04

PAGE 198

198 Figure 6 1. Diagram of treatments according to d ays in milk ( 3). Treatments were control, with no administration of PGF 1PGF in which cows received a sin gle injection of PGF on d 39 postpartum, and 2PGF in which cows received an injection of PGF on d 25 and another on d 39 postpartum. All cows were inseminated at fixed time following the double Ovsynch protocol. MS = mucus score; UC = uterine cytology.

PAGE 199

199 Figure 6 2. Effect of one or two treatments of PGF on the prevalence of subclinical endometritis on d 32 and 46 postpartum in dairy cows. The analysis included the effect of one or two treatments of PGF on the prevalence of subclinical endometritis on d 32 and 46 postpartum in dairy cows. Control = no treatment with PGF before enrollment in the timed AI protocol; 1PGF = a single treatment with PGF d 25 and 39 postpartum. Purulent vaginal dischar ge (PVD) and subclinical endometritis on d 25 postpartum were the baseline prevalence at enrollment in the study. For subclinical endometritis on d 32 and 46, the analysis included the effects of treatment (P = 0.98), day postpartum (P = 0.001), and intera ction between treatment and day postpartum (P = 0.04). a,b different superscripts among treatments denote statistical difference (P < 0.05). A,B different superscripts among treatmen ts tended to differ (P < 0.10).

PAGE 200

200 CHAPTER 7 EFFECTS OF INTRAUTER INE INFUS ION OF TRUEPERELLA P YOGENES ON ENDOMETRIAL m RNA EXPRESSION OF GE NES ASSOCIATED WITH LUTEOLYSIS AND CORPUS LUTEUM LI FESPAN IN DAIRY COWS Objectives were to determin e the effects of intrauterine infusion of Trueperella pyogenes on endometrial expression of genes affecting luteolysis and luteal lifespan. E strous cycle s were synchronized on 32 healthy Holstein cows. O n d 4 after ovulation cows were allocated randomly to receive one of three treatments: TP (n=13), intrauterine infusion of 10 mL of saline soluti on containing 10 9 CFU /mL of T pyogenes ; TNF (n=9), intrauterine necrosis infusion of 10 mL of saline solution. Five cows per treatment had uterine biopsies collected at 6, 12 and 24 h after treatment to evaluate th e endometrial expression of i prostaglandin E synthase (PGES), prostaglandin F synthase (PGFS) and oxytocin receptor (OXR). The remaining cows had ova ries scanned and blood samples collected for progesterone evaluation. R eal time quantitative PCR (RT PCR) was used to measure gene expression. E xpression s of IL6, PGES, PGFS and OXR genes were not affected by treatment, time or their interaction. Interleukin 1B gene expression was not different for treatment and time, but there was an interaction between treatment and time. Cows receiving TP had increased expression of IL1B than TNF cows at 24 h. Moreover, lL6 expression tended to be greater for TP than control at 12 h. O xytocin receptor gene expression tended to be gr eater for TP and TNF than for control cows at 12 h. No difference in mean concentration of progesterone and CL size occurred among treatments. Although no differences in mean time for luteal lifespan were detected, the percentage of cows with early demise of the CL was greater for TP cows than control

PAGE 201

201 (42.9% vs 0). In conclusion, although intrauterine infusion of T pyogenes led to early demise of CL in greater percentage of cows, the differences on endometrial expression of genes associated with inflammat ion and luteolysis were minor. Introductory Remarks Uterine diseases affect nearly half of the dairy cows after parturition leading to disruption of uterine and ovarian function which frequently results in hindered fertility, increased involuntary culling and remarkable economic losses for dairy producers (Sheldon et al., 2009). The economic losses caused by metritis alone are striking and it has been calculated at $380 per affected cow due to reduced milk production, delayed conception, treatment and incre ased culling ( Drillich et al., 2001). Thus, if we consider a conservative incidence rate of 20% for the 8.5 million dairy cows in US the annual cost of metritis alone is $6 46 million, which does not include endometritis another presentation of uterine dise ases with remarkable detrimental effects on fertility ( Dubuc et al., 2011 ). Therefore, understanding the mechanism by which microbes subvert host innate immunity disrupting ovarian and uterine function is fundamental to develop ing preventatives to mitigate the negative impacts of uterine diseases. Trueperella pyogenes is considered one of the most relevant pathogens involved in uterine diseases, especially endometritis. This is due to its relative ly high prevalence in the environment, persistence in the ut erus, severity of lesions on the endometrium, resistance to treatment, and synergistic action with gram negative anaerobes (Ruder et al., 1981; Huszenicza et al., 1999; Mateus et al., 2002a, b; Williams et al., 2005,). However, the mechanism by which T. py ogenes affects the endometrium and reproductive events in dairy cows such as length of the estrous cycle and concentration

PAGE 202

202 of ovarian steroids remain elusive (Williams et al., 2007; Kaneko and Kawakami, 2009; Kaneko et al., 2013). Several studies in recen t years have reported that intrauterine (IU) infusion of live T. pyogenes disrupts luteal development leading to early demise of the CL and ovulation of dominant follicle of first follicular wave (Kaneko and Kawakami, 2007; Kaneko and Kawakami, 2007; Kanek o et al., 2013). Cows receiving an intrauterine infusion of T. pyogenes on d 3 after ovulation had a peak of prostaglandin F metabolite (PGFM) 3 d later, followed by the regression of the newly formed CL and ovulation of the dominant follicle of the first follicular wave in approximately 50% of the time (Kaneko and Kawakami, 2008; Kaneko and Kawakami, 2009; Kaneko et al., 2013). The mechanism by which T. pyogenes disrupts normal luteal function leading to short cycles is still unclear. Culture of endometria l cells with a bacteria free filtrate of T. pyogenes induces synthesis of PGF (Miller et al., 2007). This bacterium possesses a number of virulence factors that may contribute to its pathogenic potential. One of the most important is a cholesterol dependent cytolysin, pyolysin, which is a haemolysin cytolytic for macrophages (Jos t and Bilington, 2005). A second important virulence factor is peptidoglycan, which is a pathogen associated molecular pattern that induces pro interleukin 6 (Timmerman, et al., 199 3; Stewart et al., 200 3; Bromfield and Sheldon, 2011) that can stimulate endometrial synthesis of PGF (Davidson et al., 1995; Hansen et al., 2004; Skarzynski et al., 2000). However, a possible stimulation of inflammatory mediators and its direct relation ship with luteolytic cascade factors has never been investigated with an in vivo model of intrauterine induced infection with T. pyogenes

PAGE 203

203 Therefore, a study investigating the possible molecular mechanisms by which intrauterine infusion with live T. pyogen es leads to shortening of luteal phase in dairy cows is still n eeded. Our hypothesis is that intrauterine inoculation of live T. pyogenes in cows with a newly formed corpus luteum would increase endometrial expression of genes affecting the inflammation an d the luteolytic cascade leading to an acute endometrial production of PGF and early demise of the newly formed CL. The objectives of this study wer e to determine the effects of intrauterine infusion of T. pyogenes in dairy cows with newly formed CL on e ndometrial mRNA expression of genes affecting the luteolytic cascade, plasmatic concentration of progesterone, PGFM and CL lifespan M aterials and M ethods The University of Florida Institutional Animal Care and Use Committee approved the use of all animals procedures conducted in this study. Animals, Housing and Diets The study was conducted between November of 2011 and March of 2012 in the University of Florida Dairy Unit (Hague, FL). Thirty two lactating Holstein cows w ere enrolled in the study in two ex periments Cows were housed in freestall barns with sand bedded stalls equipped with sprinklers and fans for forced evaporative cooling and ventilation. Cows were fed twice daily, immediately after the morning milking at 0830 h and again at 1230 h. Diets w ere mixed twice daily as base mixture containing corn silage, alfalfa hay and a base concentrate mix, and the additional grain supplement. This base mixture contained 54% forage and was designed to meet the nutrient needs of a 650 kg cow consuming 23 kg of diet DM and producing 40.0 kg of milk with 3.5% fat and 3.0% true protein (CPM Dairy ver. 3.0.10 software; www.cpmdairy.net ).

PAGE 204

204 Study Design and Treatments At 21 3 day postpartum Holstein cows free of calving rela ted disorders (dystocia, stillbirth, twins, retained placenta or metritis) had their estrous cycle GnRH i.m. (2 mL Cystorelin, Merial Ltd., Duluth, GA) followed 7 d later by one injection of PGF (25 mg of dinoprost tromethami ne, Lutalyse, Zoetis Animal Health, Madison, NJ) at 283 days postpartum. Two days later at day 30 3 postpartum a second dose of GnRH was given to in duce ovulation, as depicted in F igure 4 1. Ovarian structures were scanned at days 21 3, 28 3, 30 3 and 32 3 to determine follicle turnover, luteolysis and ovulation. Additionally, at day 30 3 postpartum all cows had two uterine cytology samples collected for evaluation of subclinical endometritis and bacterial culture growth using the cytobrush technique as previously described with minor modifications (Kasimanickam et al., 2004). Briefly, for sample collection the vulva disinfected with alcohol and iodine wipes and dried with paper towel before having the cytology tool passed through the cervix The autoclaved cytology tool was protected with a disposable sheath protector (Continental Plastic Corp., Delavan, WI) during vaginal transit The first uterine cytology sample collected was placed directly into a transport media vial (BD Diagnostic Syst ems, Sparks, MD, USA) for safe transportation of the biological specimen from the farm to the microbiology laboratory of the Department of Animal Sciences where samples were streaked onto a blood agar plate and cultured aerobically for 48 hours. The cultur e plates were evaluated for microbial growth 48 hours later and questionable results were submitted to the College of Veterinary Medicine Diagnostic Laboratory at University of F lorida for further evaluation. For t he second uterine sample the cytobrush wa s directly smeared onto the slide for cytologic evaluation of subclinical endometritis.

PAGE 205

205 Briefly, the percentage of polymorphnuclear leukocytes (PMNL) was determined after a total of 200 cells were counted with the microscope using 400 times magnification a nd subclinical endometritis at this stage (Kasimanickam et al., 2004). The criteria for inclusion of cows in the study were occurrence of ovulation in response to GnRH given on day 30 3 postpartum, negative culture for microbial growth and < 10% PMNL in the uterine cytology examination. The 32 eligible cows were allocated randomly (divide d in to experiment 1 and experiment 2.) to receive one of three treatments on day 4 after ovul ation (35 3 day postpartum): TP (n=13), intrauterine infusion of 10 mL of sterile saline solution containing 10 9 colony forming unit (CFU)/mL of T. pyogenes ; TNF (n=9), intrauterine infusion of 10 mL control (n=10), intra uterine infusion of 10 mL of saline solution. The T. pyogenes used in this study was isolated from a cow diagnosed with clinical metritis at the Dairy Un it confirmed by PCR and the bacteria cultivated and prepared in batches of 10 9 CFU by the College of Veterinary Medicine Diagnostic Laboratory at University of Florida for intrauterine infusion. The treatment TNF was added to provide a positive control to the study. The cyto been reported as a potent luteolytic agent capable of shortening the length of the estrous cycle when administered intravenously (Skarzynski et al., 2003b; Skarzynski et al., 2009). At the current study the treatment were gi ven intrauterine to have a consistent site of application for all treatments

PAGE 206

206 Experiment 1 A subset of 5 cows per treatment was enrolled at experiment 1 to collect endometrial sample for evaluat ion of mRNA expression of genes associated with the luteolytic cascade and to characterize tissu e inflammation post intrauterine infusions in all the treatments A sample for biopsy was not collected for all cows because we aimed to avoid any possible effects of the intervention and the potential inflammation generate d for it on luteal lifespan and conce n trations of progesterone and PGFM Uterine Tissue Sample for Biopsy Uterine tissue was collect by an uterine forceps instrument passed through the cervix and positioned in the mid portion of the uterine horn ipsilatera l to the CL The instrument was pressed against the uterine wall and a sample weig hing between 100 and 250 mg was collected at 6, 12 and 24 hours after treatments administered on day 4 post ovulation for all cows. Two set of samples were collected at hour 6 post intrauterine infusion of treatments. The first set of samples at hour 6 and the samples coll ected at 12 and 24 hours post intrauterine infusion of treatments were immediately placed in a cryovial and snap frozen in liquid nitrogen until arrival at t hat laboratory, when samples were stored at 80 o C until RNA extraction and PCR analysis. The frozen samples were later evaluated to measure the gene expression of pro inflammatory cytokines IL1B, IL6 and TNF the chemokine IL8 and the factors oxytocin receptor (OXR), prostaglandin E synthase (PGES) and prostaglandin F synthase (PGFS) gene expression. The second set of sam ples collected at hour 6 post intrauterine infusion s was immediately placed in a solutio n of 4% of paraformaldehyde (Sigma Aldrich, Saint using an automatic rotary microtome. The sliced sample s were stained with hematoxylin

PAGE 207

207 and eosin (HE) stain to evaluate the development of inflamm ation on uterine tissue after intrauterine infusions. A modified scheme of grading wa s used (Snider et al., 2011). Briefly, grade 1 was characterized by normal endometrium or minimum mild focal inflammation or fibrosis. Grade 2 was characterized mild (2 A) to moderate (2 B) inflammation and or multifocal fibrosis with 1 to 4 layers of fibroblasts surrounding the fibrotic nest. Grade 3 was characterized as severe inflammation and/or diffuse fibrosis with 5 or more fibrotic nests per 5 mm linear field with or without occurrence of severe periglandular fibrosis and ectasia with disruption of gland epithelium. RNA Extraction and Quantitative, Real Time, Reverse Transcription PCR The RNA from uterine tissues was extracted using TRIzol re agent (Invitrogen Corp., Carlsbad, CA) according to instructions provided by the manufacturer. Samples were purified (PureLink RNA Mini Kit, Invitrogen, Carlsbad, CA) and RNA concentrations were determined using NanoDrop Spectrophotometer 200 (Thermo Scien tific, Rockford, IL). Abundance of mRNA for IL1B, IL6, IL8, TNF PGES, PGFS and OXR were determined by qRT PCR from uterine biopsy tissues collected 6, 12 and 24 hours after inocula tion with T. pyogenes RNA samples (50 ng/reaction, A 260/280 treated with RNase free DNase I (Applied Biosystems, Foster City, CA) for 15 min at 37 o C, heat denatured (75 o C for 10 min), and then reverse transcribed using High Capacity cDNA Reverse Transcription Kit and random hexamers (Applied Biosystems, Foster C ity, CA). Primers specific for the selected transcripts were chosen and designed using basic local alignment search tool (BLAST, http://blast.ncbi.nlm.nih.gov/Blast.cgi) and made on demand by Invitrogen (Invitrogen, Carlsbad, CA), as shown in Table 7 1.

PAGE 208

208 Q uantitative, reverse transcription PCR was completed using SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA) and the ABI 7300 Real Time PCR System (Applied Biosystems, Foster City, CA). Amplification of each gene was obtained by running a mix ture of 2.5 L of the cDNA product of each primer and 22.5 L SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA). The following cycling conditions were applied: initial activation/denaturation (60 o C for 2 min; 95 o C for 10 min); 40 cycles of 2 steps amplification protocol (95 o C for 15 s and 60 o C for 1 min) and for dissociation (55 to 95 o C). Each PCR was performed in triplicate, and the specificity of amplification was verified by melting curve analysis. A reaction lacking reverse transcript ase was included to verify the absence of genomic DNA contamination in reactions. Mitochondrial ribosomal protein S15 ( MRPS15 ) was chosen as house keeping gene because it was reported as the most stably expressed gene in uterine tissue (Wathes et al., 2009) Experiement 2 For remaining 17 cows enrolled in the study blood samples were collected and ovaries were scanned daily to evaluate plasmatic concentration of progesterone and luteal lifespan, respectively. Three ( one for each treatment ) of the 17 cows e nrolled in the second experiment did not complete the study and were excluded from final analysis because two cows acquired severe mastitis and one cow suffered a severe leg injury. Therefore, the data presented in this section was from 14 cows with 3, 4 a nd 7 cows for control, TNF and TP treatments respectively.

PAGE 209

209 Blood Sampling, Analysis of Progesterone and PGFM Concent rations and Ultrasound Scanning Blood samples were collected from the coccygeal artery or vein using plasma K 2 EDTA vacutainer tubes dail y starting at 353 days postpartum until 22 d after ovulation or two days after luteolysis was detected for evaluation of progesterone concentration on plasma. Samples were placed immediately on ice and kept refrigerated until arrival to the laboratory. Bl ood tubes were centrifuged at 2,000 x g for 15 min, and an aliquot of 2 mL of plasma was frozen at 20 C until analysis. Concentration of progesterone in plasma was analyzed in all samples by RIA using a commercial kit (Coat a Count, Siemens Healthcare Di agnostics, Los Angeles, CA). The sensitivity of the assay was 0.05 ng/mL calculated at 2 SD below the mean counts per min at maximum binding. All samples were analyzed in a single assay. Two plasma samples with progesterone concentrations of 1.5 and 2.5 ng /mL were included throughout the sequence of samples in the assay for quality control. The intra assay CV s were 3.9 and 6.7% for the samples containing 1.5 and 2.5 ng/mL, respectively. Additional blood samples were collected from 48 to 120 hours each 12 h ours intervals after intrauterine infusion with treatments to measure plasmatic concentration of PGFM following the procedures described above for progesterone. This time frame for collection of samples for PGFM was chosen to match the window on which rece nt studies identified a peak of PGFM as response to intrauterine infusion with T. pyogenes (Kaneko and Kawakami, 2008; Kaneko and Kawakami; 2008; Kaneko et al., 2013). The concentrations of PGFM were measured by an enzyme immunoassay as described previousl y (Ginther et al., 2010). The intra and inter assay CV was 4.59% and 5.92%, respectively.

PAGE 210

210 Ovaries were scanned daily using a portable ultrasound equipped with a 5.0 MHz transrectal probe (Aloka SSD 500, Aloka Co. Ltd., Wallingford, CT) starting at 353 da ys postpartum until 22 d after ovulation or two days after luteolysis to monitor ovarian structures dynamics and to determine CL lifespan. Statistical Analys i s All responses were analyzed using GLIMMIX procedure of SAS version 9.3 (SAS/STAT, SAS Inst it ute Inc., Cary, NC). The models for all analyzes included the effects of treatment (control vs. TNF vs. TP), time and the interaction treatment by time. Continuous data were analyzed with models fitting a Gaussian or a Logarithmic distribution, and residuals w ere tested for normality. For repeated measures, models included the effects of time, interaction between treatment and time, and the random effect of cow nested within treatment. For endometrial mRNA expression a comparative method developed by Livak and Schmittgen, (2001) was applied to generate the data for presentation and samples from control cows at hour 6 post intrauterine infusions were used as reference for comparison. Relative expression values were obtained by raising the PCR amplification values to the power of delta statistical analysis were generated by normali zation of CT values from target genes with the mean of CT value from reference gene according to Vandesompele et al., (2002). Confidence interval for graphical representation of relative expression were generated from the lower and upper confidence limits differences as described by Yuan et al., (2006).

PAGE 211

211 Statistical significance was declared when P value < 0.05 and tendency for P value < 0.10. Interactions were considered if P value < 0.15. R esults Experiment 1 An interact ion between treatment and time ( P = 0.05) for IL1B (Figure 7 2). Cows treated with TP had an increased ( P = 0.03) endometrial mRNA expression of IL1B in comparison with cows treated with TNF at 24 h post intrauterine infusions. The endometrial mRNA express ion of IL6 (Figure 7 3) was not affected by treatment ( P = 0.57), time ( P = 0.19) or the interaction between treatment by time ( P = 0.34). Nonetheless, mRNA expression of IL 6 at hour 12 post intrauterine tended to be greater for TP cows ( P = 0.06) than for control cows. The mRNA expression for the chemokine IL8 was not different among treatments ( P = 0.32), time ( P = 0.16) or interaction treatment by time ( P = 0.37) as shown in Figure 5 4. For TNF once again no differences in mRNA expression were identified among treatments ( P = 0.80), time (P = 0.59) or the interaction treatment by time ( P = 0.73) as depicted in Figure 7 5. Likewise, the mRNA expression for PGES was not altered by treatments ( P = 0.11) or time ( P = 0.68), as shown in Figure 7 6. Additionall y, there was not an interaction between treatment and time ( P = 0.46) for mRNA expression of PGES. Moreover, endometrial mRNA expression for PGFS, the other luteolytic cascade transcript evaluated, did not differ among treatments ( P = 0.27) and time ( P = 0 .47) as shown in Figure 7 7. Additionally, there was not an interaction between treatment and time (P = 0.99) for endometrial mRNA expression of PGFS. The endometrial mRNA expression OXR was not different among treatments ( P = 0.12) or time (P = 0.22) as s hown in Figure 7 8. Additionally, there was not an interaction between treatment and time (P = 0.98) for mRNA expression of OXR. However, mRNA expression

PAGE 212

212 of OXR tended to be increased for TP ( P = 0.09) and TNF ( P = 0.05) in comparison with control cows at hour 12 post intrauterine infusions. A histological evaluation of HE slides of uterine tissues was conducted revealed inflammation (grade 3) with transmigration of neutrophils into the uterine glands, separatio n of the epithelium from the stratum compactum and ectasia (dilation) of uterine glands with disruption o f gland epithelium as shown in F igure 7 9B, 7 9C and 7 9D. On the other hand, only 1 of the 5 cows in control tr eatment had inflammation of the endomet rium (grade 3 ) while the others were grade 1 F igure 7 9A. Experiment 2 The criterion to determine occurrence of luteal regression was reduction on plasmatic concentration of progesterone below 1.0 ng/mL. The criterion to determine early luteolysis was red uction of progesterone concentration below 1.0 ng/mL before day 14 of the estrous cycle. The mean day of the estrous cycle for luteal regression were 17.3, 16.2 and 14.0 days for control TNF and TP, respectively, and there were not different among treatme nts ( P = 0.38). Although, no differences were observed among treatments for mean day of the estrous cycle for luteal regression the percentage of cows with early luteolysis was increased ( P =0.001) for TP cows in comparison with control and TNF and control being 42.9 %, 25.0 % and 0 % for control, TNF and TP treatments, respectively. A total of 3 cows infused with T. pyogenes had an earlier than usual luteolysis at days 10, 11 and 12 of the estrous cycle. Additionally, one cow from the TNF group also had an earlier than usual luteolysis at day 12 of estrous cycle. Moreover, one cow infused with T. pyogenes did not present luteolysis until day 22 of the estrous cycle. The mean concentration of progesterone were 3.84, 3.19 and 2.39

PAGE 213

213 ng/mL for control, TNF and T P treatment, respectively, with no differences among treatm ents detected (P = 0.33). Additionally, no interaction between treatment and time were identified (P = 0.60) as shown in Figure 7 10. However, at day 12 of the estrous cycle the concentration of pr ogesterone tended to decreased for TNF and TP treatments compared to the control treatment (Figure 7 10). The peak of plasmatic concentration s of progesterone were 9.59, 6.54 and 6.24 ng/mL for control, TNF and TP treatment, respectively, and no difference s among treatments were observed ( P = 0.31). Likewise, the day of the estrous cycle to achieve the peak of progesterone concentration was no different among treatments ( P = 0.85) being 12.7, 12.2 and 11.3 days for control, TNF and TP treatments, respective ly. Similarly, the size of corpus luteum also did not differ among treatments ( P = 0.22). H owever an interaction between treatment and size of CL was present ( P < 0.01), as depicted in F igure 7 11. The size of CL was increasing for control at days 16 and 1 7 of the estrous cycle, whereas TNF and TP had the CL reducing in size at the same days (Figure 7 11). In fact, CL size on days 16 and 17 of the estrous cycle was greater ( P < 0.05) for control cows than for TNF and TP counterparts. At days 18 and 19 of th e estrous cycle the size of the CL was smaller ( P < 0.05) for TP cows than for TNF and control cows (Figure 7 11). The mean concentration of PGFM was not different among treatments ( P = 0.66) and no differences of time ( P = 0.58) or an interaction treatme nt by time were found ( P = 0.19) as shown in Figure 7 12. Discussion The current study was designed to test the hypothesis that uterine infusion with T. pyogenes could lead to acute inflammation eliciting the luteolytic cascade and early demise of the cor pus luteum. T. pyogenes is one the most important bacterium

PAGE 214

214 associated with uterine diseases (Sheldon et al. 2006) and previous studies with intrauterine infusion of T. pyogenes on d 3 after ovulation induced a peak of PGFM 3 d later, followed by the regr ession of the newly formed CL in approximately 50% of the cows (Kaneko and Kawakami, 2008, Kaneko and Kawakami, 2009, Kaneko et al., 2013). Although no difference in mean time of luteal regression occurred in this study, our data supports, although not fol lowing the same pattern, the findings of previous studies with three of seven cows infused with T. pyogenes having an early demise of the CL. Thus, it is reasonable to suggest that T. pyogenes disrupts luteal function and reduces lifespan of corpus luteum in at least part of the cows. It is unknown the reason why some cows respond with an early demise of CL, whereas some have normal estrous cycle length s and s ome have extended luteal cycles. C onsidering that only clinically healthy cows were included in th e study and the exposure to the pathogen load was the same it is reasonable to speculate that individual immune competence and the individual subsequent host pathogen interaction might be playing a role in the regulation of how cows respond to the infectio n of T. pyogenes In vitro culture of endometrial cells in a bacteria free filtrate from T. pyogenes was able to induce synthesis of PGF (Miller et al., 2007); however, it is not clear which molecular mechanism the T. pyogenes elicits to induce release o f PGF Some evidence in the literature suggest that uterine inflammation induced by PAMPS molecules such as the pep tidoglycan present in cell wall of T. pyogenes can induce release of pro inflammatory cytokines such as TNF 20 03; Timmerman, et al., 1993). These cytokines have been shown to be capable of directly stimulating

PAGE 215

215 endometrial synthesis of PGF (Davidson et al., 1995; Hansen et al., 2004; Skarzynski et al., 2000). In the current study, cows received uterine infusion o f T. pyogenes and uterine samples for biopsy were collected at critical times to evaluate this potential interplay between major cytokines shown to be altered by T. pyogenes and critical factors of the luteolytic cascade that could potentially be altered by these mediators of the inflammatory response. Additionally, a group of cows received intrauterine infusion with a luteolytic dosage of TNF of luteolysis (Skarzynski et al., 2009). The histolo gical evaluation at 6 h after intrauterine infusion clearly showed that an acute and severe inflam mation was present in the uterine tissue of cows infused with T. pyogenes and TNF cows had normal to mil d inflammation as a result of intrauterine infusion of sterile saline solution. However, the evaluation of gene expression of pro infl ammatory cytokines, chemokines and luteolytic factors did not consistently support our hypothesis that inflammation was directly provoking stimulation of luteolytic cascade factors. Some minor differences in gene expression occurred in path expected with i ncreased IL1B expression in cows infused with T. pyogenes when compared to TNF counterparts at hour 24 post intrauterine infusion; and a tendency for increased expression of IL6 in TP treatment in comparison with control counterparts at hour 12 post intrau terine infusions. Additionally, expression of OXR tended to be increased for TP and TNF treatments in comparison with cont rol treatment at hour 12 post intrauterine infusions. The other inflammatory mediators and luteolytic factors were not affect ed by ute rine infusion with T. pyogenes

PAGE 216

216 Given the degree of inflammation observed in cows that were infused with T. pyogenes and TNF chemokine IL8, which is a bonafide neutrophil chemoattractant. Although no statistical differences were identified for IL8 the expression for cows infused with T. pyogenes was 4 to 8 times greater than in control cows suggesting that the lack of difference may be an artifact of the limitations of this study such as high variability in the individual response to T. pyogenes Additionally, there is the possibility that the sample co llected had other cells such as fibroblast, endothelial and muscle that may hamper the chance of identi fying difference s in expression of endometrial and immune cells such as previously demonstrated on in vitro studies (Jost and Bilington, 2005; Miller et al., 2007; Bromfield and Sheldon, 2011). Hitherto remain elusive how T. pyogenes major virulence factor pyolysin elicit immune response, but evidence from some other similar pore and hemolysin) suggest that pyolysin might activat e the cytosolic nod like receptor NLRP3 inflammasome that activates caspase 1 that leads to cleavage of pro and IL18 and subsequent release of these cytokines (Muoz Planillo et al., 2009; Embry et al. 2011). Therefore, it is reasonable t o speculate that T. pyogenes potentially can induce similar mechanism, which could explain the molecular basis of induced m igration of neutrophils after intrauterine infusion with T. pyogenes documented in Figure 7 9. Uterine gene expression of PGES PGFS and OXR some of the major factors associated with the occurrence of luteolysis, w ere not remarkably altered by intrauterine infusio n with T. pyogenes n o r by intrauterine infusion of a luteolytic dosage of TNF Although some minor effects on OXR expression were induced by TP and TNF, we

PAGE 217

217 failed to observe the interplay between inflammation and luteolysis that we anticipated. Considering that some numerical trends also followed the path expected we cannot exclude the possibility that the limitations of this study might have jeopardized the likelihood of identifying the expected response. Ano the r unexpected outcome was that intrauterine infusion of luteolytic dosage of TNF influence uterine expression of PGES and PGFS major luteolytic cascades factors known to be altered by this dose of TN F that the uterus maybe resilient and does not allow enough amounts of TNF s, blood stream and corpus luteum cells to mimic the effect of an intravenous a luteolytic dosages of TNF Collectively the results of the current study indicate that the inflammation induced by T. pyogenes does not consistently alter mRNA expression o f major molecules of the inflammatory responses and key factors of the luteolytic cascade. It suggests that an alternative factors not investigated under the sco p e of this study might be involved in the induction or early demise of CL. Additionally, there is the possibility that the limita tions of the study hindered the ability to clearly show the anticipated response. On the other hand, the results of the current study s upport previous findings that intrauterine infusion of T. pyogenes may disrupt luteal f unction leading to early demise of the newly formed corpus luteum. Future studies should consider the investigation of the interplay of virulence factors of T. pyogenes with inflammation and luteolysis; and perhaps elaborate a model on which cows that resp onded to T. pyogenes with early demise of corpus luteum are compared to cows not responding to

PAGE 218

218 T. pyogenes and normal cows to determine which factor might be regulating this different response C onclusion I ntrauterine infusion with T. pyogenes disrupts lute al function and lead to early demise of CL at least in part of the cows. Although, T. pyogenes clearly induced inflammation we were unable to identify a consistent response in endometrial mRNA expression of mediators of the inflammation and luteolytic fact ors that could support a direct anticipated interplay between the bacterium induced inflammation and factors triggering the early demise of the corpus luteum

PAGE 219

219 Table 7 1. Primer reference and sequences for genes investigated by quantitative real time PCR T arget g ene NCBI Sequence Primer Primer s equence MRPS15 NM_001192201.1 Forward AGATGACCCGCCCCCTTCCA Reverse GGGAGCTGGTGTCCTTCGGGT TNF NM_173966.2 Forward CCAGAGGGAAGAGTCCCCAG Reverse TCGGCTACAACGTGGGCTAC IL1 B NM_174093.1 Forward ATTCTCTCCAGCCAACCTTCATT Reverse TTCTCGTCACTGTAGTAAGCCATCA IL6 NM_173923.2 Forward TGAGTGTGAAAGCAGCAAGGA Reverse TCGCCTGATTGAACCCAGAT IL8 NM_173925.2 Forward TGTGAAGCTGCAGTTCTGTCAA Reverse TTTCACAGTGTGGCCGACTCT PGES NM_174443.2 Forward ATCGTGACGGTCCGTCTCTAA Reverse GCCCTTTGAGATTGTGACAGG PGFS NM_001035367 Forward TGTGGTGCACGTATCACGACA Reverse AATCACGTTGCCGTCCTCATC OXR NM_174134.2 Forward GCACCTGAGCATAGCCGACC Reverse GTGGCAAGGACGATGACGGG MRPS15 = mitochondrial ribosomal protein S15; TNF = tumor necrosis factor ; IL1B = interleukin 1 ; IL6 = interleukin 6; IL8 = interleukin 8; PGES = prostaglandin E synthase; PGFS = prostaglandin F synthase; OXR = oxytocin receptor.

PAGE 220

220 Figure 7 1. Diagram of experimental activities for experiments 1 and 2. 1 E ligible cows ovulated to GnRH given on day 32 3 postpartum and had no subclinical endometritis and bacterial growth. 2 T reatments included intrauterine infusion (IUI) of 10 mL of sterile saline solution containing 109 CFU /mL of T. pyogenes (TP); or IUI of 10 mL of (control). 3 Blood samples were collected daily from IUI until day 22 post ovulation or confirma tion of luteolysis. US = ultrasonographic examination.

PAGE 221

221 Figure 7 2. Relative endometrial mRNA expression of treatments and time Nonpregnant cows from the control group were used as reference for comparison. TP = intrauterine infusion of 10 mL of saline solution containing 10 9 cfu/mL of T. pyogenes ; TNF = intrauterine infusion of 10 mL of saline ontrol = intrauterine infusion of 10 mL of saline solution Within same hour differe nt letters means treatment difference ( a, b )

PAGE 222

222 Figure 7 3. Relative endometrial mRNA expression of IL6 according to treatments and time. Nonpregnant cows from the control group were used as reference for comparison. TP = intrauterine infu sion of 10 mL of saline solution containing 10 9 CFU /mL of T. pyogenes ; TNF = intrauterine infusion of 10 mL of saline intrauterine infusion of 10 mL of saline solution. Within same hour aft er treatment different capital letters means tendency to be different ( A, B ; P < 0.10).

PAGE 223

223 Figure 7 4. Relative endometrial mRNA expression of IL8 according to treatments and time Nonpregnant cows from the control group were used as reference for compar ison. TP = intrauterine infusion of 10 mL of saline solution containing 10 9 cfu/mL of T. pyogenes ; TNF = intrauterine infusion of 10 mL of saline ontrol = intrauterine infusion of 10 mL of saline solution

PAGE 224

224 Figure 7 5. Relative endometrial mRNA expression of treatments and time Nonpregnant co ws from the control group were used as reference for comparison. TP = intrauterine infusion of 10 mL of saline solution containing 10 9 cfu/mL of T. pyogenes ; TNF = intrauterine infusion of 10 mL of ; and c ontrol = intrauterine infusion of 10 mL of saline solution

PAGE 225

225 Figure 7 6. Relative endometrial mRNA expression of PGES according to treatments and time Nonpregnant cows from the control group were used as reference for comparison. TP = intraute rine infusion of 10 mL of saline solution containing 10 9 cfu/mL of T. pyogenes ; TNF = intrauterine infusion of 10 mL of ontrol = intrauterine infusion of 10 mL of saline solution

PAGE 226

226 Figure 7 7. Relative endometrial mRNA expression of PGFS according to treatments and time Nonpregnant cows from the control group were used as reference for comparison. TP = intrauterine infusion of 10 mL of saline solution containing 10 9 cfu/mL of T. pyogenes ; TNF = intrauterine infusion of 10 mL of ontrol = intrauterine infusion of 10 mL of saline solution

PAGE 227

227 Figure 7 8. Relative endometrial mRNA expression of OXR according to treatments and time Nonpregnant cow s from the control group were used as reference for comparison. TP = intrauterine infusion of 10 mL of saline solution containing 10 9 cfu/mL of T. pyogenes ; TNF = intrauterine infusion of 10 mL of saline and c ontrol = intrauterine infusion of 10 mL of saline solution Within same hour after treatment different capital letters means tendency to be different ( A, B ; P < 0.10 )

PAGE 228

228 Figure 7 9. Uterine tissu es pictures stained with hematoxylin and eosin. A endometrium of control cow with arrow pointing intact epithelium and absence of inflammation; B endometrium of T. pyogenes cow with arrow pointing damaged epithelium and neutrophils in the periglan dular area; C endometrium of T. pyogenes in a smaller magnification with arrow pointing massive accumulation of neutrophils; D endometrium of T. pyogenes with arrow pointing massive accumulation of neutrophils.

PAGE 229

229 F igure 7 10 Concentrations of prog esterone in plasma according to day of the estrous cycle TP = intrauterine infusion of 10 mL of saline solution containing 10 9 CFU /mL of T. pyogenes ; TNF = intrauterine infusion of 10 mL of saline intrauterine infusion of 10 mL of saline soluti on. Within day, concentrations of progesterone tend to be differed ( A, B C ; P < 0. 10 ).

PAGE 230

230 F igure 7 11 Size of the corpus luteum according to day of the estrous cycle TP = intrauterine infusion of 10 mL of saline solution containing 10 9 CFU /mL of T. p yogenes ; TNF = intrauterine infusion of 10 mL of saline solution containing 1 intrauterine infusion of 10 mL of saline solution. *Means an interaction between treatment and day (P < 0.001). Within day, size of the corpus luteum differed ( a b; P < 0. 10 ).

PAGE 231

231 Figure 7 1 2 Concentrations of prostaglandin F metabolite (PGFM) in plasma according to day of the estrous cycle. TP = intrauterine infusion of 10 mL of saline solution containing 10 9 CFU /mL of T. pyogenes ; TNF = intrauterine infusion o f control = intrauterine infusion of 10 mL of saline solution.

PAGE 232

232 CHAPTER 8 EFFICACY OF AMPICILL IN TRIHYDRATE FOR TR EATMENT OF METRITIS AND SUBSEQUENT FERTILITY IN DAIRY COWS Objectiv es were to evaluate the efficacy of ampicillin trihydrate for treatment of metritis in dairy cows compared with ceftiofur hydrochloride and the subsequent effects on pregnancy per artificial insemination (P/AI) for the 1st postpartum AI. R ectal temperature was measured daily for the first 12 d in milk (DIM), and fever was characterized by rectal t 39.5 o C. Vaginal discharge was scored at 4, 6 and 8 DIM, and on any day a cow had fever. Cows with vaginal discharge score 5 (reddish/brownish foul smell) were diagnosed with metritis and cows with metritis and rectal temperature 39.5 o C wer e diagnosed as puerperal metritis. Cows with metritis (n = 528) were blocked by parity and type of metritis (metritis only or puerperal metritis) and, within each block, assigned randomly to receive 11 mg/kg of ampicillin (n = 259) or 2.2 mg/kg of ceftiofu r (n = 269) once daily for 5 d. Day of diagnosis of metritis was considered study d 1. A cohort of 268 cows without metritis was selected randomly as controls at 12 DIM based on the s ame parity and day of calving. In cows with metritis, rectal temperature was measured on study d 1 to 7, and 12, and vaginal discharge was scored on study d 5, 7, and 12. Metritis cure was characterized by vaginal discharge < 5. At 32 DIM, vaginal discharge was scored for diagnosis of purulent vaginal discharge (PVD, vaginal di scharge score > 2, mucopurulent discharge). At day 39 DIM a uterine cytology samples w ere collected to evaluate subclinical endometritis. At 53 and 67 DIM ovaries were scanned to determine cyclic status. Pregnancy status was diagnosed on d 34 and 62 after 1st AI. Data were analyzed using the PROC GLIMMIX of SAS. Cure rates of metritis for ampicillin were greater than for ceftiofur on d 5 (37.1% vs. 25.2%) and 7 (57.2% vs. 46.3%) after metritis diagnosis, but not different on d 12 (82.0% vs.

PAGE 233

233 85.0%). Cows wit h puerperal metritis had lower cure arte than cows with metritis only on d 5 (39.7 vs. 23.2%), 7 (62.9 vs. 40.5%), and 12 (88.1 vs. 77.7%). Incidence of fever after treatments, did not differ between treatments (ampicillin = 20.8% vs. ceftiofur = 19.3%), b ut mean rectal temperature tended to be less for ceftiofur than ampicillin cows (ampicillin = 39.15 C vs. ceftiofur = 39.10 C). Cows receiving ampicillin had lower prevalence of PVD than those treated with ceftiofur (57.7 vs. 67.8%), but they were both g reater than no metritis cows (21.9%). Subclinical endometritis incidence was the same for ampicillin and ceftiofur (30.0 vs. 25.4%), but they were both greater than no metritis cows (14.5%).The proportion of cyclic cows at 67 DIM did not differ among treat ments (ampicillin = 74.8% vs. ceftiofur = 75.0% vs. no metritis = 75.1%). Fist service P/AI did not differ among treatments at 34 (ampicillin = 28.9% vs. ceftiofur = 29.1% vs. no metritis = 32.0%) and 62 d after AI (ampicillin = 28.0% vs. ceftiofur = 28.3% vs. no metritis = 30.5%). Ampicillin was an efficacious therapy for metritis. Rate of cure was faster for ampicillin than for ceftiofur, but on d 12 both treatments resulted in similar cure rates. Although ampicillin reduced the prevalence of PVD, and no metritis cows had less PVD and subclinical endometritis than those with metritis, P/AI for the 1st insemination did not differ among treatments. Introductory Remarks Metritis is a prevalent postpartum disease in lactating dairy cows characterized by abnorm ally enlarged uterus and a fetid, watery red brown fluid to viscous off white purulent uterine disch arge that can be accompanied or not of fever within 21 days postpartum, but more frequently diagnosed in the first week postpartum (Sheldon et al., 2006). T he incidence rates of dairy cows developing metritis ranges from 10 to 36% (Goshen and Shp igel, 2006; Santos et al., 2010 ; Chapinal et al., 2011).

PAGE 234

234 The economic losses caused by metritis are strikin g ranging from to $328 to $380 per affected cow, and the lo sses are caused by reduced milk production, delayed pregnancy cost with treatment and increased culling and death ( Drillich et al., 2001) Additionally, c ows diagnosed with metritis have an increased ri sk to develop both clinical and subclinical endometr itis (Galvo et al., 2009; Martinez et al., 2012 ). The main bacteria isolated from cases of uterine infection include Escherichia coli Trueperella ( formerly Arcanobacterium ) pyogenes and anaerobic bacteria such as Prevotella (formerly Bacteroides ) specie s and Fusobacterium necrophorum ( Griffin et al., 1974; Noakes et al., 1989; Sheldon et al., 2002). Recently, the e xpression s of some specific virulence factors by these bacteria were associated with increase risky for development of uterine disease s (Bical ho et al., 2012 ). Escherichia coli express ing the adhesin type I fimbriae fimH identified in the uterus of cows in the first 3 days postpartum was significantly associated with de velopment of metritis and endometritis Fusobacterium necrophorum expressing the luekotoxin/hemolysin lktA in the first 3 days or between days 8 and 12 postpartum w as associated with endometritis. T rueperella pyogenes expressing the type I fimbriae adhesin fimA and the pyolysin plo between 8 and 10 d or between 34 and 36 d postpar tum was associated with en dometritis (Bicalho et al., 2012 ). Therefore, it has been suggested that the presence of E. coli expressing virulence factor fimH in the uterus of cows i n the first few days postpartum paves the way for the other bacterial infecti on coordinating the initial process of tissue damage and development of uterine diseases. Thus, it is reasonable to suggest that a reduction on the extent of E. coli load in the uterus of metritic cows might mitigate the negative

PAGE 235

235 impact of the disease and minimize the risk of subsequent chronic uterine infections s uch as clinical and subclinical endometritis. A mpici llin is a beta lactam antibiotic that acts as an irreversible inhibitor of dd transpeptidase, an essential enzyme that bacte ria use to make thei r cell walls. Therefore, ampicillin generally inhibits the third and final stage of bacterial cell wall synthesis in binary fission, which ultimately leads to cell lysis. Ampicillin has received FDA approval for use in dairy cattle and it is indicated for therapy of infections caused by E. coli (Burrows, 1993; Lehtolai nen et al., 2003) but to date, no published study has evaluated efficacy of ampicillin treatment of metritis in dairy cows. We hypothesized that ampicillin would be an effective therapy for m etritis resulting in similar clinical cure and subsequent reproductive performance compared with cows treated with cef tiofur, a common antibiotic labeled and prescribed for treatment of metritis in the United States. The objective s of study were to evaluat e the efficacy of ampicillin trihydrate for treatment of metritis in dairy cows compar ed with ceftiofur hydrochloride and subsequent effect on pregnancy per AI (P/AI) to the first service. M aterials and M ethods The University of Florida Institute of Food a nd Agricultural Sciences Animal Research Committee approved all procedures in this study. Cows, Housing, and Diets The study was conducted on a single dairy farm located in central Florida. The lactating herd was composed by approximately 4,500 cows with a yearly rolling herd average milk yield of approximately 11,000 kg. A total of 528 cows diagnosed with metritis were enrolled in the study from October of 2012 to January of 2013.

PAGE 236

236 Additionally, a cohort of 268 herdmates without metritis was enrolled in th e study at 12 DIM to be used as controls. Metritic primiparous (n = 264) and multiparous cows (n = 264), and non metritic primiparous cows (n = 134) and multiparous cows (n = 134) cows were housed together during the first 2 wk postpartum, and separately t hereafter. Cows were housed in free stall barns with sand bedded stalls and equipped with sprinklers and fans for forced evaporative cooling. Cows received the same TMR to meet or exceed the nutrient requirements for a lactating Holstein cow producing 45 k g/d of milk with 3.5% fat and 3.2% true protein when DM intake is 25 kg/d (NRC, 2001). Diets earlage, ground corn, citrus pulp, solvent extracted soybean meal, expell er soybean meal, corn gluten feed, m olasses, minerals, and vitamins Experimental Design, Trea tments and Body Condition Scor ing Rectal temperature was measured using an electronic thermometer (GLA Agricultural Products, San Luis Obispo, CA), immediately after the morning milking. retrieved using the Metricheck (Metricheck, Simcro, New Zealand) was scored at 4, 6 and 8 DIM, and on any day a cow had fever. A vaginal discharg e scoring system based on Sheldon et al. (2006) was used in the current study and as follows: 1 = clear or translucent mucus; 2 = mucus containing flecks of white or off white pus; 3 = discharge white mucopurulent material; 4 = discharge containing > 50% purulent material; and 5 = watery, reddish/brownish color of foul smell. Cows with vaginal discharge score 5 were diagnosed as having metritis. Cows with a vaginal discharge score 5 and rectal temperature 39.5 C were diagnos ed as puerperal metritis. Cows with metritis (N = 528) were blocked by parity (primiparous vs.

PAGE 237

237 multiparous) and type of metritis (only metritis = 312 or puerperal metritis = 216) and, within each block, assigned randomly to receive 11 mg ampicillin/kg of B W i.m. (n = 259) as ampicillin trihydrate (Polyflex, Boehringer Ingelheim Vetmedica, St. Joseph, MO) or 2.2 mg of ceftiofur/kg of BW i.m. (n = 269) as ceftiofur hydrochloride (Excenel RTU sterile suspension, Zoetis, Madison, NJ) once daily for 5 d. The day of diagnosis of metritis was considered study d 1. A cohort of 268 cows without metritis was selected randomly at 12 DIM based on the same day of calving and same parity to match herdmates diagnosed with metritis. The body condition of all cows was assess ed at enrollment in the study using a scoring system 1 (emaciated) to 5 (obese) according to Ferguson et al. (1994) as depicted in the Elanco BCS chart (Elanco, 2009). Cows that had dystocia based on any type of assistance during delivery, a stillbirth cal f, twin calves, or occurrence of retained fetal membranes were classified as havin g calving related disorders Rectal T emperatures, Vaginal Discharge Evaluation, and Cure Definitions Cows diagnosed with metritis had rectal temperature recorded from study d 1 to 7, and again on study d 12 after morning milking (Figure 8 1). Additionally, all cows that had vaginal discharge scored using the same 1 to 5 scoring system mentioned above on study d 5, 7 and 12. Cure of metritis was defined based on vaginal dischar ge and rectal temperature. Initially, cure was determined based solely on a vaginal discharge < 5, or absence of watery, reddish/brownish color of foul smell discharge on d 5, 7 and 12 of study. In addition, cure of was also determined based on vaginal dis charge < 5 and rectal temperature < 39.5 o C. Finally, cure on d 12 of the study was also determined based on vaginal discharge score < 5, rectal temperature < 39.5 o C, and no additional

PAGE 238

238 health problem or concurrent antimicrobial other than the specified re spective treatment before d 12 in the study. Evaluation of Purulent Vaginal Discharge, Subclinical Endometritis and Estrous Cyclicity Samples of vaginal discharge and uterine endometrial cytology were collected from all cows diagnosed with metritis and f rom the cohort of cows without metritis. Vaginal discharge was collected at 32 3 DIM using the metricheck device to determine occurrence of purulent vaginal discharge (PVD), formerly known and classified as clinical endometritis (Sheldon et al., 2006; Du buc et al., 2010). Cows with vaginal discharge score > 2 were classified as having PVD. Uterine cytology samples were collected on d 39 3 postpartum using the cytobrush technique ( Kasimanickam et al., 2005 ) using a stainless steel gun protected by a one way plastic tube protector (Continental plastics, Delaval, WI). After collecting the endometrial cytology, the cytobrush was rolled onto a slide and air dried immediately. The slides were transported to the laboratory and stained using diff quick stain kit (IMEB, San Marcos, CA). Three technicians blinded to the treatments read the slides. Two hundred cells were counted in each slide using a microscope at 400 x magnifications to determine the proportion of PMNL relative to the total leukocytes and endometri al cells counted. Cows with a Gilbert et al., 2005 ). Estrous cyclicity was evaluated at 50 3 and 64 3 DIM by ultrasonographic examination of the ovaries using a portable ultr asound scanner equipped with a 7.5 MHz transrectal probe (Easi Scan, BCF Technology, Rochester, MN). Cows with at least a CL > 15 mm recorded on one of the two examination days were considered to

PAGE 239

239 be estrous cyclic, whereas those without a visible CL > 15 m m in both examinations were considered anovular Reproductive Management Cows enrolled in this study were subjected to a reproductive program as described in Figure 7 2. Cows receiving the first postpartum AI were presynchronized with 2 i.m. injections of 25 mg of PGF (5 mL of Lutalyse, 5 mg/mL of dinoprost as tromethamine salt; Zoetis, Madison, NJ) administered 14 d apart, at 50 3 and 64 3 DIM. At the second injection of PGF those identified in est rus by removal of tail chalk were artificially inseminated on the same morning. Cows not observed in estrus within 12 d of the second PGF of the presynchronization protocol were enrolled in the 5 d timed AI program at 76 3 DIM (Figure 8 2). Briefly, th e 5 GnRH (2 mL of Cystorelin sterile solution, gonadorelin diacetate tetrahydrate equivalent PGF on d 5 an d 6 of the protocol. A second i.m. injection of GnRH was administered concurrently with AI at 72 h after the first PGF Pregnancy was diagnosed by transrectal ultrasonography on d 34 3 after AI. The presence of an amniotic vesicle containing an embryo with a heartbeat was used as the criteria to determine pregnancy. Pregnant cows on d 34 3 were reexamined for pregnancy by transrectal palpation 4 wk later, on d 62 of gestation. Pregnancy per AI was calculated by dividing the number of cows diagnosed pr egnant at d 34 or 62 after AI by the number of cows receiving AI. Pregnancy loss was calculated as the number of cows that lost a pregnancy between d 34 and 62 after AI divided by the number of cows diagnosed pregnant on d 34 after AI. Cows that were detec ted in estrus before study d 32 were reinseminated and

PAGE 240

240 considered nonpregnant. Inseminations were performed by six technicians with semen from 13 Holstein sires Statistical Analysis Sample size calculation was performed using the POWER procedure of SAS v ersion 9.3 (SAS Institute Inc., Cary, NC). The sample size was calculated to provide sufficient experimental units to cure of metritis by d 7 and d 14 differ by 10 percentage units (51 vs. 41%) and by 7 percentage units (84 vs. 77%), respectively. It was assumed that 41% and 77% of the cows treated with ceftiofur w ould experience clinical cure on d 7 and 14, respectively, after initiation of treatments (Chenault et al., 2004). Additionally, sample size calculation aimed to allow identification of an increase of 8 percentage units (39 vs. 30%) in P/AI after the first insemination when considering that 30% of the cows treated with ceftiofur would become pregnant following the first postpartum AI. Under these assumptions, between 210 and 240 experimental units per treatment were deemed necessary. Because of potential at trition, a minimum of 250 cows were planned for enrollment. Categorical data were analyzed by logistic regression using the GLIMMIX procedure of SAS version 9.3 (SAS/STAT, SAS Institute Inc., Cary, NC) fitting a binary distribution. Treatment was forced i n the final models, but covariates and the interaction between treatment and covariates were sequentially removed from the model if P > 0.10. The models for cure of included the fixed effects of treatment (ampicillin vs. ceftiofur), type of metritis (only metritis vs. puerperal metritis), parity (primiparous vs. multiparous), calving related disorders (yes or no), and interactions between treatment and parity, treatment and type of metritis type, and treatment and calving related disorders, and the random e ffect of block. Two models were built for analysis of purulent

PAGE 241

241 vaginal discharge, subclinical endometritis, and estrous cyclic status. The first model included only cows randomly assigned to treatments and excluded cows without metritis, and the statistica l models were identical to those described for cure of metritis. The second model evaluated all cows, including cows without metritis, and type of metritis was removed from the explanatory variables. The models for pregnancy per AI and pregnancy loss were similar to those used to analyze purulent vaginal discharge, but also included type of insemination (after detected estrus or timed AI), technician, and sire. The continuous data with repeated measures over time were analyzed using the GLIMMIX procedure of SAS (SAS/STAT version 9.3; SAS Institute Inc., Cary, NC) with models fitting a Gaussian distribution. Data were tested for normality of residuals, and non normally distributed data were transformed before analysis. Rectal temperature was analyzed with the fixed of treatments (ampicillin vs. ceftiofur), type of metritis (metritis only vs. puerperal metritis), parity, the interactions between treatment and type of metritis, treatment and parity, parity and type of metritis, and treatment and type of metritis and parity. Block and cow nested within treatment were the random terms in the criterion was selected for the model. 0.10 were considered tendencies Results The prevalence of metritis in the farm throughout the period of the study was 36.1 % (528/1,463). Of the 528 cows enrolled in the study, 40.9% were considered to

PAGE 242

242 have puerperal metritis on the day of diagnosis (216/528), resulting in an overall incidence of puerperal metritis of 14.8% (216/1,463). Rectal Temperatures and Incidence of Fever On the day of study enrollment, the rectal temperature was greater (P < 0.05) for ampicillin than cefti ofur. Nevertheless, the mean rectal temperature of cows after treatments were initiated did not differ between treatments (Table 8 1). An interaction (P < 0.001) between treatment and day in the study was observed for rectal temperature becaus e cows receiv ing ceftiofur had low er temperatures between d 2 and 3 of treatment, whereas cows receiving ampicillin had low er temperatures on d 6 and 7 of treatment (Figure 8 3). As anticipated, type of metritis influenced (P < 0.001) rectal temperature, and it was gre ater for cows with puerperal metritis than those with metritis only. Incidence of fever did not differ between treatments and averaged 20.2% throughout the 12 d post treatment evaluation (Table 8 1). Cows with puerperal metritis had greater (P < 0.001) inc idence of fever than those with metritis only. Incidence of fever from study d 2 to 12 decreased (P < 0.001) over time, from 23.8% on study d 2 to 17.0 on study d 12. No interaction between treatment and type of metritis, or parity, or day in the study wer e observed for incidence of fever. Metritis Cure Based on Vaginal Discharge Score < 5 Of the 528 cows randomly assigned to treatments, 526 were evaluated for cure on d 5, 525 were evaluated on d 7, and 517 on d 12. Before study d 12, 11 cows were not evalu ated for vaginal discharge score, 5 ampicillin and 6 ceftiofur. The reasons were because they were culled (1 ampicillin, 2 ceftiofur) or died (4 ampicillin, 4 ceftiofur). Clinical cures based on the criterion of vaginal discharge score < 5 were greater fo r ampicillin than ceftiofur on d 5 ( P < 0.01) and 7 ( P = 0.02) of the study; however, no

PAGE 243

243 difference ( P = 0.40) in cure rates on d 12 were detected between treatments (Figure 8 4A). Cows with puerperal metritis had reduced cure on d 5 ( P < 0.001), 7 ( P < 0. 001), and 12 ( P < 0.01) compared with cows with metritis only (Figure 8 5A). Nevertheless, no interaction between treatment and type of metritis was observed for cure based on vaginal discharge < 5 on d 5, 7 and 12 after initiation of treatments. Multiparo us cows had increased (P < 0.01) cure than primiparous on study d 5, but not on d 7 and 12 (Figure 8 6A). Cows with calving related problems had reduced (P = 0.02) cure on d 5 and tended (P = 0.07) to have reduced cure on d 7 compared with cows with norma l calving (Figure 7 7A). However, on d 12 of the study the cure of metritis was the same for cows with and without calving related disorders. There were no interactions between treatment and type of metritis, parity, or calving related disorders for cure b ased on vaginal discharge score < 5 on d 5, 7 and 12 of the study. Metritis Cure Based on Vaginal Discharge Score < 5 and Rectal Temperature < 39.5 o C When cure rates were analyzed according to the criteria of vaginal discharge score < 5 and concurrent rec tal temperature < 39.5 C, cows treated with ampicillin had increased (P < 0.01) cure of metritis on and 7 compared with cows treated with ceftiofur, but the proportion of cows with metritis cured on d 5 (P = 0.18) and d 12 (P = 0.76) did not differ betwe en treatments (Figure 8 4B). Cows with puerperal metritis had reduced (P < 0.01) clinical cure throughout the 12 d observational period compared with cows with metritis only (Figure 8 5B). Multiparous cows tended (P = 0.07) to have increased clinical cure on d 5 and 12 after the initiation of treatments than primiparous cows (Figure 8 6B). Cow diagnosed with calving related problems had reduced (P = 0.001)

PAGE 244

244 clinical cure on study d 5 than those with normal calving, but this difference was no longer present o n study d 7 and 12 (Figure 7 7B). Metritis Cure Based on Vaginal Discharge Score < 5, Rectal Temperature < 39.5 o C, and no Additional Antimicrobial Therapy When cows that received any additional antimicrobial therapy during the 12 d observational period we re considered a failure of the original treatment, then clinical cure declined for both treatments. On d 12, the proportion of cows considered clinically cured was similar (P = 0.63) between ampicillin and ceftiofur (58.2 vs. 60.4%). Cows with puerperal me tritis had reduced (P = 0.02) clinical cure than those with metritis only (54.0 vs. 64.3%). On the other hand, primiparous cows had greater (P = 0.03) proportion of cows cured than multiparous cows (63.9 vs. 54.5%). Cows with calving related problems had s imilar clinical cure on d 12 compared with cows with normal calving, and it averaged 59.2%. Purulent V aginal Discharge and Subclinical Endometritis A total of 760 cows were evaluated for PVD on d 32 postpartum, 248 cows in ampicillin, 252 ceftiofur, and 2 60 no metritis. Cows receiving ampicillin had reduced (P = 0.03) prevalence of PVD on d 32 postpartum compared with those treated with ceftiofur, but they were both greater (P < 0.01) than cows not diagnosed with metritis (Figure 8 8A). The benefits of amp icillin in reducing PVD were observed in cows with only metritis (ampicillin = 49.1 vs. ceftiofur= 68.4%), but not in those with puerperal metritis (ampicillin = 68.5 vs. ceftiofur = 66.9%). The prevalence of subclinical endometritis on d 39 postpartum did not differ (P = 0.38) for cows treated between cows treated with ampicillin and ceftiofur, but they were both greater (P < 0.01) than no metritis cows (Figure 8 8B) and not interactions with other variables were detected

PAGE 245

245 Estrous Cyclicity, Pregnancy per AI and Pregnancy Loss Of the initial 796 cows enrolled in the study (259 Ampicillin, 269 Ceftiofur, and 268 no metritis), 29 cows in Ampicillin (14 sold, 7 dead, 4 pyometras, and 4 with adhesions of the reproductive tract), 42 Ceftiofur (21 sold, 13 dead, 3 pyometras, and 5 adhesions of the reproductive tract), and 21 no metritis (16 sold, 4 dead, 1 pyometra) did not receive the first AI or contributed with analysis of P/AI and pregnancy loss. Therefore, 749 cows contributed with data for resumption of est rous cyclicity and 704 cows contributed with data for P/AI. The percentage of estrous cyclic by 64 3 DIM did not differ among treatments and averaged 75% (Table 8 2). A greater (P < 0.01) proportion of multiparous cows were cyclic at 64 DIM than primipar ous cows (79.1 vs. 70.9%). For cows randomly assigned to treatments, type of metritis did not influence (P = 0.91) resumption of estrous cyclicity and 75.5 and 75.0% of cows with metritis only and cows with puerperal metritis resumed ovulation by 64 DIM. Pregnancy per AI to first service on d 34 and 62 after first insemination did not differ among treatments (Table 8 2). For cows randomly assigned to treatments, type of metritis did not influence (P < 0.95) P/AI on d 34 or 62 after insemination. On d 64, P /AI were 26.7 and 26.4% for cows with metritis only and cows with puerperal metritis, respectively. Similar to P/AI, pregnancy loss between 34 and 62 d of gestation did not differ among treatments. D iscussion As anticipated, ampicillin was an efficacious alternative treatment for metritis and showed clinical efficacy similar to or better than ceftiofur. Metritic cows treated with

PAGE 246

246 ampicillin trihydrate had faster cure rate than those treated with ceftiofur hydrochloride. Although the study design does not allow for detection of spontaneous cure, it is known that a large proportion of cows diagnosed with puerperal metritis not receiving antimicrobial therapy have remission of the symptoms within 14 d of diagnosis (Chenault et al., 2004; McLaughlin et al., 20 12). Treatment with ampicillin lead to greater proportion of cows clinically cured on d 7 after initiation of treatments when the criteria for cure was either vaginal discharge < 5 or a combination of vaginal discharge < 5 concurrent with rectal temperatur e < 39.5 o C. This benefit was observed in both, cows with metritis only and cows with puerperal metritis, which indicates that ampicillin cured metritis and puerperal metritis at a faster rate than ceftiofur. Early in the course of treatment, cows receivi ng ceftiofur had low er rectal temperature than those treated with ampicillin, but this difference reversed after study d 5 after which cows treated with ampicillin had low er rectal temperature than those treated with ceftiofur. The low er body temperature f or cows treated with ampicillin after the course of therapy is another indication that clinical efficacy in curing metritis was at least similar to that of ceftiofur. As anticipated, cows presenting puerperal metritis, a more acute debilitating form of the disease, had reduced cure rates at 5, 7 and 12 d after diagnosis, and antimicrobial therapy with either ampicillin or ceftiofur were equally efficacious at resolving puerperal metritis within the first 12 d after diagnosis. Ampicillin and ceftiofur are b oth beta lactam antibiotics that bind irreversibly to bacterial enzyme dd transpeptidase blocking the formation crosslinks between this enzyme and peptidoglycan, which compromises the formation of rigid cell wall synthesis in binary fission, which ultimate ly leading to cell lysis (Katzung, 2007). Ampicillin is

PAGE 247

247 classified as amino penicillin because of the presence of amino group as part of its core. This amino group is critical for ampicillin to penetrate the outer membrane of gram negative bacteria, which gives the drug its broad spectrum by killing both gram positive and gram negative bacteria (Katzung, 2007). Ceftiofur has also effectiveness against gram positive and gram negative bacteria and is resistant to beta lactamase, therefore, preventing the acti on of these enzymes on the degradation of the beta lactam ring, which inactivates many antibiotics of this class (Collatz et al., 2006). Ampicillin and ceftiofur are both effective against gram negative bacteria such as E. coli (Lehtolainen et al., 2003; S heldon et ., 2004), which is thought to initiate the infection and promote the initial tissue damage that facilitates establishment of other bacterial groups in the uterus. Efficacy against E. coli, particularly those expressing the virulence factor fimH ( Bicalho et al., 2012), might limit the extent of pathogenic bacteria colonization of the uterus and reduce the intensity of disease. It is interesting to note that 12.6% of the cows with metritis only developed fever after the antimicrobial therapy had bee n established. These cows were not diagnosed o C despite treatment with ampicillin or ceftiofur. In general, treating cows affected by puerperal metritis with antimicrobials r educes body temperature (Chenault et al., 2004; McLaughlin et al., 2012). Interestingly, the decline in body temperature is also observed in metritic cows that remain untreated (Chenault et al., 2004; McLaughlin et al., 2012), indicating normal resolution the proportion of cows that initiated antimicrobial therapy that develops fever has not

PAGE 248

248 been reported and it seems acceptable that a small portion of the treated cows might become worse or not respon d to the antibiotics. Administration of ceftiofur at the prescribed dose of 2.2 mg/kg results in concentrations of ceftiofur derivatives above those capable of inhibiting the growth of utero pathogenic bacteria (Drillich et al., 2006 c ; Sheldon et al., 20 04). When ampicillin is administered to cattle, concentrations increase immediately and reach approximately t 24 h (Gehring et al., 2005). Therefore, the sim ilar clinical efficacy of ampicillin and ceftiofur are likely related to their ability to maintain concentrations of antibiotics sufficient to inhibit the growth of the major pathogens that cause uterine disease. Many studies evaluating the clinical effica cy of antibiotics for therapy of metritis usually only evaluated the remission of the disease with reduction in fever and improvements in uterine discharge score (Chenault et al., 2004; McLaughlin et al., 2012). However, because of extensive spontaneous cu re even in cows with puerperal metritis (Chenault et al., 2004; McLaughlin et al., 2012), it is critical that the evaluation of therapy goes beyond clinical cure. In fact, proper treatment of metritis can influence milk yield and reproductive performance ( Goshen and Shpigel, 1996). Despite a faster cure and a small reduction in the prevalence of PVD, cows treated with ampicillin had similar prevalence of estrous cyclicity and P/AI at first postpartum insemination compared with cows treated with ceftiofur. I t has been reported consistently in the literature that PVD has detrimental impacts on fertility (Gilbert et al., 2005; Dubuc et al., 2011 ), and one would anticipate that reductions in the prevalence of PVD and

PAGE 249

249 subclinical endometritis would benefit fertil ity at first AI. However, it is possible that the changes in prevalence of PVD and cytological endometritis caused by either metritis or treatment of metritis were insufficient to impact fertility. All cows in the study were subjected to a presynchronized timed AI protocol, but it is unlikely that hormonal treatments, particularly PGF benefited uterine health that would influence P/AI ( Dubuc et al., 2011 ). Therefore, the improved cure rates and rectal temperature on d 7 after metritis diagnosis and reduc ed prevalence of PVD identified in cows treated with ampicillin in comparison with ceftiofur did not translate in improved reproductive performance in the current study. Although cows without metritis had a remarkable reduction in the incidence of PVD and subclinical endometritis in comparison with metritic cows treated with ampicillin and ceftiofur, no difference in the prevalence of estrous cyclic cows and P/ AI for the first postpartum insemination were identified between cows without metritis and those diagnosed with metritis. These results were surprising considering the well known negative associations among between uterine diseases and subsequent fertility (Dubuc et al., 2011). At this point, it is unclear why cows with metritis had similar P/AI and risk of pregnancy loss compared with those not diagnosed with metritis. It is possible that early diagnosis and prompt therapy might have minimized the negative impact of metritis on fertility. Goshen and Shpigel (2006) allocated cows diagnosed with metrit is to receive 4 treatments with 5 g of chlortetracycline as intrauterine boluses over the course of two weeks or to remain as untreated controls. The P/AI at first postpartum insemination were 38.3, 42.5, and 18% in cows without metritis, cows with metriti s treated with chlortetracycline, and cows with metritis that remained as untreated

PAGE 250

250 controls, respectively. Therefore, it is possible that proper antimicrobial therapy immediately after the diagnosis of metritis might reestablish fertility similar to that of unaffected cows. Conclusion Ampicillin was an efficacious alternative therapy for metritis resulting in faster cure rates than ceftiofur; however, by d 12 after the diagnosis of metritis, no differences between treatments were observed. Ampicillin reduc ed the prevalence of cows with PVD on d 32 postpartum compared with cows receiving ceftiofur, but treatment did not affect the prevalence of cytological endometritis in cows previously diagnosed with metritis. Similarly, treatment did not affect the resump tion of estrous cyclicity by 64 d postpartum, P/AI at first AI, and the risk of pregnancy loss. Although, cows without metritis had reduced prevalence of PVD and subclinical endometritis compared with metritic cows, estrous cyclicity and P/AI did not diffe r between those without metritis and metritic cows treated with ampicillin or ceftiofur. Type of metritis is an indicator of the severity of disease and cows with puerperal metritis have poorer cure rates than those with metritis alone. Nevertheless, type of metritis did not influence subsequent estrous cyclicity, P/AI or risk of pregnancy loss

PAGE 251

251 Table 8 1. Effect of antibiotic treatment on mean rectal temperature (RT) and incidence of fever in cows with only metritis or cows with puerperal metritis (metri tis and o C) Treatment 1 AMP CEFT P 2 Parameter Metritis 3 Puerperal Metritis Metritis Puerperal Metritis TRT Type TRT* Type Cows, n 152 107 160 109 Mean RT, 4 o C 39.04 39.31 39.02 39.22 0.09 0.01 0.20 Fever incidence, 5 % 12.5 33.2 12. 9 28.2 0.47 0.01 0.34 1 Cows with metritis were blocked by type of metritis and assigned randomly to receive 11 mg/kg of ampicillin or 2.2 mg/kg of ceftiofur once daily for 5 d. 2 TRT = effect of treatment; TM = effect of type of metritis; TRT x TM = inte raction between TRT and TM. 3 Metritis = cows with metritis only based on vaginal discharge score 5 (watery, reddish/brownish color of foul smell) and rectal temperature < 39.5 o C; P. metritis = puerperal metritis based on vaginal discharge score 5 and rec o C. 4 RT = mean rectal temperature from d 2 to 12 after enrollment. 5 Fever = incidence of fever (RT > 39.4 o C) from d 2 to 12 after enrollment.

PAGE 252

252 Table 8 2. Effect of treatment on estrous cyclicity, pregnancy per AI and pregnancy los s following the 1st postpartum insemination 1 Cows with metritis were blocked by type of metritis and assigned randomly to receive 11 mg/kg of ampicillin or 2.2 mg/kg of ceftiofur once daily for 5 d. No metritis cows were randomly selected at 12 d postpartum based on the day of calving and parity. 2 Estrous cyclic by 64 d postpartum based on the presence of a CL in at least one of the two ovaries examined on d 50 3 and 64 3 DI M

PAGE 253

253 Figure 8 1. Diagram of treatments for metritis, monitoring of cure, and evaluation of uterine health. Cows diagnosed with metritis were blocked by parity and type of metritis (only metritis vs. puerperal metritis) and, within each block, assigned rando mly to receive 11 mg/kg of ampic i llin or 2.2 mg/kg of ceftiofur once daily for 5 d. Metritis was defined as vaginal discharge (VD) score of 5 based on discharge of watery brown/red foul smell. Puerperal metritis was determined by VD score of 5 and rectal t discharge

PAGE 254

254 Figure 8 2. Diagram of reproductive program used for first insemination and pregnancy diagnosis. Cows diagnosed with metritis were blocked by parity and type of metritis (only metritis vs. p uerperal metritis) and, within each block, assigned randomly to receive 11 mg/kg of ampic i llin or 2.2 mg/kg of ceftiofur once daily for 5 d. Metritis was defined as vaginal discharge (VD) score of 5 based on discharge of watery brown/red foul smell. Puerpe ral cohort of herdmates without metritis were subjected to a presynchronized 5 d timed AI program. Cows observed in estrus any time after the se cond 64 3 d postpartum were inseminated

PAGE 255

255 Figure 8 3. Rectal temperature s on d 2, 3, 4, 5, 6, 7 and 12 after the diagnosis of metritis according to treatment. Cov = covariate value measured on the day of study enrollment. Cows diagnosed with metritis were blocked by parity and type of metritis (only metritis vs. puerperal metritis) and, within each block, assigned randomly to receive 11 mg/kg of ampic i llin or 2.2 mg/kg of ceftiofur once daily for 5 d. Metritis was defined as vaginal discharge (VD) sco re of 5 based on discharge of watery brown/red foul smell. Effect of treatment (P = 0.09), type of metritis (P < 0.001), day (P < 0.001), interaction between treatment and type of metritis (P = 0.20), and treatment and day (P < 0.001). Within a day, differ ent letters denote statistical difference ( a,b between treatments.

PAGE 256

256 Figure 8 4 Adjusted proportions ( SEM) of cows with metritis cured on d 5, 7 and 12 after initiation of treatments with ampicillin or ceftiofur. Cows diagnosed with metritis were blocked by parity and type of metritis ( only metritis vs. puerperal metritis) and, within each block, assigned randomly to receive 11 mg/kg of ampic i llin or 2.2 mg/kg of ceftiofur once daily for 5 d. Metritis was defined as vaginal discharge (VD) score of 5 based on discharge of watery brown/red foul smell. On panel A, cure of metritis was based on VD < 5. On panel B, cure of metritis was based on VD < 5 and rect al temperature (RT) < 39.5 C. Within a day, different letters denote statistical

PAGE 257

257 Figure 8 5 Adjusted proportions ( SEM) of cows with metritis cured on d 5, 7 and 12 after initiation of treatments according to initial diagnosis of metritis only or puerperal metritis. Cows diagnosed with metritis were blocked by parity and type of metritis (only metritis vs. puerperal metritis) and, within each block, assigned randomly to receive 11 mg/kg of ampic i llin or 2.2 mg/kg of ceftiofur once daily fo r 5 d. Metritis was defined as vaginal discharge (VD) score of 5 based on discharge of watery brown/red foul smell. On panel A, cure of metritis was based on VD < 5. On panel B, cure of metritis was based on VD < 5 and rectal temperature (RT) < 39.5 C. Wi

PAGE 258

258 Figure 8 6 Adjusted proportions ( SEM) of cows with metritis cured on d 5, 7 and 12 after initiation of treatments according to parity. Cows diagnosed with met ritis were blocked by parity and type of metritis (only metritis vs. puerperal metritis) and, within each block, assigned randomly to receive 11 mg/kg of ampic i llin or 2.2 mg/kg of ceftiofur once daily for 5 d. Metritis was defined as vaginal discharge (VD ) score of 5 based on discharge of watery brown/red foul smell. On panel A, cure of metritis was based on VD < 5. On panel B, cure of metritis was based on VD < 5 and rec tal temperature (RT) < 39.5 C. Within a day, different letters denote statistical dif ference ( a,b difference ( A,B

PAGE 259

259 Figure 8 7 Adjusted proportions ( SEM) of cows with metritis cured on d 5, 7 and 12 after initiation of treatments according to diagnosis of calving related disorders (dystocia, stillbirth, twins, and retained placenta). Cows diagnosed with metritis were blocked by parity and type of metritis (only metritis vs. puerperal metritis) and, within each block, assigned rando mly to receive 11 mg/kg of ampic i llin or 2.2 mg/kg of ceftiofur once daily for 5 d. Metritis was defined as vaginal discharge (VD) score of 5 based on discharge of watery brown/red foul smell. On panel A, cure of metritis was based on VD < 5. On panel B, c ure of metritis was based on VD < 5 and rectal temperature (RT) < 39.5 C. Within a day, different letters denote statistical difference ( a,b treatments. Within a day, different letters denote tendency for statistical difference ( A,B 0.10).

PAGE 260

260 Figure 8 8. Adjusted proportions ( SEM) of cows with purulent vaginal discharge on d 32 3 postpartum (panel A) or with cytological subclinical endometris on d 39 3 postpartum (panel B). Cows diagnosed with metritis were blocked by parit y and type of metritis (only metritis vs. puerperal metritis) and, within each block, assigned randomly to receive 11 mg/kg of ampic i llin or 2.2 mg/kg of ceftiofur once daily for 5 d. Metritis was defined as vaginal discharge (VD) score of 5 based on disch arge of watery brown/red foul smell. For panel A, effect of treatment (P < 0.001). For panel B, effect of treatment (P < 0.001). Within a day, different letters denote statistical difference ( a,b ,c

PAGE 261

261 CHAPTER 9 CONCLUSION S AND FUTURE DIRECTION S The studies presented in this dissertation contributed to advance the fields of reproductive physiology of dairy heifers, reproductive management of dairy cows, and uterine health of dairy cows having direct implication to dairy sciences and the dairy industry. The results presented in C hapter 3 revealed that In spite of the long re insemination interval for second and third AI, cows receiving 3TAI became pregnant at a faster rate than cows recei ving a single timed AI before introduction to natural service. The improved reproductive performance of 3TAI cows resulted in 15% greater hazard of pregnancy, 17% greater risk of pregnancy, and 9 fewer days nonpregnant than 1TAI cows. The faster pregnancy rate was likely the result of a combination of increased breeding of nonpregnant cows associated with improved probability of pregnancy to a breeding, which improved reproduction during the first 84 d in the study. Therefore, in herds in which detection of estrus is not carried out, a combination of AI and natural service is used, bulls are managed to optimize their fertility, and the resynchronized timed AI is implemented using the Ovsynch protocol with a CIDR insert, it is advantageous to allow cows multi ple inseminations before bull exposure for natural service to optimize pregnancy rate. Results from this study indicate that in herds in which detection of estrus is not carried out, and a combination of AI and natural service is used, cows should receive at least three timed AI before bull exposure for natural service. The results of C hapter 4 showed that administration of GnRH on the first day of the 5 d timed AI protocol did not influence P/AI or pregnancy loss of dairy heifers when

PAGE 262

262 a single treatment with PGF on day 5 of the protocol is used Although ovulation on study day 0 and presence of new CL 5 d later was low, treatment with GnRH increased ovulation rate in heifers compared with no GnRH treatment A s consequence of the increased ovu l ation and formation new CL the proportion of heifers with low progesterone at AI when receiving a single PGF treatment 5 d later was less for those receiving GnRH than controls Timing of induction of ovulation with GnRH relative to AI influenced P/AI of heifers not displaying estrus on the day of the timed AI In general, P/AI was better when heifers received the ovulatory stimulus concurrent with AI at 72 h after PGF than 16 h before timed AI. Therefore, when heifers are subjected to the 5 d timed AI program with a single treatment of PGF it is suggested that the initial GnRH is not necessary and the period of proestrus should be 72 h with administration of GnRH to induce ovulation concurrent with timed AI. The results of C hapter 5 complement th ose of stud y 1 in C hapter 4 showing i ncreased follicle turnover at initiation of 5 d timed AI program by using GnRH combined with two doses of PGF administered on d 5 and 6 to optimize luteolysis successful ly improved P/AI in dairy heifers. Results of the current s tudy demonstrate similar concepts of previous work with lactating dairy cows reinforcing the need for estrous and ovulation synchronization protocols to incorporate physiological principles to optimize fertility in dairy heifers. Follicle turnover by induc ing ovulation with GnRH, although low in dairy heifers, was beneficial to fertility. However, the benefit of GnRH to optimize fertility requires two doses of PGF administered 24 h apart to increase regression of a newly formed CL. The P/AI of approximately 60% obtained in the current study supports the use of the 5 d timed AI protocol as an alternative breeding program for reproductive

PAGE 263

263 management of heifers when detection of estrus is not used. Finally, it was demonstrated that high concentrations of progesterone when GnRH was administered suppressed the LH release and impaired ovulation. Further research is needed to determine if additional increase in ovulation to the initial GnRH of the 5 d timed AI protocol can further improve fertility in dairy heifers. The results f r o m C hapter 6 demonstrated that t reatment with one or two injections of PGF in early lactation before cows were subjected to a presynchronized timed AI protocol was unable to improve uterine health and measures of fertility in lactating dairy cows. Subclinical endometritis impaired P/AI and maintenance of pregnancy in lactating dairy cows, particularly when associated with PVD, and the negative effect of subclinical endometritis was observed when the inflammatory process persisted until 46 DIM. Interestingly, when both PVD and subclinical endometritis were associated or when sub clinical endometritis persisted by 46 DIM, pregnancy loss increased. Future research should focus on understand ing the mechanism leading to persistent inflammatory diseases in dairy cows and to develop new s trategies that can mitigate the negative impact o f subclinical endometritis on fertility. Chapter 7 revealed that intrauterine infusion with T. pyogenes disrupts luteal function and le a d s to early demise of CL at least in part of the cows. Although, T. pyogenes clearly induced endometrial inflammation w e were unable to identify a consistent response in endometrial mRNA expression of genes linked to inflammation and the luteolytic cascade that could support a direct anticipated interplay between the bacterium induced inflammat ory response and factors trig gering the early demise of the

PAGE 264

264 CL Future research should focus on investigat ing the immune response and activation of the endometrial luteolytic cascade in cows that undergo early demise of the CL. The results from C hapter 8 lead to the co n clusion that a m picillin was an efficacious alternative therapy for metritis resulting in faster cure rates than ceftiofur; however, by d ay 12 after the diagnosis of metritis, no differences between treatments were observed. Ampicillin reduced the prevalence of cows with PVD on d ay 32 postpartum compared with cows receiving ceftiofur, but treatment did not affect the prevalence of cytological endometritis in cows previously diagnosed with metritis. Similarly, treatment did not affect the resumption of estrous cyclicity by 64 d ays postpartum, P/AI at first AI, and the risk of pregnancy loss. Although, cows without metritis had reduced prevalence of PVD and subclinical endometritis compared with metritic cows, estrous cyclicity and P/AI did not differ between those without me tritis and metritic cows treated with ampicillin or ceftiofur. Type of metritis is an indicator of the severity of disease and cows with puerperal metritis have poorer cure rates than those with metritis alone. Nevertheless, type of metritis did not influe nce subsequent estrous cyclicity, P/AI or risk of pregnancy loss. Future research should investigate the cost benefit of using ampicillin in comparison to ceftiofur and characterize the microbiome of cows before, during and after treatment with ampicillin to determine how this antimicrobial therapy affect s the microbial uterine flora and leads to cure of metritis.

PAGE 265

265 LIST OF REFERENCES Acosta, T. J., N. Yoshizawa, M. Ohtani, A. Miyamoto. 2002. Local changes in blood flow within the early and midcycle corpus luteum after prostaglandin F(2 alpha) injection in the cow. Biol. Reprod. 66:651 658. Adams, G. P., R. Jaiswal, J. Singh, and P. Malhi. 2008. Progress in understanding ovarian follicular dynamics in cattle. Theriogenology 69:72 80. Adams, G. P., R. L. M atteri, J. P. Kastelic, J. C. H. Ko, and O. J. Ginther. 1992. Association between surges of follicle stimulating hormone and the emergence of follicular waves in heifers. J. Reprod. Fertil. 94:177 188. Al Sadi, H. I., A. F. Majeed, and A. M. Ridha. 1994. Histopathology of retained bovine fetal membranes. Theriogenology 42:273 278. Amann, R. I., W. Ludwig, K. H. Schleifer. 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143 169. Ashkar, A. A., K. L. Mossman, B. K. Coombes, C. L. Gyles, R. Mackenzie. 2008. FimH Adhesin of Type 1 fimbriae is a potent inducer of innate antimicrobial responses which requires TLR4 and type 1 interferon signaling. PLoS Pathog. 4(12) e1000233. Atli, M O., W. R. W. Bender, V. Mehta, M. R. Bastos, W. Luo, C. M. Vezina, and M. C. Wiltbank. 2012. Patterns of gene expression in the bovine corpus luteum following repeated intrauterine infusions of low doses of prostaglandin F ( 2alpha ) Biol. Reprod. 86:1 13. Aust, G., C. Simchen, U., Heider, F. A. Hmeidan, V. Blumenauer, K. Spanel Borowski. 2000. Eosinophils in the human corpus luteum: the role of RANTES and eotaxin in eosinophil attraction into periovulatory structures. Mol. Hum. Reprod. 6:1085 91. Bauer, M., I. Reibiger, K. Spanel Borowski. 2001. Leucocyte proliferation in the bovine corpus luteum. Reproduction 121:297 305. Bazer, F. W., T. E. Spencer, and T. L. Ott. 1997. Interferon tau: a novel pregnancy recognition signal. Am. J. Reprod. Immunol. 37:4 12 420. Bazer, F. W., R. C. Burghardt, G. A. Johnson, T. E. Spencer, G. Wu. 2008: Interferons and progesterone for establishment and maintenance of pregnancy: interactions among novel cell signaling pathways. Reprod Biol 8:179 211.

PAGE 266

266 Beal, W. E., R. A. Mil vae, and W. Hansel. 1980. Oestrous cycle length and plasma in the bovine oestrous cycle. J. Reprod. Fertil. 59:393 396. Beindorff, N., K. Nagai, K. Shirasuna, K. Herzog K. H oeffmann, M. Sasaki, H. Bollwein, A. Miyamoto. 2010. Vascular changes in the corpus luteum during early pregnancy in the cow. J. Reprod. Dev. 256:263 70. Bello, N.M., J.P. Steibel, and J.R. Pursley. 2006. Optimizing ovulation to first GnRH improved outcom es to each hormonal injection of Ovsynch in lactating dairy cows. J. Dairy Sci. 89:3413 3424. Benyo, D. F., G. K. Haibel, H. B. Laufman, and J. L. Pate. 1991. Expression of major histocompatibility complex antigens on the bovine corpus luteum during the e strous cycle, luteolysis, and early pregnancy. Biol Reprod 45:229 234. Benzaquen, M. E., Risco, C. A., Archbald, L. F., Melendez, P., Thatcher, M. J., and Thatcher W. W. (2007). Rectal temperature, calving related factors, and the incidence of puerperal metritis in postpartum dairy cows. Journal of Dairy Science 90:2804 2814. Berisha, B., D. Schams, M. Kosmann, W. Amselgruber, R. Einspanier 2000. Expression and tissue concentration of vascular endothelial growth factor, its receptors, and localizatio n in the bovine corpus luteum during estrous cycle and pregnancy. Biol Reprod 63:1106 14. Bicalho, M. L. S., V. S. Machado, G. Oikonomou, R. O. Gilbert, R. C. Bicalho. 2012. Association between virulence factors of Escherichia coli, Fusobacterium necroph orum, and Arcanobacterium pyogenes and uterine diseases of dairy cows. Vet. Microbiol. 157:125 131 Bicalho, R. C., S. H. Cheong, L. D. Warnick, and C. L. Guard. 2007. Evaluation of progesterone supplementation in a prostaglandin f2 based presynchronization protocol before timed insemination. J. Dairy Sci. 90:1193 1200. Bicalho, R.C., V. S. Machado, M. L. Bicalho, R. O. Gilbert, A. G. Teixeira, L. S. Caixeta, R. V. Pereira, R.V., 2010. Molecular and epidemiological characterization of bovine intrauterine Escherichia coli. J. Dairy Sci. 93:5818 5830. Binelli, M., P. Subramaniam, T. Diaz, G. A. Johnson, T. R. Hansen, L Badinga and W. W. Thatcher. 2001. Bovine interferon tau stimulates the Janus kinase signal transducer and activator of transcription pathway in bovine endometrial epithelial cells. Biol. Reprod. 64:654 665.

PAGE 267

26 7 Bisinotto, R. S, E. S Ribeiro, L T Martins, R. S. Marsola, L. F Greco, M. G Favoreto, C. A Risco, W. W. Thatcher and J. E. P Santos. 2010. Effect of interval between induction of ovulation and AI and supplemental progesterone for resynchronization on fertility of dairy cows subjected to a 5 d timed AI program. J Dairy Sci. 93:5798 5808. Bisinotto, R. S., E. S. Ribeiro F. S. Lima N. Martinez L. F. Greco L. F. S. P. Barbosa P. P. Bueno L. F. S. Scagion W. W. Thatcher and J. E. P. Santos 2013. Targeted progesterone supple mentation improves fertility in lactating dairy cows without a corpus luteum at the initiation of the timed artificial insemina tion protocol. J. Dairy Sci. 96 :2214 2225 Bleach, E.C.L., R.G. Glencross, and P.G. Knight. 2004. Association between ovarian fo llicle development and pregnancy rates in dairy cows undergoing spontaneous oestrous cycle. Reproduction 127:621 629. Bodensteiner, K. J., K. Kot, M. C. Wiltbank, and O. J. Ginther. 1996. Synchronization of emergence of follicle waves in cattle. Theriogen ology 43:31 40. Boer H. M. T., R. F. Veerkamp, B. Beerda, and H. Woelders. 2012. Estrous behavior in dairy cows: identification of underlying mechanisms and gene functions. Animal 3:446 453 Bolinder, A., B. Seguin, H. Kindahl D. Bouley, and D. Otterby 1988. Retained fetal membranes in cows: Manual removal versus non removal and its effect on reproductive performance. Theriogenology 30:45 56. Boos, A., J. Kohtes, A. Stelljes, H. Zerbe, H. H. Thole.2000. Immuno histochemical assessment of progesterone, oestrogen and glucocorticoid receptors in bovine placentomes during pregnancy, induced parturition, and after birth with or without retention of fetal membranes. J. Reprod. Fertil. 120:351 360. Borsberry, S., and H. Dobson. 1989. Peripartal diseases and their effect on reproductive performance in five dairy herds. Vet. Rec. 124:217 219. Brnnstrm, M., B. Fridn. 1997. Immune regulation of corpus luteum function. Semin Reprod Endocrinol 15:363 70. Braw Tal, R., and S. Yossefi. 1997. Studies in vivo and in vitro on the initiation of follicle growth in the bovine ovary. J. Reprod. Fertil. 109:165 171. Bretzlaff, K.N., R. S.Ott, G. D. Koritz, R.F. Bevill, B. K. Gustafsson, and L. E. Davis. 1983. Distribution of oxytetracycline in genital tract tissues of p ostpartum cows given the drug by intravenous and intrauterine routes. Am. J. Vet. Res. 44:764 769.

PAGE 268

268 Brick, T. A., and G. M. Schuenemann, S. Bas J. B. Daniels C. R. Pinto D. M. Rings and P. J. Rajala Schultz. 2012. Effect of intrauterine dextrose or antibiotic therapy on reproductive performance of lactating dairy cows diagnosed with clinical endometritis. J. Dairy Sci. 95 :1894 1905. Bridges, G.A., L.A. Helser, D.E. Grum, M.L. Mussard, and M.L. Day. 2008. Decreasing the interval between GnRH and PG F2a from 7 to 5 days and lengthening proestrus increases timed AI pregnancy rates in beef cows. Theriogenology 69:843 851. Britt, J. H. 1985. Enhanced reproduction and its economic implications. J. Dairy Sci. 68:1585 1592. Bromfield, J. J., I. M. Sheldon 2011. Lipopolysaccharide initiates inflammation in bovine granulosa cells via the TLR4 pathway and perturbs oocyte meiotic progression in vitro. Endocrinology 2011; 152:5029 5040. Bromfield, J. J., I. M. Sheldon. 2013. Lipopolysaccharide Reduces the Pri mordial Follicle Pool in the Bovine Ovarian Cortex Ex Vivo and in the Murine Ovary In Vivo. Biol. Reprod. 88:1 9. Brusveen, D. J., A.P. Cunha, C.D. Silva, P.M. Cunha, R.A. Sterry, E.P.B. Silva, J. N. Guenther, and M.C. Wiltbank. 2008. Altering the time of the second gonadotropin releasing hormone injection and artificial insemination (AI) during Ovsynch affects pregnancies per AI in lactating dairy cows. J. Dairy Sci. 91:1044 1052. Buratini Jr., J., and C. A. Price. 2011. Follicular somatic cell factors and follicle development. Reprod. Fertil. Dev. 23:32 39. Burke, J. M., R. L. De La Sota, C. A. Risco, C. R. Staples, E. J. P. Schmitt, and W. W. Thatcher. 1996. Evaluation of timed insemination using a gonadotropin releasing hormone agonist in lactating d airy cows. J. Dairy Sci. 79:1385 1393. Burrows, G. E. R. J. Morton, and W. H. Fales. 1993. Microdilution antimicrobial susceptibilities of selected gram negative veterinary bacterial isolates. J. Vet. Diagn. Invest. 5:541 547. Butler, S. T., A. L. Marr S. H. Pelton, R. P. Radcliff, M. C. Lucy, and W. R. Butler. 2003. Insulin restores GH responsiveness during lactation induced negative energy balance in dairy cattle: effects on expression of IGF I and GH receptor 1A. J. Endocrinol. 176:205 217. Callaha n, C.J., and L. A. Horstman. 1987. Treatment of early postpartum metritis in a dairy herd: response and subsequent fertility. Bov. Prac. 22:124 128.

PAGE 269

269 Campbell, B. K., R. J. Scaramuzzi, and R. Webb.1996. Induction and maintenance of oestradiol and immunor eactive inhibin production with FSH by ovine granulosa cells cultured in serum free media. J. Reprod. Fertil. 106:7 16. Caraviello, D. Z., K. A. Weigel, P. M. Fricke, M. C. Wiltbank, M. J. Florent, N. B. Cook, K. V. Nordlund, N. R. Zward, and C. L. Rawson 2006. Survey of management practices on reproductive performance of dairy cattle on large US commercial farms. J. Dairy Sci. 89:4723 4735. Cerri, R.L.A., H.M. Rutigliano, R.G. Bruno, and J.E.P. Santos. 2009b. Progesterone concentration, follicular devel opment and induction of cyclicity in dairy cows receiving intravaginal progesterone inserts. Anim. Reprod. Sci. 110:56 70. Cerri, R. L. A., H. M. Rutigliano, R. C. Chebel, and J. E. P. Santos. 2009a. Period of dominance of the ovulatory follicle influences embryo quality in lactating dairy cows. Reproduction 137:813 823. Cerri, R. L. A., H. M. Rutigliano, F. S. Lima, D. B. Arajo, and J. E. P. Santos. 2009c. Effect of source of supplemental selenium on uterine health and embryo quality in high producing da iry cows. Theriogenology 71:1127 1137. Champagne, J. D, J. H. Kirk and J. P. Reynolds. 2002. Bull management practices on California dairies: Implications for education and veterinary services. Pages 15 21 in Proc. 15th Annual Fall Symposium University of California, Davis. Chandra, V., J. Jasti, P. Kaur, C. Betzel, A. Srinivasan, and T. P. Singh. 2002. First structural evidence of a specific inhibition of phospholipase A2 by alpha toco pherol (vitamin E) and its implications in inflammation: crystal str ucture of the complex formed between phospholipase A2 and alpha tocopherol at 18 resolution. J Mol. Biol. 320:215 222. Chapinal, N., Carson, M., Duffield, T.F., Capel, M., Goden, S., Overton, M., Santos, J.E.P., & LeBlanc S.J. (2011). Journal of Dairy Science 94:4897 4903. Chebel, R. C., J. E. P. Santos, R. L. A. Cerri, H. M. Rutigliano, and R. G. S. Bruno. 2006. Reproduction in dairy cows following progesterone insert presynchronization and resynchronization protocols. J. Dairy Sci. 89:4205 4219. Che nault, J.R., J. F. McAllister, T. Chester, K. J. Dame, F, M. Kausche, E. J. Robb. 2004. Efficacy of ceftiofur hydrochloride administered parenterally for the treatment of acute postpartum metritis in dairy cows. J. Am. Vet. Med. Assoc. 224:1634 1639. Chen ault, J. R., W. W. Thatcher, P. S. Kalra, R. M. Abrams, and C. J. Wilcox. 1974. Transitory changes in plasma progestins, estradiol, and luteinizing hormone approaching ovulation in the bovine. J. Dairy Sci. 58:709 717.

PAGE 270

270 Chenoweth, P. J. 1992. A new bull br eeding soundness evaluation form. Pages 63 70 in Proc. Annual Meeting, Society for Theriogenology, San Antonio, TX. Christ, W. J., O. Asano, A. L. C. Robidoux, M. Perez, Y. Wang, G. Dubuc, W. E. Gavin, L. D. Hawkins, P. D. McGuinness, M. A. Mullarkey, M D. Lewis, Y. Kishi, T. Kawata, John R. Bristol, Jeffrey R. Rose, D. P. Rossignol, S. Kobayashi, l. Hishinuma, A. Kimura, N. Asakawa, K. Katayama, l. Yamatsu E5531, a pure endotoxin antagonist of high potency. Science 1995; 268, 80 83. Clarke, I. J., and S. Pompolo. 2005. Synthesis and secretion of GnRH. Anim. Reprod. Sci. 88:29 55. 141. Cochran, S. D., J. Cole D. J. Null, and P. J. Hansen1. 2013. Discovery of single nucleotide polymorp hisms in candidate genes associated with fertility and production traits in Holstein cattle. BMC Genetics. 14:49. Cochet, M ., D. Vaiman, F. Lefe`vre. 2009. Novel interferon delta genes in mammals: cloning of one gene from the sheep, two genes expressed by the horse conceptus and discovery of related sequences in several taxa by genomic database screening. Gene 433:88 99. Colazo, M. G., and D.J. Ambrose. 2011. Neither duration of progesterone insert nor initial GnRH treatment affected pregnancy per timed insemination in dairy heifers subjected to a Co synch protoc ol. Theriogenology 76: 578 588. Colazo, M. G., J. P. Kastelic, H. Davis, M. D. Rutledge, M. F. Martinez, J. A. Small, and R. J. Mapletoft. 2008. Effects of plasma progesterone concentrations on L H release and ovulation in beef cattle given GnRH. Domest. Anim. Endocrinol. 34:109 117. Connell, I., W. Agace, P. Klemm, M. Schembri, S. Marild, and C. Svanborg. 1996. Type 1 fimbrial expression enhances Escherichia coli virulence for the urinary tract. Proc. Natl. Acad. Sci. 93:9827 9832. Connolly, D. T., 1991. Vascular permeability factor: a unique regulator of blood vessel function. J. Cell Biochem 47:219 23. Cortvrindt, R., J. Smitz, and A. C. Van Steirteghem. 1997. Assessment of the need for fo llicle stimulating hormone in early preantral mouse follicle culture in vitro. Hum. Reprod. 12:759 768.

PAGE 271

271 Cronin, J. G., M. L. Turner, L. Goetze, C. E. Bryant, I. M. Sheldon. 2011. Toll like receptor 4 and MyD88 dependent signaling mechanisms of the innate immune system are essential for the response to lipopolysaccharide by epithelial and stromal cells of the bovine endometrium. Biol. Reprod. 86:1 9. Cruppe, L. H., L. A. Souto, M. Maquivar, P. Gunn, M. L. Mussard, D. Wolfenson, A.V. Pires, G. A. Bridges, a nd M. L. Day. 2010. Use of two coincident doses of PGF2 with the 5 d CO Synch plus CIDR estrous synchronization program. J. Anim. Sci. 88 (E Suppl. 2):768 Davidson, J. A., U. Tiemann, J. G. Betts and P. J. Hansen. 1995. DNA synthesis and prostaglandin s ecretion by bovine endometrial cells as regulated by interleunkin 1. Reprod. Fert. I. Dev. 7:1037 1043. De Vries, A. 2004. Economics of delayed replacement when cow performance is seasonal. J. Dairy Sci. 87:2947 2958. De Vries, A., and C. A. Risco. 2005. Trends and seasonality of reproductive performance in Florida and Georgia dairy herds from 1976 to 2002. J. Dairy Sci. 88:3155 3165. De Vries, A., C. Steenholdt, and C. A. Risco. 2005. Pregnancy rates and milk production in natural service and artifici ally inseminated dairy herds in Florida and Georgia. J. Dairy Sci. 88:948 56. Deluyker, H.A., J. M. Gay, L.D. Weaver, A. S. Azari. 1991. Changes in milk yield with clinical diseases for a high producing dairy herd. J. Dairy Sci. 74:436 445. Dewey, S.T., L. G. Mendona, G. Jr. Lopes, F. A. Rivera, F. Guagnini, R. C. Chebel, T. R. Bilby. 2010. Resynchronization strategies to improve fertility in lactating dairy cows utilizing a presynchronization injection of GnRH or supplemental progesterone: I. Pregnancy rates and ovarian responses. J Dairy Sci. 93:4086 95. Dias, F. C. F., M. G. Colazo, J. P. Kastelic, R. J. Mapletoft, G. P. Adams, and J. Singh. 2010. Progesterone concentration, estradiol pretreatment, and dose of gonadotropin releasing hormone affect gon adotropin releasing hormone mediated luteinizing hormone release in beef heifers. Domest. Anim. Endocrinol. 39:155 162. Diskin, M.G., 2008. HeatWatch: a telemetric system for heat detection in cattle. Vet. Q. 30:37 48. Dohmen, M.J.W., J. A. C. M. Lohuis G. Huszenicza, P. Nagy, and M. Gacs. 1995. The relationship between bacteriological and clinical findings in cows with subacute/chronic endometritis. Theriogenology 43:1379 1388.

PAGE 272

272 Dohmen, M.J., Joop, K., Sturk, A., Bols, P.E., Lohuis, J.A., 2000. Relatio nship between intra uterine bacterial contamination, endotoxin levels and the development of endometritis in postpartum cows with dystocia or retained placenta. Theriogenology 54:1019 1032. Donofrio, G., L. Ravaneti, S. Cavirani, S. Herath, A. Capocefalo, I. M. Sheldon. 2008. Bacterial infection of endometrial stromal cells influences bovine herpesvirus 4 immediate early gene activation: a new insight into bacterial and viral interaction for uterine disease. Reproduction 136:361 366. Dransfield, M. B. G., R. L. Nebel, R. E. Pearson, and L. D. Warnick. 1998. Timing of insemination for dairy cows identified in estrus by a radiotelemetric estrus detection system. J. Dairy Sci. 81:1874 1882. Drillich, M., S. Arlt, S. Kersting, A. A. Bergwerff, P. Scherpenisse and W. Heuwieser. 2006c. Ceftiofur derivatives in serum, uterine tissues, cotyledons, and lochia after fetal membrane retention. J. Dairy Sci. 89:3431 3438. Drillich, M., Beetz, O., Pfutzner, A., Sabin, M., Sabin, H. J., Kutzer, P., Nattermann, H., Heuw ieser, W., 2001. Evaluation of a systemic antibiotic treatment of toxic puerperal metritis in dairy cows. J. Dairy Sci. 84:2010 2017. Drillich, M., O. Beetz, A. Pfutzner, M. Sabin, H. J. Sabin, P. Kutzer, H. Nattermann, W. Heuwieser.2001. Evaluation of a systemic antibiotic treatment of toxic puerperal metritis in dairy cows. J. Dairy Sci. 84:2010 2017. Drillich, M., M. Mahistedt, U. Reichert, B. A. Tenhagen, and W. Heuwieser. 2006a. Strategies to improve the therapy of retained fetal membranes in dairy c ows. J. Dairy Sci 89:627 635. Drillich, M., D. Raab, M. Wittke, and W. Heuwieser. 2005. Treatment of chronic endometritis in dairy cows with an intrauterine application of enzymes. A field trial. Theriogenology 63:1811 1823. Drillich, M., U. Reichert, M. Mahlstedt, and W. Heuwieser. 2006b. Comparison of two strategies for systemic antibiotic treatment of dairy cows with retained fetal membranes: Preventive vs. selective treatment. J. Dairy Sci. 89:1502 1508. Drillich, M., Sabin, M., Sabin, H. J., Heuwies er, W., 2003. Comaparison of two protocols for the treatment of retained fetal membranes in dairy cattle. Theriogenology 59, 951 960.

PAGE 273

273 Drillich, M., D. Tesfaye, F. Rings, K. Schellander, W. Heuwieser, and M. Hoelker. 2012. Effects of polymorphonuclear neut rophil infiltration into the endometrial environment on embryonic development in superovulated cows. Theriogenology 77:570 578. Drillich, M., D. Voigt, D. Forderung, and W. Heuwieser. 2007. Treatment of acute puerperal metritis with flunixin meglumine in addition to antibiotic treatment. J. Dairy Sci. 90, 3758 3763. Dubuc, J., T. F. Duffield, K. E. Leslie, J. S. Walton, and S. J. LeBlanc. 2010. Definitions and diagnosis of postpartum endometritis in dairy cows. J. Dairy Sci. 93:5225 5234. Dubuc, J., T. F Duffield, K. E. Leslie, J. S. Walton, and S. J. LeBlanc. 2011. Randomized clinical trial of antibiotic and prostaglandin treatments for uterine health and reproductive performance in dairy cows. J. Dairy Sci. 94:1325 1338. Eiler, H., 1997. Retained pla centa. In: Youngquist, R.S. (Ed.), Current Therapy in Large Animal Theriogenology. W.B. Saunders Co., Philadelphia, PA, pp. 340 348. Eiler, H., and F. M. Hopkins. 1993. Successful treatment of retained placenta with umbilical cord injections of collagenas e in cows. J Am. Vet. Med. Assoc. 203:436 443. El Zarkouny, S. Z., J.A. Cartmill, B. A. Hensley, and J. S. Stevenson. 2004. Pregnancy in dairy cows after synchronized ovulation regimens with or without presynchronization and progesterone. J. Dairy Sci. 87 :1024 1037. Elanco Animal Health. 2009. The 5 point body condition scoring system. Bulletin AI 10752. Elanco Animal Health, Greenfield, IN. Embry, C.A., L. Franchi, G. Nunez, and T. C. Mitchell. 2011. Mechanism of impaired NLRP3 inflammasome priming by m onophosphoryl lipid A. Sci. Signal. 4, ra28. Esslemont, R. J., and M. J. Bryant. 1976. Oestrous behaviour in a herd of dairy cows. Veterinary Record 99:472 475. Ettema, J., 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. Evans, A. C., J. E. Fortune. 1997. Selection of the dominant follicle in cattle occurs in the absence of differences in the expression of messenger ribonucleic a cid for gonado tropin receptors. Endocrinology 138:2963 29 71.

PAGE 274

274 Fairchild, D. L., and J. L Pate. 1991. Modulation of bovine luteal cell synthetic capacity by interferon gamma. Biol. Reprod. 44:357 363. Fairchild, D. L., and J. L. Pate. 1989. Interferon gam ma induction of major histocompatibility complex antigens on cultured bovine luteal cells. Biol. Reprod. 40:453 457. Fang, J. Using SAS procedures FREQ, GENMOD, LOGISTIC, and PHREG to estimate adjusted relative risks a case study. 2011. In Proc. SAS Glo bal Forum; Paper 345 2011. Farin, C. E., C. L. Moeller, H. R. Sawyer, F. Gamboni, and G. D. Niswender. 1986. Morphometric analysis of cell types in the ovine corpus luteum throughout the estrous cycle. Biol. Reprod. 35:1299 1308. Ferguson, J. D, D. T. G alligan, and N. Thomsen. 1994. Principal descriptors of body condition score in Holstein dairy cattle. J Dairy Sci. 77:2695 2703. Ferrara, N., T. Davis Smyth. 1997. The biology of vascular endothelial growth factor. Endocr Rev 18:4 25. Figueiredo, M. D. J. N. Moore, M. L. Vandenplas, W. C. Sun, T. F. Murray. 2008. Effects of the second generation synthetic lipid A analogue E5564 on responses to endotoxin in [corrected] equine whole blood and monocytes. Am. J. Vet. Res. 69:796 803. Findlay J. K., and A. E. Drummond. 1999. Regulation of the FSH receptor in the ovary. Trends Endocrinol. Metab. 10:183 188. Fortune, J. E. 2003. The early stages of follicular development: activation of primordial follicles and growth of preantral follicles. Anim. Reprod. Sci 78:135 163. Fortune, J. E., S. Kito, S. A. Wandji, and V. Srsen. 1998. Activation of bovine and baboon primordial follicles in vitro. Theriogenology 49:441 449. Fortune, J. E., M. Y. Yang, and W. Muruvi. 2010. The earliest stages of follicle developme nt: follicle formation and activation. Reproduction in domestic ruminants VII. Nottingham University Press. 67:203 216. Fourichon, C., H. Seegers, and X. Malher. 2000. Effect of disease on reproduction in the dairy cow: a meta analysis. Theriogenology. 53 :1729 1759. Fourichon, C., H.Seegers, N. Bareille, F. Beaudeau. 1999. Effects of disease on milk production in the dairy cow: a review. Preventive Veterinary Medicine 41:1 35.

PAGE 275

275 Frandson, R., W. L. Wilke, A. D. Fails. 2003. Anatomy and physiology of farm animals. Lippincott Williams and Wilkins, Baltimore 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. Galvo, K. N. P. Federico A. De Vries, and G. M. Schuenemann. 2013. Economic comparison of reproductive programs for dairy herds using estrus detection, timed artificial insemination, or a combination. J. Dairy Sci. 96 :2681 2693 Galvo, K. N., M. Frajblat, S. B. B rittin, W. R. Butler, C. L. Guard, and R. O. Gilbert. cows. J. Dairy Sci. 92:4906 4913. Galvo, K. N., M. Frajblat, W. R. Butler, S. B. Brittin, C.L. Guard and R.O. Gilb ert. 2010. Effect of early postpartum ovulation on fertility in dairy cows. Reprod. Domest. Anim. 45:207 11. Galvo, K. N., L. F. Greco, J. M. Vilela, M. F. S Filho, and J. E. P. Santos. 2009b. Effect of intrauterine infusion of ceftiofur on uterine heal th and fertility in dairy cows. J. Dairy Sci. 92:1532 1542. Galvo, K. N., M. F. S Filho, and J. E. P. Santos. 2007. Reducing the interval from presynchronization to initiation of timed artificial insemination improves fertility in dairy cows. J. Dairy S ci. 90:4212 4218. Gehring, R., D. van der Merwe, A.N. Pierce, R.E. Baynes, A.L. Craigmill, and J.E. Riviere. 2005. Multivariate meta analysis of pharmacokinetic studies of ampicillin trihydrate in cattle. Am. J. Vet. Res. 66:108 112. Gifford, C. A., K. R acicot, D. S. Clark, K. J. Austin, T. R. Hansen, M. C. Lucy, C. J. Davies, T. L. Ott. 2007. Regulation of interferon stimulated genes in peripheral blood leukocytes in pregnant and bred, nonpregnant dairy cows. J. Dairy Sci. 90:274 280 Gilbert, R. O., an d W. S. Schwark. 1992. Pharmacologic considerations in the management of peripartum conditions in the cow. Vet. Clin. North Am.: Food Ani. Prac. 8:29 56 Gilbert, R. O., S. T. Shin, C.L. Guard, H. N. Erb, and M. Frajblat. 2005. Incidence of endometritis and its effects on reproductive performance of dairy cows. Theriogenology 64:1879 1888. Ginther, O. J. 1998. Ultrasonic Imaging and Animal Reproduction: Cattle. Book 3. Equiservices Publishing, Cross Plains, WI.

PAGE 276

276 Ginther, O. J., D. R. Bergfelt, L. J. Kuli ck, and K. Kot. 2000. Selection of the dominant follicle in cattle: role of estradiol. Biol. Reprod. 63:383 389. Ginther, O. J., E. L. Gastal, M. O. Gastal, and M. A. Beg. 2007. Effect of prostaglandin F2alpha on ovarian, adrenal, and pituitary hormones a nd on luteal blood flow in mares. Domest Anim Endocrinol 32:315 28. Ginther, O. J., L. Knopf, and J. P. Kastelic. 1989. Temporal associations among ovarian events in cattle during oestrous cycles with two and three follicular waves. J. Reprod. Fertil. 87: 223 230. Ginther O. J., K. Kot, L. J. Kulick, and M. C. Wiltbank. 1997. Emergence and deviation of follicles during development of follicular waves incattle. Theriogenology 48:75 87. Ginther, O. J, H. K. Shrestha, M. J. Fuenzalida, A. K. Shahiduzzaman, M. A. Hannan, and M. A. Beg. 2010. Intrapulse temporality between pulses of a metabolite of prostaglandin F2a and circulating concentrations of progesterone before, during, and after spontaneous luteoly sis in heifers. Theriogenology 74:1179 1186. Ginther, O. J., H. K. Shrestha, M. J. Fuenzalida, S. Imam, and M. A. Beg. 2010. Stimulation of pulses of 13,14 dihydro 15 keto PGF2 (PGFM) with estradiol 17 and changes in circulating progesterone concentrations within a PGFM pulse in heifers. Theriogenology 74:3 84 392. Ginther, O. J., Shrestha, H. K., Fuenzalida, M. J., Imam, S., Beg, M. A., 2010. Stimulation of pulses of 13,14 dihydro 15 keto PGF2 (PGFM) with estradiol 17 and changes in circulating progesterone concentrations within a PGFM pulse in heifers. Th eriogenology 74, 384 392. Giordano, J. O., P. M. Fricke, M. C. Wiltbank, and V. E. Cabrera. 2011. An economic decision making support system for selection of reproductive management programs on dairy farms. J. Dairy Sci. 94:6216 6232. Giordano J. O., M. C. Wiltbank J. N. Guenther R. Pawlisch S. Bas A. P. Cunha and P. M. Fricke. 2012. Increased fertility in lactating dairy cows resynchronized with Double Ovsynch compared with Ovsynch initiated 32 d after timed artificial insemination. J. Dair y Sci. 95 :639 653. Goshen, T., & Shpigel, Y. 2006. Evaluation of intrauterine antibiotic treatment of clinical metritis and retained fetal membranes in dairy cows. Theriogenology 66:2210 2218.

PAGE 277

277 Gougeon, A., and B. Lefevre. 1983. Evolution of the diameter s of the largest healthy and atretic follicles during the human menstrual cycle. J. Reprod. Fertil. 69:497 502. Griffin, J. F. T, P. J. Hartigan, W. R. Nunn. 1974. Non specific uterine infection and bovine fertility. I. Infection patterns and endometritis during the first seven weeks post partum. Theriogenology 1:91 106. Grohn, Y.T., H. N. Erb, C. E. McCullogh, H. S. Saloniemi. 1990. Epidemiology of reproductive disorders in dairy cattle: associations among host characteristics, disease and production. P rev. Vet. Med. 8:25 39. Gross, T. S., and W. F. Williams. 1988. Bovine placental prostaglandin synthesis: principal cell synthesis as modulated by the binucleate cell. Biol. Reprod. 38:1027 1034. Grunert, E. 1986. Etiology of retained bovine placenta. In Current Therapy in Theriogenology. 2nd edn. Ed D. A. Morrow. W. B. Saunders. pp 237 242. Gustafsson, B.K. 1984. Therapeutic strategies involving antimicrobial treatment of the uterus in large animals. J. Am. Vet. Med. Assoc. 185:1194 1198. Handelsman, J 2004. Metagenomics: Application of Genomics to Uncultured Microorganisms. Microbiol. Mol. Biol. Rev. 68:669 685. Hansen, T. R., L. K. Henkes, R. L. Ashley, R. C. Bott, A. Q. Antoniazzi, H. Han. 2010. Endocrine actions of interferon tau in ruminants. Soc Reprod Fertil Suppl 67:325 340. Hansen, P. J., P. Soto, and R. P. Natzke. 2004. Mastitis and fertility in cattle possible involvement of inflammation or immune activation in embryonic mortality. Am. J. Reprod. 60:1104 1109. Hayashi. K., and A. Miyamoto Angiotensin II interacts with prostaglandin F2alpha and endothelin 1 as a local luteolytic factor in the bovine corpus luteum in vitro. Biol. Reprod. 60:1104 1109. Henricks, D. M., J. T. Long, J. R. Hill, and J. F. Dickey. 1974. The effect of prostagla Reprod. Fertil. 41:113 120. Herath, S., S. T. Lilly, N. R. Santos, R. O. Gilbert, L. Goetze, C. E. Bryant, J. O. White, J. Cronin, and I. M. Sheldon. 2009. Expression of genes associ ated with immunity in the endometrium of cattle with disparate postpartum uterine disease and fertility. Reprod. Biol. Endocrinol. 7:55 68.

PAGE 278

278 Herbison, A. E., and J. R. Pape. 2001. New evidence for estrogen receptors in gonadotropin releasing hormone neuron s. Front Neuroendocrinol. 22:292 308. Hill, J., and R. Gilbert. 2008. Reduced quality of bovine embryos cultured in media conditioned by exposure to an inflamed endometrium. Aust. Vet. J. 86: 312 316. Hixon, J. E., and W. Hansel. 1974. Evidence for prefe rential transfer of prostaglandin F2 Reprod. 11:543 552. Hoelker, M., D. Salilew Wondim, M. Drillich, G. Christine, N. Ghanem, L. Goetze, D. Tesfaye, K. Schellander, and W. Heuw ieser. 2012. Transcriptional response of the bovine endometrium and embryo to endometrial polymorphonuclear neutrophil infiltration as an indicator of subclinical inflammation of the uterine environment. Reprod. Fert. Dev. 24:778 793. Hoffmann, B., and G Schuler. 2002. The bovine placenta; a source and target of steroid hormones: observations during the second half of gestation. Dom. An. End. 23:309 320. Hojo, T., M. O. Al 2009. Changes in the vascul ature of bovine corpus luteum during the estrous cycle and prostaglandin F2alpha induced luteolysis. J. Reprod. Dev. 55:512 517. Huszenicza, G., M. Fodor, M. Gacs, M. Kulcsar, M. J. W. Dohmen, M. Vamos, L. Porkolab, T. Kegl, J. Bartyik, J. A. C. M. Lohuis S. Janosi, and G. Szita. 1999. Uterine bacteriology, resumption of cyclic ovarian activity and fertility in postpartum cows kept in large scale dairy herds. Reprod. Dom. Anim. 34, 237 245. Ireland, J. J., G. W. Smith, D. Scheetz, F. Jimenez Krassel, J. K. Folger, J.L.H. Ireland, F. Mossa, P. Lonergan, and A.C.O. Evans. 2011. Does size matter in females? An overview of the impact of the high variation in the ovarian reserve on ovarian function and fertility, utility of anti Mullerian hormone as a diagnost ic marker for fertility and causes of variation in the ovarian reserve in cattle. Reprod. Fertil. Dev. 23:1 14 Jaiswal, R.S., J. Singh, and G.P. Adams. 2004. Developmental pattern of small antral follicles in the bovine ovary. Biol. Reprod. 71:1244 1251. Jiemtaweeboon, S., K. Shirasuna, A. Nitta, A. Kobayashi, H. J. Schuberth, T. Shimizu, A. Miyamoto. 2011. Evidence that polymorphonuclear neutrophils infiltrate into the developing corpus luteum and promote angiogenesis with interleukin 8 in the cow. Repr od. Biol. Endocrinol.9:79.

PAGE 279

279 Jooston, I., P. Van Eldik, and L. Elving. 1987. Factors related to the etiology of retained placenta in dairy cattle. Anim. Reprod. Sci. 14:251 262. Jooston, I., M. F. Sanders, and E. J. Hensesn. 1991. Involvement of major histo compatibility complex class I compatibility between dam and calf in the aetiology of bovine retained placenta. Anim Genet. 22:455 463 Jost, B. H, and S. J. Billington. 2005. Arcanobacterium pyogenes: molecular pathogenesis of an animal opportunist. Anton ie Van Leeuwenhoek 88:87 102. Juengel, J. L., and K. P. McNatty. 2005. The role of proteins of the transforming growth factor beta superfamily in the intraovarian regulation of follicular developme nt. Hum. Reprod. Update 11, 143 160. Juengel, J. L., H. A. Garverick, A. L. Johnson, R. S. Youngquist and M. F. Smith. 1993. Apoptosis during luteal regression in cattle. Endocrinology 132:249 54. Julien, W.E., and H. R. Conrad. 1976. Selenium and vitamin E and incidence of retained placenta in parturient d airy cows. J. Dairy Sci. 59:1954 1959. Kaneko, K., and S. Kawakami. 2008. Influence of experimental intrauterine infusion of Arcanobacterium pyogenes solution on ovarian activity in cycling cows. J. Vet. Med. Sci. 70, 77 83. Kaneko, K., and S. Kawakami. 2009. The roles of PGF (2alpha) and PGE (2) in regression of the corpus luteum after intrauterine infusion of Arcanobacterium pyogenes in cows. Theriogenology 71: 858 863. Kaneko, K., S. Kawakami, M. Miyoshi, T. Abukawa, S. Yamanaka, M. Mochizuki, S. Yosh ihara. 1997. Effect of retained placenta on subsequent bacteriological and cytological intrauterine environment and reproduction in Holstein dairy cows. Theriogenology 48:617 24. Kaneko, K., M. Nakamura, and R. Sato. 2013. Influence of Trueperella pyogene s in uterus on corpus luteum lifespan in cycling cows. Theriogenology 79:803 808. Kankofer, M., J. W ierciski, and H. Z erbe. 2002. Prostagla ndin E2 9 keto reduct ase activity in bovine retained and not retained placenta. Prostaglandins, Leukotrienes and Ess ential Fatty Acids 66: 413 417 Kankofer, M., J. Wierciski, and M. Fidecki. 1998. Activity of hyaluronidase in placental tissues from cows with and without retained fetal membranes. Zentralbl. Veterinrmed. A 45:337 341.

PAGE 280

280 Kasimanickam, R., M. L. Day, J. S Rudolph, J. B. Hal, and W. D. Whittier. 2009. Two doses of prostaglandin improve pregnancy rates to timed AI in a 5 day progesterone based synchronization protocol in beef cows. Theriogenology 71:762 767. Kasimanickam, R., T. F. Duffield, R. A. Foster, C J. Gartley,K. E. Leslie, J. S. Walton, W. H. Johnson. 2004. Endometrial cytology and ultrasonography for the detection of subclinical endometritis in postpartum dairy cows. Theriogenology 62:9 23. Kasimanickam, R., T. F. Duffield, R. A. Foster, C. J. G artley, K. E. Leslie, J. S. Walton, and W. H. Johnson. 2005. The effect of a single administration of cephapirin or cloprostenol on the reproductive performance of dairy cows with subclinical endometritis. Theriogenology 63:818 830. Kastelic, J. P., R. B. Cook, G. H. Coulter. 1997. Scrotal/testicular thermoregulation and the effects of increased testicular temperature in the bull. Vet Clin N orth Am Food Anim Pract. 13:271 82. Kato, S., A. Shiratsuchi, K. Nagaosa, Y. Nakanishi. 2005 Phosphatidylserine and integrin mediated phagocytosis of apoptotic luteal cells by macrophages of the rat. Dev. Growth. Differ. 47:153 61. Katzung, B. G. 2007. Basic and Clinical Pharmacology, 10th edition. New York, NY: McGraw Hill Medical p. 733. Kaufmann, T. B., S. Wester mann, M. Drillich, J. Plntzke, and W. Heuwieser. 2010. Systemic antibiotic treatment of clinical endometritis in dairy cows with ceftiofur or two doses of cloprostenol in a 14 d interval. An. Rep. Sci.121:55 62. Kelton, D. F., K. E. Leslie, W. G. Etherin gton, B. N. Bonnett, and J. S. Walton. 1991. Accuracy of rectal palpation and of a rapid milk progesterone enzyme immunoassay for determining the presence of a functional corpus luteum in subestrous dairy cows. Can. Vet. J. 32:286 291. Khoury, E. L., and L. A. Marshall. 1990. Luteinization of human granulosa cells in vivo is associated with expression of MHC class II antigens. Cell Tissue Res. 262:217 224. King, S. S., E. M. Carnevale, L. G. Nequin, and J. J. Crawford. 1998. Inhibition of bacterial endome tritis with mannose. J. Equine Vet. Sci. 18:332 334. Knight, P. G., and C. Glister. 2001. Potential local regulatory functions of inhibins, activins and follistatin in the ovary. Reproduction 121:503 512. Knowlson, J. C. 1834. Yorkshire Cattle Doctor. Wi lliam Walker and Sons. pp 108 109.

PAGE 281

281 Kobayashi, S., B. Berisha, W. M. Amselgruber, D. Schams, A. Miyamoto. 2001. Production and localisation of angiotensin II in the bovine early corpus luteum: a possible interaction with luteal angiogenic factors and prost aglandin F2 alpha. J Endocrinol. 170:369 80. Komar, C. M., A. K. Berndtson, A. C. O. Evans, and J. E. Fortune. 2001. Decline in circulating estradiol during the periovulatory period is correlated with decreases in estradiol and androgen, and in messenger RNA for P450 aromatase and P450 hydroxylase, in bovine preovulatory follicles. Biol. Reprod. 64:1797 1805. Kossaibati, M. A., and R. J. Esslemont. 1997. The costs of production diseases in dairy herds in England. Vet. J. 154:41 51. Kuhn, M. T., J. L. Hutchison, and G. R. Wiggans. 2006. Characterization of Holstein heifers fertility in the United States. J. Dairy Sci. 89:4907 4920. parental antibiotic therapy for retained fetal membranes in cows. Indian J Anim Reprod 25:154 155. Laben, R. L., R. Shanks, P. J. Berger, and A. E. Freeman. 1982. Factors affecting milk yield and reproductive performance. J. Dairy Sci. 65:1004 1015. Langermann, S., S. Palaszynski, M. Barnhart, G. Auguste, J. S. Pinkner, J. Burlein, P. Barren, S. Koenig, S. Leath, C. H. Jones, and S. J. Hultgren. 1997. Prevention of mucosal Escherichia coli infection by FimH adhesin based systemic vaccination. Science 276:607 611. Laven, R. A. 2006. NADIS Cattl e Diseases. Retained foetal membranes. www.nadis.org.uk/DiseasesCattle/RFM/Retained%20Foetal%20Membranes.htm. Accessed April 26, 2013. Laven, R.A., and A. R. Peters. 1996. Bovine retained placenta: aetiology, pathogenesis and economic loss. Vet. Rec. 139: 465 471. Leblanc, S. 2010. Assessing the association of the level of milk production with reproductive performance in dairy cattle. J Reprod Dev. 56:S1 S7. LeBlanc, S. J, T. F. Duffield, K. E. Leslie, K. G. Bateman, G. P. Keefe, and J. S. Walton. 2002b. The effect of treatment of clinical endometritis on reproductive performance in dairy cows. J Dairy Sci. 85:2237 2249. LeBlanc, S. J. 2008. Postpartum uterine disease and dairy herd reproductive performance: A review. Vet. J. 176:102 114.

PAGE 282

282 LeBlanc, S. J., T. F. Duffield, K. E. Leslie, K. G. Bateman, G. P. Keefe, J. S. Walton, and W. H. Johnson. 2002a. Defining and diagnosing postpartum clinical endometritis and its impact on reproductive performance in dairy cows. J. Dairy Sci. 85:2223 2236. Lehtolainen, T., A. Shwimmer, N. Y. Shpigel, T. Honkanen Buzalski, and S. Pyorala. 2003. In vitro antimicrobial susceptibility of Escherichia coli isolates from clinical bovine mastitis in Finland and Israel. J. Dairy Sci. 86:3927 3932. Lei, Z. M., N. Chegini, C. V. R ao. 1991. Quantitative cell composition of human and bovine corpora lutea from various reproductive states. Biol Reprod 44:1148 56. Leung, S. T., Z. Cheng, E. L. Sheldrick, K. Derecka, A. P. Flint, D. C. Wathes. 2001. The effects of lipopolysaccharide an d interleukins 1alpha, 2 and 6 on oxytocin receptor expression and prostaglandin production in bovine endometrium. J. Endocrinol. 168:497 508. Lewis, G. S. 2004. Steroidal regulation of uterine immune defenses. Anim. Reprod. Sci. 82 83:281 294. Li, J. and R. M. Roberts. 1994. Interferon tau and interferon alpha interact with the same receptors in bovine endometrium. Use of a readily iodinatable form of recombinant interferon tau for binding studies. J Biol.Chem. 269:13544 13550. Lima, F. S., A. De Vri es, C. A. Risco, J. E. P. Santos and W. W. Thatcher. 2010. Economic comparison of natural service and timed artificial insemination breeding programs in dairy cattle. J. Dairy Sci. 93:4404 13. Lima, F. S., C. A. Risco, M J. Thatcher, M. E. Benzaquen, L. F Archbald, J. E. P. Santos, and W. W. Thatcher. 2009. Comparison of reproductive performance in lactating dairy cows bred by natural service or timed artificial insemination. J Dairy Sci. 92:5456 66. Lima, F. S., H. Ayres, M. G. Favoreto, R.S. Bisinotto, L.F. Greco, E.S. Ribeiro, P.S. Baruselli, C. A. Risco, W.W. Thatcher, and J.E.P. Santos. 2011. Effects of gonadotropin releasing hormone at initiation of the 5 d timed artificial insemination (AI) program and timing of induction of ovulation relative to A I on ovarian dynamics and fertility of dairy heifers. J. Dairy Sci. 94:4997 5004. Lima, J. R., F. A. Rivera, C. D. Narciso, R. Oliveira, R. C. Chebel, and J. E. P. Santos. 2009b. Effect of increasing amounts of supplemental progesterone in a timed artific ial insemination protocol on fertility of lactating dairy cows. J. Dairy Sci. 92:5436 5446 Lindell, J. O., and H. Kindahl. 1983. Exogenous prostaglandin F2 alpha promotes uterine involution in the cow. Acta Vet. Scand. 24:269 274.

PAGE 283

283 Liszewska, E., R. Rekaw iecki, and J. Kotwica. 2005. Effect of progesterone on the expression of bax and bcl 2 and on caspase activity in bovine luteal cells. Prostag. Lipid. Mediat. 78:67 81. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression data usi ng real time quantitative PCR and the 2 408. Lopes Jr., 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 outc omes of dairy heifers inseminated at induced estrus or at fixed time after a 5 day or 7 day progesterone insert based ovulation synchronization protocol. J. Dairy Sci. 96 :1612 1622 Lopez, H., L. D. Satter,and M. C. Wiltbank. 2004. Relationship between l evel of milk production and estrous behavior of lactating dairy cows. Anim. Reprod. Sci. 81:209 223. Lovendahl, P., and M. G. G. Chagunda. 2009. Short communication: Genetic variation in estrus activity traits. Journal of Dairy Science 92, 4683 4688. Luc y M. C. 2001. Reproductive loss in high producing dairy cattle: where will it end? J. Dairy Sci. 84:1277 93. of granulosa cells in small follicles of sheep. J. Reprod. F ertil. 115:251 262. Luo, W., A. Gmen, J. M. Haughian, and M. C. Wiltbank. 2011. The role of luteinizing hormone in regulating gene expression during selection of a dominant follicle in cattle. Biol. Reprod. 84:369 378. Lv, L., F. Jimenez Krassel, A. Se n, A. Bettegowda, M. Mondal, J. K. Folger, K. Lee, J.J. Ireland, and G. W. Smith. 2009. Evidence supporting a role for cocaine and amphetamine regulated transcript (CARTPT) in control of granulosa cell estradiol production associated with dominant follicl e selection in cattle. Biol. Reprod. 81:580 586. Machado, V. S., M. L. S. Bicalho, R. V. Pereira, L. S. Caixeta, J. H. J. Bittar, G. Oikonomou, R. O. Gilbert and R. C. Bicalho. 2012. The effect of intrauterine administration of mannose or bacteriophage on uterine health and fertility of dairy cows with special focus on Escherichia coli and Arcanobacte rium pyogenes. J. Dairy Sci. 95 :3100 3109. Maj, J. G., and M. Kankofer. 1997. Activity of 72 kDa and 92 kDa matrix metallo proteinases in placental tissue s of cows with and without retained foetal membranes.Placenta 18:683 687.

PAGE 284

284 Martinez N., C. A. Risco, F.S. Lima, Bisinotto, L. F. Greco, E. S. Ribeiro, F. Maunsell, K. N. Galvo, and J. E. P. Santos. 2012. Evaluation of peripartal calcium status, energetic profile and neutrophil function in dairy cows at low or high risk of d eveloping uterine disease. J. Dairy Sci. 95:7158 7172. Martinez, M. F., G. P. Adams, D. R. Bergfelt, J. P. Kastelic, and R. J. Mapletoft. 1999. Effect of LH or GnRH on the dominant follicle of the first follicular wave in beef heifers. Anim. Reprod. Sci. 57:23 33. Mateus L., L. Lopes da Costa, F. Bernardo, and J. Robalo Silva. 2002a. Influence of puerperal uterine infection on uterine invo lution and postpartum ovarian activity in dairy cows. Reprod. Dom. Anim. 37:31 35. Mateus L., L. Lopes da Costa, H. Carvalho, P. Serra, and J. Robalo Silva. 2002b. Blood and intrauterine leukocyte profile and function in dairy cows that spontaneously recovered from postpartum endometritis. Reprod. Dom. Anim. 37:176 180. Mazerbourg, S., C. A. Bondy, J. Zhou, J., and P. Monget, P. 2003. The insulinlike growth factor system: a key determinant role in the growth and selection of ovarian follicles? A comparative species study. Reprod. Domest. Anim. 38:247 258. Mazerbourg, S., M. T. Overgaard, C. Oxvig, M. Christiansen, C. A. Conover, I. Laurendeau, M. Vidaud, G. Tosser Klopp, J. Zapf, and P. Monget. 2001. Pregnancy associated plasma protein A (PAPP A) in ovine, bovine, porcine and equine ovarian follicles: involvement in IGF binding protein 4 proteolytic degradation and mR NA expression during follicular development. Endocrinology 142:5243 5253. McCracken J. A., E. E. Custer, and J.C. Lamsa. 1999. Luteolysis: A neuroendocrine mediated event. Physiol. Rev. 45:263 323. McDougall, S. 2001. Effect of intrauterine antibiotic t reatment on reproductive performance of dairy cows following periparturient disease. New Zeal. Vet. J. 49:150 158 McLaughlin, C. L., E. Stanisiewski, M. J. Lucas, C. P. Cornell, J. Watkins, L. Bryson, J. K. S. Tena, J. Hallberg, and J. R. Chenault. 2012. Evaluation of two doses of ceftiofur crystalline free acid sterile suspension for treatment of metritis in lactation dairy cows. J. Dairy Sci. 95:4363 4371. McLaughlin, C.L., E. Stanisiewski, C. A. Risco, J. E. P. Santos, G. E. Dahl, R. C. Chebel, C. Lagr ow, C. Daucherty, L. Bryson, D. Weisel. 2013. Evaluation of ceftiofur crystalline free acid sterile suspension for control of metritis in high risk lactating dairy cows. Theriogenology 79:725 734.

PAGE 285

285 McNatty, K. P., K. Reader, D.A. Heath, and J. L. Juengel 2007. Control of ovarian follicular development to the gonadotrophin dependent phase: a 2006 perspective. Pages 55 Nottingham University Press: Nottingham, UK. McNaughton, A. P., and R. D. Murray. 2009. S tructure and function of the bovine fetomaternal unit in relation to the causes of retained fetal membranes. Vet Rec. 165:615 622. McVey, W. R. Jr., and G. L. Williams. 1991. Mechanical masking of neurosensory pathways at the calf teat interface: Endocrin e, reproductive and lactational features on the suckled anestrous cow. Theriogenology 35:931 941. Meadows, C., P. J. Rajala Schultz, and G. S. Frazer. 2005. A spreadsheet based model demonstrating the nonuniform economic effects of varying reproductive pe rformance in Ohio dairy herds. J. Dairy Sci. 88:1244 1254. Mellieon Jr., H. I., S. L. Pulleya, G. C. Lambb, J. E. Larsonc, J. S. Stevenson. 2012. Evaluation of the 5 day versus a modified 7 day CIDR breeding program in dairy heifers. Theriogenology 78:199 7 2006 Mihm, M., A. Baguisi, M. P. Boland, and J. F. Roche. 1994. Association between the duration of dominance of the ovulatory follicle and pregnancy rates in beef heifers. J. Reprod. Fertil. 102:123 130. Mihm, M., P. J. Baker, J. L. Ireland, G. W. Sm ith, P. M. Coussens, A. C. Evans, and J. J. Ireland. 2006. Molecular evidence that growth of dominant follicles involves a reduction in follicle stimulating hormone dependence and an increase in luteinizing hormone dependence in cattle. Biol. Reprod. 74:1 051 1059. Mihn, M., and A.C. O. Evans. 2008. Mechanisms for dominant follicle selection in monovulatory species: a comparison of morphological, endocrine and intraovarian events in cows, mares and women. Reprod. Dom. Anim. 43:48 56. Miller, A. N. A. PhD Thesis. University of London; London, UK: 2009. The effect of Arcanobacterium pyogenes in the bovine uterus. Miller, A. N. A., E. J. Williams, K. Sibley, S. Herath, E. A. Lane, J.Fishwick, D. M. Nash, A. N. Rycroft, H. Dobson, C. E. Bryant, and I. M. Shel don. 2007. The effects of Arcanobacterium pyogenes on endometrial function in vitro, and on uterine and ovarian function in vivo. Theriogenology 68:972 980. Miller, J. K., and E. Brzezinska Slebodzinska. 1993. Oxidative stress, antioxidants, and animal fu nction. J Dairy Scie 76:2812 2823.

PAGE 286

286 Miller, R. H., H. D. Norman, M. T. Kuhn, J. S. Clay, and J. L. Hutchison. 2007. Voluntary waiting period and adoption of synchronized breeding in Dairy Herd Improvement herds. J. Dairy Sci. 90:1594 1606. Miyamoto, A., K. Okuda, F. J. Schweigert, D. Schams. 1992. Effects of basic fibroblast growth factor, transforming growth factor beta and nerve growth factor on the secretory function of the bovine corpus luteum in vitro. J Endocrinol 135:103 14. Miyamoto, A., K. Shira suna, T. Shimizu, H. Bollwein, and D. Schams. 2010. Regulation of corpus luteum development and maintenance: specific roles of angiogenesis and action of prostaglandin F2 Nottingham University Press. 67:289 304. M iyamoto, A., K. Shirasuna, T. Shimizu, K. Sasahara. 2009. Local regulation of corpus luteum development and regression in the cow: Impact of angiogenic and vasoactive factors. Dom. An. Endocrinol. 37:159 169. Miyamoto, A., K.Shirasuna, M. P. Wijayagunawar dane, S. Watanabe, M. Hayashi, D. Yamamoto, M. Matsui, snd T. J. Acosta. 2005. Blood flow: a key regulatory component of corpus luteum function in the cow. Domest. Anim. Endocrinol. 29:329 39. Miyamoto, A., S. Kobayashi, S. Arata, M. Ohtani, Y. Fukui, and D. Schams. 1997. Prostaglandin F2 alpha promotes the inhibitory action of endothelin 1 on the bovine luteal function in vitro. J. Endocrinol. 152:R7 11. Miyoshi, M., Y. Sawamukai, and T. Iwanaga. 2002. Reduced phagocytic activity of macrophages in the bo vine retained placenta. Reprod. Dom. Anim. 37:53 56. Molenda Figueira, H. A, C. A. Williams, A. L. Griffin, E. M. Rutledge, J. D. Blaustein and M. J. Tetel. 2006. Nuclear receptors coactivators function in estrogen receptor and progestin receptor depend ent aspects of sexual behavior in female rats. Hormones and Behavior 50:383 392. Moreira, F., C. Orlandi, C. A. Risco, R. Mattos, F. Lopes, and W. W. Thatcher. 2001. Effects of presynchronization and bovine somatotropin on pregnancy rates to a timed artif icial insemination protocol in lactating dairy cows. J. Dairy Sci. 84:1646 1659. Muller, L. D., and M. J. Owens. 1974. Factors associated with the incidence of retained placentas. J Dairy Sci 57:725 728. Muoz Planillo, R., L. Franchi, L. S. Miller, and G. Nunez. 2009. A critical role for hemolysins and bacterial lipoproteins in Staphylococcus aureus induced activation of the Nlrp3 inflammasome. J. Immunol. 183:3942 3948.

PAGE 287

287 National Animal Health Monitoring System (NAHMS). 2002. Part 1: Reference of dairy health and management in the United States. Ctr Epidemiol Anim Health, Fort Collins, Co. National Animal Health Monitoring System (NAHMS). 2009. Part IV: Reference of Dairy Cattle Health and Management Practices in the United States, 2007. #N494.0209. Cen ters for Epidemiology and Animal Health, USDA: APHIS: VS, Fort Collins, CO. National Research Council. 2001. Nutrient Requirements of Dairy Cattle. 7th rev. ed. Natl. Acad. Sci., Washington, DC. Navanukraw, C., D. A. Redmer, L. P. Reynolds, J. D. Kirsch, A. T. Grazul Bilska, and P. M. Fricke. 2004. A modified presynchronization protocol improves fertility to timed artificial insemination in lactating dairy cows. J. Dairy Sci. 87:1551 1557. Neuvians, T. P., D. Schams, B. Berisha, and M. W. Pfaffl. 2004. I nvolvement of pro inflammatory cytokines, mediators of inflammation, and basic fibroblast growth factor in prostaglandin F2alphainduced luteolysis in bovine corpus luteum. Biol. Reprod.70:473 80. Nimz, M., M. Spitschak, F. Schneider, R. Frbass, and J. V anselow. 2009. Down regulation of genes encoding steroidogenic enzymes and hormone receptors in late preovulatory follicles of the cow coincides with an accumulation of intrafollicular steroids. Domest. Anim. Endocrinol. 37:45 54 Niswender, G. D., J. L. Juengel, P. J. Silva, M. K. Rollyson, and E. W. McIntush. 2000. Mechanisms controlling the function and life span of the corpus luteum. Physiol. Rev. 80:1 29. Nitta, A., K. Shirasuna, S. Haneda, M. Matsui, T. Shimizu, S. Matsuyama, K. Kimura, H. Bollwein, A. Miyamoto A. 2011. Possible involvement of IFNT in lymphangiogenesis in the corpus luteum during the maternal recognition period in the cow. Reproduction 142:879 92. NOAA. 1996. National Oceanic and Atmospheric Administration. Pages C 31 36. Livestock hot weather stress. US Dept. Commerce, Natl. Weather Serv. Central Reg., Reg. Operations Manual Lett. Nogueira, M. F., J. Jr. Buratini, C. A. Price, A. C. Castilho, M. G. Pinto, C. M. Barros. 2007. Expression of LH receptor mRNA splice variants in bovine granulosa cells: changes with follicle size and regulation by FSH in vitro. Mol Reprod Dev 74:680 686.

PAGE 288

288 Norman, H. D., J. R. Wright, S. M. Hubbard, R. H. Miller, and J. L. Hutchison. 2009. Reproductive status of Holstein and Jersey cows in the United Stat es. J. Dairy Sci. 92:3517 3528. corpus luteum of the cow. J Reprod Fertil ;85:483 7. Okker, H., E. J. Schmitt, P. L. A. M. Vos, P. Scherpenisse, A. A. Bergwerff, and F H. Jonker. 2002. Pharmacokinetics of ceftiofur in plasma and uterine secretions and tissues after subcutaneous postpartum administration in lactating dairy cows. J. Vet. Pharmacol. Ther. 25:33 38. Okuda, K., A. Korzekwa, M. Shibaya, S. Murakami, R. Nish imura, M. Tsubouchi, I. Woclawek Potocka, and D.J. Skarzynski. 2004. Progesterone is a suppressor of apoptosis in bovine luteal cells. Biol. Reprod. 71:2065 2071. Okuda, K., and R. Sakumoto. 2003. Multiple roles of TNF super family members in corpus luteu m function. Reprod. Biol. Endocrinol. 95:1 10. Oltenacu, P. A., A. Frick, and B. Lindhe. 1990. Epidemiological study of several clinical diseases, reproductive performance and culling in primiparous Swedish cattle. Prev. Vet. Med. 9:59 74. Ott, T. L., C. A. Gifford. 2010. Effects of early conceptus signals on circulating immune cells: lessons from domestic ruminant Amer J. Reprod. Immunol. 64:245 254. Overton, M. W., and W. M. Sischo. 2005. Comparison of reproductive performance by artificial inseminatio n versus natural service sires in California dairies. Theriogenology 64:603 13. Pate, J. L. 1995: Involvement of immune cells in regulation of ovarian function. J Reprod. Fert. Supp.l 49:365 377. Pate, J. L., and P. L. Keyes. 2001. Immune cells in the co rpus luteum: friends or foes? Reproduction 122:665 676. Penny, L. A, D. Armstrong, T. A. Bramley, R. Webb, R. A. Collins, E. D. Watson. 1999. Immune cells and cytokine production in the bovine corpus luteum throughout the oestrous cycle and after induced luteolysis. J Reprod Fertil 115:87 96. Peter, A.T., H. Levine, M. Drost, and D.R. Bergfelt. 2009. Compilation of classical and contemporary terminology used to describe morphological aspects of ovarian dynamics in cattle. Theriogenology 71:1343 1357 Pet ers, A. R., and R. A. Laven 1996. Treatment of bovine retained placenta and its effects. Vet. Rec.139:539 541.

PAGE 289

289 Petroff, M., K. M. Coggeshall, L. S. Jones, J. L. Pate.1997. Bovine luteal cells elicit major histocompatibility complex class II dependent T cell proliferation. Biol. Reprod. 57:887 93. Pru, J. K., M. P. Lynch, J. S. Davis, B. R. Rueda. 2003. Signaling mechanisms in tumor necrosis factor alpha induced death of microvascular endothelial cells of the corpus luteum. Reprod. Biol. Endocrinol. 1:1 7. Pryce J. E., S Bolormaa, A. J. Chamberlain, P. J. Bowman, K. Savin, M. E. Goddard, B. J. Hayes. 2010. A validated genome wide association study in 2 dairy cattle breeds for milk production and fertility traits using variable length haplotypes. J. Dairy Sci. 93:3331 3345. Pursley J.R., M.C. Wiltbank, J.S. Stevenson, J.S. Ottobre, H.A. Garverick, and L.L. Anderson. 1997. Pregnancy rates per artificial insemination for cows and heifers inseminated at a synchronized ovulation or synchronized estrus. J. Dai ry Sci. 80:295 300. Rabaglino, M.B, C.A. Risco, M. J. Thatcher, F. Lima, J.E.P. Santos, and W.W. Thatcher. 2010b. Use of a five day progesterone based timed AI protocol to determine if flunixin meglumine improves pregnancy per timed AI in dairy heifers. T heriogenology 73:1311 1318. Rabaglino, M.B., C.A. Risco, M. J. Thatcher, I.H. Kim, J.E.P. Santos, and W. W. day Co Synch + CIDR protocol for estrous synchronization and resynchronization of dairy heifers. J. Dairy Sci. 93:1050 1058. Raetz, C. R., and C. Whitfield. Lipopolysaccharide endotoxins. Annu. Rev. Biochem. 2002; 71, 635 700. Rahe, C.H., R.E. Owens, J.L. Fleeger, H.J. Newton, P.G. Harms. 1980. Pattern of plasma luteinizing hormone in the cyclic cow : dependence upon the period of the cycle. Endocrinology 107:498 503. Rajala, P. J, and Y. T. Grohn. 1998. Effects of dystocia, retained placenta, and metritis on milk yield in dairy cows. J. Dairy Sci. 81:3172 3181. Reibiger, I., K. Spanel Borowski. 200 0. Difference in localization of eosinophils and mast cells in the bovine ovary. J. Reprod. Fertil. 118:243 9. Rekawiecki, R., M. K. Kowalik, D. Slonina, and J. Kotwica. 2008. Regulation of progesterone synthesis and action in bovine corpus luteum. J. Phy siol. Pharmacol. 59:75 89.

PAGE 290

290 Ribeiro, E. S., A. P. A. Monteiro, F. S. Lima, H. Ayres, R. S. Bisinotto, M. Favoreto, L. F. Greco, R. S. Marsola, W. W. Thatcher, and J. E. P. Santos. 2012 c Effects of presynchronization and length of proestrus on fertility of grazing dairy cows subjected to a 5 day timed artificial insemination protocol. J. Dairy Sci. 95:2513 2522. Ribeiro, E. S., R. S. Bisinotto, M. G. Favoreto, L.T. Martins, R. L. A. Cerri, F. T. Silvestre, L. F. Greco, W. W. Thatcher, and J. E. P. Santos. 2012b. Fertility in dairy cows following presynchronization and administering twice the luteolytic dose of prostaglandin F2 day timed artificial insemination protocol. Theriogenology 78:273 284. Ribeiro, E. S., K. N. Gal vo, W. W. Thatcher, and J. E. P. Santos. 2012a. Economic aspects of applying reproductive technologies to dairy herds. Anim. Reprod. 9:370 387. Risco, C. A, F. Moreira, M. DeLorenzo, and W. W. Thatcher. 1998. Timed artificial insemination in dairy cattle Part II. Comp. Cont. Educ. Pract. Vet. Food. Anim. 20:1284 1289. Risco, C. A., and J. Hernandez. 2003. Comparison of ceftiofur hydrochloride and estradiol cypionate for metritis prevention in dairy cows affected with retained fetal membranes. Theriogen ology 60: 47 58. Rivera, G. M., Y. A. Chandrasekher, A. C. Evans, L. C. Giudice, and J. E. Fortune. 2001. A potential role for insulin like growth factor binding protein 4 proteolysis in the establishment of ovarian follicular dominance in cattle. Biol. Reprod. 65:102 111. Rivera, G. M., and J. E. Fortune. 2003. Selection of the dominant follicle and insulin like growth factor (IGF) binding proteins: evidence that pregnancy associated plasma protein A contributes to proteolysis of IGF binding protein 5 in bovine follicular fluid. Endocrinology 144:437 446. Rivera, G. M., and J. E. Fortune. 2001. Development of co dominant follicles in cattle is associated with a follicle stimulating hormonedependent insulin like growth factor binding protein 4 protease Biol. Reprod. 65:112 118 Rivera, H., H. Lopez, and P.M. Fricke. 2004. Fertility of Holstein dairy heifers after synchronization of ovulation and timed AI after removed tail chalk. J. Dairy Sci. 87:2051 2061. Roa, J., and M. Tena Sampere. 2007. KiSS 1 s ystem and reproduction: comparative aspects and roles in the control of female gonadotrpic axis in mammals. 2007. Ge. Comp. Endocrinol. 152:132 140.

PAGE 291

291 Roberts, S. J. Veterinary Obstetrics and Genital Diseases. Woodstock, VT. 373 393. Robinson, R. S, A.J. H ammond, G. E. Mann, M. G. Hunter. 2008. A novel physiological culture system that mimics luteal angiogenesis. Reproduction 135:405 13 Roche, J. F., 1996. Control and regulation of folliculogenesis a symposium i n perspective. Rev. Reprod. 1:9 27. Roelofs a, J., F. Lpez Gatius, R. H. F. Hunterd, F. J. C. M. van Eerdenburge, and Ch. Hanzen. 2010. When is a cow in estrus? Clinical and practical aspects. Theriogenology 74:327 44. Rowson, L. E., R. Tervit, and A. Brand. 1972. The use of prostaglandins for syn chronization of oestrus in cattle. J. Reprod. Fertil. 29:145. Royal M.D., R.F. Smith, and N.C. Friggens. 2008. Fertility in dairy cows: bridging the gaps. Animal 2:1101 1103. Ruder, C. A., R. G. Sasser, R. J. Williams, J. K. Ely, R. C. Bull, and J. E. Bu tler. 1981. Uterine infection in postpartum cows II possible synergistic effect of fusobacterium necrophorum and Corynebacterium pyogenes. Theriogenology 15:573 580. Rutigliano, H. M., F. S. Lima, R. L. A. Cerri, L.F. Greco, J. M. Villela, V. Magalhes, F T. Silvestre, W. W. Thatcher, and J. E. P. Santos. 2008. Effects of method of presynchronization and source of selenium on uterine health and reproduction in dairy cows. J. Dairy Sci. 91:3323 3336. Saacke, R.G. 2008. Insemination factors related to time d AI in cattle. Theriogenology 70:479 484. Sako, D., X. J. Chang, K. M. Barone, G. Vachino, H. M. White, G. Shaw, G. M. Veldman, K. M. Bean, T. J. Ahern, B. Furie, et al. 2003. Expression cloning of a functional glycoprotein ligand for P selectin. Cell 75 : 1179 86. Sakumoto, R., M. Vermehren, R. A. Kenngott, K. Okuda, and F. Sinowatz. 2011.Localization of gene and protein expressions of tumor necrosis factor {alpha} (TNF), and TNF receptor types I and II in the bovine corpus luteum during the estrous cycl e. J. Anim. Sci. 89:3040 3047. Sangsritavong, S., D. K. Combs, R. Sartori, and M. C. Wiltbank. 2002. High feed intake increases blood flow and metabolism of progesterone and estradiol cattle. J. Dairy Sci. 85:2831 2842.

PAGE 292

292 Santos, J. E. P., C. D. Narciso, F. Rivera, W. W. Thatcher, and R. C. Chebel. 2010a. Effect of reducing the period of follicle dominance in a timed AI protocol on reproduction of dairy cows. J. Dairy Sci. 93:2976 2988. Santos, J. E. P., H. M. Rutigliano, and M. F. S Filho. 2009. Risk factors for resumption of postpartum cyclicity and embryonic survival in lactating dairy cows. Anim Reprod Sci. 110:207 21. Santos, J. E. P., R. S. Bisinotto, E. S. Ribeiro, F. S. Lima, L. F. Greco, C. R. Staples, and W. W. Thatcher. 2010b. App lying nutrition and physiology to improve reproduction in dairy cattle. Soc. Reprod. Fertil. Suppl. 67:387 403. Santos, N R ., G. C. Lamb, D. R. Brown, R. O. Gilbert. 2009. Postpartum endometrial cytology in beef cows. Theriogenology 71:739 745. Santos, T. M. A, L. S. Caixeta, V. S. Machado, A. K. Rauf, R. O. Gilbert, R. C. Bicalho. 2010c. Antimicrobial resistance and presence of virulence factor genes in Arcanobacterium pyogenes isolated from the uterus of postpartum dairy cows. Vet. Microbiol. 145: 84 8 9. Santos, T. M. A, R. O. Gilbert, L. S. Caixeta, V. S. Machado, L. M. Teixeira, R. C. Bicalho. 2010d. Susceptibility of Escherichia coli isolated from uteri of postpartum dairy cows to antibiotic and environmental bacteriophages. Part II: In vitro antimi crobial activity evaluation of a bacteriophage cocktail and several antibiotics. J. Dairy Sci. 93:105 114. Santos, T. M. A., R. C. Bicalho. 2012. Diversity and Succession of Bacterial Communities in the Uterine Fluid of Postpartum Metritic, Endometritic a nd Healthy Dairy Cows. PLoS ONE 12:e53048. Sartori, R., J. M. Haughian, R. D. Shaver, G. J. Rosa, and M. C. Wiltbank. 2004. Comparison of ovarian function and circulating steroids in estrous cycles of Holstein heifers and lactating cows. J. Dairy Sci. 87: 905 920. Sartori, R., P. M. Fricke, J. C. Ferreira, O. J. Ginther, and M. C. Wiltbank. 2001. Follicular deviation and acquisition of ovulatory capacity in bovine follicles. Biol. Reprod. 65:1403 1409. Sartori, R., P. M. Fricke, J. C. P. Ferreira, O. J. G inther, M. C. Wiltbank. 2001. Follicular deviation and acquisition of ovulatory capacity in bovine follicles. Biol Reprod. 65:1403 1409. Savio, J.D., M.P. Boland, and J.F. Roche. 1990. Development of dominant follicles and length of ovarian cycles in post partum dairy cows. J. Reprod. Fertil. 88:581 591.

PAGE 293

293 Schams, D., W. Amselgruber, R. Einspanier, F. Sinowatz, D. Gospodarowicz. 1994. Localization and tissue concentration of basic fibroblast growth factor in the bovine corpus luteum. Endocrine 2:907 12. Sc hmitt, E. J., J. F. Boucher, C. A. Van den Eede, J. R. Chenault, S. N. Van Oye, and V. J. Maes. 2001. Comparison of clinical efficacy of ceftiofur and oxytetracylcine for treatment of acute puerperal metritis in dairy cows. Proceedings Amer. Assoc. Bov. P rac. 34, 199 Schmitt, E.J, P.T. Diaz, M. Drost, and W.W. Thatcher. 1996. Use of a gonadotropin releasing hormone agonist or human chorionic gonadotropin for timed insemination in cattle. J. Anim. Sci. 74:1084 1091. Senger, P. L. 1994. The estrus detecti on problem: new concepts, technologies, and possibilities. J Dairy Sci. 77:2745 53. Senior, J., R. Sangha, G. S. Baxter, K. Marshall, and J. K. Clayton. 1992. In vitro characterization of prostanoid FP DP IP and TP receptors on the non pregnant human myometrium. Br. J. Pharmacol. 107:215 221. Sheldon, I. M., A. N. Rycroft, B. Dogan, M. Craven, J. J. Bromfield, A. Chandler, M. H. Roberts, S. B. Price, R. O. Gilbert, and K. W. Simpson. 2010. Specific strains of Escherichia coli are pathogenic for the e ndometrium of cattle and cause pelvic inflammatory disease in cattle and mice. PLoS ONE 5:e9192. Sheldon, I. M., D. E. Noakes, A. N. Rycroft, D. U. Pfeiffer, and H. Dobson. 2002. Influence of uterine bacterial contamination after parturition on ovarian do minant follicle selection and follicle growth and function in cattle. Reprod. Cambridge England 123:837 845. Sheldon, I. M., G. S. Lewis, S. LeBlanc, and R. O. Gilbert. 2006. Defining postpartum uterine disease in cattle. Theriogenology 65:1516 1530. She ldon, I. M., J. Cronin, L. Goetze, G. Donofrio, H J. Schubert. 2009. Review. Defining postpartum uterine disease and the mechanisms of infection and immunity in the female reproductive tract in cattle. Reprod Domest Anim. 44:1 9. Sheldon, I. M., M. Bushne ll, J. Montgomery, and A. N. Rycroft. 2004. Minimum inhibitory concentrations of some antimicrobial drugs against bacteria causing uterine infections in cattle. Vet. Rec. 155:383 387. Sheldon, I.M., and D. E. Noakes,. 1998. Comparison of three treatments for bovine endometritis. Vet. Rec. 142, 575 579.

PAGE 294

294 Shimasaki, S., R. K. Moore, F. Otsuka, and G. F. Erickson. 2004. The bone morphogenetic protein system in mammalian reproduction. Endocr. Rev. 25:72 101. Shirasuna, K., D. Yamamoto, K. Morota, T. Shimizu, M. Matsui, A. Miyamoto. Prostaglandin F 2 alpha stimulates endothelial nitric oxide synthase depending on the existence of bovine granulosa cells: analysis by co culture system of endothelial cells, smooth muscle cells and granulosa cells. Reprod. Domest. Anim. 43:592 59 8. Shirasuna, K., S. Jiemtaweeboon, S. Raddatz, A. Nitta, H. J. Schuberth, H. Bollwein, T. Shimizu, and A. Miyamoto. A. 2012. Rapid accumulation of polymorphonuclear neutrophils in the corpus luteum during prostaglandin F(2 ) induced lute olysis in the cow. PLoS ONE 7:e29054. Silva E., M. Gaivao, S. Leitao, B. H. Jost, C. Carneiro, C. L. Vilela, da Costa L Mateus L. Genomic characterization of Arcanobacterium pyogenes isolates recovered from the uterus of dairy cows with normal puerperiu m or clinical metritis. Veterinary Microbiology 2008;132:111 118. Silva, E., S. Leitao, T. Tenreiro, C. Pomba, T. Nunes, L. Lopes da Costa, and L. Mateus. 2009. Genomic and phenotypic characterization of Escherichia coli isolates recovered from the uterus of puerperal dairy cows. J. Dairy Sci. 92:6000 6010. Silva, W. J., G. D. Niswender. 1986. Maintenance of the corpus luteum of early pregnancy in the ewe. IV. Changes in luteal sensitivity to prostaglandin F2a throughout early pregnancy. J. Anim Sci. 63:1 201 1207. Sirois, J., and J. E. Fortune. 1988. Ovarian follicular dynamics during the estrous cycle in heifers monitored by real time ultrasonography. Biol. Reprod. 39:308 317. Skarzynski, D. J., M. M. Bah, K. M. Deptula, I. Woclawek Potocka, A. Korzekwa M. Shibaya, W. Pilawski, K. Okuda. 2003b. Roles of tumor necrosis factor a of the estrous cycle in cattle: an in vivo Study. Biol. Reprod. 69:1907 1913. Skarzynski, D. J., J. J. Jaroszewski, M .M. Bah, K. M. Deptula, B. Barszczewska, B. Gawronska, W. Ha nsel. 2003. Administration of a nitric oxide synthase inhibitor counteracts prostaglandin F2 induced luteolysis in cattle. Biol. Reprod. 68:1674 81 Skarzynski, D. J., S. Kobayashi, K. Okuda. 2001 Influence of nitric oxide and noradrenaline on prostagla ndin F(2)(alpha) induced oxytocin secretion and intracellular calcium mobilization in cultured bovine luteal cells. Biol. Reprod. 63:1000 1005.

PAGE 295

295 Skarzynski, D. J., A. Miyamoto, and K. Okuda. 2000. Production of prostaglandin F2 alpha by cultured bovine en dometrial cells in response to tumor necrosis factor alpha: cell type specificity and intracellular mechanisms. Biol. Reprod. 62:1116 1120. Skarzynski, D. J., K. K. Piotrowska, M. M. Bah, A. Korzekwa, I. Woclawek Potocka, K. Sawai, and K. Okuda. 2009. Ef fects of Exogenous Tumour Necrosis Factor a on the Secretory Function of the Bovine Reproductive Tract Depend on Tumour Necrosis Factor a Concentrations. Reprod. Dom. Anim. 44:371 379. Skinner, D. C., A. Caraty, R. Allingham. 2001. Unmasking the progester one receptor in the preoptic area and hypothalamus of the ewe: no colocalization with gonadotropin releasing neurons. Endocrinology 142:573 579. Smith, B. I., G. A. Donovan, C. A. Risco, R. Littell, C. Young, L. H. Stanker, and J. Elliott. 1998. Compariso n of various antibiotic treatments for cows diagnosed with toxic puerperal metritis. J. Dairy Sci. 81:1555 1562. Smith, G. W., A. Sen, J. K. Folger, and J. J. Ireland. 2010. Putative role of cocaine and amphetamine regulated transcript (CARTPT) in domina nt follicle selection in cattle. Reproduction in domestic ruminants VII. Nottingham University Press. 67:105 118. Smith, J. W., L. O. Ely, W. D. Gilson and W. M. Groves. 2004. Effects of artificial insemination vs. natural service breeding on production a nd reproduction parameters in dairy herds. The Prof. Anim. Sci. 20:185 90. Soto, P., P. R. Natzke, and P. J. Hansen. 2003. Actions of tumor necrosis factor oocyte maturation and embryonic development in cattle. Am. J. Rep. Immunol. 50:380 388. Souza, A. H., H. Ayres, R. M. Ferreira, and M. C. Wiltbank. 2008. A new presynchr onization system (Double Ovsynch) increases fertility at first postpartum timed AI in lactating dairy cows. Theriogenology 70:208 215. Spencer, T. E., and F. W. Bazer. 1995 Temporal and spatial alterations in uterine estrogen receptor and progesterone rec eptor gene expression during the estrous cycle and early pregnancy in the ewe. Biol. Reprod. 53:1527 1543. Spencer, T. E., C. A. Gray, T. L. Ott, G. A. Johnson, W. S. Ramsey,and F. W. Bazer. 1999. Differential effects of intrauterine and subcutaneous admi nistration of recombinant ovine interferon tau (roIFNs) on endometrial gene expression of cyclic ewes. Biol. Reprod. 61:464 470. Spiegelman, D., and E. Hertzmark E. 2005. Easy SAS calculations for risk or prevalence ratios and differences. Am. J. Epidemio l. 162:199 200.

PAGE 296

296 Staats, D. A., D. P. Lohr, and H. D. Colby. 1988. Effects of tocopherol depletion on the regional differences in adrenal microsomal lipid peroxidation and steroid metabolism. Endocrinol. 123:975 980. Stabenfeldt, G. H., L. L. Ewing, and L E. McDonald. 1969. Peripheral plasma progesterone levels during the bovine oestrus cycle. J. Reprod. Fertil. 19:433 442. Steffan, J., Agric, M., Adriamanga, S., Thibier, M., 1984. Treatment of metritis with antibiotics or prostaglandin F2a and influence of ovarian cyclicity in dairy cows. Am. J. Vet. Res. 45:1090 1094. Stevens, R. D., and R. P. Dinsmore. 1997. Treatment of dairy cows at parturition with prostaglandin F2 a or oxytocin for prevention of retained fetal membranes. J. Am. Vet. Med. Assoc. 21 :1280 1284. Stevenson, J. S. 2011. Alternative programs to presynchronize estrous cycles in dairy cattle before a timed artificial insemination program. J. Dairy Sci. 94:205 217. Stevenson, J. S., J. C. Dalton, J. E. P. Santos, R. Sartori, A. Ahmadzadeh, and R. C. Chebel. 2008. Effect of synchronization protocols on follicular development and estradiol and progesterone concentrations of dairy heifers. J. Dairy Sci. 91:3045 3056. Stevenson, J. S., J. F. Smith, D. E. Hawkins. 2000 .Reproductive outcomes for dairy heifers treated with combinations of prostaglandin F2alpha, norgestomet, and gonadotropin releasing hormone. J. Dairy Sci. 83:2008 2015. Stewart, A. B., E. K. Inskeep, E. C. Townsend, R. A. Dailey. 2003. Effects of gram positive bacterial pathogens in ewes: peptidoglycan as a potential mediator of interruption of early pregnancy. Reproduction 125:295 299. Sunderland, S.J., M. A. Crowe, M. P. Boland, J. F. Roche, J. J. Ireland. 1994. Selection, dominance and atresia of follicles during the oestrous cycle of heifers. J. Reprod. Fertil. 101:547 555. Takagi, M., S. Fujimoto, M. Ohtani, A. Miyamoto, M. P. B. Wijagunawardane T. J. Acosta, K. Miyakawa, and K Sato. 2002. Bovine retained placenta: hormonal concentrations in fetal and maternal placenta. Pl acenta 23:429 437. Takagi, M., D. Yamamoto, S. Ogawa, T. Otoi, M. Ohtani, and A. Miyamoto. 2008. Messenger RNA expression of angiotensin converting enzyme, endothelin, cyclooxygenase 2 and prostaglandin synthases in bovine placentomes during gestation an d the postpartum period. Vet. J. 177:398 404.

PAGE 297

297 Taniguchi, H., Y. Yokomizo, and K. Okuda. 2002. Fas Fas ligand system mediates luteal cell death in bovine corpus luteum. Biol. Reprod. 66:754 759. Terblanche, H. M., L. J. Kritzinger, and J. S. Van Heerden. 1976. Induced parturition in cattle. 1. Clinical studies. J. S. Afr. Vet. Assoc. 47:113 115. Thatcher, W. W., A. Guzeloglu, R. Mattos, M. Binelli, T. R. Hansen, and J. K. Pru. 2001. Uterine conceptus interactions and reproductive failure in cattle. Theri ogenology 56:1435 1450. Thurmond, M.C., C. M. Jameson, J. P. Picanso. 1993. Effect of intrauterine antimicrobial treatment in reducing calving to conception interval in cows with endometritis. J. Am. Vet. Med. Assoc. 203:1576 1578. Timmerman, C. P., E. Mattson, L. Martinez Martinez, L. De Graaf, J. A. G. Van Stripj, H. Verbrugh, J. Verhoef, A. Fleer. 1993. Induction of release of tumor necrosis fator from human monocytes by staphylococci and staphylococcal peptidoglycans. Infec. Immun. 61: 4167 4172. To wnson, D. H., A. R. Liptak. 2003. Chemokines in the corpus luteum: implications of leukocyte chemotaxis. Reprod. Biol. Endocrinol. 1:94. Tsai, S. J., and M. C. Wiltbank. 1998. Prostaglandin F2alpha regulates distinct physiological changes in early and mid cycle bovine corpora lutea. Biol. Reprod. 58:346 352. Turner, E. C., J. Hughes, H. Wilson, M. Clay, K. J. Mylonas, T. Kipari, W. C. Duncan, H. M. Fraser. 2011. Conditional ablation of macrophages disrupts ovarian vasculature. Reproduction 141:821 31. Ul ug, U., S. Goldman, I. Ben Shlomo, and E. Shalev. 2001. Matrix metalloproteinase (MMP) 2 and MMP 9 and their inhibitor, TIMP 1, in human term decidua and fetal membranes: The effect of prostaglandin F(2alpha) and indomethacin. Mol. Hum. Reprod. 7:1187 1193 Van Werven, T., Schukken, Y.H., Lloyd, J., 1992. The effects of duration of retained placenta on reproduction, milk production postpartum disease and culling rate. Theriogenology 37:1191 1203. Vandesompele, J., K. De Preter, F. Pattyn, B. Poppe, N. V. Roy, A. De Paepe, and F. Speleman. Accurate normalization of real time quantitative RT PCR data by geometric averaging of multiple internal control genes. Genome Biology 2002; 3:7.

PAGE 298

298 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 national genetic evaluation for cow fertility. J. Dairy Sci. 87:2285 2292. Vasconcelos, J. L. M., R. W. Silcox, G. J. M. Rosa, J. R. Pursley, and M. C. Wiltbank. 1999. Synchronization rate, size of th e ovulatory follicle, and pregnancy rate after synchronization of ovulation beginning on different days of the estrous cycle in lactating dairy cows. Theriogenology 52:1067 1078. von Krueger, X. P. Scherpeniss S. Roiger, and W. Heuwieser. 2013. Determin ation of ceftiofur derivatives in serum, endometrial tissue, and lochia in puerperal dairy cows with fever or acute puerperal metritis after subcutaneous administration of ceftiofur crystalline free acid. J. Dairy Sci. 96 :1054 1062. Voss, A. K., and J. E Fortune. 1993a. Levels of messenger ribonucleic acid for cytochrome P450 17 alpha hydroxylase and P450 aromatase in preovulatory bovine follicles decrease after the luteinizing hormone surge. Endocrinology 132:2239 2245. Voss, A. K., and J. E. Fortune. 1993b. Levels of messenger ribonucleic acid for cholesterol side chain cleavage cytochrome P 450 and 3 beta hydroxysteroid dehydrogenase in bovine preovulatory follicles decrease after the luteinizing hormone surge. Endocrinology 132:888 894. Walter, I., and A. Boos. 2001. Matrix metalloproteinases (MMP 2 and MMP 9) and tis sue inhibitor 2 of matrix metalloproteinases (TIMP 2) in the placenta and interplacental uterine wall in normal cows and in cattle with retention of fetal membranes. Placenta 22:473 4 83. Washburn, S. P., W. J. Silvia, C. H. Brown, B. T. McDaniel, and A. J. McAllister. 2002. Trends in reproductive performance in southeastern Holstein and Jersey DHI herds. J. Dairy Sci. 85:244 251. Wathes D. C., G. E. Mann, J. H. Payne, P. R. Riley, K. R. Stevenson, and G. E. Lamming. 1996. Regulation of oxytocin, oestradiol and progesterone receptor concentrations in different uterine regions by oestradiol, progesterone and oxytocin in ovariectomized ewes. J. Endocrinol. 151:375 393. Wathes, D. C. Z Cheng W. Chowdhury, M. A. Fenwick, R. Fitzpatrick, D. G. Morris, J. Patton,and J. J. Murphy. 2009. Negative energy balance alters global gene expression and immune responses in the uterus of postpartum dairy cows. Physiol. Genomics 39:1 13. Webb, R., and B. K. Campbell. 2007. Development of the dominant follicle: mechanisms of selection and maintenance of oocyte quality. Soc. Reprod. Fertil. Suppl. 64:141 163.

PAGE 299

299 Webb, R., P. C. Garnsworthy, J. G. Gong, and D. G. Armstrong. 2004. Control of follicular gr owth: local interactions and nutritional influences. J. Anim. Sci. 82 (E Suppl.):63 74. Weigel, K.A. 2006. Prospects for improving reproductive performance through genetic selection. Anim. Reprod. Sci. 96:323 330 Wettemann, R. P., H. D. Hafs, L. A. Edge rton, and L.V. Swanson. 1972. Estradiol and progesterone in blood serum during the bovine estrous cycle. J. Anim. Sci. 34:1020 1024. Williams, E. J, D. P. Fischer, D. U. Pfeiffer, G. C. W. England, D. E. Noakes and I. M. Sheldon. 2005. Clinical evaluation of postpartum vaginal mucus reflects uterine bacterial infection and the immune response in cattle. Theriogenology 63: 102 117. Williams, E. J., D. P. Fischer, D. E. Noakes, G. C. England, A. Rycroft, H. Dobson, I. M. Sheldon. 2007. The relationship betw een uterine pathogen growth density and ovarian function in the postpartum dairy cow. Theriogenology 68:549 559. Wiltbank, M. C., H. Lopes, 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. Woad, K. J., A. J. Hammond, M. Hunter, G. E. Mann, M. G. Hunter, R. S. Robinson. 2009. FGF2 is crucial for the development of bovine luteal endothelial networks in vitro. Reproduction 138: 581 58 8. W oessner, J. F. 1998. The matrix metalloproteinase family. In Matrix Metalloproteinases. Eds W. C. Parks, R. P. Mecham. Academic Press. pp 1 4. Wolf, E., G. J. Arnold, S. Bauersachs, H. M. Beier, H. Blum, R. Einspanier, T. Frohlich, A. Herrler, S. Hiendled er, S. Kolle, K. Prelle, H. D. Reichenbach, M. Stojkovic, H. Wenigerkind, and F. Sinowatz. 2003. Embryo maternal communication in bovine strategies for deciphering a complex cross talk. Reprod. Dom. Anim. 38:276 289. Xu, Z., H. A. Gaverick, G. W. Smith, M F. Smith, S. A. Hamilton, and R. S. Youngquist. 1995. Expression of follicle stimulating hormone and lutenizing hormone receptor messenger ribonucleic acid in bovine follicles during the first follicular wave. Biol. Reprod. 65:951 957.

PAGE 300

300 Yamashita, H., D. Kamada, K. Shirasuna, M. Matsui, T. Shimizu, K. Kida, B. Berisha, D. Schams, A. Miyamoto A. 2008. Effect of local neutralization of basic fibroblast growth factor or vascular endothelial growth factor by a specific antibody on the development of the corpu s luteum in the cow. Mol. Reprod. Dev. 75:1449 56. Yang, L., X. L. Wang, P. C. Wan, L. Y. Zhang, Y. Wu, D. W. Tang, S. M. Zeng. 2010. Up regulation of expression of interferon stimulated gene 15 in the bovine corpus luteum during early pregnancy. J. Dair y Sci. 93:1000 1011. Yuan, J. S., A. Reed, F. Chen, and C. N. Stewart Jr. 2006. Statistical analysis of real time PCR data. BMC Bioinformatics 7:85. Zemjanis, R. 1970. Diagnostic and Therapeutic Techniques in Animal Reproduction. Williams and Wilkins Co ., Baltimore, MD. Zhang, Z., Q. Fang, J. Wang 2008. Involvement of macrophage colony stimulating factor (M CSF) in the function of follicular granulosa cells. Fertil Steril. 90:749 54. Zhao, Y., H. Rong, N. Chegini. 1995. Expression and selective ce llular localization of granulocyte macrophage colony stimulating factor (GM CSF) and GM CSF alpha and beta receptor messenger ribonucleic acid and protein in human ovarian tissue. Biol. Reprod. 53:923 30. Zuluaga, J. F., and G. L. Williams. 2008. High pre ssure steam sterilization of previously used CIDR inserts enhances the magnitude of the acute increase in circulating progesterone after insertion in cows. Anim. Reprod. Sci. 107:30 35.

PAGE 301

301 BIO GRAPHICAL SKETCH Fbio Soares de Lima was born on January 30, to Gaspar de Lima and Dolores Soares de Lima in Muzambinho, Minas Gerais, Brazil. He is the third of four children that grew up o n a dairy farm surrounded by cows from an early age. After graduation from high school he wa s accepted and enrol led in the College of Veterinary M edicine at So Paulo State University, where he had an opportunity to work under the supervision of Dr. Jos Luiz Moraes Vasconcelos. In 2004, during the clinical year of his veterinary studies he spent 5 months at the Un iversity of Wisconsin, in Madison working on research projects that involved physiology of reproduction, health and reproductive management of dairy cows under the supervision of Dr. Milo Wiltbank. After his graduation from So Paulo State University, he w orked for 18 months at the University of California, Davis in the Veterinary Medicine Teaching and Research Center in Tulare, under the supervision of Dr. Jos Eduardo P. Santos, conducting research in the areas of dairy cow nutrition, reproduction and he alth. In July 2006, he was accepted as clinical resident in the Food Animal Reproduction and Medicine Program in the C ollege of Veterinary Medicine at the University of Florida. In 2007, he started a Master of Science program in Clinical Sciences under the supervision of Dr. Carlos A. Risco that was concluded in 2009. After completion of his MS c he remained at University of Florida and received his PhD degree from the Department of Animal Sciences under advisement of Dr. Jos Eduardo P. Santos. After comple tion of his PhD degree he will move to Cornell University to work as post doctoral associate with Dr. Rodrigo Bicalho